Assessment of children, revised and updated third edition

Historical Landmarks in Cognitive and Educational Assessment

Date

Event

2200

Chinese emperors set up civil service programs.

circa

Fitzherbert (1470-1538) proposed a test of mentality consisting of counting 20 pence, telling one.'s age, and identifying one's father.

B.C.

1510

1575

circa 1610

1692

Date 1860

Fechner (1810-1887) gave mathematical form to the relationship Weber had found between stimulus strength and the strength of sensation.

1869

Galton (1822-1911) published Classification of Men According to Their Natural Gifts, which stimulated the study of

Huarte (1530-1589), a Spanish physician, published Examen de Ingenios (The Tryal of Wits) in which he suggested formal mental testing and defined intelligence as docility in learning from a master, understanding and independence of judgment, and inspiration without extravagance.

1879

W undt (1832-1920) founded the first psychological laboratory in Leipzig, Germany.

1888

Swinburne (1560-1623) proposed that persons held on criminal charges be examined for their ability to measure a yard of cloth or name the days of the week.

J. M. Cattell (1860-1944) opened a testing laboratory at the University of Pennsylvania; his work helped to establish the foundations of mental measurement in the United States.

1897

Ebbinghaus (1850-1909) studied tests of arithmetic, memory span, and sentence completion.

1904

Pearson (1857-1936) formulated the th�ory of correlation. Spearman (1863-1936) introduced a two-factor theory of intelligence that posited a general factor (g) and specific factors (s).

T homasius (1655-1728), a German philosopher, published books on the use of quantitative data to obtain knowledge of men's minds.

mental inheritance and individual differences; he is considered the founder of individual psychology.

1799

Itard (1775-1838), in his work with the W ild Boy of Aveyron, assessed differences between normal and abnormal cognitive functioning.

1904

1809

Gross (1777-1855) developed a theory concerning errors in observations.

1905

1823

Event

Bessel (1784-1846) investigated differences between the observations of two astronomers and accounted for the differences by means of each observer's "personal equation."

Binet (1857-1911) and Simon (18731961) developed a useful intelligence test

for screening school children. 1905

E. L. T horndike (1874-1949) studied animal intelligence, formulated laws of learning, and developed principles of test construction.

1834

Weber (1795-1878) studied thresholds of awareness, particularly how observers detect differences between stimuli.

1909

Goddard (1866-1957) translated the Binet-Simon Scale from French into English.

1835

Quetelet (1776-1874) studied normal probability curves.

1912

Stem (1871-1938) introduced the concept of mental quotient.

1837

Seguin (1812-1880) established the first successful school for mentally retarded children and devised the Seguin Form Board.

1916

1838

Esquirol (1772-1840) proposed several levels of mental retardation and differentiated mental retardation from mental illness.

Terman (1877-1956) published the Stanford Revision and Extension of the BinetSimon Intelligence Scale; with Merrill (18887.1978), in 1937, he issued � revision called the Stanford-Binet Intelligence/S�ale; other revisions were published in 1960 and 1972.,

I

Event

Date 1917

1918

1919

Yerkes (1876-1956), with colleagues, published the Army Alpha and Army Beta tests, which were group­ administered intelligence tests used for the assessment of, military recruits in the United States.

Date

Event

1938

Gesell (1880-1961) published the Gesell Maturity Scale.

1939

Wechsler (1896-1981) published the Wechsler-Bellevue Intelligence Scale; re­ visions were issued in 1955 and 1981 under the titles Wechsler Adult Intel­ ligence Scale and Wechsler Adult Intel­ ligence Scale - Revised.

Otis (1886-1964) published the Absolute Point Scale, a group intelligence test. Monroe (1863-1939) and Buckingham ) published theIllinois Ex­ amination, a group achievement test.

1940

(1899-

Cattell (1893) published the Cattell Infant Intelligence Scale.

1949

WechslerIntelligence Scale for Children was published; a revision was issued in 1974 under the title Wechsler Intel­ ligence Scale for Children-Revised.

Kohs (1890-1984) published the Kohs Block Design Test, a test of nonverbal reasoning.

1959

Guilford (1897) proposed a Structure of Intellect model of intelligence based on factor analytic methods.

Porteus (1883-1972) published the Por­ teus Maze Test.

1961

Kirk (1904-

1923

Kelly (1884-1961), Ruch (1903-1982), and Terman (1877-1956) published the Stanford Achievement Test.

1923

1924

1926

p.

Goodenough (1886-1959) published the Draw-A-Man Test.

(1927-

) and J. J. McCarthy

) published the Illinois Test

of Psycholinguistic Ability. 1963

R. B. Cattell (1905) proposed a theory of fluid and crystallized intelligence.

1928

Arthur (1883-1967) published the Point Scale of Performance Tests.

1931

Stutsman (1894-1980) published the Merrill-Palmer Scale of Mental Tests.

1969

Bayley (1899) published the Bayley Scales ofInfant Development.

1933

T hurstone (1887-1955) proposed a mul­ tiple factor analytic approach to the stlldy of human abilities.

1972

D. McCarthy (1906-1974) published the McCarthy Scales of Children's Abilities.

1975

U.S. Public Law 94-142 passed, pro­ claiming the right to equal education for all handicapped children.

1933

T iegs (1891-1970) and Clark (18951964) published the Progressive

Achievement Tests, later renamed the California Achievement Tests. 1936

Lindquist (1901-1978), with colleagues, published the Iowa Every-Pupil Tests of Basic Skills, later renamed the Iowa Tests of Basic Skills.

1936

Piaget (1896-1980) published Origins of

1979

tests used for the assessment of black children for classes for the educable mentally retarded are culturally biased.

1980

Judge Grady inIllinois ruled in Parents in Action on Special Education v. Joseph P. Hannon that intelligence tests are not racially or culturally biased and do not discriminate against black children.

1986

R. L. T horndike. (1910), Hagen (1915), and Sattler (1931) published the Stanford-BinetIntelligence Scale: Fourth Edition, in which a point­ scale format replaced the age-scale for­ mat of the Stanford-Binet Intelligence Scale: Form lrM.

Intelligence. 1936

1938

1938

Doll (1889-1968) published the Vine­ land Social Maturity Scale; a revision by Sparrow (1933), Balla (19391982), and Cicchetti (1937) was published in 1985 under the title Vine­ land Adaptive Behavior Scales. Bender (1897-1987) published the Bender Visual Motor Gestalt Test. Buros (1905-1978) published the first

Mental Measurements Yearbook.

Judge Peckham in California ruled in

Larry P. v. Wilson Riles that intelligence

ASSESSMENT OF CHILDREN _____

Revised and Updated Third Edition

Jerome M. Sattler

San Diego State University

;;r .---..-. . -..

.. � =

Jerome M. Sattle

.

Publisher, Inc.

San Dliego

_____

Copyright © 1988, 1990. and 1992 by Jerom'e M. Sattler. Pu lisher, Inc . P.O. Box 151677, San Diego, Cali fOt'llia 921 7 5 . A l l rights reserved, N o part of t h e material protected by thiS copyright notice may be reproduced or util ized in any form or by any means . electronic or mechanical, includ ing photocopying, recording, or by any information storage and r (formulate recommendations) communicate assessment findings both in writing and ora 'y read and interpret research in the field of clinical and psychoeducational assessment understand laws and government regulations concern­ ing the assessment and placement of special children •

•

•

•

•

•

•

•

•

This text will assist you in obtaining these skills. How­ ever. two caveats are in order. First, this text is not a substitute for test manuals or for texts on child psycho­ path logy; it supplements material contained in test man­ uals and summarizes some major findings in the area of chil ' psychopathology. Second, this text cannot substitute for clinically supervised experiences. Each student in training should receive supervision in all phases of assess­ ment, including test administration, scoring, report writ­ ing , and consultation. Ideally every student should examine, in a variety of settings, many different types of children - normal and emotionally disturbed, retarded and gifted, as well as the physically handicapped .

checklists is the Ninth Mental Measurements Yearbook (Mitchell , 1985 ) . Started by Buros, this yearbook is now in its ninth edition. The Standards for Educational and Psy­ chological Testing (American Psychological Association , 1985) is a useful source for information about technical and professional standards for test construction and use . In deciding whether to administer tests individually or on a group basis, you must consider the nature of the referral question and whether alternative assessment routes are available. How important is it that tests be administered individually? Would group tests be as effec­ tive as the individual battery in answering the referral question? Are there any motivational, personality, lin­ guistic, or physically disabling factors that may impair the examinee's performance on group tests? Individual tests should be administered when there is reason to question the validity of the results of group tests or when an ex­ tremely careful evaluation of the examinee's performance is needed . You should discuss the reasons for individual assessment with referral sources in order to familiarize them with its values and limitations. Such discussions may lead to a reduction in the number of children unnecessarily referred for individual assessment, making it easier to provide prompt and intensive services to children most in need of individual assessment . Before beginning the assessment, you may need to con­ fer with the referral source, especially if the referral infor­ mation is incomplete. In preparation for the formal assess­ ment it is helpful to read background informatio n ; interview teachers , parents, o r other individuals familiar with the child; and study cumulative records . The tests and other assessment procedures finally selected - the assess­ ment battery - should be geared to the needs of a specific child . The tests selected may be changed from child to child, depending on the referral problem. Additionally, you may want to change or add tests or other assessment procedures as a result of the information you obtain from the initial assessment procedures administered. Establishing Rapport and Administering and Scoring the Assessment Battery

Evaluating and Selecting an Assessment Battery

A prerequisite for effective assessment is the ability to plan asses,ment strategies and to choose tests to meet specific needs. Table 1-1 presents some useful guidelines for eval­ uating tests and other assessment procedures. These guidelines indicate that it is important to consider the information contained in test manuals, including reliabil­ ity, validity, and normative data. An excellent source to consult for information on a vast number of tests and

This text contains information on establishing rapport and administering tests, including sections on working with special children, completion of record booklets (pro­ tocols), general administrative procedures and specific test procedures, and examiner-examinee variables . This infor­ mation should be used in conjunction with test and assess­ ment manuals. Practical administrative suggestions, based on clinical experience and research findings, are also pro­ vided for various assessment procedures.

8

CHAPTER I

CHALLENGES IN ASSESSIN G CHIILDREN

Table I - I Guidelines for Evaluating a Test InfoT7TUltion About the Test 1 . What is the name of the test? 2 . Who are its authors? 3 . Who published it?

4. 5. 6.

When was it published? Is there an alternative form available? How much does it cost?

7. How long does it take to administer? 8. Is there a test manual?

9. How recently has the test been revised? 1 0 . What was the standardization group?

3. Is the vocabulary level of the test's directions approplriate for the examinee?

4.

How are the test items presented?

5. 6.

What stated and unstated adaptations can be made im

How are the test items responded to? presentation and response modes?

7. Is the test free of sex and ethnic biases?

8. Are the test materials interesting to the examinee?

9. Is the test suitable for individual or group administn.ation?

Reliability and Validity

Aids to Interpreting Test Results

1 . How reliable are the test and its component parts? 2. How valid is the test for its stated purposes?

I.

Administration and Scoring

Does the manual provide a clear statement of (a) the purposes and applications for which the test is intended and

(b)

the qualifications needed to administer the test and

interpret it properly?

2 . Do the test, manual, record forms, and accompanying materials guide users toward sound and correct interpreta­ tion of the test results?

3 . Are the statements in the manual that express relationships presented in quantitative terms , so that the reader can tell how much precision or confidence to attach to them?

1.

Scales and Norms 1 . Are the scales used for reporting scores clearly and carefully described?

2. Are norms reported in an appropriate form, usually standard scores or percentile ranks?

3 . Are the populations to which the norms refer clearl� defined and described?

Examinee Considerations 1 . What prerequisite skills are needed by the examinee to complete the test?

2 . In what languages or modes of communication can the test be administered?

Are the directions for administration complete and dear?

2 . Are the scoring procedures clear?

4.

If more than one form is available, are tables a:vailalble

5.

Does the manual discuss the possible value of local norms

showing equivalent scores on the different forms? and provide any help in preparing local norms?

Source: Adapted from Sal end ( 1 984) and Thorndike and Hagen ( 1 977).

When you administer a test, you must score the child's responses to specific items as you progress. To score tests accurately, you need to be aware of the research findings concerning scoring bias, "halo" effects (the tendency in making a judgment about one characteristic of a person to be influenced by another characteristic or general impres­ sion of that person) , common errors in test scoring , and use of scoring criteria. These findings are reviewed i n this text. Of course it is imperative that you understand fully the scoring principles and criteria described in the test manuals. Similar considerations are involved in interview­ i ng , observing behavior, and i n formal assessment procedures . Observing Behavior

An important part of the assessment process is observing the child's behavior during the testing session. Knowing

how the child performs on the various tasks helps im indi­ vidualizing the clinical evaluation and supplemremts the more objective evaluation. Behavioral observatioms , focus on such areas as the child's interpersonal relati'oms , at­ titude, language , motivation , and motor skill s . Whiile ad­ ministering intelligence tests and other types of formlal and informal ability tests, you can observe how childnem solve conventional and novel problems, how well they atttend to tasks, and how persistent they are in solving pr·otblems. Specific suggestions are made i n this text for obserwiIng and evaluating the child's behavior during assessment. . Interpreting the Assessment Results

After the tests have been administered and the imteerview and observation completed, you must interpret tthee find­ ings . You will find this one of the most challengi:ngg of all the assessment activities, as it will draw on youII' k knowl-

9

BEHAVIORAL OBJECTIVES OF THIS TEXT

edge of children , test theory, developmental psychology, personality theory, and psychopathology. Various parts of this text will assist you with interpreting assessment results.

u

Formulating Recommendations

t

I

In addition to interpreting the assessment findings, you will often be asked to formulate recommendations . Your knowledge of clinjcal psychology, assessment theory, de­ velopmental psychology, psychopathology, and psycho­ educational remediation procedures will aid you in per­ forming this task. This text provides some information on the formulation of recommendations. However, it is not designed to systematically cover remediation procedures or i ntervention strategies. Such knowledge must be gained from texts and journals that cover clinical psychology, rehabilitation, and educational strategies for use with spe­ cial children, and from clinical experience. Communicating the Assessment Results

The findings from formal and informal tests , interviews, and observations can make a unique contribution to the study of child development and of exceptional children. The value of the contribution you make, however, will depend on your ability to communicate your findings clearly and meaningfully. The assessment report may be read by any number of professionals (including teachers, p sychiatrists, probation officers , pediatricians, ne u ­ rologists , social workers , and attorneys), by the child's parents, and in some cases by the older child as well . It should be written shortly after the assessment battery has been administered, scored, and interpreted and should contain your findings, interpretations, and recommenda­ tions . Chapter 23 provides detailed guidelines for writing reports. One of the best ways to learn how to write a report is to study reports written by competent clinicians . Several reports are included in this text to demonstrate different styles and approaches to the evaluation task. In spite of their differences they all exhibit clear writing, skill ful analysis , and good clinical judgment. However, these re­ ports should be used only as general guides ; you should develop your own style and approach to report writing . This text also presents examples of common communica­ tion problems encountered in report writing. After the report has been written , you may want to meet with the child, the child's parents, and the referral source to discuss the results. You may be called on to present your results at a staff conference, which will require skill in

"It's about Benny, doctor. He 's just come from school with an

10 of

1 041 Should I put him right to bed? "

Copyright 1 955, Cowles Magazines, Inc.

explaining your findings and recommendations . If chil­

dren or parents are at the conference they may be anxious

and defensive, in which case you will need to be able to deal with their reactions . Being thoroughly familiar with the subject matter presented in this text will enable you to present and defend your findings clearly and systemat­ ically. The assessment should not stop with the report. Every attempt should be made to work with the child, the parents, and the referral source on a continuous basis . The interventions recommended in the report need to be monj ­ tored carefully and modified as necessary. Conducting Research on Assessment

An understanding of research findings will add greatly to your understanding of assessment procedures and the as­ sessment process; progress in assessment cannot occur without research . Many factors bear investigation - for example, the reliability and validity of tests, interviews , and observations for various populations need t o b e deter­ mined, clinical and educational theories should be evalu­ ated , procedural changes in test administration require further study, and different remediation strategies need to

10

be evaluated . Familiarity with research findings and with the problems involved in such research will provide you with a base from which you can evaluate your own assess­ ment techniques and published research reports . Finally, you may find yourself doing research on intelligence test­ ing, tests of special abilities, interviewing, and observa­ tion. Much of the material in this text is relevant to this task. A knowledge of research and research problems may stimulate you to design and conduct your own investigations . Understanding Laws and Government Regulations

Working with special children requires knowledge of state and federal regulations concerning assessment pro­ cedures, particularly if the assessment involves educa­ tional programs. Although only a small part of this text addresses regulations such as Public Law 94-142, those of you who plan to work with school systems as psychologists should become thoroughly familiar with relevant state and federal regulations. Regulations currently cover such areas as nonbiased assessment, classification of handicapping conditions, eligibility for special programs, individualized educational programs, rights of parents, and confiden­ tiality of records .

CHAPTER I

CHALLENGES IN ASSESSING CH LDREN

ral reinforcers, and distractors - is one of the hal l marks of the behavioral approach . The multi method approach to assessment in this text attempts to wed the best traditions of the normative-developmental and behavioral assessment approaches . Normative data are usefu l in various ways (cf. Edelbrock, 1984) . First, in epidemiological surveys, nor­ mative data help us to determine what constitutes normal or deviant functioning for particular age groups and sexes. Data on the distribution of scores on various te ts and on the incidence and prevalence of behavioral problems help set standards for determining significant cognitive, affec­ tive, and behavioral deviations . Second, normative data help us to select appropriate target areas and behaviors that need change and remediation . Normative data facilitate the process of identifying the child's areas of strength and weakness and behaviors that are normal or deviant. Third, normative data allow us to make comparisons among dif­ ferent tests and different informants. For example, com­ paring data from different ability and achievement tests helps us to identify those children who are gifted or re­ tarded across divergent areas , and comparing data from parents and teachers helps us to identify children who are deviant both at home and at school .

CONCLU D I N G C O M M E N T O N P H I LOSOPHY O F T H I S TEXT

This text integrates the normative-developmental perspec­ tive with the behavioral perspective in the assessment of children. Children change as they grow, and these changes are reflected in the cognitive, affective, and behavioral domains (Edelbrock, 1984) . The normative-developmental perspective emphasizes the need to evaluate these domains in relationship to (a) the child's age, gender, ethnicity, socioeconomic status (SES), and other potentially impor­ tant demographic variables and (b) the child's prior devel­ opment in these domains and its influence on current and future development. Thus, the normative-developmental approach evaluates the child's cognitions, affect , and be­ havior in relation to a reference group (usually composed of children of the same age as the referred child) and attempts to account for changes as the child grows older. The behavioral perspective focuses on the importance of (a) specifying intraindividual (within-the-individual) changes in behavior, (b) selecting specific target behaviors for change, and (c) accounting for environmental deter­ minants that may shape and control behavior. The careful specification of environmental contingencies that relate to behavior- such as factors associated with the setting, natu-

CHALLE N G E S I N ASSESS I N G CHILDREN

An individual assessment is a rare situation in that ;a highly able and skilled professional devotes his or her e)xclusive attention to one child for a period ranging from 1 tOJ several hours. This may never have happened before in thle child's life, and it may never happen again . As an examimer, you have the opportunity to gather valuable information about the child that will be useful for assessment and intelTvention purposes. The assessment procedures covered in this book< serve a number of different purposes, including screening, classi· fication/placement, and program planning/remediation . Assessment procedures vary in the extent to whichl they an useful in meeting one or more of these purposes .. As yOl read about these assessment procedures, keep in rmind thf different purposes that they may serve and beconne aware of the purposes for which each procedure is b-es;t suited This text provides guidelines designed to prormote thl usefulness and fairness of assessment procedure s. I believI that clinical and psychoeducational assessment re!present one of the most effective ways of promoting h{e menta health and educational needs of children from 3111 ethni·

I I

SUMMARY

backgrounds. Each child r ep re se n ts a separate challe'nge for the examiner ; this book aims to i nc rea se your abil ity to rise effectively to the challenge .

S U M MARY

I . This text introduces i mportant assessment procedures needed to evaluate children . 2 . The multiple assessment approach advocated in the text covers four important assessment procedures: norm-referenced tests, interviews, observations , and informal tests. 3 . Norm-referenced tests provide valuable information about many areas of development, but should be used only in conjunction with other sources of data. 4. I nterviews are a useful way to obtain information in a less structured format. 5. Observations provide valuable information about how the child functions in various settings. 6 . Informal assessment procedures may provide additional information about the child's learning ability, language style, and other areas of functioning. 7. I n formation obtained from all sources used in the assess­ ment should be evaluated and major discrepancies resolved be­ fore diagnostic decisions and recommendations are made. 8. Developmental norms are critical in the assessment of children. 9 . Parents and significant others must be included in any diagnostic work with children. 1 0 . The assessment process should never focus exclusively on test scores. I I . Tests are powerful tools and must be thoroughly under­ stood before they are used as assessment instruments. 1 2 . The child's competencies as well as deficiencies should be evaluated in any assessment . 1 3 . Much remains to b e learned about the assessment of intelligence. 1 4 . Label i ng and classifying, although important, must not impair your ability to see the child as a unique individual . 15. Ethnic and cultural differences must be considered in evaluating assessment results. 16. Testing has become a public concern, particularly in rela­ tion to the assessment and placement of ethnic minority children. The attacks on testing during the last 15 years should make us doubly vigilant about how we use tests.

1 7 . Competencies needed to become a skilled clinical as­ sessor include the ability to evaluate, select, administer, and i nterpret assessment tools; the ability to communicate one's find­ ings orally and in writing ; and the ability to interpret research findings. 1 8 . This text emphasizes the integration of the nornlative­ developmental and behavioral perspectives in the assessment of children .

KEY T E RM S , CONCE PTS, A N D NAMES

Daniel Hoffman v. The Board of Education of the City of New York (p. 2 ) Norm-referenced tests (p. 3 ) Norm group (p. 3 ) Interviews ( p . 4) Observations (p. 4) Informal assessment (p. 4) Multiple assessment approach (p. 5) Labels and classifications (p. 5) Cultural diversity (p. 6) Clinician (p. 6) Normative-developmental perspective (p. 10) Behavioral perspective (p. 10)

STUDY Q U ESTI O N S I . What relevance does the case of Daniel Hoffman have t o the practice of school and clinical psychology? 2. What are the fou r pillars of assessment and how do they complement one another? 3 . What are some important guidelines for using intelligence and special ability tests? 4. Why has testing come under attack and how has it been defended? 5. What technical and clinical skills are needed to become a competent clinical assessor? 6. What are the major parts of the assessment process? 7. What are some important guidelines for evaluating tests? 8 . Compare and contrast the normative-developmental per­ spective with the behavioral perspective.

USE F U L STATISTI CAL AN D M EASU REM E NT CONCEPTS

Descriptive Statistics

We conquer the facts of nature when we observe and experiment upon them. When we measure them we have made them our servants. A little statistical insight trains them for invaluable work.

Multiple Correlation Norm-Referenced Measurement

-Edward L. Thorndike

Derived Scores Statistical Significance Reliability Validity Factor Analysis Test Your Skill Summary

12

13

DESCRIPTIVE STATISTICS

This chapter covers statistical and measurement concepts that will enable you to evaluate more effectively the psy­ chometric properties of intelligence tests, special ability tests , and interview and observational data. (Psycho­ metrics is the quantitative assessment of an individual's psychological traits or attributes . ) Additionally, your knowledge of these concepts will enhance your under­ standing of test manuals and research reports . This chapter reviews basic statistical and measurement concepts ; it is not meant to be a substitute for texts in s tatistic s or testing and measurement.

DESCRIPTIVE STATISTICS

Descriptive statistics summarize data obtained on a sample of individual s . Areas covered in a study of descriptive

statistics include scales of measurement, measures of cen­ tral tendency, measures of dispersion, the normal curve, and correlations. The more commonly used symbols in statistics and psychometrics are shown in Table 2-1. These symbols are a shorthand way of describing important char­ acteristics of a test formula or norm group. Scales of Measurement

Data can be ordered by various methods. In most cases, one of four types of scales is used: nominal , ordinal , interval , or ratio. A scale is a system for assigning values or scores to some measurable trait or characteristic. These values then can be subjected to various mathematical pro­ cedures in order to determine relationships between the traits or characteristics of interest and other measured behaviors .

Table 2- 1 Some Common Statistical and Psychometric Symbols and Abbreviations

Definition

Symbol a

Intercept constant in a regression equation Slope constant in a regression equation

b

c

Any u nspecified constant

CA

Chronological age

DQ

Developmental Quotient

f

Frequency

F

Test statistic in analysis of variance and covariance

Symbol S2

SEes, SEm , SEM, SEmeas' Sm' Sm ' smeas' or sefT

Definition Variance of the sample Standard error of estimate

Standard error of measurement

t test T score;

IQ

Intelligence Quotient

M

Mean (see also

MA

Mental age

Md or Mdn n N p P

Median

Percentile

Y

Q

Semi-interquartile range; half the difference

z

z

a

Standard deviation of a population

E

"The sum of"; EX means add up all the Xs

cp

Phi coefficient; a correlation coefficient for a

X2

Chi square

>

Greater than

X)

Raw score

X

Mean

Third quartile score (75th percentile score) Pearson correlation coefficient Coefficient of determination; the proportion of

'p b

or p

r xx

rxy

variance in Y attributable to X Point biserial correlation coefficient Spearman rank-difference correlation coefficient (also referred to as rho) Reliability coefficient Validity coefficient (X represents test score and Y the criterion score)

R S, s , or SD

Coefficient of multiple correlation Standard deviation of the sample

X;

10 indicates how far the

score falls above or below the mean of the X

Q3

-

group

Number of cases in a sample

First quartile score (25th percentile score)

's

Deviation score X

(a) Probability; (b) Proportion

Q, ,2

x

standard score with a mean of 50 and

standard deviation of

Number of cases in a subs ample

between Q3 and Q,

,

T

a2

< 2:: :S

±

..r

A second raw score score; standard score with a mean of 0 and standard deviation of 1

Variance of a population (scores) 2

x

2 contingency table

Less than Greater than or equal to Less than or equal to Plus or minus Square root

CHAPTER 2

14

Nominal scales. A nominal measurement scale consists of a set of nonordered categories , one of which is assigned to each item being scaled. The numbers, letters, or names ordinarily represent mutually exclusive categories and thus cannot be arranged in any meaningful order. An example of nominal scaling is the assignment of numbers to baseball players or the designation of examinees as male or female. The scale is of limited usefulness because it allows only for classification. Ordinal scales. An ordinal measurement scale has the property of order. The variable being measured is ranked or ordered without regard for differences in the distance between scores. An example of ordinal scaling is the ranking of persons from highest to lowest on the basis of class standing. An ordinal scale tells us who is first, second, and third; it does not tell us whether the distance between the first- and second-ranked scores is the same as the distance between the second- and third-ranked scores. Interval scales. An interval measurement scale has an arbitrary zero point and equal units. Examples of interval scales are the Celsius scale, which gives temperature , and the Wechsler Intelligence Scale for Children - Revised (WISC-R) , which reports intelligence test scores. On the WISC-R, an increase of lO IQ points from 100 to 110 means the same amount of change as an increase from 120 to 130 or from any other point to the point 10 points higher on the scale. However, it makes no sense to say that a child with an IQ of 150 is twice as intelligent as a child with an IQ of 75 . I nterval scaling does not permit such comparative statements . Ratio scales. A ratio measurement scale has a true zero point, equal units, and equality of ratios. Because there is a

FJ
MOMMA, l

AND VOLA ONL.Y HAVE DINNER WITH ME ONCe A WEEK !

meaningful zero point, a true ratio exists between mea­ surements made on a ratio scale. Weight is one examp.}e of a characteristic measured on a ratio scale; an indivil dual who weighs 150 pounds is twice as heavy as one who weighs 75 pounds. Because most psychological chrurac­ teristics do not permit the measurement of an absol ute zero point (such as "zero intelligence") , ratio scales are found infrequently in psychology. In many cases we must be content with interval scales or the weaker ordinal sc'ales .

Measures of Central Tendency

The three most commonly used measures of central ten­ dency - the typical or representative score - are the mlean , median, and mode . Mean. The mean is the arithmetic average of alii the scores in a set of scores. To obtain the mean, divide the� sum of all the scores by the total number of scores in the set · (N) . The formula is as follows:

where

=

mean of the scores sum of the scores N = number of scores. Example: The mean for the four scores 2 , 4, 6, 8 is

M

EX

=

M

=

20

4

=

5

The mean is responsive to the exact position of each score in a distribution, but it is also sensitive to aa few

YE5, etA. HEI. Infant scales have limited : for hahahahahahahalan ana(, 1aJ h'handicapped children, be­ lin are P! p� P! P! p! P! P pi pI pt pre e primarily of a perceptual­ intelligegegegeg egege�e ge ge rge l ig(gence tests have more con­ ability, y, y, y, y, y, y, 1, c,
92

CHAPTER 5

TESTING C H I LDREN

Table S-2 Behavior and Attitude Checklist Behavior and Attitude Checklist

Child's name: Age: Test(s) administered : IQ:

Examiner: Date of report: Date of examination: Grade:

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

__ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __

__ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __

X

__ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

Instructions: Place an

on the appropriate l ine for each scale,

I, Allitude toward examiner and test situation: I , cooperative 2, passive 3 , tense 4, gives up easily II, Attitude toward self: 5 , confident 6 , critical of own work Ill, Work habits: 7 , fast 8 , deliberate 9 , thinks aloud 1 0 , careless IV Behavior: I I , calm V. Reaction to failure: , 1 2 , aware of failure 1 3 , works harder after failure 1 4 , calm after failure 1 5 , apologetic after failure VI, Reaction to praise: 1 6 , accepts praise gracefully 1 7 , works harder after praise VII, Speech and language: 1 8 , speech poor 1 9 , articulate language 20 , responses direct 2 l , converses spontaneously 2 2 , bizarre language VJll, Visual-motor: 2 3 , reaction time slow 24, trial-and-error 25 , skillful movements lX, Motor: 2 6 , defective motor coordination Overall test results: 2 7 , reliable 2 8 , valid -

· '

-

-

· '

-

-

'

-

-

· '

-

-

· '

-

-

· '

-

· '

-

-

· '

. '

. '

. ' . '

-

. '

. '

. '

. '

-

. '

. '

-

. '

-

. '

-

. '

-

. '

-

. '

-

'

-

-

. '

-

. '

-

· '

-

. '

-

. '

-

'

-

· '

-

· '

-

· '

-

· '

-

· '

-

-

-

-

-

· ' · '

· '

· '

· '

. '

-

. '

-

-

-

-

,

'

. ' . '

-

. '

-

. '

-

-

-

. '

. '

. ' . ' . '

-

,

'

. '

-

. '

-

. '

-

. '

-

-

. '

-

. '

-

. '

-

. '

-

. '

-

. '

-

. '

-

. '

-

. '

-

hyperactive

'

-

unaware of failure gives up easily after failure agitated after failure not apologetic after failure

-

'

. '

-

. '

-

-

-

-

. '

. '

. '

. ' . '

. '

-

. '

'

-

'

-

. '

-

'

-

-

'

-

. '

· '

-

. '

-

. '

-

. '

-

. '

-

'

-

-

. '

not confident accepting of own work

-

-

. '

-

. '

. '

. '

-

uncooperative aggressive relaxed does not give up easily

-

-

-

-

. '

-

-

-

-

-

. '

-

'

. '

. '

-

-

. '

-

. '

-

-

-

-

. '

. '

. '

-

. '

-

. '

-

. '

. '

. '

-

. '

-

. '

-

. '

. '

. '

-

slow impulsive thinks silently neat

. '

-

. '

-

-

· '

. '

. '

-

. '

. '

. '

-

. '

-

-

'

-

-

--

-

· '

· '

--

. '

-

-

X.

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

__ __ __ __ __ __ __ __ __ __ __ _ __ __ __ __ __ __ __ __

___ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __

-

need to meet with extremely shy children on more than one occasion before they are ready to respond to the test questions ,

-

-

-

. '

-

. '

'

. '

. '

. '

. '

. '

. '

-

-

-

. '

. ' . '

-

-

-

. '

-

-

. '

-

-

-

. '

-

. '

-

. '

-

. '

-

. '

. '

. '

'

. '

. '

-

. '

. '

-

. '

-

-

accepts praise awkwardly retreats after praise speech good inarticulate language responses vague only speaks when spoken to reality-oriented language

-

. '

-

. '

-

reaction time fast careful and systematic awkward movements

. '

-

. '

-

. '

-

good motor coordination

. '

-

. '

-

. '

-

unreliable invalid

. '

. '

-

-

. '

. '

-

-

. '

. '

-

-

Delinquent children are likely to experience not only the situational stress induced by the authoritarian setting in which testing usually takes place (such as a juvenile deten-

93

TESTING AUTISTIC CHILDREN

tion center or jail) , but also the real and immediate threat posed by the examination itself. Delinquent children may feel that they have everything to lose and nothing to gain by revealing themselves to you . Testing might lead to place­ ment in an institution or foster home, or some other dis­ position they regard as punishment. As a consequence, they may seek to defend themselves against self-revela­ tion. You must make every attempt to understand and work through these difficulties; if not, a valid picture of the delinquent's abil ities will not be obtained.

TESTI NG AUTISTIC C H I LDREN

The inherent handicaps of autistic children - such as their difficulty in establishing social relationships, their im­ paired communication skills, and their unusual responses to sensory stimuli - are likely to create difficulties in the testing situation and tax your resources as an examiner. These children may show little or no desire to do the tasks or to interact with you. Your normal methods of encour­ agement, such as smiling , may be ineffective (Baker, 1983). Talk slowly and simply to the autistic child - use short sentences and omit unnecessary words and complex grammatical forms . "Be concrete and specific . . . . If the child does not respond, repeat your question or direction in the same or a simpler way" (Lord & Baker, 1977, p. 184) . Make sure you have the child's visual attention when you speak; visual cues help the child attend to and process your speech.

Learn About the Autistic Child's Communication Skills Before the assessment begins, find out as much as possible about the child's communication skills from parents, teach­ ers , and your own classroom observations. Consider the following questions (Lord & Baker, 1977) : Can the child follow simple directions? Can the child answer yes or no? • Does the child understand gestures or pictures or signing? Can the child read? Does the child have any individual idiosyncrasies, such as using code words or phrases ("bye-bye" for "no" or "look, look" for a favorite toy)? •

•

• •

Also, Watch how people who know the child talk to him or her. Parents or teachers may assure you that the child understands everything

they say. However, they may actually use frequent dramatic gesture, or physically guide the child through tasks. When work­ ing with the child , experiment . . . with directions or questions of varying complexity. (Lord & Baker, 1 977 , p . 184)

Consider the Autistic Child's Resources In conducting the assessment, consider the autistic child's preferences, abilities, handicaps, and individual style (Baker, 1983). Select tests that match the autistic child's developmental level. If necessary, start tests at a point well below the child's chronological age . Because of the autistic child's language problems, select tests that provide sepa­ rate measures of language and nOnlanguage skills or use a battery that contains both language and nonlanguage tests.

Language Deficits of Autistic Children The language deficits of autistic children tend to manifest themselves in four ways (Baker, 1983) . 1 . Receptive language skills may be better developed than expressive language skills. A child may understand directions , such as "Give me the pencil," but not be able to say "pencil" when he or she wants it bac k . 2 . Repetition o f rote mechanical phrases o r delayed echolalia (repetition of what you have said after a short delay) may not constitute meaningful language. A child who repeats the phrase "Go away" may not be able to use it in context. Just the same, immediate and delayed echolalia should not be viewed solely as empty repetition of words , phrases, or sentences with little or no social or commu­ nicative function. Research with a group of four autistic boys (Prizant & Duchan, 1981) indicated that many of their echolalic utterances showed evidence of comprehension, and that unfocused echoes were relatively rare , about 4 percent of the total number of utterances. The results suggest that immediate and delayed echolalia may have functional communicative elements and should be evalu­ ated in the context of other communicative elements, such as eye gaze, posture, and gesture . 3 . Language skills demonstrated in one setting may not generalize to another setting . For example, an object may be identified correctly only if it is like one the child has at home . 4. The child's language skills may not follow normal or expected developmental patterns . A child who does not talk still may be able to read . Because of these language problems , you may need to use alternative methods to present stimulus materials to autistic children (Baker, 198 3 ) . Visually clear and self-

CHAPTER 5

94

TESTING CHILDREN

explanatory materials are preferred , with gestures and direct physical guidance used when needed . Alternative modes of communication - such as pointing to pictures, sign language, written words, or other symbols - also may be used with children familiar with these methods. When you interpret test results and formulate recommendations, consider the autistic child's interest in completing tasks, ability to work independently, responses to social and behavioral interventions, strengths, and weaknesses .

material during the examination, may show perseveration (a persistent continuation of a response despite its inap­ propriateness given a change in test conditions), become emotionally labile, display inappropriate anger and hostil­ ity toward the materials or the examiner, withdraw from the situation, or give aberrant response s . These behaviors may represent attempts by the child to cope with a difficult situation. Although these behaviors may i nterfere with the goals of the examination , such reactions may serve as coping mechanisms to help the child avoid further stress.

TEST I N G BRAI N-I NJ U R E D C H I L D R E N

Useful Procedures for Handling Perseverative and Avoidance Behaviors

Brain-injured children may have a wide range of reactions during testing and show a variety of cognitive deficits (see Chapter 22) . Some may react to the evaluation like normal children, whereas others may be more fearful or reticent or show emotional lability (easy arousal and shift from one emotion to another) . If you know that a child is likely to be fearful of testing, be prepared to devote additional time prior to the formal testing to helping reduce the child's anxieties . You might try to gain the child's trust and confi­ dence by working with him or her on simple puzzles or game-like material s . In this regard, it i s often helpful to ask the parent about the child's reaction to learning that an evaluation was being planned. With the brain-injured child, praise, encouragement to keep going, and constructive comments should be used more frequently than would be necessary with the normal child. Start the assessment battery with relatively easy tests so that the child will experience an initial success that may help him or her to feel more at ease. Give more difficult tests in the middle of the session , after initial apprehension is dealt with but before fatigue sets in. On occasion, brain-injured children who know the an­ swer to a problem may require an inordinately long time to organize an appropriate response. They may sit quietly for a long time before responding, or they may make tentative, hesitant responses . If these behaviors occur, permit the children to proceed at their own pace ; do not urge them to respond. However, when the delay is excessive (over a minute) , the problem should be repeated if standard ad­ ministration allows, because the children may have forgot­ ten it. You may also wish to repeat the item at a later point in the battery. Before beginning the examination, minimize all poten­ tial sources of d istraction in the testing roo m . The room should be quiet , with objects and toys removed . Some brain-damaged children, when confronted with difficult

Perseverative and avoidance behaviors can be minimized by using the following procedures: I . Introduce testing procedures slowly and casually, and permit the child to play with toys similar to the test material s . 2 . Avoid any suggestions o f inadequacy in the child's performance . 3 . Avoid sudden movements or noises. 4. Introduce new materials gradually, reassuring the child that the new activities will be pleasant. 5 . When an emotional reaction occurs, do not overstim­ ulate the child by trying to talk him or her into feel ing better. Provide a relaxing respite to allow the child to work through his or her moment of anxiety. 6 . If perseveration occurs, redirect the child . 7 . If the emotional lability becomes too severe , stop testing and then sit quietly or go back to a test that the child has previously passed . 8 . Remind the child, as often as necessary, of a specific test instruction , such as "Point to the one that is not alike ," when he or she fails to recall the directions.

Useful Procedures for Working with Language­ Impaired Brain-Injured Children Communication will be difficult with brain-injured chil­ dren who have speech or language difficultie s . You may have to experiment with different communication meth­ ods, rates of communication, and types of content to find the most effective mode of communication . The following guide l ines w i l l promote better communication w ith language-impaired brai n-inj u red c h i ldren (Lubinski , 1981) : I.

Face the brain-injured child when speaking with

TESTING MENTALLY RETARDED C H I LDREN him or her. Eye contact promotes attention and help� the child take advantage of nonverbal cues. 2. Alert the child that communication is about to oc­ cur. For example, say the chi ld's name and a few word� of g reet i n g b e fore i n t roduc i n g a topic , question , or instruction . 3 . Speak slowly and clearly to the child . 4. Talk about concrete topics, such as objects and people in the immediate environment . The child may have di fficulty comprehending abstract ideas. 5. Keep related topics together. 6. Use short, syntactically complete utterances. The language-impaired brain-injured child may respond more accurately to individual ideas than to a lengthy string. This concept is particularly important when you are giving the child i nstructions or asking the child to respond to questions . 7 . Pause between utterances t o give the child time to comprehend and interpret the message. 8. Check the child's comprehension before proceed­ ing. Ask a question based on the information presented or h ave the child demonstrate his or her understanding of the information . 9. Repeat important ideas several ways. Redundancy helps comprehension. 1 0 . Use nonverbal cues to augment spoken communica­ tio n . Combining speech with nonverbal cues (such as ges­ ture, signing, and pictures) may facilitate comprehension. I I . Ask questions that require short responses or that can be responded to nonverbally. This will allow the child to have a sense of active participation even if his or her responses are only single words or gestures. 1 2 . Ask the child to repeat a word when speech is unintel ligible or confusing. Although the child may be­ come frustrated if asked to repeat a response many times , one or two repetitions can be helpful . 1 3 . Encourage the child to express ideas in several way s . This may facilitate his or her own thinking and verbal production as well as your comprehension . 1 4 . Encourage the child to use nonverbal cues, such as gesturing and pointing, to supplement attempts to speak about an object verbal ly. 1 5 . Present the child with a multiple-choice array (if necessary) from which the response can be chosen. For example, in trying to get the child to identify the source of pai n , you might ask questions like "Do you have a head­ ache?" and "Do you have pain in your chest?" until the child signals his or her choice . You may want to point or gesture as you provide the alternatives.

95

1 6 . Repeat what the child has said thus far as a means of focusing the conversation . 1 7 . When the child becomes frust rated by communica­ tion failures, discuss the difficulty openly. Recognize that the child is having a difficult time and go on to another topic. 1 8 . Monitor your nonverbal cues for indications of im­ patience or hostil ity. When the child has difficulty, try to show a relaxed demeanor. Nonverbal cues may tell the child a great deal about the effectiveness of his or her communication or lack of it.

TESTING MENTALLY RETA R D E D C H I LDREN

In cases where mentally retarded children are uncooper­ ative during the examination, evaluate the possible adap­ tive significance (as in the case of brain-injured children) of the uncooperative behavior. Negativistic behavior, for example, may enable the retarded child to maintain sel f­ esteem in the face of difficult intellectual or social de­ mands. Aggressive , hyperactive behavior may represent a child's emergency reaction to a novel situation involving difficult tasks . Echolalia may serve as a way of establ ishing and maintaining a relationship, even though it is maladap­ tive in other ways. Persistent questioning may represent an e ffo rt to ensure stabil ity. Perseveration may help the child to manage the situation . Denial may be used to cover vu l nerability. You can take a number of actions to cope with these behaviors . If mentally retarded chi ldren try to reverse roles, you can agree to take turns aski ng questions and thereby help them become more at ease and diminish their need for control . Aggressive and hyperactive behavior may be reduced by beginning the examination with easy questions . To prevent children from being continually con­ fronted with inadequacy, you can alternate between diffi­ cult tasks and easy ones. (Modifying test procedures , however, may produce less reliable results; see the end of this chapter for further discussion of test modifications . ) A s with all children, development o f a warm, accepting relationship will probably reduce unacceptable behavior.

Evaluating Severely and Profoundly Retarded Children Severely and profoundly retarded children may be difficult to evaluate because of self-stimulating behavior, self-

96

injury, brief attention span, destructive behavior, temper tantrums, seizures, or noncompliance with instructions or requests . Additionally, conventional standardized tests may prove to be of limited usefulness because they have a limited number of appropriate items. If this is the case, you may want to use a functional assessment approach (see Chapter 21), which links assessment directly to in­ struction. The challenge of assessing profoundly handicapped children is captured in the following commentary : One might say that profoundly handicapped children constitute the . . . most challenging group of humans to assess. . . . The profoundly handicapped do not readily respond to one's immedi­ ate suggestions, commands, and gestures. They are often immo­ bile, unattentive, disfigured, unpredictable, self-destructive, un­ responsi v e , and l ack comprehensible l anguage . They do constitute a chal lenge to those who elect to interact with them. Sometimes, just when one decides that they are "untestable, unteachable, or hopeless," they respond by displaying the behav­ ior which has been taught them for days or months! One asks what can be so rewarding as witnessing the child's first smile, or word l ike "milk," or a step forward, or dual response of "yes" or "no," or perhaps the swallowing of a semi-solid food? (Finkle, Hanson, & Hostetler, 1983, pp. 79-80)

Observing Profoundly Handicapped Children When you observe a profoundly handicapped child , pay close attention to the fol lowing (Finkle et a! . , 1983) :

Attention. Does the child attend to things in his or her environment? If so, what things? To what degree? For how long? • Movement. How does the child move from one part of the room to another? To what extent does the child need assistance in moving? Handedness. Which is the child's preferred hand? Does the child use two hands for activities? Deficits. What are the child's deficits? Strengths. What are the child's strengths? Compensatory behaviors. What compensatory behav­ iors are present? Reinforcers . What things in the child's environment serve as reinforcers? • Tonicity. How would you rate the child's general health and physical strength? Endurance. How long can the child take part in an activity? How readily does fatigue set in? •

•

•

•

•

•

•

CHAPTER 5

TESTIN G CHILDREN

TEST I N G PHYSICALLY H A N D I C A P P E D C H I LDREN: G E N E RA L SUG G ESTIONS

Many of the techniques used with normal children can be used to test physically handicapped children, but their application is more demanding when the examinees are physically handicapped . Normal children often need little encouragement. They are accustomed to answering test questions and are likely to find the tasks challenging. Physically handicapped children, in contrast, may feel at a disadvantage in the test situation. Their physical limita­ tions may make them appear clumsy and awkward, result­ ing in feelings of self-consciousness. They must cope not only with specific deficits associated with their disability (such as inadequate sight or heari ng ) , but also with the anxiety and uncertainty of their parents. Anxiety associ­ ated with repeated medical examinations, peer difficulties, and related factors may also have influenced their affective development and, indirectly, cognition and learning. Their reactions to the test situation may depend largely on how they perceive themselves outside of the test situation. They may be aware of their handicaps and reluctant to expose their disabilities in the examination . Patience and encour­ agement will be required to elicit optimal performance .

Potential Sources of Difficulty in Testing Physically Handicapped Children Testing physically handicapped children poses various problems. First, false impressions of the child's intellec­ tual ability may be induced by communication difficulties, such as speech and hearing deficiencies . Second, because handicapped children may be unaccustomed to concen­ trated work for long periods of time , they may fatigue easily. Third , it may be d ifficult to decide whether attention difficulties, when present, are associated with physical deficiencies, with medication (if used), or with cognitive deficiencies . Fourth, it may be difficult to establish rapport with physically handicapped children who have height­ ened dependency. Fifth, physically handicapped children may feel extreme pressure when a test is timed. Finally, it may require increased effort to ensure that physically handicapped children understand the test directions fully. In evaluating the cognitive abilities of physically handi­ capped children, do not confuse sensory deficits (for ex­ ample, visual , auditory, or tactile deficits) with cognitive deficits. A child may have one or more sensory or motor difficulties and yet have adequate cognitive abilities. If timed tests are used, supplement them with tests that are

TESTING PHYSICALLY HANDICAPPED CHILDREt\l: G r: N E RA L SUGGE:STI ONS not timed but that still adequately assess the children's skill repertoire . If children show undue stress when wor1�ing under time pressure, it may be preferable to discontinue administering the timed test. It is imperative that phys­

ically handicapped children not be penalized because of their sensory or motor deficits when they are administered ability tests. The tests selected must be such that handi­ capped children can respond at their optimal level . The tests should assess their cognitive abilities and not the extent of their physical abilities .

Informal Evaluation of Sensory Functioning Before you test a handicapped child, have the child's vi­ sion, heari n g , physical conditio n , and health status screened . Ask the parents of severely handicapped chil­ dren about signs, signals , or gestures that the child under­ stands or uses and their meaning. Through observation and informal testing, determine the degree to which the child is physically able to respond to the tests. This in­ volves informally evaluating the child's (a) vision, hearing, speech, sitting balance, and arm-hand use, (b) reading and writing skills (for a school-age child) , and (c) ability to indicate yes or no by either verbal or nonverbal means. The informal evaluation can be done by observing how the child performs such tasks as describing a picture , putting a puzzle together, reading and then writing a short sentence, and responding to simple questions. During these ac­ tivities, note how the child responds to your questions, performs motor tasks, and succeeds on the tasks.

Selection of Tests After you have become familiar with the child's problems and the limitations associated with his or her disabil ities. you should select tests that are appropriate and are geared to the child's strengths and limitations. It may be necessary to omit tests or items that require physical or sensory abil ities that the child does not have . For example, items that require deftness in object manipulation or drawing may have to be eliminated when children with severe handicaps of the upper extremities are tested.

Other Testing Considerations Additional valuable suggestions for working with phys­ ically handicapped children are as follows (cf. Ramirez , 197 8 ; Shontz, 1977 ; Wright, 1983) : •

Plan t o assess physically handicapped children on

97

more than one occasion and, if possible, in at least one setting that is familiar to the chil d . Blind children, for example, may demonstrate many skilled behaviors in an environment that is familiar to them but not in an un­ familiar setting. Give handicapped children time to accli­ mate themselves to the surroundings . Schedule , if needed, several shorter test sessions spaced over several days to counter the problem of fatigue . Position the handicapped child in his or her preferred way. Be sure that the child uses adaptive equipment that was made for him or her. Give credit for correct responses given in any form of communication, including sign language , Braille, fin­ gerspelling, and teletouch. Converse with a handicapped child as you would with any other child, with the same spirit , content, and approach . Simply ask the child who may need assistance , "Do you need help" or "How should I help you?" Do not shout at a blind child. The child has a visual problem, not a hearing problem. • Do not be embarrassed about common expressions that might seem awkward, such as asking a blind child whether he or she has seen a specific event. The English language is fi lled with these terms, and you are likely to be more sensitive to them than is the child with whom you are talking. Moreover, blind children commonly use the term see in reference to themselves. Use graphic language when directing an older, blind child. Instead of saying "The library is right over there," say "From where you are standing now, walk straight about 20 paces up a ramp, then walk 20 paces to the building, and then enter two sets of double doors that are about two paces apart and swing out . " Do not "talk over" o r provide the words for a child who stutters or has difficulty speaking. Be patient, listen , and let the child speak for himself or herself. You may repeat the thought back to the child to confirm communication of the idea, but it is not appropriate to outguess the child or assume that you know what the child is trying to express . Speak directly to a disabled child. D o not direct con­ versation to an attendant, assistant, or nearby companion as if the child did not exist. Always face a child with a hearing impairment. Be sure that the child can see your lips; speak clearly without exaggerating your lip movements . Do not call special attention to a disabled child. Ap­ proach the child as one who happens to have a handicap ­ not as a handicap that belongs to a child . •

•

•

•

•

•

•

•

•

•

•

CHAPTER 5 TESTING CHI LDREN

98

Do not allow the child's particular disability to bias your perception of the child's functioning in other areas (for example , hearing, general health, or emotional matu­ rity in the case of a blind child) . Do not overlook behaviors that do not fit in with your expectations. Remember that psychological reactions to the dis­ ability are not uniformly disturbing or distressing, do not necessarily result in maladjustment, and are not related in a simple way to the physical properties of the disability. • Recognize that the disabil ity is only one of many factors affecting the total l i fe of a child with a disability, and often its influence may be relatively minor. •

•

•

TESTING VISUALLY I M PA I R E D C H I LD R E N

Visual Impairment Defined Before you administer any tests to a visually impaired child, identify the extent of the child's visual impairment, the quality ofthe usable vision, and the extent of the field of vision. The term blindness encompasses a range from total lack of vision to the ability to see at 20 feet what persons of normal vision can see at 200 feet (20/200) . In some school districts special programs for visually handicapped chil­ dren may include those with 20170 acuity. Individuals may also be classified as legally blind if their acuity is good but their field of vision is so narrow that it subtends an angle no greater than 20 degrees .

Classification Scheme Visually impaired children can be classified according to three broad groupings (Bauman, 1974) : I . Vision is of no practical use in the test situation . This group includes the totally blind, those who can differenti­ ate only between l ight and dark, and those who can dis­ tinguish shapes, but only when the shapes are held between the eyes and the source of l ight . 2 . Vision can be used in handling large objects , locating test pieces, or following the examiner's hand movements during a demonstration, but cannot be used in reading even enlarged print effectively. These children may be able to use vision and touch in working with form boards, but are at a great disadvantage if they are required to read . 3 . Vision can be used to read print efficiently, but only

when the type is large, the page is held close to the eyes, or a magnifier or other special visual aid is used.

Assessment Considerations It is permissible to offer your arm to older blind children, or even to take the hand of young blind children, to lead them to your office . Inform them about the general layout of the room and about other details, such as the presence of a tape recorder, if one is being used . Take into account the amount of useful vision the partially sighted child has, adjusting the testing procedures accordingly. Be sure that partially sighted children wear their glasses when they are tested and that the lighting in the testing room is adequate (bright but without glare ) . In order to enable the child to form figure-ground discriminations, be sure that the sur­ face of the table is not white or the same color as the objects the child is expected to see.

Signs of visual difficulty. Signs that are suggestive of visual difficulty include the following: BEHAVIOR • Rubs eyes excessively

Shuts or covers one eye, tilts head, or thrusts head forward • Has difficulty in reading or in other work requiring close use of the eyes Blinks abnormally or is irritable when doing close visual work • Moves head excessively when reading Holds books too close or too far from eyes Is unable to see distant things clearly Squints eyel ids or frowns Loses place while reading Avoids close visual work Has poor sitting posture while reading • Has difficulty judging distances • Crossed eyes Red-rimmed , encrusted, or swollen eyelids Inflamed or watery eyes • Recurring sties •

•

•

•

• •

•

COMPLAINTS •

Eyes itch, burn , or feel scratchy

• Cannot see well • Dizziness, headaches, or nausea following close vi­ sual work • Blurred or double vision • Tires easily after visual work

TESTING HEARING-IMPAIRED CHILDREN Lack of these signs should not lead you to assume that the child's vision is adequate . Defects such as slight astig­ matism or a binocular visual defect can impede reading without observable symptoms being present . Whenever there is any doubt about the child's vision , referral to an optometrist or ophthalmologist is warranted .

Selecting tests.

If the visual handicap is not too severe , intelligence tests such as the WISC-R, Stanford-Binet : Fourth Edition, WPPSI , and WAIS-R may be used. If the handicap is severe , administer only verbal items. However, keep in mind that the ability to answer even verbal items may be reduced if information required by the items de­ pends in part on visual experiences . The Hays-Binet, Perkins-Binet, and Blind Learning Aptitude Test are also available for evaluating blind children.

TESTING H EARING·I M PA I R E D C H I LDREN

Hearing Impairment Defined

Hearing impairment is a general term that refers to hearing losses ranging from mild to profound. A deaf child is one whose hearing disability precludes successful processing of li nguistic information through audition . A hard-of­ hearing child has residual hearing sufficient for successful processing of linguistic information through audition, gen­ erally with the use of a hearing aid . One of the most serious problems of prelingually deaf children (children whose deafness was present at birth or occurred at an age prior to the development of speech and language) is that they cannot acquire speech and language normally.

Classification Scheme Hearing is represented by a continuum ranging from very acute perception, such as that of a gifted musical conduc­ tor who can detect an out-of-tune instrument in an orches­ tra, to total deafness, such as that of an individual who can detect only strong vibrations through tactile sensations. The following classification scheme is useful in evaluating hard-of-hearing individuals. It is based on the extent to which the individual needs a higher than average level of intensity of sound in order to hear.

1 . Minimal hearing loss (below 15 dB loss). Children with a loss of less than 15 decibels (dB) are the least hard-

99

of-hearing . They can be given the same tests administered to normally hearing childre n . 2 . Mild hearing loss (15-30 dB loss). Children with a mild hearing loss may not be recognized as having a problem unless communication problems develop, i n which case they may b e referred for an audiological eval­ uation . Standard test procedures usually are satisfactory unless communication problems are evident. 3 . Moderate hearing loss (30-50 dB loss). Children with a moderate hearing loss may hear speech in a one-to­ one situation but not in a classroom. Some children in this group may require special testing procedures because of their communication problems. The procedures selected should be based on the individual child's degree of communication. 4. Severe hearing loss (50-80 dB loss). Children with a severe hearing loss hear only the loudest speech sounds. Their articulation, vocabulary, and voice quality will differ from that of normally hearing children. Because many will have severe communication problems, special testing pro­ cedures will be required. 5 . Profound hearing loss (greater than 80 dB). Children with a profound hearing loss do not hear speech . They need special testing procedures devised for deaf children or for those with profound hearing losses. It is important to know i f frequency ranges are affected differentially by the hearing loss, in which case particular attention should be given to the ranges that encompass speech sQund§.

Assessment Considerations Before you begin the formal assessment of a hearing­ impaired child, consider the type of loss, degree of loss , age of onset, and etiologic components . Hearing-impaired children are not a homogeneous group. If hearing loss was a result of a neurological disorder such as meningitis, there may be other concomitant dysfunctions that you will need to consider in administering the tests. If the child wears a hearing aid , determine when it was last checked, whether the child uses it consistently, and , at the time of the evalua­ tion, whether it is turned on . Try to learn about the child's functioning level - that is, the methods by which the child receives information and how he or she communicates (receptive and expressive skills) . Interviewing the parents and teachers and observing in the classroom can help you obtain information about the child's preferred modes of communication. Use these modes, if possible, to present

1 00

CHAPTER 5 TESTING CHILDREN

the test items , and encourage children to respond in their preferred modes.

Does the child respond once to a sound and not again? Does the child understand speech? Does the child confuse similar sounding words? If the child does not speak, how is communication carried out (for example, pointing, gesturing, shifting eye gaze)? •

• •

Signs of hearing difficulty.

Some signs of possible hear­

ing difficulty are as follows: GENERAL SIGNS • • • • •

Lack of normal response to sound Inattentiveness Difficulty in following oral directions Failure to respond when spoken to Frequent requests to have speaker repeat what was

said Intent observation of speaker's l ips ( l ipreading, speech reading) Habit of turning one ear toward the speaker Cupping hand behind ear Unusual voice quality (for example, monotonous) Speech too loud or too soft Faulty pronunciation Poor articulation Frequent earaches or discharges from ears •

•

•

•

•

•

• •

EXAMPLES OF POSSmLE HEARING D IFFICULTIES

Has difficulty discriminating consonant sounds; for example, hears mat for bat, tab for tap Has difficulty discriminating and learning short vowel sounds Has difficulty sounding out a word, sound by sound; for example, has difficulty saying k-a-t for cat Has difficulty relating printed letters such as "f," "pi ," and "ide" to their sounds Has difficulty separating sounds that make up blends; for example, has difficulty determining that "fl" has the sounds f-f . 1-1 Spells and reads sight words better than phonetic words •

•

•

•

•

.

.

•

Watch for these signs during all examinations . These signs, coupled with answers to the fol lowing questions, w i l l g u i d e you in u n c ov e r i n g p o s s i b l e h e a r i n g impairments : Are the volume and pitch of the child's voice appropri­ ate to the situation? • Is the child's pronunciation intelligible, consistent, and age appropriate? Is the child's speech fluent or are there unusual pauses? • Does the child grope for words? Are the child's replies timely or are there unusual delays? •

•

•

In addition , note whether the child becomes frustrated when you do not understand him or her and, if so, how the child then behaves (for example, w ithdraws or acts out) . If you detect any behavior that suggests a possible hearing deficit, refer the child to an audiologist.

Special communication skills needed.

Examiners who work with deaf childre n need special communication skills. If you have not received special training, prepare for the testing of deaf children by observing classes for the deaf, noting how teachers communicate with their pupils. It is important to use a total communication approach with those deaf children familiar with it. This approach entails the simultaneous use of speech, sign language (a manual system of communication such as American Sign Lan­ guage or Signed English), and fi ngerspelling. Generally, in administering tests to deaf children a variety of tech­ niques can be used, including speec h , gesture, panto­ mime, writing, signs, fi ngerspelling, and drawing. In one study with severely and profoundly hearing­ impaired children, use of the total communication ap­ proach resulted in the highest WISC-R Performance Scale IQs (Sullivan, 1982 ) . The total communication approach was found to be superior to pantomime and visual-aid modifications and to verbal statements . These results indi­ cate that if you plan to assess hearing-impaired children, you should be skilled in a manual system of communica­ tion . Pantomime and visual aids are poorer administration procedures; they should be used only when you cannot communicate in sign language , or when the hearing­ impaired child is not versed in sign language or other special communication modalities. Pantomime and visual aids are , however, preferable to a verbal-only administra­ tion. If you cannot administer the test in the child's pre­ ferred communication modality, request a psychologist who has the requisite skills to perform the evaluation or hire an interpreter.

Effects of visual cues. Testing children with impaired hearing requires a high degree of skill and wide range of experiences with deaf childre n . Because sight is the chief means by which deaf children receive stimuli, they are likely to seek visual clues, such as facial expressions or movements of hands, to gain understanding about their

TESTING HEARING-IMPAIRED CHILDREN performance . You must realize that any movements you make may fu rnish cues to deaf children . Facial ex­ pressions, rather than tone of voice, will convey your mood , and a frown or a grimace of impatience will be quickly noted and interpreted unfavorably. Smile to reward the children's efforts, but not to reward a correct respon se. Smiling when a wrong response has been given should be avoided, so that children are not encouraged to make similar responses.

Comprehension problems . Although hearing-im­ paired children may give the impression of being able to understand directions and test questions, closer inspection may reveal that they are feigning comprehension to obtain your approval . They may have learned how to play a role in order to avoid confronting potentially embarrassing situa­ tions. In tum , you may have difficulty understanding the answers of hearing-impaired children, particularly those with expressive difficulties . Responses given in panto­ mime by the child should be given credit only when you have no doubt about the accuracy of the answer. Techniques for giving instructions.

You must be able to make instructions understood without giving away an­ swers in the process. Pantomime can be used, but it does not always convey to the child what you intend to commu­ nicate . A simple demonstration will sometimes suffice, but the demonstration itself may indicate the answer to the problem. If you use speech, be sure that you are looking at the child while you are reading the test instructions and ques­ tions and that the child is watching your face . Speech should be clear, distinct, and natural . Maintain a pleasant face. Be sure that there are no obstructions blocking the child's view of your lips, and that the lighting in the room is appropriate. Children who wear hearing aids should do so during the examination; be sure the hearing aids are in good working order and are turned "on . " In some cases it may b e necessary t o use a n interpreter skilled both in sign language and in English . Anyone who tests hearing-impaired children routinely should learn sign language .

Use of standardized tests.

The assessment of hearing­ impaired children poses special problems. Arriving at a valid measure of ability is difficult because many tests have large verbal components . Verbally based tests usually do not give an accurate picture of the hearing-impaired child's level of mental ability. They are more likely to measure the extent of the hearing-impaired child's language deficiency.

101

Verbally oriented tests should be used with extreme cau­ tion. if at all . The performance tests selected for hard-of-hearing chil­ dren should not depend on verbal directions . Timed tests may be less valid for hard-of-hearing children because the added stress of being timed may interfere with their perfor­ mance to a greater extent than it does with children with normal hearing . Representative performance tests include the WISC-R, WPPSI, and WAIS-R Performance Scales; the AbstractlVisual Reasoning subtests of the Stanford­ Binet: Fourth Edition; the nonverbal subtests of the K­ ABC ; the Ontario School Ability Examination; the Leiter International Performance Scale; the H iskey-Nebraska Test of Learning Aptitude; the nonauditory/nonverbal sub­ tests of the lllinois Test of Psycholinguistic Ability ; and Raven's Progressive Matrices . These tests differ as to their reliability, validity, and datedness ; consequently, your se­ lection of a test should be carefully considered. Standardized tests designed for hearing children are likely to pose difficulties for hearing-impaired children because such tests do not take into consideration the ex­ traordinary communication requirements of hearing­ impaired children . Obviously any test or subtest that is even partially dependent on oral instructions will place the hearing-impaired child at a disadvantage . Therefore, ad­ ministration and scoring procedures must take into ac­ count the child's deficits .

Modification of standard procedures.

Modifications in test procedures for deaf children include omitting verbal tests, adding printed or signed words, and using panto­ mime, demonstration, and manual communication. Modi­ fications that maximize assessment information include testing of limits, reinforcing responses , practicing test­ type items, eliminating time limits , and demonstrating task strategies. Appendix 0 presents special instructions for administering the WISC-R Performance Scale to hear­ ing-impaired children. In examining hard-of-hearing chil­ dren, select only the most appropriate tests and use at least one performance measure of cognitive ability.

Observation guidelines. Although the audiologist and speech-language pathologist are the specialists in deter­ mining the communication skills of hard-of-hearing chil­ dren, the psychoiogist also makes observations, both for­ mal and informal , of the child's communication skills . Notes should indicate where the observations occurred (for example, in a quiet one-to-one situation, in a class­ room, or in a job or social setting), because the child's communication skills may differ in various settings. As-

1 02

sess the child's skills in reading and writing, speech (intel­ ligibility and pleasantness), and lipreading. Note the ex­ tent to which the child is able to understand conversation during the examination. Do not assume that the commu­ nication difficulties of hard-of-hearing children indicate that they have limited intelligence . A congenital hearing loss can interfere with the development of many different skills that may or may not be related to cognitive ability.

A primary assessment goal.

A primary goal in evaluat­ ing hard-of-hearing children is to determine the extent to which their performance resembles that of a normally hearing child or that of a deaf child . Vernon (1974 , p. 209, with changes in notation) explains why this effort is needed: By interrelating IQ , educational , and other facts about a deaf or hard-of-hearing child it is possible to derive a picture which reveals the role played by his or her hearing loss. If the youth's profile is similar to that of the normally hearing, his or her loss and the way it has been coped with is not particularly disabling. By contrast, if the profile is similar to that of a deaf child, then the loss has had major effects on communication, language develop­ ment, and education. Appropriate planning for the two kinds of children varies drastically. What woul d be constructive for one would in certain cases be devastating for the other. An evaluation that does not fully address itself to this issue has failed to serve one of its major functions. The issue cannot be handled without comprehensive information. Shortcuts will not suffice and hasty, inadequately done evaluations are actually unethical and wasteful of human resources.

Vernon and Brown (1964) described the tragic case of a hearing-i mpaired child who was adm i n i stered the Stanford-Binet Intelligence Scale : Form I..r M for her initial intellectual assessment. Because no one realized that the child was deaf, she received an IQ of 29, which led to her commitment to a hospital for the mentally retarded, where she remained for five years. It was only after she was administered a performance intelligence test and obtained an IQ of 113 that the staff realized that she was not retarded . Upon discharge from the institution, she entered a school for the deaf and made good progress. The case demon­ strates what can occur when there has been a failure in the diagnostic process.

TESTING C ER EBRA L PALSI ED C H I LD R E N

The testing of cerebral palsied children presents particular difficulties . Their motor, speech, visual , and auditory diffi­ culties may limit the applicability of standardized tests and

CHAPTER 5 TESTIN G CHILDREN make caution mandatory in interpreting test results . Be­ cause cerebral palsied children frequently perform motor tasks in a slow and laborious manner, they may be at a particular disadvantage when time l imits are imposed . Furthermore , it is sometimes difficult to determine whether their test failures are due to physical disabilities or to cognitive deficits. In spite of the difficulties inherent in using standard tests, however, it is important to compare the cerebral palsied child's performance with that of the normal child, because the latter sets the standards in the world at large. Many cerebral palsied children talk adequately and have good use of at least one hand; modifications of test admin­ istration procedures may not be necessary for these chil­ dre n . However, when serious physical limitations exist, modifications are needed. Tests requiring the least modifi­ cation should be selected . Use performance tests if the child has good hand usage and poor speech , or use verbal tests if the opposite is true . Tests requiring motor skills are likely to be inappropriate for children with motor hand­ icaps. When subtests are eliminated, scores may be pro­ rated, if prorating is permitted by the test manual . The following case illustrates the value of employing test modification s . Tommy, a n 8-year-old severely handicapped boy with athetoid­ type cerebral palsy, had no speech and communicated only by painstakingly pointing out numbers that corresponded to words or phrases taped to the tray of his wheelchair. Yet Tommy ap­ peared to the examiner to be alert, curious, and eager to partici­ pate in the assessment. The examiner attempted to admin i ster the Pictorial Test of Intelligence, a test that requires the child to point to one of four pictures that corresponds to the correct answer to a verbal question. However, Tommy was unable to make the move­ ments necessary to differentiate between the pictures, even though they were printed on a large card. As a modification of the standard procedure, the examiner read each test item and then pointed in turn to each of the four pictures. When the examiner pointed to the choice Tommy wished to make, the child wiggled in his wheelchair. Because of the problem of interpreting Tommy's movements, many of the items had to be administered several times. A fter an hour and a half of this exhaustive pro­ cedure , both examiner and child finished the test with a sense of real accomplishment. Tommy, the boy who at first glance ap­ peared to be mentaJly retarded, had earned an IQ of 120.

Research studies and follow-up reports indicate that the initial test results obtained by cerebral palsied children have a satisfactory degree of reliability and validity (Crowell & Crowell , 1954; Klapper & Birch , 1966 ; Kogan, 195 7 ; Portenier, 1942 ; Taylor, 1961) . However, it is still recC1mmended that the first scores not be used as the sole criterion in long-range planning, not only for cerebral

1 03

SUGGESTIONS FOR ADMINISTERING TESTS palsied children but for all children examined by an intel­ ligence test.

SUGG ESTIONS F O R A D M I N I ST E R I N G T ESTS

You are l ikely to be anxious when you are fi rst learning to administer tests . You must become a juggler in order to successfully carry out a multitude of tasks - establish rap­ port, administer the appropriate items, keep the materials ready, respond appropriately to the child, precisely record the child's responses, observe the child's behavior, and score the child's responses. In time many of these pro­ cedures will become routine , but even the most experi­ enced examiners should review their test procedures peri­ odically. Test manuals usually present valuable suggestions for administering tests; these suggestions should be stud­ ied carefully. The following tips are designed to supple­ ment or reemphasize a number of points that are often found in test manual s . These suggestions, coupled with the instructions in the test manuals , will help you to administer tests better. Some tests have specific guidelines that may differ from those presented below; in such cases, the guidelines in the respective manuals should always be followed. Table 5-3 shows an excellent general outline of testing procedures.

General Considerations

1 . Know your task well enough that the test flows almost automatically, leaving you maximally free to ob­ serve all aspects of the child's behavior. If you stop to read a paragraph in your manual , the child is not likely to wait for you . He or she may be out of the chair, playing games, or wanting to leave . Know your materials so that you can reach for them without looking away from the child; work on perfecting ways to record responses that do not inter­ rupt the relationship. 2 . Adhere to test instructions exactly. 3 . Be purposeful in what you do and keep control of the situation. Let there be no doubt as to who is in charge. If control becomes an issue, stop the testing until the matter has been settled. You may need to play a game, take a break, enlist the parent's help, retreat to easier items, or use some other tactic , but do not let things get out of hand. 4. Take your time and suit your tempo and decibel level to the child . Some appreciate a boisterous approach , whereas others are quiet and fragile . Use words and a tone of voice that w i l l help the child feel confident and reassured.

5 . Give the child a choice only when you intend to leave the situation up to him or her. 6. Do not do or say anything that may diminish the child's self-respect. 7. Avoid motivating the child by making comparisons between the child and another child or by encouraging competition . If the parents are present, discourage them from using such techniques. 8 . Try to maintain the child's cooperation at all times . Remember that you need to earn the child's attention and cooperation . Keep the situation friendly, interesting, and rewarding . Do not let the relationship degenerate into a power struggle. Sometimes you cannot avoid a modest degree of unpleasantness, especially with unhappy, re­ bellious children, but it is your responsibility to establish the best rapport you can and to elicit the child's best effort . 9 . Establish and maintain rapport with the parents, and enlist their support of the assessment effort . 1 0 . Use your own reaction to the child as a cue . If you find yourself continually trying to improve rapport with an older preschooler, this may suggest excessive manipula­ tion by the child. If the child's behavior makes you feel irritable or angry, then others may feel the same way. (Recognize, however, that a defiant child is an unhappy one who needs help . ) Also, watch the complexity of your own language (and that of the mother or father) as a guide to what the child is understanding; you should be adjusting automatically to the cues the child gives you . 1 1 . Do not change the directions because you think the child can do a particular task with the altered directions . You must follow standard procedures. Under certain cir­ cumstances, if the directions are not part of the task - for example, if you want to see whether a child can draw a circle, and not whether he or she will draw a circle when given certain directions - then you may be able to change words . These circumstances are unusual , however, and are most often found in informal assessments rather than for­ mal assessments . 1 2 . Be attentive to any signs of visual or hearing diffi­ culties, as noted in earlier sections of this chapter.

Physical Arrangements I . Use a testing room that has minimal distractions . You and the test materials should be the most salient stimuli . Ideally the examination room should both minimize exter­ nal sources of distraction and maximize the child's moti­ vation . The many different conditions you will meet on the job will invariably fall short of the ideal . Although you may have to settle for conditions that approximate the

1 04

CHAPTER 5

TESTING CHILDREN

Table 5·3 A General Outline of Testing Procedures

I. General testing precautions. A. Read, learn, and reread instructions. B. Always adhere to standardized procedures . I . Use exact wording.

2 . Maintain accurate timing.

3. Present materials in the prescribed manner. 4 . Follow scoring instructions rigidly. 5 . Do not depend solely on reading the printed direc­ tions, but do have them available for ready reference. C. Be objective. 1 . Give no indication of the correctness or incorrectness of the child's responses. 2. Give no clues about the answer you expect; watch your verbal intonation; remember you are testing, not teaching. D. Be natural . 1 . Be warm but impersonal . 2 . Learn to use standardized wording in a natural and informal manner. 3 . Achieve rapport and verbal give-and-take before the test begins; take a listening attitude. E. Prepare the environment. I . Avoid distractions. a. Visual : Have the child face away from doors and windows where movement and activity are going on. Have him or her face away from large open spaces that have, for example, distracting pictures, colors, or toys. Avoid clutter. b. Auditory : Avoid noisy areas. c. Emotional : Avoid testing when the child is hurried , troubled, or ill . 2 . Provide optimum conditions for good performance. a. See that the child is in a comfortable position and has a clear view of the materials . b . Provide a well-lit room with adequate ventilation and comfortable temperature. c. Avoid glaring lights, reflections from the pages, and other distractions . Face the child away from the window if possible. d. Speak in a clear, audible voice at a moderate rate of speed . e. Maintain interest through enthusiasm, attention to the child, and smooth presentation of the material . f. Commend and encourage for general performance

II.

but never on specific items. g. Let the child know you want to see how well he or she can do.

Administering and scoring the test.

I . Provide an efficient arrangement and method of ma­ nipulating materials for a. Recording. b . Viewing the manual without its becoming a barrier

berween you and the child. c. Putting away and bringing out materials . d. Avoiding delays and distractions for the child. 2 . Make a smooth transition from test to test and from item to item. You must know at each point in the test what your next presentation will be. 3 . Know your materials and scoring well enough that you do not extend the test unnecessarily. Overtesting may create fatigue and disinterest. a. Know the scoring standards. b. Begin at the appropriate point. 4. Learn to handle extraneous behavior. a. Disregard or redirect irrelevant remarks. b. Minimize extraneous movements by developing interest, motivation, and task orientation. If extra­ neous movements do not interfere with the child's functioning, ignore them. If necessary, provide the child with a positive outlet such as grasping the edges of the desk or folding his or her hands. c. Foresee fatigue and distraction. B . Scoring. I . It is essential that you know the scoring standards well . This requirement applies particularly to the un­ derstanding of the intent of each test . 2 . Remember that the scoring standards are just what the term says: they are "standards" for scoring rather than all-inclusive right-or-wrong answers . It is often nec­ essary to evaluate equivalent responses in the light of other responses listed in the scoring standards , since not all possible responses are included.

3. Check all answers with the manual in order to verify any doubtful responses.

4 . Recheck every step in the scoring process. 5 . Double-check all figures and calculations - the chron­ ological age, the number of correct items, additions, and arithmetic calculations. C . Care of materials. 1 . Whenever any of the materials presented to the child become marked or defaced in any way that might influence the child's response, replace them. 2. If you must point to the pictures , be sure not to mark the page . Use the back end of your pen or pencil .

3 . If any materials are lost or damaged, replace them with objects identical to the original.

A. Efficiency.

Source: Reprinted, with changes in notation, by pennission of the publisher and author from W D. Kirk, Aids and Precautions in Administering the Illinois Test of Psycholinguistic Abilities, pp. 5-7, 1 3- 14 . Copyright 1 974, University of Illinois Press.

1 05

SIGGESTIONS FOR ADMINISTERING TESTS ictal , you must never test when the conditions woul d aeversely affect the child's performance . 2. Control your materials . Do not allow the examinee to tun pages in the test manual , to play with the test mate­ rils, or to hold pencils, pens , or toys, except when these ac needed for a test. Unless specifically prohibited, the chid may be permitted to turn pages in test booklets that cotain items (for example, the pictures in the Wechsler p' ture Completion subtest) . This permits the child to pnceed at his or her own pace and encourages rapport; hCNever, it does not have to be a routine procedure . Nith infants up to 6 months, your materials can be in sijht, but after that, the only materials visible or accessibie shJUld be those needed for the task at hand. Place your test ki'on a low chair next to your side such that when the kit is opn the back ofthe l id faces the child. Stay between the kit ani the child , keeping the kit out of the child's reach so that th( child does not see any answers in the test manual . Ceck to see that the testing kit contains all necessary mzerials before you begin testing. . It is often a good idea to have one attractive object alcady on the table when you enter the room with the ch:d . This object should not be one you will use for tesing. '. Make sure the child is comfortable. For example , the char and table should be supportive and at appropriate heshts, the l ighting should be sufficient , and the ventila­ tiol adequate . If the child's feet do not touch the floor, ar nge a solid box or block for them . A high chair is optmal for young children who are able to sit upright on ther own (usually by about 6 months of age)- they are acustomed to it, it is comfortable, and it provides foot suport. 5 Children up to 2 or 3 years of age will usually do betcr with a parent present . If parents are present , seat then behind the child, out of the child's line of vision. If a pannt is to hold a small child on his or her lap, make sure that he chair is low enough so that together their knees can fit Llderneath the table. 6 Unless the examinee is an infant , no other children shoud be present. Having another child in the room can be dis ptive, and the feel ings of the child who is not getting you attention can be hurt . Avoid these problems by re­ quering, at the time the appointment is made, that the pamt make arrangements for siblings . 7 If the examinee is older than 3 years of age, the parflt generally should not be in the room , unless the child insi�s on having a parent there or a parent is needed to faciltate the testing . An observation room with a one-way mirnr allows parents to view the testing and facilitates theil Understanding of the results you obtain.

Making the Child's Acquaintance

1 . Keeping in mind that the parent prepares the child for testing, allay parental anxieties as best you can when you arrange the appointment. Tel l the parent what you will be doing with the child -- such as looking at how the child works with materials , which activities are easy for the child and which are more difficult, and how the child expresses ideas and interests. Avoid words like test, exam­ ination , and evaluation . Assure the parent that most chil­ dren find the activities pleasant and that you are looking forward to meeting the chil d . 2 . Meet the child in the waiting room, and - with a young child especially - spend some time there. When you greet the child, do not overwhelm him or her. If the child is young, you might bring a toy and at fi rst try to keep the parents seated so that you can kneel at the child's level . Initially give most of your attention to the parent and let the child look you over, gradually making friends as the child tolerates your attention . When the child seems comfort­ able, invite the child and, depending on the child's age , the parent into your room. To a young child you might say something like "1 left a book on my table for you to look at ; let's go see it" or "Time to do some things now; I'll show you where they are . " Do not ask "Do you want to come into my room now?" unless you are prepared to accept no as an answer. 3 . Do not remove the child abruptly from an interesting activity. 4 . Fol low the test guidelines for introducing the test. You can set an older child at ease with a few introductory remarks. The following sample may be a useful model for introducing a test : We will be doing some things together, and most children enjoy doing them . Some of the things that we will be doing will be easy and some will be hard. That's because some of the things are for children younger than you and some of the things are for children older than you . So don't worry i f you can't get all of the answers right I don't expect you to know all of the answers. But I do want you to do the very best you can. -

Watch how the child handles the items. Praise with your voice and tone (for example, "You know how to do these things !"). Do not let the child play with the test materials to get comfortable. Recognize that it may take a l ittle while for a young child to understand that he or she should do what you ask. 5. You will probably have to teach a young child that, once he or she has worked on one activity, something else interesting will ensue . Children of l or 2 years of age need time to understand that you expect them to do something specific with the materials. After a few items, even quite

CHAPTER 5

1 06

young infants get this message. It may take a while for some children (particularly those from large families who have learned to defend their toys) to give up one object in exchange for another. Generally you should present the new object as you remove the old one, but you may need to let the child become interested in the new one before requiring that the old one be relinquished . With children who have not yet achieved object permanence (see Chapter 3 ) , covering the old toy with your hand while presenting the new one may facilitate the exchange. A comfortable rhythm will soon develop if you are running the session smoothly. 6 . If the parents are present, let them know that you do not expect the child to pass all of the test items; acknowl­ edge from the beginning that some items are simply too difficult. Never tell a parent that a child 1ailed" an item .

Reinforcement 1 . Positive reinforcement is your most potent tool ; use it. Attend to behavior you like . Nodding, smiling, and appreciative noises may be more effective than words . Be sure to reinforce effort, which may or may not be accom­ panied by success. 2 . Try to ignore or circumvent the behaviors you do not like ; in some cases you may have to distract the child . Parents may be able to help you, but some parents inter­ fere . Generally, handle problems by trying to foresee and forestall them . 3 . Avoid using tangible reinforcers as motivators, ex­ cept with autistic or profoundly retarded children who are accustomed to receiving them. In such cases, be sure you have parental permission to use primary reinforcers, such as Cheerios, candy, or raisins . One of the best motivators, in addition to positive reinforcement, is to arrange the ac­ tivities so that the child experiences successes inter­ mittently. 4. Remember to keep giving positive messages . Build an image as a kindly, fairly exciting but predictably firm adult who likes the child and is (will be) pleased with his or her effort. S . Clearly define and consistently maintain l imits on allowable behavior. Be sure that the child understands the l imits of behavior. Although consistency is necessary, do not be inflexible. Accept the child's need to test l i mits , and try to adapt your limits to the child's needs, giving him or her time to accept the m . This approach will demonstrate your respect for his or her feelings. 6 . Reinforce parents . Most parents respond positively to kind words about their caretaking and teaching, as well as their patience and effort. Show them that you recognize

TESTI NG CHILDREN

their affection , their sense of humor, and the problems with which they are coping . These recommendations are es­ pecially important for interactions with anxious or defen­ sive parents .

Management

1 . Do not confuse the child . The younger the child, the more important it is that your gestures and use of materials be clear and precise . With babies, it helps to exaggerate your movements a little so that they see what you want (for example , pat the Bayley dolly rather fi rmly and then mo­ tion for the baby to do the same) . Through the tone of your voice and your actions (such as handing him or her mate­ rials) make clear to the child when you expect a response . 2 . Do not ignore any remarks made by the child . Attend and be sensitive to the child's needs , such as for a drink or physical activ ity. Watch for early signs of boredom , fatigue, physical discomfort , or emotional dis­ tress, and take appropriate action before such conditions become acute . The child may prefer that the parent, rather than you, wipe his or her nose or take him or her to the bathroom . But do not take too many breaks, because this will lengthen the overall time and fatigue the child. 3. As an alternative to a formal break in testing, you may administer gross motor items as a break from verbal items . 4. Give the child a chance to get acquainted with the materials if necessary (for example, let an 18-month-old see the "bunny" you are going to hide under the cup On the Bayley) , but do not let the child play with the materials . Parents sometimes object t o your adherence t o structure when the child is attracted to a toy. Explain that if you do not adhere to structure the child will become confused about what the test expectations are and the session will run so long that the child will grow tired . 5 . Be sure that the child is ready and expectant when you present a new task . Tapping on the table may help draw a young child's attention to a toy if the child is inattentive. You may need to institute a signal such as "Get ready ! " Avoid urging the child to respond before h e or she is ready. 6. Redirect activities in a way that is consistent with the child's motives and interests. 7 . Recognizing that many of the tasks that you ask the child to perform may be too difficult, make them as pleas­ ant as you can. For example, you might say, with a smile, "Here's another real hard one !" or "Here's one for older kids but let's try it!" Learning to encourage effort takes practice. 8. Always be truthful. Do not say that you are almost through if you are not. 9. Finish with some easy items, even if you do not

SUGGESTIONS FOR ADMIN ISTERING TESTS

need to - this leaves both the child and the parent w ith a good feeling . 1 0 . Particularly with infants and young children, you may have sessions in which testing cannot be completed . The reasons for this outcome are numerous. You may never make friends . The child may remain unhappy or suspicious for the entire session. If things are going badly, reschedule the appointment and reassure the parent that such things occasionally happen . 1 1 . In general , you should accept children as they are , knowing that there are reasons for the ways they feel and act . Help them find acceptable outlets for their feeling , and try to meet their needs. Skill s you acquire in handling children can aid you in helping them gain confidence in their abilities . Increased confidence may enhance their cooperativeness and willingness to respond to the tests.

Test Procedures Arranging materials.

A general rule of thumb is to administer items from the child's left to right in the order given in the test manual . Even demonstration items should be so arranged. This rule simply brings some uniformity to the test administration . Stimulus materials should be placed in front of the child one at a time so that the child can see the blocks, cards, or other material clearly. Some beginning examiners have been known to inadvertently cover materials as they placed them on the table. The child should be able to see your complete movements when you rearrange materials in demonstration items, unless the test manual says otherwise . Be sure to record the child's ar­ rangements as soon as the item is completed. In constructing sample block designs it is a good idea to complete one row at a time , starting each row at the child's left and working to the right. This will tend to ensure that you administer the sample items in the same way to each child. Be sure that the child can easily reach the blocks or other stimulus materials . Do not remove the materials until the child says that he or she has finished the task. When administering items that appear in booklets, you need not hold the booklet, unless the manual so states. The booklet can be left flat on the table and pages (or cards) turned over for each succeeding item. Keep the testing table clear of all extraneous materials to avoid needless distractions . Return all materials to the testing kit immediately after they are used .

Giving instructions and asking questions. The in­ structions for the various tests should be read verbatim, in an even and relaxed manner; avoid a robotic or machine-

1 07

like approach . Most beginning examiners have a tendency to read the test instructions and questions too quickly or too slowly. Pace your reading and conversation to the child's age and ability leve l , striving for clarity and a natural quality. Avoid unusual facial expressions or modu­ lations in voice that may give cues regarding performance. The instructions used for the first trial of a test can be repeated on successive trials or items . It is not permissible to explain any of the words that are used in the directions or in the questions on the various test items unless the manual explicitly permits explanation . Be sure not to add any additional words to the instructions, such as "Now here is . . . . " Memory items should not be repeated. You should know the directions well ; ideally, they should be memo­ rized. You should become so familiar with the test mate­ rials that you can introduce and remove them without breaking the interaction between you and the child . Unless specifically prohibited by the test manual, phras­ ing requests as mild commands (such as "Tell me another reason . . . ") is preferable to asking the child if he or she knows the answer. "Can you tell me . . . " and "Do you know . . . " often result in a simple no from a child, whereas a mild command may encourage the child to try to answer the question .

Timing items and encouraging replies. On tests that have time limits, do not tell children how much time is allotted to the task unless the manual says otherwise . If they ask, you might tell them to give their answer as soon as they know it. Timing should begin as soon as the item is shown to the child. Seldom accept the first "I don't know" response given by the child . Ask the child to try to answer, unless the question appears to be too difficult for him or her. It is permissible to say something like "Try to give an answer" or "Try to answer it in some way, if you can . " Clarifying responses. If you are uncertain about what a child has said, ask him or her to repeat the response . To minimize the need for repetition, pay close attention to everything the child says. The purpose of questioning is to clarify ambiguous responses . Obviously incorrect re­ sponses should not be probed. When a response contains one part that is correct and one that is incorrect, ask the child for his or her preference and score the response accordingly. Avoid giving the child the impression that a test is being discontinued because of repeated failure . If a child asks to have part of a question repeated, repeat the entire question . Providing feedback. You should not tell the examinee whether responses are correct. If asked, you can say that it

CHAPTER 5 TESTING CHILDREN

1 08

is against the rules to tell the answers, but that you can answer their questions when you have finished .

Recording responses.

Immediately record the child's answers in the appropriate spaces provided in the test record booklet . Be as accurate as possible . Exercise judg­ ment in recording parts of the response that have limited relevance to the test question . In the margin of the record booklet, note behaviors of interest, other observations, hypotheses, and other relevant information .

Scoring responses. The child's responses should be scored as soon as they are given. Try to shield the scores in the record booklet as unobtrusively as possible . The rec­ ord booklet should be positioned in such a way that the child does not see answers to questions or get feedback on prior or current responses. You might want to place the record booklet on a clip board and hold it at an angle to the child. Any doubts about scoring should be resolved by refer­ ring to the scoring section of the test manual . On the WISC-R, WPPSI, WAIS-R, Stanford-Binet: Fourth Edi­ tion, and other tests that employ entry and discontinuance points, it is absolutely necessary to score the responses immediately at these points so that the correct administra­ tion procedures can be followed. As you are administering a test, place a question mark to the left of any responses that were difficult to score. Give special attention to these items on recheck. All scores should be rechecked carefully after the examination. The scoring examples shown in test manuals are guides . They usually are not meant to represent all possible correct or incorrect answers, except in the case of those items that have only one correct answer or for which the manual l ists the only acceptable responses. You should study the scor­ ing criteria for each test so that you know what are correct, questionable, and incorrect answers. Try to discern the kind of response called for by the item. The scoring sec­ tions occasionally provide guidelines. When guidelines are not provided, study the sample responses to ascertain the kind of response that should receive credit . Such study will allow you to make appropriate decisions regarding the correctness or incorrectness of a response . Acknowledging performance. When the examination has been completed, children should be praised for their efforts . It is not necessary to give them trinkets. Unless arrangements have been made with the parents , do not give candy at the completion of the examination.

Informal Observation It is often helpful to combine informal observation with standardized testing. It is preferable to do the formal testing first because it is easier to move from a structured situation to a nonstructured one than vice versa . Do not overlap them, however, because you may lose the essential structure of the standardized test. The parent (or a sibling) may be especially helpful here . When you have finished testing, you might, for example, bring out some toys and ask the parent to play with the child . Leave the situation unstructured so that you can hear how the child expresses himself or herself when more relaxed . If the parent begins to ask too many questions of the child (in an effort to elicit verbalizations) , suggest that the parent just comment on what the child is doing . By noticing nuances in the parent­ child interaction you can get an idea of the child's behavior. Watch the child's reactions to parental expectations and control tactics; note contrasts between the child's behavior in this non structured situation and in the structured situa­ tion required for testing; observe the child's ability to use toys in a constructive/imaginative/appropriate way; and try to get an idea of the chjld's attention span . (See Chapter 17 for further discussion of systematic observational pro­ cedures and Chapter 16 for play interview procedures . )

Talking t o Parents Handling a complex situation involving the parents and the child is not easy. Most of your attention should be focused on the child. You may need to schedule a separate appoint­ ment to talk with the parents about their concerns, any information you need, or issues you want to discuss . A young child may be put on the floor to play while you talk after the session, but an older child will be interested in your conversation . Even with a very young child, parents may feel restrained when talking in the child's presence, especially if they are discussing emotionally laden mate­ rial and feel sad or angry, feelings an infant or toddler may pick up. Such cases need to be handled individually. Remember that you are the parents' ally in providing the best possible care for the child. Resist the temptation to assign blame , remembering that any behavior you see is a product of a long series of complex interactions. Remain friendly, helpful , warm , objective, and credible. Whatever you discover in the evaluation session will be useless unless your views are (eventually) respected and accepted by the adults who are responsible for the child's welfare . Finally, be aware of the feelings of the parents . Parents

SUGGESTIONS FOR ADMINISTERING TESTS

have limited experience with testing; they may be unin­ formed about the meaning of test results, anxious abo t their child's behavior, and defensive about their child's need for testing. By reassuring parents in an honest manner and addressing their concerns whenever possible, you can help to alleviate some of their anxieties .

Pitfalls in Administering Tests The following examples illustrate test administration pro­ cedures that will interfere with establ ishing rapport (Teglasi & Freeman , 1983).

Example 1.

An examiner interacted the "obligatory" few minutes with an ll-year-old girl . Just then, the girl felt comfortable enough to bring up a concern that was directly germane to the referral problem. The examiner's response was , "Well , we've spent enough time talking; shall we get started with the test?" Comment: This example violates the principle that you always need to be attentive to the childs needs and concerns. Delay beginning the test, if necessary, when there are more pressing matters.

Example 2. The examiner asked, "Do you want to get started?" When the child responded "no!" the examiner said , "Well, we're going to get started . " Comment: This example violates the principle that ifyou offer choices, be

prepared to honor them . Example 3. The examiner said , ''These are getting hard. They are so hard that even I don't know some of the answers. Do you sometimes wonder if your teachers know the answers to tests that they give you? So you're not really expected to know all of them . . . now even I don't know the answers to some of these. I look them up in a book." Comment: This example violates the principle that you should not introduce extraneous material that may in­

crease childrens anxiety, perplex them, or reduce their motivation . Example 4. The examiner asked, "Can you try to re­ member any others?" When the child responded "no," the examiner said enthusiastically, "Well , you did very well ." Comment: This example violates the principle that you

should give praise for childrens efforts and not after either an incorrect response or a lack of response. You can acknowledge the child's statements by saying "Okay" or "Let's try the next one . "

1 09

Example 5. The child asked, "When can I take a break?" to which the examiner responded, "Any time you want . " But when the child said "now," the examiner said, "Well , let's finish this fi rst. " Comment: This example violates the principle that ifyou say something, be prepared tofollow it

literally. Example 6. A child was reading the words on a reading test so quickly that the examiner could not keep up. The examiner frequently said , "Tell me again . . . back up . . . . " Comment: This example violates the principle that you

should pace the examination so that you can comfortably and accurately record the information. The examiner in this example should have provided structure by saying "I'd like you to wait for a little while in between words" or "Say the next word when I say okay. "

Departures from Standard Procedures Departures from standard procedures during test adminis­ tration may change the meaning of test scores . When standard procedures have been altered , scores based on the test norms may be inappropriate. However, standard test procedures are not applicable to all children, as we saw earlier in this chapter, and on some occasions a more accurate estimate of intellectual ability may be obtained by adaptation of standard procedures, particularly when chil­ dren with severe physical disabilities or severe brain damage are being tested . For example, you can read to blind children items that they would have had to read for themselves, allow children with major speech problems to use a typewriter or write their answers, and direct children who cannot point or speak to give their answers to multi­ ple-choice items through eye movements.

When do departures violate test norms?

In order for the test norms to be used with confidence, the test items must be administered according to standard procedures. This means, for example, that you must use the exact words of the questions, the specific test materials , and the specified time l imits . The effects of small deviations in test procedures, however, are not known . Each test has a stan­ dard error reflecting deviations caused by the nuances in the examiner-examinee relationship, the conditions of the examination, the time of day, and other similar factors . It is hoped that when small changes occur in rapport or in other aspects of the test situation, the resulting score will remain within the range of the standard error, but this will never be known.

1 10

Whether norms are violated when test procedures are modified depends on the extent of the modifications. Most of the procedures suggested in the previous sections do not appear to preclude the use of standardized norms. How­ ever, more serious modifications, such as alterations in the wording of questions, changes in the scoring criteria, extensions of time l imjts, or the use of a multiple-choice procedure for items that call for definitions, are likely to jeopardize the use of test norms. Although standard administrative procedures should be the rule, occasionally exceptions may be made in order to obtain some estimate of the child's ability without regard for norms. On those occasions when modifications must be used, the resulting scores should be reported as "estimates" obtained by departures from standardized testing pro­ cedures and should be interpreted cautiously in l ight of the modification s . The scores are likely to be less precise and predictions about future levels of functioning cruder than would otherwise be the case .

Incentives and their effects. The effect of incentives, such as praise, candy, or money - referred to as social or token reinforcement - on children's test performance has been studied extensively. Results of 34 different studies, with normal and handicapped children of various ethnic groups, present a mixed picture : Fourteen indicated that incentives o r feedback did not affect performance (Busch & Osborne, 1976; Clingman & Fowler, 1975; Cohen, 1970; Cook, 1973; Galdieri , Bar­ cikowski , & Witmer, 1972 ; Goh & Lund, 1977 ; Graham, 1971; Jackson, Farley, Zimet , & Gottman, 1979; Lyle & Johnson, 1973; Miller, 1974; Quay, 1971; Sweet & Ring­ ness, 1971 ; Tiber & Kennedy, 1964; Thfano, 1976) . Thi rteen indicated that the effects were mixed (Bergan, McMani s , & Melchert, 1971; Bradley-Johnson, Johnson, Shanahan, Rickert, & Tardona, 1984; Breuning & Zella, 1978; Galbraith, Ott, & Johnson, 1986; Johnson, Bradley-Johnson, McCarthy, & Jamie, 1984; Kieffer & Goh , 1981 ; Klugman, 1944; Moran , McCullers, & Fabes, 1984; Saigh & Payne, 1976; Sweet & Ringness, 1971; Terrell , Terrell, & Taylor, 1981 ; Willis & Shibata, 1978; Young , Bradley-Johnson, & Johnson, 1982). Seven indicated improvements or decreases (Ali & Costello, 1971 ; Bradley-Johnson, Graham, & Johnson, 1986 ; Edlund, 1972 ; Piersel , Brody, & Kratochwill, 1977; Saigh, 1981a, 1981b; Terrell, Taylor, & Terrell , 1978) . •

•

•

The differences between studies finding significant ef­ fects and those finding nonsignificant effects are difficult to

CHAPTER 5

TESTING C H ILDREN

discern . In some studies token reinforcement increased scores on the WISC-R Verbal Scale but not on the Perfor­ mance Scale. There is no readily available explanation to account for such differential effects . One could hypoth­ esize that incentives are more effective in situations calling for speed and quick reaction time . Yet some findings indi­ cate the opposite effect . Social and token reinforcers ap­ pear to operate in similar ways in the various ethnic and social-class groups studied . The nonsignificant findings suggest that many children view the standard test situation as inherently rewarding . They may enjoy the special attention of and acceptance by an adult examiner. The standard testing condition in all of these experiments that reported nonsignificant findings was apparently one in which the children were motivated to perform at their maximum level; they did not need additional social or token reinforcement or feedback to perform more effectively. The significant findings , in con­ trast, suggest that for many other children reinforcers do increase motivation. If knowing whether a child can per­ form better on a specific reinforcement schedule would be helpful in designing remediation programs, you mjght consider testing lirmts by evaluating the child under rein­ forcement conditions after you complete the standard ad­ ministration . This procedure, however, will i ntroduce practice effects, which will make the testing-of-limits find­ ings more difficult to evaluate. The research on incentives and their effects on intel­ l igence test scores fails in one maj or respect - there is little information about the validity of the scores obtained under the incentive conditions . It may be that the most val id scores are those obtained under conditions similar to those found in the child's environment. If teachers and parents do not use incentives in teaching chi1dren, it is doubtful that the scores obtained using such reinforcement procedures will be more valid than those obtained under standard conditions. Scores obtained using specific reinforcements may be more valid in environments that use reinforce­ ments . These hypotheses need confirmation . At present there is l ittle evidence that scores obtained using tangible or social reinforcement are more valid than those obtained under standard conditions.

Testing-or-limits. Generally, the only modifications made to standard administrative procedures should be those discussed in the test manuals (for example, changing the order of tests if necessary or eliminating spoiled tests) or those necessary to test handicapped children. However, there may be times when you want to go beyond the

SUGGESTIONS FOR ADMINISTERING TESTS

standard test procedures in order to gain additional in� r­ mation about the child's abilities . The information from testing-of-limits procedures can occasionally be helpful, especially in clinical or psychoeducational settings. Any successes obtained during testing-of-limits, of course , cannot be credited to the child's scores . Testing-oj-limits techniques should be used only after the entire test has been administered using standard pro­ cedures. Otherwise additional cues may facilitate the child's performance on the remaining items of a test. Such score increases have been reported for the Block Design and Picture Arrangement subtests of the WISe and Wechsler-Bellevue Intelligence Scale (Sattler, 1969) . The following procedures may be used in testing-of­ limits : I . Providing additional cues. To determine how much help is necessary for the child to solve a problem, you may want to provide a series of cues to the child . One approach is simply to readminister failed items, telling the examinee that there is another solution or arrangement and asking him or her to try to find it. Or you might reproduce the examinee's construction (for example, a block design pat­ tern or picture arrangement layout), tell the examinee that there is an error, and ask him or her to find it and correct it . If this procedure does not lead to a correct solution, you can show the child the first step in solving the problem, after which you can provide a series of additional steps if needed . A second approach is to begin by asking the child how he or she went about trying to solve the problem (for exam­ ple, "How did you get that answer?") . Before suggesting that the item be attempted again, you might want to provide the first step in reaching the correct solution or tell the child which part of the original method was incorrect. Another approach is to provide additional structure (for example, if the child becomes disoriented when asked to throw a ball, you might put a line on the floor to better orient the child ) . Overall, you can break down and sim­ plify the tasks until it becomes clear to you what the child can and cannot do with help and under what conditions. The above procedures will help you to determine the extent to which the examinee can benefit from additional cues . The more cues that are needed before success is achieved, the greater the possible degree of learning disor­ der or cognitive deficit . This information will be helpful in planning remedial efforts. 2. Changing modality. To determine the influence of the modality, you may want to change it . For example, if the

I I I

child fails problems in oral form , you might see if he or she can solve them in written form . 3 . Establishing methods used by the examinee. There are many different ways of solving the test questions. On Digit Span memory tests , for example, the task can be solved by grouping the digits in sets of two, three , or more digit sequences; by recalling them as a number (4-1-3 as four hundred and thirteen); or by recalling them as distinct digits in sequence. The method used may be related to learning efficiency or to personality features, or it may have no particular import. Learning how the child solved the problem may give you additional insight into how well the task was understood and the particular memory strat­ egy used by the child. To learn how the child went about solving the problem, you may simply ask what method was used. Some children will be able to verbalize their method , but others will not, even though they have answered correctly. 4. Eliminating time limits. When the child does not complete a test because of time limits , you can read­ minister the test (after the examination) without time limits and delete references to speed or time l imits in the direc­ tions . This will help you determine whether the child can solve the problem . 5 . Asking probing questions. Occasionally examinees will give responses or make constructions with blocks or other materials that are vague or idiosyncratic . If you want to, you can go back to these items , repeating the item and response and then saying "Tell me more about what you mean , " or reconstructing the design and saying "Tell me about what you assembled (or made) ; explain it to me . " Such probing questions may give you insight into how the examinee approaches tasks. As a result of the help provided during the testing-of­ limits phase , the child may pass tests . During the test proper, too, the child may solve a problem after the time limit has been reached. In such cases, you can include in your report that the child benefited from additional help or time, but do not change the test scores. One of the problems associated with helping procedures introduced during the testing-of-limits phase is that these procedures may invalidate the results of retesting occur­ ring at a later date. Therefore , you must carefully consider the benefits and costs of testing limits . If retesting with the same test may be needed in the near future, then testing-of­ limits procedures probably should not be used. However, if the goal is to evaluate the limits of the child's abilities on the test or determine problem-solving approaches and there is no reason to plan on a retest in the near future (say within

1 12

the next 12 to 24 months) , testing-of-limits may be quite useful .

EXAM I N E R CHARACTE RISTICS

Examiner Skills

A competent examiner must be flexible, v igilant, and self­ aware and must genuinely enjoy working with children. Other traits that will prove helpful are a sense of humor and the ability to work under less than favorable conditions. A preschool child who is uncooperative, pushes away toys offered to him or her, and remains silent will test the skills of even the most competent examiner. Realizing that sub­ jective factors - such as your personal style, the child's physical and mental condition, the setting, interruptions during testing, your preparation, and the child's language facility - invariably affect the testing situation and may reduce its objectivity, your task is to obtain cooperation from even the most intractable child under conditions that depart as little as possible from standardized procedures . Flexibility. When you test children, you must be pre­ pared to adjust your testing techniques. You may need to have frequent rest breaks, more than one testing session, or additional time for the child to explore and become famil­ iar with the surroundings . Considerable effort may be required to establish and maintain rapport . It is especially important to use tact, diplomacy, and ingenuity, and to have patience and understanding . These procedures will reduce fatigue and help to alleviate the child's anxiety. Order the battery so as to begin with tests that do not accentuate the child's weakness, particularly in the case of physically disabled children or language-impaired chil­ dren . Selecting appropriate tests will facilitate rapport and lead to more reliable and valid test results . If a child proves untestable, the child's parents may be able to provide information about the child's language ability, motor abil­ ity, social skills, self-sufficiency, mental development, and overall development. Vigilance. Test administration should not be routine or automatic. You must become familiar with the test material and procedures so that you are free to exercise constant vigilance, making certain that the child has the necessary physical abilities to proceed with the test, observing when

CHAPTER S

TESTING C HILDREN

the child is or is not making his or her best effort , and deciding when to offer encouragement and praise . When­ ever the child's interest in the tasks wanes, offer encourage­ ment and support. Addressing problem behaviors may be taxing . For example, how do you decide whether a particu­ lar behavior reflects true helplessness or is manipulative? And once you decide, what actions should you take? What­ ever difficulties you may encounter, keys to successful testing include appropriate timing of supportive comments and actions and maintenance of an appropriate flow of test materials. Self-awareness. You should strive to understand your own temperament and attitudes toward handicapped, ex­ ceptional, and ethnic minority children. Not all of you will be (or should be expected to be) equally effective in work­ ing with all types of exceptional children, every age group, and all ethnic groups . If you are having difficulty establish­ ing rapport with certain children, try to determine why. Whenever you recognize that you are not fully capable of establishing rapport with a child, you should disqualify yourself from testing the child and ask a colleague to complete the evaluation . It is only through encounters with a variety of children and through a willingness to be open to self-evaluation and feedback from others that you will learn the l imits of your own abilities . You should contin­ ually seek such knowledge by monitoring your own behav­ iors and reactions and by requesting feedback from colleagues . Examiner Expectancy Effects

Your test administration should not be influenced by your personal impressions of the examinee - a reaction known as the halo effect. This may occur when you overrate the responses of a child whom you perceive as bright or underrate the responses of a child who appears dull. In the testing of handicapped children, the halo effect may occur in the following way: Motivated by a feel ing of sympathy often reinforced by seeing the physical energy expended by so many palsied children in fol l ow­ ing instructions, the examiner easily believes his [or her) hope, i . e . , that the child knows more than h e [or she) can express , and hence overestimates the child's abil ity. (Burgemeister, 1962, p. 117)

Binet and Wechsler carefully sought to diminish halo ef­ fects by their standardization procedures, but examiners too must take precautions to avoid them .

1 13

EXAMINER CHARACTERISTICS

G U I DA N C E OFF ICE

' /1I1r. Merrill 's real strength is his ability to empathize in one-to-one situations. " Cartoon first appeared in New Era. Used by permission.

Social-psychological research indicates that an experi­ menter's hypotheses or expectancies may exert sOme sub­ tle influence over a subject's performance by affecting how the experimenter behaves with a subject. Early data re­ turns in an experiment may lead to the development of experimenter expectancies that can subtly affect the exper­ imenter's behavior in later interactions with the subject (Rosenthal , 1966) . Administering an individual intel­ ligence test is somewhat analogous to conducting an ex­ periment. Background information on the examinee may lead the examiner to formulate a hypothesis, albeit vague , regarding the examinee's level of intelligence, and this hypothesis may affect the exami ner-examinee rela­ tionship. Consequently, it is imperative that examiners be on guard against the halo effect when administering tests. Research on expectancy effects. Studies have exam­ ined whether information given to examiners affects their scoring of intelligence test responses provided on ques­ tionnaires or given by children in actual testing activities. Expectancy has been manipulated by providing various

kinds of case history information , such as prior test scores, grades , academic achievement history, socioeconomic sta­ tus, ethnicity, sex, and behavioral ratings. Most questionnaire studies indicate that examiners are influenced by pretest information in scoring responses, especially if the responses are ambiguous (Auffrey & Robertson, 1972; Babad, Mann, & Mar-Hayim, 1975; Donahue & Sattler, 197 1 ; Egeland, 1969; Fiscus , 1975; Grossman, 1978 ; Sattler, Hillix, & Neher, 1970; Sattler & Winget, 1970; S imon , 1969) . Generally, examinees de­ scribed as being bright are likely to receive more credit than examinees described as being dull for the exact same response. No significant scoring differences were found for protocols of black and white children (Mishra, 1983) or in a study where responses were presented in accented or nonaccented speech by means of a tape recorder (Rap­ paport & McAnulty, 1985) . Findings from studies in which examiners actually tested children are not so clear. In some cases positive expectancies led to children's obtaining higher scores (Hersh, 1971; Larrabee & Kleinsasser, 1967 ; Schroeder &

1 14

Kleinsasser, 1972), but in others they did not (Dangel, 1972 ; E kren , 1962 ; Gillingham, 1970; Saunders &. V itro, 1971 ; Sneed , 1976) . In the latter cases, examiners probably resol ved any discrepancies created by the referral informa­ tion by re l yi ng on the children's actual test performance. In actual test situations, the pretest information may color the ex aminer's attitudes initially, but these attitudes are likely to be les s salient than the information the examiner is continual l y acquiring. The examiner adjusts his or her own judg ments as additional information is acquired . For an expec t ancy effect or self-fulfilling prophecy to occur, the exam i ner must ignore the child's actual performance. The st\ldies reviewed indicate that pretest information plays a role in the overall evaluation of the examinee and, on occas ion , in the scoring of responses and the adminis­ tration of the test. The research does not suggest that the existence of expectancies regarding an examinee's proba­ ble level o f performance means that scoring bias will necessarily occur. Rather, the research confirms that ex­ aminers must guard against the occurrence of halo effects w hen admi nistering intelligence tests. The processes by which pretest information affects the examiner's perfor­ mance are still unclear.

Awareness of expectancy effects. The elimination of halo effects in administering , scoring , and evaluating intel­ lige nce :md other ability tests is a difficult goal . It is probabl i mpossible to eliminate completely your positive or negat ive evaluations of the child ; however, it is possible and necessary to minimize the influence of these reactions o n your test administration and scoring of the child's test responses . Be sure that you do not smile, nod your head, l ean forward , sustain additional eye contact, offer more support, be friendlier, give more praise, repeat questions more frequently, or create a warmer atmosphere with ex­ aminees who you believe are bright. You must become aware of your reactions to each child, and be especially alert to possible halo effects in your scoring of a child's ambiguc>us responses.

Variability Among Examiners

Research conclusively indicates that examiners differ in h ow theY score the same test responses (Bradley, Hanna, & Lucas, '.980; Brannigan, Calnen, Loprete , & Rosenberg, 197 6; Brannigan, Rosenberg , Loprete , & Calnen, 1977; Cuenot & Darbes, 1982 ; Curr & Gourlay, 1956; Davis, Peacod , Fitzpatrick , & Mulhern , 1969; Franklin, Still-

CHAPTER 5

TESTING C H I L D REN

man, Burpeau, & Sabers, 1982; Jordan , 1932 ; Kaspar, Throne , & Schulman , 1968; LaCrosse , 1964 ; M ahan, 1963 ; Masling, 1959; Massey, 1964; M iller & Chansky, 1972 ; M iller, Chansky, & Gredler, 1970; Plumb & Charles, 1955; Ryan , Prifitera, & Powers, 1983 ; Sattler, Andres, Squire, Wisely, & Maloy, 197 8 ; Sattler & Ryan , 1973a; Sattler & Squire , 1982 ; Sattler, Winget, & Roth, 1969; Schwartz, 1966; Scottish Council for Research in Educa­ tion , 1967 ; Smith , May, & Lebovitz, 1966; Walker, Hunt, & Schwartz, 1965 ; Wrightstone , 1941) . These reports indi­ cate that the examiner's experience in giving tests probably makes only a minor contribution , if any, to this type of examiner variability (Curr & Gourlay, 1956; Davis et aI . , 1969; Jordan, 1932; Kaspar et al . , 1968; LaCrosse , 1964; Masling, 1959; Plumb & Charles, 195 5 ; Ryan et al . , 1983; Sattler & Ryan , 1973a; Schwartz, 1966; Smith et al . , 1966) . In addition , examiners occasionally differ in the responses they obtain from children (Bennett, 1970; Cattell , 1937; Cieutat, 1965; Cieutat & Flick, 1967 ; Cohen, 1950 , 1965 ; Curr & Gourlay, 1956; Davis et al . , 1969; Di Lorenzo & Nagler, 1968; Green, 1960-62 ; Kaspar et al . , 1968; Krebs, 1969; Nichols, 1959; Oakland, Lee , & Axelrad, 1975; Rothman, 1974; Sattler, 1966, 1969; Sattler & Theye , 1967 ; Schachter & Apgar, 195 8 ; Schwartz & Flanigan , 1971; Smith & M ay, 1967 ; Thomas , Hertzig, Dryman , & Fer­ nandez, 1971). In some studies children performed better for female examiners than for male examiners , but no systematic trends are evident (Back & Dana, 1977; Brad­ bury, Wright, Wal ker, & Ross, 1975 ; Cieutat & Flick, 1967; Pedersen, Shinedl ing, & Johnson, 1968; Quereshi, 1968; Smith et al . , 1966) . With ethnic minority children , the examiner's race plays a negligible role (Sattler & Gwynne, 1982b) . (See Chapter 19 for further information about the examiner's race as a variable . ) Little is known about the temperament variables that are related to exam­ iners' scoring or administration styles (Egeland, 1967 ; Sattler, 1973b; Sattler & Martin , 1971). The preceding research confirms clinical impressions that the examination setting has elements of subjectivity. Every attempt should be made to reduce any sources of subjectivity related to the examiner that may affect the validity of the test score. You should become aware of your scoring standards when confronted with ambiguous re­ sponses - whether you are lenient or rigorous in your judg­ ments - and always try to be fair and consistent in your application of the scoring criteria. Table 5-4 was designed to assist you in obtaining information about your adminis­ trative techniques . During the early phases of your career, you should answer the questions in Table 5-4 after every assessment you complete. The list can also be used as a

SUMMARY

checklist by an observer. An understanding of child psy­ chology and exceptionali ty, of test manuals, and of mate­ rial in this book, coupled with a willingness to eva1 ua t e and reflect on your testing skills, will enable you to reac h a high level of competence in testing children.

A F I NAL THOU G H T-THROUGH A C H I LD ' S EYES

As repeatedly observed in this chapter, children play an active role in the testing process. They will react to you . just as you react to them. Sometimes you may actually receive feedback from the children you test. Exhibit 5-1 presents a tongue-in-cheek example of such feedback.

S U M MARY

I . Every effort should be made to establish rapport with children. This can be done by using praise, understanding com­ ments, and meaningful explanations . The assessment aims to elicit children's best efforts . 2 . Research studies suggest that familiarity with the exam· iner tends to enhance the performance of low SES children (b) about 7 . 6 points when the test has an SD of I S ) . 3 . Observing the c h i l d carefully during t h e evaluation is a n important part o f the evaluation process . Make notes o f the child's behavior, rather than rely ing on memory. Attend to the child's attitude toward you and the test situation, attitude toward self, work habits, reaction to test items, reaction to failure and praise, language, visual-motor skil l s , and motor skill s . 4 . Some o f the keys to successful testing o f infant and pre­ school children include l istening to the children's communica­ tions and timing your communications appropriately. 5 . In testing ethnic m i nority children, openness and fran­ kness in acknowledging miscommunication may help to facilitate rapport . 6. The key to testing emotionally disturbed children is under­ standing their particular problems and adapting the testing tech­ niques to their needs. 7. Because delinquent childre n are l i kely to be defensive, attempts must be made to maximize their cooperative effort s . 8 . Autistic children are l i kely to b e extremely difficult to test . Before testing, learn about the child's preferred mode of commu­ nication. Flexibility will be needed to obtain an estimate of the autistic child's abil ities. 9 . In testing brain-inj ured children (as well as all other child ren), it is important to reduce any anxiety that is generated by the test situation. 1 0 . Mentally retarded children may try to avoid difficult ques­ tions. By showing acceptance and reducing the threatening as-

1 15

pects of the test situation, you can enhance thei r per fonnance and reduce avoidant behaviors . I I . The administration of standardized tests to physical ly handicapped children requires patience, understandin.g , and flex­ ibil ity. The test items administered to physically hundicapped children must be ones to which they can respond. Informally evaluate the child's sensory functions before you begi n the evalua­ tion . Be careful not to allow your expectations to bias your judgments. 1 2 . In testing visually impaired children, you m ust first iden­ tify the extent of their impairment and how it may inte rfere with their abil ity to take tests. You can then select appropril:\te tests, or portions of tests, based on your fi ndings. 1 3 . The assessment of hearing-impaired children is partic u­ larly difficult because of the unique communication problems associated with their disability. You should recogni ze your own l imitations as an examiner and be prepared to learn t he necessa ry communication skills if you w i l l be working regu larly with hearing-impaired children. Although hearing-impaired children differ as to their degree of hearing loss, nonverbal tests usual ly provide the most valid means of assessing their level ()f i ntellec­ tual functioning. 14. Flexibility is needed i n the testing of cerebr al palsi ed children. If tests are administered carefully, the results obtained are likely to provide an accurate estimate of the level of cogniti ve functioning of cerebral palsied children . 1 5 . Skill in the administration of tests develops wi th experi ­ ence. A n examiner must become adept at developi ng empathy and understanding, maintaining l imits, arranging an appropriate environment, using appropriate reinforcements, and d eveloping a harmonious relationship with parents . Practice is i m portan t , especially in learning how t o read the directions, ha ndl e the te st materials, and score the responses . The guidelines presented in this chapter are designed to supplement those that apPe ar in test manuals . Remember that the scoring guidelines in t est manuals are just guides , and that in most cases the answers are (lot meant to be the only correct answers. 1 6 . Departures from standard procedures may be needed when handicapped children are tested. However, the resulting scores may yield less precise estimates of the child's level of ability, depending on the extent of the departure. 1 7 . Incentives, such as special praise, candy, or money, some ­ times lead to higher test scores, but l ittle is known abo ut which specific procedures are effective with which groups of children . There i s I ittle evidence t o support the position that reinfo rcers act differentially for different ethnic groups or social c lilsse s . 1 8 . O n some occasions i t i s useful t o test l i mits in order t o determine whether the child c a n solve problems when given helping cues or special procedures . Testing-of-limill; Should be carried out only after the standard administration of t he entire test. Testing-of-limits should not be used when rete sting i s planned soon after the initial testing , because it may increase the chances of obtaining invalid scores on the retest. 1 9 . A willingness to be open to self-evaluation and to receive

CHAPTER 5

1 16

TESTING C H ILDREN

Table 5-4 Checklist for General Test Administration Practices General Test Administration Practices C hecklist __ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __

Examiner:

Date:

Examinee :

Age:

Observer:

Test administered:

__ __ __ __ __ __ __ __ __ __ __ __ __ __ __ _

__ _ __ __ __ __ __ __ __ __ _ _ _

Circle one yes yes yes yes

no no no no

yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes

no no no no no no no no no no no no no no no no

yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes

no no no no no no no no no no no no no no no

I. 2. 3. 4. 5. 6. 7. 8. 9. 10. II.

12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35.

Established rapport before beginning the test Prepared child for examination Arranged test environment to minimize distractions Avoided distracting mannerisms Showed interest in child Gave child ample encouragement and support Wore appropriate, nondistracting attire Spoke at appropriate volume for the setting Appeared open and accepting of child's feelings Seemed at ease with child Maintained frequent eye contact with child Had all necessary materials present Arranged materials conveniently Placed manual so that child could not read it Arranged materials so that child could not view test items other than the one(s) in use Manipulated materials with ease and confidence Read all directions verbatim Used accurate timing procedures Used unobtrusive timing procedures Recorded responses in record booklet Paced examination to suit child's ability Explained test procedures adequately Used vocabulary suited to child Showed awareness of signs of fatigue Handled fatigue appropriately Showed awareness of emotional upsets Handled emotional upsets appropriately Took needed breaks Handled child's manipulations appropriately Gave appropriate explanations or clarifications Used additional questions to clarify, not to improve, child's answers Made inquiries in a nonthreatening manner Praised child appropriately (e.g. , did not praise correct answers, praised effort) Handled disruptions adequately Responded honestly and positively to child's questions in ways consistent with the testing

feedback from others will help beginning examiners develop

scores to examinees purported to be bright than to examinees

clinical skills. A sense of humor can also be helpful. 20. The halo effect is a problem that may arise in the course of conducting an examination. There is substantial evidence that when examiners are provided with information about the exam­ inee's probable level of performance, they may assign better

the overall evaluation of the examinee and , on occasion, in the scoring of responses and administration of tests . Recognizing the possibility of their occurrence helps to reduce halo effects .

purported to be dull . Thus pretest information may play a role in

2 1 . Examiners have been found to differ among themselves in

SUMMARY •

1 17

Exhibit 5- 1 II •

•

•

On the Other Foot"

My name is Timmy Jones and I am in the third grade. I am as smart as anyone in the Elm Street Schoo l , including grown­ ups . I even learned their language. Take that woman that tested me last wee k . Psychological Report

Name: Audrey O'Neill Age: Uncertai n Observation of behavior: Mrs. O'Neill is below average height and weight. It was a cold, wet day and she was limping slightly. Her grooming leaves something to be desired . Her clothing was dark and several years behind the style, "com­ fortable" shoes evidenced lack of concern with the impression made on others, long straight hair would be more appropriate on a younger person, and fingernails were dirty - she looks as if she spends her spare time refinishing furniture. The bag she carried her test equipment in has seen better days. The whole impression she gave was one of mild involutional depression. M rs . O'Neill actually lost her way to the supply closet where she had set up makeshift quarters even though she has been using it all semester. This was my first time and I found it right away. Attempts to establish rapport were clumsy and interview techniques emphasized the negative . Of course I don't beat up on my l ittle brother. And what did she mean by asking if everyone in my family has the same last name? She was distracted by the sound of the band practicing in the next room and preoccupied by the three phone calls that inter­ rupted the testing session. Test performance: Mrs . O'Neill made three errors administer­ ing the Stanford-Binet: Form L-M . All the better examiners use the WISC-R anyway. During a 90-minute testing session she managed to misplace her reading glasses once, her glass case twice, and her pen three times. Her handwriting is almost completely illegible, and she dropped three of the blocks. She transposed numbers twice, once administering digits forward and once transcribing a phone message. She definitely be­ came nervous when I went into my hyperactive act, and has a hard time tolerating drumming fingers, tapping feet, etc. Strengths and weaknesses: She is high in long-term memory and i nformational background . She remembered my sister and brother, and asked about my grandmother's health. She is

their scoring of test responses and in the responses that they obtain from examinees. However, the sources of these differences have not been established. The examiner's experience in giving tests does not appear to play a crucial role in the scoring of test

low in short-term memory, concentration, and auditory skills: auditory reception and discrimination and auditory figure­ ground. She is also low in sequencing, both auditory and visual . She had problems with fine-motor skills and spatial orientations. Her motor coordination is gross. Social relationships: Both child and peer relationships are in need of improvement. The children who greeted her in the hall are all either failing in school, from broken homes, or from the socioeconomic level that my parents do not allow me to associate with. Nor does she handle peer relationships well . I t took us 1 2 minutes and 4 8 seconds by the new stopwatch I got for my birthday to get from one end of the hall to the supply closet because every grown-up we passed had to stop and tell her something . No wonder she is not more task­ oriented. Summary: Mrs. O'Neill is of approximately average native endowment for her cultural group membership. However, she is not functioning well because of (a) mild depression and (b) learning disabilities, which are developmental and therefore at her age will get worse . Peer relationships distract her from her work . Recommendations: 1 . Mrs . O'Neill could probably continue to perform her duties adequately in a simpler and more protected setting . Perhaps she could be transferred to the Gilmanton Corners V illage School . 2 . A p rog ram for id�nti fy ing the gifted would improve her child contacts and broaden her perspective. 3 . She should stop e ating lunch alone while working on records. Not only did she get three mayonnaise stains on my cumulative folde r. but she is heading for a good case of professional burn-out .

Timothy Jones Grade 3 The Elm Street School Reprinted. II ith changes in notation, with permission of the publisher and author from Audrey Myerson O'Neill, " . . . On the Other Foot," JOllrnal 0/ School Psychology, 1 98 1 , 1 9, pp. 7 1 -72. Copyright 1 98 1 , Pergamon Journals, Ltd. Source:

responses. The examiner's sex is not usually a critical factor in the child's performance. Ex.aminers' personality variables have not been shown to be associ ated significantly with either scoring or administration styles.

1 18

KEY T E R M S, CONCEPTS, A N D N A M ES Rapport ( p . 86) Ech olal ia (p. 93) Emotional labi lity (p. 94) Perseveration (p. 94) Blindnes s (p. 98) Hearing impa irme nt (p. 99) Deaf child (p. 99) Hard -of-hearing child (p. 99) Mini m al hearing loss (p. 99) Mild hearing loss (p. 99) Moderate hearing loss (p. 99) Severe hearing loss (p. 99) profound hearing loss (p. 99) Social reinforcement (p. 110) Token reinforcement (p. 110) Testing-of-lim its (p. 110) Expectancy effects (p. 112) Halo effect (p. 112)

STUDY QU ESTIO N S

1 . Explain how a n examiner should g o about establishing rapport with children. Include a discussion of research studies.

CHAPTER S

TESTING CHILDREN

2. What are some useful strategies for testing infant and preschool children? 3. What are some factors to consider in testing ethnic minor­ ity, emotionally disturbed, delinquent, autistic, brain-injured, and mentally retarded children? Include in your discussion both unique and common testing principles for these groups. 4. What special adjustments may be needed to test physically handicapped children? 5. What are some useful techniques for testing visually im­ paired children? 6. What are some factors to consider in testing hearing­ impaired children? 7. What are some useful techniques for testing cerebral palsied children? 8. What are some general suggestions for administering tests? 9. Discuss the problems associated with departures from standard procedure . Include relevant research on the use of incentives and feedback. 10. Discuss the concept of testing-of-limits. I I . Discuss the concept of halo effects. Include relevant research.

WECHSLE R I NTE LLI GE NCE SCALE

FOR C H I LDREN - REVISED (WISC-R) : DESC RI PTION

Standardization

The concept [of intelligence} has as much scientific status as does the concept of gravity, and the global measures of intelligence, although limited for diagnosTic purposes, do help tell us where all our other more experimental measures are located in a genera! correlational map of human cognitive variations.

Deviation IQs, Scaled Scores, and Test-Age Equivale nts Reliability Validity lntercorrelations Between Subtests and Scales

-Richard E. Snow

WISC-R IQs and Stratification Variables Factor Analysis Administering the WISC-R Short Forms of the WJ SC-R Choosing Between the W ISC-R and the WPPSI and Between the WISC-R and the WAIS-R ..\dministering the WISC-R (and WPPSI and WAIS-R) to Handicapped Chil dren Assets of the WISC-R Limitations of the WISC-R Concluding Comment on the WISC-R Summary

1 19

Exhibit 6- 1 WISC-R-Like Items

Information (30 questions)

Picture Arrangement ( 1 2 items)

How many legs do you have? What must you do to make water freeze? Who discovered the North Pole? What is the capital of France?

The task is to arrange a series of pictures into a meaningful sequence. The photograph of the WISC-R (Figure 6- 1 ) shows a Picture Arrangement item. Block Design

Similarities ( 1 7 questions)

In In In In

what way are what way are what way are what way are

( I I items)

The task is to reproduce stimulus designs using four or nine blocks. An example of a Block Design item is shown below.

pencil and crayon alike? tea and coffee alike? inch and mile alike? binoculars and microscope alike?

Arithmetic ( 1 8 questions)

If I have one piece of candy and get another one, how many pieces will I have? At 1 2 cents each, how much will 4 bars of soap cost? If a suit sells for 1f2 of the ticket price, what is the cost of a $ 1 20 suit?

Object Assembly (4 items)

The task is to arrange pieces into a meaningful object. An example of an Object Assembly item is shown below.

Vocabulary (32 words)

ball summer

�

poem obstreperous

. •, ��

Comprehension ( 1 7 questions)

Why do we wear shoes? What is the thing to do if you see someone dropping his packages?

� �

I '

In what two ways is a lamp better than a candle? Why are we tried by a jury of our peers?

�.

Courtesy of The Psychological Corporation.

Digit Span

Coding

Digits Forward contains seven series of digits, 3 to 9 digits in length (Example: 1 -8-9).

The task is to copy symbols from a key (see below) .

Digits Backward contains seven series of digits, 2 to 8 digits in length (Example: 5-8- 1 -9) . Picture Completion (26 items)

The task is to identify the essential missing part of the picture. A picture of a car without a wheel. A picture of a dog without a leg. A picture of a telephone without numbers on the dial .

SAMPLE

An example of a Picture Completion task is shown below.

I

Courtesy of The Psychological Corporation.

Mazes

The task is to complete a series of mazes . The photograph of the WISC-R (Figure 6- 1 ) shows an example of a maze. Note.

The questions resemble those that appear on the WISC-R but

are not actually from the test. Chapter 7 describes each subtest in Courtesy of The Psychological Corporation.

more detail.

1 20

121

DEVIATION IQS. SCALED SCORES. AND TEST-AGE EQU 'VALE NTS

The Wechsler Intelligence Scale for Children - Revi sed (WISC-R) (Wechsler, 1974) was published in 1974, 25 years after the original publication of the WISC . Its prede­ cessor, the WISC (Wechsler, 1949), was developed as a downward extension of the adult intelligence test, the Wechsler-Bellevue Intelligence Scale. To make the origi­ nal adult scale more suitable for children, easier items were added to the beginning of the subtests. The WISC-R covers an age range from 6-0 to 16-11 years and contains 12 subtests (see Figure 6-1) . Six of the tests form the Verbal Scale (Information, Similarities, Arithmetic, Vocabulary, Comprehension , and Digit Span) and the other six form the Performance Scale (Picture Completion, Picture Ar­ rangement, Block Design, Object Assembly, Coding, and Mazes). Items similar to those on the WISC-R are shown in Exhibit 6-1 . A total of 72 percent of the WISC items are retained in the WISC-R, either intact (64 percent) or with substantial modification (8 percent) ; the items in the Cod­ ing subtest are the same i n the WISC and the WISC-R.

STA N DARDI ZAT I O N

The WISC-R was standardized on a sample of 2 ,200 American children selected as representative of the popu­ lation on the basis of the 1970 U . S . Census. In the stan­ dardization sample, there were 11 different age groups, ranging from 6-6 to 16-6 years, with 200 children in each group. Unlike the WISC , which included only whites in the standardization group, the WISC-R included non­ whites (blacks, American Indians, Asians, Puerto Ricans, and Mexican Americans) . The proportions in the WISC-R sample approximate those in the 1970 Census more closely for whites than for nonwhites_ (The discrepancy between the standardization sample and the census data is no greater than 1 percent for whites, but is as high as 4 . 5 percent for nonwhites . ) The extent of the effect o f these discrepancies on scaled test scores cannot be assessed . Given the relatively small size of the discrepancy, however, the effects should not be large .

D EVIAT I O N IQS, SCALED SCORES, AND T EST·AG E EQU IVALENTS

The WISC- R , like other Wechsler scales, provides three separate IQs: a Verbal Scale IQ, a Performance Scale IQ, and a Full Scale IQ . All three are Deviation IQs obtained by comparing the examinees' scores with those earned by a representative sample of age peers. Because Deviation IQs

are standard scores, at each age level each of the three IQs has the same mean of 100 and standard deviation of 15 . When the WISC-R is administered, raw scores are ob­ tained on each subtest. The raw scores are then converted to normalized standard scores (or scaled scores) within the examinee's own age group through use of a table in the WISC-R manual . This table is based on four-month age i ntervals between 6-0-0 (years , month s , days) , and 16-11-30. The scaled scores for each subtest have a mean of 10 and standard deviation of 3 . Prorating Procedure

The IQ tables in the WISC-R manual are based on only 10 of the 12 subtests. The two supplementary subtests, Digit Span and Mazes, are excluded from the calculation of the IQ even when they have been administered. When fewer than 10 subtests are administered , prorating is necessary to compute the IQ. The WISC-R manual provides a table for prorating the scores when four of the subtests are adminis­ tered in each scale . When fewer than four subtests are administered in each scale, IQs should be computed using the special short-form procedure described later in this chapter. Test-Age Equivalents

In developing the WISC-R and the other Wechsler scales, Wechsler did not use mental age to calculate IQs because he believed that the concept was potentially misleading. (See Chapter 4 for criticisms associated with the mental­ age concept . ) He rejected the notion that mental age repre­ sents an absolute level of mental capacity or that the same mental age in different children represents identical intel­ ligence levels . Soon after the initial publication of the WISC , however, he recognized that mental-age or test -age equivalents (average age associated with a score) would be useful. In subsequent publications of the WISC and WISC-R, Wechsler provided a table oftest-age equivalents to facilitate interpretation . Test-age equivalents are essen­ tially mental age (MA) scores. The WISC-R test-age equivalent scores can be com­ pared with scores on other tests that use mental ages or test ages, and reported to parents, teachers , and other indi­ viduals . Test ages are obtained directly from raw scores on each subtest . An average test-age equivalent can be ob­ tained by summing the individual subtest age equivalents and dividing by the number of subtests . A median test age can be obtained by rank ordering all of the test ages from high to low and finding the median age. The WISC-R test

CHAPTER 6

1 22

Figure 6- 1 .

WECHSLER I NTELLIGENCE SCALE FOR CHILDREN-REVISED (WISC-R): DESCRIPTION

Wechsler I ntel ligence Scale for C h ildren-Revised. Courtesy of The Psychological Corporat ion.

ages have adequate validity, as indicated by high correla­ tions with the Stanford-Binet: Form L-M mental age (r = . 88) (Sutton, Koller, & Christian, 1982) and the Pea­ body Individual Achievement Test (r = . 82) (Huberty & Koller, 19 84) . R E LIA 8 1 LITY

The WISC-R has outstanding reliability. Each of the three IQ scales has an internal consistency reliability coefficient of . 89 or above in the standardization group over the entire age range covered by the scale. Average internal consis­ tency rel iability coefficients, based on the 11 age groups, are . 96 for the Full Scale IQ, . 94 for the Verbal Scale IQ, and .90 for the Performance Scale IQ . Table 6-1 summarizes average reliabilities and standard

errors of measurement for the 12 subtests and three scales . Internal consistency subtest reliabilities , although ade­ quate, are less satisfactory than scale reliabil ities . The average subtest reliability coefficients range from a low of . 70 for Object Assembly to a high of . 86 for Vocabulary. The highest reliabilities are generally found among the Verbal Scale subtests, although one Performance Scale subtest, Block Design, has a reliability coefficient of . 85 . The average reliability coefficients range from . 77 to . 86 (Mdn = . 80) for the Verbal Scale subtests and from . 70 to . 85 (Mdn = . 72) for the Performance Scale subtests . When the reliabil ity coefficients are considered for each of the 11 age groups separately, they range from a low of .57 for Mazes at the 16-6 age level to a high of .92 for Vocabu­ lary at the 16-6 age level . The reliability coefficients are similar, for the most part, across the 11 age groups .

1 23

VALIDITY Table 6- 1 Average Reliability Coefficients and Standard Errors o·f Measurement for WISC-R Subtests and Scales

Average reliability coefficient

Average standard error of measuremenr

Infonnation S i milarities Arithmetic Vocabulary Comprehension Digit Span

. 85 .81 . 77 .86 . 77 .78

1 . 19 1 . 34 1 . 38 1 . 15 1 . 39 1 .44

Picture Completion Picture Arrangement Block Design Object Assembly Coding M azes

.77 .73 . 85 . 70 . 72 . 72

1 .45 1 . 57 1 . 17 1 .70 1 . 63 1 . 70

Verbal Scale IQ Perfonnance Scale IQ Full Scale IQ

. 94 . 90 .96

3 . 60 4 . 66 3 . 19

Subtest or scale

Note.

Reliabil ity coefficients for 1 0 of the

1 2 subtests (except Digit Span

Stability

The WISC-R provides stable IQs for each of the three scales . In the standardization sample the stability of the WISC-R was assessed by retesting a group of 303 children from three age g roups a fter a one-month interval (Wechsler, 1974) . For the retest sample, the stability coeffi­ cients were . 95 for the Full Scale IQ, . 93 for the Verbal Scale IQ , and . 90 for the Performance Scale IQ; those for the 12 subtests ranged from . 65 (Mazes) to . 88 (Informa­ tion) ( Mdn .78) . The mean test-retest IQs and srandard deviations for the Verbal , Performance, and Full Scales for the three groups are shown in Table 6-2 . The average increase in IQ from the first to the second testing wa S about 7 IQ points for the Full Scale, about 4 IQ points for the Verbal Scale, and about 10 IQ points for the Performance Scale. These differences, which can be considered to be the result of practice effects, are much greater for the Performance Scale than for the Verbal Scale. Other studies (Haynes & Howard , 1986; Naglieri & Pfeiffer, 1983a) also report relatively stable WISC-R IQs over a two-year inter­ val (mean Verbal , Performance, and Full Scale IQ changes were less than 3 points). =

and Coding) are split-half correlations. For Digit Span and Coding the reliability coefficients are test-retest coefficients obtained on a sample of about 50 children in six different age groups who were retested after a one-month interval . Verbal , Performance, and Full Scale reliability co­ efficients are based on a formula for computing the reliability of a composite group of tests.

Source: Adapted from Wechsler ( 1 974).

Standard Errors of Measurement

The standard errors of measurement (SEm) in IQ points, based on the average of the 11 age groups, are 3 . 19 for the Full Scale, 3 . 60 for the Verbal Scale, and 4 . 66 for the Performance Scale (see Table 6-1) . Thus more confidence can be placed in an IQ based on the Full Scale than in an IQ based on either the Verbal or the Performance Scale. The standard errors of measurement for the subtests in scaled score points, based on the average of the 11 age groups, range from 1 . 15 to 1 .44 for the Verbal Scale sub­ tests and from 1 . 17 to I. 70 for the Performance Scale subtests . Within the Verbal Scale, Vocabulary and Infor­ mation have the smallest SEm ( 1 . 15 and 1 . 19 , respectively ) . With in the Performance Scale, Block Design and Picture Completion have the smallest SEm ( 1 . 1 7 and 1 . 45 , respectively ) .

Precision Range

Table C-I in Appendix C shows the confidence intervals for the 68 . 85, 90, 95 , and 99 percent levels for each scale. The confidence intervals are provided for separate age levels as well as for the average of the age levels. The child's specific age group -not (he average o/the 11 age groupS ­ should be used (0 obtain the most accurate confidence interval. The preci sion ranges of the WISC-R IQs are similar throughout the age levels covered by the scaJe. For further discussion of precision range (or confidence inter­ vals) , see Chapters 2 and 2 3 .

VALI D I TY

A variety of studies have investigated the criterion validity of the WISC-R by correlating the WISC-R with the WPPSI, WAIS-R, WAlS, Stanford-Binet: Fourth Edition, Stanford-Binet: Form L-M, other intelligence test , and measures of achievement and school grades. Table 6-3 shows the median correlations between the WISC-R Ver­ bal , Performance, and Full Scales and other intelligence tests , achievement tests, and school grades . Although the table is not exhaustive, the representative studies listed show that the WISC-R has satisfactory concurrent validity.

CHAPTER 6

1 24

WECHSLER INTELLIGENCE SCALE FOR CHILDRE N - REVISED (WISC-R): D ESCRI PTION

Table 6-2 Test-Retest WISC-R IQs for Three Groups of Children First testing

Second testing

Scale

Mean IQ

SD

Mean IQ

SD

Change

6 '/2 -7 '12 ( N 97)

Verbal Performance Full

98.3 97.9 97.9

12.3 14.5 13.2

1 02 . 2 106.5 1 04 . 5

12.7 15.0 14.0

+3.9 + 8.6 + 6. 6

1 0 '/2 - 1 1 '/2 (N 1 02 )

Verbal Performance Full

99.0 98.5 98.6

1 3 .4 13.9 13.7

1 02 . 4 1 09 . 3 1 06 . 2

13.8 16.3 15. 1

+ 3 .4 + 10.8 +7.6

1 4 '/2 - 1 5 '12 (N 1 04 )

Verbal Performance Full

96. 8 96 . 2 96 . 1

16. 1 14.4 15.4

1 00.0 105.4 103.0

17. 1 18.2 1 8 .0

+3.2 +9.2 + 6. 9

Age

=

=

=

Reprinted, with a change i n notation, from the WISC-R manual, pages 3 2 - 3 3 . Reproduced b y permission. Copyright 1 974 by The Psychological Corporation, New York, N. Y. All rights reserved.

Source:

Median correlations range from the upper . 30s to the low . 80s . WISC-R and WPPSI

Because the WISC-R overlaps with the WPPSI in the age range of 6-0-0 to 6-7 -15 , either of the two tests can be used to evaluate childre n i n this age range . I n one study (Wechsler, 1974), a representative sample of 50 male and female 6-year-olds were given the two tests in counter­ balanced order. The correlations were . 80 for the Verbal Scales, . 80 for the Performance Scales, and . 82 for the Full Scales. Although the mean IQs on the Verbal , Perfor­ mance, and Full Scales were higher on the WPPSI than on the WISC-R, the differences were small ( 1 . 5 , 2 . 8 , and 2 . 5 I Q points, respectively) . Another study (Quereshi & McIntire, 1984) o f 72 6-year-old normal children, which also used a counter­ balanced design, indicated that WISC-R and WPPSI IQs were comparable - WISC-R IQs were slightly higher than WPPSI IQs (differences were . 73 on the Verbal Scale, 2 . 25 on the Performance Scale, and 1 .45 on the Full Scale) . Correlations were uniformly high for the three IQs (r's = . 86, .77, and . 85 , respectively) . In a study (Rasbury, McCoy, & Perry, 1977) in which 5to 6-year-old middle-class children were administered both tests, with a one-year test-retest interval , the WISC-R yielded IQs that were 5 points lower on the average than those of the WPPSI . Correlations between the two tests were . 8 1 for the Verbal Scales, . 80 for the Performance

Scales, and . 94 for the Full Scales . Individual subtest scores on the WISC-R were approximately 1 scaled-score point lower than those on the WPPSI. In 10 percent of the cases, differences between the two tests were larger than 10 IQ points. Consequently, in retest situations where the WISC-R follows the WPPSI or vice versa, differences in the two IQs may reflect errors of measurement rather than any changes in the child's ability. WISC-R and WAlS-R

Because the WISC-R overlaps with the WAIS-R in the range 16-0-0 to 16-11-30, either of the two tests can be used to test children in this age range. Table 6-4 summarizes the results of studies in which seven different samples were given both the WISC-R and the WAIS-R and one sample was randomly assigned either the WISC-R or the WAIS-R . Firm conclusions about the comparabi lity of the two scales are difficult because some of these studies had retest inter­ vals as long as six years whereas others were conducted within the same day. Populations were also varied , includ­ ing the mentally retarded, special education students , the hearing impaired, delinquents , and nonreferred children; few studies had representative samples. Still, some trends appear in the findings. The WAIS-R yielded higher IQs than the WISC-R i n some groups, particularly in groups of mentally retarded or moderately below average students . O ne sample of mentally retarded children obtained WAIS-R IQs that were , on the average, 11 points higher than their WISC-R IQs (Rubin et al . , 1985 ) . The findings

1 25

INTERCORRELATIONS BETWEEN SUBTESTS AND SCALES

Table 6-3 Criterion Validity Studies for the WISC-R

Median correia lions Crilerion

Stanford-Binet: Fourth Edition Stanford-Binet: Form L-M Siosson Intelligence Test McCarthy Scales of Children's Abilities K-ABC Woodcock-Johnson Broad Cognitive Ability Group intelligence tests Peabody Picture Vocabulary Test - R Quick Test Wide Range Achievement Test Reading Spelling Arithmetic Peabody Individual Achievement Test Other achievement tests Reading Arithmetic School grades

Verbal Scale

Performance Scale

Full Scale

. 78

.75 .75

.68 .51

.68 . 50

. 62 . 65

. 72 . 70

. 77 .61

.55 .59

. 77 .66

.72 .76

.47 . 68

.68 . 72

.57 . 50 . 62

.34 .26 .46

.56 .59 .52

.75

.45

.71

. 66 . 56

.47 .48

. 82 .61

quent, and hard of hearing, the two scales yielded com­ parable IQs . The Performance Scales may be more com­ parable than the Verbal Scales, but this hypothesis needs further verification . Little is known about the com­ parability of the two scales for gifted adolescents. WISC-R and Stanford-Binet: Fourth Edition

Four studies comparing the WISC-R and the Stanford­ Binet: Fourth Edition are reported in the Technical Manual of the Fourth Edition (Thorndi ke , Hagen , & Sattler, 1986b) . As Table 6-5 indicates , the WISC-R yielded slightly higher IQs than did the Stanford-Binet: Fourth Edition, even though the latter has a larger standard devi­ ation than the WISC-R (16 vs. 1 5 ) . Correlations ranged from . 66 to . 83 between the WISC-R Full Scale IQ and the Fourth Edition composite . These results suggest that the two tests yield scores that are approximately equal , but they are not interchangeable. In another study (Carva­ jal & Weyand, 1986) the correlation between the WISC-R and the six-subtest General Purpose Abbreviated Battery on the Fourth Edition was .78; the Full Scale IQ and Composite Score were very close - 1 15 . 0 and 1 1 3 . 3 , respectively.

.65

.58 . 39

SOUTee: This table is based on studies cited in Sattler ( 1 982b) and on the fol lowing studies: A1gozzine and Ysseldyke ( 1 98 1 ) ; Altepeter and Handal ( 1 986); Appelbaum and Thma ( 1 982) ; Arinoldo ( 1 982); Bracken . Prasse, and Breen ( 1 984); Breen ( 1 98 1 ) ; Breen and Siewert ( 1 983); Coleman and Harmer ( 1 985) ; Crofoot and Bennett ( 1 980); Davis and Kramer ( 1 985); Estabrook ( 1 984); Grossman and Johnson ( 1 982) ; Had­ dad ( 1 986); Hollinger and Sarvis ( 1 984); Hutton and Davenport ( 1 985); Hynd, Quackenbush, Kramer, Connor, and Weed ( 1 980); Kaufman and Kaufman ( 1 983); Kitson and Vance ( 1 982); KJanderman, Devine, and Mollner ( 1 985); Marshall , Hess, and Lair ( 1 978); Mask and Bowen ( 1 9 84); McCallum, Karnes, and Edwards ( 1 984); McGrew ( 1 983); Naglieri ( 1 98 2 , 1 984, 1 985a, 1 985b); Naglieri and Anderson ( 1 985); Nag1ieri and Haddad ( 1 984); Naglieri and Yazzie ( 1 983); Oakland and Dowling ( 1 983); Obrzut, Obrzut, and Shaw ( 1 984); Paramesh ( 1 982) ; Phelps, Rosso, and Falasco ( 1 985) ; Pommer ( 1 986); Prasse and Bracken ( 1 98 1 ) ; Reilly, Drudge, Rosen, Loew, and Fischer ( 1 985); Rosso, Falasco, and Koller ( 1 984); Rust and Lose ( 1 980); Smith and Smith ( 1 986); Sutter and Bishop ( 1 986b); Thompson and Brassard ( 1 984) ; Tramill, Tramill , Thornthwaite, and Anderson ( 1 9 8 1 ) ; White ( 1 979); Worthing, Phye, and Nunn ( 1 984); Wurtz, Sewell, and Manni ( 1 985); Zins and Barnett ( 1 984).

suggest that some children who are classified as mentally retarded on the WISC-R may not be so classified on the WAIS-R . This hypothesis is based on test-retest intervals of about three or four years . In studies with nonreferred children, as well as in studies with children who are somewhat below average, del in-

I N TE RCORRE LAT I O N S B E TW E E N SUBTESTS AND SCALES

lntercorrelations provide i nformation about the rela­ tionships of WISC-R subtests to each other and to the scales. Average intercorrelations between the 12 subtests range from a low of . 19 to a high of . 69 (Mdn = .40) . The six highest subtest average intercorrelations are between Vocabulary and Information ( . 69), Vocabulary and Sim­ ilarities ( . 67), Vocabulary and Comprehension ( . 66) , In­ formation and Similarities (.62), Block Design and Object Assembly ( . 60), and Similarities and Comprehension ( . 59) . The six lowest subtest average intercorrelations are be­ tween Picture Completion and Coding ( . 19), Picture Com­ pletion and Digit Span ( . 21) , Object Assembly and Digit Span (.21), Coding and Mazes ( . 2 1 ) , Picture Arrangement and Digit Span ( . 22) , and Mazes and Digit Span ( . 22). Thus Verbal Scale subtests are more highly intercorrelated than are the Performance Scale subtests. Average correlations between the Verbal Scale subtests and the Verbal Scale range from .45 to .78 (Mdn = .70) ; those between the Performance Scale subtests and the Performance Scale range from . 3 3 to . 68 (Mdn = .53) . The intercorrelations between the individual subtests and the Full Scale exhibited a similar trend. Vocabulary has the

CHAPTER 6

1 26

WECHSLER INTELLIGENCE SCALE FOR CHILDREN-REVISED (WISC-R): DESCRIPTION

Table 6-4 Studies Comparing the W I SC-R and the WAIS-R

N

Study

Braden & Paquin ( 1 985) Grace ( 1 986) Meacham ( 1 984) Rubin , Goldman, & Rosenfeld ( 1 985) Sattler, Polifka, Polifka, & Hilsen ( 1 984) Wechsler ( 198 1 ) Zimmerman, Covin, & Woo-Sam ( 1 986)

32 b 55 e 37 b 41d 3()e 80r 5()e 40e

Verbal Scale

Test-retest interval

WISC-R

WAlS-R

Diff '

7 8 .08

8 1 . 95

+ 2 . 87

5 8 . 80 8 1 .00 99.40

70.05 8 1 . 30 99.60

75 . 80 70. 60

8 1 . 20 75 . 30

+ 1 1 . 25 + . 30 + .20 + 5 .4 +4.7

r

1 to 6 years

same day same day 3 t o 7 years 4 years 1 to 6 weeks 4 years 4 years

.80 . 76 . 89 . 84 .57

a

In the difference column (diff. ) , a plus ( + ) sign means that WAIS-R IQs were higher than WISC-R IQs, and a minus ( - ) sign means that WAlS-R IQs were lower than WISC-R IQs. b Subjects were in a residential school for the deaf. c S u bjects were i n c a rcerated male d e l i nquent s . This w a s not a test-retest study. S u bj ects were random l y a s s i g ned to e i t her the WISC-R (N = 25) or the WAIS-R (N = 30) .

highest correlation of any of the subtests with the Full Scale ( . 74) , followed by Information ( . 70) and Block De­ sign ( . 68) . Digit Span (.43) and Coding ( . 3 8) have the lowest correlations with the Full Scale. Within their re­ spective scales, Vocabulary ( . 78) and Block Design ( . 68) have the highest correlations with the Verbal and Perfor­ mance Scale IQs.

WISC·R IQS AND STRATI F I CATIO N VARIABLES

Table 6-6 shows the relationship between WISC-R IQs and the demographic characteristics of the standardization

sample. Differences between the boys' and girls' mean IQs on the three scales were less than 3 points. Thus sex differences are not large enough to assume any practical significance on the scal e . Mean IQs of white children were about 1 standard deviation higher than those of black chil­ dren (102 vs. 86) . Mean IQs show a clear relation to parental occupation group: for the total group, children of professional and technical workers obtained Full Scale IQs that were 21 points higher, on the average, than those of children of unskilled workers (108 . 2 vs. 87.3). Urban­ rural differences on the three scales were smal l - less than 3 points. IQs of children from the South were on the average about 6 points lower than those of children from the West (96 . 6 vs . 102 . 6) .

Table 6-5 Comparison of WISC-R and Stanford-Binet Intelligence Scale: Fourth Edition

Stanford-Binet: Fourth Edition

WISC-R

Sample

N

23 Normalb 205 NormalC 19 Gifted' Learning disabled' 90 61 Mentally retarded'

Verbal Performance Scale Scale IQ IQ

Full Scale IQ

Verbal Reasoning

\03.9 1 17.6 85.2 66 . 2

1 15.0 \ 05 . 2 1 1 7.7 87 . 8 67 . 0

104 . 4 1 13.5 88.8 69.4

\ 05 . 3 1 14 . 5 92 . 3 7 1 .9

Abstract Visual Reasoning

98 . 9 \ 09 . 6 87. 7 7 1 .6

ShortQuantitative Term Reasoning Memory

\ 02 . 1 1 1 7.2 86. 3 73.9

102 . 4 1 1 6. 1 84 . 8 67.5

Composite

Diff '

r

1 13.3 \ 02 . 4 1 16.3 84 . 8

+ 1 .7 +2.8 + 1 .4 +3.0

.78 .83 . 69

66 . 2

+ .8

, In the difference (diff.) column, the plus ( + ) sign indicates that WISC-R Full Scale IQs were higher than Fourth Edition composite scores. Carvajal & Weyand ( 1 986). C Thorndike et al. ( 1 986b). b

. 87 . 66

1 27

FACTOR ANAL YSIS

Perfo mulIlce Scale WISC-R

WAIS-R

Diff a

94. 87 88.88 99.08 60. 66 84. 70 1 04 . 30 7 9 . 70 7 2 . 60

97 . 2 3 8 3 . 80 1 00 . 4 1 65 . 54 86.47 1 02 . 80 8 3 . 30 7 2 . 30

+2.38 - 5 .08 + 1 .33 +4.88 + 1 . 27 - 1 .50 + 3.6 - .30

d

c

f

Full Scale r

. 74 . 75 . 82 . 82 .76 . 85 .75

Diff a

WISC-R

WAIS-R

82 . 3 2

82. 7 5

+ .43

5 5 . 83 8 1 . 13 1 0 1 .90 7 5 . 90 69.60

66. 93 82. 70 1 0 1 .00 8 1 .20 73.00

+ 1 1 . 10 + 1 . 57 - . 90

+5.2 + 3 .4

r

. 83 . 86 .88 .88 . 70

Subjects were in a residential school for the mentally retarded. Subjects were in special education. Nonreferral sample.

FACTOR ANALYSIS

A factor analysis of the standardization group indicated that three factors could efficiently describe the WISC-R (see Table 6-7) (Kaufman, 1975a) . These factors are la­ beled Verbal Comprehension , Perceptual Organization, and Freedom from Distractibility. The term Verbal Com­ prehension describes the hypothesized ability underlying the factor for both item content (verbal) and mental pro­ cess (comprehension). This factor appears to measure a variable common to the Verbal Scale subtests. Vocabulary, Information, Comprehension, and Similarities have the highest loadings on the Verbal Comprehension factor, fol­ lowed by Arithmetic, which has a moderate loading . 1wo Performance Scale subtests - Picture Completion and Pic­ ture Arrangement - also have moderate loadings on the Verbal Comprehension factor, suggesting that these two subtests may require verbal mediation to a greater degree than do the other Performance Scale subtests. The term Perceptual Organization describes the hypoth­ esized ability underlying the factor for both item content (perceptual) and mental process (organization) . This fac­ tor appears to measure a variable common to the Perfor­ mance Scale subtests. Block Design, Object Assembly, and Picture Completion have high loadings on the Percep­ tual Organization factor; Mazes and Picture Arrangement have moderate loadings . Although researchers had some difficulty naming the third factor, Freedom from Distractibility appears to be most appropriate . The name focuses on the ability to concentrate or remain attentive . The Arithmetic and Digit

Span subtests have high loadings on the Freedom from Distractibility factor; Information and Coding B have moderate loadings . (Coding A has only a minimal loading on this factor. ) The factor analytic results give strong empirical support to interpretation of the Verbal and Performance IQs as separately functioning entities in the WISC-R . The factor structure of the WISC-R closely agrees with the actual organization of the subtests. As for the measurement of g , the WISC-R subtests cluster into three groups (see Table 6-8): those with high loadings (Vocabulary, Information, Similarities, Block Design, and Comprehension), moder­ ate loadings (Arithmetic, Object Assembly, Picture Com­ pletion, and Picture Arrangement) , and low loadings (Digit Span, Mazes, and Coding) . Other Factor Analytic Studies

Other factor analytic studies of the WISC-R with black children , Mexican-American childre n , lower-middle­ class children, mentally retarded children, adolescent psy­ chiatric patients, epileptic children, children referred for academic difficulties , learning disabled children, slow learners , delinquents , and emotionally disturbed children indicate that the factors found in these groups are generally similar to those found in the standardization group (Blaha & Vance , 1979; Carlson, Reynolds, & Gutkin, 1983 ; Dean , 1980 ; DeHorn & Klinge, 197 8 ; Groff & Hubble, 1982 ; Hodges, 1982; Hubble & Groff, 1981a; Johnston & Bolen , 1984; McMahon & Kunce , 1981 ; Naglieri , 1981c; Reynolds & Gutki n , 1980b; Richards, Fowler, Berent, & Boli, 1980 ;

CHAPTER 6

1 28

WECHSLER INTELLIGENCE SCALE FOR CH ILDREN-REVISED (WISC-R): DESCRIPT ION

Table 6-6 Relationship of WISC-R IQs to Sex, Race, Occupation of Head of Household. Urban-Rural Residence, and Geographic Residence

Verbal lQ Demographic variables

Performance IQ

Full Scale IQ

N

M

SD

M

SD

M

SD

1 , 1 00 1 , 1 00

1 0 \ .2 98 . 8

1 5 .4 14.3

1 00 . 4 99 . 8

15. 1 14.9

1 00 . 9 99. 1

15.4 14.6

Sex

Boys Girls

Race

White boys Black boys White girls Black girls

945 1 43 925 1 62

\03 . 3 87.6 1 00 . 7 88.0

14.7 1 2 .4 13.5 13.7

1 02 . 2 87.9 \ 02 . 1 86.5

14.3 13.7 1 3 .9 1 3 .0

\ 03 . 1 86.7 1 0 1 .4 86.2

14.5 12.5 13.5 12.9

Parental occupation group

Whites 1 . Professional and technical 2 . Managerial , clerical , sales 3 . Skilled 4 . Semi-skilled 5 . Unskilled

308 538 446 499 79

1 09 . 7 1 04 . 2 1 00 . 5 97.9 92 . 3

12.8 14.0 13.2 13.5 13.3

1 07 . 1 1 04 . 0 101.3 99 . 2 93 . 4

1 3.7 14.6 13. 1 1 3.4 1 3.8

1 09 . 4 1 04 . 4 1 00 . 9 98.3 92. 1

12.9 14. 1 12.9 13.2 13.3

Blacks 1 . Professional and technical 2 . Managerial, clerical , sales 3 . Skilled 4 . Semi-skilled 5 . Unskilled

20 51 47 1 29 58

92 . 0 9 1 .7 89.6 87.4 82.6

12.7 1 1 .5 10.6 1 3 .6 13.5

90. 8 90. 9 87.4 87.0 82 . 8

1 0. 4 12.5 I \. I 14.2 13.3

90. 7 90. 5 87.5 86.0 81 .5

1 1 .4 10.2 13.6 12.5

Total Group I . Professional and technical 2 . Managerial , clerical , sales 3 . Skilled 4 . Semi-skilled 5 . Unskilled

329 594 495 639 143

108.6 1 03 . 1 99. 5 95 . 7 87.8

13.5 14.2 13.3 14.3 14.3

1 06 . 1 1 02 . 9 1 00 . 1 96.9 88.8

14. 1 14.8 13.6 14.6 1 4. 6

108.2 103.2 99 . 6 95.9 87.3

13.6 14.4 13.3 14.4 14. 1

1 ,557 643

1 00 . 6 98.5

14.7 15.4

1 00 . 5 99 . 0

15.1 14.8

1 00 . 6 98.6

14.9 15.2

478 64 1 696 385

1 02 . 8 1 00 . 0 96. 9 101 .8

14.6 14.0 15.7 14.3

101 .3 1 0 1 .0 96. 7 1 03 . 1

14.4 14.6 15.9 13.6

1 02 . 3 1 00 . 5 96 . 6 1 02 . 6

14.6 14.2 15.9 13.8

Urban-rural residence

Urban Rural

Geographic residence

Northeast Northcentral South West

II. I

Source: Adapted from Kaufman and Doppelt ( 1 976) .

Schooler, Beebe, & Koepke, 1978 ; Shiek & Miller, 1978; Sutter & Bishop, 1986a; Snow, Cohen, & Holliman, 1985 ; Swerdlik & Schweitzer, 1978; Vance & Wallbrown, 1977; Wallbrown, Blaha, Wallbrown, & Engin, 1975) . Subtest Specificity

Subtest specificity refers to the proportion of a subtest's variance that is both reliable (that is, not due to errors of

measurement) and distinctive to the subtest (see Chapter 2). Although the subtests overlap in their measurement properties (that is, the majority of the reliable variance for most subtests is common factor variance) , many of the subtests also have a relatively high degree of subtest specificity that allows for interpretations of specific subtest functions (see Table 6-9) . Although the relatively high degree of subtest specificity provides a fi rm ground for profile analysis of scaled scores , constraints still must be placed

1 29

FACTOR ANALYSIS

Table 6-7 Factor Loadings of WISC-R Subtests for I I Age Groups (Va\rirnax Rotation)

Subtest

6!1a

7!1

8!1

35

49

65 63 31

57 64 29

74

9!1

lO!1

J l !1

1 3 !1

1 4 !1

1 5 !1

16!1

Mdn.

65 67 33

77 74 53

60

72 65 40

57 62 32

64

82 70

62 37 75 69

27

27

36 42 39

1 2 !1

Factor A - Verbal Co'mprehension Information Similarities Arithmetic Vocabulary Comprehension Digit Span Picture Completion Picture Arrangement Block Design Object Assembly Coding Mazes

44

32 58 49

60 38 69

67 62 23

63 69 41 73 63 13

63 65 37 2

34 26 22 21

18

55 23

39

29

22

27

40

38 28 19

31 30 24 12 08

44

44

27 20 23 18

30 34 10 06

27 26 33 14 08

06 21

67

04

64

17

16 20

67 16 35 34 33 22 17 15

20 23 -06

78

71

63 20

66 16

63 64 37 72 64 18

31 33 23 20 13 08

35 25 20 14

39 25

35 33

22 22

15 13

31 12

27 21 15 12

34 38 19

22 34 19

24 31 07

18 35 12

32 34 16 29

25 34 20 24

32 13

30 12

60 29

57 49 72 64

54 41 76 70

57 41 66 65

24

20 48

47

Factor B - Perceptual Organization Information Similarities Arithmetic Vocabulary Comprehension Digit Span

18 23 21 17 28 18

33 30 20

Picture Completion Picture Arrangement Block Design Object Assembly Coding Mazes

27 38

52 49

57 12

66 66 23

48 51 58 69 24

60

60

47

Information Similarities Arithmetic Vocabulary Comprehension Digit Span

55 37 59

51 27 59

31 28 51

36 33 54

31 28 50

16

48 38 63 53

08 56

28 52

32 32 48 25 12 57

Picture Completion Picture Arrangement Block Design Object Assembly Coding Mazes

13 27

06 28 27 25 25

33 09

18 12

01 19

32 19

36 16

46 15

42 24

35 12 45 22

44

22 28 20

20 24 19 24 30 14

25

35 35 27

55 41 66 58 17

60 41 73 70

62 61 54 61

57 41 70 68

13

22

47

42

16 47

70 64 12

47

32

48

24

47 30 58

23

22 33 19

26 37 12

33 26 11

63 49 63 65 20 49

19

38 44 38 31 33 07

38 24

36 27 14

26 23 19 21 04

20

Factor C - Freedom from Distractibility

25 18 20 21

23

41 23

28 26 61

33 30 54

44

23 45 29

28 59

22 59

02 25 28 12 53 12

20 12 50 22 42 26

05 07 27 05 45

44

30

33 19 39 11 16 18 12 15

04

28 48 39 15 56 08 10 33 11 40 24

46 31 64 45 24 62

41 28 58

12 09

11 12 28 12 42

27 09 31 18

33 24 56

22

Note. Decimal points omitted. a

This row indicates age leve l .

Source: From A . S . Kaufman, "Factor Analysis o f the WISC-R a t I I Age Levels between 6 '/2 and 1 6 '/1 Years," Journal of Consulting and Clinical Psychology, 1 975, 43, pp. 1 38- 1 40 . Copyright 1 975 by the American Psychological Association . Reprinted by permission.

CHAPTER 6

1 30

WECHSLER INTELLIGENCE SCALE FOR CHILDREN -REVISED (WISC-R): DESCRIPTION

Table 6·8 WISC·R Subtests as Measur-es of g

Fair measure of g

Good measure of g'

Subtest

Median loading ofg

Vocabulary Information Similarities Block Design Comprehension

. 80 . 76 . 76 . 73 .72

Proportion of variance attributed to g ( % ) 64

58 58 53 52

Poor measure of g

Subtest

Median loading ofg

Proportion of variance attributed to g (%)

Arithmetic Object Assembly Picture Completion Picture Arrangement

. 65 .62 .61 . 60

42 38 37 36

Subtest

Median loading ofg

Proportion of variance attributed to g (%)

Digit Span Mazes Coding

.49 .45 .41

24 20 17

Note. The square o f the median coefficients provides the proportion of each subtest's variance that may b e attributed to g .

For children ages 6 t o 7 years, there are only three good verbal measures of g (Information, Arithmetic, and Vocabulary) and two good nonverbal measures of g (Block Design and Picture Arrangement). Source: Adapted from Kaufman ( l 975a).

a

on interpreting subtest functions . These constraints in­ clude determining which subtest scaled scores are signifi­ cantly different from one another and analyzing all rele­ vant subtests before drawing any conclusions about unusual ability or weakness . (See Chapter 8 for a discus­ sion of profile analysis.) For example, low scores on Digit Span, Arithmetic, and Coding and average or high scores on the other subtests may indicate that the child is highly distractible. A low score on Digit Span but not on Arith­ metic and Coding, however, may suggest difficulty with auditory memory, or anxiety. Specific interpretations can be given to all subtests, with the exception of (a) Sim­ ilarities for children above 8 '/2 years of age and (b) Object Assembly for all children; patterns involving these two

subtests probably are attributable to measurement error (Kaufman, 1975a) . Factor Scores

Factor scores also can be obtained from the WISC-R, permitting the identification of meaningful psychological dimensions (Kaufman , 1975a) . The Verbal Comprehen­ sion factor score measures verbal knowledge and under­ standing obtained informally and through formal educa­ tion. It reflects the application of verbal skills to new situations . The Perceptual Organization factor, a nonver­ bal score, reflects the ability to interpret and organize visually perceived material within a time limit. The Free-

Table 6·9 Specificity for- W ISC·R Subtests

Adequate specificity

Ample specificiry

Inadequate specificity

Subtest

Ages

Subtest

Ages

Subtest

Ages

Information Similarities Arithmetic Digit Span Picture Completion Picture Arrangement Block Design Coding Mazes

all ages 6 112 to 8 '12 all ages all ages 6 '/2 to 8 '12 all ages all ages all ages all ages

Vocabulary Comprehension Picture Completion

all ages all ages 9 112 to 1 6 '/2

Similarities Object Assembly

9 '/2 to 1 6 '/2

Source: Adapted from Kaufman ( 1 979c).

all ages

131

ADMINISTERING THE WISC-R

dom from Distractibility factor score measures the

8 � DDDDDD 0 � U

Month

Age

�

0>

Year Dote Tested

cludini.

.errievol ,y,rem, wilhoul pe,mi"ion i n A.

.... r i r i n g

bul nol limiled 10. Oholocooyin9. oud,o.i,uol recording ond

O l

ele( 1 ro n i(

4

lesh, if necenary.

o. mechonicol. in.

Iron,mlnion. ond oorlroyol or duolico1;on in ony inlo.mo,ion " oroge ond

.

I,om Ihe oubl i'her . See (ololog 101 !urlher inlormOI;on.

The i",ychologicol (or O ol;on

New YOI�. N Y. 1 0 0 1 7 .

7"·I03AS

Cover page of WISC-R record booklet. Copyright © 197 1 , 1974 by The Psychological Corporation, San Antonio,

Rights Reserved.

TX. All

1 34

CHAPTER 6

WECHSLER INTELLIGENCE SCALE FOR CHILDREN-REVISED (WISC-R): DESCRIPTION

child's stealing . Later, when questioned on the Com­ prehension subtest about finding a wallet in the street, the child looked distressed, but the examiner failed to recog­ nize the distress . Comment. Be sensitive to nonverbal cues as well as verbal ones. An alert exarruner would have said something like "Now, this question has nothing to do with our previous discussion. This is one of the questions I ask everyone. " Example 2 . The examiner wanted t o say something sup­ portive to the examinee after the Digit Span subtest. Exam­ iner: "Are you aware that you have a very good memory?" (Child scored in the below average range on the Digit Span subtest. ) Child: "No, I have a lousy one. I forget things all the time ." Comment. It is important that reinforcements be congruent with the examinee's performance and be given at appropriate times . Example 3 . The examiner watched a lO-year-old assem­ ble the horse on the Object Assembly subtest, leaving one piece out. When the child said "finished, " the exarruner pointed to the extra piece. The child quickly corrected the error and was given full credit. Comment. Nonverbal cues may inappropriately help examinees solve problems. Example 4. Only when the child's arrangement on the Picture Arrangement subtest was correct did the examiner ask "finished?" If the child was still checking an incorrect sequence, the examiner was silent. The child soon caught on. Comment. Do not give verbal cues that may alert examinees to how well they are doing. Example 5. The examiner, noting that a child had mis­ placed only one block in a complicated design on the Block Design subtest, said, "Be sure to check your answer. " Comment. It is inappropriate to add directions that are not in the test manual.

These examples illustrate just a few of the ways in which an exarruner may inappropriately influence test adminis­ tration. It behooves each of you to learn about possible sources of error - both overt and covert - in your adminis­ tration of the WISC-R and other tests, and take appropriate steps to prevent these errors from occurring. Subtest Sequence

The WISC-R manual states that, although it should be convenient to administer the subtests in the order indicated in the manual , it is permissible to change the order when

the needs of a particular child must be met, or when the examiner has a personal preference . You are strongly urged to administer the subtests in the order specified in the manual unless there is some compelling reason to use another order. (For instance, children who are extremely bored or frustrated with some subtests may be more moti­ vated if they are given a different subtest or a subtest of their choice . ) Because the order in the manual was used in obtaining the standardization data, any other order con­ stitutes a departure from standard procedures , and there is evidence that such departures do affect children's scores (Exner, 1966; Morri s , Marti n , Johnson , B i rc h , & Thompson, 1978 ; Sattler, 1969) . Starting Rules

In some situations, an examiner may have doubts about whether the items at the entry point were passed and thus may decide to administer earlier items in the series. (The entry-point items are those that the child must pass in order for the examiner to administer more difficult items . ) Wechsler (1974, p . 59) provided a specific rule for scoring failures on items below the entry point : "If subsequent scoring of the test reveals that some of the earlier items were administered unnecessarily, the child should be given full credit for these items - even if he [or she} earned partial or no credit" (italics added) . Therefore, when an item below the entry point is failed or partially credited , give full credit for the item if further checking indicates that, in fact, the child correctly answered the items at the entry point. This rule applies only to normal children 8 years old and above (because they have entry points above the first item on certain subtests) . Here is an example . After administering words 4 and 5 on the Vocabulary subtest to a normal 8-year-old child, you are uncertain of the scoring of the responses that were given to these two entry items. You therefore decide to administer words 3 and 2 , and then word I because word 3 was definitely failed . The child passes words 1 and 2, and the subtest is continued with word 6. After the examina­ tion, a check of the responses indicates that the child should receive full credit for words 4 and 5 . The rule requires that the child receive full credit for defining word 3 (as well as words 1 and 2), even though the definition of word 3 was incorrect, because word 3 occurs below the entry words that were defined correctly. This rule favors the child, by ensuring that the child is not penalized for your decision to administer the earlier word that the child subsequently failed . The rule is an attempt to maintain standardized scoring procedures.

1 35

ADMINISTERING THE WISC-R

Discontinuance Rules

The discontinuance scoring rule applies to situations in which the examiner has some doubts as to whether the items at the discontinuance point were failed and subse­ quently decides to administer additional items in the series (Wechsler, 1974 , p. 59) . (The discontinuance-point items are those that the child fails in a consecutive series, indicat­ i ng , according to the manual , that the subtest should be discontinued . ) The rule is as follows: If subsequent scor­ ing of the items indicates that the additional items were administered unnecessarily, the child should not be given credit for items passed after the discontinuance point. Here is an example . After administering the first 15 words of the Vocabulary subtest, you are uncertain of the _ scoring of some of the responses that the child gave to words 11 through 15 . You therefore decide to administer additional words , and the child definitely passes words 16 and 17 but fails words 18 through 22. The subtest is, therefore, discontinued after word 22. After the examina­ tion, checking of the responses indicates that the child should not receive credit for definitions of words 11 through 15 . The rule requires that the child not receive credit for the definitions of words 16 and 17 , even though these defini­ tions were correct, because these words occur after the test should have been discontinued. In contrast to the starting rule, this rule does not favor the child. The rule prevents the child from receiving additional points because of your uncertainty in scoring items during the regular administra­ tion of the subtests . This rule constitutes another attempt to maintain standardized scoring procedures. Repetition of Items

The WISC-R manual encourages examiners to use judg­ ment in deciding whether an "I don't know" response reflects lack of knowledge or lack of desire to respond . If you decide that it reflects the latter, repeat the question or ask it again at some later point ..,- especially if the "I don't know" response is given to an easy question. Better yet, the first time a child says "I don't know," say something like "I want you to try your hardest on each ofthese. Try your best to answer each question . " Give credit if the child correctly answers the question . Questions should not be repeated on the Digit Span subtest . Use of Probing Questions and Queries

The WISC-R manual encourages flexibility in administer­ ing test items. Acknowledge , for example, negativistic or

mistrustful responses and continue to probe to determine whether the child knows an answer. For example , if the child responds to the Comprehension question "Why are criminals locked up?" with "Society labels people crimi­ nals who shouldn't be so labeled," you can say "Well, try to give some answers that other people think are reasonable . " You will have to be alert to recognize responses that indi­ cate a need for these kinds of probes. Probes are also required on verbal responses that are incomplete, indefi­ nite, or vague . Queries should be used when you are unsure of how to score the response; they should not be used to elicit a higher quality response. Spoiled Responses

An explicit scoring rule on the WlSC-R is that spoiled responses are scored O. A spoiled response is one that initially was partially right, but was spoiled by the child's incorrect elaboration on his or her initial response. For example, the response "Goes ticktock" (for the Vocabulary word clock) would be spoiled by the elaboration "It's the engine on a motorcycle," which reveals the child's miscon­ ception about the word clock. Modifying Standard Procedures

Deviations from procedures for administering the subtests are likely to result in scores that differ from those obtained with standard administrative procedures . Studies (Herrell & Golland, 1969; Post, 1970; Sattler, 1969; Schwebel & Bernstein, 1970) indicate that children obtained higher scores on some of the subtests when they were encouraged to talk about their problem-solving procedures, think about their answers before responding, explain their pic­ ture arrangements, or solve problems after receiving a series of cues. Such modifications, i f used, should be employed only after the standard administration . They may be helpful in clinical assessment of the child's poten­ tial for learning. Scoring WISC-R Responses

Arriving at a score for some WISC-R subtests is by no means a simple matter. Judgment is important, especjally for handling ambiguous responses that occur on the Vocab­ ulary, Comprehension, and Similarities subtests. To be­ come a skilled examiner, you need to study carefully the scoring criteria, scoring guidelines, and scoring examples in the WlSC-R manual . The supplementary WISC-R scor-

CHAPTER 6

1 36

WECHSLER I NTELLIGENCE SCALE FOR CHILDREN-REVISED (WISC-R): DESCRIPTIO N

ing manual can also assist you in scoring these three subtests (Massey, Sattler, & Andres, 197 8 ) . Even authorities who have published scoring manuals do not always agree on the scoring of responses. Eleven scoring discrepancies (see Table 6-10) were found between the WISC-R manual and two other scoring guides for Vocabulary and Similarities items (Sattler, Squire, & An­ dres, 1977 ) . Fortunately, however, such differences appear to be relatively few in number. Some examiners are more lenient than others in giving credit, and at times examiners may not consistently adhere to their own relative standards. For example, examiners may be strict on some occasions and lenient on others. Dramatic differences in the scoring standards of examiners have been illustrated in a number of reports . In one study, 99 school psychologists gave IQs ranging from 63 to 117 to the same WISC protocol (Massey, 1964) . In other studies, graduate-student examiners (Miller et al . , 1970) and mem­ bers of the American Psychological Association (Miller & Chansky, 1972) differed by as much as 17 points in scoring the same test protocol .

Extrapolated IQs

When the sum of scaled scores obtained by a child i s beyond those shown in the WISC-R manual, Table C - 6 i n Appendix C can b e used t o obtain extrapolated I Q s . This table extends the IQ table down to the minimum possible sum of scaled scores and up to the maximum possible sum

of scaled scores. I n using Table C-6, remember that Wechsler recommends that a child's Full Scale IQ not be determined unless he or she obtains raw scores greater than 0 on at least three Verbal and three Performance Scale subtests. The values presented in Table C-6, which were obtained by using linear regression methods, are helpful in obtain­ ing a specific intelligence quotient . These values, however, are outside the range of the children who were tested in the standardization sample . Therefore the possibility of error in extrapolations such as these may be disconcertingly large. Nevertheless, these values may be of some use in reporting results for children who fall in the very low or very high IQ ranges . If possible, use another test (such as the Stanford-Binet Intelligence Scale: Fourth Edition) to evaluate children whose IQs fall outside the range covered by the WISC-R .

S HORT FORMS OF T H E W ISC-R

Short forms of the WISC-R and other Wechsler scales are occasionally used as screening devices . (Their advantages and disadvantages will be discussed at the end of this section . ) The selection of a short form is usually based on such criteria as acceptable reliability and validity, the power of the short form to answer the referral question and provide clinically useful information , the examinee's phys­ ical capabilities, and the amount of time available for administering the test.

Table 6· 1 0 Eleven Scoring Discrepancies Between WISC·R Manual and Two Scoring Guides

Subresr

Item and response

Vocabulary

8. donkey - "a mule" 1 2 . diamond - "can cut glass" (or "a rock or an object that can cut glass") 1 2 . diamond - "mineral" 1 3 . gamble - "a game" 2 1 . stanza - "a phrase" 2 2 . seclude - "opposite of include" 25. belfry - "a tower"

Similarities

7 . cat - mouse - "both pets" 1 3 . mountain - lake - "both in country"

1 5 . first - last - "places" 1 7 . salt - water - "chemicals"

WISC-R score

lastak and lasrak score

Scouish Council score

2 I 1 (Q)

o (Q) o o 1

o o (Q) o

2 2

2

I 2 2

Source: Reprinted with permission of the publisher and authors from J . M. Sattler, L. Squire, and J . Andres, "Scoring Discrepancies between the WISC-R Manual and 'TWo Scoring Guides , " Journal of Clinical Psychology, 33, p. 1059 . Copyright 1977, Clinical Psychology Publishing Co . , Inc.

1 37

SHORT FORMS OF THE WISC-R

The validity of short forms is usually eva I uatcd by c arre· lating the short form IQ with the Full Scale IQ, evaluating mean differences between the two IQs, and determining the extent of agreement in the intell igence classifications provided by the two IQs. Silverstein (l985a) argued , how­ ever, that these three criteria are not useful . First, it is virtually certain that there will be high correlations be­ tween short-form and Full Scale IQs . Second, with suffi­ ciently large samples , a significant difference between long and short IQs is likely to occur, making this criterion nearly meaningless. Third, it is virtually certain that the short-form and Full Scale IQs will yield different classifi­ cations. [Goh (1978), for example , found that short-form WISC-R IQs misclassified 45 percent of a group of 142 children . ] Silverstein suggested that other considerations be used to determine the appropriateness of a short form. If a specific classification must be obtained for the clinical or psychoeducational purpose, then short-form IQs will be inappropriate . Selecting the Short Form

The 10 best short-form combinations of two, three, four, and five WISC-R subtests, arrived at by using the stan­ dardization data and a formula that takes i nto account subtest unreliabil ity, are shown in Table C-1O of Appendix C. Because the validities of the various short forms are high , clinical considerations should influence short-form selection . For example, when using a tetrad, you may want to select a combination consisting of two Verbal a nd t w o Performance Scale subtests in order to obtain some repre­ sentation of both verbal and performance skil ls in the short form . An examinee's physical capabilities may also guide you in selecting a short form. Examinees with marked visual impairment or severe motor dysfunction of the upper ex­ tremities will have difficulty with Performance Scale tasks. In such cases , the Verbal Scale can serve as a useful short form. For hearing-impaired examinees, the Performance Scale alone is a useful short form. These short forms should be administered using the child's preferred mode of communication and should be supplemented by other tests designed to accommodate the special physical abilities of these children. Converting Short-Form Scores into Deviation Quotients

After the specific combination of subtests has been admin­ istered, it is necessary to convert the short-form scores to a Full Scale IQ estimate . Simple prorating and regression

procedures are not applicable because they do not deal adequately with the problem of subtest reliability (Tellegen & Briggs , 1967). The more acceptable procedure is to tran s form the short- fo r m scores i nto the fam i l i a r Wechsler-type Deviation Quotient, which has a mean of 100 and a standard deviation of 15. Exhibit 6-3 shows the procedure for converting the short-form scores into a Deviation Quotient. Although this approach does not elim­ inate the many problems associated with short forms , it does appear to provide fairly reliable IQs .

Yudin's Abbreviated Procedure

In the Yudin (1966) WISC short-form procedure, which also applies to the WISC-R, every other item on most subtests is administered . The specific procedures , modi­ fied by Silverstein (l968a), are shown in Table C-U of Appendix C . After the test has been administered, the scaled scores and IQs are obtained from the manual in the usual way. The Yudin procedure differs from other short­ form procedures in that all of the subtests are used. Its advantages are that a representative sample of items is administered, profile analysis can be applied, and approxi­ mately 56 percent of the items are used. Although the Yudin procedure has satisfactory reliabil­ ity (Reid , Moore, & Alexander, 1968 ; Yudin, 1966) , short­ comings have been noted. These include a moderate loss of validity and reliability, less reliable profile data, and IQs that differ from those obtained on the Full Scale (Dean, 1977f; Erikson, 1967 ; Finch , Kendall , Spirito , Entin, Montgome ry, & Schwartz, 1979 ; Gayton, Wi lson , & Bernstein , 1970; Goh , 197 8 ; Rasbury, Falgout, & Perry, 1978; Satz, Van de Riet, & Mogel , 1967 ; Tel legen & Briggs , 1967) . The assets and liabilities of the Yudin ab­ breviated procedure should be considered carefully before it is used . Hobby ( 1980) described a WISC-R procedure in which only odd items are administered on most subtests. It is similar to the Yudin procedure , but has more specific basal and ceil ing procedures and correction factors.

Vocabulary and Block Design Short Form

A popular screening short form consists of Vocabulary and Block Design . These two subtests have excellent reliabil­ ity, correlate highly with the Full Scale over a wide age range, and are good measures of g. Table C-37 in Appen­ dix C can be used to convert the sum of scaled scores on these two subtests directly to an estimate of the Full Scale IQ.

CHAPTER 6

1 38 •

WECHSLER INTELLIGENCE SCALE FOR CH ILDREN -REVISED (WISC-R): DESCRIPTION

Exhibit 6-3 Obtaining Deviation Quotients for Short Forms

The following formula is used to compute the Deviation Quotient for a short form:

l

where

Deviation Quotient = (\5/Sc)(Xc

-

Mc) +

100

I

I S, . n + 2f.rjk (standard deviation of composite score) Xc composite score (sum of subtest scaled scores in the short form) Me normative mean, which is equal to IOn S5 subtest standard deviation, which is equal to 3 n number of component subtests f.rjk = sum of the correlations between component subtests. Sc

=

=

=

=

=

This equation considers the number of subtests in the short form, the correlations between the subtests, and the total scaled-score points obtained on the short form . A more straightforward computational formula for obtain­ ing the Deviation Quotient is as follows:

I

Deviation Quotient

where

a b

=

15/Sc

=

100

-

=

(composite score x a) + b

n(\50)/Sc'

Table C-36 in Appendix C can be used in obtaining the appropriate a and b constants. In using Table C-36, first select the heading corresponding to the number of subtests in the short form. The first column under each heading is f.rjk ' This term represents the sum of the correlations between the sub­ tests making up the composite score. To obtain f.rjk' use the WISC-R correlation table of the group closest in age to the examinee (Table 14 on pages 36 through 46 of the WISC-R manual ) . With two subtests in the short form, only one corre­ lation is needed . With three sub tests in the short form, three correlations are summed (\ with 2 , I with 3 , and 2 with 3 ) . With four subtests in the short form, s i x correlations are

Arithmetic, Vocabulary, Picture Arrangement, and Block Design Short Form Another useful screening short form is Arithmetic , Vocab­ ulary, Picture Arrangement, and Block Design (Kaufman , 1976b) . It takes longer to administer than Vocabulary and Block Design, but provides more clinical and diagnostic information .

summed (I with 2 , I wtih 3 , I with 4, 2 with 3 , 2 with 4 , and 3 with 4) . With five subtests in the short form, 10 correlations are summed (I with 2 , I with 3, I with 4, I with 5, 2 with 3, 2 with 4, 2 with 5, 3 with 4, 3 with 5 , and 4 with 5 ) . After f.rjk is calculated, the values for the two constants are obtained under the appropriate heading. The procedure used to obtain the Deviation Quotient can be summarized as follows: I.

Sum the scaled scores of the subtests in the short form to obtain the composite score. 2 . Sum the correlations between the subtests to obtain f.rjk . 3 . Find the appropriate a and b constants in Table C-36 in Appendix C after f.rjk has been obtained. 4. Compute the Deviation Quotient by using the composite score and the a and b constants. Example: A three-subtest short form composed of the Arithmetic, Vocabulary, and Block Design subtests is admin­ istered to a 6-year-old child . The child obtains scaled scores of 7, 12, and 13 on the three subtests. The four steps are as follows: 1.

The three scaled scores are summed to yield a composite score of 3 2 . 2 . The correlations between the three subtests are obtained from Table 14 (page 36) of the WISC-R manual (Arith­ metic and Vocabulary, .52; Arithmetic and Block Design, .47; Vocabulary and Block Design, . 43 ) . These are sum­ med to yield 1 .42 (f.rjk) . 3 . The appropriate row in Table C-36 in Appendix C is the fourth one under the heading "3 Subtests ." The values for the constants a and b are 2 . 1 and 37, respectively. 4 . The formula Deviation Quotient

=

(composite score x a) + b

is used to obtain a Deviation Quotient of 104 [(32 x 2 . 1) + 37) .

Other Useful Short Forms Other short forms discussed in the literature are Sim­ ilarities and Vocabulary (Fell & Fell, 1982 ) ; Similarities and Object Assembly (Fell & Fell , 1982) ; Similarities, Vocabulary, and Block Design (Karnes & Brown, 1981) ; Similarities, Vocabulary, Block Design, and Object As­ sembly (Karnes & Brown, 1981) ; Similarities , Vocabulary,

CHOOSING BETWEEN THE WISC-R AND THE WPPSI A ND B ::TWE:EN THE WISC-R A N D THE WAIS-R

Block Design, and Picture Completion (Clarizio & Vew" 1984); Similarities, Object Assembly, and Vocabulary (Dirks , Wessels, Quarfoth , & Quenon , 1980) ; and Infor­ mation, Comprehension , Block Design , Picture Arrange­ ment and Coding (Kennedy & Elder, 1982). Comment on Short Forms of the WISC-R (and Other Tests as Well)

Short forms save time and are useful screening devices. but they have many disadvantages. First, short-form IQs may be less stable than those obtained with the standard form. Second, information about cognitive patterning is lost. Third, opportunity to observe the examinee's problem­ solving methods is lost. Fourth, some short forms, such as the Verbal Scale alone , do not allow for the assessment of performance abilities . Fifth, when there is intersubtest variability, the savings in time may be expensive in terms of lost information . Finally, reliability of the obtained IQ is reduced when subtests are eliminated. Those interested in using short forms must weigh the time saved against the validity lost. In addition, it is impor­ tant to consider what kind of decision will be made on the basis of the short-form scores. The most efficient testing strategy for a particular s ituation will depend, in part, on the goal of the evaluation - whether it is for a general assessment of intell igence , classification , selection, or screening . Even when all of the subtests are administered , the IQ obtained on the WISC-R (and all other intelligence tests) is but an estimate of the different kinds of abil ities possessed by a child. When a small number of subtests is used, the estimate may be far less adequate than that provided by the Full Scale. Educational and clinical situations call for more, rather than less, extensive cognitive evaluation. You are encouraged 10 administer the Full Scale, unless there is some compelling reason to administer a short form. In­ cluded among these reasons would be situations in which the child was ready to quit testing or the physical ca­ pabilities of the examinee made some of the subtests inap­ propriate . The Full Scale should be administered so as 10 maximize diagnostic information and minimize placement errors. Short forms are not recommended for any place­ ment, educational, or clinical decision-making purpose.

C HOOSING BETWE E N T H E WISC-R AND TH E WPPSI A N D B ETWE E N THE W ISC-R AND T H E WAIS-R

The WISC-R overlaps with the WPPSI for the age period 6-0-0 to 6-7-15 and with the WAIS-R for the age period

1 39

16-0-0 to 16-\1-30. The overlap in ages between the WISC-R and the WPPSI and between the WISC-R and the WAIS-R is especially helpful in retest situations. A child first administered the WISC-R at age 6-0-0 can be retested with the WPPSI at any time during the next seven months . Similarly, a child tested with the WAIS-R at age 16-0-0 can be retested with the WISC-R up until about his or her seventeenth birthday. For these overlapping age periods, the WISC-R manual indicates that "the examiner should choose the Scale that is most appropriate for his [her] purposes" (p. 53). Unfortunately, it is difficult to see how this statement can guide your choice. In order to select a scale, one needs information about the advantages and disadvantages of each scale. The choice of a test should depend on the validity of the inferences that can be made from scores on it . To this end , it would be helpful to consult validity studies that compare the WISC-R w ith the WPPSI and the WISC-R with the WAIS-R in their overlapping age ranges, using samples of both normal and exceptional childr�n . Rather than the personal preferences of the exammer, the choice ought to depend on which test yields the small­ est standard errors of measurement for scores at the levels obtained. Because standard errors of measurement are provided in the Wechsler manuals for age but not for ability level, the needed information is not available . In the case of the WISC-R versus the WPPSI, the standard errors of measurement for the Verbal , Performance, and Full Scale IQs are smaller for the WPPSI than for the WISC-R at age 6 1f2 years. In the case of the WISC-R versus the WAIS-R, the standard errors of measurement for the Verbal and Full Scale IQs (but not Performance Scale IQs) are slightly smaller for the WAIS-R than for the WISC-R at 16 112 years. The child's estimated level of intelligence should also be considered in selecting a test. A 6 1/2 -year-old child who obtains an IQ that is below the normal level will be admin­ istered a greater number of WPPSI items than WISC-R items. (For example, one needs 9 correct WPPSI Informa­ tion items but only 3 correct WISC-R Information items to obtain a scaled score of 5 . ) Similarly, a 16-8-year-old child needs many more successes on the WISC-R than on the WAIS-R to obtain the same scaled score. (For example, one needs J3 correct WISC-R Information items but only 4 correct WAIS-R Information items to obtain an age­ corrected scaled score of 5 . ) Consequently, for children with below-normal ability, a more thorough sampling of ability can be obtained from the WPPSI than from the WISC-R and from the WISC-R than from the WAIS-R in their overlapping age ranges. For normal and gifted chil­ dren, all three tests appear to provide an adequate sam­ pling of ability.

1 40

CHAPTER 6

WECHSLER INTELLIGENCE SCALE FOR CHILDREN-REVISED (WISC-R): DESCRIPTION

A D M I N IST E R I N G THE WISC-R (AN D WPPSI A N D WAIS-R) TO H A N DICAPPED

is possible when there is impairment of arm-hand use . The cards can be arranged in the order indicated by the child.

C H I LDREN

Various physical abilities are necessary in order for a child to respond to each WISC-R subtest (see Table 6-11) . Vision and/or hearing are necessary for most of the Verbal Scale subtests; vision and arm-hand use are necessary for the Performance Scale subtests. In administering the WISC-R (and WPPSI and WAIS-R) to children with physical dis­ abilities, you must attempt to fi nd new ways to give the test without, in the process, providing cues to the child. When needed modifications go beyond simply permitting the child to respond in a different manner, however, it is likely that the results must be reinterpreted in the light of the modifications. The point-scale format of the WISC-R (and WPPSI and WAIS-R) reduces the need to shift continually from one subtest to another, thus making the WISC-R (and WPPSI and WAIS-R) more convenient than the Stanford­ Binet: Form lrM (but not the Stanford-Binet: Fourth Edi­ tion) for adaptive administration.

Verbal Scale Subtests All of the Verbal Scale subtests can be administered orally if the child can hear. If the child cannot hear but can read, the Information, Comprehension , Similarities , and Vocab­ ulary questions can be typed on cards and presented one at a time . Visually presenting the Arithmetic and Digit Span items poses more difficulties because of the time limits involved in these subtests and because visual presentation of the items seems drastically different from oral presenta­ tion , especially with Digit Span items . Therefore these two subtests may have to be omitted from the battery in testing deaf children. Detailed instructions for administering the Performance Scale subtests to deaf children appear in Appendix D. If the child cannot respond orally, written replies to any of the Verbal Scale subtests can be accepted .

Performance Scale Subtests WISC-R (and WPPSI and WAIS-R) Performance Scale subtests require the child to have adequate vision. Adapta­ tions center on the child's methods of responding. The Picture Completion subtest can be given only to a child who has adequate vision and who can either describe the missing part orally or in writing or point to it. Block Design , Object Assembly, Coding, Digit Symbol , Mazes, Animal House , and Geometric Design are not easily adaptable for the child whose arm-hand use is severely impaired. Adaptation of the Picture Arrangement subtest

Advantages of lWo Separate Scales The testing of handicapped children is facilitated by the division of the subtests into the Verbal and Performance Scales . The Verbal Scale can be administered to blind children and to children with severe motor handicaps, whereas the Performance Scale can be administered to hearing-impaired children and to children who have little or no speech. When the Verbal Scale can also be adminis­ tered to a hearing-impaired child , a comparison between the two scales may reveal the extent of the child's verbal deficit.

Unknown Effects of Modifications W ithout empirical findings, there is no way of knowing how the suggested modifications affect the reliability and validity of the scores. Yet, when standard procedures can­ not be used because sensory handicaps prevent the child from comprehending the instructions, such modifications may be needed. When modifications are used, the resulting

score should be considered only as an approximate esti­ mate of the child's test score. Timed Subtests Speed of correct response on the WISC-R Picture Ar­ rangement, Block Design, and Object Assembly subtests is significantly related to chronological age and to prob­ lem-solving ability (Kaufman, 1979b) . Older children solve the tasks more quickly than younger children, and those who solve the problems quickly also tend to solve more problems than those who solve them slowly. Because speed plays only a limited role in enabling children below 10 years of age to earn bonus points, it is reasonable to administer these subtests to 6- to IO-year-old orthopedi­ cally handicapped children who are able to manipulate the materials; these children will not be unduly penalized for failure to earn bonus points (Kaufman, 1979b).

ASSETS OF T H E WISC-R

The WISC-R is a well-standardized test, with excellent reliability and adequate concurrent validity. Like the WAlS-R and the WPPSI, it divides the 12 subtests into two sections and provides three IQs - Verbal , Performance,

141

ASSETS OF THE WISC-R

Table 6- 1 1 Physical Abilities Necessary and Adaptable for WISC-R, WPPSI , and WAIS-R Subtests

Vision

Hearing

Oral speech

Arm-hand use

Infonnation Comprehension Arithmetic S imilarities Vocabulary Digit Span

Xr Xr Xr Xr Xr

Xa Xa Xa Xa Xa X

Xa Xa Xa Xa Xa Xa

Xw Xw Xw Xw Xw Xw

Picture Completion Picture Arrangement Block Design Object Assembly Coding, Digit Symbol Mazes

X X

Xa XO

X X X

Xa Xa Xa Xa Xa Xa

Xp or Xw Xa X X X X

Sentences Animal House Geometric Design

Xr X X

Xa Xa Xa

Xa XO

Xw Xa X

Subtest

X

Note. The code is as follows:

X - This ability is required. Adaptation is not feasible if this function is absent or more than mildly impaired. Xa - This ability is required for standard administration , but the subtest is adaptable. X o - Examinees who are able to speak can say their answers. X p - ExaTPjnees who are able to point can point to their answers. Xr- Examinees who are able to read can be shown the questions. If the examinee cannot read, hearing is necessary. If neither the ability to read nor the ability to hear is present, the subtest should not be administered . Xw - Examinees who are able to write can write their answers.

and Full Scale. This division is especially helpful in clinical and psychoeducational work and aids in the assess­ ment of brain-behavior relationships. A valuable feature of the scale is that all children take a comparable battery of subtests. The following are assets of the WISC-R : 1 . Good validity. The WISC-R has adequate concurrent validity with a variety of ability and achievement mea­ sures. 2. High reliabilities. The internal consistency reli­ abilities of the WISC-R Full Scale IQs are extremely high (average rxx = . 96) , with standard errors of measurement of less than S points on the three scales. Because the WISC-R manual provides reliability data, standard errors of measurement, and intercorrelations of subtest scores by one-year age intervals as well as for the average of the 1 1 a g e groups, t h e scale's properties c a n be evaluated throughout its entire age range. Confidence intervals can be established for IQs for each of the 11 separate age groups, thereby providing estimates that are specifically applicable to each child's chronological age.

3. Excellent standardization. The standardization pro­ cecjures were excellent, sampling four geographic regions, both sexes, white and nonwhite populations, urban and rural residents, and the entire range of socioeconomic classes. 4. Good administration procedures. The prescribed procedures for administering the WISC-R are excellent. The examiner actively probes the child's responses in order to evaluate the breadth of the child's knowledge and deter­ mine whether the child really knows the answer. On items that require two reasons for maximum credit, the child i s asked for another reason when only one reason is given. These procedures ensure that the child is not penalized for not understanding the demands of the questions. The em­ phasis on probing questions and queries is extremely desirable. S. Good manual and test materials. The WISC-R man­ ual is easy to use ; it provides clear directions and tables. Reading of the directions is facilitated by the fact that the examiner's instructions are printed in a different color.

1 42

CHAPTER 6

WECHSLER INTELLIG ENCE SCALE FOR CHILDREN-REVISED (WISC-R): DESCRIPTION

Helpful abbreviations are provided for recording the child's responses, such as "Q" for Query, "DK" for Don't Know, "Inc . " for Incomplete, and "NR" for No Response . The test materials are interesting to children . 6. Helpful scoring criteria. The criteria for scoring re­ plies have been carefully prepared . The Similarities and Vocabulary scoring guidelines, for example, detail the rationale for 2 , 1, and 0 scores. A number of examples demonstrate the application of the scoring princ iples . Many typical responses are scored, and those deemed to need further inquiry are indicated by a "Q." 7 . Extensive research and clinical literature. There is a vast amount of research and case material on the WISC-R that can be used to aid interpretation.

LIMITATI O N S O F T H E W ISC-R

Although the WISC-R is an excellent instrument, there are a number of difficulties with the test and with the manual that should be recognized. 1 . Limited applicability of norms for children younger than 6-4 years and older than 16-8 years. One of the major

questions about the WISC-R is the applicability of its norms for children between 6-0 and 6-3 years of age and between 16-8 and 16-11 years of age . In the standardization of the WISC-R, the only children tested were those whose birthdates were at mid-year, plus or minus 1 . 5 months. Consequently, at the 6-year-old age leve l , no children who were between 6-0 and 6-3 years of age were included. (A footnote on page 48 of the WISC-R manual indicates that there was a small norms group [N = 50] at age 6 years, 0 months , which was used as a guide in extrapolating norms for the 6-0 to 6-3 age range . ) The same type of gap occurs for children between 16-8 and 16-11 years of age . Neverthe­ less , the WISC-R manual presents norms for these two three-month age periods. Therefore the standardized nor­ mative scores, arrived at by extrapolation , may not yield correct results for these two age groups. 2 . Limitedfloor and ceiling. Another difficulty with the WISC-R is that the range of Full Scale IQs (40 to 160) is insufficient for both severely retarded children and ex­ tremely gifted children. The test is designed so that a 6year-old child receives up to 3 scaled-scored points for giving no correct answers on some subtests. Throughout the entire age range of the test, children receive at least 1 scaled-score point on every subtest, even if they have a raw score of O . Wechsler recognized that this can present a problem in computing IQs and therefore recommended that IQs for each scale be computed only when the child

obtains a raw score greater than 0 on at least three of the subtests on each of the scales. Similarly, a Full Scale IQ should not be computed unless raw scores greater than 0 are obtained on three Verbal and three Performance sub­ tests. These are only recommendations, however, and must be considered as such until validity data show that other procedures for computing IQs are not valid . If Wechsler's recommended procedure is followed, what is the lowest possible IQ that a 6-year-old child can re­ ceive? If the child obtained raw scores of 1 on the Informa­ tion , Vocabulary, Comprehension, Picture Completion, Object Assembly, and Coding subtests and a raw score of 0 on each of the remaining four subtests, the resulting IQs would be as follows: Verbal Scale IQ = 57 (15 scaled­ score points) , Performance Scale IQ = 48 (11 scaled-score points) , and Full Scale IQ = 48 (26 scaled-score points) . Six I-point successes yield an IQ of 48 . Therefore the WISC-R may not provide precise IQs for young children who are functioning at 2 or more standard deviations below the mean of the scale. Even for IQs between 70 and 80, only a very small sample of a child's ability is tested because so few items are administered . The WISC-R does not appear to sample a sufficient range of cognitive abilities for low-functioning children. If a child fails all or most of the items on the WISC-R, a different test should be admin­ istered to obtain a more accurate estimate of the child's abilities. The highest IQ that can be obtained by children aged 16-8 years and older is 158 , a ceiling score that probably is too low to make the test appropriate for use with extremely gifted children. 3. Nonuniformity of scaled scores. The range of scaled scores on all subtests is not uniform throughout the age range covered by the scale. Only in the 6- through 10-year age range can children receive up to 19 scaled-score points on all subtests. At age 11 years, for example, the highest raw score on the Arithmetic subtest (18 points) corre­ sponds to a scaled score of 1 8 . At age 16- 8 , this same raw score corresponds to a scaled score of 16. On only 5 of the 12 subtests can children aged 16-8 and older receive the highest scaled score. The nonuniformity of available scaled scores throughout the entire age range makes profile analysis difficult to apply to high-scoring children 11 years of age or older. Profile analysis techniques can be applied appropriately throughout the entire range of scaled scores for five sub­ tests only (Similarities, Vocabulary, Digit Span, Picture Arrangement, and Coding) . For older gifted children, profile analysis can be applied only when all scaled scores are 16 or below. Applying profile analysis uniformly to all

SUM MARY

subresrs would be misleading for individual cases because the same number ofscaled-score points cannot be obtained on all subtests. 4 . Difficulty in scoring responses. When responses on the Similarities, Comprehension, and Vocabulary subtests differ from those that appear in the WISC-R manual , they may be difficult to score. Such difficulties may lead to halo effects in scoring and contribute to examiner bia s (cf. Sattler & Ryan, 1973b) . 5 . Difficulty in interpreting norms when a supplemen­ tary subtest is substituted for a regular subiest. With the norms based on only the 10 regular subtests, there is no way to know precisely what the scores mean when one of the supplementary subtests (Digit Span or Mazes) is sub­ stituted for a regular subtest. A substitution of this kind, therefore , should be made only in unusual circumstances and the results labeled "tentative" when the scores are reported. 6. Lack ofnormative datafor raw scores. The WISC-R manual fails to give means, standard deviations, and fre­ quency distributions for the raw scores. 7. Failure to describe procedure for establishing cutoff criteria. The WISC-R manual fails to provide information concerning how the cutoff criteria (the number of items that should be administered before the test is discontinued) were determined (that is, empirically or intuitively) .

CONCLU DI N G C O M M E N T ON T H E WISC-R

The WISC-R has been well received by those who use tests to evaluate children's intellectual ability. It has excellent standardization, reliability, and concurrent validity, and much care has been taken to provide useful administrative and scoring guidelines. Although some minor problems exist in the WISC-R manual, it is, on the whole, excellent. A valuable addition to the manual would have been data about the standard errors of measurement of IQ scores on the Verbal, Performance , and Full Scales at IQ levels of 70, 100, and 130 (if not others) . The WISC-R will serve as a valuable instrument in the assessment of children's intel­ ligence for many years to come .

SUM MARY I . The WISC-R was published in 1974, 25 years after the original WISC . The WISC-R is similar to its predecessor, with 72 percent of the items retained , plus the original Coding subtest.

1 43

The WISC-R is applicable to children from 6-0-0 to 16-11-30 years of age. Standardization of the scale was excellent and included both white and nonwhite children. 2. The WISC-R provides Deviation IQs for the Verbal, Per­ formance, and Full Scales (M = 100, SD = 15) and standard scores for the 12 subtests (M = 10, SD 3) . 3 . Although Wechsler objected to the use of mental ages in the calculation of IQs, the WISC-R manual includes a table of test-age equivalents for the scaled scores; these are essentially mental-age scores. 4. The internal consistency reliabilities of the Verbal , Perfor­ mance, and Full Scales are excellent (average of .94, .90, and . 96, respectively), with a standard error of measurement for the Full Scale of about 3 IQ points. Subtest reliabilities range from . 70 to .86. 5 . Shifts in IQ due to practice effects (after a one-month test­ retest interval) are about 7 IQ points on the Full Scale, 4 IQ points on the Verbal Scale, and 10 IQ points on the Performance Scale. Average changes in Full Scale IQs over a two-year period are about 3 points. 6. The WISC-R has acceptable criterion validity. Median correlations with measures of achievement and school grades range from the upper . 30s to the low . 80s. 7 . The WISC-R has acceptable concurrent validity. Correla­ tions with other Wechsler scales and with the Stanford-Binet : Fourth Edition are in the . 70s to . 80s . 8. The WISC-R tends to provide lower IQs than do the WPPSI and the WAIS-R. The various Wechsler scales do not appear to provide interchangeable IQs . 9. The relationships between WISC-R IQs and various strati­ fication variables were as follows. Mean differences between girls and boys were less than 3 points. IQs of white children were about I standard deviation higher than those of black children (102 vs. 86). Children's mean IQs varied by up to 21 points as a function of their parents' occupational group (108 vs. 87). U rban­ rural differences were smal l , and geographic region differences were largest between the West and the South (103 vs. 96). 10. A factor analysis of the WISC-R standardization data indicated that three factors account for the scale's structure : Verbal Comprehension, Perceptual Organization, and Freedom from Distractibility. The Verbal Scale subtests load primarily on Verbal Comprehension; Performance Scale sub tests load pri­ marily on Perceptual Organization; and Arithmetic, Digit Span, and Coding load primarily on Freedom from Distractibility. The best measures of g are four Verbal Scale subtests - Vocabulary, Information, Similarities, and Comprehension - and one Perfor­ mance Scale subtest - Block Design. I I . The WISC-R factor structure found in a variety of ethnic groups and exceptional populations generally is similar to that found in the standardization sample. 1 2 . Because most WISC-R subtests have an adequate degree of subtest specificity, interpretation of profiles of subtest scores generally is on firm ground . 1 3 . The Deviation IQs associated with the Verbal and Perfor=

1 44

CHAPTER 6

WECHSLER INTELLIGENCE SCALE FOR CHILDREN -REVISED (WISC-R): DESCRIPTION

mance Scale IQs can be used as factor scores. Somewhat purer factor scores can be obtained by using Information, Similarities, Vocabulary, and Comprehension for the Verbal Comprehension factor and Picture Completion, Picture Arrangement, Block De­ sign, and Object Assembly for the Perceptual Organization fac­ tor. The three subtests that comprise Freedom from Distrac­ tibility - Arithmetic, Digit Span, and Coding - also provide a factor score. 1 4 . It is important to develop proper administrative pro­ cedures early in your testing career. 1 5 . Beginning examiners tend to make a variety of adminis­ trative errors. They should be especially careful to complete the record booklet properly, adhere to direction s , probe ambiguous responses, and follow discontinuance procedures. 1 6 . The standard order of administering the subtests should be followed in all but the most exceptional circumstances. 1 7 . The WISC-R manual describes a number of adminis­ trative procedures that must be followed to ensure standardized scoring. When the entry-point items are passed, give credit to items failed below the entry-point items. Conversely, do not give credit to items passed above the discontinuance-point items. 1 8 . The WISC-R requires the use of many probing questions and queries. Spoiled responses are scored O . 1 9 . Modifications i n test procedures have been found t o in­ crease children's scores. Use modifications only after the stan­ dard administration . 20. Scoring WISC-R Vocabulary, Comprehension, and Sim­ ilarities subtests requires considerable s ki l l . A careful study of the scoring criteria can help to reduce errors in scoring . 2 1 . Extrapolated IQs are difficult to interpret because of er­ rors associated with such scores. If extrapolated IQs are needed, however, Table C-6 in Appendix C can be consulted. 22. Short forms of the WISC-R, although practical , have serious d isadvantages . Short-form IQs may be less stable, im­ pede profile analysis, and result in misclassifications. If short forms are needed for screening purposes , the procedures advo­ cated by Tellegen and Briggs should be fol lowed to determine Deviation IQs. Table C-1O in Appendix C shows the best com­ binations of two, three, four, and five WISC-R subtests. Other tables in Appendix C also provide information about short forms. 2 3 . The WISC-R and the WPPSI can be viewed as alternative forms for children aged 6-0 to 6-7 years, and the WISC-R and WAIS-R for children aged 16-0 to 16-11 years. Although the choice of a test in the overlapping age groups will likely depend on the examiner's personal preference, for children of below-normal ability the WPPSI is slightly preferred to the WISC-R and the WISC-R is slightly preferred to the WAIS-R. In the overlapping age levels, the standard errors of measurement are smaller for WPPSI IQs than for WISC-R IQs and smaller for WAIS-R IQs than for WISC-R Verbal and Full Scale IQs. 24. Children must be able to hear to take most WISC-R (and WPPSI and WAIS-R) Verbal Scale subtests, although vision may be used as a substitute modality for some subtests. Arm-hand use is a prerequisite for almost all of the Performance Scale subtests, although some adaptations are possible. The Verbal and Perfor-

mance Scale arrangement of the subtests facilitates the selection of scales for testing handicapped children. Special procedures, described in Appendix D, are usually needed to administer the Performance Scale to deaf children. 25. The assets of the WISC-R include its excellent reliability, validity, and standardization; good administrative procedures; good manual ; and helpful scoring criteria. 26. The l i m itations of the WISC-R include l im ited ap­ plicability of norms for ages 6-0 to 6-3 years and for ages 16-8 to 16-11 years, l i mited range of IQs (40 to 160) , nonuniformity of scaled scores, d ifficulty i n scoring some subtests, difficulty in interpreting norms when a supplementary subtest is substituted, lack of normative data for raw scores, and failure to describe procedures for establishing cutoff criteria. 2 7 . Overall , the WISC-R represents a major contribution to the field of intelligence testing of children. It serves as one of the most important instruments for this purpose.

KEY TERMS, C O N C E PTS, A N D N A M E S

WJSC-R standardization sample (p. 121) WISC-R Deviation IQ (p. 121) WISC-R scaled scores (p. 121) WJSC-R test-age equivalents ( p . 121) Reliability of the WJSC-R (p. 122) Standard errors of measurement of the WJSC-R (p. 123) Stabil ity of the WJSC-R (p. 123) Val idity of the WJSC-R (p. 123) WJSC-R subtest intercorrelations (p. 125) Stratification variables on the WISC-R (p. 126) WISC-R Verbal Comprehension factor (p. 127) WJSC-R Perceptual Organization factor (p. 127) WJSC-R Freedom from Distractibility factor (p. 127) Subtest specificity on the WISC-R (p. 128) Factor scores on the WISC-R (p. 130) Entry-point items on the WISC-R (p. 134) Discontinuance procedure on the WISC-R (p. 135) Extrapolated IQs for the WISC-R (p. 136) Short forms of the WISC-R (p. 136) Yudin's abbreviated procedure for the WISC-R (p. 137)

STUDY Q U ESTI O N S I . Discuss the WISC-R . Include in your discussion t h e follow­ ing issues: standardization, Deviation IQs , test-age equivalents, reliability, and validity. 2 . Describe and interpret the intercorrelations between WISC-R subtests and scales. 3 . Describe and interpret WISC-R IQs with respect to the stratification variables used in the standardization sample. 4 . Describe and interpret WISC-R factor analytic findings. 5 . Discuss WISC-R administrative considerations.

STUDY QUESTIONS

6 . Discuss WISC-R short forms, including their values and l imitations. 7. For overlapping ages how would you go about choo'5ing between the WISC-R and the WPPSI, and between the WISC-R and WAIS-R?

1 45 8. Identify the most important factors to consider in adminis­ tering the WISC-R (and other Wechsler tests) to handicapped children. 9. Discuss the assets and l imitations of the WISC-R.

WISC-R SU BTESTS

Information

Wit is brushwood, judgment is timber. The first makes the brightest flame, but the other gives the most lasting heat.

Similarities Arithmetic

-Hebrew proverb

Vocabulary The knowledge of words is the gate to scholarship.

Comprehension

-Woodrow Wilson

Digit Span Picture Completion

The true art of memory is the art of attention. -Samuel Johnson

Picture Arrangement Block Design Object Assembly Coding Mazes Summary

1 46

1 47

INFORMATION

This chapter provides the necessary background for ad­ ministering, scoring, and interpreting the 12 WISC-R sub­ tests. Included in the description of each subtest is the rationale, factor analytic findings , reliability and corre la­ tional highlights , and administrative and interpretive con ­ siderations. The factor analytic findings (from Kaufman , 1975a, 1979c) and reliability and correlational data dis­ cussed in this chapter are based on the entire standardiza­ tion sample. Reliabilities for the Digit Span and Coding subtests are test-retest correlations, whereas those for the remaining 10 subtests are split-half correlations correcled by the Spearman-Brown formula. The abilities measured by each WISC-R subtest, back­ ground factors influencing performance, implications of high and low scores, and suggested training activities to improve a child's scores are summarized in Table C-13 in Appendix C . Table C-13 is a useful reference for report writing and deserves careful study and evaluation . Readers interested in the WISC-R Structure-of-Intellect classifica­ tions may refer to Table C-12 in Appendix C ; a supplemen­ tary scoring guide by Massey, Sattler, and Andres (1978) presents additional responses to aid in the scoring of the Similarities, Vocabulary, and Comprehension subtests. Although many of the subtests have enough subtest spec­ ificity (see Chapter 6) to provide reliable estimates of some specific abilities, combinations of individual subtests will produce the most reliable estimates. For example, the Verbal Scale IQ, which is derived from a combination of five subtests , yields more accurate data about a child's verbal skills than does a single subtest score, such as the Vocabulary scaled score. In the discussions of test administration procedures, a number of questions are posed to aid you in observing children's performance . The answers to these questions will serve as a database for testing clinical hypotheses once the test has been completed. Although you may not have many questions at the beginning of the testing , you should have some specific ones in mind at the end . Precise record­ ing as you proceed will enable you to answer questions that arise later.

I N F O RMATION

The Information subtest contains 30 questions that sample a broad range of general knowledge. Included are ques­ tions concerning names of objects, dates, historical and geographical facts , and other such information . The child's age determines which item is used to start testing: Chil-

dren 6 to 7 years old start with item 1 ; 8 to 10 years old, item 5; II to 13 years old, item 7; and 14 to 16 years old, item H . All items are scored 1 o r 0 (pass-fail ) , and the subtest i s discontinued after five consecutive failures. The questions usually can be answered correctly with a brief, simply stated fact . Children need only demonstrate that they know specific facts ; they need not find rela­ tionships between these facts .

Rationale The amount of know ledge children possess may depend on their natural endowment, the extent of their education (both formal and informal) , and their cultural oppor­ tunities and predilections. In general , the Information sub­ test samples the knowledge that average children with average opportunities should be able to acquire through normal home and, school experiences . The child's re­ sponses and comments provide clues about the child's general range of information, alertness to the environ­ ment, social or cultural background, and attitudes toward school and school-like tasks ("Those questions are hard, just like my teacher asks") . High scores should not be interpreted as indications of mental efficiency and compe­ tence, since the fact that individuals have acquired isolated facts does not mean that they know how to use them appropriately or effectively. However, intellectual drive may contribute to higher scores. Successful performance on the Information subtest requires memory for habitual ,

H O W W O U LD Y O U S C O R E T H I S ?

CHAPTER 7

1 48

overlearned responses (that is, information that the child has likely been exposed to over and over again), especially in older children. Thus Information provides clues about the child's ability to store and retrieve old data.

Factor Analytic Findings The Information subtest is tied with Similarities as the second-best measure of g (58 percent of its variahce may be attributed to g) . It has ample subtest specificity to permit specific i nterpretation of its functions and contributes sub­ stantially to the Verbal Comprehension factor (Mdn load­ ing = .63).

WISC-R SUBTESTS

ones may suggest poor motivation, anxiety, temporary inefficiency, or an environment that has not been consis­ tent. Alternatively, this pattern may indicate a problem with retrieval of information from long-term memory. When you suspect such a problem , analyze the content of the failed items - do they deal with numerical information, history, science, or geography? Content analysis may pro­ vide clues to areas of interest or suggestions for points of inquiry after the test has been completed. Question : What comes in a bottle? Answer: Genies . (Flumen & Flumen, 1 979)

Reliability and Correlational Highlights Information is a reliable subtest (rxx = . 85 ) . It correlates more highly with Vocabulary (r = . 69) than with any other subtest. It correlates moderately with both the Full Scale (r = . 70) and Verbal Scale (r = . 74) , but to a lesser degree with the Performance Scale (r = . 56 ) .

Administrative and Interpretive C onsiderations The Information subtest is easy to administer. The ques­ tions are simple and direct, and timing is not required. Credit should be given if questions are answered correctly at any time during the course of administering the scale. Scoring is usually straightforward : a correct response re­ ceives 1 point; an incorrect response, O . If two or more answers are given to a question, the examinee should be asked to choose the best answer. Answers should be re­ corded verbatim . Children who are hesitant to respond should be encouraged to guess or take a chance . Note the quality of a child's answers. Is the child thinking the questions through or simply guessing? Are answers precise or wordy? Overly long responses or responses filled with extraneous information may suggest an ob­ sessive-compulsive orientation - children with this orien­ tation sometimes feel compelled to prove how much they know. Alternatively, excessive responses may simply re­ flect the examinee's desire to impress you. The child's entire test protocol plus other relevant information should be considered in interpreting such behavior. Inhibition, too, should be noted, as inability to recall an answer may suggest that the question is associated with conflict-laden material. (For example, a child may not be able to recall the number of legs on a dog because of a traumatic experi­ ence with dogs . ) Examine the pattern o f successes and failures . Failures on easy items coupled with successes on more difficult

SIMILARITIES The Similarities sub test contains 17 pairs of words ; the child must explain the similarity between the two items in each pair. All children begin with the first item. The first four items are scored 1 or 0 (pass-fail) ; items 5 through 17 are scored 2 , 1 , or 0 , depending on the conceptual level of the response. The subtest is discontinued after three con­ secutive failures .

Rationale In addition to perceiving the common elements of the paired terms, children must bring these common elements together in a concept in order to answer the questions on the Similarities subtest. Thus the Similarities subtest may measure verbal concept formation - the ability to place objects and events together in a meaningful group or groups . Although concept formation can be a voluntary, effortful process, it can also reflect well-automatized ver­ bal conventions (Rapaport , Gil l , & Schafer, 1968) . Perfor­ mance on the Similarities subtest may be related to cultural opportunities and interest patterns. Memory may also be involved.

Factor Analytic Findings The Similarities subtest is tied with Information as the second-best measure of g (58 percent of its variance may be attributed to g). It has ample subtest specificity for children 6 \t2 , 7 1/2 , and 8 1f2 years old, but not for older children. Thus specific interpretation of the subtest's func­ tions is appropriate only for the three earliest ages. The Similarities subtest contributes substantially to the Verbal

ARITHMETIC Comprehension factor (Mdn loading = .64) and thus is interpretable as a measure of verbal comprehension.

Reliability and Correlational Highlights Similarities is a reliable subtest (rxx = . 81) . It correlates more highly with Vocabulary (r = .67) and Information (r = . 62) than with any other subtests. It correlates moder­ ately with the Full Scale (r = . 71) and the Verbal Scale (r = . 72), and to a somewhat lesser degree with the Perfor­ mance Scale (r = . 58).

Administrative and Interpretive Considerations Responses to the first four questions are generally easy to score , but scoring of items 5 to 17 is more difficult. On the latter items (5-17), a conceptual response, such as a gen­ eral classification, receives a 2 ; a more concrete response, such as a specific property of the item, receives a I; and an incorrect response receives a O. This scoring system is designed to take into account whether the responses are essential likenesses or superficial likenesses. Many Similarities responses are difficult to score, as was shown by a study of the way 110 psychologists and graduate students scored 187 ambiguous WISC-R Similarities, Vo­ cabulary, and Comprehension responses (Sattler, Andres, Squire . Wisely, & Maloy, 1978) . A level of 80 percent agreement in scoring was achieved for only 51 percent of the ambiguous Similarities responses (95 out of 187) . In practice, however, it is unlikely that any one protocol would include such a large number of ambiguous re­ sponses. Scoring difficulties arise in part from the limited number of examples in the manual and the difficulty of establishing precise criteria that apply to all responses , including idiosyncratic ones.

Scoring criteria.

A careful study of the scoring guide in the manual will help you become more proficient in scor­ ing . lWo parts of Appendix A in the WISC-R manual ­ "Scoring Criteria" for the Similarities subtest - are es­ pecially important. First, you should thoroughly master the general scoring principles, which elucidate the ra­ tionale for 2, I , and 0 scores. Second, you should carefully study the "Sample Responses" section , which lists types of responses that should be queried [as shown by a "(Q)"] , so that you can recognize responses that need probing . Noting children's u nderstanding of the task.

Note whether children understand the task. On items I and 2 , give children the correct responses if they answer incor­ rectly. Children who state that they do not know the answer

1 49

to a question should be encouraged to think about the question, but they should not be pressed unreasonably. When children give multiple acceptable answers to an item, score their best response. When both correct and incorrect responses are given to an item, say "Now which one is it?" The score should be based on the answer to your follow-up question . If the child gives a I -point answer to item 5 or 6, you should tell the child the 2-point answer, to provide a model response . This procedure may encourage the child to give 2-point responses on later items .

Interpreting responses. Responses to the Similarities subtest may provide insight into the logical character of the child's thinking processes. Observe the child's typical level of conceptualization throughout the subtest. Are the an­ swers on a concrete , functional , or abstract level? Con­ crete answers typically refer to qualities of the objects (or stimuli) that can be seen or touched (apple-banana: "Both have a skin") . Functional answers typically concern a func­ tion or use of the objects (apple-banana: "You eat them") . Finally, abstract answers typically refer to a more univer­ sal property or to a common classification of the objects (apple-banana: "Both are fruits"). A 2-point abstract response does not necessarily reflect abstract thinking ability. It may simply be a conventional , overlearned response. For example, there is a difference between the 2-point response "Both fruits" for apple-ba­ nana and the 2-point response "Social qualities" for liberty­ justice . The former may be a conventional response, whereas the latter likely reflects a more abstract level of conceptualizing ability. Observe how the examinee handles any frustration in­ duced by the subtest questions. Is the examinee nega­ tivistic and uncooperative , or is he or she genuinely unable to see the similarity involved? Responses such as "They are not alike" may be an indication of negativism, an attempt to avoid the task demands, suspiciousness, a coping mecha­ nism, or failure to know the answer. To determine which of these factors may account for a child's responses, compare the child's style of responding on the Similarities subtest with that on other subtests. Question : In what way are an orange and a pear alike? Answer: Both give me hives.

ARITH M ETIC

The Arithmetic subtest contains 18 problems , 15 of which are presented orally and 3 of which are presented on cards.

CHAPTER 7

1 50

The last three Arithmetic items appear in the Picture Com­ pletion/Block Design booklet. Many of the problems are similar to those commonly encountered by children. An­ swers must be given without the use of paper and pencil . Children 6 to 7 years old (and older children suspected of mental retardation) begin the test with item 1 ; 8 to 10 years old, item 5 ; 11 to 13 years old, item 8 ; and 14 to 16 years old , item 10. The problems are timed, with the first 13 items having a 3D-second time limit; items 14 and 15 , a 45-second time limit; and items 16 to 1 8 , a 75-second time limit. Items are scored l or 0, with the exception of items 2 and 3 , for which V2 point can be given. The subtest is discontinued after three consecutive failures. Problems on the Arithmetic subtest test various skills. Problems 1 , 2 , and 3 require direct counting of concrete quantities . Problems 4, 6 , 7 , 8 , 9, 11, and 12 require simple addition or subtraction . Problems 5 , 13, and 15 involve simple division. Problem 10 involves multiplication. Prob­ lems 14 , 16, 17, and 18 require the use of automatized number facts and subtle operations, such as identifying relevant relationships at a glance. The answers to problems 13 and 15 may come intuitively to children experienced with facts. For those who have not yet automatized simple arithmetic facts, the problems may require reflection and mental operation.

WISC-R SUBTESTS

presented in the task) and recognizing when an appropriate response has been made (in order to change incorrect patterns or strategies) . The mathematical ski l l s include the ability to comprehend and integrate verbal information presented in a mathematical context and numerical ability.

Factor Analytic Findings The Arithmetic subtest is a fair measure of g (42 percent of its variance may be attributed to g). It has ample subtest specificity to permit specific interpretation of its functions. Arithmetic has a high loading on the Freedom from Dis­ tractibility factor (Mdn loading = . 5 8) and a moderate loading on the Verbal Comprehension factor (Mdn load­ ing = . 37 ) .

Reliability and Correlational Highlights

Arithmetic is a relatively reliable subtest (rxx = . 77). It correlates more highly with Information (r = . 54) and Vocabulary (r = . 52) than with the other subtests. It has a relatively low correlation with the Full Scale (r = . 58) and Verbal Scale (r = .58), and an even lower correlation with the Performance Scale (r = .48).

Administrative and Interpretive Considerations Rationale The problems on the Arithmetic subtest require the child to follow verbal directions, concentrate on selected parts of questions, and use numerical operations. Children must have knowledge of addition, subtraction, multiplication, and division operations. The emphasis of the problems is not on mathematical knowledge per se, but on mental computation and concentration. C oncentration is es­ pecially important for the complex problems. The Arithmetic subtest measures numerical reasoning ­ the ability to solve arithmetical problems. It requires the use of noncognitive functions (concentration and attention) in conjunction with cognitive functions (knowledge of numerical operations) . Success on the subtest is influenced by education, interests, fluctuations of attention, and tran­ sient emotional reactions. Like the Vocabulary and Infor­ mation subtests, Arithmetic taps memory and prior learn­ ing; however, it also requires substantial concentration and the active application of select skills to new and unique situations (Blatt & Allison, 1968) . Information-processing strategies as wel l as mathe­ matical skills may underlie performance on the Arithmetic subtest (Stewart & Moely, 1983). These strategies may include rehearsal (in order to remember the information

Testing-of-limits procedures can facil itate interpretation of the test results. Examinees should be given additional time to complete the problems when they need it, even though no credit is given for a late response. Any correct late responses should be recorded , along with the amount of time that elapsed between expiration of the time limit and their submission. Such information may help you differ­ entiate between failures due to temporary inefficiency and those due to limited knowledge . Successful delayed per­ formance may indicate temporary inefficiency or a slow, painstaking approach to problem solving . To determine reasons for children's failure , use inquiry after the test is completed. You might say, for example , "Let's try this one again. Tell m e how you solved the problem." Failure may be caused by poor knowledge of arithmetical operations, inadequate conceptualization of the problem, temporary inefficiency or anxiety, poor con­ centration , or carelessness. Allowing the child to use paper and pencil is another testing-of-limits procedure that may help you to differenti­ ate inadequate arithmetical knowledge from attentional or concentration difficulties. If the child is able to solve the items with pencil and paper, the failure is not associated with lack of arithmetical knowledge; the errors may be

VOCABULARY

151

associated with attention o r concentration difficulties that inhibit mental computation . If the child fails the ite m s in both situations, the failures more l ikely reflect difficul ties with arithmetical knowledge, although attention and con­ centration difficulties may be interfering with the exam­ inee's ability to solve written arithmetic problems . Inspect the written work to see whether numbers are misaligned, steps are sequenced incorrectly, or there is evidence of inadequate mastery of basic arithmetical operations . Question: If I cut a pear in thirds, how many pieces will I have? Answer: One. Question: (Testing-of-I imits) Are you sure I will have only one piece? Answer: Yes , and I will have the other two pieces.

VOCABU LARY

The Vocabulary subtest contains 32 words arranged in order of increasing difficulty. The child is asked to explain orally the meaning of each word (for example, "What is a ?" or "What does mean?" ) . Children 6 to 7 years old (and older children suspected of being mentally retarded) begin with item 1 ; 8 to 10 years old, item 4; 11 to 14 years old, item 6; and 14 to 16 years old, item 8. All items are scored 2 , 1, or O. The subtest is discontinued aft r fiv{! consecutive failures. __

__

Rationale The Vocabulary subtest, a test of word knowledge, may tap a variety of cognition-related factors - including learning ability, fund of information , richness of ideas, memory, concept formation , and language development - that may be closely related to children's experiences and educational environments. Because the number of words known by children is correlated with their ability to learn and to accumulate information, the subtest provides an excellent estimate of intellectual ability. Performance on the subtest is stable over time and relatively resistant to neurological deficit and psychological disturbance (Blatt & Allison, 1968) . Performance on the Vocabulary subtest is a useful index of the examinee's general mental ability.

Factor Analytic Findings The Vocabulary subtest is the best measure of g in the scale (64 percent of its variance may be attributed to g) . It has

adequate subtest specificity to permit specific interpreta­ tion of its functions across all ages. It contributes substan­ tially to the Verbal Comprehension factor (Mdn load­ ing = . 72) .

Reliability and Correlational Highlights

Vocabulary is the most reliable subtest (rxx = . 86) in the sc ale . It c orrelates m o re h ighly with I n formation (r = . 69 ) , Si milarities ( r = . 67 ) , and Comprehension (r = . 66) than with the other subtests. It correlates moder­ ately with the Full Scale (r = . 74) and the Verbal Scale (r = . 68), and to a lesser degree with the Performance Scale (r = . 58) .

Administrative and Interpretive Considerations Care should be taken to pronounce each word clearly and correctly. Pronunciation is especially important on the Vocabulary subtest because the test items are single words and you are not allowed to spell them. When you suspect that children have not heard a word correctly, have them repeat it to you . Carefully record the examinees' responses to the words .

Scoring responses. The scoring system (2, 1, or 0 for all items) takes into account the quality of the response. Two points are awarded for good synonyms , major uses, or general classifications , whereas one point is given for vague responses, less pertinent synonyms, or minor uses . Elegance of expression is not taken into account. Vocabulary is one of the more difficult subtests to score. Often it is not easy to implement Wechsler's scoring crite­ ria . In the study by Sattler et al . (1978) , 80 percent of the raters gave the same score to only 38 percent of the 352 ambiguous Vocabulary responses . Probing borderline re­ sponses and studying carefully the scoring guidelines in the WISC-R manual and the scoring examples in the Mas­ sey et aI . (1978) manual will help you resolve some of the scoring problems that arise in the course of administering the subtest. You must try to do the best job possible with the available guides. Qualitative analysis.

A qualitative analysis of a child's responses to the Vocabulary subtest may reveal something about the examinee's background , cultural milieu, social development, life experiences, responses to frustration , and thought processes . The basis for incorrect responses should be determined, since it is important to distinguish among guesses, clang associations, idiosyncratic associa­ tions , and bizarre associations . Inquiry is especially

CHAPTER 7

1 52

important whenever peculiar responses, mispronuncia­ tions , or peculiar inflections are given . Among children with schizophrenia or other severe forms of mental disor­ der, language disturbances occasionally can be seen in word definitions .

Further inquiry. When young children or older chil­ dren who may be mentally retarded give a 0- or I-point response to the first word of the Vocabulary subtest, tell them the 2-point answer. This procedure , designed to encourage 2-point responses , is not followed on subse­ quent items. Responses suggestive of regionalism or slang should be probed further (for example, "Give me another meaning for ") . The "Sample Responses" section of Appendix B of the WISC-R manual lists many responses that should be queried [as shown by a "(Q)"] ; study this section care­ fully so that you can recognize responses that should be probed. The nature of the response should determine whether the inquiry occurs during or after the standard administration . For example, if the answer clearly defines a homonym of the test item, repeat the question by saying "What else does mean?" However, if the response is possibly indica­ tive of a thinking disorder, further probing should be delayed until testing has been completed. During the test­ ing-of-limits phase you might say "To the word you said Tell me more about your answer. "

they seem uncertain about how best to express what they think? Do they use gestures to illustrate their statements or even depend on them exclusively? Note the content of definitions . Are the words chosen synonyms for the stimulus word (thief- "a burglar") , or do they describe an action (thief- "takes stuff" )? Do the chil­ dren describe some special feature of the object (donkey ­ "it has four legs") , or do they try to fit it into some category (donkey - "a living creature that is kept in a barn")? Note any emotional overtones, personal experiences, or feelings (alphabet - "I hate to write") . •

•

Question : What is a chisel? Answer: When you are cold you get the chisels. (Flumen & Flumen, 1 979)

__

__

__

.

COMPREHENSION

The Comprehension subtest consists o f 17 questions that deal with problem situations involving knowledge of one's body, interpersonal relations, and social mores. All chil­ dren begin the subtest with the first item, and all items are

__

Observing responses.

The following guidelines are useful for observing and evaluating Vocabulary responses (Taylor, 1961) . Write down all responses, whether correct or not . Note whether children are definitely familiar with the word or only vaguely familiar with it. If children explain a word, do they try to be precise and brief or embark on lengthy explanations? Are their responses objective or do they relate to personal experiences? Note whether children confuse the word with another one that sounds like it. If they do not know the meaning of a word, do they guess? Do they readily say "I don't know" and shake off further demands, or are they puzzled? In testing-of-limits , note whether displaying the printed word improves recognition. Watch for possible hearing difficulties by listening carefully to how children repeat words. Have the words been heard correctly or with some distortion? Note how the children express themselves. Do they find it easy or difficult to say what they mean? Do they have mechanical difficulties pronouncing words properly? Do •

WISC-R SUBTESTS

-

T HOSE A R E A D U LT "WORDS. WHY DON 'T YO U A S K ME SOME K I O WOR DS L I K E �'DORK�' �\5W J RLY,/I A N D \\ W H I PPET ?//

\\w , MP/, ):::.=� /

.

'��1I T{3f

/)o1' Ot>�

•

•

•

•

•

Courtesy Herman Zielinski .

1 53

COMPREHENSION

scored 2, I, or O. The subtest is discontinued after four consecutive failures.

Rationale The Comprehension subtest involves understanding given situations and providing answers to specific problems. Success depends , in part, on possession of practical in­ formation plus an ability to draw on past experiences in reaching solutions. Responses may reflect the child's knowledge of conventional standards of behavior, exten­ siveness of cultural opportunities, and level of develop­ ment of conscience or moral sense. Success suggests that the child has social judgment , or common sense, and a grasp of social conventionality. These characteristics im­ ply an ability to use facts in a pertinent, meaningful , and emotionally appropriate manner.

Factor Analytic Findings The Comprehension subtest is a good measure of g (52 percent of its variance may be attributed to g). It has adequate subtest specificity to permit specific interpreta­ tion of its functions across the entire age range and has a high loading on the Verbal Comprehension factor (Mdn loading = . 64) .

Reliability and Correlational Highlights C o m p re h e n s i o n is a reasonably re l i a b l e subtest (rxx = . 77 ) . It correlates more highly with Vocabulary (r = . 66) and S imilarities (r = . 59) than with the other subtests . It correlates moderately with the Full Scale ( r = . 66) and the Verbal Scale (r = . 68), and t o a lesser degree with the Performance Scale (r = . 5 3 ) .

Administrative and Interpretive Considerations The Comprehension subtest is difficult to score because children give many responses that differ from those pro­ vided in the manual . In the SattIer et al . (1978) study, 80 percent of the raters gave the same score to only 49 percent of the 187 ambiguous Comprehension responses . The most complete or best response receives a score of 2 ; a less adequate response , I ; and an incorrect response , O. If an inquiry about a response alters the meaning of the initial response, the reply to the inquiry determines the amount of credit given . On the first item, you are supposed to tell children the correct 2-point response if they give a less adequate re­ sponse . This procedure is meant to encourage children to

give 2-point responses and is allowed only on the fi rst item. On the nine items (3 , 4, 7, 8, 9, 12 , 14, 16, and 17) that require two ideas for full credit (2 points), you are required to ask for a second idea when only one idea is given, so that children are not automatically penalized for not giving two reasons . However, on the other items, for which an ade­ quate one-idea answer is scored 2 points , obvious I-point responses should not be probed in an attempt to improve the score .

Scoring criteria. Carefully study the "Sampie Re­ sponses" section of Appendix C of the WISC-R manual so that you will know which response types need further inquiry [these are labeled "(Q)"] . The examples indicate that many 0- and I-point responses should be queried. The inclusion of a "Q" with the 2-point responses indicates that the entire response (including the elaboration) is worth 2 points. Additional queries offer you an opportunity to evaluate more thoroughly the extensiveness of the child's knowledge. Interpreting responses.

Responses to the Comprehen­ sion questions may provide valuable information about the child's personality style, ethical values, and social and cultural background. Unlike the Information questions , which usually elicit precise answers, the Comprehension questions may elicit more complex and idiosyncratic re­ plies. Because the questions involve judgment of social situations, answers may reflect the child's attitudes. Some responses reveal understanding and acceptance of social mores, whereas others reveal understanding but not accep­ tance of social mores. A child may know the right answers but not practice them . Some children may maintain that they do not have to abide by social conventions, believing that such matters do not pertain to them personally. Initiative, self-reliance, independence, self-confidence, helplessness, and other traits may be revealed in children's replies . For example, children may reveal dependent per­ sonality styles by indicating that they would seek help from their mothers or others when confronted with the various problem situations . Replies to questions 6 and 9, which ask the examinee what should be done if a younger child starts a fight with the examinee and why criminals are locked up, may reveal independence, manipulative tendencies , na·ive perceptions of problems, cooperative solutions, hostility, or aggression (Robb, Bernardoni, & Johnson, 1972) . Note how children respond to the questions (Taylor, 1961) : Do children's failures indicate misunderstanding of the meaning of a word or the implications of a particular phrase? •

CHAPTER 7

1 54 •

Do the children give complete answers or just part of a

phrase? Do they respond to the entire question or only to a part of it? Do the children seem to be objective, seeing various possibilities and choosing the best way? Are they indecisive - unable to come to firm answers? Are the children's responses too quick , indicating failure to consider the questions in their entirety? Do the children recognize when their answers are suffi cient? •

•

•

•

•

B ecause Comprehension requires considerable verbal expression , the subtest may be sensitive to mild language impairments and to disordered thought processes. Be alert to language deficits, such as word-finding difficulties, cir­ cumstantial or tangential speec h , or other expressive difficulties .

Further inquiry. Probe unusual responses by asking children to explain further. Although inquiry may provide insight into the examinee's thought processes, it should not be conducted routinely after every response. Extensive inquiry can be conducted as part of testing-of-limits after the examination has been completed. Record the exam­ inee's responses verbatim during the initial presentation of the items and during the inquiry phase in order to facilitate restudy and qualitative evaluation. Question: Why should children who are sick stay home? Answer: To take their antibionic s . (Flumen & Flumen , 1 979)

DI GIT SPAN

On Digit Span, a supplementary subtest, the child listens to a series of digits given orally by the examiner and then repeats the digits . The Digit Span subtest has two parts : Digits Forward, which contains series ranging in length from three to nine digits, and Digits Backward, which contains series ranging in length from two to eight digits . There are two series of digits for each sequence length . Digits Forward is administered fi rst, followed by Digits Backward . Digit Span is not used in the computation of the IQ when the five standard Verbal Scale subtests are administered . All items are scored 2 , 1 , or O. The subtest is discontinued

WISC-R SUBTESTS

when the child fail s both trials on any one item, both on Digits Forward and on Digits Backward. Separate scaled scores are not provided for Digits Forward and Digits Backward. Although Digit Span is a supplementary subtest, admin­ istering it may make the WISC-R more useful diag­ nostically and programmatically, especially for obtaining factor scores. Considering the small investment of time and energy required to give this subtest, and the fact that it is used in calculating the Freedom from Distractibility factor, Digit Span should be administered routinely.

Rationale Digit Span is a measure of short-term auditory memory and attention. Performance may be affected by one's ability to relax. A child who is calm and relaxed may achieve a higher score on the subtest than one who suffers from excessive anxiety. The task assesses the child's ability to retain several elements that have no logical relationship to one another. Because auditory information must be re­ called and repeated orally in proper sequence , the task has been described as a sequencing task. Digits Forward primarily involves rote learning and memory, whereas Digits Backward requires considerably greater transformation of the stimul us input prior to recall. The mental image of the numerical sequence not only must be held longer (usually) than in the Digits Forward se­ quence, but must be manipulated before it is restated . High scores on Digits Backward may indicate flexibility, good tolerance for stress, and excellent concentration . Digits Backward involves more complex cognitive processing than does Digits Forward and has higher loadings on g than does Digits Forward (Jensen & Osborne, 1979). Because of differences between the two tasks, it is useful to consider Digits Forward and Digits Backward sepa­ rately. Digits Forward appears to involve primarily se­ quential processing, whereas Digits Backward appears to involve both planning ability and sequential processing. Additionally, Digits Backward may involve the ability to form mental images and the ability to scan an internal visual display formed from an auditory stimulu s . H ow­ ever, more research is needed to support the hypothesis about the role of visualization in Digits Backward perfor­ mance. Raw score differences of 3 points between Digits Forward and Digits Backward may be considered notewor­ thy. Gardner (1981) presents separate means, standard de­ viations, and percentile ranks for Digits Forward and Digits Backward for children ages 5-0 to 15-11.

1 55

DIGIT SPAN

Factor Analytic Findings The Digit Span subtest is a poor measure of g (24 percent of its variance may be attributed to g). It has ample subtest specificity to permit specific interpretation of its functions across the entire age range. It contributes substantially to the Freedom from Di stractibi l ity factor (Mdn load­ ing = . 56) .

Reliability and Correlational Highlights

Digit Span is a relatively reliable subtest (ru = . 78). It correlates more highly with Arithmetic (r = . 45 ) than with any other subtest. It has a low correlation with the Full Scale (r = .43), the Verbal Scale (r = . 45 ) , and the Perfor­ mance Scale (r = . 34) .

Administrative and Interpretive Considerations Be sure that the examinee cannot see the digits in the manual or on the record blank. The digits should be read clearly at the rate of one per second, with the inflection dropped on the last digit in the series. It is a good idea to practice reading speed with a stopwatch . It is never per­ missible to repeat any of the digits on either trial of a series. Both trials of each series are routinely administered. The child receives credit for each trial that he or she passes . On Digits Backward, if the child passes the sample three-digit series (on either the fi rst or the second trial), proceed to th two-digit series. If the child fails the sample series, read the specific directions in the manual that explain how the series should be repeated correctly. Whenever there is any doubt about the child's auditory acuity, an audiological examination should be requested.

Recording performance.

The number of digits in each series correctly recalled may be recorded in the record booklet by placing either a mark designating a correct answer above or on each digit correctly recalled or a mark designating an incorrect answer on each digit missed. An even better procedure is to write out the exact sequence given by the child in the available space (whether the digits are right or wrong) . These procedures provide qualitative information that can prove valuable when you write your report. An examinee who consistently misses the last digit in the first series and then successfully completes the second series differs from one who fails to recall any of the digits in the first series but successfully completes the

second. Similarly, a child who responds to the sequence 3-4-1-7 with 3-1 -4-7 is quite different from the child who says 9-8-5-6. The scoring system does not distinguish among failure patterns . For example, the same score is given to both the examinee who misses one digit in the eight-digit sequence and the examinee who misses all eight digits, even though the second exami nee's performance is more inefficient , perhaps because of lapses in attention associated with anxiety or other factors.

Observing performance. Observe whether failures in­ volve leaving out one or more digits , transposing digits , interjecting incorrect digits , or producing more digits than were given. Children who recall the correct numbers but in an incorrect sequence are more likely to have a deficit in auditory sequential memory than in auditory memory. Note also whether failures usually occurred on the fi rst trial and successes on the second trial . This pattern may reflect a learning-to-learn process or be indicative of a need for a warm-up in order to achieve success. Consider the following questions: Is performance effortless or is a great deal of COn­ centration used? Do the children view the task as interesting , boring, or difficult? When errors are made, do the children notice , or do they think that their answers are correct? •

•

•

When the Digits Backward series is presented, do the children understand the di fference between this task and Digits Forward? Are the errors made on Digits Backward similar to or different from those made on Digits Forward? As the Digits Backward series proceeds , do the chil­ dren become stimulated and encouraged, or tense and anxious? •

•

•

Examinees use various methods of recalling the digits . They may visualize the digits; say the digits to themselves and reproduce them by use of verbal, motor, or auditory techniques; or group the digits. Some grouping techniques introduce meaning into the task so that separate digits become numbers grouped into hundreds, tens, or other units. If grouping occurs, the function underlying the task may be changed from one of attention to one of concentra­ tion . After the subtest has been completed, you might ask the examinee how he or she went about remembering the numbers . If you do, be sure to record the response .

CHAPTER 7

1 56

WISC-R SUB TESTS

Reliability and Correlational Highlights Question: Now I am going to say some more numbers, but this time when I stop I want you to say them backwards . For example, if I say 8-4-6, what would you say? Answer: I'd say, you've got to be kidding! (Adapted from Flumen & Flumen, 1 979)

Picture C o m p l etion is a relatively rel ia b l e subtest ( r:ex = . 77) . It correlates more highly with Block Design (r = . 52) than with any other subtest. It has a somewhat low correlation with the Full Scale ( r = . 5 7 ) , the Perfor­ mance Scale (r = . 54) , and the Verbal Scale (r = . 50) .

Administrative and Interpretive Considerations

PICTU RE COMPLETION

The Picture Completion subtest consists of26 drawings of objects from everyday l ife , each of which lacks a single important element. The pictures are shown one at a time . The child's task is to discover and name (or point to) the missing element within the 20-second time limit. Children 6 to 7 years old (and older children suspected of mental retardation) begin with item 1 ; those 8 to 16 years old start with item 5 . All items are scored 1 or 0 (pass-fail) , and the subtest is discontinued after four consecutive failures . Most children enjoy this subtest.

Rationale The Picture Completion subtest involves recognizing the object depicted, appreciating its incompleteness , and de­ termining the missing part. It is a test of v isual discrimina­ tion - the ability to differentiate essential from nonessen­ tial details. Picture Completion requires concentration, reasoning (or visual alertness) , visual organization, and long-term visual memory (as the items require the child to have stored information about the complete figure). Picture Completion may measure perceptual and con­ ceptual abilities involved in visual recognition and identifi­ cation of familiar objects. Perception, cognition, judg­ ment, and delay of impulse all may influence performance . The time limit on the subtest places additional demands on the examinee. The richness of children's life experiences also may affect performance on the subtest.

Factor Analytic Findings

The Picture Completion subtest is a fai r measure of g (37 percent of its variance may be attributed to g). It has ample to adequate subtest specificity to permit specific interpreta­ tion of its functions at different ages and contributes sub­ stantially to the Perceptual Organization factor (Mdn load­ ing = . 57) .

Picture C ompletion is easy to administer. S imply leave the booklet flat on the table and tum the cards over to show each succeeding picture. If children have speech difficul­ ties, such as those that occur in aphasia, the subtest can be administered by having examinees point to the place where the part is missing . As you administer the subtest, consider the following: Do the examinees understand the task? Do they say anything that comes to mind , or do they search for the right answer? When they fail , do they find fault with themselves or with the picture? Are they aware of being timed? If so, does the timing make them anxious or prompt them to change the pace of their responding? • •

•

•

If a child's performance leaves any doubt about his or her visual skills, a visual examination should be requested.

Observing perseveration.

Observe whether persevera­ tion occurs . A child who says "mouth" for each picture portraying a person (pictures 2, 12 , 1 5 , 17, and 19) is displaying perseveration . "Mouth" is the correct answer for picture 2, but not for the subsequent pictures depicting people .

Recording performance.

Examinees should be aware that they are being timed , because it is important for them to realize that speed is expected . Usually, allowing them to see the stopwatch is all that is necessary. Record each incorrect response as well as the time taken to make the response. Examinees who usually respond in less than five seconds may be more impulsive, more confi­ dent, and, if correct, brighter than those who take more time . Examinees who generally respond correctly after the time limit (for which they do not receive credit) may be brighter than those who respond incorrectly before the time limit is reached. Because the pass-fail scoring makes no provision for such qualitative factors, individual varia-

1 57

PICTURE ARRANGEMENT

tions should be carefully evaluated in each case and d:s­ cussed in the report . Delayed correct responses may suggest temporary inefficiency or depression, whereas extre mely quick but incorrect responses may reflect impulsivity.

Further inquiry.

The WISC-R manual indicates that, if necessary, you may give each of three guiding statements once to help children understand the requirements of this subtest . (a) If children mention a nonessential missing part, you can ask for the most important part that is missing. (b) If children name the object pictured , you can ask what is missing. (c) If children name a part that is off the card , you can ask what is missing. On five items (6, 14, 22, 2 3 , and 24) , children should be asked to point to the missing part on the card if ambiguous responses are given. In other cases , whenever there is any doubt about children's verbal or pointing responses, ask for clarification. After the subtest has been completed, you may follow up by inquiring about children's perceptions of the task. "How did you go about coming up with the answer?" and "How did you decide when to give an answer?" are possible questions. Any peculiar answers should be queried . Chil­ dren's behavior during this subtest may provide insight into how they react to time pressure. As a testing-of-limits procedure , children can be asked to look again at those pictures that they missed. You might say "Look at this picture again. Before , you said that was missing. That's not the part that's missing. Look for something else . "

years old are gi ven item 3 . Items 1 to 4 are scored 2 , 1, or 0; items 5 to 12 are scored 3 points for the correct arrange­ ment , with up to 2 additional time-bonus points. The subtest is discontinued after three consecutive failures.

Rationale The Picture Arrangement subtest measures children's abil­ ity to comprehend and evaluate a total situation . In order to accomplish the task, children must grasp the general idea of a story. Although trial-and-error experimentation is sometimes involved, an appraisal of the total situation depicted in the cards is necessary for successfui com­ pletion . The subtest is primarily a nonverbal reasoning test which may be viewed as a measure of planning ability. Anticipation, visual organization, and temporal sequenc­ ing are involved . The ability to anticipate the conse­ quences of initial acts or situations is tested, as well as the ability to interpret social situations. Some children may generate covert, analytical , verbal descriptions of alter­ native story sequences to guide them in arranging the stimulus cards. In such cases, verbal sequencing processes are measured by the subtest as well . The capacity to anticipate, judge, and understand the possible antecedents and consequences of events is important in lending mean­ ingful continuity to everyday experiences (Blatt & Allison, 1968).

__

PICTU RE ARRANGEM E NT

In the Picture Arrangement subtest, children must place a series of pictures in logical sequence. The 12 series, or items, are similar to short comic strips. Individual cards, each containing a picture, are placed in a specified disar­ ranged order, and the child is asked to rearrange the pictures in the "right" order to tell a story that makes sense. The number of pictures per set ranges from three to five. One set of cards is presented at a time . Time limits vary from 45 seconds for items 1 through 8 to 60 seconds for items 9 through 12 . The only motor action required is to change the position of the pictures. All children begin with the sample item, after which children 6 to 7 years old (and older children suspected of mental retardation) are given item I and children 8 to 16

Factor Analytic Findings

The Picture Arrangement subtest is a fair measure of g (36 percent of its variance may be attributed to g). It has ample subtest specificity to permit specific interpretation of its functions across the entire age range and a moderate load­ ing on the Perceptual Organization factor (Mdn load­ ing = .41).

Reliability and Correlational Highlights Picture A rrangement is a relatively rel iable subtest

(r.u = . 7 3 ) . It correlates more highly with Block Design (r = .46) than with any other subtest. It has a relatively low correlation with the Full Scale (r = . 55 ) , the Performance Scale (r = . 52), and the Verbal Scale (r = .49) . Administrative and Interpretive Considerations Arrange the Picture Arrangement items from the exam­ inee's left to right in the order given in the manual . As you

CHAPTER 7

1 58

present the demonstration items, be sure not to inadver­ tently cover the picture s with your hand - e xaminees should be able to see all of the pictures and follow your movements when you are rearranging the pictures. Record the examinees' Picture Arrangement sequence as soon as you pick up the cards . Children should be coached if they fail any of the first four items. Children can earn bonus points for speed on items 5 through 12 . In order to help them understand the impor­ tance of speed, encourage them to work quickly. If chil­ dren do not tell you when they are finished, ask them . If necessary, you can add to the directions "Tell me when you are finished . " When children are extremely compulsive , tell them that the cards do not have to be perfectly straight (or aligned) so that they will not be penalized (that is, lose time-bonus points) for their compulsiveness .

Observing performance.

The Picture Arrangement subtest gives you the opportunity to observe how children approach performance tasks involving planning ability. Do they examine the cards , come to some decision, and then reassess it while they arrange the cards (Taylor, 1961)? Or do they proceed quickly without stopping to reconsider their decision? Are their failures due to lack of understand­ ing of the task (revealed by their leaving the pictures in their original order)? What types of errors do they make? (For example, are cards placed in a perfunctory manner or is one card always moved to the same position?) Note persistence, trial-and-error patterns, discourage­ ment, impulsiveness, or rigidity. Compare children's ap­ proaches to the Picture Arrangement items with their ap­ proaches to the Block Design and Object Assembly items. Are the same patterns consistently employed in searching for solutions? If the same patterns are not employed, what might account for the differences? Consider the extent to which task content, fatigue, and mood changes may have influenced examinees' approaches to the various items . Note children's response to any coaching on the first four items (Zimmerman & Woo-Sam, 1985) : Do they appear to grasp the point of the arrangement or story? Does coach­ ing help them to understand the task requirements?

Further inquiry.

Although inquiry into children's ar­ rangements - particularly o n those items they have failed - may elicit useful material, such inquiries are best reserved until testing has been completed, so that stan­ dardization of test administration is safeguarded. Select for inquiry items that may lend insight into an examinee's thought patterns. The items need not be ones the child failed; the fact that an item was correctly arranged does not

WISC-R SUBTESTS

necessarily mean that the child interpreted it correctly. Because the last two items attempted are likely to be the most complex, they may be selected if you have no specific choices. Arrange the Picture Arrangement cards for each item separately, in the order given by the child . Then ask the child to "tell what is happening in the pictures" or to "make up a story" or to "tell what the pictures show. " Consider the following in evaluating the stories (Taylor, 1961) : Are they logical , fanciful, or bizarre? Are they creative or conventional? Are attitudes revealed, such as self-oriented or so­ cially oriented themes? Are incorrect arrangements a consequence of incor­ rect perceptions of details in the pictures or of failure to consider some details? Did the child consider all of the relationships in the pictures? •

•

•

•

•

Testing-of-Iimits.

Useful testing-of-limits procedures include giving the child additional time to complete the arrangement and arranging one or more pictures on items that were failed . Children who solve the problems with the aid of cues may have greater ability than those who fail in spite of additional guidance. Graded help should be intro­ duced only after the standard examination has been com­ pleted, because it has been shown that such help during testing significantly raises Picture Arrangement scores (Sattler, 1969) .

B LOCK D E S I G N

The Block Design subtest contains 11 items, consisting of two-dimensional , red-and-white pictures of abstract de­ signs. (A model is constructed by the examiner for the first two item s . ) Examinees must use red and white blocks to assemble a design identical to that in the picture . Children 6 to 7 years old (and older children suspected of being mentally retarded) begin with item 1; children 8 to 16 years old start with item 3 . The patterns are arranged in order of increasing difficulty - four blocks are used for the first eight designs, and nine blocks are used for the last three designs. All items are timed: items I to 4 are given a maximum of 45 seconds each; items 5 to 8, 75 seconds; and items 9 to 11, 120 seconds. Items 1 to 3 are scored 2 , 1, or 0; items 4 to I I receive 4 points for a correct completion and up to 3 additional time-bonus points for quick execution . The sub­ test is discontinued after two consecutive failures.

1 59

BLOCK DESIGN

Rationale Block design involves the ability to perceive and analyze forms by breaking down a whole (the design) into its component parts and then assembling the components into the identical design, a process referred to as analysis and synthesis. The subtest combines visual organization with the reproductive aspects of visual-motor coordination. Success involves the application of logic and reasoning to spatial relationship problems. Consequently, Block De­ sign can be considered a nonverbal concept formation task requiring perceptual organization, spatial visualization, and abstract conceptualization. It is also a constructional task involving spatial relations and figure-ground separation . Performance may be affected by rate of motor activity and vision. Inadequate performance should not be inter­ preted as direct evidence of inadequate visual form and pattern perception, because the ability to discriminate block designs (that is, to perceive the designs accurately at a recognition level) may be intact even though the ability to reproduce them is impaired .

Place before the examinee the exact number of blocks needed for the item . Children should be instructed to tell you when they have completed each item. You might say "Tell me when you have fi nishe d . " This instruction should be given not only on items 4 through 11, but on items 1, 2, and 3 as well .

Observing performance.

Block Design i s an excellent subtest for observing children's problem-solving ap­ proach . The following issues should be considered: Are the examinees hasty and impulsive or deliberate and careful? Do they give up easily or become disgusted when faced with possible failure, or do they persist and keep on working even after the time limit has been reached? Do children use only one kind of approach, or do they alter their approach as the need arises? Do they use trial and error? Do they study the designs first? Do they have a plan? Do they construct units of blocks, or do they work i n piecemeal fashion? •

•

•

•

•

• •

Factor Analytic Findings The Block Design subtest is the best measure of g among the Performance Scale subtests and is the fourth-best mea­ sure of g among all 12 subtests (53 percent of its variance may be attributed to g) . It has ample subtest specificity to permit specific interpretation of its functions across the entire age range . It contributes substantially to the Percep­ tual Organization factor (Mdn loading = . 66 ) .

Reliability and C orrelational Highlights

Block Design is a reliable subtest (rxx = . 85) . It correlates more highly with Object Assembly (r = . 60) and Picture Completion (r = . 52) than with the other subtests. It cor­ relates moderately with the Full Scale (r = . 68) and the Performance Scale (r = .68) and to a lesser degree with the Verbal Scale (r = . 5 8 ) .

Administrative a n d Interpretive Considerations Be sure that the area being used for block arrangement is clear of other blocks and materials. Construct the demon­ stration design by completing, from the examinee's left to right, first the first row and then the second row. Be careful that your hand does not block the examinee's view. Scram­ ble the blocks before each new design is administered.

Excessive fumbling or failure to check the pattern sug­ gests anxiety. V isual-perceptual difficulties may be indi­ cated if children twist their bodies to improve their per­ spective on the design or if they leave space between the blocks in the assembled design. Try to differentiate be­ tween excessive cautiousness as a personality style and excessivp. slowness as a possible indication of depression.

Testing-of-Iimits.

If testing-of-limits is indicated or de­ sired, the specific item(s) should be presented after the entire examination has been completed . Research has shown that administering a series of cues to children dur­ ing the standard administration of the Block Design subtest significantly raises subtest scores (Sattler, 1969) . A useful procedure to follow in testing-of-limits is to place one block (or one row) in its correct position for those designs that were failed. When you arrange the top row, you might say "Let's try some of these again. I'm going to put together some of the blocks . I will make the top row. Now you go ahead and finish it . Now make one like this. Tell me when you have finished . " If examinees still fai l , additional blocks can be arranged. The amount of help needed to reproduce the designs accurately should be recorded . Children who need many cues may have less developed spatial reasoning ability than those who need few cues.

CHAPTER 7

1 60

Cautionary note. Prior experience with the commercial game Trac 4, which uses block design patterns, has been reported to increase bright lO-year-old children's WISC-R Block Design scores by about 3 scaled score points (Dirks, 1982) . Increases did not occur on other subtests. These results suggest that WISC-R short forms using the Block Design subtest should be avoided if children have played Trac 4, because the scores are likely to be inflated. Stan­ dard IQs may also be slightly inflated if children have played this game.

WISC-R S U BTESTS

is required; and long-term v isual memory (having stored information about the object to be formed) .

Factor Analytic Findings The Object Assembly subtest is a fair measure of g (38 percent of its variance may be attributed to g). Its inade­ quate subtest specificity prevents interpretation of the spe­ cific underlying ability it measures. Since it contributes substantially to the Perceptual Organization factor (Mdn loading = . 65 ) , however, it may be used as a measure of perceptual organization .

OBJ ECT ASSE M BLY

The Object Assembly subtest requires that children put j igsaw pieces together to form common objects : a girl (seven pieces), a car (seven pieces) , a horse (six pieces) , and a face (eight pieces). There is one sample item: an apple (four pieces) . Items are given one at a time, with the pieces presented in a specified disarranged pattern . All children receive all items, beginning with the sample and continuing with items I through 4 . All items are timed . The fi rst item is given a maximum of 120 seconds; the next two items, 150 seconds each; and the fourth item, 180 seconds. Scores for perfect perfor­ mance are 6 points each for the girl and the face and 5 points each for the horse and the car. Bonuses of up to 3 points are awarded for quick performance . The girl , horse, and car items each have a maximum score of 8, and the face item, 9. Points are awarded for partially correct performances.

Rationale The Object Assembly subtest is mainly a test of the exam­ inees' skill at synthesis - putting things together to form familiar objects . It requires visual-motor coordination, with motor activity guided by visual perception and sen­ sorimotor feedback. Object Assembly is also a test of visual organizational ability. Visual organization is needed to produce an object out of parts that may not be immedi­ ately recognizable . In order to solve the jigsaw puzzles , examinees must b e able t o grasp a n entire pattern by anticipating the relationships among its individual parts. The tasks require some constructive ability as well as perceptual skill - children must recognize individual parts and place them correctly in the incomplete figure. Perfor­ mance may also be related to rate and precision of motor activity ; persistence, especially when much trial and error

Reliability and Correlational Highlights O bj e c t A s s e m b l y is a r e l at i v e l y r e l i a b l e s ub t e s t (r..:x = . 70) . I t correlates more highly with Block Design (r = . 60) than with any other subtest. It has a somewhat low correlation with the Full Scale (r = . 56) , a moderate correlation with the Performance Scale (r = . 60) , and a low correlation with the Verbal Scale (r = .46) .

Administrative and Interpretive Considerations Make sure that children do not see the WISC-R manual, which contains pictures of the correctly assembled ob­ jects . The screen used to set up the individual puzzle parts can be used to cover the manual, if desired . lf the screen is used , tape, glue, or staple a piece of cardboard or part of a manila folder to the back side in order to prevent the child from seeing the diagrams on the card. Place the pieces close to examinees so that they do not waste time reaching for the pieces. As in other subtests, you may have to ask examinees to tell you when they are finished. Do not give any cues to the examinees that indicate approval or disap­ proval of their performance .

Observing performance. Object Assembly is an es­ pecially good subtest for observing children's thinking and work habits. Some children envision the complete object almost from the start and either recognize the relations of the individual parts to the whole or have an imperfect understanding of the relations between the parts and the whole. Others merely try to fit the pieces together by trial­ and-error methods. Still others may have initial failure, followed by trial-and-error and then sudden insight and recognition of the object . Observe how children respond to errors and how they handle frustration . Do they demand to know what the

161

CODING

object is before they construct it, or insist that piece are missi ng, or say that the object doesn't make sen e ( Z im­ merman & Woo-Sam, 1985) ? Are low scores due to tempo­ rary inefficiency, such as reversal of two parts, which results in loss of time-bonus credits? A child who spends a long time with one piece, trying to position it in an incor­ rect location, may be revealing anxiety or rigidity.

Testing-of-Iimits.

After the subtest has been completed, inquire about any constructions that appear to be peculiar or unusual (such as pieces placed on top of each other) . Testing-of-limits procedures similar to those described for the Picture Arrangement and Block Design subtests can be used after the entire scale has been administered . A series of graduated cues can be introduced, such as placing one or more pieces in the correct location, and then the amount of help children need to complete the task successfully can be noted. Children needing only a few cues to complete the object may have latent perceptual organization skills. An­ other approach is to see whether asking children to visu­ alize the completed object ( for example, a car) mentally first helps them to assemble the puzzle. Question: What i s gasoline? Answer: To put on the thing what takes your temperature so it don't hurt you . (Flumen & Flumen, 1 979)

CODING

The Coding subtest requires that children copy symbols that are paired with other symbols . The subtest consists of two separate and distinct parts . Coding A is administered to children under 8 years of age , and Coding B to those 8 years of age and over. Each part uses a sample, or key. In Coding A, the sample consists of five shapes - star, circle , triangle, cross, and square . Within each shape, a special mark appears (a vertical line, two horizontal lines , a horizontal line, a circle , and two vertical lines, respec­ tively) . Children are required to place within each shape the mark that appears within it in the sample. There are 5 practice shapes, followed by 43 shapes in the subtest. In Coding B, the sample consists of boxes containing the numbers I through 9 in the upper part and a symbol in the lower part. The stimuli are boxes containing just a number

in the upper part and an empty space in the lower part. Children must write in the space the symbol that is paired with the number in the sample. There are 7 practice boxes, followed by 93 boxes in the subtest proper. The time limit for each Coding task is 120 seconds.

Rationale Coding taps the ability to learn an unfamiliar task and involves speed and accuracy of visual-motor coordination, attentional skills, short-term memory, cognitive flexibility (in shifting rapidly from one pair to another) , and, possi­ bly, motivation. The subtest also involves speed of mental operation (psychomotor speed) and , to some extent, visual acuity. Success depends not only on comprehending the task, but also on using pencil and paper skillfully. Coding B may involve a verbal-encoding process if children attach verbal descriptions to the symbols . For example, a " + " symbol may be labeled as a "plus sign" or "cross" and the "V" symbol as the letter "V" Performance may be enhanced when the symbols are recoded in terms of verbal labels . Consequently, Coding B can also be de­ scribed as measuring the ability to learn combinations of symbols and shapes and the ability to make associations quickly and accurately. Coding A can also involve a verbal-encoding process, but to a lesser degree. Coding A and Coding B thus may involve separate information­ processing modes. The speed and accuracy with which the task is per­ formed are a measure of the child's intel lectual abil ity. At each step in the task the [child] must inspect the next digit, go to the proper location in the table, code the information dis­ tinguishing the symbol found, and carry this information in short-term memory long enough to reproduce the symbol in the proper answer box. (Estes, 1974, p. 745)

Coding thus can be conceptualized as an information­ processing task involving the discrimination and memory of visual pattern symbols .

Factor Analytic Findings The Coding subtest is the poorest measure of g in the scale (17 percent of its variance may be attributed to g) . It has ample subtest specificity to permit specific interpretation of its functions across the entire age range and contributes moderately to the Freedom from Distractibility factor (Mdn loading .42) . =

CHAPTER 7

1 62

Reliability and Correlational Highlights

Coding is a relatively reliable subtest (ra = . 72) . It corre­ lates more highly with Block Design (r = . 33 ) than with any other subtest. It has a low correlation with the Full Scale (r = . 3 8) , the Performance Scale (r = . 33 ) , and the Verbal Scale (r = . 36) .

WISC-R SU BTESTS

Does the child understand the task? Are the child's failures due to inadequate form percep­ tion or poor attention? Does the child check each figure with the samples or remember the samples? Does the child recheck every symbol before moving on to the next one? Does the child pick out one figure only and skip others? Does the child work s moothly, or does he or she seem confused at times? Does the child understand and proceed correctly after explanations have been made? Is the child aware of any errors? Is the child persistent? Does the child need repeated urging? Answers to the above questions will provide valuable information about various characteristics, including atten­ tion span. An increase in speed, coupled with correct copying of symbols , suggests that the child is adjusting to the task well. A decrease in speed, coupled with correct copying of symbols, suggests fatigue. This subtest is par­ ticularly useful for evaluating children's attention when attentional difficulties are suspected , such as after a head injury. Distortion of forms may mean that the child has difficul­ ties with perceptual functioning . Ask about any symbol that is peculiarly written to find out whether it has some symbolic meaning to the child. •

•

•

•

Administrative and Interpretive Considerations The Coding subtest appears in a separate booklet (the last page of the Mazes booklet). It should be administered on a smooth drawing surface . Both you and the child should use a pencil with red lead, without an eraser. One point is allotted for each correct item, with up to 5 additional time­ bonus points given to children with a perfect score on Coding A . There are no time-bonus points for Coding B . Children with visual defects or speci fic motor dis­ abilities may be penalized on this subtest. Generally the subtest should not be given to such children. I f it is given, it should not be counted in the final score. Left-handed children also may be penalized on the Cod­ ing subtest. If the way they write causes them to cover the sample i mmediately above the line of writing , they have to lift their hands repeatedly during the task. Either of two procedures can be used to counteract this difficulty. One method is to show the child the sample from another record booklet. A more satisfactory procedure is to cut the sample from another record booklet, mount it on a strip of card­ board, and then position it for convenient viewing .

Observing performance. Examinees' methods of pro­ ceeding with the task should be observed. Children who stop working after the fi rst line should be told "Continue on the next line . " These instructions should be counted as part of the 2-minute time limit. If examinees skip around, filling in symbols for like shapes or like numbers first, tell them to proceed in order. Useful observational guidelines are as follows: Is the child impulsive or meticulous? Is tremor evident? Does the child's speed increase or decrease as he or she proceeds? Are the child's marks well done , just recognizable, or wrong? Are there any distortions? If distortions are present, do they appear once only, occasionally, or each time the figure appears? How many different figures are affected? Is the child being penalized for lack of speed , for inaccuracy, or for both? •

•

•

•

• ·

•

•

•

•

• •

MAZES

Mazes is a supplementary test; it is not used in the com­ putation of the IQ when the five standard Performance Scale subtests are administered . Mazes consists of nine maze problems and one sample problem . Children are required to draw a line from the center of each maze to the outside without crossing any of the lines that indicate walls. Each maze is presented separately. Children 6 to 7 years old (and older children suspected of mental retarda­ tion) begin with the sample maze, followed by maze I ; children 8 to 1 6 years old start with maze 4 . All items are timed . The first four mazes are given a maximum of 30 seconds each; the fi fth maze , 45 seconds; the sixth maze, 60 seconds; the seventh and eighth mazes, 120 seconds; and the last maze, 150 seconds . The number of errors made determines the child's score . The range of scores is from 0 to 5 points, with mazes I to 3 having a maximum score of 2; mazes 4 and 5 , a maximum score of 3; mazes 6 and 7 , a maximum score of 4; and mazes 8 and

1 63

SUMMARY

9, a maximum score of 5. The subtest is discontinued after two consecutive failures. Although the subtest doe� not have to be routinely administered, administering it to chIl­ dren who are either language-impaired or cul tu ra lly differ­ ent may generate valuable information .

Consider the following questions as you observe perfor­ mance. Does the child understand the task? Does the child study the mazes and extensively plan a route before proceeding? Does the child show signs of tremor, difficulty in controlling the pencil , or difficulty in drawing uniform lines? Does the child solve the mazes correctly after the time limit has expired? Is crossing of lines related to poor visual-motor coor­ dination or to impulsivity? Does the child say anything that suggests anxiety (for example, "The little boy is trapped in the center of the maze")? •

•

•

Rationale

In order to successfully complete the Mazes subtest, chil­ dren must (a) attend to the directions, which include locat­ ing a route from the entrance to the exit, avoiding blind alleys, crossing no lines, and holding pencil on paper; and (b) execute the task , which involves remembering and following the directions , displaying visual-motor coordi­ nation, and resisting the disruptive effect of an implied need for speed (Madden, 1974 ) . The Mazes subtest ap­ pears to measure planning ability and perceptual organiza­ tion (following a visual pattern) . Success requires visual­ motor control and speed combined with accuracy. Factor Analytic Findings

The Mazes subtest is a poor measure of g (20 percent of its variance may be attributed to g) . It has ample subtest specificity to permit specific interpretation of its functions across the entire age range and contributes moderately to the Perceptual Organization factor (Mdn loading = .47 ) . Reliability and Correlational Highlights

Mazes is a relatively reliable subtest (rxx = . 72 ) . It corre­ lates more highly with Block Design (r = . 44) than with any other subtest. It has a low correlation with the Full Scale (r = .44 ) , the Performance Scale (r = .47), and the Verbal Scale (r = . 34) . Administrative and I nterpretive Considerations

The subtest should be administered on a smooth drawing surface. A pencil with black lead is used to demonstrate the sample item, but examinees use pencils with red lead, without erasers. Cues that you should provide after certain errors are made are listed in the manual . When children make these errors for the first time, inform them that an error has been made. The cues should be helpful to children , especially those who do not fully understand the task requirements. A table in the manual clearly shows the number of points allotted for various types of performance. The sample responses , which illustrate the rules regarding the scoring of errors , should help you score children's performance .

•

•

•

Question: Listen, say just what I say : "Eating too much cake and ice cream can give you a stomach ache . " Answer: S o you have t o take a n Alka Seltzer, right? (Adapted from Flumen & Flumen, 1 979)

S U M MARY I . Information measures the child's available information ac­ quired as a result of native ability and early cultural experience. Memory is an important aspect of performance on the subtest. The subtest ties with Similarities as the second-best measure of g and contributes to the Verbal Comprehension factor. Subtest specificity is ample at all ages. Information is a reliable subtest (rxx . 85) and is relatively easy to administer. 2 . Similarities measures verbal concept formation and ties with Information as the second-best measure of g. Subtest speci­ ficity is adequate only for ages 6'12 to 8'/2 years. Similarities is a reliable subtest (ru = . 81), but considerable skill is needed to score some items correctly. 3 . Arithmetic measures numerical reasoning ability. It is a fai r measure o fg , and contributes t o the Freedom from Distractibility and Verbal Comprehension factors. Subtest specificity is ade­ quate at all ages. Arithmetic is a reasonably reliable subtest (ru . 77) and is relatively easy to administer. 4. Vocabulary measures language development, learning abil­ ity, and fund of information. The subtest is an excellent measure of g and contributes to the Verbal Comprehension factor. Subtest specificity is adequate at all ages. Vocabulary is a reliable subtest (r", = . 86), but skill is needed to administer and score it correctly. 5 . Comprehension measures social judgment: the ability to use facts in a pertinent, meaningful , and emotionally appropriate manner. The subtest is a moderately good measure of g and =

=

CHAPTER 7

1 64

contributes to the Verbal Comprehension factor. Subtest specific­ ity is adequate at all ages. Comprehension is a relatively reliable subtest (ra = . 77) , but it requires considerable skill to score. 6. Digit Span is a supplementary subtest that measures short­ term memory and attention; it forms part of the Freedom from D istractibility factor. It is a poor measure of g. Subtest specificity is ample at all ages . Digit Span is a reasonably reliable subtest (rxx = . 78) and is relatively easy to administer. It should be administered routinely, even though it is not used in the computation of the IQ when the five standard Verbal subtests are administered. 7. Picture Completion measures the ability to differentiate essential from nonessential details. It requires concentration, visual organization, and visual memory. The subtest is a fair measure of g and contributes to the Perceptual Organization factor. Subtest specificity is ample or adequate at all ages. Picture Completion is a reasonably reliable subtest (r = . 77) and is relatively easy to administer. 8 . Picture Arrangement measures nonverbal reasoning abil­ ity. It also may be viewed as a measure of planning ability - that is, the ability to comprehend and size up a . total situation. The subtest is a fair measure of g and contributes to the Perceptual Organization factor. Subtest specificity is ample at all ages. Picture Arrangement is a relatively reliable subtest (r = .73) and i s easy to administer. 9. Block Design measures visual-motor coordination and per­ ceptual organization . The subtest is the best measure of g among the Performance Scale subtests and contributes to the Perceptual Organization factor. Subtest specificity is ample at all ages. Block Design is a reliable subtest (rxx = . 85 ) , but skill is needed to administer the subtest correctly. 1 0 . Object Assembly measures perceptual organization abil­ ity. The subtest is a fair measure of g and contributes to the Perceptual Organization factor. Because it has an inadequate amount of subtest specificity, its specific underlying abil ity should not be interpreted. It can be interpreted more generally, however, as a measure of perceptual organization. Object Assem­ bly is a relatively reliable subtest (ru = . 70), but skill is needed to administer it correctly. xx

xx

WISC-R SUBTESTS

I I . Coding measures visual-motor coordination, speed of mental operation, and short-term memory. The subtest is the poorest measure of g in the scale, but it contributes to the Freedom from Distractibility factor. Subtest specificity is ample at all ages. Coding is a relatively reliable subtest (r.... = . 72) and is easy to administer. 1 2 . Mazes is a supplementary subtest that measures planning ability and perceptual organization. The subtest is a poor mea­ sure of g, but it contributes to the Perceptual Organization factor. Subtest specificity is ample at all ages . Mazes is a relatively reliable subtest (r = . 72) and is easy to administer. xx

KEY T E R M S, C O N C EPTS, A N D N A M E S

WISC-R Information (p. 147) WISC-R Similarities (p. 148) WISC-R Arithmetic ( p . 149) WISC-R Vocabulary (p. 151) WISC-R Comprehension (p. 152) WISC-R Digit Span (p. 154) WISC-R Picture Completion (p. 156) WISC-R Picture Arrangement (p. 157) WISC-R Block Design (p. 158) WISC-R Object Assembly (p. 160) WISC-R Coding ( p . 161) WISC-R Mazes ( p . 162)

STUDY Q U EST I O N

Discuss the rationale, factor analytic findings, reliability and correlational highlights, and administrative and interpretive con­ siderations for each of the following WISC-R subtests: Informa­ tion , Similarities , A rithmetic , Vocabulary, Comprehension , Digit Span, Picture Completion, Picture Arrangement, Block Design, Object Assembly, Coding, and Mazes .

I NTE RPRETING TH E WISC-R

Profile Analysis

The first ofour senses which we should take care never to let rust through disuse is that sixth sense, the imagination . . I mean the wide-open eye which leads us always to see truth more vividly, to apprehend more broadly, to concern ourselves more deeply, to be , all our life long, sensitive and awake to the powers and responsibilities given to us as human beings. .

A Successive Level Approach to Test Interpretation

.

Comparisons Between WISC-R Verbal and Performance Scales That Can Guide Interpretations Comparisons Between WISC-R Factor Scores That Can Guide Interpretations

- C hristopher Fry

Comparisons Between WISC-R Subtests That Can Guide Interpretations Analysis of Three Wechsler Subtests from a Piagetian Perspective Estimated Percentile Rarlks and Test-Age Equivalents for Raw Scores Ulustrations of Qualitative Analysis on the WISC-R Comment on Interpreting the WISC-R Psychological Evaluation Test Your SIdlI Summary

1 65

1 66

An understanding of the underlying properties of the WISC-R subtests and scales, how the subtests cluster, Verbal-Performance discrepancies, and patterns of subtest scores will help you in interpreting a child's WISC-R performance . Because the WISC-R and other Wechsler scales have Deviation IQs with a mean of 100 and standard deviation of 15 , and subtests with a mean of 10 and standard deviation of 3 , statistical approaches to eval uating profiles can be readily applied . After the Verbal-Performance discrepancy and profiles have been statistically evaluated, you will need to interpret the findings , particularly when significant differences are established between various sets of scores . The informa­ tion presented in Chapters 6 and 7 forms the basis for such interpretations . The present chapter provides important additional guidelines for interpreting Verbal-Performance differences, making within-scale comparisons, and mak­ ing between-subtest and among-subtest comparisons. In­ terpretation of the WISC-R is also facilitated by a knowl­ edge of child psychopathology, exceptional children, and clinical and psychoeducational approaches to test inter­ pretation . This chapter provides an introduction to these approaches, which are discussed in more detail in Chap­ ters 1 8 , 20, 2 1 , 22, and 23 . As in all clinical and psycho­ educational assessments, a child's performance on any one assessment procedure must be interpreted in light of the child's background and other data obtained during the evaluation.

PRO F I LE A NALYSIS

Profile analysis, or scatter analysis, refers to interpreting or analyzing the pattern of scaled scores and Deviation IQs obtained by an individual examinee . Some profiles show extreme variability (for example, subtest scores that range from 1 to 19), others moderate variability (for example , subtest scores that range from 5 t o 15), and still others minimal variability (for example, subtest scores that range from 8 to 12) . The pattern of subtest scores within each scale can also be examined, as can the relationship (a) be­ tween the Deviation IQs on the two scales and (b) among the three factor deviation scores. In the early days of the Wechsler and Wechsler-type scales, it was hoped that profile analysis would increase diagnostic precision. Unfortunately, research studies have not provided any firm basis for making diagnostic classifi­ cation decisions from profile analysis . Using profile analy­ sis with the WISC-R, WPPSI, and WAIS-R is problematic because the subtests are not as reliable as the Deviation IQs and do not measure unique processes . Still , profiles may

CHAPTER B

I NTERPRETING THE W ISC-R

point out strengths and weaknesses, and these patterns allow for the development of hypotheses that can contrib­ ute to an understanding of the child.

Aim of Profile Analysis The Deviation IQ does not provide any information about the combination of high and low subtest scores earned by a particular child. The goal of profile analysis is to describe the child's unique ability pattern. Determining which tasks a child handles best may help in describing the child's strengths and in planning programs of instruction that util ize those strengths . A flat and significantly above aver­ age profile on an intelligence or special ability test may indicate that the child is gifted and would profit from instruction that capitalized on exceptional intellectual skills. Conversely, a flat and significantly below average profile probably indicates limited intellectual ability. Spe­ cial instructional programs can be tailored to meet the needs of children with this type of profile as well . Profiles with peaks and valleys may be indicative of special strengths and weaknesses and may provide clues about the child's cognitive style and possible remedial directions. The goal of profile analysis is not to classify or cate­ gorize children; rather, it is to find clues about their abili­ ties. Ideas generatedfrom profile analysis must be viewed

simply as hypotheses to be checked against other informa­ tion about the examinee. By clarifying the functional nature of a child's learning problems, profile analysis may assist you in arriving at recommendations for cl inical treat­ ment, educational programs , or vocational placement. Some configurations of subtest scaled scores have great variability. Intersubtest variability may result from tempo­ rary inefficiencie s , permanent incapacity, or disturbed school experiences. Which, if any, of these interpretations is appropriate must be determined on an individual basis. Psychologists believe that profiles showing high variability indicate more potential ability than do profiles showing limited variability (Sattler & Kuncik, 1976) . There is little , if any, research to support this belief, however. In each case, you will have to seek out the best explanation of the child's profile, using all of the test data and the clinical h istory. A profile of subtest scores that is within normal l imits (scaled scores of 8 through 12) should not be considered diagnostic of any exceptionality. Even variability that is outside of "normal limits" does not necessarily indicate the presence of pathology or exceptionality, however. It may simply reflect the child's cognitive style. There is no evi­

dence to support the assumption that pathology and sub­ test variability are necessarily linked.

PROFILE ANALYSIS

Need to Establish Significant Differences In profile analysis, you cannot simply look at twO !>c re and say that one is meaningfully higher than the other. You must determine whether differences represent chance vari­ ation or variation associated with factors other than chance. The procedures used to make such determinations are discussed in the following sections .

Profile analysis is dependent o n the presence of statis­ tically significant differences between the scales, factor scores , or subtests that are being compared. Before mak­ i n g a ny statements about the examinee's cog nitive strengths and weaknesses , you must determine whether (a) the IQs on the Verbal and Performance Scales are significantly different from each other; (b) the subtest scaled scores are significantly different from the mean of their respective scal es; and (c) the subtest scaled scores (of interest) are significantly different from one another. Dif­ ferences among factor scores must be determined to be significant before they can be considered to be meaningful .

Primary Methods of Profile Analysis The methods most commonly used for profile analysis are the following: I . comparing the Verbal and Performance IQs 2. comparing each Verbal Scale subtest scaled score w ith the examinee's mean Verbal Scale scaled score 3 . comparing each Performance Scale subtest scaled score with the examinee's mean Performance Scale scaled score 4 . comparing each subtest scaled score with the mean subtest scaled score based on all subtests administered 5 . comparing sets of individual subtest scores 6. comparing the Verbal Comprehension, Perceptual O rganization , and Freedom from Distractibility factor scores 7 . comparing subtest scaled scores in each factor with their respective mean factor scores Notice that the first and sixth methods involve comparing IQs , whereas the others involve comparing scaled scores. Now let us examine each of these methods in more detail.

Method 1: Comparing Verbal and Performance Scale IQs. Table C-2 in Appendix C gives the differences be­ tween the Verbal and Performance IQs needed to satisfy the .05 and .01 significance levels. Because the differences required to reach the . 05 or .01 significance levels gener­ all y are similar throughout the WISC-R age levels, values

1 67

based on the average of the 11 age groups can be used. The critical values are as follows:

. 05

. OJ

12

IS

Thus a 12-point difference between the Verbal and Perfor­ mance Scale IQs is statistically significant at the . 05 leve l , whereas a IS-point difference is significant a t the .01 leve l . A statistically significant difference reflects a high proba­ bil ity that skill levels in verbal and nonverbal intellectual functioning are different. Exhibit 8-1 describes the pro­ cedure used to obtain the critical values shown in Table C-2 in Appendix C . This procedure can be used to determine the needed critical values for any comparison involving two scales, two subtests, or two tests . (Probabilities asso­ ciated with various Verbal-Performance Scale differences are shown in Table C-8 of Appendix C . )

Methods 2 , 3 , and 4 : Comparing subtest scaled scores with various mean scaled scores. To compare subtest scaled scores with various mean scaled scores, consult Table C-3 in Appendix C. This table shows the deviations from the Verbal Scale mean, Performance Scale mean, and overall mean that are needed for comparisons involving 5 , 6 , 10, 1 1 , o r 1 2 subtests t o reach the . 05 and .01 significance levels. The procedure for using Table C-3 in Appendix C is as follows: I . Write the names of the subtests and their respective scaled scores on a sheet of paper. 2 . Sum the six (or five) Verbal Scale subtests scaled scores. 3 . Compute the mean of the Verbal Scale subtests by dividing the sum of the Verbal Scale subtests by 6 (or 5 ) . 4 . Calculate the deviation from the mean for each Verbal subtest by subtracting each Verbal subtest scaled score from the Verbal Scale mean . Enter these deviations, with the appropriate sign (+ or - ) , opposite the scaled scores. 5. Sum the six (or five) Performance Scale subtests scaled scores. 6 . Compute the mean of the Performance Scale sub­ tests by dividing the sum ofthe Performance Scale subtests by 6 (or 5). 7 . Calculate the deviation from the mean for each Performance subtest by subtracting each Performance

CHAPTER 8

1 68

I NTERPRETING THE WISC-R

'" Exhibit 8- 1 Procedure Used to Determine Whether Two Scores in a Profile Are Significantly Different

In order to establish whether differences between scores in a profile are reliable, it is necessary to apply some statistical procedures to the profile. As always, we cannot be 1 00 percent certain that the differences between any two subtest scores are reliable. A confidence level must therefore be selected, such as a 95 percent level of certainty that the differences are significant. In order to determine whether the difference between two scales or subtests or tests is reliable, the following formula can be used:

whether there is a significant difference between two scaled scores. Let us say that we are interested in determining the value needed to represent a significant difference between the WISC-R Verbal and Performance Scales for children i n the standardization group. The average standard errors of mea­ surement associated with these two scales are 3 . 60 and 4 . 66 , respectively, a s indicated in the WISC-R manual . We know from a normal curve table that at the 95 percent confidence level the z value is J . 96. Substituting these values into the formula yields the following:

Difference Score = Z.JSEmA 2 + SEmB2

Difference Score = J . 96 .J3 . 602 + 4 . 662 = 12

I

I

The Difference Score refers to the magnitude of the difference between scales or subtests A and B. The z refers to the normal curve value associated with the desired confidence leve l . If we select the 95 percent level, the associated z value is 1 . 96. The terms under the square root sign refer to the standard error of measurement associated with each scale or subtest (or test ) . Many test manuals provide these standard errors of measurement. The fol lowing example illustrates how to determine

subtest scaled score from the Performance Scale mean . Enter these deviations, with the appropriate sign ( + or - ) , opposite the scaled scores. 8 . Compute the mean of the overall subtests by divid­ ing the sum by IO (or 11 or 12) . 9 . Calculate the deviation from the overall mean for each subtest scaled score by subtracting each scaled score from the overall mean. Enter these deviations, with the appropriate sign ( + or - ) , opposite the scaled scores . 1 0 . Determine whether the deviations from the Verbal and Performance means are significant by using Table C-3 in Appendix C. The values in Table C-3 reflect significant differences at the . 05 and .01 levels of probability. Be sure to use the appropriate column in Table C-3 to obtain the significant deviations. Column 1 is used when the five standard Verbal Scale subtests are administered, and col­ umn 2 is used when Digit Span is also administered as a sixth Verbal subtest . 1 1 . Place an asterisk next to each subtest deviation that is significantly above or below the mean . 1 2 . After the asterisk write an S to indicate a strength or a W to indicate a weakness.

Differences between these two scales that are at or above this value are significant at the 95 percent level of confidence. A larger difference ( 1 5 ) is needed for the 99 percent confidence level. These values appear in the small rectangle of Table C-2 in Appendix C. All values i n Tables C-2, C- 1 5 , and C-26 in Appendix C were obtained by following the above procedu re . For the 9 9 percent confidence level , t h e z value of2 .58 is used in the equation.

The above steps are illustrated in Table 8-1, which shows how to determine whether the subtest scaled scores differ significantly from the mean of the Verbal Scale scaled scores, Performance Scale scaled scores, and all the scaled scores. In Table 8-1 , none of the Verbal Scale subtest scores differed significantly from the Verbal Scale mean. On the Performance Scale, the Coding score was significantly lower than the Performance Scale mean . With respect to the overall mean, the Information score was significantly lower, whereas Picture Completion and Object Assembly scores were significantly h igher. (In determining whether a scaled score is significantly different from the mean , dis­ regard the plus or minus sign in Table 8-1 . ) You are now in a fi rm position to infer that the difference between the Cod­ ing score and the child's mean Performance Scale scaled score is not a chance difference. A comparison of the subtest scaled scores with the overall scaled score indicates strengths in Picture Completion and Object Assembly and weakness in Information . The critical values used in the preparation of Table C-3 are based on the assumption that the scores on all subtests in a scale are to be compared with the mean score for that

1 69

PROFILE ANALYSIS

Table 8- 1 An Example of Profi le Analysis on the W ISC-R-Comparing Each Subtest Scaled Score to Various Mean Scaled Scores

Deviation from

Subtest

Scaled score

Verbal average

4 7 6 8 7 5

-2.2 + .8 -.2 + 1 .8 + .8 - 1 .2

Information Similarities Arithmetic Vocabulary Comprehension Digit Span Picture Completion Picture Arrangement Block Design Object Assembly Coding Mazes Mean Verbal scaled score :

13

Performance average

-4.3* W - 1.3 -2 3 -.3 - 1 .3 -3.3 +2.2 - 1.8 + 1 .2 +3.2 - 4 . 8* W

9

12 14 6

Overall average

+4.7* S + .6 +3.7 + 5 .7* S -2.3

37 = 6.2 6

54 Mean Performance scaled score : 5

Mean overall scaled score: 2.l II

=

=

10. 8

8.3

* Significant at the .01 level . = strength, W = weakness. See Table C-3 in Appendix C to obtain deviations that are significant.

Nore. S

scale. Therefore only one significance level (either . 05 or .01) should be used to determine the critical values ; mixing levels of significance is not recommended (A. B. Silver­ stein, personal communication , 1980) .

Method 5 : Comparing sets of individual subtest scores. Table C-2 in Appendix C shows the differences between sets of scaled scores on the 12 WISC-R subtests thaI are required for the average of the 1 1 age groups to reach the .05 and .01 significance levels. The table indicates, for example, that a 4-point difference between Information and Similarities is significant. When multiple comparisons are made between the subtests , the values in Table C-2 are overly l iberal (that is, lead to many significant differ­ ences) . The values are more accurate when a priori planned comparisons are made of pairs of subtests , such as Information versus Comprehension or Digit Span versus Arithmetic . The procedure described in Exhibit 8-1 was

followed in obtaining the needed critical values between subtests . Before multiple subtest comparisons are made (that is, more than one set of subtest scores is compared) , the difference between the highest and lowest subtest scores should be determined (A. B. Silverstein, personal commu­ nication, 1980) . A difference of 7 scaled-score points be­ tween the highest and lowest scaled scores is significant at the . 05 level . Differences that are 7 scaled-score points or greater can then be interpreted . If the difference between the highest and lowest subtest scaled scores is less than 7 scaled-score points, multiple comparisons between indi­ vidual subtest scores should not be made .

Method 6: Comparing Verbal Comprehension, Percep­ tual Organization, and Freedom from Distractibility factor scores. Table C-2 in Appendix C presents the differences between sets of Verbal Comprehension , Per-

1 70

ceptual Organization, and Freedom from Distractibility factor scores (in the form of Deviation IQs) that are needed to reach the . 05 and .01 significance levels. Chapter 6 discusses how factor scores are obtained . Because the Verbal and Performance Scale IQs contain subtests that are also contained in the Freedom from Distractibility factor, the Freedom from Distractibility factor score should be compared with the Verbal Comprehension and Perceptual Organization factor scores and not with the Verbal Scale and Performance Scale IQs . If the Verbal Scale and Perfor­ mance Scale IQs were used , the comparison would involve overlapping subtests.

Method 7: Comparing subtest scaled scores in each factor with their respective factor scores. It is useful to compare each scaled score to the mean of its respective factor score . This procedure allows you to determine whether the subtests that comprise each factor signifi­ cantly differ from the mean factor scaled score. Consult Table C-5 in Appendix C to obtain the critical values. For Verbal Comprehension, they range from 2 . 58 to 2 . 93 at the 5 percent level ( 3 . 12 to 3 . 54 at the I percent level). For Perceptual Organization, they range from 2 . 78 to 3 . 52 at the 5 percent level (3 . 36 to 4 . 26 at the I percent level). For Freedom from Distractibil ity, they range from 2 . 80 to 3 . 04 at the 5 percent level (3.43 to 3 . 72 at the I percent level) . The steps involved in this procedure are similar to those described in the previous section using the Verbal and Performance subtests (see methods 2 and 3 ) .

Other Approaches t o Profile Analysis Supplementary approaches to profile analysis examine the kinds of variability found in the normative group, allowing features of an individual child's profile to be compared with those of the normative group. The three base rates de­ scribed below can be used for examining different kinds of variability.

Base rate subtest scaled-score ranges.

In the standard­ ization group, the median scaled-score range - that i s , the size of the difference between the highest and lowest WISC-R subtest scaled scores - was 7 points for the Full Scale , 4 points for the Verbal Scale, and 5 points for the Performance Scale (Kaufman, 1976c ) . These median ranges were similar across age groups, sexes, races, pa­ rental occupations , and intelligence levels. The range scaled-score index is not very helpful because it is difficult to interpret and little research is available to guide its interpretation . It deals with only 2 scores, and therefore fails to take into account the variability among all 10 (or 11

CHAPTER 8

INTERPRETING THE WISC-R

or 12) subtest scores. The range index should not be dis­ carded, however, because it provides base rate information about what occurred in the standardization sample.

Base rate differences between each subtest score and an average subtest score in the WISC-R standardization sample. Table C-7 in Appendix C gives the frequencies with which various differences between subtest scores and average WISC-R Verbal , Performance, or overall scores occurred in the standardization sample . The table shows, for example , that a difference as large as 3 . 4 points be­ tween the scaled score on Information and the Verbal Scale average was obtained by 5 percent of the standardization sample. This table should be used only when differences have been found to be statistically significant . (See num­ bers 2, 3, and 4 in the preceding section , Primary Methods of Profile Analysis . ) Differences of approximately 3 to 5 points between each subtest score and the respective aver­ age Verbal Scale or Performance Scale score were ob­ tained by 5 percent of the standardization sample.

Base rate Verbal-Performance differences (the prob­ ability-of-occurrence approach) . Determi ning how frequently a Verbal-Performance IQ difference of a given magnitude occurred in the standardization sample is re­ ferred to as the probability-of-occurrence approach. The frequencies with which a variety of Verbal-Performance discrepancies occurred in the normative standardization sample are given in the expectancy table in Table C-9 of Appendix C. The table shows, for example , that a 14 . 01 difference in either direction between the Verbal and Per­ formance IQs occurred among 25 percent of the children in the standardization sample. The relationship of Verbal-Performance discrepancies to IQ level and various demographic variables has been determined for the standard ization sample . Verbal ­ Performance I Q differences were not significantly related either to sex or to race, but they were significantly related to intelligence level and parental occupation (Kaufman , 1976d) . The median discrepancy for the total group was 8 IQ points (Kaufman, 1976d) . A lthough brighter children (IQs of 110 and above) had higher discrepancies (Mdn discrepancy = 9 IQ points) than did less able children (IQs of 89 and below) (Mdn discrepancy = 6 to 7 points), neither the Verbal Scale IQ nor the Performance Scale IQ tended to predominate at any level of intelligence . Chil­ dren of professional parents , however, tended to have higher Verbal than Performance Scale IQs, and children of semiskilled and unskilled workers tended to have higher Performance than Verbal Scale IQs .

171

A SUCCESSIVE LEVEL APPROACH TO TEST INTERPRETATlm �

C omment o n the Verbal-Performance difference ap­ proaches. In evaluating Verbal-Performance Scale dif­ ferences, it is important to determine whether the differ­ ence is likely to have occurred solely by chance (referred to as the reliability-of-difference approach ) . Differences that are not significantly different from those likely to occur by chance should not be interpreted . Verbal-Performance Scale differences may be signifi­ cant and yet occur with some frequency in the population . Thus the discrepancy may be reliable but not unusual . Whether a significant difference has practical significance is open to question. Although a statistically significant difference may have less diagnostic relevance if it occurs in a large proportion of the standardization sample (this is where the base rate information about Verbal-Performance Scale differences may be helpful in interpretation) , it is stil l likely to reflect different skili levels in verbal and nonverbal intellectual functioning . The reliability-of-difference approach is based on the standard error of measurement of each scale, whereas the probability-of-occurrence approach is dependent on the correlation between the two scales . Both of these ap­ proaches are aids to clinical judgment and should not be

used in a mechanical fashion or as a replacement for clinical judgment. When a reliable (significant) difference occurs between the child's Verbal and Performance Scale scores, hypoth­ eses about the child's cognitive strengths and weaknesses should be formulated. Such discrepancies may provide a meaningful profile of abilities, even though they may 0 c U i in a large segment of the population . Any significant dis­ crepancy should be interpreted in light of all the data that you have obtained about the examinee.

Comment on Profile Analysis The assignment of statistical significance to a difference between subtests or scales tells us that the difference is large enough that it cannot be attributed to measurement error. The WISC-R manual recommends using the . 15 level of significance as the minimum level for determining whether there are significant differences between scaled scores and between Verbal and Performance Scale IQs. In contrast, I recommend that the . 05 level of significance be used in order to reduce the chances of making a Type I error. By using the more stringent confidence level, one can avoid accepting differences between scores as true differences when they are not in fact true differences . T h e W I S C - R manual a l s o states that a Ve rba l ­ Performance difference of 15 points o r more calls for "further investigation . " No further information is given

about how to follow up, however. Should a child be admin­ istered additional special tests or be referred for neurologi­ cal examination? The fact that a Verbal-Performance Scale difference that is significant at the . 05 or .01 level is somewhat unusual does not necessarily mean that there is a need for further investigation. A child's entire perfor­ mance and clinical history should always be evaluated carefully before further investigation is recommended. The difference between an examinee's subtest score and the mean scaled score is a more stable measure than the difference between pairs of subtest scores. Use of the mean scaled score has the additional advantage of reducing the chance of errors associated with multiple comparisons . Several factors may account for a profile of scores. These include the child's cognitive skill development, age , sex, racial or ethnic group , socioeconomic status, educa­ tion, special training, social and physical environment, family background, nationality, temperament, personality, and psychopathology. The possibility also must be consid­ ered that the variability of scores is simply a reflection of the unreliability of the individual subtest scores, examiner variability, or situational variability. Profile analysis is a useful tool for comparing intrain­ dividual differences in various ability and achievement areas . Profile differences may represent only uneven skill development, however; taken alone, they do not constitute a sufficient basis for making decisions about pathological conditions or about possible causes of and cures for the uneven development. Profile analysis should be viewed as a clinical tool to be used in conjunction with other assess­ ment strategies.

A SUCCESSIVE LEVE L APPROACH TO TEST I N T E R PRETAT I O N

A successive level approach is useful in integrating the various kinds of information obtained from Wechsler scales. The six levels of analysis are as follows: I. The Full Scale IQ. The first level focuses on the Full Scale IQ. In many cases the Full Scale IQ is the most reliable and valid estimate of intellectual ability provided by the scale . It is the primary numerical and quantitative index , providing information about the child's relative standing in the general population, as represented by the standardization group. The Full Scale IQ is a global esti­ mate of the child's level of cognitive ability. The child's classification level is based on the Full Scale IQ (see Table BC-2 on the inside back cover). Converting the Full Scale IQ to a percentile rank (see Table BC-l on the inside back

1 72

cover) will help you to interpret this score for personnel who may not be familiar with standard scores. IIa. Verbal and Performance IQs. The second level focuses on the Verbal and Performance IQs and the extent to which there is a significant difference between the two. The Verbal Scale IQ provides information about verbal comprehension skills, whereas the Performance Scale IQ covers perceptual organization skills (see Chapter 6) . Table BC-l can also be used to obtain the percentiles associated with the Verbal and Performance Scale IQs . lib. Factor scores. An alternative second level pro­ cedure is to focus on the factor scores and the extent to which there are significant differences among the three factor scores. The Verbal Comprehension factor (informa­ tion, Similarities, Vocabulary, and C omprehension) pro­ vides a somewhat purer measure of verbal comprehension than does the Verbal Scale IQ , whereas the Perceptual Organization factor (Picture Completion, Picture Ar­ rangement, Block Design, and Object Assembly) provides a somewhat purer measure of perceptual organization than does the Performance Scale IQ . The Freedom from Dis­ tractibility factor (Arithmetic , Digit Span, and Coding) measures a variety of complex processes involving atten­ tion , concentration , and problem-solving strategies (see the discussion in a later section on factor scores) . When the three subtests comprising Freedom from Distractibility significantly deviate from their scale in the same direction , either positively or negatively, you are o n relatively firm ground in comparing the three factor scores . III . Subtest variability within scales. The third level focuses on deviations of the various subtests from the mean of the Verbal Scale or Performance Scale . Hypotheses about strengths and weaknesses can be developed from these analyses. IV Intersubtest variability. The fourth level focuses on comparisons between sets of subtests or among clusters of subtests . Although these comparisons are open to the errors associated with multiple comparisons, they are val­ uable for generating hypotheses. (Table CAl in Appendix C provides the percentile ranks for the subtest scaled scores . ) V Intrasubtest variability. The fi fth level focuses on the pattern of performance within each individual subtest . Because the items are arranged in order of difficulty, pat­ terns in which successes and failures are interspersed need to be evaluated carefully. For example, a child who passes the first item, fails the next four, passes the next one, fails the next four, and overall passes a total of four items is showing a different pattern from one who passes the first four items and fails the remainder, although both children receive 4 raw-score points. The child with the markedly

CHAPTER 8

INTERPRETING THE WISC-R

uneven pattern may have cognitive or attentional inefficien­ cies that should be explored further. VI . Qualitative analysis. The sixth level focuses on specific item failures and the content of the responses, or qualitative analysis. Inspecting responses to spec ific items can aid you in understanding the child's knowledge of specific information, such as knowledge of the concept half on the Arithmetic subtest or knowledge of the four seasons of the year on the Information subtest . Careful attention to unique or highly personal responses may be especially informati ve . Both verbal and nonverbal re­ sponses should be evaluated . For example, a child with paranoid tendencies may give querulous, distrustful, or legalistic responses , whereas one who is depressed may give slow, hesitant , and blocked responses, interspersed with self-deprecatory remarks. Qualitative aspects of test interpretation are further discussed toward the end of this chapter and in Chapter 18 .

C O M PARISONS BETW E E N WISC·R VERBAL AND P E R F O R M A N C E SCALES THAT CAN G U I D E I NT E R PRETAT I O N S

The evaluation o f the WISC-R Verbal and Performance Scales depends primarily on the hypotheses made about the individual subtests that comprise the respective scales. Some general observations can be made concerning the two scales, however.

Interpretation of Verbal and Performance Scales The Verbal Scale is dependent on the child's accumulated experience . It usually requires the child to give a response that is in his or her repertoire. The questions (input) are presented verbally, and the responses (output) are given orally. The Verbal Scale might be considered to be an index of verbal ability and crystallized intelligence . The Performance Scale, in contrast , is more dependent on the child's immediate problem-solving ability. It re­ quires the child to meet new situations and apply past experience and previously acquired skills to a new set of demands . The stimuli (input) are nonverbal , and most are presented visually. Solutions (output) require motor re­ sponses and , to a lesser extent, verbal responses. The Performance Scale might be considered to be an index of nonverbal ability and fluid intelligence . Solving problems on the Verbal and Performance Scales involves both verbal and nonverbal strategies. After exam­ ining the extent to which the Verbal Scale subtests involve visualization or other nonverbal processes and the extent

COMPARISONS BETWEEN WISC-R FACTOR SCOR ES THAT C A N GUIDE INTERPRETATIONS

to which the Performance Scale subtests involve language activity in the form of overt verbal responses or mediating symbolic activity, Lawson and Ingl is (1983) concluded that some Verbal subtests are more verbal than others , and some Performance subtests are more nonverbal than others . Within the Verbal Scale , some solutions used by exam­ inees on the Arithmetic and Digit Span subtests may in­ volve visualization strategies to a greater extent than do solutions on other verbal subtests. Within the Performance Scale , solutions on the Block Design and Object Assembly subtests may involve the ability to visualize configurations in space and therefore may not depend greatly on verbal processing . In contrast, performance on the Coding , Pic­ ture Completion, and Picture Arrangement subtests may depend on both verbal and nonverbal processing . Perfor­ mance on Coding B depends on the ability to learn associa­ tions between digits and symbols that can be encoded verbally. Performance on Picture Completion depends on a knowledge of "the way the woild is," and this knowledge may be imparted by verbal means. Consequently, there are no pure tests of either verbal or nonverbal ability on the WISC-R and other Wechsler scales. Table C-42 in Appen­ dix C presents a summary of the interpretive rationales, possible implications of high and low scores, and instruc­ tional impl ications for the Verbal , Performance, and FuJI Scales and for the Freedom from Distractibility factor score.

Formulating Hypotheses About Verbal­ Performance Discrepancies Significant Verbal and Performance Scale differences may indicate the following : interest patterns cognitive style psychopathology (such as emotional disturbance or brain damage) deficiencies or strengths in processing information deficiencies or strengths in modes of expression deficiencies or strengths in the ability to work under time pressure (such as time constraints on the Performance Scale) sensory deficiencies • •

•

•

•

1 73

ample , that you would not say that a child with a Verbal Scale IQ of ISO and a Performance Scale IQ of 125 had a performance deficit. In this case both abilities are well developed, with verbal skills even better developed than performance skills. Similarly, a child with a Verbal Scale IQ of 70 and a Performance Scale IQ of 50 should be viewed as having deficits in both verbal and nonverbal areas; the higher Verbal Scale IQ of 70 should not be considered an absolute strength.

1. Illustrative Hypotheses for Verbal > Performance: a. Verbal skills are better developed than performance skills. b . Auditory-vocal processing skills are better devel­ oped than visual-motor discrimination skills. c . Knowledge acquired through accumulated experi­ ence is better developed than immediate problem-solving ability. d. The examinee may have difficulty with practical tasks . e . Performance deficits may exist , including deficits in copying skill s . f . Limitations in visual-motor integration may b e influ­ encing performance. g . Difficulties may be experienced in performing speeded tasks . 2.

Illustrative Hypotheses for Performance > Verbal:

a. Performance skills are better developed than verbal skills. b. Visual-motor discrimination skills are better devel­ oped than auditory-vocal processing skills . c. Immediate problem-solving ability is better devel­ oped than knowledge acquired as a result of accumulated experience. d. The examinee may have difficulty with reading and academic achievement. e. A language deficit m'lY exist . f. Limitations in auditory conceptual skills and auditory processing skills may be influencing performance. g . Difficulties may be experienced in working effec­ tively without time pressure.

•

•

Which interpretations are pertinent must be decided on the basis of the child's entire performance and clinical history. Hypotheses about Verbal-Performance Scale differ­ ences should be formulated in relationship to the child's absolute Verbal , Performance, and Full Scale IQs and only when the differences are significant . This means, for ex-

COM PARISONS BETW E E N WISC·R FACTOR SCORES T H AT CAN G U I D E I NTERPRET ATI O N S

Only when factor scores are significantly different from each other should hypotheses be formulated about a child's strengths and weaknesses (see Table C-2 in Appendix C ) . The frequencies with which given discrepancies between

CHAPTER 8

1 74

the WISC-R factor scores occur in the standardization sample are shown in Table 8-2 . A summary of interpretive rationales for Wechsler factor scores can be found in Table C-42 in Appendix C . Interpretations o f the Verbal Comprehension and Per­ ceptual Organization factors are similar to those of the Verbal and Performance Scales, respectively. Interpreta­ tion of the Freedom from Distractibility factor is based on a number of processes that may underlie performance on this factor. A high Freedom from Distractibility score suggests the ability to sustain attention, good short-term memory, numerical ability, encoding ability (such as se­ quencing ability) , use of rehearsal strategies, ability to shift mental operations rapidly on symbolic material , and ability to self-monitor. A low Freedom from Distractibility score suggests difficulty in sustaining attention, distrac­ tibility, anxiety, short-term retention deficits, encodi ng deficits, poor rehearsal strategies, difficulty in rapidly shifting mental operations on symbolic material , and inad­ equate self-monitoring skills. A careful analysis of the examinee's entire performance on the scale, other test scores, and case history information is required to deter­ mine the best explanation of the examinee's performance. The list of processes above provides a useful starting point for developing hypotheses about Freedom from Distrac­ tibility scores. The term Freedom from D i stractibi l ity may over­ simplify the complex processes underlying performance on this factor and therefore may not be appropriate (Ownby & Matthews, 1985). In some cases , distractibility may not be the key factor interfering with performance on the subtests that comprise this factor. Digit Span, Arith­ metic , ·and Coding may involve , to some extent, the use of efficient task strategies to solve problems. These subtests

INTERPRETING THE WISC-R

are not the only ones that may require efficient task strat­ egies, however. Block Design , Picture Arrangement, and Object Assembly are also likely candidates. Further re­ search will be needed to clarify what is measured by the third WISC-R (and WAIS-R) factor. If Arithmetic , Digit Span, and Coding scores are not consistent (for example, the Digit Span score is high and Arithmetic and Coding scores are low) , the Freedom from Distractibility factor score may be especially difficult to interpret. If these three subtest scores are consistently low or high, more confidence can be placed in interpretation of the Freedom from Distractibility factor. Factor scores should not be included in a psychological report . They should simply be used in generating hypoth­ eses about the child's performance .

COM PARISONS BETW E E N WISC-R SU BTESTS THAT CAN G U I D E INTERP RETAT I O NS

Once differences between subtest scaled scores have been found to be significant (see Table C-2 in Appendix C), the findings must be translated into meaningful descriptions . Interpreting differences between subtest scores is not an easy matter. The material in Chapter 7 and in this chapter will help you in making interpretations. Table 8-3 charts suggested abilities and factors associated with the 12 WISC-R subtests. You should view the interpretations that follow as hypotheses that may prove to be useful in the evaluation of a child's performance. These hypotheses, however, need to be further investigated through a study of the child's entire test performance and clinical history. The following list of subtest comparisons does not in-

Table 8·2 Percentage of Population Obtaining Discrepancies Between WISC·R Factor Deviation Quotients (DQ)

Percentage obtaining given or greater discrepancy in either direction 50 20 10 5 2

Freedom from Distractibility DQ vs. Verbal Comprehension DQ

Freedom from Distractibility DQ vs. Perceptual Organization DQ

Verbal Comprehension DQ vs. Perceptual Organization DQ

Percentage obtaining given or greater discrepancy in a specific direction

9. 1 8 1 7.47 22.41 26. 7 1 3 1 . 69 35 . 1 0

1 0 . 39 1 9 . 77 2 5 . 36 30.22 3 5 . 86 3 9 . 72

8.38 1 5 . 94 20.45 24. 3 6 28.91 32 .02

25 10 5 2.5 1 0.5

Source: Adapted from Clampit, Adair, and Strenio ( 1 983).

COMPARISONS BETWEEN WISC-R SUBTESTS THAT CAN GV OE INTERPRETATIONS

1 75

Table 8-3 Suggested Abilities or Factors Associated with the 1 2 WISC-R Subtests

s: s::

.g

�

..g, ..:;

� "" .� .E: -S:! ·E

D S : H igh Similarities and low Digit Span may reflect good conceptual izing ability coupled with pocr rote auditory memory for digits . S < D S : Low Similarities and high Digit Span may suggest poor conceptualizing ability coupled with good rote auditory memory for digits .

COMPARING SUBTESTS IN VERBAL AND PERFORMANCE SCALES

9. Similarities (S) and Block Design (BD). Similarities and Block Design both reflect abstract reasoning ability. They both require the abstraction of relations among stim­ ulus items . S > B D : High Similarities and low Block Design may suggest that abstract reasoning ability is better with verbal stimuli than with nonverbal materials . S < B D : Low Similarities and high Block Design may suggest that abstract reasoning ability is better with non­ verbal materials than with verbal stimuli . 1 0 . Comprehension (C) and Picture Arrangement (PA). The Comprehension and Picture Arrangement subtests both contain stimuli that are concerned with social interac­ tion . The comparison relates knowledge of social conven­ tions (Comprehension) to the capacity to anticipate and plan in a social context (Picture Arrangement) . PA > C : H igh Picture Arrangement coupled with low Comprehension may suggest sensitivity to interpersonal nuances, but a disregard for social conventions. PA < C : Low Picture Arrangement coupled with high Comprehension may suggest an understanding of social situations in the abstract, but difficulty in deciding what the situations mean or what actions to take once the examinee is involved in the situation s . 1 1 . Picture Completion (PC) and Arithmetic (A). Pic­ ture Completion and Arithmetic both involve concentra­ tion . On Picture Completion, however, the concentration is directed to an externalized form - a visual stimulus ­ whereas on Arithmetic the concentration is directed to an internalized stimulus - a memory trace .

1 77

PC > A: High Picture Completion and low Arithmetic may suggest adequate concentration for a visual stimulus , but not for an auditory stimulus. PC < A : Low Picture Completion and high Arithmetic may suggest adequate concentration for an auditory stim­ ulus, but not for a visual stimulus . COMPARING SUBTESTS IN PERFORMANCE SCALE

1 2 . Picture Completion (PC) and Picture Arrangement (PA). Th is comparison provides an estimate of attention to detail versus organization of detail . Both Picture Comple­ tion and Picture Arrangement involve perception of de­ tails, but Picture Arrangement also requires logical order­ ing of details, or sequencing . PC > PA : High Picture Completion and low Picture Arrangement may suggest that perception of details is better developed for nonsequencing tasks than for tasks requiring sequencing and organization . PC < PA : Low Picture Completion and high Picture Arrangement may suggest that perception of details is less well developed for nonsequencing tasks than for tasks requiring sequencing and organization .

T H E: N UM8ERj 4'1 AN D 1 2 1 A � E A L I KE" BECAvSE I F YOV MULTI PL.Y THEM BOTH BY 3 , 4, 5, OR. 6, THE SVM O F TH E I R D I G I T S I S THE SAME. U K " :3 T I M E r 4 9 1 5 1 4 7 . I P L v S 4 PL u 5 7 15 1 2. . 3 1 1/'-{ e S 1 2 1 1 5 .3 6 3 . :3 PLv.! " PL vS 3 1 5 A L S O 1 2 . IF YOU U5E 4, • • •

H E RM

Courtesy Herman Zielinski.

1 78

1 3 . Picture Completion (PC) and Block Design (BD) . This comparison relates visual perception to visual-motor­ spatial coordination . PC > BD: H igh Picture Completion and low Block Design may suggest adequate nonspatial visual perceptual ability but inadequate spatial visual ization abil ity. PC < BD: Low Picture Completion and high Block Design may suggest inadequate nonspatial visual percep­ tual ability but adequate spatial visualization abil ity. 1 4 . Object Assembly (OA) and Picture Arrangemenr (PA). This comparison relates inductive reasoning (work­ ing from parts to a whole) to sequenc ing . Both tasks require synthesis into wholes without a model, but Picture Arrangement also involves sequencing . OA > PA : High Object Assembly and low Picture Ar­ rangement may suggest that visual inductive reasoning skills are better developed than visual sequencing skills. OA < PA : Low Object Assembly and high Picture Ar­ rangement may suggest that visual inductive reasoning skills are less well developed than visual sequencing ski l l s . 1 5 . Block Design (BD) , Object Assembly (OA) , and Mazes (MA). The Block Design, Object Assembly, and Mazes subtests require visual-motor coordination and in­ volve motor activity guided by perceptual organization. The role of visual organization differs in the three subtests. On the Block Design subtest , visual organization is in­ volved in a process of analysis (breaking down the pattern) and synthesis (building the pattern up again out of the blocks ) . On the Object Assembly subtest, visual organiza­ tion is used in the arrangement of parts into a meaningful pattern . On Object Assembly, children may develop a mental concept of the whole from parts without the visual representation of the whole , whereas on Block Design the drawing of the whole is provided . On the Mazes subtest, visual organization is involved in planning and foresight . Thus the term visual organization refers to different func­ tions on each of these three subtests. BD > M A : High Block Design and low M azes may suggest that visual organization skills involving analysis and synthesis are better developed than those involving planning and foresight . BD < M A : Low Block Design and high Mazes may suggest that visual organization skills involving analysis and synthesis are less well developed than those involving planning and foresight . OA > BD: High Object Assembly and low Block De­ sign may suggest that nonverbal inductive reasoning s kills (working from parts to a whole) are better developed than nonverbal deductive reasoning skills (working from a

CHAPTER 8

INTERPRETING T H E WISC-R

whole to parts) . Additionally, it may suggest d i fficulty in interpreting figure-ground relationships. OA < BD: Low Object Assembly and high Block De­ sign may suggest that nonverbal inductive reasoning skills are less well developed than nonverbal deductive reason­ ing skills, or that a deficit exists in the ideation required to form a visual representation of the objects .

COMPARlNG THREE OR MORE SU BTESTS

1 6 . Information (I) , A rithmetic (A) , and Comprehen­ sion (C) versus Similarities (S) , Vocabulary (V) , and Digit Span (DS). This comparison contrasts subtests that have relatively long verbal questions (Information, Arithmetic , and Comprehension) with those that have relatively short verbal questions (Similarities, Vocabulary, Digit Span) (Kaufman , 1979a) . I, A, C > S , V, DS: High Information , Arithmetic, and Comprehension coupled with low Similarities, Vocabu­ lary, and Digit Span may suggest that the child performs better when the verbal stimuli are long than when they are short . The child may put forth more effort to attend to verbal material that is of a relatively long duration. The pattern may also reflect the child's ability to benefit from the contextual cues contained in the longer questions. I, A, C < S, V, DS: Low Information, Arithmetic , and Comprehension coupled with high Similarities , Vocabu­ lary, and Digit Span may suggest that the child performs better when verbal stimuli are short than when they are long. The child may put forth more effort to attend to verbal material that is of a relatively short duration. This pattern also may suggest an auditory reception deficit associated with deriving meaning from spoken language . 1 7 . Similarities (S) , Vocabulary (V) , and Comprehen­ sion (C) versus Information (I) , A rithmetic (A) , and Digit Span (DS). This comparison contrasts subtests that require a fair amount of verbal expression (Similarities , Vocabu­ lary, Comprehension) with those that require relatively little verbal expression (Information, Arithmetic , Digit Span) (Kaufman, 1979a) . S , Y, C > I , A, DS: High Similarities, Vocabulary, and Comprehension coupled with low Information , Arith­ metic, and Digit Span may suggest that the child performs better when the tasks require a fair amount of verbal expression than when they require relatively l ittle verbal expression. One possibility is that the child may put forth the extra effort in situations that require verbal expression but not in those that require minimal verbal expression.

ANALYSIS OF THREE WECHSLER SUBTESTS FROM A

PI.A GE TIA�,

PERSPECTIVE

S , V, C < I, A , DS : Low Similarities, Vocabu lary. and Comprehension coupled with high Informatio n , Arith­ metic, and Digit Span may suggest that the child performs better when the tasks require relatively little verbal ex­ pression than when they require a fair amount of verbal expression. One possibility is that the child may put forth e ffort only when tasks require minimal verbal effort . Addi­ tionally, this pattern may be associated with communica­ tion problems or shyness associated with speaking in rela­ t ively long sentences . 1 8 . Picture Completion (PC) , Pir.tu re A rrangemel1t

(PA) , and Object Assembly (OA) versus Block Design (BD) and Coding (CD). This comparison contrasts subtests that contain relatively meaningful perceptual stimuli (Picture C ompletion, Picture Arrangement, Object Assembly) with those that have relatively abstract perceptual stimuli (Block Design , Coding) (Kaufman, 1979a) . PC , PA , OA > BD , CD: H igh Picture Completion, Picture Arrangement, and Object Assembly coupled with low Block Design and Coding may suggest that the child performs better when the visual stimuli are meaningful than when they are abstract or nonmeaningful . PC , PA, OA < BD, C D : Low Picture Completion , Pic­ ture Arrangement, and Object Assembly coupled with high Block Design and Coding may suggest that the child performs better when the visual stimuli are abstract than when they are concrete . 1 9 . Picture Completion (PC) and Picture Arrangemel1l

(PA) versus Block Design (BD) , Object Assembly (OA) , Coding (CD), and Mazes (MA). Five of the six Pcrfor

mance Scale subtests involve perceptual organization (the exception is Coding) . This comparison of the Performance Scale subtests distinguishes those subtests that tap pri­ marily perceptual organization (Picture Completion and P icture Arrangement) from those that involve both percep­ tual organization and v isual-motor coordination (Block Design , Object Assembly, Coding , Mazes) . PC, PA > BD, OA , C D , MA: H igh Picture Completion and Picture Arrangement coupled with low Block Design, Object Assembly, Coding, and Mazes may suggest that the child performs better when the tasks requi re only percep­ tual organization than when they require both perceptual organization and visual-motor coordination. PC , PA < BD, OA, CD, MA: Low Picture Completion and Picture Arrangement coupled with high Block Design, Object Assembly, Coding, and Mazes may suggest that the child performs better when the tasks require both percep­ tual organization and visual-motor coordination than when they require only perceptual organization .

1 79

ANALYSIS OF T H R E E WECHSLER SUBTESTS FROM A PIAGETIAN P E RSPECTIVE

Piaget's distinction between figurative knowing and oper­ ative knowing may be used in analyzing performance on the Wechsler scales and other cognitive tests. Figurative knowing refers to knowledge of static states (knowledge of content, such as names of objects), whereas operative knowing refers to knowledge of dynamic transformations (knowledge of how to apply content to arrive at a solution , such as putting together three sticks to reach an object) . This distinction appears to be similar to the crystallized/ fluid distinction in the Cattell/Hom approach (see Chapter 3 ) . Elkind (1979) applied the figurative/operative distinc­ tion to three WISC-R subtests: Information, Similarities, and Block Design .

Information The items on the Information subtest are cognitively quite heterogeneous. Some items assess figurative learning, oth­ ers operative learning, and still others openness to learning extraneous bits of information . For example, "What do you call this finger?" requires simple figurative knowl­ edge, whereas "How many ears do you have?" may involve "some operative understanding - to the extent that number and counting are involved" ( p . 244) . Items also tap the preoperational level ("What do you call this finger?") , concrete operational level ("How many pennies make a nicke!?") , and formal operational level ("What causes iron to rust?") . To understand the kind of learning shown by a child's responses , we have to study each Information item sepa­ rately. Ideally, the child's responses should be used to gain some insight into how much of the information was ac­ quired figuratively, how much was acquired operatively, and what was the highest level of operativity involved . A simple construct like "range of information" does not ade­ quately explain tasks found on the Information subtest .

Similarities The Similarities subtest is far from a homogeneous mea­ sure of the processes involved in concept formation. The wheel and bal l item is a figurative tas k , whereas the piano - guitar item involves concrete operations . Early items are preoperational , intermediate ones reflect con­ crete operations, and later ones reflect formal operations.

1 80

Some items even measure divergent thinking (sal t ­ water) . Thus the Similarities subtest measures both figur­ ative and operative knowing as well as divergent thinking. It would be simplistic to interpret the Similarities subtest as a straightforward measure of concept attainment ability. Again, each response should be studied to evaluate the type of knowledge shown by the chil d .

Block Design Elkind views the Block Design subtest as "almost entirely a test of perceptual regulations - concrete operations in the service of perceptual analysis" (p. 246). It is largely an operative task, measuring primarily concrete operations.

Comment on the Piagetian Perspective The figurative/operative distinction offers a promising tool for analyzing many different types of content, including those found on cognitive tests and in school curricula. Unfortunately, as Elkind (1979) points out , it is not always possible to determine whether a correct response repre­ sents a rote answer (figurative knowing) or genuine prob­ lem solving (operative knowing) . For example, "How many ears do you have?" can be answered purely by rote or by problem solving. Consequently, the reason for success or failure on some items cannot be known without further inquiry. This is wpy testing-of-limits is helpful . By asking follow-up questions, you may be able to determine the quality of thinking involved in the child's responses.

CHAPTER 8

years . As noted in Chapter 2, test-age equivalents have many drawbac k s ; therefore their routine u s e is not recommended.

I LLUSTRAT I O N S OF Q U A LITATIVE A NA LYSIS O N T H E W ISC·R

This section provides a few interpretations based on item content, test responses and behavior, test patterns, and testing-of-limits procedures . An assumption underlying some of the interpretations in this section is that the cog­ nitive variables involved in the WISC-R (and WPPSI and WAlS-R) subtest s - such as planning , anticipation, atten­ tion, and concentration - are reflective of personality orga­ nization. If this assumption holds true, then the pattern of scores may reveal general modes of adaptation . Interpreta­ tions should be guided by the consistency present among the various aspects of test performance - profile of scores, content of responses, and style of responses . All inter­

pretations formulated through these procedures should be viewed as tentative and should be evaluated in light of the child's performance on the entire test, other assessment results, and clinical history. The examples in this section usually apply only in cases where other information can confirm the hypothesi s .

A Variety of Hypotheses About Interpreting the WISC·R Some children show a pattern of missing easy items and succeeding on more difficult items. This pattern may occur among bright children who are bored by the easy items and thus give careless or even nonsense replies, only to become challenged by more difficult items that allow them to demonstrate their skills (Palmer, 1983) . This pat­ tern may also suggest inconsistent attention or effort re­ sulting from anxiety or other factors . Various questions and words may arouse associations with violence or hostility. 1\.vo such Vocabulary subtest words are knife and affliction. On Similarities, a response such as "Apple and banana are alike because they both can be used to feed a person poison" may be suggestive of psychological difficulties. Reactions to Picture Completion tasks may provide information about a child's difficulties. A child with inca­ pacitating anxiety about separation issues may be unable to answer any item and may ask you why something has to be missing (Brooks, 1979) . A child suffering from dysphasia (a form of speech impairment due to central nervous sys­ tem impairment) may exhibit difficulty in finding words •

ESTI MAT E D P E RC ENTI LE RANKS A N D TEST·AGE EQU IVALENTS FOR RAW SCORE S

In some settings and on some occasions it may be useful to convert subtest scaled scores to percentile ranks or to convert raw scores to test-age equivalents. For example, when discussing assessment results with a teacher or par­ ent or physician, you may fi nd it useful to use derived scores. Table C-41 in Appendix C gives the estimated percentile ranks for each WISC-R (and W PPSI and WAIS-R) scaled score and qualitative descriptions for the scaled score. You should never estimate an IQ on the basis ofonly one subtest score, however. Table 21 in the WISC-R manual provides the test-age equivalents of raw scores on each subtest. The test-age equivalents provide approxi­ mate developmental levels for achievement on the subtest . For example, a raw score of 6 on the Informa�ion subtest is roughly equivalent to a developmental age level of 6-6

INTERPRETING THE WISC-R

•

•

lSI

ILLUSTRATIONS OF QUALITATIVE ANALYSIS ON THE VI!ISC · R

(anomia) . For example, the child may point to the '11iss· ng part instead of naming it, describe the missing part or its use, or indicate the part's function with gestures (Small,

1982). • Children who continually recheck their work with the Block Design model may be revealing obsessive ten­ dencies. The painstaking Coding task may be more difficult for the alert , creative, and intuitive child than for the more pedantic , passive , and slow-moving one (Taylor, 1961). • Errors o n Mazes may b e indicative o f impulsivity (overshooting alleys and exits) ; loss of control or poor coordination (walls cut through) ; or excessive caution, perfectionism, or depression (correct completion after time l imits) (Zimmerman & Woo-Sam, 1985 ) . • I n a paranoid individual the content o f responses, style of responses, and pattern of scores may be interrelated (Blatt & Allison, 1968) . The content of the examinee's responses may reflect suspiciousness regarding the record­ ing of the responses or feelings of being tricked . The style may be cautious, rigid, and legalisti c . Scores may be highest in areas related to hyperalertness to details (Picture Completion or Picture Arrangement) and to bringing to­ gether and relating disparate things (Similarities). •

Interpreting Comprehension Subtest Responses The following are possible interpretations of Comprehen­ sion subtest responses that may reveal characteristic of the child's social judgment .

Question 1.

"What is the thing to do if a boy (girl) much smaller than yourself starts to fight with you?" (All exam­ ples are from Taylor, 196 1 . ) A nswer: "He should not hit you" ( a possible indication of a moralistic response ) . A nswer: " I should not fight" ( a possible indication of a defensive response ) . Answer: " I would tell m y mother" ( a possible indication of dependency) .

Question 2 .

"What i s the thing t o d o when you cut your

finger?"

Answer: "You should be more careful and not cut your finger" (a possible indication of a moralistic response , Tay lor, 1961). Answer: "Fingers can fall off or be cut off and maybe they could be sewed on" (a possible indication of preoc­ cupation with aggression and retribution, Brooks, 1979) .

Question 3 . "What are you supposed to do if you find someone's wallet or pocketbook in a store?" (All examples are from Zimmerman & Woo-Sam , 1985 . ) Answer: "Give it back, say you didn't take it" ( a possible indication of guilt) . Answer: "Give it to your mother" (a possible indication of passivity). Answer: '"Take the money" (a possible indication of sociopathy) . Answer: "Don't do nothing" (a possible indication of difficulty in coping). Answer: "I never stole anything" (a possible indication of denial) . Diagnostic Aspects of the Picture Arrangement Subtest Correct as well as faulty solutions on the Picture Arrange­ ment subtest allow you to evaluate the developmental level of a child's reasoning (Taylor, 1961) . Most failures are due to difficulties in logical reasoning . Faulty solutions may also result from inattention ( for examp l e , on the SLEEPER ite m , failure to see the time on the clock) . Reasoning difficulties are suggested when a child who is more than 8 years old regularly makes young child-like arrangements . Reasoning difficulties tend to occur among mentally retarded children and brain-injured childre n , and sometimes among severely emotionally disturbed chil­ dren. Mildly emotionally disturbed children usually pro­ duce logical stories. Their anxieties and preoccupatio ns, however, may be revealed by unusual arrangements , such as neglecting details for the sake of a particular theme . There are two methods that can be used to obtain thema­ tic material from the Picture Arrangement subtest. These techniques should be used only after the entire scale has been administered . Both are screening devices; they are not intended to replace other thematic instruments, such as the Thematic Apperception Test (TAT) or the Children's Apperception Test. These techniques permit the child to project feelings and experiences , however, and may reveal a range of experiences and affect . One method is to arrange one or more Picture Arrange­ ment items in the order given by the child during the test proper and then ask the examinee to make up a story. Only items that were originally given to the child should be selected. Standard TAT instructions can be used , such as "Make up a story. Tell what is happening , what happened before, what will happen next, and how the people are feeling . " The methods usually applied to thematic material can then be used to interpret the stories. A second method involves randomly placing the FIGHT

1 82

sequence of cards before the child and then asking the child to make up a story about one ofthe cards, in standard TAT fashion (Craig, 1969) . After the story has been completed and the cards have been removed, the PICNIC sequence of cards is placed randomly before the child and the child is asked to tell a story about one of the cards. This procedure can be followed for any other items that were administered to the child during the test proper.

COM M E N T O N I NTERPRET I N G T H E WISC·R

The child's performance on the WISC-R should be inter­ preted in relation to all other sources of data. It is impor­ tant to verify the validity of the test scores with other information obtained about the child. If the Full Scale IQ does not appear to provide a valid estimate of the child's ability, consider the Verbal Scale and Performance Scale IQs as separate pieces of information that may have greater validity. Any hypotheses about the validity of the Full Scale IQ, Verbal Scale IQ, or Performance Scale IQ should be developed on the basis of all the available information you have about the child . When a significant Verbal-Performance split occurs , the Full Scale IQ may be a misrepresentation of the indi­ vidual's functioning level . In this situation, the Full Scale IQ may merely represent some kind of forced average of rather disparate primary skills. What meaning , for exam­ ple, can we attach to an IQ of 100 that results from a Verbal Scale IQ of 130 and a Performance Scale IQ of 70? Al­ though the IQ of lOO may be the best overall estimate of the child's cognitive level, the child is not likely to be average either in situations calling for verbal reasoning or in those requiring nonverbal reasoning . Unfortunately, there is lit­ tle research that can help us understand how examinees with large Verbal-Performance discrepancies function out­ side of the test situation. To a lesser extent, a similar problem exists in interpreting the Verbal and Performance IQs when there is an exceptionally large amount of vari­ ability among the subtests within each scale. Performance on each WISC-R scale and subtest is multi­ determined . This means that any one scale or subtest likely measures many different abilities. Consequently, a high score or a low score does not tell you what particular functions are well developed or impaired . This informa­ tion will come only from a sifting of all WISC-R scores, scores obtained from other tests, qualitative information, testing-of-limits , and clinical history. Good clinical skills constitute the sturdy thread from which the tapestry is woven.

CHAPTER 8

INTERPRETING T H E WISC-R

As noted in Chapter 6, the WISC-R may not be the instrument of choice for evaluating the cognitive abilities of children who function at either an extremely low or an extremely high level . In the case of low-functioning chil­ dren, too few items may be administered. In the case of high-functioning children, there may not be a sufficient number of challenging items . Profile analysis is also ham­ pered by the fact that the highest scaled scores are not available on each subtest throughout all of the age ranges covered by the scale. In Chapters 20, 21, and 2 2 , brief WISC-R case studies are presented for various groups of exceptional childre n . A study of these cases will help you i n interpreting the WISC-R. Interpreting the WISC-R is a challenging activity. The WISC-R gives an estimate of the child's level of intellectual functioning . The word estimate should be emphasized . The WISC-R provides useful, but limited, information about the range , depth, and real-world applications of a child's intellectual ability. Many clinicians are not satisfied with merely estimating intellectual ability from the Wechsler scales. They want to use these scales to evaluate personality and temperament, to diagnose various forms of psychopathology, to deter­ mine brain lateralization , and so forth . Once one goes beyond the confines of the IQs provided by the Full, Verbal , and Performance Scales, however, the ground be­ comes loose and wobbly. Interpretations become more impressionistic and may be less reliable and valid . When on this ground, step carefully, continually getting your bearings from research findings and clinical experience.

PSYCH O LO G I CAL EVALUAT I O N

The psychological evaluation in Exhibit 8-2 illustrates how the WISC-R can contribute to the assessment of an emo­ tionally disturbed youngster. For illustrative purposes the report focuses on only the WISC-R. A thorough assess­ ment would be based on a battery of assessment pro­ cedures.

TEST Y O U R SKI LL

Exhibit 8-3 presents a series of exercises designed to sharpen your skill in writing reports and in interpreting the WISC-R (and WPPSI and WAIS-R) . In each excerpt shown in Exhibit 8-3 , some inadequacy of description or interpretation exists. Find the mistakes in the sentences . After you have completed your analysis , check your eval­ uations with those shown in Appendix A .

1 83

T E ST YOUR SKILL

Exhibit 8-2 Analysis of and Line-by-Line Commentary on a Psychological Evaluation of an Emotionally Disturbed 7-Year-Old Examined with the WISC-R

INTRODUCTORY REMARKS

Shown on the following pages is a case study of an emo­ tionally disturbed child. Jim, a 7-year-old boy referred for evaluation because of antisocial behavior, was administered the WISC-R. His Full Scale IQ of 1 1 3 is classified in the High Average range of intell igence. There was a significant 1 5point difference between his Verbal Scale IQ ( 1 1 9) and Perfor­ mance Scale IQ ( 1 04) . Most subtest scores were above aver­ age , with the exception of those on two performance tasks (Object Assembly and Coding) . These results suggest that his nonverbal skills may be more variable than his verbal skills. Notice in the report the several references to Jim's persever­ ance and need for reassurance during the test session. A lso note that several of his answers were related to his antisocial behavior pattern. This case illustrates that emotional distur­ bance may not necessarily affect cognitive functioning. ANALYSIS OF THE REPORT Identifying Data

The report begins with the traditional identification data . In an actual report, the child's last name would be included. Test Administered

This part of the report cites the test name (or names) and test scores. It is optional because the name of the test an major WISC-R scores (Verbal, Performance, and Full Scale IQsl can be included in the body of the report . Often the suhtest scores are not included in the report . Reason for Referral

This section, which usually begins the narrative portion of the report , explains the reason for the evaluation. It documents what the psychologist sees as the purpose for the evaluation and helps to develop the focus of the recommendations. It also may contain information about the child that is related to the referral question. Background I nformation

This section describes portions of Jim's background that might be pertinent to the issues under consideration. The first para­ graph sets the stage for a historical understanding of the problem and provides information as to possible resources available within the family for remediation. The second para­ graph cites in more detail those behaviors that led to Jim·s

referral . The third paragraph focuses on school behavior and recent changes in problem behaviors. Behavioral Observations

This section describes Jim's behavior during the examination, with particular emphasis on his approach to the test and behavior reflective of his unique style. Assessment Results and Clinical Impressions

This section begins with a description of Jim's overall test performance . Normative data and confidence levels are then reported . The confidence bands for IQs are found in Table C- I in Appendix C. The percentile ranks for IQs are found in Table BC- I on the inside back cover of this text. The second paragraph discusses the discrepancy between his Verbal and Performance Scale IQs. Table C-2 in Appendix C gives values that indicate significant differences between scaled scores and between individual subtest scores. Table C-3 in Appendix C gives values that indicate significant di fferences between individual subtest scores and their re­ spective mean scaled scores. The third paragraph describes Jim's strengths and weak­ nesses based on his subtest scores. Jim's numerical reasoning, range of knowledge, and language usage are well developed, as suggested by his scores on Arithmetic, Information , and Vocabulary, respectively. Less adequate are his visual-motor coordination and psychomotor speed , as ind icated by his scores on Object Assembly and Coding. (Table C- 1 3 in Ap­ pendix C provides a summary interpretation of skills associ­ ated with the individual WISC-R subtests . ) The fourth paragraph describes some idiosyncratic re­ sponses that may be reflective of Jim's behavior problem, and the last paragraph provides a brief summary of Jim's overall level of functioning. Recommendations

Recommendations are made before the final summary. Note that because the evaluation did not include personality tests, the recommendations call for further testing and evaluation. Summary

The final section of the report summarizes the major findings and recommendations.

(Exhibit conrinues next pag/.:)

1 84

CHAPTER 8

INTERPRETING TH E WISC-R

Ex hibit 8·2 (cont.) THE REPORT WITH L INE-BY-LINE COMMENTA RY

Date of examination: November 20, 1 985 Date of report: November 2 1 , 1 985 Grade: Second

Name: Jim Dare of birth : July 1 7 , 1 97 8 Chronological age: 7-4

Test Administered

Wechsler Intelligence Scale for Children - Revised (WISC-R) : VERBAL SCALE

I n formation Si rnilarities Arithmetic Vocabulary Co mprehension

PERFORMANCE SCALE

14 12 15 13 12

Picture Completion Picture Arrangement Block Design Object Assembly Coding

12 13 14 8 6

Verbal Scale IQ = 119 Performance Scale IQ = 104 Fu ll Scale IQ 113 ± 7 at the 95 % confidence level =

Reason for Referral I 2 3

Jim, a 7-year, 4-month-old boy, was referred to the clinic for evaluation because of involuntary defeca­ tion, fecal smearing, enuresis, and stealing.

I

2 2-3

who was referred to whom he was referred specific behaviors leading to referral

4-7

early infancy and family background

Background 4 6

�

10 II

12

i .l 1 ·\ 15

16 17

18 19

20 21 22 23

2� 25

26

27

28

29 �O

According to Jim's aunt, Jim was born out of wedlock and never knew his natural father. He was separated from his mother at 6 months of age, when she devel­ oped leukemia from which she died a year later. Since that time Jim has lived with a paternal aunt and her three children, who range in age from 8 to 1 8 years. The aunt has been divorced twice. Jim call s his aunt "mother" and thought of her first husband as his father. This man, who had been in the family as long as Jim, has not been involved with the family since he di­ vorced the aunt. Jim was 2 '/2 -years-old at the time of the divorce. His aunt described having made in­ complete and ineffective attempts at toilet training during this period of turmoi l . Last year Jim's aunt remarried. During the six months that the marriage lasted, Jim formed no attachment to his second step­ uncle. Jim's aunt reported that he has bowel movements in the bathtub; he smears feces on the walls or leaves them in trash cans around the house. He also soils himself frequently. He wanders around the house at night, and sometimes vanishes for hours whiJe in the park or on his way home from school . His aunt stated that he stole a stopwatch from the principal's office, and food and money from other places. Apparently, he seems to make an effort to be discovered when he steals.

8- 1 0 family constellation 1 0- 1 1 attachment to parental figures 1 2 - 1 5 time of early separation from parental figure

1 5 - 1 7 past parental behaviors that may relate to Jim's current problems 1 7 - 1 8 recent family changes 1 8-20 Jim's relationship to new famiJy member

2 1 -24 encopresis

24-26 "wandering" behavior 26-30 stealing behavior

(Exhibit continues next page)

TEST YOUR SKILL

1 85

Exhibit 8·2 (cont.) 31 32 33 34 35 36

Jim is an excellent student, although he is difficu lt to handle because of his opposition and defiance. H:s behavior in the past three months has been changing . There has been a decrease in encopresis and fecal smearing, and an increase in stealing and aggressive behavior.

3 1 -3 2 level of academic achievement and deviant classroom behavior 33-36 recent changes in behavior

Behavioral Observations 37 38 39 40 41 42 43

44

45 46 47 48 49 50 51

Jim is a smal l , thinly built, energetic child. He was cooperative and friendly during the testing session . H i s test behavior was characterized by competitive­ ness, tenacity, and anxiety. He seemed to want to answer all of the questions correctly and was reluctant to g ive up on any question. On the Information sub­ test, he responded to "What are the four seasons of the year?" with "Spring" and "Fall ," but could not re­ member the other seasons. He had to be encouraged to go on to the next question, and three times later he spontaneously returned to the season question, adding "W inter" and "Summer." Jim seemed to need continual assurance from the examiner that he was answering the items correctly. He often asked, "Have I gotten them all right?"

37 appearance 37-38 overall response to being tested 39-42 response to test materials

42-48 example of response to a test question

48-50 style of relating to examiner 50-5 1 example of test behavior

Assessment Results and Clinical Impressions 52 53 54 55 56 57 58 59

60 61

62 63

64

65 66 67 68 69 70 71 72 73 74 75 76 77 78 79

W ith a chronological age of7A, Jim achieved a Verbal Scale IQ of 1 1 9 , a Performance Scale IQ of 1 04 , and a Full Scale IQ of 1 1 3 ± 7 on the WISC-R. His overall performance is classified in the High Average range and is ranked at the 8 1 st percentile. The chances that the range of scores from 1 06 to 1 20 includes his [rUe IQ are about 95 out of 1 00. The present measure of his level of intellectual functioning appears to be valid. Although there was a I S -point difference between the Verbal and Performance Scales, this difference was associated primarily with his low scores on two Performance Scale subtests. Whereas his verbal skills are uniformly well developed, his performance skill s show more variability, with some abilities well devel­ oped and others less well developed . Overall (with the exception of two of the Performance subtest scores) Jim consistently demonstrated above-average skills as assessed by the test. W ithin the verbal area, his numerical reasoning ability, range of knowledge, and language usage are excellent. Social comprehension and concept forma­ tion ability are also strong. Within the performance area, his analytic and synthetic ability, planning and anticipation ability, and ability to differentiate essen­ tial from nonessential details are all well-developed. Less adequate are skills associated with visual-motor coordination and with psychomotor speed. It is diffi­ cult to account for his lower scores in these two areas

52 53 54 55 56

chronological age Verbal Scale IQ and Performance Scale IQ Full Scale IQ and confidence band normative classification based on Full Scale IQ percentile rank 35-58 con fidence range of IQ 58-59 validity of tcst results 60-6 1 discrepancy between IQs 6 1 -63 variabil ity noted in Performance area 63-64 description of verbal skills 64-66 description of nonverbal skills 66-69 summary of performance on WISC-R

70-73 verbal skill strengths

73-75 nonverbal skill strengths

77 -78 nonverbal skill weaknesses 78-80 relation of weaknesses to overall ability (Exhibit continues next page)

1 86

CHAPTER 8

INTERPRETIN G T H E WISC-R

Exhibit 8·2 (cont.) 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 1 00 101 1 02 1 03 1 04 1 05 1 06 1 07 1 08 1 09

in l ight of his overall above-average ability. Perhaps the scores reflect temporary inefficiency due to fa­ tigue, or perhaps they simply indicate that his abilities in these areas are not as well developed. Overall , he approaches problems by giving detailed answers that are descriptive and meaningfu l . A number of Jim's responses were particularly interest­ ing. In response to the question "Why are criminals l ocked up?" he said, "Because they steal things that are expensive . " In reference to another question asking why it was better to give money to a charity than to a beggar, he said, "Because the beggars might be a burglar." His mention of "stealing" and the specifica­ tion of "expensive" and "burglar" are notable re­ sponses in view of his behavior pattern, which in­ cludes stealing. A preoccupation with stealing may be intruding into his outlook toward l ife. One response differed from his general pattern of rather clearcut, detailed, well-oriented, and d irect responses . He said that apple and banana are alike because they "feel the same. " The above responses were the only ones that were idiosyncratic. The test results suggest that Jim's behavioral prob­ lems, for the most part, have not interfered with his intellectual functioning. His psychomotor speed and v isual-motor coordination are less-adequately devel­ oped than his other abilities, but his overall function­ ing is better than average. There were suggestions of preoccupation with stealing, but it is difficult to deter­ mine the extensiveness of this preoccupation.

80-83

possible reasons for low scores

84-85

response style

86-92

examples of unique responses

92-96

relation of responses to current behavior

96- 1 00 instance of unusual response sty Ie

1 00- 1 0 I observation of unusual responses 1 02- 1 04 effect of behavior problems on cognitive functioning 1 04- 1 06 weaknesses 1 06 - 1 07 level of overall functioning 1 07- 108 relation of specific responses to behavior problems 1 08- 1 09 l imits of i nterpretation

Recommendations 1 10 111 1 12 1 13 1 14 1 15 1 16 1 17 1 18 1 19

On the basis of the present limited evaluation, it is recommended that a personality evaluation be con­ ducted. Furthermore, the seriousness of his behav­ ioral d isturbance suggests that therapy should be initi­ ated for both Jim and his aunt. Every attempt should be made to obtain further i nformation about his home environment and to determine which factors i n the home may be reinforcing his deviant behavior pattern. His aunt should be actively engaged i n the develop­ ment of a treatment program.

1 10 limitations of current testing 1 1 1 - 1 1 2 recommendation for personality testing 1 1 2- 1 1 4 therapeutic interventions 1 14- 1 1 7 needed additional information about home environment

1 1 8- 1 1 9 suggested intervention

Summary 1 20 121 1 22 1 23 1 24 1 25 126 1 27 1 28

Jim, with a chronological age of7-4, achieved an IQ of 1 1 3 ± 7 on the WISC-R. This IQ is at the 8 1 st percen­ tile and in the High Average range. The chances that the range of scores from 1 06 to 1 20 includes his true IQ are about 95 out of 100. The test results appear to give a valid indication of his present level of intellec­ tual functioning . Jim's verbal skil l s were uniformly well developed, but there was some variability in his performance skills. The findings suggest that his be-

1 20 121 1 2 1 - 1 22 1 22- 1 24 1 24- 1 26

chronological age IQ and name of test percentile rank and normative classification of IQ confidence limits associated with IQ validity of test results

1 26- 1 2 7 verbal skills 1 27 - 12 8 variability of nonverbal skills (Exhibit continues next page)

1 87

TEST YOUR SKILL

Exhibit 8·2 (cont.) 1 29 1 30 131 1 32

havioral problems are not significantly interferI ng with his cognitive skills. A personality evaluation was recommended, along with a treatment progra that would involve lim and his aunt.

1 29- 1 30 possible e ffects of behavior problems on intellectual functioning 1 30- 1 32 recommendations

(Signature) 10

Lynn Mack, M A

Exhibit 8·3 Test-Your·Skill Exercises for the WISC·R

Read each item to determine where it is inadequate . Check your evaluation with the comments in Appendix A . Unnecessary Technical I nformation I . A difference of 4 points between her Information and

Comprehension scores is significant at the 5 % level . 2 . Mark did not score zero on any Comprehension items. He had 12 2-point responses and 5 I-point responses. 3. On the Comprehension subtest, Bill scored 18; IO is average and 19 i s the ceiling. 4. On Block Design , she failed items 3, 4, and 5 . 5 . Her Object Assembly score was significantly different from the mean of her other performance scores. 6. Bill scored 5 points on the Similarities subtes!. 7. He missed the Picture Arrangement item about rain . 8. The Digit Span is an optional subtest and was not used in computing the lQ. 9. On the WISC-R, the majority of her scores hovered around 12 . 10. A total scaled score of 52 yielded a Performance IQ score of 1 02 . I I . O n the Information subtest she earned a scaled score of 13, which is 1 standard deviation above the mean of 10. 1 2 . Bannatyne's recategorization system reveals particular strength in the V isuo-Spatial area (prorated IQ = 127 ) . 1 3 . Frank appears to b e weak i n the areas dealing with Freedom from Distractibility. 1 4 . A review of Glenda's Verbal scores indicates significance at the .05 level in her Vocabulary and Comprehension tests. 1 5 . Her score on Coding was 4 points lower than her score on Block Design and 5 points lower than her score on Object Assembly. 1 6 . There was a significant difference between lohn's Verbal and Performance IQs at the .05 leve l . 1 7 . A mean scaled score o f 9 was obtained o n the 10 subtests . 1 8 . Intersubtest scatter was minimal .

Poor Writing

19. His score on the WISC-R was equivalent to an IQ of approximately 9 8 . 20. Within the Verbal Scale, her abilities ranged from aver­ age to very superior when compared to other children in her chronological age group. 2 1 . Average abilities were indicated in Pat's attention and concentration and how well they are u sed in conjunction with solving basic arithmetic problems, and in her au­ ditory vocal sequencing memory. 22. The examination of the results displayed a significant amount of point discrepancy between the Verbal and Performance IQ scores of 37 points. 23. The Verbal scores were close-knit in the solidly average range from 9 through II. 24. I n the area of math, the regular classroom teacher might make use of concrete situations to add meaning and to reinforce the total experimental background . 25. All of Mary's scores were respectable and adequate, with the exception of Digit Span where she received a 7 . 26. Bill has better mental than nonverbal abilities . 2 7 . Statistical factors and the tenor of his test performance indicate excellent chance (95 %) that his test performance would fall consistently (other things being equal) within the range of 117 to 129. 28. lim's high scores on the Performance Scale suggest that he has well developed abilities in the conception of his environment . 29. The following description was written about a 6-year­ old's performance on the Block Design subtest: "She gave up easily even after putting the blocks into the design but not recognizing it as it was supposed to be , thus disman­ tling the design she made . " 30. The following sentence was written for Verbal subtest scores between 9 and 1 1 : "Her verbal subtest scores ap­ pear to be within the average range . "

(Exhibit continues next page)

1 88

CHAPTER 8

INTERPRETING T H E WISC-R

Exhibit 8-3 (cant.) 3 1 . The accuracy of his intrasubtest scores were intermittent on many of his subtests. He missed items in such propor­ tions that he was able to complete all of the subtests. 3 2 . His other outstanding score that was indicated was his performance in the Picture Completion subtest. 3 3 . In reviewing her Performance Scale subtest scores, there appears to be a significance at the . 05 level between her Object Assembly and Block Design . 34. The Verbal test profile showed a peek for the Comprehen­ sion subtest. 3 5 . She showed a retarded score on the ability to see spatial relationships. Technical Errors

36. A lower score on Information (scaled score 9) shows poor range of knowledge. 3 7 . Ave rage to above-average scores were obtained by George in the sequencing area, w ith a prorated IQ of 103 in the Sequencing I area and a prorated IQ of 120 in the Sequencing II area. 3 8 . Henry scored in the average intellectual range on the WISC-R, with a mental age of 7-2 and a chronological age of 7-6. 39. Five of the 12 WISC-R subtests were 2 standard devia­ tions from the mean, indicating that her performance was in the very superior range. 40. The IO-point difference between Brandon's Verbal and Performance Scale IQs approaches significance at the 5 percent level, suggesting that his verbal skill develop­ ment is somewhat ahead of his nonverbal development. 4 1 . Her Full Scale IQ of 109 ± 6 just barely reaches the Above Average classification. 42 . The range of scores from 6 to 16 indicates good performance. 43 . Her scores ranged from a classification of Very Superior to Borderline Mentally Deficient. 44. The Picture Completion score was significantly lower than the Picture Arrangement score. Because these two subtests are somewhat similar in the testing of detail, reasoning ability, and perceptual organization, the Pic­ ture Completion subtest may have been spoiled. 45 . Bill's IQ of 114 ± 9 classifies him in a range from average, high average, to superior intellectual functioning. 46. On the basis of a range of scores from 12 to 19 (Verbal IQ = 142 , Performance IQ = 1 3 1 , Full Scale IQ = 141 ) , the following statement was made: "His subtest scores show great variability, indicating he has definite strengths and weaknesses . " 4 7 . Bill achieved a Verbal Scale score of 65 , a Performance Scale score of 60, and a Full Scale IQ of 1 1 8 .

48. Her scaled score of 3 on Information places her in the Mentally Retarded range. 49. The following statement was made on the basis of a Verbal Scale IQ of 107 and a Performance Scale IQ of l l l : "Her Performance Scale IQ is higher than her Verbal Scale I Q . " I naccurate or I ncomplete I nterpretations

50. H i s low I nformation score reflects potential repressive mechanisms at work. 5 1 . In response to the question "In what way are anger and joy alike?" she said , "tell ing about something that you do." Clinical i nterpretation of this statement suggests that it is probably an emotional indication of aggression. 52. Two subtests deviate significantly from his average Ver­ bal subtest score: Vocabulary ( h igh) and Arithmetic (low ) . 5 3 . A high Picture Completion subtest score and a l o w Cod­ ing subtest score may predict difficulty in reading . 54. Bill's Full Scale IQ was achieved with a l O-point differ­ ence between his Verbal and Performance Scale scores, in favor of the latter. This difference suggests an action­ oriented person. 55. A high score on Coding reflects a high and sustained energy level . 56. Discrepancies between this WISC-R administration and the one given three years ago may be due to Henry's variable attention and memory span or possibly examiner differences. 57. She scored significantly higher on Information and Vo­ cabulary than on Digit Span . 5 8 . The 1 3-point difference between Helen's Verbal and Per­ formance Scale scores indicates that her Verbal subtest scores are significantly higher than her Performance subtest scores. 59. Lack of social judgment and immature responses were noted on the Comprehension subtest (scaled score 8 ) . 60. Bill scored high on Object Assembly because he was persistent in his attempt to assemble the objects. 6 1 . The 15-point discrepancy between Mary's Verbal and Performance Scale IQs indicates that she has a learning disability. 62 . The consistency of his scores probably indicates that John is functioning at his capacity. 63 . His lowest score was on the Picture Completion subtest, where he received a scaled score of 12. 64 . A weakness in another modality - i . e . , hearing - was evi­ dent i n the Digit Span subtest score.

(Exhibit continues next page)

1 89

SUMMARY Exhibit 8·3 (cont.) 65 . Her high Verbal subtest scores indicate that she i able to express herself with little difficulty. 66. Based on his scores, he would be considered a slow learner or a low average student. 6 7 . The following statement was based on a Similarities scaled score of 15 and a Digit Span scaled score of 10: "She has good conceptual izing ability and poor rate memory for digits . " 6 8 . The 40-point difference between Greg's Verbal and Per­ formance IQs can probably be accounted for by the fact that at age 6, Greg has not yet developed the visual-motor skills he needs to do his best on the nonverbal part of the WISC-R . 69 . Comprehension requires social adaptation, practical judgment, and self-direction. 7 0 . The i n t rasubtest scatter may i nd icate a lack of persistence. 7 1 . The following sentence was based on scaled scores of 13 or higher, with the exception of a 10 on Block Design: "His weaknesses seemed to be in visual-motor coordina­ tion and spatial orientation. His lowest and only average score was on the Block Design subtest. " 72 . Another relatively low score occurred in the Perfor­ mance Scale, indicating attentional skill weakness . 7 3 . Her verbal skills appear significantly better developed than her performance skills, suggesting that her ability to respond automatically with what is already known may be more developed than her ability to use past experi­ ences and previously acquired skills to solve new problems. 74. He scored relatively high on Digit Span and Coding. 75 . The fact that her Similarities score was lower than her Comprehension score impl ies some difficulty involving fear and emotional excesses. 76. The following interpretation was based on scaled scores of 8 on Picture Completion, 7 on Picture Arrangement, 10 on Block Design, 7 on Object Assembly, and 9 on Coding: "On two tests, Picture Arrangement and Object Assembly, she scored in the low average range. Because these tests measure s kills similar to those measured by tests on which she scored higher, it appears probable that

S U M MARY

I . Profile analysis is a method of generating hypotheses about the organization of intellective functions. It is a statistica.1 approach to evaluating differences between various scores. After the significance of such differences has been established, it i s

77.

78. 79.

80.

81.

82.

83. 84.

85.

she did not apply the same effort on these tests or that there is some other intervening factor such as poor visual acuity. " The following statement was based on a Verbal IQ o f l29, Performance IQ of 100, and Ful l Scale IQ of 120: "Monica's verbal skills appear to be somewhat more developed than her performance skill s . " A review o f her Verbal tests does not appear t o indicate any areas of significance. A third factor on the test generally denotes poor atten­ tion, poor concentration, and poor ability to screen out extraneous influences. In the area of acquired knowledge, which assesses life experiences, social-educational exposure, and learning, he obtained an average score. The following sentence was based on a Verbal IQ of98 , a Performance IQ of 131, and a Full Scale IQ of 109: "The discrepancy between her verbal and performance scores is significant and may suggest that she is compensating for her lack of verbal abilities with her superior perfor­ mance abilities to achieve good grades in school . " The following statement was based on a Verbal IQ of96, a Performance IQ of 131, and a Full Scale IQ of 113: "An examination of the overall results of this WISC-R session indicates that Frank is very strong in areas where using one's hands is important ." His low functioning on Coding may relate to his appar­ entiy weak background in school-related tasks. On the Coding subtest the child is asked to attach a meaningless symbol to a number. Because she tried hard, he passed many items.

I nappropriate Recommendations

86. The significant difference between his Verbal and Perfor­ mance IQs indicates a need for further investigation into his intellectual ability. 87. Because of the extreme scatter in her nonverbal abilities, Julie should be examined by a neurologist and given appropriate medical tests to confirm or eliminate the possibility of neurological impairment .

necessary to consider a variety of factors that may account for a specific profile. Profile analysis is one form of test interpretation, and it must be employed in relation to the entire test performance and cl inical history. 2 . The primary approaches to profile analysis include com­ paring Verbal and Performance Scale IQs , comparing Verbal and

1 90

Performance subtest scaled scores with the average scores on the respective scales, comparing sets of individual subtest scaled scores, comparing factor scores, and comparing subtest scaled scores with the average factor scores on each factor. 3 . Other approaches to profile analysis - based on frequency of occurrence in the population - include an evaluation of subtest scaled-score ranges, deviations of subtest scores from the child's own average, and Verbal-Performance Scale differences. 4. Profile analysis requires considerable judgment and skil l . S . Scale differences and subtest differences should b e inter­ preted only when the differences are statistically significant. The assignment of statistical significance to a difference means that you can be relatively confident that the difference is due to factors other than chance. The .05 significance level is recommended as a minimum. 6 . A successive level approach to test interpretation is help­ ful in the development of hypotheses. The six levels of the approach are as follows: (a) the Full Scale IQ, (b) Verbal and Performance IQs or factor score s , (c) intersubtest variability, (d) subtest variability within scales, (e) intrasubtest variability, and (f) qualitative analysis. 7. The Verbal Scale is dependent on the child's accumulated experience, whereas the Performance Scale is more dependent on the child's immediate problem-solving ability. Both scales i nvolve verbal and nonverbal strategies in the solution of problems. 8. Verbal-Performance Scale differences may reflect the child's interests or cognitive style, or they may be an indication of some type of deficit. A difference in favor of the Verbal Scale IQ may indicate that verbal and auditory processing skills are better developed than performance and visual discrimination skills. A difference in favor of the Performance Scale IQ may indicate that performance and visual discrimination skill s are better developed than verbal and auditory processing skills . 9 . Factor scores should be compared with other factor scores and not with IQs. 10. The Freedom from Distractibility factor appears to mea­ sure attention, memory, numerical ability, encoding ability, use of rehearsal strategies, ability to shift mental operations rapidly, and sel f-monitoring ability. I I . Any hypotheses generated from subtest comparisons should be treated as tentative and formulated in relation to the child's absolute level of scaled scores. 1 2 . A Piagetian perspective, emphasizing figurative knowing (knowledge of static states) and operative knowing (knowledge of dynamic transformations) , may be usefu l in analyzing the Wechsler scales. 1 3 . It is sometimes helpful to translate scaled scores into percentile ranks or test-age equivalents.

CHAPTER 8

INTERPRETING THE WISC-R

1 4 . Qualitative analysis of WISC-R responses and patterns should be helpful in interpreting performance. 1 5 . A child's performance on the WISC-R should never be interpreted in isolation; it should always be v iewed in relation to all other sources of data. 1 6 . The WISC-R provides an est;l1Ulte of a child's level of intellectual functioning; it does not measure the complete domain of intellectual functioning. 1 7 . In cases of large Verbal-Performance discrepancies, the Full Scale IQ may be more difficult to interpret than the separate Verbal and Performance Scale IQs.

KEY TERMS, C O N C E PTS, AND N A M ES

Profile analysis (p. 166) Primary methods of profile analysis ( p . 167) Base rate subtest scaled-score ranges (p. 170) Base rate Verbal-Performance differences (or probability-of­ occurrence approach) ( p . 170) Successive level approach to interpreting the Wechsler scales (p. 1 7 1) I ntersubtest variability (p. 172) I ntrasubtest variability (p. 172) Elkind's Piagetian perspective for subtest analysis (p. 179) Figurative knowing (p. 179) Operative knowing ( p . 179) Qualitative analysis ( p . 180)

STUDY Q U ESTIO N S I . Discuss the intent o f profile analysis, methods of profile analysis, and approaches to profile analysis on the WISC-R . 2 . Describe the successive level approach to interpreting the WISC-R. 3. Discuss how to interpret differences between the WISC-R Verbal and Performance Scales and between factor scores. 4. Discuss how to interpret d i fferences between WISC-R subtests . C ite at least seven subtest comparisons i n your presentation . 5 . Discuss Elkind's Piagetian analysis of Wechsler subtests. 6. Give five illustrations of qual itative analysis on the WISC-R . What precautions must be taken in a qualitative analysis? 7. What are some general considerations in interpreting the WISC-R?

WECHSLE R PRESCHOO L AN D PRI MARY SCALE O F I NTELLIG ENCE

(WPPSI )

From the child offive (0 myself is but a step. But from the new-born baby (0 the child offive is an appalling distance.

Standardization Deviation IQs, Scaled Score s , and Test-Age Equivalents Reliability

-Tolstoy

Val idity Intercorrelations Between Subtests and Scales WPPSI IQs and Stratification Variables Factor Analysis Administering the WPPSI WPPSI Subtests Interpreting the WPPSI Assets of the WPPSI Limitations of the WPPSI Psychological Evaluation Test Your Skill Summary

191

1 92

CHAPTER

9

WECHSLER PRESCHOOL AND PRIMARY SCALE OF INTELLIG E N C E (WPPSI)

In 1967 the Wechsler Preschool and Primary Scale of Intelligence (WPPSI) was published for use with children between the ages of 4 and 6 1/2 years (Wechsler, 1967) (see Figure 9-1). It is separate and distinct from , although similar in form and content to, the WISC-R. The WPPSI contains 1 1 subtests (see Exhibit 9-1) , 8 of which (Informa­ tion , Vocabulary, Arithmetic, Similarities, Comprehen­ sion , Picture Completion , Mazes, and Block Design) also appear on the WISC-R and 3 of which (Sentences, Animal House, and Geometric Design) are unique to the WPPSI. The WPPSI does not contain the 4 WISC-R subtests: Digit Span , Picture Arrangement, Object Assembly, and Cod­ ing. The methods of computing IQs and evaluating scores are similar to those employed on the WISC-R. Thus, the WPPSI can be considered a downward extension of the WISC-R. (The WPPSI-R is discussed in Appendix G . )

STAN DARDIZAT I O N

The WPPSI was standardized on 1 , 200 children, 100 boys and 100 girls in each of six age groups, ranging by half-

•

years from 4 to 6 1/2 years. The 1960 U . S . census data were used to select representative children for the normative sample . Whites and nonwhites were included in the sam­ ple , based on the ratios found in the census for four geographic regions in the United States .

DEVIATION IQS, SCALED SCO RES, A N D T EST-AGE E Q U I VA L E N TS

The WPPSI, like the WISC-R and the WAIS-R, employs the Deviation IQ (M = 100, SD = 1 5 ) for the Verbal , Per­ formance , and Full Scale IQs and scaled scores (M = 10, �'D = 3 ) for the subtests. The IQs are obtained by compar­ ing the examinee's scores with the scores earned by a representative sample of his or her own age group . A raw score is first obtained on each subtest and then converted to a scaled score w ithin the examinee's own age group through use of a table in the WPPSI manual . Age g roups are divided into three-month intervals from 3-10-16 (years, months, days) to 6-7-15 . The IQ tables in the WPPSI

•

•

----'�

,

Figure 9- 1 .

;

Wechsler Preschool and Primary Scale of I ntelligence. Courtesy of The Psychological Corporation .

DEVIATION IQS, SCALED SCORES, AND TEST-AGE EC _'I'IAL[HTS I'"

Exhibit 9- 1

1 93

____________________�n � � �·aa � ______________________________________�

WPPSI-Like Items

Information (23 questions)

Mother loves me. Ted l i kes to eat apples. Martha l ikes to visit the museu m . She will go there today.

Show me your eyes. Touch them. How many legs does a cat have? In what kind of store do we buy meat? What is the color of an emerald?

Animal House

The task is to place appropriate colored cylinders in the corresponding holes on a board. The colored cylinders are matched with four different animals. The Animal House task is shown in the photograph of the WPPSI (see Figure 9- 1 ) .

Vocabulary (22 words)

boot book

nice annoy

Picture Completion (23 items)

Arithmetic (20 problems)

Card with squares of different sizes . Card is placed in front of child. Examiner says, "Here are some squares. Which one is the biggest? Point to it." Bill had one penny and his mother gave him one more. How many pennies does he now have? Judy had 4 books. She lost 1 . How many books does she have left?

The task is to identify the essential missing part of the picture (see Figure 9- 1 ) . A picture o f a tricycle without handlebars. A picture of a doll without a leg . A picture of a swing without a seat. Mazes ( 1 0 mazes)

Jimmy had 7 bananas and he bought 8 more. How many bananas does he have altogether?

The task is to complete a series of mazes. An example of a maze is shown in the lower left-hand corner of the photograph in Figure 9- 1 .

Similarities (16 questions)

Geometric Design ( 1 0 designs)

You can read a book and you can also read a Apple pie and ice cream are both good to In what way are a quarter and a dollar alike? In what way are a cow and a pig alike?

__ .

__

.

Comprehension ( 1 5 questions)

Why do you need to take a bath? Why do we have farms? What makes a sailboat move? Sentences ( 1 0 sentences)

The task is to copy geometric designs which are shown on printed cards. The designs include a circle , square, triangle , and diamond . Block Design ( 1 0 designs)

The task is to reproduce stimulus designs using three or four blocks (see Figure 9- 1 ) . Note. The sample questions resemble those that appear on the WPPSI but are not actually from the test. The subtests are described in detail in this chapter.

The task is to repeat sentences given orally by the examiner.

manual are based on 10 of the 11 subtests. The supplemen­ tary subtest, Sentences, is excluded from the calculation of the IQ unless another Verbal Scale subtest is skipped. Also excluded from the IQ tables is Animal House Retest.

tered in each scale, IQs should be computed by using the special short-form procedure described later in this chap­ ter. This procedure can be used whenever fewer than 10 subtests are administered.

Prorating Procedure

Test-Age Equivalents

When fewer than 10 subtests are administered, prorating is necessary. The WPPSI manual provides a table for prorat­ ing scores when four of the subtests are administered in each scale. When fewer than four subtests are adminis-

The WPPSI manual provides a table of test-age equivalents to fac il itate interpretation of a child's performance . Chap­ ters 2 and 4 discuss the assets and limitations associated with their use .

CHAPTER 9

1 94

WECHSLER PRESCHOOL AND PRIMARY SCALE OF INTELLIGENCE (WPPSI)

RELIABI LITY

The WPPSI has excellent reliabil ity. Internal consistency reliabilities for each of the three IQs range from . 91 to .96 over the range covered by the scale . Average internal consistency reliability coefficients , across the six age groups, are as follows: . 96 for the Full Scale IQ , .94 for the Verbal Scale IQ , and . 93 for the Performance Scale IQ .

Subtest Reliabilities The reliabilities for the subtests are not so satisfactory as those for the three scales (see Table 9-1 ) . The average subtest reliabilities range from a low of . 77 for Animal House to a high of . 87 for Mazes. Reliabilities are similar for Verbal and Performance Scale subtests . The reliability coefficients are similar across the six age groups . They range from a low of .62 for Animal House at 4 years of age to a high of . 9 1 for Mazes at 5 th years of age .

Standard Errors of Measurement Based on the average of the six age groups , the standard errors of measurement (SEn,) in IQ points are 2 . 88 for the Table 9- 1 Average Reliability Coefficients and Standard Errors of Measurement for WPPSI Subtests and Scales

Full Scale , 3 . 57 for the Verbal Scale, and 3 . 85 for the Performance Scale (see Table 9-1). Thus more confidence ' can be placed in the IQ based on the Full Scale than in an IQ based on either the Verbal or the Performance Scale. The standard errors of measurement for the subtests in scaled score poi nts , again based on the average of the six age groups , range from 1 . 1 8 to 1 . 34 for the Verbal Scale subtests and from 1 . 08 to 1 .46 for the Performance Scale subtests. Within the Verbal Scale, Sentences and Vocabu­ lary have the smallest SE", (1 . 18 and 1 . 2 1 , respectively) . Within the Performance Scale , Mazes and Picture Com­ pletion have the smallest SE", (1 .0� and 1 . 20 , respectively) .

Reliability with Special Populations Satisfactory split-half Full Scale reliabilities have been reported for a variety of populations, including Mexican­ American children (r = . 95 ) (Henderson & Rankin, 1973), gifted children (r = .93) (Ruschival & Way, 1971), mentally retarded children (r = . 88) (Richards, 1970) , and British children (r = .97) (Brittai n , 1969) . A test-retest study over a one-year time period revealed satisfactory reliability (r = . 89) for the WPPSI Full Scale IQ for a group of 25 black, lower-socioeconomic-class children (Croake, Keller, & Catlin, 197 3 ) .

Test-Retest Reliability A verage reliability coefficient

Average standard error of measurement

Information Vocabulary Arithmetic Similarities Comprehension Sentences

.81 . 84 . 82 . 83 .81 . 85

1 . 34 1 .21 1 .23 1 . 24 1 .32 1.18

Animal House Picture Completion Mazes Geometric Design Block Design

. 77 . 83 . 87 . 82 . 82

1 .46 1 . 20 1 .08 1 . 29 1 . 26

Verbal IQ Performance I Q Full Scale IQ

. 94 .93 . 96

3 .57 3 . 85 2 . 88

Subtest or scale

Note. Reliabi l ity coefficients for all subtests except Animal House are odd-even correlations corrected by the Spearman-Brown formula. For Animal House the reliability coefficients are test-retest coefficients. Source: Adapted from Wechsler ( 1 967).

When the test was readministered to 50 children in the standardization group after a period of approximately 1 1 weeks, average increases o f 3 . 0 , 6 . 6 , and 3 . 6 I Q points were found in Verbal , Performance , and Full Scale scores, respectively (see Table 9-2 ) . Respective test-retest correla­ tions were . 86 , . 89, and .91 for the Verbal , Performance, and Full Scales. The changes in subtest scaled scores, which probably were due for the most part to practice effects , ranged from - . 2 (Sentences) to + 1 . 3 (Mazes); changes were significant on the Information ( + . 7 ) , Sim­ ilarities ( + . 7 ) , Animal House ( + . 4) , Picture Completion ( + .7), Mazes ( + 1 . 3 ) , and Block Design ( + 1 . 0) subtests , but not on the others (Wasik & Wasik, 1970) . Generally, the Performance Scale subtests showed greater practice e ffects than did the Verbal Scale subtests.

Precision Range The precision ranges (or confidence intervals) on the WPPSI are similar throughout the age levels covered by the scale. Table C-14 in Appendix C shows the confidence intervals for the 6 8 , 85 , 90 , 95, and 99 percent levels for each scale. Confidence intervals are provided for indi-

1 95

VALIDITY

Table 9·2 Test·Retest WPPSI IQs for 50 Children Between and 53;' Years of Age

First testing

5 '1.

Second testing

Scale

M IQ

SD

M IQ

SD

Change

Verbal IQ Performance IQ Full Scale IQ

1 04 . 5 103 . 8 105 . 6

14.7 14. 1 14.8

1 07 . 5 1 1 0.4 109 . 2

13.9 13.7 13.3

+ 3 .0 + 6. 6 + 3.6

Source: Adapted from Wechsler ( 1 967) .

vidual age levels a s well a s for the average of the age levels.

The child's specific age group, and not the average o/ the six age groups, should be used to obtain the most accurate confidence interval. For further discussion of precision ranges and how to report them, see Chapters 2 and 23 , res pecti vel y.

VALI DITY

The WPPSI manual presents minimal information about the validity of the scale . Since the publication of the WPPSI, however, a number of concurrent and predictive validity studies have been reported .

WPPSI and Stanford-Binet : Form L-M Most of the studies correlating the WPPSI and the Stan­ ford-Binet: Form L-M were done before the 1972 norms became available , so the 1960 Stanford-Binet norms were used. Studies using the 1960 Stanford-Binet : Form L-M norms indicated that the WPPSI had satisfactory concur­ rent validity (Mdn r = . 82) (Anthony, 197 3 ; Austin & Carpenter, 1970; Bach , 1968; Barclay & Yater, 1969; Crockett, Rardin , & Pasewark , 1975 ; Dokecki, Frede , & Gautney, 1969 ; Fagan, Broughton, Allen, Clark, & Emer­ son , 1969; Kaufman, 1973b; Oakland, King, White, & Eckman , 197 1 ; Pasewark , Rardin, & Grice, 197 1 ; Plant, 1967; Prosser & Crawford , 197 1 ; Rellas, 1969; Richards, 1968 ; Ruschival & Way, 1971 ; Sewell, 1977; Zimmerman & Woo-Sam , 1970) . The W PPSI Verbal Scale correlates more highly with the Stanford-Binet: Form L-M than does the Pe rfo r m a n ce S c a l e ( Mdn r's = . 8 1 a n d . 6 7 , respectively ) . Before i t was renormed in 1972 , the Stanford-Binet: Form L- M in many cases yielded substantially higher IQs than did the WPPS I . Therefore IQs provided by the 1960

Stanford-Binet: Form L-M norms were not interchange­ able with those of the WPPS I . With the 1972 norms, for children in the 4- to 6-year-old age range the Stanford­ Binet: Form L-M yields IQs that are on the average about 7 to 10 IQ points lower than they were with the 1960 norms. Therefore it is likely that the two scales now yield more similar IQs . There are few published research studies comparing the 1972 Stanford-Binet: Form L-M norms with the WPPSI, however. In one study (Sewell, 1977) with 35 black children, the WPPSI yielded somewhat higher Full Scale IQs than did the Stanford-Binet: Form L-M (M Full Scale IQ = 96 vs . 91; r = . 7 1 ) .

WPPSI and Stanford-Binet Intelligence Scale: Fourth Edition In a comparison of the WPPSI and the Stanford-Binet Intelligence Scale : Fourth Edition , a sample of75 children (M age = 5 years, 6 months) were administered the two tests in counterbalanced order (52 percent received one order, 48 percent the other) (Thorndike et aI . , 1986b) . On the WPPSI, the three mean IQs were 108 . 2 (Verbal Scale) , 110 . 3 (Performance Scale) , and lIO . 3 (Full Scale) . On the Fourth Edition, the scores were 109 . 8 (Verbal Reasoning) , 100 .4 (AbstractIVisual Reasoning) , 102 . 2 (Quantitative Reasoning ) , 104 . 5 (Short-Term Memory ) , and 105 . 3 (Composite) . Correlations between the Fourth Edition composite score and the Verbal , Performance , and Full Scale IQs were . 78, . 71 , and . 80, respectively. The Fourth EJition thus yielded a mean composite score that was 5 points lower than the WPPSI Full Scale IQ. These two tests do not appear to yield interchangeable scores.

WPPSI and WISC-R: Chronological Ages in Common As we saw in Chapter 6, the WPPSI and the WISC-R overlap for a six-month period, from 6 to 6 '/2 years. The three studies cited in that chapter (Quereshi & McIntire, 1984; Rasbury et aI . , 1977; Wechsler, 1974) indicated that the meaning of differences between WPPSI and WISC-R IQs is not clear. WPPSI and WISC-R IQs may or may not be directly interchangeable, depending on the population . Further research is needed to explore the comparability of these two tests .

WPPSI and WISC-R: Predictive Validity Studies A remarkable longitudinal study (Yule, Gold , & Busch, 1982) examined the relationship of W PPSI scores to WISC-R and achievement test scores obtained 11 years

CHAPTER 9

1 96

WECHSLER PRESCHOOL AND PRIMARY SCALE O F INTELLIGENCE (WPPSI)

later. Eighty-five normal British children, believed to be representative of their community, were administered the WPPSI at 5 V2 years of age , and then the WISC-R and achievement tests at l 6 1h years of age . Correlations be­ tween the two intelligence tests were extremely high, rang­ ing from . 73 to . 86 for the Verbal, Performance , and Full Scales; in addition, IQs on the three scales differed by fewer than 4 points (see Table 9-3) . Correlations between the WPPSI and three achievement tests were also ex­ tremely high (see Table 9-4) . For example, the correlation between the Full Scale IQ and Mathematics was an ex­ tremely high r = . 7 2 , accounting for 50 percent of the variance. Table 9-3 Means, Standard Deviations, and Correlations Between W PPSI and WISC-R (N 85) =

WISC-R

WPPSI Scale

M IQ

SD

M IQ

SD

r

Verbal IQ Performance IQ Full Scale I Q

1 03 . 26 1 07 . 97 1 06 . 0 1

1 3.91 1 4 . 24 14. 1 8

99.44 1 04 . 78 1 02 .09

1 6 .06 1 5 . 30 1 5 .40

. 82 .73 . 86

Source: Adapled from Yule, Gold, and Busch ( 1 982 ) .

Another l l-year longitudinal investigation studied a sam­ ple of 51 Welsh children with spina bifida (Tew & Lau­ rence , 1983) . The children were given the WPPSI at 5 years of age and the WISC-R at 16 years of age . Correla­ tions were extremely high : . 88 for the Verbal Scale IQ, . 90 for the Performance Scale I Q , and .92 for the Full Scale IQ . Mean IQ differences were less than 6 points (for example, WPPSI M Full Scale IQ = 8 1 . 3 , WISC-R M Fuli Scale IQ = 75.6). These results again attest to the remark­ able stability of IQs obtained by children over many years. In still another longitudinal study of the WPPSI, 139 British children were administered the WPPSI at 4 1/2 years of age and the WISC-R at 8 1h years of age (Bishop & Butterworth , 1979) . Correlations between the two tests once again were extremely high (r = .66 for the Verbal Scale, r = . 70 for the Performance Scale , r = .75 for the Full Scale). The Verbal-Performance discrepancy correla­ tion was weaker (r = .42) than the correlations between IQs . The IQs on the three scales differed by 4 or fewer points and were remarkably stable over the four-year pe­ riod covered in the investigation.

Table 9-4 Correlations Between WPPSI and Scores on T h ree Achievement Tests

WPPSI subtest or scale

Achievement tests Reading

Spelling

Mathematics

Information Vocabulary Arithmetic Similarities Comprehension

. 39 .62 .56 .48 .43

.25 .49 .55 .47 .22

. 42 .57 . 68 . 52 .48

Animal House Picture Completion Mazes Geometric Design Block Design

.48 . 39 . 34 . 34 .45

.29 . 36 . 44 .39

.47 .43 .53 .56 . 63

Verbal IQ Performance I Q Full Scale IQ

.61 .53 .61

.48 .5 1 .53

. 65 . 69 . 72

.44

Note. The achievement tests were the Sentence Reading Test NS6, Ver­ non's Graded Spelling, and Vernon's Graded Arithmetic-Mathematics Test. Source: Reprinted, with changes in notation , with permission of the publisher and authors, from W Yule, R. D. GOld, & C . Busch, "Long­ Term Predictive Validity of the WPPS I : An I I -Year Follow-Up Study," Personality and Individual Differences, 1 982 , 3, p. 69 . Copyright 1982 by Pergamon Journals, Ltd.

WPPSI and McCarthy Scales In a study of the WPPSI and McCarthy Scales of Children's Abilities, the WPPSI y ielded Full Scale IQs that were on average 7 points higher than the McCarthy Scales' General Cognitive Index scores (Ms = 107 . 5 2 and 100 . 7 1 , respec­ tively ; r = . 86), a difference that was significant (Schmits & Beckenbaugh, 1979). The sample consisted of 2 1 pre­ dominantly white middle-class children between the ages of 3-10 years and 5-7 years . These results suggest that the two scales do not yield interchangeable standard scores.

WPPSI and Achievement Tests White middle-class children.

WPPSI predictive valid­ ity coefficients are satisfactory for white middle-class pop­ ulations . For example, White and Jacobs (1979) reported a coefficient of . 5 8 between the WPPSI Full Scale and the Gray Oral Reading Test, administered one to three years after the WPPSI, for a group of 28 white middle-class

1 97

VALIDITY

children. Kaufman (I973b) found correlations of . 30 and . 3 7 between the WPPSI and the Mathematics and Reading parts of the Metropolitan Achievement Tests , respectively. The achievement tests were administered approximately four months after the WPPSI to a group of 31 white middle­ class 6-year-olds. Pasewark , Scherr, and Sawyer (\974) reported a significant correlation between the WPPSI and the Metropolitan Achievement Tests total score (r = .58) for a sample of 30 normaI 6-year-olds . Reynolds, Wright, and Dappen (1981) reported significant correlations be­ tween the WPPSI FulJ Scale and the Wide Range Achieve­ ment Test Reading (r = . 3 6) , Spell ing ( r = . 36), and Arithmetic (r = . 60) in a sample of 60 6-year-olds (mostly white and middle class) who had been referred for psycho­ logical evaluation. Krebs (1969) investigated the effectiveness of the WPPSI in predicting reading scores. The WPPSI was adminis­ tered to 70 kindergarten-age children (34 boys and 36 girls ) , equally divided between lower and upper so­ cioeconomic status groups. One year later, when the chil­ dren were in first grade, the reading section of the Stanford Achievement Test and the Gilmore Oral Reading Para­ graphs Test were administered . As Table 9-5 shows, all of the WPPSI subtest and scale scores were significantly related to reading scores on both tests . When the data were reanalyzed based on socioeconomic class, the WPPSI scores were found to have higher correlations with reading scores of the lower socioeconomic status group than with those of the upper socioeconomic status group. For exam­ ple, in the lower socioeconomic status group, the correla­ tions between the total reading score on the Stanford Achievement Test and the Verbal, Performance, and Full Scales were . 59, .61, and . 66, respectively, whereas in the upper socioeconomic status group, the correlations were . 3 2 , . 3 5 , and .40, respectively. In the total sample, Arith­ metic and Geometric Design were the two subtests that best predicted read ing achievement on the Stanford Achievement Test (R = . 63 ) . The WPPSI, however, was not found t o discriminate between good and poor readers in a sample of children initially tested with the WPPSI at 5 years of age and given a reading test 3 years later (Badian, 1984). The children tested were from the lower middle to the middle class in a predominantly white area and were at risk for learning disabilities when initially tested.

Ethnic minority children and low SES children. With ethnic minority children and c h ildren of lower so­ cioeconomic status, the predictive validity of the WPPSI is

Table 9-5 Correlations Between WPPSI and Scores on Two Reading Tests

Gilmore Oral Reading Paragraphs Test

Stanford Achievement Test: Reading

Information Vocabulary Arithmetic Similarities Comprehension Sentences

.49 . 52 . 54 .48 . 36 . 54

. 52 . 53 . 58 .53 .38 .55

Animal House Picture Completion Mazes Geometric Design Block Design

.4 1 .43 .42 .52 .44

.46 .47 .47 . 54 .49

Verbal IQ Performance IQ Full Scale IQ

. 57 .58 . 62

.61 .63

WPPSI subtest or scale

. 68

Note. All correlation coefficients are significant.

Source: Reprinted by permission of the author from E. G. Krebs, "The Wechsler Preschool and Primary Scale of Intelligence and Prediction of Reading Achievement in First Grade" ( Doctoral dissertation. Rutgers State University). Ann Arbor, Mich . : University Microfilms. 1 969, pp. 73a-73b, No. 70-336 1 . Copyright 1 970, E. G. Krebs.

more variable than it is with white middle-class children. In one study (Crockett, Rardin, & Pasewark, 1976), a sample of Head Start children were administered the Met­ ropolitan Achievement Tests three to four years after ad­ ministration of the WPPSI. The only significant correla­ tions were between Mathematics and the Full Scale IQ (r = .43) and between Mathematics and the Performance Scale IQ (r = . 5 2 ) . The investigators indicated that WPPSI Verbal Scale scores must be viewed with caution in the evaluation of Head Start children because many of these children are of lower socioeconomic status, bi­ lingual , and evidence atypical development of language skills during primary school years. In another study with a group of 50 black, 32 white, 4 Hispanic, and 1 Oriental child, all of whoin were of lower socioeconomic status and had been referred to a special setting because of personal and social difficulties, the WPPSI Full Scale IQ was signif­ icantly correlated with speech development (r = . 6 1 ) , motor development (r = .41) , and perceived improvement

1 98

CHAPTER 9

WECHSLER PRESCHOOL AND PRIMARY SCALE OF INTELLIGENCE (WPPSI)

in a specialized program ( r = . 73) (Dlugokinski , Weiss, & Johnston , 1976) . Based on their work with Mexican-American 5 '/2 -year­ old children from economically depressed areas , Hender­ son and Rankin (1973) suggested that it is ill advised to use the WPPSI routinely in making decisions as to whether to place these children in special classes . They found that there was an 18-point difference between the children's Verbal and Performance Scale IQs (74 vs. 92) , and that the predictive validity of the WPPSI, using third-grade Metro­ politan Reading Tests scores, was poor ( r = .27) . The WPPSI, therefore , appears to have dubious utility as a predictor of later school performance for this group of children . Unfortunately, there is no way of knowing which of the scales is the more valid predictor of future academic performance, because separate correlations for the Verbal and Performance scales were not reported .

WPPSI and Other Tests In 11 studies, correlations between the WPPSI and tests other than those previously covered (English P icture Vo­ cabulary Test, Frostig Developmental Test of V isual Perception , Goodenough-Harris Drawing Test, Hiskey­ Nebraska Test of Learning Aptitude, Ill inois Test of Psy­ cholinguistic Abilities , Metropolitan Readiness Tests, Peabody Picture Vocabulary Test , Pictorial Test of Intel­ l igence, Primary Mental Abilities Test, Progressive Ma­ trices , Stanford Achievement Test, Torrance Tests of Cre­ ative Thinking, Vane Kindergarten Test) have ranged from a low of . 2 8 (Metropolitan Readiness Tests) to a high of . 82 (Primary Mental Abilities Test) (Mdn r = . 64) (Austin & Carpenter, 1970; Bach , 196 8 ; Dokecki et a1 . , 1969; Kaplan, 1985 ; Krebs , 1969 ; McNamara, Porterfield, & Miller, 1969; Plant, 1967 ; Plant & Southern, 196 8 ; Tsushima & Stoddard, 1986; Yater, Barclay, & Leskosky, 197 1 ; Yule, Berger, Butler, Newham, & Tizard, 1969 ) . In nine of these investigations , lower-class, ethnic minority group, or Head Start children were studied; therefore , it is difficult to know whether the findings are generalizable to other groups of children . WPPSI scores also have been found to be related significantly both to perceptual development and to creativity (Lichtman , 1969; Yule et a1 . , 1969) .

The predictive validity of the WPPSI may be related to the specific scale (Verbal , Performance, or Full Scale) and the length of time between administration of the WPPSI and that of the criterion measure . Overall , recent research suggests that the WPPSI administered at kindergarten age may serve as an excellent long-term predictor of intel­ l igence and school performance measured during late adolescence.

I NTERCORRE LA T I O N S B E TWE E N SU BTESTS A N D SCALES

Intercorrelations between subtests and scales permit us to observe the degree of relationship between various parts of the WPPS I . Average intercorrelations between the 11 sub­ tests range from a low of . 2 8 to a high of . 60 (Mdn = . 42 ) . The s i x highest subtest average intercorrelations are be­ tween Information and Vocabulary ( . 60) , Information and Comprehension ( . 60) , Information and Arithmetic ( . 5 8 ) , Vocabulary and Comprehension ( . 57), Similarities and Comprehension ( . 55 ) , and Sentences and Comprehension ( . 5 3 ) . The six lowest subtest average intercorrelations are between Similarities and Mazes ( . 28), Similarities and Geometric Design ( . 30), Sentences and Mazes ( . 30), Sim­ i larities and Animal House ( . 3 1 ) , Comprehension and Mazes ( . 3 3 ) , and - tied at . 34 - Animal House and Com­ prehension and Sentences and Geometric Design. Average correlations between the Verbal Scale subtests and the Verbal Scale range from . 62 to . 73 (Mdn = . 65 ) , whereas those between the Performance Scale subtests and the Performance Scale range from . 50 to . 60 (Mdn = . 57) . Intercorrelations between the individual subtests and the Full Scale show similar trends. Vocabulary has the highest correlation of any of the subtests with the Full Scale ( . 70) , followed by Arithmetic ( . 68), and Comprehension ( . 65) . Animal House ( . 5 3 ) and Mazes ( . 54) have the lowest correlations with the Full Scale . Within their respective scales, Vocabulary ( . 73) and Geometric Design ( . 60) have the highest correlations with the Verbal and Perfor­ mance IQs.

WPPSI I Q S AND STRAT I F I CATIO N

Comment on Concurrent and Predictive Validity of the WPPSI The studies reviewed in this section attest to the concurrent and predictive validity of the WPPSI with most popula­ tions. Further work is needed to determine its validity with Mexican American and other ethnic minority children .

VARIABLES

Geographic and Socioeconomic Differences A study of the relationship of WPPSI IQs to socio­ economic status (SES) (as determined by father's occupa­ tion), urban v s . rural residence , and geographic region of

1 99

WPPSI IQS AND STRATIFICATION VARIABLES

children in the standardization group revealed s( me important trends (Kaufman, 1973c ) . With respect to SES, the largest difference, about 18 points, was found between children of unskilled laborers and those of professional men . Differences among SES levels 2, 3 , and 4 were not significant (see Table 9-6) . Urban-rural residence was not a significant factor, but geographic region was . Children from the West had significantly higher IQs than did chil­ dren in the other three regions; IQs of children in the other three regions did not differ significantly. There is no read­ ily available explanation tel account for the regional differ­ ence . SES differences found on the WPPSI corroborate those found on the WISC-R (Kaufman & Doppell. 1976) , WISC (Seashore, Wesman , & Doppelt, 1950), and Stan­ ford-Binet Intelligence Scale: Forms L and M (McNemar, 1942) , Stanford-Binet Intelligence Scale: Fourth Edition (Thorndike et a1 . , 1986b) , and WAIS-R (Chastain & Rey­ nolds , 1984) .

Gender Differences Gender differences have been found on some WPPSI sub­ tests. In the standardization sample, boys obtained signifi­ cantly higher scores than did girls on the Mazes subtest , but girls obtained higher scores than did boys on the Animal House , Geometric Design, Block Design, an d

Sentences subtests (Herman, 1968 ) . Thus, Performance IQs may be achieved in different ways by boys and girl s . In a sample of 5 1/2 -year-old British children, boys obtained higher WPPSI IQs than did girls (M IQs of 104 v s . 98) (Brittain, 1969 ) . Significant sex differences in IQ were not found by Ruschival and Way (1971) in an American sam­ ple , however. Generally, for purposes of individual assess­ ment, sex differences on the WPPSI do not appear to play an important role .

Ethnic Differences Kaufman (1973a) compared the scores of 132 matched pairs of black children and white children in the WPPSI standardization group . The children were matched on age , sex, geographic region , father's occupation , and urban­ rural residence . The white children obtained significantly higher Verbal IQs (99 vs. 88), Performance IQs (97 vs. 88) , and Full Scale IQs (98 vs. 87) at all of the age levels studied; differences averaged about 11 IQ points. On the Verbal Scale , significant differences between the two ethnic groups were found at each age level. On the Performance Scal e , however, differences between the two ethnic groups were significant at the 4-year-old level, but not at the 5- and 6-year-old levels . The differences de­ creased with age , almost in a l inear fashion : 16 points at 4

Table 9-6 Means and Standard Deviations o( WPPSI IQs (or SES, Residence, and Region

Demographic variable SES (Father's occupation) 1 . Professional & Technical 2 . M anager, Clerical, & Sales 3 . Skilled 4 . Semiskilled 5 . Unskilled Residence Urban Rural Region Northeast North Central South West

Verbal lQ

Performance IQ

N

M

SD

M

SD

M

SD

78 78 78 78 78

1 10 102 100 98 93

13 13 15 14 14

1 08 1 02 1 00 99 93

12 13 14 15 16

1 10 1 02 1 00 99 92

12 12 15 14 15

325 325

99 99

14 14

98 99

15 14

98 99

15 14

1 78 1 78 1 78 178

101 101 98 105

14 15 15 13

102 1 00 98 1 04

14 15 16 13

1 02 1 02 98 1 05

14 16 15 13

Full Scale IQ

Source: Reprinted, with a change in notation, by permission of the publisher and author from A . S . Kaufman, "The Relationship ofWPPSI IQs to SES and Other Background Variables," Journal of Clinical Psychology. 29, p. 356. Copyright 1 973 , Clinical Psychology Publishing Co. , Inc .

200

CHAPTER

9

WECHSLER PRESCHOOL AND PRIMARY SCALE O F INTELLIGENCE (WPPSI)

years, 11 points at 4 1/2 years, 8 points at 5 years, 7 points at 5 112 years and 6 years, and 5 points at 6 Y2 years . This trend suggests that the deficit was not cumulative . It is not easy to explain why the older black children performed at a level so much higher than that of their younger peers.

FACTOR ANALYSIS

A factor analysis of the WPPSI, using the standardization sample , y ielded two principal factors: Verbal and Perfor­ mance (Carlson & Reynolds , 198 1 ) . The factor loadings associated with the two factors are shown in Table 9-7 . The Verbal factor is best represented by the six Verbal Scale subtests (Information , Vocabulary, Comprehensio n , Arithmetic , Similarities, and Sentences ) , whereas the Per­ formance factor is best represented by the five Perfor­ mance Scale subtests (Block Design, Mazes, Geometric Design, Picture Completion , and Animal House) . Al­ though the factor loadings vary across age levels, the general trends noted can be applied to the various age levels covered by the scale. The results support the divi­ sion of the WPPSI into Verbal and Performance Scales, provide construct validity for the test, and suggest that, for children between the ages of 4 and 6 , the WPPSI may be a more sensitive instrument for the purpose of assessing the

structure of intelligence than is the Stanford-Binet: Form L-M , which provides only a global index of intelligence.

WPPSI Subtests as Measures of g A hierarchical factor analysis, using the WPPSI standard­ ization sample, indicated that all 1 1 WPPSI subtests have strong loadings on the general intell i gence factor (g) (Wallbrow n , Blaha , & Wherry, 1973) (see Table 9-8) . The loadings are relatively similar across the age levels of the test, suggesting that differentiation of abilities is not appar­ ent during the ages covered by the scal e . Although the Verbal subtests tend to have the highest g loadings, the range of median loadings for the 1 1 subtests is relatively narrow ( . 55 to . 72) . The g factor accounts for the largest percentage of subtest variance (39 percent) .

WPPSI Subtest Specificity With the exception of Information and Comprehension, WPPSI subtests have enough specificity to warrant specific clinical interpretation across the age range covered by the scale (Carlson & Reynolds, 1981) . Thus most subtests allow for individual interpretation of a child's strengths and weaknesses.

Table 9·7 Factor Loadings of WPPSI Subtests for Six Age Groups (Varimax Rotation)

Faclor A

-

Verbal

-

4a

40

5

50

6

60

Mdn

4a

40

Information Vocabulary Arithmetic S i milarities Comprehension Sentences

71 63 50 75 71 66

74 70 56 69 77 66

66 63 54 57 73 53

74 68 61 66 71 61

69 70 54 61 75 55

74 62 55 72 71 65

72 65 54 67 72 63

32 31 45 17 29 27

A nimal House Picture Completion Mazes Geometric Design Block Design

46 43 15 27 32

40 49 07 27 29

25 37 29 24 22

32 37 30 17 35

25 42 42 25 45

25 31 10 20 34

28 40 22 24 33

29 51 77 62 50

Sublesl

FaclOr 8 Performance 5

50

6

60

Mdn

29 18 42 26 17 28

37 31 51 13 28 29

40 31 51 17 31 30

34 32 46 31 25 39

31 35 54 23 22 21

33 31 48 21 27 28

43 42 65 57 56

60 50 57 73 72

56 53 61 80 68

63 54 58 75 55

42 49 65 62 61

49 50 63 67 58

Note. Decimals omitted. " This row indicates age leve l . Source: Reprinted. with changes in notation, with permission from the publisher and authors, from L . Carlson & c . R . Reynolds, "Factor Structure and Specific Variance of the WPPSI Subtests at Six Age Levels," Psychology ill the Schools, I 98 I , J 8, p. 5 1 .

20 1

ADMINISTERING THE WPPSI

Table 9-8 W PPSI Subtests as Measures of g

Median loading

Proponion of variance attribured to g

Information Vocabulary Arithmetic Similarities Comprehension Sentences

. 72 .64 .68 . 63 . 68 . 64

52 41 46 40 46 41

Animal House Picture Completion Mazes Geometric Design Block Design

. 55 . 60 .57 .64 . 62

30 36 32 41 38

Subtest

Nore. The square of the median coefficient provides the proportion of each subtest's variance that may be attributed to g . SOl/ree: Adapted from Wal lbrown, Blaha, and Wherry ( 1 97 3 ) .

children as it does for white children . These results suggest that WPPSI is a fair instrument in the sense that it measures the same dimensions of intellect in both ethnic groups.

Factor Structure for Two White Subcultural Groups A study (Heil , Barclay, & Endres, 1978) of the WPPSI factor structure of two groups of white preschool children , one labeled "educationally deprived" (educationally handi­ capped as a result of poverty, neglect, del inquency, or cultural or l i nguistic isolation from the community at large) and the other normal , found essentially the same three factors in both groups: a Verbal factor, a Perfor­ mance factor, and a factor limited to the Picture Comple­ tion subtest . These results, which generally support those found for samples of black and white children, indicate that sociocultural differences do not produce different in­ tellectual structures on the WPPSI . Other studies also support the Verbal and Performance dichotomies on the WPPSI (Haynes & Atkinson , 1984; Ramanaiah & Adams, 197 9 ; Silverstein, 1986a) .

Factor Analysis of WPPSI Compared to WISC-R The factor structure of the WPPSI is generally similar to that of the WISC-R, particularly in regard to the Verbal Comprehension and Perceptual Organization factors . The primary difference between WPPSI and WISC-R factor analytic findings is that a Freedom from Distractibi l ity factor emerges on the WISC-R but not on the WPPSI . Conceivably, sustained directed attention, partly measured by the Freedom from Distractibility factor, is a part of every subtest at younger age levels and emerges as a separate factor only at older age levels. Alternatively, it may be that the Freedom from Distractibility factor does not emerge on the WPPSI because the test does not include either Digit Span or Coding ; it is these two subtests, along with Arithmetic, that have generally defined Freedom from Distractibility on the WISC-R.

WPPSI Factor Structure for Black Children and White Children For the black children and white children in the WPPSI standardization sample , the same two factors, Verbal and Performance , were found in each group (Kaufman & Hol ­ lenbeck, 1974) . Various statistical procedures used to test the similarity of factor loadings also indicated that the WPPSI has virtually the same factor structure for black

ADM I N ISTERI N G T H E W PPSI

The general administrative suggestions described for the WISC-R in Chapter 6 are also appropriate for the WPPSI. The two scales have common problems in adm inistration and scoring . Some 5ubtest names are the same in both scales, so be careful not to substitute WISC-R directions for WPPSI directions or vice versa . The suggestions shown in Exhibit 9-2 , which supplement those given in other parts of this chapter, should aid you in learning to administer the WPPSI . Figure 9-2 shows the cover of the WPPSI record bookJet .

Physical Abilities Necessary for the WPPSI The physical abilities children need in order to take the WPPSI are , for the most part, the same as those required for the WISC-R (see Table 6-11 in Chapter 6) . Adequate visual-motor skills, in particular, are needed to handle the Performance Scale materials. Alternative ways of admin­ istering the WPPSI items are restricted by the fact that young children have l i mited writing and reading ski l l s . Children who cannot speak usually w i l l not be able t o write their answers, and those who cannot hear usually will not be able to read the questions . The specific suggestions for

202 �

CHAPTER 9

WECHSLER PRESCHOOL AND PRIMARY SCALE O F INTELLIGENCE (WPPSI)

Exhibit 9-2 Adm inistering the WPPSI

I . Complete the top of the record booklet. 2. Using date of testing and date of birth, calculate the chronological age (CA) and put it in the box provided . On the WPPSI CA must be stated in years, months, and days. 3. Administer the subtests in the order presented in the manual , except in rare circumstances . Do not change the wording on any subtest . Read the directions exactly as shown in the manual . Do not ad lib . 4. Start with the appropriate item on each s ubtest and follow discontinuance criteria. You must know correct scoring criteria before you give the test. 5. Write out all responses completely and legibly. Do not use unusual abbreviations. Record time accurately. 6. Question all ambiguous or unscorable responses, writing a (Q) after each questioned response. 7. Be patient when working with children in the WPPSI age group. Several breaks may be needed d uring the testing. 8. O n Comprehension , i f a child only gives one reason to Questions 7 , 8 , 9, 10, 14, and 15, request a second reason. These questions are marked with an asterisk ( * ) in the manual . 9. Carefully score each protocol , recheck the scoring, and transfer subtest scores to the front of the record booklet u nder Raw Score. If you have failed to question a re­ sponse when you should have and the response is ob­ viously not a 0 response, give the child the most appro­ priate score. 1 0 . If a subtest was spoiled, write spoiled by the subtest total score and on the front cover where the raw and scaled scores appear. If for some reason a subtest was not administered, write NA in the margin in the record book­ let and on the front cover. I I . Raw scores are transformed into scaled scores through use of Table 21 on page 90 of the WPPSI manual . Be sure to use the page of the table that is appropriate for the child's age and the correct row and column for each transformation. 1 2 . The Verbal Score is based on the total of the scaled scores on the five standard Verbal Scale subtests. The Perfor-

administering the WISC-R to handicapped children are also useful for the WPPSI, however; you should review the material in Chapter 6 carefully before administering the WPPSI to physically handicapped children.

13.

14.

15.

16.

17.

18.

mance Score is based on the total of the scaled scores on the five Performance Scale subtests. Do not use Sen­ tences to compute the Verbal Score unless you substitute it for another Verbal subtest. Add the Verbal Score and the Performance Score together to get the Full Scale Score. If it should be necessary to prorate either the Verbal or the Performance section, follow the explicit directions g iven on page 43 of the WPPSI manual. You will also need to use Table 24 on page 106 of the manual . The IQs are obtained from Table 22 of the W PPSI man­ ual . Be sure to use the correct section of the table for each of the three IQs. There is a Verbal section and a Perfor­ mance section on page 101, and a Full Scale section on page 102 . Record the IQs. Next , recheck all of your work. If the IQ(s) was (were) prorated, write PRO beside the appropriate IQ(s). Look up the confidence intervals for the Full Scale IQ, Verbal Scale IQ, and Performance Scale IQ in Table C-14 in Appendi x C . Look up the percentile rank and classification for each of the IQs in Tables BC-I and BC-2 on the inside back cover of this text. I f desired, use the material on pages 104 and 105 ( Table 2 3 ) of the WPPSI manual to obtain Test-Age Equiv­ alents. They can be placed (in parentheses) in the right­ hand margin of the cover page of the record booklet next to the scaled score. For Test-Age Equivalents above those in the table, use the highest Test-Age Equivalents and a plus sign. For Test-Age Equivalents below those in the table, use the lowest Test-Age Equivalent and a minus sign. I n summary, be sure to read directions verbatim, pro­ nounce words clearly, query at the appropriate times, start with the appropriate item, discontinue at the proper place, place items properly before the child, use correct timing, and follow the specific guidelines in the manual for administering the test.

Source: Courtesy of M. L. Lewis

Testing-or-Limits on the WPPSI The general testing-of-limits suggestions presented in Chapter 5 are also useful with the WPPSI .

ADMINISTERING

THE

203

WPPSI

WPPSI

_ N A M E ______ AGE ____ SEX __

RECORD

ADDRESS __ __ __ __ __ __ __ __ __ __ __

FORM

__ __ __ PARENT'S N A M E __ __ __ __ __ __ __ __ __ __ __ __ __

Wechsler Preschool and Primary

SCHOOL _______ G RADE__ __ __ __ _

Scale of Intelligence

__ __ PLACE OF TESTING ______ TESTED BY __ _

REFERRED BY

N O T E S Year

Monlh

Day

Dale Tesled Dale of Birth Age

Scaled Score

Raw Score VERBAL TESTS Information Vocabulary Arilhmelic Simila rities Comprehension ISentences) Verbal Score PERFORMA NCE TESTS Animal House Picture Complelion Mazes Geometric Design Block Design IAnimal House Retest)

_ _ _

Performance Score

Scaled Score

Verbal Score

IQ

---- *--*

Performance Score Full Scale Score

*Prorated jf necessary

Copyright 1 949, renewed 1 9 7 6 ;

©

1 9 6 3 , 1967 by T h e P s y c h o l o g i c a l C o r p o r a t i o n .

A l l r i gh IS reserved. N o p a r t of t h i s r e c o r d f o r m may b e reproduced i n a n y form o f p r i n t i n g or by a n y other

means,

recording

Printed i n U. S. A.

Figure 9-2.

and

electronic

or mechanical,

in

transmission,

system. without permission

and

i n c l u d i ng,

POrtrayal

bUI

not

l i mited

to,

phOtocopying,

audiovisual

or d u p l i c a t i o n i n a n y i n f o r m a t i o n Storage and

retrieval

writing f r o m t h e publisher. See Catalog f o r further i n f o r m a t i o n .

A l l r i g h l S reserved under t h e B e r n e Conve n t i o n . T h e PSYChological C o r p o r a t i o n . N e w Y o r k

7 7 - 1 53AS

Cover page of W PPSI record booklet. Copyright 1 949, renewed 1 976; © 1 963, 1 967 by The Psychological Corporation, San

Antonio, TX. All Rights Reserved.

204

CHAPTER 9

WECHSLER PRESCHOOL AND PRIMARY SCALE OF INTELLIGENCE (WPPSI)

Extrapolated IQs Extrapolated IQs are provided in Table C - 17 in Appendix C for scaled scores that are either below or above those shown in the WPPSI manual . Extrapolated IQs, however, must be used cautiously, as noted in Chapter 6 .

WPPSI Short Forms Short forms of the WPPSI have the same disadvantages as those of the WISC-R (see Chapter 6) and should never be used for classification or selection purposes. A short form may be useful for screening or research studies, however. The information in Table C-21 in Appendix C can aid you in the selection of a short form . This table, based on the standardization data, shows the best WPPSI short forms for combinations of two, three, four, and five subtests. Because the short forms of a given length are , for all practical purposes, mutually interchangeable , you can use clinical or other considerations to select the short form .

Obtaining a Deviation Quotient from the short form . After the short form has been selected , follow the pro­ cedures outlined in Chapter 6 for converting WISC-R composite scores to Deviation Quotients (see Exhibit 6-3 ) . Table C-36 in Appendix C can b e used t o obtain the appropriate a and b constants . The only difference be­ tween the procedure for the WISC-R and the one for the WPPSI is that for the WPPSI the correlation table of the group that is closest in age to the examinee should be used to obtain f.rjk (that is, one of the tables on pages 26 through 31 of the WPPSI manual ) .

A variety of WPPSI short forms. Dokecki et al . (1969) found that the four subtests that comprised the best short form for middle-class children were Information, Com­ prehension , Arithmetic , and Geometric Design (r = . 94 with the Full Scale) . The best two-subtest short form for predicting reading achievement in a group of lower-class children consisted of Information and Geometric Design (Plant & Southern, 1968) . Kaufman 's short form. Kaufman ( 1972a) proposed a four-subtest short form composed of the Arithmetic, Com­ prehension, Picture Completion, and Block Design sub­ tests . This short form was highly related (r = . 96) to the Full Scale in a sample of 116 outpatient children, with mean IQs showing a difference of less than 3 points (Full Scale IQ = 1 12 . 1 1 , four-subtest short form IQ = 109 . 76) (Haynes & Atkinson, 1983) . There was only a 61 percent agreement in IQ classifications between the short form and the Full Scale, however.

Yudin 's WPPSI short form . Yudin's (1966) short­ form method for the WISC-R, in which the number of items within the subtests is reduced , has been applied to the WPPSI by Silverstein (1968a) . Table C-22 in Appendix C shows the specific procedures. Silverstein suggested that Sentences be excluded from the short form because the subtest was omitted in establishing the IQ tables. The fol lowing reliabilities were reported for Yudin's short form: .91 for the Verbal Scale, . 9 1 for the Performance Scale , and . 94 for the Full Scale . WPPSI Vocabulary plus Block Design short form. The Vocabulary plus Block Design combination is useful as a two-subtest screening short form . When this combina­ tion is used, Table C-37 in Appendix C can be used to convert the sum of scaled scores on the two subtests di­ rectly to an estimate of the Full Scale IQ. Correlations of . 82 and . 89 have been reported between the Vocabulary plus Block Design short form and the Full Scale IQ (Sil­ verstein, 1970b, and Haynes & Atkinson, 1983 , respec­ tively) . Studies (Haynes & Atkinson , 1983 ; King & Smith, 1972) also report that the Vocabulary plus Block Design short form agreed with the IQ classifications obtained from the Full Scale in only about 3 3 to 35 percent of the cases .

These results reinforce my recommendation that short forms never be used for classification or selection purposes.

WPPSI S U BTESTS

This section describes the 11 WPPSI subtests . The factor analytic findings are from Carlson and Reynolds (1981) and from Wallbrown et al . (1973) .

Information The Information subtest contains 23 questions, 12 of which, with minor changes in wording, come from the WISe . Most questions require the child to give a simply stated fact or facts . All items are scored 1 or 0 (pass-fail) , and the subtest is discontinued after five consecutive failures.

Rationale.

The rationale presented for the WISC-R In­ formation subtest appears to apply to the WPPSI Informa­ tion subtest (see Chapter 7 ) . The WPPSI questions, how­ ever, appear to assess that part of the child's knowledge of the environment that is gained from experiences rather than from education, especially formal education .

205

WPPSI SUBTESTS

Factor analytic findings.

The Information subtest i the best measure of g in the scale (52 percent of its variance may be attributed to g). The subtest contributes substan­ tially to the Verbal factor (Mdn loading = . 72) .

Reliability and correlational highlights.

Information is . 81 ) ; it correlates more highly with Vocabulary and Comprehension (r's = . 60) than with any of the other subtests . It correlates moderately with the Full Scale (r = . 70) and Verbal Scale (r = . 73) and to a lesser degree with the Performance Scale (r = . 56) .

a reliable subtest (rxx

=

A d m inistrative considerations.

The administrative considerations presented for the WISC-R Information sub­ test are also relevant for the WPPSI Information subtest . In addition , some WPPSI items require special scoring con­ sideration s . For example, the answers to question 4 ("What comes in a bottle?") do not mention things that come in plastic bottles . Because the term "etc . appears in the scoring criteria appendix of the manual , however, it seems logical to assume that credit should be given for references to substances that come in plastic bottles, such as shampoo and liquid soap . Likewise , the scoring criteria referring to question 9 ("What shines in the sky at night?") do not include "planet, " yet a planet shines in the sky at night. It is recommended that credit be given for "planet," "comet," and other astronomical terms .

Reliability and correlational highlights.

Vocabulary is a reliable subtest (rxx = . 84 ) . It correlates more highly with Information (r = . 60) than with any other subtest. It correlates moderately with the Full Scale (r = . 65) and Verbal Scale (r = . 66) and to a lesser degree with the Performance Scale (r = . 5 1) .

Administrative considerations. A l l examinees start the subtest with the first word. This procedure differs from the one used for the WISC-R, where the starting word depends on the child's age. The general administrative guideiines presented for the WISC-R Vocabulary subtest should be followed for the WPPSI Vocabulary subtest. Scoring re­ quires considerable judgment, and the WPPSI manual provides too few sample responses . Because many of the Vocabulary words have near homonyms (words I , 1 3 , 14 , 15 , 16, 17 , and 22) , you must carefully articulate each word .

n

Question: What do you call a baby goat? A nswer: Matilda would be a nice name .

Vocabulary The Vocabulary subtest contains 22 words , 14 of which are from the WISe . The child is asked to explain orally the meaning of each word . Each word is scored 2 , 1, or 0, and the subtest is discontinued after five consecutive failures.

Rationale.

The rationale presented for the WISC-R Vo­ cabulary subtest generally applies to the WPPSI Vocabu­ lary subtest (see Chapter 7) . Formal education , however, is less likely to be an influence in vocabulary development for preschool children than for older children. Experi­ ences are l ikely to be the major contributing factor to vocabulary development of preschool children.

Factor analytic findings. The Vocabulary subtest is a moderately good measure of g (41 percent of its variance may be attributed to g) . The subtest contributes substan­ tially to the Verbal factor (Mdn loading = . 65 ) .

Arithmetic The Arithmetic subtest consists of 20 problems, six of which are from the WISe . For the first 4 items , the child points to the correct answer on a card; for the next 4 items, blocks are used ; and for the last 12 items, oral answers are required . The subtest is started at different points , depend­ ing on the child's age and possible level of intellectual functioning. Children under 6 years of age (and older children who may be mentally retarded) start with item I ; children 6 years o f age and older begin with item 7 . The first 8 problems have no time l imit , but the last l2 problems have a 30-second time limit . Each item is scored I or 0, and the subtest is discontinued after four consecutive failures . The problems on the Arithmetic subtest reflect various skills. Problems I through 4 entail perceptual judgments involving the concepts of biggest, longest , most, and same. Problems 5 through 8 require direct counting of concrete quantities. Problems 9 through 20 involve simple addition or subtraction , although simple division or multi­ plication also can be used.

Rationale. The rationale described for the WISC-R Arithmetic subtest appears to apply generally to the WPPSI Arithmetic subtest (see Chapter 7) . The skills required for the WPPSI Arithmetic subtest, however, are likely to be less dependent on formal education than are those required for the WISC-R Arithmetic subtest. The first four WPPSI questions, which require the child to make comparisons and perceptual discriminations, appear to measure nonverbal reasoning ability ; these four prob-

206

CHAPTER 9

WECHSLER PRESCHOOL AND PRIMARY SCALE O F INTELLIGENCE (WPPSI)

lems measure quantitative concepts without involving the explicit use of numbers.

Factor analytic findings.

The Arithmetic subtest is a moderately good measure of g (46 percent of its variance may be attributed to g) . The subtest has a high loading on the Verbal factor (Mdn loading = . 54) and a moderate loading on the Performance factor (Mdn loading = .48) . Its loading on the Performance factor may be accounted for by the fact that some items employ pictures of sets of objects that must be visually analyzed before verbal com­ parisons are made.

Reliability and correlational highlights. Arithmetic is a reliable subtest (rxx = . 82 ) . It correlates more highly with Information (r = . 58) than with any other subtest. It correlates moderately with the Full Scale (r = . 68), Verbal Scale (r = . 62 ) , and Performance Scale (r = . 60) . Administrative considerations.

The administrative considerations discussed for the WISC-R Arithmetic sub­ test generally apply to the WPPSI Arithmetic subtest . Scoring, as in the WISC-R, is for the most part easy - l or 0 points . The time taken by the child to solve each problem should be recorded . On problems 9 through 20, correct answers given after the time l imit has expired should also be noted.

Similarities The WPPSI Similarities subtest consists of 16 questions , 7 of which are found in the WISC . The first 10 questions require simple analogies; the remaining questions are sim­ ilar to those found in the WISC-R Similarities subtest. Items I through 10 are scored I or 0 (pass-fail) , and items 11 through 16 are scored 2, I , or 0, depending on the concep­ tual level of the response. The subtest is discontinued after four consecutive failures.

Rationale.

The rationale described for the WISC-R Similarities subtest generally applies to the WPPSI Sim­ ilarities subtest (see Chapter 7). Because over half of the WPPSI questions ( I to 10) require analogies , however, the subtest may be measuring logical thinking rather than verbal concept formation , especially at the earlier levels (i . e . , below 5 years of age) .

Factor analytic findings. The Similarities subtest is a moderately good measure of g (40 percent of its variance may be attributed to g). The subtest contributes substan­ tially to the Verbal factor (Mdn loading = . 67) .

Reliability and correlational highlights. Similarities is reliable subtest (rex = . 83 ) . It correlates more highly

a

with Comprehension ( r = . 55 ) than with any other subtest. Correlation is relatively low with the Full Scale (r = . 58 ) , moderate with the Verbal Scale (r = . 62 ) , and low with the Performance Scale (r = .44).

Administrative considerations. T h e administrative considerations discussed for the WISC-R S imilarities sub­ test generally apply to the WPPSI Similarities subtest. Scoring procedures, however, differ. Because responses to the first 10 questions are scored I or 0, few scoring prob­ lems should be encountered with these questions. As in the WISC-R, however, the questions dealing with similarities (items 11 through 16) are difficult to score . Scoring guide­ lines should be studied carefully. Comprehension The Comprehension subtest contains 15 questions, 6 of which, with minor changes in wording, are from the WISe . Several content areas are covered, including health and hygiene, knowledge of the environment, and knowl­ edge of activities in society. Items are scored 2, I, or O. The subtest is discontinued after four consecutive failures .

Rationale. The rationale presented for the WISC-R Comprehension subtest appears to apply generally to the W PPSI Comprehension subtest (see Chapter 7). Lin­ guistic skill and logical reasoning, however, may play a more important role on the WPPSI Comprehension subtest than on the WISC-R Comprehension subtest. Factor analytic findings.

The Comprehension subtest is a moderately good measure of g (46 percent of its variance may be attributed to g). The subtest contributes substantially to the Verbal factor (Mdn loading = 72 ) .

.

Reliability and correlational highlights. Comprehen­ sion is a reliable subtest (ra = . 81 ) . It correlates more highly with Sentences (r = . 53) than with any other sub­ test. It correlates moderately with the Full Scale (r = . 65 )

and the Verbal Scale ( r = . 69) and t o a lesser degree with the Performance Scale (r = . 44).

Administrative considerations . The administrative considerations discussed for the WISC-R Comprehension subtest generally apply to the WPPSI Comprehension sub­ test. Because Comprehension responses are occasionally difficult to score , judgment is needed to arrive at appropri­ ate scores. As in all decisions on the scoring of WPPSI and

207

WPPSI SUBTESTS

WISC-R responses , the content of the response . 101 the quality of the verbalization, should be considered.

words in the sentences may have more serious memory problems.

Animal House and Animal House Retest Sentences The Sentences subtest, which is a supplementary subtest , contains 13 sentences of increasing length, ranging from 2 to 18 words. The child is required to repeat each sentence verbatim, after listening to the examiner say it. The subtest is discontinued after three consecutive failures. When it i s administered a s a sixth Verbal Scale subtest, the subtest is not used in calculating the I Q . Items are scored 1, 2, 3 , or 4 points, depending on the length of the sentence and the number of errors made. Errors in reproducing the sen­ tences include omissions , transpositions, additions, and substitutions of words . The Sentences subtest is a memory test, measuring immediate recall and attention . Because suc­ cess may depend on verbal facility, failure may not neces­ sarily reflect poor memory ability. For children 5 years of age and older, scores may be related primarily to memory ability, but for children younger than 5 years of age , scores may reflect verbal knowledge and comprehension, rather than immediate recall ability per se.

Rationale.

Factor analytic findings.

The Sentences subtest is a moderately good measure of g (41 percent of its variance may be attributed to g). The subtest contributes s ubs tal l ­ tially to the Verbal factor (Mdn loading = .63).

Reliability and correlational highlights. Sentences i s a rel iable subtest (r.t:( = . 85 ) . It correlates more highly with

Comprehension (r = . 5 3 ) than with any other subtest. It correlates moderately with the Full Scale (r = .61) and the Verbal Scale (r = . 64) and to a lesser degree with the Performance Scale (r = .47 ) .

Administrative considerations.

Scoring the child's re­ sponses on the Sentences subtest is complicated because of the variety of errors that can occur. Responses must be analyzed carefully in order to determine the type of error that has been made. In addition to recording the number of errors , evaluate the quality of the child's responses . For example, were any idiosyncratic or peculiar words added? Did errors occur toward the beginning, middle, or end of sentences? Were sentences completely missed or were only a few errors made in each sentence? The child who misses a few words may be revealing minor temporary inefficiencies, whereas the one who cannot recall any

The Animal House subtest, a replacement for the WISC Coding subtest, requires the child to place a cylinder of the appropriate color in a hole on a board. The subtest's name is derived from the fact that four animals are depicted in various colored cylinders , or "house s . " For example, the dog has a black house and the chicken has a white house. Animal House is a l iberally timed subtest (maximum time of 5 minutes) in which a premium is placed on speed. A perfect performance in 9 seconds or less is credited with 70 raw-score points, whereas one obtained in 5 minutes is credited with 12 raw-score points. Animal House Retest is exactly the same test as Animal House. The word "retest" simply indicates that separate normative scaled scores are available for retest perfor­ mance . Although any subtest can be administered a second time , Animal House is the only one in the scale for which separate retest scores are provided. It is the only subtest in the battery specifically designed to evaluate learning ability.

Rationale.

Animal House requires the child to associate signs with symb ol s . M e mory, attention span , goal awareness, concentration, and finger and manual dexterity may all be involved in the child's performance (Herman , 196 8 ; Wechsler, 1967) . The subtest may also be a measure of learning ability (Wechsler, 1967) . Animal House has been found to correlate significantly with a measure of learning (r = .71) and a measure of motor skill (r = - . 69) in a sample of 36 children 5 to 6 years old (Sherman , Chinsky, & Maffeo, 1974). (The nega­ tive correlation results from lower motor skill scores re­ flect faster reaction times . ) The combination of learning and motor scores leads to a better prediction of Animal House scores than do the learning scores by themselves . The results suggest that motor abilities, in addition to learning abilities, may be involved in performance on the Animal House subtest . Yule et al . (1969) cautioned against accepting Wechsler's statement that performance on the Animal House Retest may di ffe rentiate between slow and fast l e a rners . Wechsler's statement is potentially misleading because it is difficult to assess and predict rate of learning and because there are few validation studies to support Wechsler's posi­ tion . An assessment of learning ability using the Animal House Retest must take into account the child's age and initial score. For example, children who are 4 1/2 years old

208

CHAPTER 9

WECHSLER PRESCHOOL AND PRIMARY SCALE OF INTELLIGENCE (WPPSI)

and who achieve a raw score of3 on initial testing maintain their initial status on retest by improving their perfor­ mance by 2 raw score points, whereas children of the same age who achieve a raw score of 18 on the initial subtest need an increase of about 10 raw score points to maintain their position. In addition to the child's age and initial score, motivational factors , the magnitude of the retest change, and the overall level of ability must be considered as factors that may affect learning ability. Yule and his col­ leagues concluded that empirical work is needed before scores on the Animal House Restest can be meaningfully interpreted.

Factor analytic findings.

The Animal House subtest is a moderately good measure of g (30 percent of its variance may be attributed to g) . The subtest has a moderate loading on the Performance factor (Mdn loading = . 49 ) .

Reliability and correlational highlights. Animal House is a relatively reliable subtest (rxx = . 77) . It correlates

more highly with Geometric Design (r = .43) than with any other subtest. It has a low correlation with the Full Scale (r = . 5 3 ) , the Performance Scale (r = . 50) , and the Verbal Scale (r = . 46).

Administrative considerations.

Note whether the child is right- or left-handed before administering this subtest. Children should be encouraged to use the hand they prefer. As on all timed subtests , do not stop timing once the subtest has begun. If the subtest is spoiled, do not include it in the final calculations. (This is true, of course, for all subtests in the scale . ) The WPPSI manual does not provide adequate guidance about what to do when a child stubbornly reinserts a peg in the wrong hole or refuses to take the examiner's hints during the demonstration items. If a child puts a yellow peg under the same dog for the second time , simply say "No, it should be a black one , " insert the peg yourself, and go on to the next sample item. The Animal House Retest is not used in the calculation of the IQ. Comparing the child's two performances, how­ ever, can provide some indication of the child's learning ability or ability to benefit from practice. In the WPPSI manual , Table 20 is used to obtain raw scores for the Animal House and Animal House Retest , but a separate part (last column) of Table 21, "Scaled Score Equivalents of Raw Scores," is used to obtain the Animal House Retest scaled scores. An inspection of the scaled scores for Animal House and Animal House Retest indicates that the child needs a

higher raw score on the Animal House Retest in order to obtain a scaled score equivalent to the one earned on the initial Animal House administration. For example , a raw score of 16 on Animal House is equivalent to a scaled score of 10 , but the same raw score on Animal House Retest is equivalent to a scaled score of 9, resulting in a loss of 1 scaled-score point .

Picture Completion The Picture Completion subtest consists of 23 drawings of common objects (for example, dol l , roses, and door) , each of which lacks a single important element . Twelve of the drawings appear on the WISe . The child's task is to dis­ cover and name or point to the essential missing portion of the incompletely drawn picture. Although there is no exact time l imit for each picture, the next picture is shown after 15 seconds if there is no response .

Rationale. The rationale described for the WISC-R Pic­ ture Completion subtest appears to hold for the WPPSI Picture Completion subtest (see Chapter 7) . Factor analytic findings. The Picture Completion sub­ test is a moderately good measure of g (36 percent of its variance may be attributed to g) . The subtest has a moder­ ate loading on the Performance factor (Mdn loading = . 50) and on the Verbal factor (Mdn loading = .40) . Reliability and correlational highlights. Picture Com­ pletion is a reliable subtest (rxx = . 8 3 ) . It correlates more highly with Information (r = . 47) than with any other subtest. It correl ates moderatel y with the Full Scale (r = . 60) and to a lesser degree with the Verbal Scale (r = . 54) and the Performance Scale (r = . 5 5 ) . Administrative considerations. The administrative considerations discussed for the WISC-R Picture Comple­ tion subtest generally apply to the WPPSI Picture Comple­ tion subtest, with the exception of time limits. Unlike the WISC-R Picture Completion subtest, which allows a max­ imum of 20 seconds per card, the WPPSI Picture Comple­ tion subtest has no absolute time l imit. Even though the WPPSI Picture Completion subtest is not timed, it may prove to be valuable for qualitative analysis to record the amount of time taken by children to make each response . In presenting the second card, it is advisable to repeat the directions given for the first card: "Look at this picture. Some important part is missing. Tel l me what is missing . " Children are given credit if they correctly point to the

209

WPPSI SUBTESTS

missing part . If a pointing response is accompan ied by a verbal response, however, the verbal respon�e is g i yen precedence over the pointing response . Therefore. an in­ correct verbal response (for example, saying " hair" to number I ) accompanied by a correct pointing response (pointing to the missing tooth) receives a score of O.

Mazes The Mazes subtest consists of 10 mazes, 7 of which are from the WISe . Three new horizontal mazes intended for younger children appear at the beginning of the subtest . In the WPPS I , Mazes is a standard subtest, whereas in the WISC-R it is a supplementary one . The subtest is discon­ tinued after two consecutive failures.

Rationale.

The rationale described for the WISC-R Mazes subtest appears to apply to the WPPSI Mazes sub­ test (see Chapter 7) .

Factor analytic findings. The Mazes subtest is a moder­ ately good measure of g (32 percent of its variance may be attributed to g) . The subtest contributes substantially to the Performance factor (Mdn loading = .63) . Reliability and correlational highlights.

Mazes is a reliable subtest (rxx = . 87 ) . It correlates more highly with Geometric Design (r = . 48) than with any other subtest . It has a low correlation with the Full Scale (r = . 54) , the Performance Scale (r = . 57 ) , and the Verbal Scale

(r

=

. 44) .

Administrative considerations.

Although the adminis­ trative considerations described for the WISC-R Mazes subtest apply to the WPPSI Mazes subtest, the adminis­ trative procedures differ. Timing, scoring, and other de­ tails differ. For example, Mazes l A , l B , 2, 4, 5 , and 6 are

Example

each allowed a maximum of 45 seconds; mazes 3 and 8, 60 seconds; and maze 10, 135 seconds. Therefore you must be sure to use the procedures appropriate for the scale being administered . Allow the child to finish each maze , re­ gardless of the errors made, because interruptions may generate anxiety and confusion and leave the child with a sense of fai lure . Scoring the Mazes subtest requires considerable judg­ ment . You must become familiar with special terms , such as "blind alley," "false exit," "alley wal l , " and "false start," which designate specific features of the mazes or of the child's performance . Likewise, you must be careful to point out these features of the test to the child in the sample items . Study the child's failures carefully. Note whether there is a pattern to the child's failure , or whether there are signs of tremor or other visual-motor difficulties. After the entire examination has been administered, you may want to return to the Mazes subtest to inquire into the child's performance on any mazes of interest (for example, "Why did you go that way?") . A careful evaluation of the failures that occur on the Mazes subtest may prove to be useful . Two examples are shown in Figure 9-3 . In example 1, the girl failed to complete the maze , but made no errors as far as her performance went . In example 2, another girl entered a blind alley, thereby making an error. In the first case one wonders why the girl stopped short before reaching the goal . Perhaps her perseverance is l imited , perhaps she takes things for granted and hopes that others will under­ stand her, or perhaps she was distracted. In contrast, the second performance may be that of an impulsive girl who works well until she is about to complete the task and then is unable to do so correctly. These analyses are, of course, only tentative, subject to modification after study of the child's performance on the entire subtest and other subtests on the scale, as well as other sources of data.

1

Example

Figure 9-3.

Two examples of failures on the Mazes subtest.

2

WPPSI Mazes reprinted by permission of the publ isher. Copyright 1949;

© 1 963, 1 967, The Psychological Corporation, San Antonio, TX. All Rights Reserved.

210

CHAPTER 9

WECHSLER PRESCHOOL AND PRIMARY SCALE O F INTELLIG E N CE (WPPSI)

Geometric Design The Geometric Design subtest contains 10 designs - in­ cluding a circle, a square, and a diamond - that the child is asked to copy. There is no time limit . Scores for items I to 5 range from 0 to 2 ; for items 6 and 7 , from 0 to 3 ; and for items 8 to 10, from 0 to 4 .

Rationale. The Geometric Design subtest i s considered to measure perceptual and visual-motor organization abili­ ties . Low scores may indicate lags in the developmental process. Even bright young children may have difficulty obtaining high scores, because the motor ability needed for successful performance (the ability to grasp a pencil appropriately, make contact with paper, and draw appro­ priate lines) is associated in part with maturational pro­ cesses that may be independent of the development of cognitive processes . Factor analytic findings. The Geometric Design sub­ test is a moderately good measure of g (41 percent of its variance may be attributed to g). The subtest contributes substantial l y to the Performance factor ( Mdn load­ ing = . 67 ) . Reliability and correlational highlights.

Geometric Design is a reliable subtest (rxx = . 82 ) . It correlates more highly with Mazes and Block Design (r's = . 48) than with any other subtests. It has a low correlation with the Full Scale (r = . 5 8 ) , a moderate correlation with the Perfor­ mance Scale (r = . 60) , and a low correlation with the Verbal Scale (r = . 4 8).

Administrative considerations.

The Geometric Design subtest is difficult to score . At least eight different general criteria must be used to score the designs; each design also has special scoring criteria. There is seldom unanimous agreement among examiners on the scores given to the drawings of normal children. Sattler (1976) reported that unanimous agreement occurred among a sample of 18 school psychologists on only 7 out of 50 drawings. Lower­ ing the agreement criterion to 78 percent of the group (14 out of 18) still resulted in agreement on only 23 of the 50 drawings. Similar results were found for a group of 14 inexperienced graduate students. Morsbach, McGoldrick, and Younger (1978) also found that interscorer reliability was inadequate on the Geometric Design subtest, particu­ larly on designs 3, 4, 6, 8, 9, and 10. You will need to study

the scoring criteria carefully in order to become proficient in scoring Geometric Design . The special copyrighted blank paper obtained from the test publisher for the administration of the Geometric Design subtest is not necessary (Yule et aI . , 1969) . You need only fold a sheet of paper in half, writing on each drawing the number of the design and "top" and "bottom" relative to the child's frame of reference. Only half of the paper should be shown to the child at a time to avoid distraction from the design drawn on the other half of the paper.

Block Design The Block Design subtest contains 10 items. The child is shown a model constructed by the examiner for the fi rst seven items and designs for the last three. Flat blocks are used to reproduce the designs. Children under 6 years of age (and older children who may be mentally retarded) start with item I ; children 6 years of age and older start with item 3 . All of the items are timed . The first four items are given a maximum of 30 seconds; the next two, 45 seconds; the next two, 60 seconds; and the last two, 75 seconds . There are no time-bonus credits as there are on the WISC-R. A score of 2 is given for a successful performance on the first trial , a score of 1 for a successful performance on the second trial , and a score of 0 when both trials are fail ed .

Rationale. The rationale described for the WISC-R Block Design subtest appears to apply to the WPPSI Block Design subtest (see Chapter 7 ) . Factor analytic findings. The Block Design subtest i s a moderately good measure of g (38 percent of its variance may be attributed to g) . The subtest contributes substan­ tially to the Performance factor (Mdn loading = . 5 8 ) . Reliability and correlational highlights. Block Design is a reliable subtest (rxx = . 82 ) . It correlates more highly with Arithmetic (r = . 50) than with any other subtest. It correlates moderately with the Full Scale (r = . 61) and to a lesser degree with the Performance Scale (r = . 59) and the Verbal Scale (r = . 52) . Administrative considerations .

The administrative considerations described for the WISC-R Block Design subtest general ly apply to the WPPSI Block Design subtest.

INTERPRETING THE WPPSI

I NT E R PRETI N G THE W PPSI

Much of the material in Chapter 8 pertains to the WPPS l . The successive-level approach t o test interpretation , pro­ fi le analysis, Verbal-Performance Scale comparisons, and subtest comparisons are essentially the same for both t e WISC-R and the WPPSI . The information in Table C-24 in Appendix C can aid you in interpreting the WPPSI subtests, as well as in writing report s . It summarizes the abilities thought to be measured by each WPPSI subtest, background factors that may influence subtest performance, and imphcations of high and low scores. It deserves careful study. The classifications associated with WPPSI IQs are shown in Table BC-2 on the inside back cover. Table BC-I on the inside back cover shows the percentile ranks for the WPPSI Full Scal e , Verbal Scale, and Performance Scale IQs . Percentile ranks associated with subtest scaled scores are shown in Table C-41 in Appendix C . The WPPSI Structure-of-Intellect classifications can be obtained from Table C-23 in Appendix C . The individual subtests should not be viewed as a means of determining specific cognitive skills with precision . Rather, subtest scores should be used as a means of gener­ ating hypotheses about the child's abilities . The most reli­ able estimates of specific abilities are derived from the Verbal Scale IQ (verbal comprehension) and the Perfor­ mance Scale IQ (perceptual organization) , not from indi­ vidual subtest score s . Because there i s a great deal o f overlap between the WPPSI and the WISC-R, especially for the eight subtest types that they share , much of the information in this text on the WISC-R is pertinent to the WPPS I . You are encour· aged to review Chapter 7 , which discusses the WISC-R subtests, before reading the rest of this chapter.

21 1

Appendix C must be used . See Chapter 8 for an explana­ tion of each of the approache s . (Factor scores are not included in profile analysis on the WPPSI because the Verbal and Performance Scales adequately describe the organization of the scale . ) 1 . Comparing Verbal and Performance Scale IQs. Table C-15 in Appendix C provides the critical values for com­ paring the Verbal and Performance IQs: II at the .05 level and 14 at the .01 level. (Probabilities associated with vari­ ous Verbal Performance Scale differences are shown in Table C-I9 in Appendix C . )

2 . Comparing each Verbal subtest scaled score to the mean Verbal scaled score. Table C-I6 in Appendix C provides the critical values: they range from 2 . 83 to 3 . 05 at the .05 level and from 3 . 39 to 3 . 66 at the .01 level for the five standard Verbal subtests. 3 . Comparing each Performance subtest scaled score to the mean Performance scaled score. Table C-16 in Appen­ dix C provides the critical values : they range from 2 . 61 to 3 . 26 at the . 05 level and from 3 . 13 to 3 . 91 at the . 01 level for the five Performance subtests. 4. Comparing each subtest scaled score to the mean subtest scaled score. Table C-16 in Appendix C provides the critical values : they range from 2 . 94 to 3 . 84 at the .05 level and from 3 . 78 to 4 . 50 at the . 01 level for the \0 standard subtests . 5 . Comparing sets of individual subtest scores. Table C-15 in Appendix C provides the critical values: they range from 3 to 4 at the . 05 level and from 4 to 5 at the .01 leve l . The values in Table C-15 are overly liberal (that i s , lead to too many significant differences) when more than one comparison is made. They are most accurate when a priori planned comparisons are made, such as Comprehension versus Information or Block Design versus Animal House . (See Chapter 7 for additional information that can guide subtest interpretations . )

Profile Analysis Because profile analysis on the WPPSI is similar to that on the WISC - R , the mate rial in Chapter 8 describing WISC-R profile analysis should be reviewed before a pro­ file analysis is undertaken. Although much less is known about profi l e analysis on the WPPSI than on other Wechsler scales, the procedure can still be useful in gener­ ating hypotheses about a child's strengths and weaknesses. The five approaches to profile analysis on the WPPSI described here are the same as those described for the WISC-R. The only difference is that different tables in

Silverstein advises determining the difference between the highest and the lowest subtest scores before making multiple subtest comparisons ( personal communication, 1980 ) . If this difference is 6 scaled-score points or more, a significant difference at the .05 level is indicated . Differ­ ences between subtests that are 6 scaled-score points or greater can then be interpreted . If the difference between the highest and lowest subtest scaled scores is less than 6 scaled-score points, multiple comparisons between indi­ vidual subtest scores should not be made.

212

CHAPTER

9

WECHSLER PRESCHOOL AND PRIMARY SCALE O F INTELLIGENCE (WPPSI)

Other Approaches to Profile Analysis The supplementary approaches to WPPSI profile analysis described in this section are similar to those discussed for the WISC-R in Chapter 8 .

Base rate subtest scaled-score ranges. In the standard­ ization group, the median range between the scaled scores of the highest and lowest subtests was 6 points for the 10 standard subtests, 4 points for the five standard Verbal Scale subtests, and 5 points for the five standard Perfor­ mance Scale subtests (Reynolds & Gutkin, 1981) . Base rate differences between each subtest score and an average subtest score in the WPPSI standardization sample. A child's score on each subtest may be compared with his or her average WPPSI Verbal , Performance, or overall score through use of Table C-18 in Appendix C , which gives the frequencies with which various differences occurred in the standardization sample . The table shows, for example, that a difference as large as 3 .4 points be­ tween the score on Information and the Verbal Scale average was obtained by 5 percent of the standardization sample. This table should be used only for differences that have first been found to be reliable. (See numbers 2 , 3, and 4 in the profile analysis section above . ) A difference of approximately 4 points between each subtest score and the respective average Verbal or Performance Scale score was obtained by 5 percent of the standardization sample.

Base rate Verbal-Performance differences (or proba­ bility-of-occurrence approach) . Table C-20 in Appen­ dix C presents the percentage of individuals in the stan­ dardization group who obtained a given d iscrepancy between the Verbal and Performance Scales . The table shows, for example, that between 25 and 50 percent of the population in each WPPSI age group had a 10-point differ­ ence (in either direction) between the two IQs.

Clinical and Educational Uses of the WPPSI The administration of the WPPSI, like the administration of any other psychological or educational test, requires careful observation of the child's performance . For exam­ ple, behaviors suggestive of emotional disturbance, lan­ guage difficulties, perceptual problems, or visual-motor difficulties should be noted carefully and reported. (See Chapter 18 for further assessment considerations.) There is little published research on the clinically meaningful use of profile variability on the WPPSI; research in this area is needed . The statement by Wechsler (1967) that very poor

performance on the Animal House subtest will at times be associated with organic deficit should be regarded as an unconfirmed hypothesis until research studies become available .

Learning disabilities and the WPPSI . An interesting contribution by Hagin, Silver, and Corwin (1971) illus­ trates how the WPPSI can be used in the assessment of cognitive functioning of children with learning disabilities . These investigators reported on WPPSI profiles for three different subgroups of learning-disabled children: those with specific language disabil ity, brain damage, and devel­ opmental immaturity. Specific language disability. Children in the specific language disability subgroup have problems in developing body-image concepts, in establishing cerebral dominance for language, and in orienting figures in space and sounds in time (see Chapter 20) . Richard's performance (see Table 9-9) i s representative of that of this subgroup. Although a neurological examina­ tion did not indicate brain damage, Richard showed some minor soft signs (equivocal signs of brain damage) such as gross errors in right-left orientation and mild difficulties i n movement. H is visual-motor skills were good, but h e had some difficulties w ith auditory discrimination and se­ quencing . On the WPPSI, Richard earned an average Full Scale IQ (98) , but showed a 23-point spread between his Verbal and Performance Scale IQs (87 and 1 10, respec­ tively) . Within the Verbal Scale, his major difficulties were in areas requiring quantitative reasoning, logical thinking, and social judgment. His range of knowledge and his linguistic skill were adequate , although he sometimes ex­ pressed ideas awkwardly. H is above-average Performance Scale IQ suggests that he has good potential for learning . It was recommended that educational intervention stress the auditory modality. Brain damage. Brain damaged children demonstrate many of the behaviors of the specific language disability group, but in addition show abnormality on the standard neurological examination. Some children are hyperactive, whereas others are hypoactive. Generally the findings do not point to focal brain damage, and specific etiological factors are rarely found in the child's history (see Chapter 22). The children in this subgroup present special educa­ tional problems because of poor impulse control, limited attention span , inadequate motor coord ination , and anxiety. Karl (see Table 9-9) is an example of a child in the brain-

213

INTERPRETING THE WPPSI

Table 9-9 Illustrations of W P PSI Deviations from Mean Scaled Scores for Children with Learning Disabilities i n First Grade

Richard: Specific Language Disability Subtest or scale

Deviation

+ 2.2 - .8 + 3.2* .8 -2.8

7 6 7 7 7 6

+ .3 .7 + .3 + .3 + .3 .7

- 1 .8 + .2 +4.2* + 2 .2 + .2

11 7 6 7 6

+ 3 .6* - .4 - 1 .4 .4 - 1 .4

Deviation'

Score

Deviation

I nformation Vocabulary Arithmetic Similarities Comprehension Sentences

10 11 6 6 7 8

+2 +3*

- .8

-2 -1 0

11 14 11 15 11 9

Arumal House Picture Completion Mazes Geometric Design Block Design

10 13 12 11 10

- 1 .2 +1.8 + .8 - .2 - 1 .2

9 11 15 13 11

M Verbal scaled score M Performance scaled score

8 1 1 .2

-2

87 1 10 98

1 1 .8 10.8 1 06 1 12 1 10

Rosemary: Developmental Immaturity Score

Score

Verbal I Q Performance IQ Full Scale IQ a

Karl: Brain Injured

-

-

6.7 7.4 80 82 79

The deviations are from the mean of the respective scales.

* These are significant deviations (see Table C- 1 6 in Appendix C).

Source: Adapted from Hagin, Silver, and Corwin ( 1 97 1 ) .

damaged subgroup. Because there is no typical brain­ damaged child , he was selected simply to ill ustrate how one brain-damaged child performed on the WPPSI . The neurological examination disclosed poor fine and gross motor coordination and severe difficulties in performing skilled movements . Karl exhibited confusion in right-left discrimination, restless motion, tremors, and hyperac­ tivity. His verbal communications were , at time s , in­ coherent and circumstantial . On the WPPSI, he obtained a Full Scale IQ of no . Conceptual thinking, memory, atten­ tion , and concentration were areas in which his perfor­ mance was below its average level . His motor problems were especially evident in his difficulty in grasping the pegs in the Animal House subtest and in his four-finger, non-oppositional grip on the pencil in the Geometric De­ sign subtest . His best performances were on the Mazes and Vocabulary subtests, subtests that reflect planning ability and word knowledge, respectively. His performance im­ proved when he became familiar with the task require-

ments . It was recommended that educational efforts em­ phasize visual-motor and organizational skills .

Developmental immaturity. Developmentally imma­ ture children exhibit slowness in reaching developmental landmarks, but no clinical or historical evidence of central nervous system damage . In physical appearance , gross and fine motor development, language , and social awareness, these children seem to be younger than their chronological ages. Low birth weight appears frequently in their histories. Rosemary (see Table 9-9) can be considered to be repre­ sentative of this subgroup . Neurological difficulties were absent. On the WPPSI , she performed in the Borderline range . The profile of deviations of her subtest scaled scores from the mean of her scaled scores is essentially flat; the only significant variation occurred on the Animal House subtest. Recommendations centered on general enrich­ ment, with particular emphasis on language stimulation .

214

CHAPTER 9

WECHSLER PRESCHOOL AND PRIMARY SCALE OF INTELLIGENCE (WPPSI)

Reading Disability and the WPPSI Attempts have been made to determine whether there are differences in patterns of intellectual abilities, as measured by the WPPSI, among preschool and school-age children who differ in their reading skills. The results suggest that the WPPSI profiles for average and poor readers are essen­ tially similar (Barron, 1971 ; Kavajec z , 1969 ) . This state­ ment holds for both Anglo-American and Mexican-Amer­ ican children. It is difficult to generalize from such re­ search findings , however, because few children in the WPPSI age range are able to read.

ASSETS OF T H E WPPSI

The WPPSI has many assets. 1 . It has excellent psychometric properties. The WPPSI has excellent reliability and validity and has been carefully standardized . The features that are common to other Wechsler tests - such as separate Verbal and Performance Scales , Deviation IQs, and a convenient manual - are also part of the WPPSI. 2. It provides useful diagnostic info rmation. The WPPSI provides diagnostic information useful for the as­ sessment of cognitive abilities of preschool children, both normal and mildly mentally retarded . It also furnishes data that are helpful in planning special school programs, per­ haps tapping important developmental or maturational fac­ tors needed for school success in the lower grades . 3 . The test has a high interest level. Children enjoy taking the test; the mixture of verbal and performance items maintains their interest .

L I M ITAT I O N S OF T H E WPPSI

The WPPSI has a number of l i mitations . 1 . It takes a long time to administer. Administration time may be too long for some children, although fatigue is not often a problem. In the standardization sample , ap­ proximately 10 percent of the children needed 90 minutes or more to complete the test. The scale often takes over 60 minutes to administer. With younger children or with handicapped children, two test sessions may be needed. When this procedure is followed , there is no way of determining whether the break between testing sessions affected a child's scores, since the procedure differs from that used in standardizing the scale. Empirical data would be helpful in clarifying the effect of two test sessions on test scores.

2 . It has a limitedjloor and ceiling. The WPPSI , like the WISC-R, is limited by the absence of an adequate fl oor and ceil ing (that is, it does not clearly differentiate abilities at the lower and upper ends of the scale) . IQ equivalents of the scaled scores range from 45 to 15 5 . This range, how­ ever, is not applicable until the 5 if2 -year-old leve l . At the 4-year-old leve l , the lowest Full Scale IQ shown i n the manual is 5 1 , the lowest Verbal Scale IQ is 5 5 , and the lowest Performance Scale IQ is 55 . A child recei ves up to 4 scaled-score points for having given no correct answers . Wechsler (1967) recognized this problem and recom­ mended that IQs for each scale be computed only when the child obtained a raw score greater than 0 on at least two of the subtests on each of the scales . Similarly, he recom­ mended that a Full Scale IQ not be computed unless raw scores greater than 0 were obtained on at least two Verbal and two Performance subtests . Following Wechsler's recommended procedure, let us calculate IQs for each of the three scales for a 4-year-old child who obtained raw scores of 1 on the Information , Vocabulary, Animal House, and Picture Completion sub­ tests and a raw score of 0 on each of the remaining six subtests. The resulting IQs are as follows : Verbal Scale IQ = 59 (17 scaled-score points) , Performance Scale IQ = 60 (21 scaled-score points) and Full Scale IQ = 55 (38 scaled-score poi nts) . Four I-point successes thus yielded an IQ of 55 . This example demonstrates that the WPPSI may not provide precise IQs for children who are functioning two or more standard deviations below the mean of the scale. Further research is needed to determine the validity of the WPPSI for moderately mentally retarded children . In one study (Rellas, 1969) of a sample of 26 gifted children , the percentages of children who obtained the maximum possible scores on the Arithmetic, Mazes, and Block Design subtests ranged from 11 to 19 percent. In a related study (Hawthorne, Speer, & Bucce llato , 1983) of a sample of 306 gifted children , the percentages of children who failed to obtain a ceiling (that is, did not reach the criterion for discontinuance of a particular subtest) were as follows: Arithmetic - 8 percent, Geometric Design - 17 percent , Block Design - 21 percent , Information - 50 percent , Mazes - 61 percent, Picture Completion - 68 per­ cent, Vocabulary - 69 percent, Comprehension - 95 per­ cent , and Similaritie s - 96 percent . Consequently, the WPPSI appears to be of limited usefulness in assessing the upper limits of the ability of gifted children . On the basis of the available evidence, the WPPSI does not appear to be appropriate for the assessment of severely retarded or highly superior children . If a child fails (or passes) all or most of the items on the WPPSI, it is

215

SlUtlMARY

ald, isable to switch to a test that can provide a more alc rate estimate of the child's abil ity. 3. Scoring of responses is difficliit. Like the WISC-R, tthe WPPSI is not always easy to score . Scoring is es­ plecially difficult on the Geometric Design subtest, in w{hlch scores may rely on examiners' subjective decisions, amd on the Vocabulary, Similarities, and Comprehension slUb:ests . Consultation with colleagues is recommended wlhen responses are difficult to score .

4. It may pose special difficulties for disadvanTaged

cJhildren.

Di sadvantages of the WPPSI, at least with lower­ cl a childre n , include the ambiguity and possible emo­ tilonal loadings of several Comprehension subtest ques­ tilons and the need to ask for additional reasons on several qUle5tions, which may make some children uncomfortable (Fagan et aI . , 1969) . 5 . No information is provided about cutoff criteria. The \-"'PPSI manual fails to provide information about whether thle cutoff criteria were determined empirically or intu­ itiIvely.

P'SYC H O LO G I CAL EVALUATION

Exhibit 9-3 illustrates the application of the WPPSI to evaluation of a developmentally immature child. The re­ port summarizes information obtained from the parents and from a k.indergarten teacher and cites both qualitative a d quantitative information obtained during the evalua­ tion. Profile analysis is used to develop some assessment i n formation, and recommendations are based on the test results and background information.

T EST Y O U R SKILL

The WISC-R Test-Your-Skill Exercises in Chapter 8 also pertain to the WPPSI. If you have not reviewed these exercises recently, you are encouraged to do so now. In addition, three exercises that pertain only to the WPPSI fol low. In each exercise, there is some inadequacy of description or interpretation . Analyze the mistakes, then check your answers with those shown in Appendix A. I . 'Tom's excellent performance on Block Design and

Geometric Design suggests that he has good ability in analyzing social situations and has high moral judgment . " 2 . "The Geometric Design subtest presented problems for her and she fell in the slow learner category. " 3 . The following interpretation was given to WPPSI Performance Scale scores on Animal House ( 12 ) , Picture

Completion ( 1 5 ) , Mazes (13), Geometric Design (14) , and Block Design (12 ) . "While her two lowest scores on the Performance Scale were above average, they may suggest some visual acuity problem s . "

S U M MARY

I . The WPPSI, designed to be used with children between 4 and 6 '12 years of age , follows the basic fermat of the WISC-R , providing Verbal, Performance, and Full Scale IQs. 2 . Three WPPSI subtests - Sentences, Animal House, and Geometric Design - do not appear on other Wechsler tests. 3. The standardization sample was representative of the U . S . population, with average reliabilities ranging from . 7 7 t o . 87 for the individual subtests and from . 93 to . 96 for the three scales. 4 . Like the other Wechsler Scales, the WPPSI employs De­ viation IQs for the Verbal , Performance, and Full Scale IQs. Similarly, scaled scores are provided for each subtest. IQs can be prorated when fewer than lO subtests are administered. 5. The WPPSI has excellent reliability for the three IQs but less satisfactory reliability for the subtest scaled scores. The WPPSI Performance Scale generally shows greater practice ef­ fects than does the W PPSI Verbal Scale. 6 . Validity studies comparing the W PPSI and the Stanford­ Binet: Form L-M and Fourth Edition, indicate that the two scales correlate highly ; Mdn r = . 8 1 for Form L-M and Mdn r = . 80 for the Fourth Edition. Scores from the two scales are not inter­ changeable, however. 7. The WPPSI appears to correlate highly with the WISC-R, but IQs may not be interchangeable. Two II-year longitudinal studies in Great Britain showed extremely high correlations be­ tween the WPPSI and WISC-R, with IQs differing by less than 6 points. 8. WPPSI IQs may not be interchangeable with McCarthy Scale scores. 9. The WPPSI has excellent to adequate predictive validity for both white and black childre n , using achievement scores as the criterion. More work is needed to establish predictive validity for Mexican-American children, however. 1 0 . The relationship between WPPSI IQs and demographic characteristics indicates that there is an 18-point difference be­ tween the highest and lowest socioeconomic status groups (llO vs. 92) . Urban-rural differences are not significant. Children from the West obtained significantly higher IQs than did those from other regions of the country. I I . Sex differences are minimal on the WPPSI subtests; when they occur, they are more pronounced on Performance than on Verbal subtests. 1 2 . The IQs obtained by white children in the standardization sample were about II points higher on average than those of a matched group of black children (98 vs. 87). 1 3 . Factor analytic studies support the division of the scale into Verbal and Performance sections. All II WPPSI subtests have

Exhibit 9-3 Psychological Evaluation: A C h i ld with Developmental Immaturity Evaluated by the WPPSI Date of examination: June 1 2 , 1 986 Date of report: June 1 5 , 1 986 Grade: K indergarten

Name: Debbie Date of birth : November 2 5 , 1 980 Chronological age: 5-6

Test Administered

Wechsler Preschool and Primary Scale of Intelligence (WPPSI) : VERBAL SCALE

Information Vocabulary Arithmetic Similarities Comprehension

PERFORMANCE SCALE

8 13 10 7 9

Ani mal House Picture Completion M azes Geometric Design Block Design

5 6 8 7 9

Verbal IQ = 96 Performance IQ = 80 Full Scale IQ 87 ± 4 at the 85 percent confidence level =

Reason for Referral

Debbie's parents requested the evaluation because they were concerned about her rate of development . Her parents de­ scribed her as being a slow learner and as having a short attention span. Debbie's developmental landmarks were all reached slightly later than is average. The results of a neuro­ logical examination were essentially negative. Her parents consider her to be a fairly well adjusted child who is generally happy at home and with other childre n . There is one other sibling i n the family, a boy, who is said by the parents to be gifted. Debbie's kindergarten teacher described her as being a willing worker when supervised by an adult. When she is on her own , however, her attention often wanders aimlessly. In class, her retention appears to be l imited ; she is distracted by anything that crosses her vision. She tends to perceive situa­ tions as parts, not wholes, and because she fixes her attention on small details she fails to understand many situations. Debbie also has speech problems. In class she speaks slowly and uses phrases that are more characteristic of a 3-year-old than a 5-year-old . General Observations

Debbie is an attractive youngster, of average height and weight for her age . Although an articulation problem was evident, her speech was understandable . She exhibited some awkwardness in motor coordination. Her walking gait was uneven, and she had some difficulty in turning the pages of a test booklet . At t i mes she was restless during the testing. She was cooperative , however, and attempted to answer the ques­ tions and do the tasks asked of her.

Debbie's c urrent level of intellectual functioning is in the Dull Normal range and i s ranked at the 20th perce nt i l e . The chances that the range of scores from 8 3 to 9 1 includes her true IQ are about 85 out of 100. The good rapport that existed between Debbie and the examine r and the child's ability to follow d i rections and to attempt to respond to the items suggest that the present results are val i d . Debbie's performance skil l s are n o t a s wel l developed as are her verbal skills. The l6-point d i fference between her scores on the verbal and performance parts of the scale suggests that visual-motor ability, perceptual ability, ability to attend to perceptual details, and persistence are at a level of develop­ ment that is below norma l . In contrast, not only do her verbal skills show more variabi lity than do her performance skill s , b u t also the overall level o f verbal development is within the normal range. Her out standing strength was her word knowl­ edge. She was able to define words at a level that was higher than the norm for her age peers and above the average of her verbal scaled scores. For example, she gave satisfactory defi­ nitions to such common vocabulary words as "fur," "join," and "diamond . " Her arithmetic skills appear to be at an average level. Debbie's answers were usually short, precise, and d irect. Her failures were manifested both by i ncorrect answers and by her saying "No" when she d id not know an answer. She seemed to experience more d ifficulty on the Similarities sub­ test , which measures logical thinking , than on most other verbal subtest s . Instead of giving analogies, she would repeat part of the question in her answer or give associations. For example, to the question "You ride i n a train and you also ride in a , she said "Choo-choo." This one verbal subtest , more than any of the other verbal subtests, reflected her difficulty in grasping concepts and suggested some i m ma­ turity i n reasoning. She at fi rst refused to complete the A nimal House subtest, but w ith encouragement and support finally proceeded with the task. This behavior, too, may suggest immaturity. __

"

Recommendations

The results suggest that Debbie's principal cognitive handicap is a gap between visual-motor skills and verbal skills, i n favor of the latter. In school situations she will not l i kely be per­ ceived as being extremely slow because of her average verbal skill s . She will need encouragement and attention, however, because she may tend to remove herself from difficult situa­ tions by i nattention or by simply refusing to try. Her parents should be helped to accept her present level of development

Test Results

The WPPSI results were as follows: Verbal Scale IQ of 96, Performance Scale IQ of 80. and Full Scale IQ of 87 ± 4 .

(Exhibit continues next page) 216

217

SUMMARY

Exhibit 9-3 (cont.) and not place unrealistic demands on her. Special programs to improve her muscle coordination and speech are recom­ mended. Summary

In summary, on the WPPS I , Debbie, with a chronological age of5-6, obtained a Full Scale IQ of87 ± 4, which is in the Dull Normal classification and at the 20th percentile rank. The chances that the range of scores from 83 to 91 includes her true IQ are about 85 out of 100. The results appear to give a valid estimate of her present level of intellectual functioning. Case h istory material suggested a pattern of developmental

strong g loadings. A Freedom from Distractibility factor does not emerge on the WPPS I . The factor structure is similar for both black and white children. 1 4 . The administrative considerations that apply to the W I SC-R generally apply to the WPPSI also. Because the WPPSI is used with a younger age group, there are some problems in adapting the subtests to alternative sensory modalities. 1 5 . Table C-21 in Appendix C shows the best combination of two, three, four, and five WPPSI subtests, and Table C-3 7 in Appendix C shows IQs for the sum of the scaled scores for the Vocabulary plus Block Design short form. 16. The i nterpretive rationale, factor analytic findings. relia­ bil ity and subtest correlations, and administ rative considerat ions for each of the 11 WPPSI subtests are presented in the chapter. The proposed interpretive rationales and possible impl ications of high and low scores are summarized in Table C-24 in Appendi)( C . 1 7 . The rationale for the WISC-R Information subtest pmb­ ably applies to the WPPSI Information subtest, although WPPSI questions may be related more to the child's experiences than to formal education . The subtest is the best measure of g in the scale and contributes to the Verbal factor. It is a rel iable subtest (r.o: = . 81 ) . Judgment is required in scoring responses . 1 8 . The rationale for the WISC-R Vocabulary subtest prob­ ably applies to the WPPSI Vocabulary subtest, although formal education probably has less influence on performance on the WPPSI. The subtest is a moderately good measure of g and contributes to the Verbal factor. It is a reliable subtest (r", = . 84) . Scoring requires considerable judgment. 1 9 . The rationale for the WISC-R Arithmetic subtest prob­ ably applies to the WPPSI Arithmetic subtest, although formal education probably has less influence on performance on the WPPSI . The subtest is a moderately good measure of g and contributes to the Verbal and Performance factors. It is a rel iable subtest (r(.< . 82 ) . Scoring is easy. 20. The WPPSI Similarities subtest appears to measure log­ ical thinking to a greater extent than does the WISC-R Sim­ ilarities subtest. The subtest is a moderately good measure of g and contributes to the Verbal factor. It is a reliable subtest (ru = . 83 ) . Judgment is required in scoring the last six items. =

immaturity. The examination revealed that she has better verbal skills than performance skills. Visual-motor coordina­ tion and other perceptual skil l s are less well developed than is her vocabulary ability. Immaturity was suggested by some of her responses and behavior patterns. She will need support and encouragement, and her parents should be helped to accept her at her present level of functioning. Special pro­ grams were recommended to i mprove her muscle coordina­ tion and speech. (Examiner's Signaulre) Examiner

2 1 . The rationale for t.he WISC-R Comprehension subtest probably applies to the WPPSI Comprehension subtest, although linguistic skill and logical reasoning may play a more significant role on the WPPSI. The subtest is a moderately good measure of g and contributes to the Verbal factor. It is a reliable subtest (r.« . 81). Scoring requ i res considerable judgment. 22. Sentences is the only supplementary subtest in the WPPSI. It is a memory test , measuring immediate recall and attention . The subtest is a moderately good measure of g and contributes to the Verbal factor. It is a rel iable subtest (ru = . 85 ) . Scoring requires considerable s k i l l . 2 3 . Animal House i s considered t o measure memory, atten­ tion span, goal awareness, concentration, and finger and manual dexterity. It is a moderately good measure of g and contributes to the Perfo rmance factor. It is a somewhat rel iable subtest (r(.< . 77 ) . Ad ministration is relatively easy. The abil ities mea­ sured by the Animal House Retest are not known at this time. 24. The rationale for the WISC-R Picture Completion subtest probably applies to the WPPSI Picture Completion subtest. The subtest is a moderately good measure of g and contributes to the Performance factor. It is a reliable subtest ( r(.{ . 83). Adminis­ tration is relatively easy. 25. The rationale for the WISC-R Mazes subtest probably applies to the WPPSI Mazes subtest . The subtest is a moderately good measure of g and contributes to the Performance factor. It i s a reliable subtest ( rxx . 87 ) . Scoring requires considerable judgment. Administrative procedures differ from those used on the WISC-R . 26. Geometric Design is considered to measure perceptual and visual-motor organization abil ities. It is a moderately good measure of g and contributes to the Performance factor. It is a . 82 ) . This subtest may be the most d ifficult reliable subtest (r of all WPPSI subtests to score. 27. The rationale for the WISC-R Block Design subtest prob­ ably applies to the WPPSI Block Design subtest . The subtest is a moderately good measure of g and contributes to the Perfor­ mance factor. It is a rel iable subtest (r", . 82). The subtest requires skill to administer. 28. Although the same considerations that apply to profile =

=

=

=

xx

=

=

218

CHAPTER

9

WECHSLER PRESCHOOL AND PRIMARY SCALE O F INTELLIGENCE (WPPSI)

analysis on the WISC-R apply to profile analysis on the WPPSI, more care should be taken i n using profile analysis with the WPPSI because fewer research fi ndings are available. 29. Some WPPSI profiles may be related to specific types of learning d isabil ities . 30. For the most part, WPPSI profiles of average and poor readers are similar. 3 1 . Although the WPPSI has some l i m itations - such as long administration time, inadequate floors and ceilings that l imit use w ith severely retarded and highly gifted children, difficult scor­ ing on some subtests, and minor admjnistration problems - it is, overal l , a well-standardized, carefully developed instrument that is a valuable tool for the assessment of children's intelligence.

WPPSI Comprehension ( p . 206) WPPSI Sentences ( p . 207) WPPSI Animal House ( p . 207) WPPSI Animal House Retest ( p . 207) WPPSI Picture Completion ( p . 208) WPPSI M azes (p. 209) WPPSI Geometric Design (p. 2 10) WPPSI Block Design ( p . 2 10) Profile analysis (p. 2 1 1 ) Specific language disability ( p . 212) Brain-damaged ( p . 2 12 ) Developmental immaturity ( p . 2 1 3 )

KEY T E RMS, CONCEPTS, A N D NAMES

STU DY QU EST I O N S

WPPSI standardization (p. 192) WPPSI reliability (p. 194) WPPSI validity (p. 195) Demographic correlates of WPPSI (p. 199) Gender differences on the WPPSI (p. 199) Ethnic differences on the WPPSI (p. 199) WPPSI factor analysis (p. 2(0) WPPSI subtest specificity (p. 2(0) Testing-of-limits (p. 202) Extrapolated IQs (p. 204) WPPSI short forms (p. 204) Yudin's WPPSI short form (p. 204) WPPSI Information (p. 204) WPPSI Vocabulary (p. 205) WPPSI A rithmetic ( p . 205) WPPSI Similarities ( p . 206)

I . Describe t h e W PPSI and then discuss i t s standardization , reliability, and validity. 2 . Describe WPPSI factor analytic fi ndings. 3. Discuss WPPSI IQs with respect to the stratification vari­ ables used in the standardization sample . 4. Discuss some general administrative considerations for the WPPSI . 5 . Discuss WPPSI short forms. 6 . Discuss the rationale, factor analytic findings, reliability and correlational h i gh l ights , and administrative considerations for each of the fol lowing WPPSI subtests : Information, Vocabu­ lary, Arithmetic, Similarities, Comprehension, Sentences, Ani­ mal House, Picture Completion, Mazes , Geometric Design, and Block Design. 7 . Briefly describe profile analysis on the WPPSI. 8. Discuss the assets and limitations of the WPPSI .

-

10 WECHSLER ADU LT I NTELLI G E N C E SCALE- REVISED (WAIS-R) Co-authored by Joseph J. Ryan

Standardization

When I was a boy of14, my faTher was so ignorant I could hardly sTand TO have The old man arollnd. But when I got TO be 21, I was aSTOnished aT how much he had learned in seven years.

Deviation IQs and Scaled Scores Reliability

- M ark Twain

Val idity Intercorrelations Between Subtests and Scales

Youth thinks intelligence a good substitute for experience, and his elders think experience a substitute for intelligence.

WAIS-R IQs and Stratification Variables Factor Analysis

--Lyman Bryson

Administering the WAIS-R WA IS-R Short Forms WAIS-R Subtests Interpreting the WAIS-R Assets of the WAIS-R Limitations of the WAIS-R Test Your Skill Summary

219

220

CHAPTER

T h e We c h s l e r A d u l t I n t e l l i ge n c e S c a l e - Rev i sed (WAIS-R; Wechsler, 1981) is the latest edition of an instru­ ment introduced in 1939. In its original version, it was called the Wechsler-Bellevue Intelligence Scale - Form I (Wechsler, 1939), after David Wechsler and Bellevue Hos­ pital in New York City, where Wechsler served as chief psychologist. A second form of the Wechsler-Bellevue, Form II, was published in 1946. Form I was revised in 1955 and again in 198 1 . The WISC-R and WPPSI are also derivatives of the 1939 adult scale. Because this textbook focuses on testing children, use of the WAIS-R with exam­ inees 16 to 17 years old will be highlighted. The WAIS-R contains II subtests grouped into Verbal and Performance sections. The six Verbal Scale subtests a re Information , Digit Span , Vocabulary, Arithmetic, Comprehension , and Similarities ; the five Performance Scale subtests are Picture Completion, Picture Arrange­ ment, Block Design, Object Assembly, and Digit Symbol . (Digit Symbol is similar to Coding B on the WISC-R . ) The WAIS-R covers an age range from 16 years, 0 months to 74 years, 11 months. It overlaps with the WISC-R from 16 years, 0 months to 16 years, 11 months. The WAIS-R is similar to the 1955 Wechsler Adult Intel­ l igence Scale (WAIS; Wechsler, 1955) . About 80 percent of the items are the same or only slightly modified . Some of the new items and revised procedures were developed on the basis of cultural considerations . For example, two of the WAIS-R Information items make reference to famous black Americans (Louis Armstrong and Martin Luther King) . WAIS items that had proved to be either too easy or too hard were eliminated from the WAIS-R . Also , scoring procedures on the WAIS-R Digit Span were altered to increase the variability of scores.

ST A N DARDI ZATI O N

T h e WAIS-R was standardized o n 1 , 880 white and non­ white A mericans equally divided with respect to gender, selected to be representative of the U . S . late adolescent and adult population during the 1970s. The demographic characteristics used to obtain a stratified sample were age, sex , race (white = 1 , 664, black = 192 , and Asians plus native A mericans = 24) , geographic region (Northeast, North Central , South, and West) , education, and urban­ rural residence . In the standardization sample , there were nine different age groups (16-17, 18-19, 20-24 , 25-34, 35-44 , 45-54, 55-64, 65-69 , and 70-74) , w ith 160 to 300 i ndividuals in each group .

10

WECHSLER ADULT INTELLIG ENCE SCALE-REVISED (WAIS-R)

DEVIAT I O N IQS A N D SCALED SCORES

The WAIS-R , like the WISC-R and the WPPS I , employs the Deviation IQ ( M = 100, SD = 15) for the Verbal , Per­ formance, and Full Scales and standard scores for the subtests (M = 10, SD = 3 ) . In order to obtain Deviation IQs, one first converts raw scores into scaled scores, using a table on the front of the WAIS-R record booklet or Table 19 in the WAIS-R manual . These scaled scores are based on a reference group of subjects in the standardization sample who were 20 to 34 years of age . The scaled scores are then summed, and the sum and the examinee's age are used to find Deviation IQs in Table 20 in the WAIS-R manual . Verbal Scale IQs are based on the sum of the six Verbal subtests, Performance Scale IQs on the sum of the five Performance subtests, and Full Scale IQs on all 1 1 subtests . Prorating Procedure The WAIS-R manual provides a table (Table 23) for prorat­ ing the scores when five of the Verbal Scale subtests are administered, or when four of the Performance subtests are administered. When even fewer subtests are adminis­ tered, IQs should be computed by using the specialized short-form procedure described later in this chapter. Scaled Score Equivalents for Age Groups The WAIS-R manual (Table 21) also provides scaled score equivalents (M = 10, SD = 3) of raw scores for each of the nine age groups in the standardization sample. This table is useful in conducting profile analysis and in comparing an individual's performance directly with that of age peers; however, the age-corrected scaled scores should not be

used to calculate lQs. Age-corrected subtest scaled scores should be used only to make subtest interpretations and comparisons . These age-corrected scores, in addition to the reference group scaled scores , should be recorded for every WAIS-R protocol to avoid interpretive errors.

R E LIABI LITY

The WAIS-R provides highly reliable IQs. Each of the three IQ scales has an internal consistency reliability co­ efficient of . 88 or above in the standardization group over the entire age range covered by the scale. Reliability coeffi­ cients, based on a formula for computing the reliability of a composite group of tests, range from . 95 to .97 (average

22 1

RELIABILITY rxx =

. 97) for the Verbal IQ, from . 88 to . 94 (average .93) for the Performance IQ, and from . 96 to . 98 (average rxx = . 97) for the Full Scale IQ . (Similar findings have been reported for a clinical sample by Ryan, Prifitera, and Larsen , 1982 . ) The reliabilities for the individual subtests are less satis­ factory than those for the IQs. They range from a low of . 52 for Object Assembly at ages 16-17 to a high of . 96 for Vocabulary at six of the nine age groups. The average reliability coefficients range from . 68 for Object Assembly to . 96 for Vocabulary (Mdn rxx = . 83 ) . Similar findings have been reported for a clinical sample (Ryan et aI . , 1982 ) . Table 10-1 presents the reliability coefficients for each subtest and scale for ages 1 6 to 17 and for the average of the standardization sample . The highest reliabilities are found among the Verbal Scale subtests (average reliabilities range from . 83 to .96) . rxx =

Block Design is the most reliable subtest (r = . 87) in the Performance Scale (average reliabilities in the Perfo r­ mance Scale range from . 68 to . 87). The reliability coeffi­ cients for 9 of the 1 1 .subtests are split-half correlations corrected by the Spearman-Brown formula. For D igit Span and Digit Symbol , reliabilities are based on test­ retest correlations because the items in these subtests do not lend themselves to the split-halfcalculation procedure . xx

Standard Errors of Measurement The standard errors of measurement (SEm) in IQ points , based on the average of the nine age groups, are 2 . 5 3 for the Full Scale IQ, 2 . 74 for the Verbal IQ, and 4 . 14 for the Performance IQ. Thus, more confidence can be placed in the Full Scale IQ than in either the Verbal or the Perfor­ mance Scale IQ. The Verbal Scale subtests (average SEn/'S

Table 1 0- 1 Reliability Coefficients and Standard Errors of Measurement for WAIS-R Subtests and Scales for 1 6- to 1 7-Year-Olds and for the Average of the Nine Age Groups in the Standardization Sample

Reliability coeffiCient WAlS-R subtest or scale

1 6- to 1 7year-olds

Standard error of measurement

Average of standardization group

1 6- to 1 7year-olds

Average of standardization group . 93 1 .2 3 .61 1 . 14 1 . 20 1 .24

I nfonnation Digit Span Vocabulary Arithmetic Comprehension S imilarities

. 90 . 70 .96 .73 .78 . 80

. 84

. 84 1 . 44 .49 1 .20 1 . 16 1 . 29

Picture Completion Picture Arrangement Block Design Object Assembly Digit Symbol

.71 . 66 . 87 . 52 .73

.81 . 74 . 87 .68 .82

1 . 43 1 . 47 .97 1 .9 1 1 .53

1 .2 5 1 .4 1 .98 1 . 54 1 . 27

Verbal IQ Perfonnance IQ Full Scale I Q

. 95 . 88 . 96

.97 .93 . 97

3 . 30 5. 1 8 2 . 96

2 . 74 4. 14 2.53

. 89 . 83 . 96 . 84 . 84

Note. Reliability coefficients for 9 of the 1 1 subtests (all but Digit Span and Digit Symbol) are split-half correlations. For Digit Span and Digit Symbol, coefficients are based on test-retest data obtained on subsamples from the standardization group of 48 to 80 individuals in four age groups , who were retested after a 1 - to 7-week interval. Verbal , Performance, and Full Scale reliability coefficients are based on a formula for computing the reliability of a composite group of tests. Source: Adapted from Wechsler ( 1 98 1 ) .

222

CHAPTER

range from .61 to 1 . 24 scaled score points) usually have smaller standard errors of measurement than do the Per­ formance Scale subtests (average SEm's range from .98 to 1 . 54 scaled score points). Within the Verbal Scale, Vocab­ ulary and Information have the smallest average SEII1's ( . 6 1 and . 9 3 scaled score points , respectively ) ; within the Per­ formance Scale, Block Design and Picture Completion have the smallest average SEII1's ( . 98 and 1 . 25 scaled score points, respectively) . Stability The stability of the WAIS-R was assessed by retesting two groups (71 individuals between the ages of 25 and 34 and 48 individuals between the ages of 45 and 54) in the standardization sample after an interval of two to seven weeks (Wechsler, 1981 ) . The stability coefficients for ages 25 to 34 were . 95 for the Full Scale IQ, . 94 for the Verbal IQ, and . 89 for the Performance IQ. For the 11 subtests, the stability coefficients ranged from . 69 for Picture Arrange­ ment to .93 for Vocabulary. The stability coefficients for ages 45 to 54 were generally similar to those for the younger group. For the three IQs, the stabilities again were high : . 96 for the Full Scale IQ, . 97 for the Verbal IQ, and . 90 for the Performance IQ. Of the 11 subtests, Obj ect Assembly had the lowest stability coefficient (rxx = . 67) and Information had the highest (ru = . 94). The Verbal, . Performance, and Full Scale test-retest IQs, standard de­ viations, and test-retest score changes for the two groups are shown in Table 10-2 . In the combined sample, the mean changes upon retest after two to seven weeks were 3 . 3 points on the Verbal IQ, 8 . 4 points on the Performance IQ, and 6.2 points on the Full Scale IQ (Matarazzo & Herman, 1984a) . These differ-

10

WECHSLER ADULT INTELLIGENCE SCALE-REVISED (WAIS-R)

ences likely reflect short-term practice effects. The prac­ tice effect was greater for the Performance Scale than for the Verbal Scale. The range of change was - 12 to + 15 for the Verbal Scale IQ, - 12 to + 2 8 for the Performance Scale IQ, and - 12 to + 20 for the Full Scale I Q . Thus , both gains and losses occurred on retest . Similar findings have been reported with adult clinical groups (Ryan , George­ miller, Geisser, & Randall , 1985). The test-retest stability coefficients for the three IQs were all highly significant, yet considerable within-subject variability was demonstrated . These findings underscore the distinction between psychometric and cli nical retest stabi lities (Matarazzo & Herman, 1984a) . H igh psycho­ metric stability is reflected by a sizable test -retest correla­ tion, whereas high clinical stability is demonstrated by the absence of meani ngful score change. For the standardiza­ tion sample, psychometric stability was satisfactory, but clinical stability was unimpressive ; over 80 percent of the subjects showed changes in Full Scale IQ that exceeded the average SE",. The WAIS-R IQ appears to be less stable than one might infer from the test-retest coefficients alone, and large changes in IQ can occur upon retesting . Additional behavioral or clinical corroborative data are needed for valid i nterpretation of test-retest changes .

Precision Range Table C-25 in Appendix C shows the 68, 85, 90, 95, and 99 percent confidence intervals for each scale for all age groups and for the average of the nine age groups in the standardization sample. The examinee's specific age group should be used to obtain the most accurate confidence leve l .

Table 1 0-2 Test·Retest WAIS·R IQs for Two Groups

First testing

Second testing

Age

Scale

M IQ

SD

M IQ

SD

Change

25-34 years (N = 7 1 )

Verbal IQ Performance IQ Full Scale IQ

1 02 . 0 103 . 0 1 02 . 4

14.0 15.6 1 5 .0

1 05 . 3 1 1 1 .9 1 09 . 0

14.3 1 7.2 16.8

+3.3 + 8.9 +6.6

45-54 years (N = 48)

Verbal IQ Performance IQ Full Scale IQ

1 0 1 .0 97 . 3 99.0

14.9 1 2 .9 1 3 .5

1 04 . 1 1 05 . 0 1 04 . 7

15.7 1 5 .6 1 6. 1

+3.1 +7.7 +5.7

Source: Adapted from Wechsler ( 1 98 1 ) .

223

VALIDITY

VALI D I TY

The criterion validity of the WAIS-R has been investigated in a variety of studies by correlating the WAIS-R with the WAIS, WISC-R, Stanford-Binet: Fourth Edition , other intell igence tests , measures of achievement, and years of school ing. Evidence of construct validity has been pro­ vided b y : (a) the level and pattern of intercorrelations between WAIS-R subtests and scales; (b) the observation that scores on the WAIS-R are distributed in a manner consistent with theoretical expectations; and (c) the results of factor analyses applied to the intercorrelations of the 11 subtests. WAIS-R and WAI S A sample o f 72 individuals i n the 35 t o 44 age group o f the standardization sample were administered the WAIS-R and WAIS in counterbalanced order within a three- to six-week period (Wechsler, 1981) . The correlations between the tests were . 9 1 for the Verbal Scale, . 79 for the Performance Scale, and . 88 for the Full Scale. For the 11 subtests, correlations ranged from a low of . 50 for Picture Arrange­ ment to a high of .91 for Vocabulary (Mdn r = .79). Subtests on the Verbal Scale have higher correlations (range of . 7 1 to . 91 , Mdn r = . 86) than do those on the Performance Scale (range of . 50 to . 85 , Mdn r = . 66) . The high correlations are IIOt surprising, because most items are the same in both tests. What is less clear is why the correlations are relatively low for some Performance Scale subtests . Shifts in scores on the Performance Scale may be due to practice effects . For each of the three scales , mean IQs were lower on the WAIS-R than on the WAIS. The difference was 6 . 9 points on the Verbal Scale (101 . 8 vs. 108 . 7 ) , 8 .0 points on the Performance Scale (105 .4 vs. 113 .4) , and 7 . 5 points on the Full Scale (103 . 8 vs. 111 . 3 ) . These results were based on a small sample of the standardization group (those aged 35 to 44) and should not be generalized to other age groups. Studies that have compared the WAIS-R and the WAIS within a variety of populations almost uniformly indicate that the WAIS-R provides lower scores than does the WAIS (Mdn differences = - 6 . 6 , - 6.4, and - 6 . 8 for the Ver­ bal, Performance, and Full Scale IQs, respectively), but correlations between the two scales are high (Mdn r = . 94 for the Verbal Scale, . 86 for the Performance Scale, and . 94 for the Full Scale) (Edwards & Klein , 1984 ; Kelly, Montgomery, Felleman , & Webb, 1984; Lewis & Johnson, 1985 ; Lippold & Claiborn, 1983 ; Mishra & Brown, 1983 ; M itchell , Grandy & Lupo, 1986; Prifitera & Ryan , 1983; Rabourn, 1983 ; Rogers & Osborne, 1984 ; Ryan , Rosen-

berg , & Heilbronner, 1984 ; Simon & Clopton , 1984; Smith , 1983 ; Urbina , Golden, & Ariel , 1982 ; Warner, 1983 ; Wechsler, 1981) . Full Scale IQ differences were larger for persons of average and low average abil ity (Mdn = - 8 . 4) than for those of high average (Mdn = - 5 . 3) or superior (Mdn = - 3 . 7) ability. In contrast, in a study with mentally retarded subjects , the WAIS-R Verbal ( + 3 . 9) and Full Scale ( + 2 . 1) IQs were found to be higher than corresponding WAIS values. Thus studies suggest that different relationships may exist between the WAIS-R and the WAIS at different points on the intell igence dis­ tribution . Overall , WAIS-R and WAIS IQs are not inter­ changeable . I n assessing individuals who have received an initial evaluation with the WAIS and are later retested with the WAIS-R, it is important to keep in mind that individuals are likely to obtain lower scores on the WAIS-R. If the WAfS-R fQ equals or exceeds the previously obtained WAfS fQ, the probability is high that an improvement in cog­ nitive efficiency has occurred. If, however, the WAfS-R fQ is lower than the WAfS fQ, do not infer automatically thaI intellectual deterioration has occurred. Consider the ex­ tent of the difference. A WAIS-R IQ that is much lower than the WAIS IQ may suggest some loss of functioning , but this hypothesis must be supported by clinical data.

WAI S-R and WISC-R The relationship between the WAIS-R and the WISC-R was discussed in Chapter 6 . The research suggests that the WAIS-R yields slightly higher IQs than does the WISC-R, particularly with low-functioning individuals. More re­ search is needed on the comparability of the two scales , however. WAIS-R and Stanford-Binet: Fourth Edition

A sample of 47 normal individuals (M age = 19-5) were administered the WAIS-R and Stanford-Binet: Fourth Edi­ tion (Thorndike et al . , 1986b) within a two-week interval . The mean composite score on the Stanford-Binet : Fourth Edition was 98 . 7 , and the WAIS-R Verbal , Performance, and Full Scale IQs were 100 . 2 , 103 . 7 , and 102 . 2 , respec­ tively. Correlations between the Stanford-Binet and WAIS-R scales were . 90 for the Verbal IQ, . 85 for the Performance IQ, and .91 for the Full Scale IQ . In another study, 2 1 mentally retarded i ndividuals (M age = 19-6) were administered the WAIS-R followed by the Stanford-Binet : Fourth Edition , with a median interval of three weeks between the two test administrations

224

CHAPTER

(Thorndike et al . , 1986b). The mean Stanford-Binet com­ posite score was 63 . 8 , and the WAIS-R Verbal , Perfor­ mance , and Full Scale IQs were 74 .0, 74 . 2 , and 7 3 . 1 , respectively. Correlations between the composite score and the Verbal, Performance, and Full Scale IQs were . 74 , . 68, and .79, respectively. Like those for normal subjects , results for retarded individuals suggest that the WAIS-R yields higher scores than does the Stanford-Binet: Fourth Edition . Other Concurrent Validity Studies Table 10-3 summarizes the results of studies that have correlated the WAIS-R with tests of ability, tests of achieve­ ment , and years of formal education (ability: Edinger, Shipley, Watkins, & Hammett, 1985 ; Fowles & Tunick, 1986; Gregg & Hoy, 1985 ; Heinemann , Harper, Friedman, & Whitney, 1985 ; Hiltonsmith , H ayman, & Kleinman, 1 9 8 4 ; Klett , Watso n , & Hoffman , 1 9 8 6 ; K l i n g & Kupersmith , 1984 ; Maxwell & Wise, 1984; Retzlaff, Shcner, & Gibertin i , 1986; Zachary, C r u mpto n , & Spiegel, 1985 ; achievement: Ryan & Rosenberg, 1983; formal education : Matarazzo & Herman, 1984b). The studies indicate that the WAIS-R has satisfactory concur­ rent validity w ith intelligence tests, picture vocabulary tests, achievement measures, and years of education. Me­ dian correlations shown in Table 10-3 between the various measures and the WAIS-R Full Scale IQ range from .43 to . 94. Construct Validity of the WAIS-R One way to assess construct validity is to determine whether scores from the WAIS-R conform to the expecta­ tions or predictions of a viable theory of intelligence . The Cattell-Horn theory (see Chapter 3) postulates that fluid i ntelligence involves the ability to solve novel problems and process new material , whereas c rystallized intel­ l igence involves the retrieval of well-learned facts . Fluid intell igence is bel ieved to decline w ith advancing age, whereas crystallized intelligence is said to show little or no age-related deterioration . The age norms for the 11 WAIS-R subtests indicate that with advancing age verbal ability declines much less than perceptual organization and motor skills (Sattler, 1982a) . Corrections for age-related decrements are minimal on the Verbal Scale, w ith the exception of the Similarities subtest (see Table 10-4) . Conversely, marked changes are shown on the Performance Scale subtests, for which additional scaled score points are awarded as a function of age . If the Performance Scale is considered a measure of fluid intel-

10

WECHSLER ADULT INTELLIGENCE SCALE-REVISED (WAIS-R)

l igence and the Verbal Scale a measure of c rystallized intelligence, then the WAIS-R age norms are consistent with the Cattell-Horn theory : fluid intelligence, but not crystallized intelligence , shows a marked decrement w ith advancing age . Factor analysis can also assist in construct validation because it provides a method for determining the structure and components of intelligence measured by a given test . Factor analysis of the WAIS-R standardization sample indicates that all 11 subtests measure general intelligence (g) with a moderate to high degree of success (Blaha & Wallbrown , 1982 ; Gutki n , Reynolds, & Galvin, 1984; O'Grady, 1983 ; Parker, 198 3 ; Silverstein, 1982a) . These results provide support for interpretation of the Full Scale IQ as a global measure of intelligence. Evidence supportTable 1 0·3 Concurrent Validity Studies of the WAIS·R

Correlations

Criterion Wechsler Adult Intel­ ligence Scale" Wechsler Intelligence Scale for Children ­ Revised" Stanford-Binet: Fourth Edition" Wonderlie Personnel Test Henmon-Nelsonb Siosson Intelligence Test Revised Beta Examination Quick Teste Shipley-Hartford Scaleb Peabody Picture Vocabulary Test Wide Range Achievement Test ( 1 978) Revision Reading Spelling Arithmetic Woodcock-Johnson Tests of Cognitive Ability (Broad Cognitive Cluster Score) Years of education

Verbal IQ

Performance IQ

Full Scale IQ

.94

. 86

. 94

. 83

. 76

.87

. 82

. 77

.77 . 83

. 62 .51

. 85 . 75 . 79 .78

.27

.53

. 80

.68

043 . 75 . 76

.78

. 62

. 76

.68 . 67 .76

AI

0 42 . 66

. 62 . 60 .76

.73 .56

, Median correlation s . Median correlation for Full Scale only. C Median of Forms 1 , 2, and 3 . b

AI AI

. 69 . 54

INTERCORRELATIONS BETWEEN SUBTESTS AND SCALES

225

Table 1 0-4 Additional Scaled Score Points Awarded on WAI S-R Subtests When the Reference Group Receives a Scaled Score of 1 0

subdivision of the WAIS-R subtests (see the section on factor anal ysis later in this chapter) . The available research provides substantial support for the construct validity of the WAIS-R .

Age group WAIS-R subtest Verbal Scale Information Digit Span Vocabulary Arithmetic Comprehension Similarities Perfonnance Scale Picture Completion Picture Arrangement Block Design Object Assembly Digit Symbol

35-44 45-54 55-64

65-69

0 0

0 0

-I 0 0

0 I

0 I

0 0 0

0 I 0

0

I 0

I 2

2 2 2 2 3

70- 74 I N T E RCORRELATI O N S BETW E E N S U BT ESTS A N D SCALES

I

I

2

3

2 4 3 4 5

4 5 4 4 6

Note. The results in Table 1 0-4 indicate that raw scores yielding a scaled score of 10 in the WAIS-R reference group yield, in nearly every case, the same or higher scaled scores in the five age groups over 34 years. The greatest change is at ages 70 to 74 years. For example, in the reference group, a raw score of 57 on Digit Symbol yields a scaled score of 1 0 . but at ages 70 to 74 L'Jis same raw score yields a scaled score of 1 6 . The 6 additional scaled-score points awarded at ages 70 to 74 change the percentile for a raw score of 57 from the 50th to the 98th. Digit Symbol exhibits the most change of any WAIS··R subtest, showing a steady increment in scaled-score points from ages 35 to 74. The greatest changes are consistently shown on the Performance Scale subtests . On the Verbal Scale subtests, the increment in caled-score points is never greater than I point, except for Similarities. For this subtest, two additional scaled-score points are awarded at ages 65 to 69, and three additional scaled-score points are awarded at ages 70 to 74 for a performance that is average in the reference group. The increments in Table 1 0-4 actually reflect a decline in performance ability. The decline is more graphically revealed when we examine the raw-score points needed at the various age groups to obtain an average scaled score . For example, in the reference group, a Digit Symbol raw score of 57 yields a scaled score of 1 0 . At ages 70 to 74, however, a raw score of only 29 is required to obtain a scaled score of 1 0 . Thus, individuals in the oldest age group need 28 fewer raw-score points than do those in the reference group to obtain average status in their age group. Source: From J. M. Sattler, "Age effects on Wechsler Adult Intelligence Scale - Revised tests , " Journal of Consulting and Clinical Psychology, 1 982, 50, p. 786 . Copyright 1 982 by the American Psychological Asso­ ciation. Reprinted by permission.

ing interpretation of the Verbal and Performance IQs as separate entities is also available (Atkinson & Cyr, 1984 ; Beck, Horwitz , Seidenberg, Parker, & Frank, 1985 ; Ryan, Rosenberg , & DeWolfe , 1984; Silverstein , 1982a) , al­ though some researchers suggest that the Verbal and Per­ formance IQs do not always constitute the most accurate

The WAIS-R manual provides tables that show the inter­ correlations between the 11 subtests and three scales for each of the nine age groups and for the average of the nine age groups. At ages 16 to 17 the subtest intercorrelations range from a low of .21 to a high of . 79 (Mdn = . 43 ) . The six highest subtest intercorrelations are for Vocabulary and I nformation ( . 79 ) , Similarities and Vocabulary ( . 74) , Comprehension and Vocabulary ( . 74), Information and Similarities ( . 7 1 ) , Similarities and Comprehension ( . 66), and Information and Comprehension ( . 65) . The six lowest subtest intercorrelations are for Object Assembly and Digit Symbol ( . 2 1 ) , Picture Arrangement and Digit Sym­ bol ( . 23 ) , Picture Arrangement and Object Assembly ( . 25 ) , Picture Arrangement and Digit Span ( . 28), Object Assembly and Arithmetic ( . 29) , and - in a tie - Object Assembly and Information ( . 30) and Block Design and Picture Arrangement ( . 30) . The intercorrelations among the six Verbal Scale sub­ tests at ages 16 to 17 range from . 36 to . 79 (Mdn = . 61), whereas the intercorrelations among the five Performance Scale subtests range from .21 to . 57 (Mdn = 36 ) The Verbal Scale subtests are therefore more highly intercorre­ lated than are the Performance Scale subtests. Correla­ tions between the 11 subtests and the Full Scale range from a low of . 46 (Picture Arrangement and Object Assembly) to a high of . 82 (Vocabulary ) . Five of the six Verbal Scale subtests (Digit Span is the exception) correlate more highly with the Full Scale than do any of the Performance Scale subtests . Of the five Performance Scale subtests, Block Design has the highest correlation with the Full Scale ( 67 ) The trends observed at ages 16 to 17 are also evident in the intercorrelations based on the average of the nine age group s . Intercorrelati (:m s range from . 3 3 to . 8 1 (Mdn = . 48) . The Verbal Scale subtests have higher inter­ correlations ( . 45 to . 8 1, Mdn = 61) than do the Perfor­ mance Scale subtests ( . 38 to . 63 , Mdn = .47). The Verbal Scale subtests correlate more highly with the Full Scale ( . 58 to . 8 1 , Mdn = . 75) than do the Performance Scale subtests ( . 57 to . 68 , Mdn = . 61 ) . Vocabulary correlates .

.

.

.

.

CHAPTER 1 0

226

WECHSLER ADULT INTElliGENCE SCALE-REVISED (WAIS-R)

Table 1 0·5 Relationship of WAIS·R IQs to Sex, Race, Occupation, Urban-Rural Residence, Geographic Residence, and Years of Education

Verbal lQ

Performance IQ

Full Scale IQ

Demographic variables

N

M

SD

M

SD

M

SD

Sex

Males Females

940 940

1 00 . 9 98.7

15. 1 14.7

1 00 . 6 99. 2

15.2 15. 1

1 00 . 9 98.7

15.3 1 5 .0

Race and sex

White males Black males White females Black females

836 93 828 99

1 02 . 3 88.2

14.7 13. 1 14.3 13.2

1 02 . 0 88.0 1 00 . 8 86.7

14.7 14.8 14.7 1 2 .4

1 02 . 4 87 . 3 1 00 . 4 86.4

1 3 .6 14.6 12.3

Occupationa

Whites I . Professional and technical 2 . Managerial, clerical, sales 3 . Craftsmen and foremen 4 . Operatives, service workers, farmers 5 . Laborers 6. Not i n labor force

378 200 329 51 515

98.6 94 .4 90. 6 1 00 . 8

12. 1 1 1 .9 1 2 .0 13.2 15.9 1 5 .2

1 09 . 2 1 04 . 2 101 . 8 96.4 93 .4 1 00 . 2

14.0 12.5 13.6 14.8 15.2 1 5.0

1 12.2 1 05 .0 99 . 8 94. 8 91.3 1 00 . 6

1 3 .0 1 2. 0 12.6 13.6 15.4 15.3

92 . 8 95.4 95 . 7 85 . 1 83 . 8 86.5

9.8 13.3 10.2 12.7 13. 1 12.5

95 . 5 92 . 1 93.9 85.9 83 .3 85 .4

9.9 15.6 1 3 .4 13.8 14.5 1 1 .5

93.2 93.2 94. 4 84 . 6 82.4 85.2

9.4 14.6 1 1.3 13.0 1 3 .4 1 1.1

1 1 1 .3

12.8 12.3 1 1 .9 13.5 15.3 15.5

1 08 . 4 1 03 . 4 1 0 1 .3 94 . 5 90.9 98.7

14.0 13. 1

1 1 1 .0 1 04 . 1

68 5 80

1 04 . 3 98.4 92.7 88.9 99.2

13.7 15.2 15.5 15.3

99 . 5 93.0 89.0 98.9

1 3 .4 12.6 1 2 .6 14. 1 15.2

1 42 1 459

1 00 . 4 98.0

15.0 14.4

1 00 . 0 99.4

15.1 15.3

1 00 . 3 98.4

1 5 .2 14.9

Blacks I . Professional and technical 2 . Managerial, clerical , sales 3 . Craftsmen and foremen 4. Operatives, service workers, farmers 5 . Laborers 6. Not i n labor force Total Group 1 . Professional and technical 2 . Managerial, clerical, sales 3 . Craftsmen and foremen 4 . Operatives, service workers, farmers 5 . Laborers 6 . Not i n labor force

191

10 29 9 68 16

60 206 409 213 404

1 00 . 2 87.5 1 1 2 .4 1 05 . 1

14.8

1 5 .6

Urban-rural residence

Urban Rural

Geographic residence

Northeast North Central South West

465 497 576 342

101.7 98.6 98.6 1 0 1 .0

14.8 14.3 15.7 14.3

1 0 1 .4 1 00 . 2 97.2 1 0 1 .9

1 5.0 1 4. 4 16.2 14.2

101.7 99. 1 97. 9 1Ol.5

1 5 .0 14.2 16.2 14.4

Years of education

I. 2. 3. 4. 5. 6.

133 158 472 652 25 1 2 14

82 . 2 90 . 2 96. 1 1 00 . 1 107.7 1 15.7

13.6 1 1 .0

84 . 5 93. 1 97 . 8 1 00 . 2 105.7 1 1 1 .2

14.9 14.3 14.8 13.5 1 2 .0 13.0

82.2 90. 7 96.4 1 00 . 1 1 07.4 1 15.3

13.6 1 2. 0 14.3 12.6 11.1 12.2

0-7 8 9- 1 1 12 1 3- 1 5 16 and u p

13.8 12. 1 10.9 1 1 .6

a For ages 1 6- 1 7 and 1 8- 1 9 , stratification of the samples was according t o occupation of head of household. Source: Adapted from Chastain and Reynolds ( 1 984).

227

FACTOR ANALYSIS

with 16 or more years of education (82 vs. 115). Table 10-6 presents a breakdown of IQs on the three scales according to sex and race for 16- to 17-year-olds . As in the total sample, sex differences are minimal . Race differences are similar to those for the total group. The correlation between years of school completed and the Full Scale IQ was . 54 for the total standardization sample (Matarazzo & Herman , 1984b) . There was a pro­ gressive increase in mean Full Scale IQ as a function of years of school completed: 8 or fewer years of education, M IQ = 86.4; 9-11 years, M IQ = 96.4; 12 years, M IQ = 100 . 1 ; 13-15 years, M IQ = 107 . 4 ; 16 or more years , M IQ = 115 . 3 . Matarazzo and Herman recommend that

more h ighly ( . 81) with the Full Scale than does any other subtes! . The subtests with the lowest correlations with the F u l l Scale are Object Assembly ( . 57) and Digit Symbol ( . 57 ) .

WAI S- R I Q S A N D STRAT I F I CATION VAR I A B LES

The relationship of WAIS-R IQs to various stratification variables used in the standardization sample is shown in Table 10-5 (Chastain & Reynolds, 1984) . Differences be­ tween the mean IQs for males and females on the three scales were less than 3 points . Likewise, less than 3 points separated the IQs for the urban and rural subjects of the normative sample. Slightly larger differences emerged when area of geographic residence was considered , the most sizable one being the 3 . 8 Full Scale IQ points that separated subjects l iving in the Northeast (M Full Scale IQ = 101 . 7 ) from those residing in the South (M Full Scale IQ = 97 . 9 ) . Differences attributed to sex , urban versus rural residence , and geographic location are not large enough to assume any practical significance and, in most cases, may be ignored for interpretive purposes. Mean IQs show a much clearer relation to race , occupa­ tion , and education . For example, white subjects in the standardization sample scored about 15 points higher than black subjects on the Full Scale (101 .4 vs. 86.8) . The IQs of those employed in professional/technical jobs were 22 p o i n t s h i g h e r on average t h a n tho se of l abore rs ( I l l vs. 89) . The most striking difference - 33 points - was between those w ith less than 8 years of education and those

when you need a crude estimate of an adult examinee's premorbid (before the onset of disease) WAIS-R IQ, you can use (he mean IQ values reported above to arrive at an estimate of such a WAIS-R IQ value. This should be done only when no other estimates are available (such as high school transcripts or test scores).

FACTOR AN ALYSI S

Numerous factor analytic investigations o f the WAIS-R have been conducted on the standardization sample (Blaha & Wal lbrow n, 1982 ; Glass, 1982; Gutkin et aI . , 1984 ; O'Grady, 1983 ; Parker, 198 3 ; Silverstein, 1982a) as well as on various clinical samples (Atkinson & Cyr, 1984; Faulstich , McAnulty, Gresham, Veitia, Moore , Bernard, Waggoner, & Howell, 1986 ; Ryan , Prifitera, & Rosenberg, 1983 ; Ryan, Rosenberg , & DeWolfe , 1984 ; Ryan & Schneider, 1986) . Results of these studies indicate that the WAIS-R may be characterized as either a two- or a three-

Table 1 0·6 Relationship of W AIS·R IQs to Sex and Race for 1 6· to 1 7· Year·Olds

Verbal lQ Demographic variables Sex

Males Females

Race and sex

White males Black males White females Black females

Source: Adapted from Chastain and Reynolds ( 1 984).

Performance IQ

Full Scale IQ

N

M

SD

M

SD

M

SD

1 00 100

1 00 . 9 99 .4

16.0 13.6

101 .2 99 . S

I S .9 14. 1

101 . 1 99 . 2

16.2 1 3 .4

8S

103.4

14.6

1 03 . 8

1 03 . 8

14.8

86.4 l OO. S 90. S

1 6.9 13.2 1 3 .9

14.6

14 87 12

8S . 6 1 00 . 8 88.0

IS.6 13.3 12.8

8S.4 100.4 88.7

16.S 12.7 13. 1

228

CHAPTER 1 0

factor battery. A three-factor solution is useful for clinical and psychoeducational tasks, but when individual pro­ tocols are being interpreted , any hypotheses based on factor analytic findings should be supplemented with clinical judgment . The three WAIS-R factors are similar to those on the WISC-R - namely, Verbal Comprehension (Information , Vocabulary, Comprehension , and Similarities), Perceptual Organization (Picture Completion [except at 18-19 years] , Block Design, and Object Assembly), and Freedom from Distractibility (Digit Span and Arithmetic) (see Table 10-7). The subtests comprising the factors, however, differ somewhat for the two scales. Coding loads on the Freedom from Distractibility factor on the WISC-R, but its counter­ part on the WAIS-R, Digit Symbol , does not . Similarly, Picture Arrangement is one of the subtests in the Percep­ tual Organization factor on the WISC-R, but not on the WAIS-R . There is no simple explanation for these find­ ings . (All factor analytic findings presented in Table 10-7 and below are based on work by the chapter authors . ) WAIS-R Subtests as Measures of g As shown in Table 10-8, all of the WAlS-R subtests are either good or fair measures of g (the general intelligence factor) . Based on the average of the nine age levels, the subtests with the highest g loadings are Vocabulary ( . 86) , Information ( . 8 1 ) , Similarities ( . 79) , Comprehension ( . 78) , Arithmetic ( . 75 ) , and Block Design ( . 72 ) . The sub­ tests with fair g loadings are Picture Completion ( . 70) , Picture Arrangement ( . 63) , Digit Span ( . 62) , Object As­ sembly ( . 61) , and Digit Symbol ( . 59). Overall , the Verbal sub tests are better measures of g than are the Performance subtests . Subtest Specificity Subtest specificity refers to the proportion of a subtest's variance that is both reliable (that is, not due to errors of measurement) and distinctive to the subtest. Although individual subtests overlap in their measurement proper­ ties (that is, the majority of reliable variance for most subtests is common factor variance) , many of them possess a relatively high degree of specificity, which justifies inter­ pretation of specific subtest functions (see Table 10-9 ) . Some subtests have either ample o r adequate specificity throughout the entire age range covered by the WAIS-R (Digit Span, Vocabulary, Picture Completion, Block De-

WECHSLER ADULT INTELLIGENCE SCALE-REVISED (WAIS-R)

sign, and Digit Symbol) , whereas others have inadequate specificity at one or more ages (Information, Arithmetic , Comprehension, Similarities, Picture Arrangement, and Object Assembly ) . Subtests with inadequate specificity should not be interpreted as measuring specific functions, and cautious interpretation is required for subtests falling within the adequate specificity category. These subtests can be interpreted as measuring g and the appropriate principal factor, however ( Verbal Comprehension, Percep­ tual Organization , and Freedom from Distractibi lity, respectively) . Factor Scores Factor scores can be obtained from the WAIS-R, permit­ ting the identification of meaningful psychological dimen­ sions. The Verbal Comprehension factor score measures verbal knowledge and understanding obtained by formal and informal education and reflects the ability to apply verbal skills to new situations. The Perceptual Organiza­ tion factor score reflects the ability to interpret and orga­ nize visually perceived material while working against a time limit. The Freedom from D istractibility factor score measures the ability to attend or concentrate, but also may involve numerical proficiency and sequencing s kills . These factor scores are discussed more fully in Chapter 8 . The preferred way t o obtain the three factor scores is to use the following combinations of subtests: Verbal Comprehension

=

Perceptual Organization

=

Freedom from Distractibility

=

Sum of age-corrected scaled scores on Information , Vocabulary, Comprehension, and Similarities at all ages Sum of age-corrected scaled scores on Block Design , Object Assembly, and Picture Completion at every age level (except 18-19, where the last subtest should be omitted from the factor score) Sum of age-corrected scaled scores on Digit Span and Arithmetic

The sums of the respective age-corrected subtest scaled scores comprising the three factors can be converted into Deviation IQs (M = 100 , SD = 15) . Table C-28 in Appen-

229

FACTOR ANALYSIS

Table 1 0-7 WAIS-R Subtest Loadings on Factor A (Verbal Comprehension), Factor B (Perceptual Organization), and Factor C ( F reedom from Distractibility) for N ine Age Levels and the Average Following Varimax Rotation

Age group WAIS-R subtesl

1 6- 1 7

1 8-19

20-24

25-34

35-44

45-54

55-64

65-69

70- 74

Av.

75 30 81

Factor A- Verbal Comprehension

Information Digit Span Vocabulary Arithmetic Comprehension Similarities

78 22 81 54 69 78

80 46 83 58 66 67

77 28 83 40 69 68

67 36 77 43 71 60

62 36 80 41 72 68

70 54 86 56 69 65

76 27 87 45 71 70

78 30 75 45 82 67

76 38 83 41 68 63

Picture Completion Picture Arrangement Block Design Object Assembly Digit Symbol

40 37 29 18 38

26 36 19 16 37

45 52 21 25 II

46 48 27 20 42

45 44 31 23 31

36 37 27 21

49 45 27 16 41

38 40 31 17 41

38 26 30 12 24

44

28 37 14 36 46

31 26 33 36 26 41

20 31 31 22 33 46

21 22 26 34 30 36

57 48 72 65 41

66 48 74 76 55

51 54 75 75 55

56 42 69 73 38

Factor B

-

44

20 23 31 25 33 25

21 31 21 53 19 30

18 II 20 27 31 30

30 19 28 30 26 38

28 17 24 42 26 31

32 34 21 49 20 29

Picture Completion Picture Arrangement Block Design Object Assembly Digit Symbol

46 24 79

19 21 77 48 52

59 33 67 67 26

64 44

43 32

67 78 24

64

41 33 74 71 45

23

71 67 42 27 19 32

Perceptual Organization

Information Digit Span Vocabulary Arithmetic Comprehension Similarities

60

44

88 25

44

Factor C - Freedom from Distractibility

Information Digit Span Vocabulary A rithmetic Comprehension Similarities

28 74 27 49 22 18

25 09 31 15 41 27

26 63 29 63 25 24

44 53 40 63 32 26

48 63 40 55 32 33

38 10 32 15 35 36

21 56 37 49 27 II

25 68 41 53 23 31

33 42 26 87 26 20

30

Picture Completion Picture Arrangement Block Design Object Assembly D igit Symbol

20 18 26 14 39

67 41 25 47 17

12 17 35 23 49

18 22 46 16 26

35 48 42 18 53

62 51 31 27 14

10 20 31 26 19

25 34 19 21 38

24 18 28 09 09

17 23 33 17 36

64

34 55 27 27

= average . Decimal points omitted. According to Parker (1983), Digit Span and Arithmetic should be included in the Freedom from Distractibility factor at ages 18-19 and 45-54 because these subtests have high loadings on this factor in a four-factor solution.

Note. Av.

CHAPTER 1 0

230

WECHSLER ADULT INTELLIGENCE SCALE-REVISED (WAIS-R)

Table 1 0-8 WAIS-R Subtests as Measures of g

Fair measure of g

Good measure of g

Subrest

Median g loading

Proportion of variance attributed to g (% )

Vocabulary I nformation Similarities Comprehension Arithmetic Block Design

. 87 . 82 . 79 . 78 .76 .73

76 67 62 61 58 53

Subtest

Median g loading

Picture Completion Picture A rrangement Object Assembly Digit Span Digit Symbol

.71 . 67 .63 . 62 . 60

Proportion of variance attributed to g

(%) 50 45 40 38 36

Note. These are median loadings based o n the nine age groups in the standardization sample. The square o f the median coefficients provides the proportion of each subtest's variance that may be attributed to g .

dix C provides for a rapid conversion of the three factor scores into Deviation IQs for ages 16 to 17 and for the average of the nine age groups.

A D M I N I S T E R I N G T H E WAIS-R

The general procedures discussed in Chapter 5 for admin­ istering psychological tests are also useful in administer­ ing the WAIS-R, as are the guidelines in Chapters 6 and 7 for administering the WISC-R. Be careful not to confuse the administration procedures of the WISC-R with those of the WAIS-R, however- different instructions and time lim­ its are used on some subtests of the same name . The general problems in administering the WISC-R also apply to the WAIS-R. The suggestions shown in Exhibit 10-1 should aid you in learning how to administer the WAIS-R. (Figure 10-1 shows the cover of the WAIS-R record booklet . ) Physical Abilities Required for the WAIS-R Adequate hearing and language functions are required for the Verbal Scale subtests , and adequate vision and visual­ motor ability are needed for the Performance Scale sub­ tests. The suggestions presented in Chapters 6 and 7 for administering the WISC-R to handicapped examinees (see particularly Table 6-11) should be carefully studied .

Subtest Sequence In most situations subtests should be administered in the standard sequence, as specified in the WAIS-R manual . If warranted, however, all of the Verbal subtests can be ad­ ministered first followed by the Performance subtests. This order may be less stressful for examinees who have minor motor problems that may interfere with the Perfor­ mance subtests . It is also permissible to adjust the order of administration for examinees who fatigue easily (for ex­ ample, the elderly or physically ill) or display marked anxiety about the testing situation. Examinees who fatigue easily may be given certain subtests (for example, Arith­ metic , Digit Span , or Digit Symbol) early in the session when their energy levels and attention-concentration skills are best. Anxious examinees may be started w ith subtests that are relatively nonthreatening and do not have strict time limits (for example, Information, Comprehension, or Vocabulary) . Some examinees with psychiatric o r neurological disor­ ders may not be able to complete the entire WAIS-R in a single session. In such cases, schedule breaks to coincide with the end of a subtest so that testing can easily be resumed at a later time. In the rare instances where a subtest must be interrupted prior to completion, resume administration of the subtest where you stopped, except on Similarities, Block Design, and Picture Arrangement . On these subtests, the easy items provide some examinees with the practice they need to succeed at more difficult items. Therefore , if Similarities , Block Design , or Picture Arrangement is i nterrupted, the first few items should be

23 1

ADMINISTERING THE WAIS-R

Table 1 0-9 Amount of Specificity for WAIS-R Subtests

WAIS-R

subtest

Ample specificity

Adequate specificity

Inadequate specificity

Ages

Ages

Ages

1 6-44, 55-74,

45-54

Information

D igit Span

1 6-74,

Average

Vocabulary A rithmetic

20-34, 45-74,

Average Comprehension

45-54

Average

1 6-74,

Average 1 8- 1 9 , 35-44

1 6- 1 7

25-44, 65-74,

1 6-24, 55-64

1 8- 1 9, 35-74,

1 6- 1 7 , 20-24

Average Similarities

25-34

Average Picture Completion Picture Arrangement

1 6-74,

Average

Average Block Design

20-24, 65-69

1 6- 1 9, 25-64, 70-74,

1 6-34, 55-69,

entry-point items as soon as they are admini stered . C redit is automatically given for the unadministered early items when the entry-point items (the first two, three, or four items administered) are answered correctly. If the entry­ point items are answered incorrectly, it is necessary to administer the earlier items before continuing the subtest . Discontinuance Rules Nine of the II subtests have discontinuance rules - the subtest is discontinued after a specified number of con­ secutive items are failed. (Object Assembly and Digit Symbol are the two subtests without discontinuance rules . ) If you administer additional items in a subtest because you are not sure whether the items at the discontinuance point were failed, use the following rule to score the additional items (Wechsler, 1981, p. 54) : lfsubsequent scoring of the

items indicates that the additional items were 'administered unnecessarily, do nOT give creditfor any items passed after The disconTilluance point. This rule , which is also followed i n the WISC - R , h e l p s to m a i n t a i n standardized procedures. Repetition of Items Use judgment in deciding when to repeat questions . The D igit Span items are the only verbal items that cannot be repeated .

35-54, 70-74

Average Object Assembly

1 8- 1 9

Digit Symbol

1 6-64, 70-74,

1 6- 1 7 . 20-74.

65-69

Average

Average

repeated at the next session so that the examinee can reestablish the mental set necessary to succeed on the harder items . Naturally, you must consider how any ad­ ministrative irregularity will affect the accuracy of the test results .

Use of Probing Questions and Queries Use probing questions when responses are ambiguous , vague, o r indefinite , o r when probing i s indicated b y a "(Q)" in the various scoring c riteria appendices in the WAIS-R manual . Testing-of-Limits Procedures Use modifications in test procedures designed for testing­ of-limits only after all of the subtests have been adminis­ tered according to standardized procedures . Scoring WAIS-R Responses

Starting Rules Eight of the II subtests are started with the first item . The remaining three subtests (Information , Vocabulary. and A rithmetic) are started with the fifth, fourth , and third items, respectively. On these three subtests , score the

Arriving at accurate scores for WAIS-R sub tests and scales is no simple matter. Scoring errors can have a significant impact on the accuracy of WAIS-R IQs . A study indicated that graduate students and experienced psychologists dif­ fered by 4 to 18 IQ points in the scores they gave to two

232 •

CHAPTER 10

WECHSLER ADULT INTELLIGENCE SCALE-REVISED (WAIS-R)

Exhibit 1 0- 1 Supplementary Instructions for Administering the W AIS-R

I . Complete the top of the record booklet. 2. Calculate the chronological age (CA) and put it in the box provided. 3. Administer the subtests in the order presented in the WAIS-R manual , except in rare circumstances . Do not change the wording on any subtest. Read the directions and material from the manual . Do not ad lib. 4. Start with the appropriate item on each subtest and follow discontinuance criteria. This means that you must know the scoring criteria before you administer the scale. 5. Write out all responses completely and legibly. Do not use unusual abbreviations. Record time accurately. 6. Question all ambiguous or unscorable responses, using the words suggested in the WAIS-R manual for ques­ tioning. You may need to question on Information, Vocabu­ lary, Comprehension, and Similarities. Whenever you ask a question, write (Q) after the questioned response. 7. Carefully score each protocol , recheck scoring, and transfer subtest scores to the front of the record booklet under Raw Score. If you have failed to question a response when you should have and the response is obviously not a 0 response, give the examinee the most appropriate score. 8. If a subtest was spoiled, write spoiled by the subtest total score and on the front cover where the raw scores and scaled scores appear. If for some reason a subtest was not administered, write NA in the margin in the record booklet and on the front cover. 9 . Raw scores are transformed into scaled scores through use of the table on the front of the record booklet (or Table 19

WAIS-R protocols (Ryan, Prifitera, & Powers , 1983) . Ad­ ditionally, both groups of raters converted scaled scores to IQs incorrectly, gave credit for individual items incor­ rectly, and added raw scores incorrectly. In another study additional scoring problems included failing to credit re­ sponses that appeared in the manual and assigning credit for symbol pairings on Digit Symbol incorrectly (Franklin et aI . , 1982 ) . Obviously, you should guard against making such errors . Judgment is important in scoring WAIS-R responses, especially when ambiguous responses are encountered on the Vocabulary, Comprehension, and Similarities subtests. Carefully study the scoring c riteria, scoring guidel ines, and scoring examples in the WAIS-R manual .

on page 90 of the WAIS-R manual). Be sure to use the correct row and column for each transformation. 1 0 . The Verbal score is based on the total of the scaled scores on the six Verbal Scale subtests. The Performance score is based on the total of the scaled scores on the five Performance Scale subtests. Add the Verbal score and the Performance score to get the Full Scale score. I I . If it should be necessary to prorate either the Verbal or Performance section, follow the explicit directions given on page 8 8 of the WAIS-R manual . 1 2 . The IQs are obtained from Table 20 (pages 92 through 109) of the WAIS-R manual. There is a Verbal section, a Performance section, and a Full Scale section. Be sure to use the correct section of the table for each of the three IQs. Record the IQs . Next, recheck all of your work. If any IQs were prorated, write PRO beside the appropriate IQs . 1 3 . If you want to compare the examinee to the reference group, make a profile of the examinee's scaled scores on the record booklet. A profile of age-corrected scaled scores can also be made. 1 4 . Look up the confidence intervals for the Full Scale IQ, Verbal Scale IQ, and Performance Scale IQ in Table C-2S in Appendix C. 1 5 . Look up the percentile rank and classification for each of the IQs in Tables BC-I and BC-2 on the inside back cover of this text. Source: Courtesy of M. L. Lewis.

Extrapolated IQs Because the lowest scaled score is 1 point, the sum of scaled scores cannot be lower than 6 for the Verbal Scale or 5 for the Performance Scale when proration procedures are used . Consequently, the minimu m sum of scaled scores is shown in the WAlS-R manual . When the sum of scaled scores is higher than the highest sum shown in Table 20 of the WAIS-R manual , report the IQ as "over 150" (Wechsler, 1981) . In some cases, however, you can extrapolate the IQ. Table C-30 in Appendix C extends the sum of scaled scores and IQs up to the maximum possible sum of scaled scores for the Verbal , Performance, and Full Scales separately for each of the nine age groups. As noted in Chapter 6 , extrapolated I Q s must b e used cautiously.

WAIS-R

R ECOR D FOR M

WECHSLER ADULT

NA ME

_ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __

A DDRESS

I NTELL I G ENCE SCALE­

__ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __

MA R I TA L SEX ____ AGE ___ RACE _____ S TA TUS

R EVISED

__ __ __ __ _

O CC UPATION

__ _ __ __ __ __ __ __ __ __ __ __ __ __ __

PLA CE OF TESTlNG

EDUCATION

__ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __

TESTED B Y

__ __ __ __ _ __ __ __ __ __ __ __ __ __ __ __ __

Year

TABLE OF SCALED SCORE EaU IVALENTS· �

0

� 't:I

B til ..

19

c

.2

c

a.

(f)

'"

E 0 'C

i5

;;;

'0,

�

c 0 'iii

c

.!'!

u

Q3 E £;

�

l'! a. E 0 0

'"

D :; '"

g

.c

'"

.� '"

PERFORMANCE TESTS c .2

·c !!1

C1l � � a. 2 E c> 0 ii: o

E i:i) .

---

C .. E '" "' 0> � c 2 ", .� t Q. «

:>< 0> c

g '�

en o

0

..

en 't:I

>-

:0

U E .9!.. � D ", 0 «

0 D

"&, E >O (f)

.-

18

-

28

70

-

32

-

-

-

51

-

93

29

27

69

-

31

28

-

-

-

41

9 1 -92

16

-

26

68

19

-

-

20

20

50

-

89-90

28

25

66-67

-

30

27

-

-

49

40

84-88

27

24

65

18

29

26

-

19

47-48

39

79-83

17

15

14

26

22-23

63-64

17

27-28

25

19

-

44-46

38

75-78

12

25

20-21

60-62

16

26

24

-

18

42-43

37

70-74

23-24

1 8- ; 9

55-59

15

25

23

18

17

38-41

35-36

66-69

10

22

17

52-54

13-14

23-24

22

17

15-16

35-37

34

62-65

19-21

1 5- 1 6

47-51

12

2 1 -22

20-21

16

14

3 1 - 34

32-33

57-61

1 7- 1 8

14

43-46

11

19-20

18-19

15

13

27-30

13 11

9

8

7

6

5

30 3 1 53 56

15-1 6

1 2- 1 3

37-42

10

17-18

16-17

14

1 1 -12

23-26

28-29

48-52

13-14

11

29-36

8-9

14-16

14-15

13

8-10

20-22

24-27

44-47

9-12

9-10

20-28

6-7

1 1 -13

1 1 -13

1 1 -1 2

5-7

1 4-19

21-23

37-43

6-8

8

14-19

5

8-10

7-10

8-10

3-4

8-13

16-20

30-36

3

5

7

1 1 -13

4

6-7

5-6

5-7

2

3-7

13-15

23-29

4

6

9-10

3

4-5

2-4

3-4

-

2

9-12

16-22

1

3

3-5

6-8

1 -2

2-3

1

2

1

6-8

8-15

0-2

0-5

0

0- 1

0

0-1

0

1

0-2

0

0-5

0-7

4

I

RAW SCORE

VERBAL TESTS

2

01 differences between

Date of Birth

SUMMARY

..

18

17

16

15

14

13 12

11

10 9

8

7

6

Raw Score

Informal ion

---

----

---

Digil Span

--

--

---

Vocabulary

--- --

----

Arilhmelic

---

--

---

Comprehension

----

---

----

S i m i larilies

---

---

----

Verbal Score

P,Clure Complelion

---

--

---

4

PIClure Arrangemenl

--

---

----

Block Design

----

---

---

2

Objecl Asse m b l y

----

--

----

D i g i t Symbol

--

--

---

5

3 1

scores on the lesls

Performance Score

FULL SCALE

Printed in U.S.A.

permiSSion All flghlS reserved

Figure 1 0- 1 .

---

PERFORMANCE TESTS

PERFORMANCE

1981, 1955, 1947 by The PsyChological CorPOrallon Standardllatlon Edition COOYlighl (I 1 976by The Psychological Corporation No cart ol \rus lorm may be cOPied by any process w :hout

Scaled Score

V E R BA L TESTS

VERBAL COpyright Cl

Day

Age

u en

19

Month

Date Tesled

Iii

·Cltntr,lans who wish todraw a profile maydo so by lOCating the sublect 5 raw scores on the table above and dtawlng a hne to connecl lhem See Chapler � 10 the Manual for a diSCUSSion ol lhe signIficance

__ _ __ __ __ __ __ __ __ __ __ __

___

Sum of Scaled Scores

IQ

---

--

---

---

--

---

9-99 1 8 29

Cover page of WAIS-R record booklet. Copyright ©

TX . All rights reserved.

233

1 98 1 , 1 955, 1 947 by The Psychological Corporation, San Antonio,

234

CHAPTER 1 0

W A I S- R S H ORT FORMS

Short forms of the WAIS-R have the same advantages and disadvantages as do short forms of the WISC-R. Chapter 6 thoroughly discusses the use of short forms; review this material as needed . Selecting the Short Forms Table C-34 in Appendix C provides the 10 best short form combinations of two, three, fou r, and five WAIS-R sub­ tests. For all practical purposes, the short forms of a given length are mutually interchangeable in terms of psycho­ metric properties. The choice of a short form should be based on the considerations discussed in Chapter 6 . After t h e short form has been selected, follow the pro­ cedures outlined in Chapter 6 (see Exhibit 6-3) to convert the composite scores to Deviation Quotients . Use Table C -36 in Appendix C to obtain the appropriate a and b constants. To obtain rjk, use the section of the WAIS-R correlation table (Table 15) that corresponds to the exam­ inee's age group. Satz-Mogel Abbreviated Procedure The Yudin short form procedure, which reduces the number of items within subtests, is known as the Satz­ Mogel short form when applied to the WAIS-R (Satz & Mogel , 1962 ) . Table C-35 in Appendix C shows the spe­ cific procedures for obtaining this WAIS-R short form . Validity coefficients and standard errors of measurement , based on the entire standardization sample, are . 94 and ± 5 . 1 for the Verbal I Q, . 89 and ± 6 . 8 for the Performance IQ, and . 95 and ± 4 . 7 for the Full Scale IQ, respectively (Silverstein, 1982d) . Although the Satz-Mogel approach is a useful screening device for estimating the Full Scale IQ, it should not be used to interpret individual subtest scores. Poor correlations have been reported between estimated subtest scaled scores and actual obtained score s . Addi­ tionally, estimated scaled scores have been found to exceed the range of ± 2 scaled score points from actual obtained scores in over 25 percent of 81 cases studied (Evans, 1985 ) . Vocabulary plus Block Design Short Form A two-subtest combination that is popular as a short form screening instrument is Vocabulary plus Block Design. These two subtests have moderate (Block Design) and high (Vocabulary) correlations with the Full Scale, have consis-

WECHSLER ADULT INTELLIGENCE SCALE-REVISED (WAIS-R)

tently high reliabilities, and are good measures of g. If this combi nation is chosen, Table C-37 in Appendix C can be used to convert the sum of age-corrected scaled scores directly into an estimated Full Scale IQ (Brooker & Cyr, 1986) . The rel iability of the composite is impressive (rxx = . 94 for the average of nine age groups) with an SE", in IQ points of 3 . 5 8 for 16- to 17-year-olds and 3 . 64 for the average of all nine age groups (Silverstein, 1985b) . The Vocabulary plus Block Design short form consistently overestimates the average Full Scale IQ of clinical samples by approximately 3 points , however ( Margolis, Taylor, & Greenlief, 1986; Roth , Hughes, Monkowski , & Crosson , 1984; Ryan , Larsen , & Prifitera , 1 98 3 ; Thompson , Howard, & Anderson, 1986) . Vocabulary, Block Design, and Information A useful three-subtest short form is Vocabulary, Block Design, and Information . Table C-38 in Appendix C can be used to convert the sum of the three age-corrected subtest scaled scores into an estimated Full Scale I Q . Vocabulary, Block Design, Arithmetic, and Similarities A useful four-subtest short form is Vocabulary, Block Design , Arithmetic , and Similaritie s . Table C-39 in Ap­ pendix C can be used to convert the sum of the four age­ corrected subtest scaled scores into an estimated Full Scale IQ. Information, Arithmetic, Picture Completion, and Block Design Short Form Another potentially useful short form combination is In­ formation, Arithmetic , Picture Completion , and Block Design (Reynolds , Willson, & Clark , 1983) . Table C-40 in Appendix C can be used to convert the sum of the four age­ corrected subtest scaled scores into an estimated Full Scale IQ. The reliability of the composite is adequate (rxx = . 88 for the average of nine age groups) , and estimated stability over time is excellent ( .95 for 71 standardization subj ects between 25 and 34 years of age; . 96 for 48 standardization subjects between 45 and 54 years of age) . The SE", in IQ points is 5 . 73 for 16- to 17 -year-olds, and 5 . 26 for the average of the nine age groups in the standardization sam­ ple . This four-subtest short form is a valid predictor of the Full Scale IQ in normal subjects (Reynolds et aI . , 1983) and neurologically impaired patients (Ryan, 1985).

235

W AIS-R SUBTESTS

W A I S-R SUBTESTS

Information The I nformation subtest contains 29 questions which sam­ ple a broad range of general knowledge, including literary, historical , and geographical facts and dates . All exam­ inees start with item 5 . All items are scored 1 or ° (pass­ fail ) , and the subtest is discontinued after five consecutive failures. The questions usually can be answered with a simply stated fact. The examinee is not required to find relationships between facts in order to receive credit . Rationale. The rationale presented for the WISC-R In­ formation subtest applies to the WAIS-R Information sub­ test (see Chapter 7 ) . Factor analytic findings. I nformation is the second­ best measure of g in the scale (67 percent of its variance may be attributed to g). It has adequate subtest specificity in every age range with the exception of 45 to 54 years, where its specificity is inadequate . Information contributes substantially to the Verbal Comprehension factor (Mdn loading = . 76) . Reliability and correlational highlights. I nformation is a reliable subtest (rXY = . 89 ) . It correlates more highly with Vocabulary (r = . 81) than with any other subtest. It correlates moderately with the Full Scale IQ (r = . 76) , the Verbal Scale IQ (r = . 79) , and the Performance Scale IQ (r = . 62 ) . Administrative and interpretive considerations. The administrative and interpretive considerations presented for the WISC-R Information subtest are also relevant for the WAIS-R (see Chapter 7) . Digit Span The Digit Span subtest has two parts : Digits Forward , which contains series ranging in length from three to nine digits, and Digits Backward, which contains series rang­ ing in length from two to eight digits . The examinee listens to a sequence of digits given orally by the examiner and then repeats the digits . There are two sets of digits of each length. Digits Forward is administered first, followed by Digits Backward . On the WAIS-R Digit Span is a regular subtest, whereas on the WISC-R it is supplementary. All series are scored 2 , 1 , or 0. On both parts of the

subtest, testing i s discontinued after failure o n both trials of any series. Scaled scores and age-corrected scores are not provided separately for Digits Forward and Digits Backward. Rationale. The rationale presented for the WISC-R Digit Span subtest applies to the WAIS-R Digit Span sub­ test (see Chapter 7). Factor analytic findings. The Digit Span subtest is a fair measure of g (38 percent of its variance may be attributed to g) . It has ample subtest specificity across the entire age range to permit specific i nterpretation of its functions. Digit Span has a high loading on the Freedom from Distractibility factor (Mdn loading = . 56) . Reliability and correlational highlights. Digit Span is a reliable subtest (ru = . 83 ) . It correlates more highly with Arithmetic ( r = . 56) and Vocabulary (r = . 52 ) than with the remaining subtests. It has a relatively low correla­ tion with the Full Scale IQ (r = .58), the Verbal Scale IQ (r = . 57), and the Performance Scale IQ (r = . 50) . Administrative and interpretive considerations. The administrati ve and interpretive considerations presented for the WISC-R Digit Span subtest are also relevant for the WAIS-R (see Chapter 7).

Vocabulary The Vocabulary subtest contains 35 words arranged in order of increasing difficulty. Each word is presented or­ ally and in writing, and the examinee is asked to explain aloud its meaning . All examinees start w ith the fourth word, except those who seem to have poor verbal ability. Each word is scored 2 , I , or 0, and the subtest is discon­ tinued after five consecutive failures. Rationale. The rationale presented for the WISC-R Vo­ cabulary subtest applies to the WAIS-R Vocabulary subtest (see Chapter 7) . Factor analytic findings. The Vocabulary subtest is the best measure of g in the scale (76 percent of its variance may be attributed to g). The subtest has an adequate amount of subtest specificity across the entire age range and contributes substantially to the Verbal Comprehension factor (Mdn loading = . 83) .

236

CHAPTER 10

Reliability and correlational highlights. Vocabulary is the most reliable subtest in the scale (rxx = . 96) . It corre­ lates more highly with Information ( r = . 81 ) than with any other subtest. It has high correlations w ith the Full Scale IQ (r = . 81 ) and the Verbal Scale IQ (r = . 85) and a moderate correlation with the Performance Scale IQ (r = .65) . Administrative and interpretive considerations. The administrative and interpretive considerations presented for the WISC-R Vocabulary subtest generally apply to the WAIS-R (see Chapter 7) . On the WAlS-R Vocabulary subtest, however, the examinee looks at a word list as the examiner pronounces each word. Also, when the exam­ inee gives a 0- or I-point definition on the first word, no help is provided by the examiner. Study carefully the "Sample Responses" section of Ap­ pendix A of the WAIS-R manual so that you will know which responses require further inquiry, indicated by "(Q) . " The examples indicate that many 0- and I-point responses should be queried. When a 2-point response is accompanied by a "(Q)," it is the entire response including the elaboration that is worth 2 points . Arithmetic The Arithmetic subtest contains 14 problems ; 13 are given orally and the other one involves blocks. All examinees start with item 3. All problems are timed , with items 1 through 4 having a time limit of 15 seconds ; items 5 through 9, 30 seconds; items 10 through 13 , 60 seconds; and item 14 , 120 seconds. All items are scored l or 0, with up to 2 additional time-bonus points possible on items 10 through 14 . Rationale. The rationale presented for the WISC-R Arithmetic subtest applies to the WAIS-R Arithmetic sub­ test (see Chapter 7). Factor analytic findings. Arithmetic is a good measure of g (58 percent of its variance may be attributed to g). The subtest has ample subtest specificity to permit interpreta­ tion of its functions at ages 20 to 34 years and 45 to 74 years . It has only adequate specificity at ages 18 to 19 years and 35 to 44 years, and it has inadequate specificity at ages 16 and 17 years. Arithmetic has a high loading on the Freedom from Distractibility factor (Mdn loading = .53) and a moderate loading on the Verbal Comprehension factor (Mdn loading = .45 ) .

WECHSLER ADULT INTELLIGENCE SCALE-REVISED (WAIS-R)

Reliability and correlational highlights. Arithmetic is a reliable subtest (rxx . 84) . It correlates best with Vocab­ ulary (r .63) and Information (r = .61). It correlates moderately with the Full Scale IQ (r = .72), the Verbal Scale IQ (r = . 70 ) , and the Performance Scale I Q ( r = . 62 ) . =

=

Administrative and interpretive considerations. The administrative and interpretive considerations discussed for the WISC-R Arithmetic subtest generally apply to the WAIS-R (see Chapter 7). A booklet is not used to present any of the WAlS-R items to the examinee, however. Comprehension The Comprehension subtest contains 16 questions covering a wide range of situations and proverbs. Questions deal with such issues as government operations and laws, health standards, and social mores. All examinees begin the subtest with item 1, and all items are scored 2 , 1, or O . The subtest is discontinued after four consecutive failures. Rationale. The rationale presented for the WISC-R Comprehension subtest applies to the WAIS-R Com­ prehension subtest (see Chapter 7 ) . Factor analytic findings. The Comprehension subtest is a good measure of g (61 percent of its variance may be attributed to g). Subtest specificity varies with age . There is ample specificity at ages 45 to 54, adequate specificity at ages 25 to 44 and 65 to 74, and inadequate specificity at ages 16 to 24 and 55 to 64. Consequently, specific inter­ pretation of the subtest's functions is appropriate only at some ages. The Comprehension subtest has a high loading on the Verbal Comprehension factor (Mdn loading = . 69) . Reliability and correlational highlights. Comprehen­ sion is a reliable subtest ( rxx = . 84). It correlates more highly with Vocabulary (r = .74) , Information (r = .68), and Similarities ( r = .68) than with any other subtests . It correlates moderately with the Full Scale IQ (r = . 74) , the Verbal Scale IQ ( r = . 76) , and the Performance Scale IQ ( r = .61). Administrative and interpretive considerations. Ad­ ministrative and interpretive considerations presented for the WISC-R Comprehension subtest are also relevant for the WAlS-R . Study carefully the "Sample Responses" sec­ tion of Appendix B of the WAIS-R manual so that you will

WAIS-R SUBTESTS

know which responses require further inquiry, i ndicated by "(Q ) . " The examples indicate that some 0- and I-point responses should be queried .

Similarities The Similarities subtest contains 14 pairs of words; the examinee is asked to explain the similarity between the two words in each pair. All examinees start with the first item. All items are 8cored 2 , 1 , or 0, depending on the conceptual level of the response . The subtest is discontinued after four consecutive failures.

Rationale. The rationale described for the WISC-R Similarities subtest applies to the WAIS-R Similarities subtest (see Chapter 7 ) . Factor analytic findings. Similarities is the third-best measure of g in the scale (62 percent of its variance may be attributed to g) . The subtest has ample subtest specificity at ages 25 to 34, adequate specificity at ages 18 to 19 and 35 to 74 , and inadequate specificity at ages 16 to 17 and 20 to 24. Specific interpretation of the subtest's functions is inap­ propriate at ages 16 to 17 and 20 to 24 . Similarities contrib­ utes substantially to the Verbal Comprehension factor (Mdn loading = . 67 ) . Reliability and correlational highlights. Similarities is a reliable subtest ( rxx = . 84) . It correlates best with Vocab­ ulary (r = . 72) and Comprehension (r = . 68). It corre­ lates moderately with the Full Scale IQ (r = .75), the Verbal Scale IQ (r = . 74) , and the Performance Scale IQ (r = .64). Administrative and interpretive considerations. Most of the administrative and interpretive considerations pre­ sented for the WISC-R Similarities subtest apply to the WAlS-R (see Chapter 7 ) . The major difference is that on the WAIS-R all items are scored 2, 1, or 0, whereas on the WISC-R a 2 may be given only on items 5 to 1 7 . Consider­ able skill is required to score Similarities responses . Appendix C in the WAIS-R manual merits careful study. First, the general scoring principles, which give the ra­ tionale for scores of 2 , 1 , and 0, should be thoroughly mastered. Second , the sample responses section should be studied carefully so that responses that should be queried, indicated by "(Q) , " can be recognized readily.

237

Picture Completion The Picture Completion subtest consists of 20 drawings of common objects (such as a door, a boat, and a leaf), each of which lacks a single essential element. The examinee's task is to name or point to the missing portion of the picture. There is a 20-second time limit for each picture. All examinees start with the first item. Each item is scored l or 0 (pass-fail), and the subtest is discontinued after five consecutive failures.

Rationale. The ratibnale described for the WISC-R Pic­ ture Completion subtest applies to the WAIS-R Picture Completion subtest (see Chapter 7) . Factor analytic findings.

The Picture Completion sub­ test is a fajr measure of g (50 percent of its variance may be attributed to g) . The subtest has ample specificity at all ages to permit specific interpretation of its functions . Pic­ ture Completion contributes substantially to the Perceptual Organization factor at all the age levels except 18 to 19 (Mdn loading = . 5 1 ) .

Reliability and correlational highlights.

Picture Com­ pletion is a reliable subtest (rxx = . 81) . It correlates more highly with Vocabulary (r = . 55) than with any other sub­ test. It correlates moderately with the Full Scale IQ (r = .67), the Performance Scale IQ (r = .65), and the Verbal Scale IQ (r = .61).

Administrative and interpretive considerations. The administrative and interpretive considerations presented for the WISC-R Picture Completion subtest generally ap­ ply to the WAIS-R (see Chapter 7). However, whereas on the WISC-R three different guiding statements may be given by the examiner when the examinee gives an incor­ rect response, on the WAIS-R only one type of guiding statement is permitted (asking for the most important part missing when an unessential missing part is given) .

Picture Arrangement The Picture Arrangement subtest requires the examinee to place a series of pictures in a logical sequence . The 10 series are similar to short comic strips. The individual pictures are placed in a specified disarranged order, and the examinee is asked to rearrange the pictures in the "right" order to tell a story. One set of cards is presented at

238

CHAPTER 10

a time . There is little motor action required, as the pictures must simply be shifted to make a meaningful story. All examinees start with item 1 . The first item has two trials and is scored 2 , 1 , or O . For the remaining items (2 to 10) , 2 points are given for each correct arrangement com­ pleted within the time limit. On items 2, 5 , 8, and 9, 1 point is given for an acceptable variation of the correct arrange­ merit. Items 1 to 4 have a 60-second time limit; items 5 to 8 , a 90-second time limit; and items 9 and 1 0 , a 120-second time limit.

Rationale. The rationale presented for the WISC-R Pic­ ture Arrangement subtest applies to the WAIS-R Picture Arrangement subtest (see Chapter 7 ) . Factor analytic findings. The Picture Arrangement sub test is a fair measure of g (45 percent of its variance may be attributed to g). The subtest has ample specificity at ages 16 to 19, 25 to 64, and 70 to 74 to warrant specific interpreta­ tion of its functions at these ages. It has inadequate specif­ icity at ages 20 to 24 and 65 to 69 . Picture Arrangement has modest loadings on the Verbal Comprehension (Mdn loading = . 40) , Perceptual Organization (Mdn load­ ing = . 3 3 ) , and Freedom from Distractibility (Mdn load­ ing = . 22 ) factors . Picture Arrangement cannot be uniquely allocated to any one of the three factors, however. Reliability and correlational highlights.

Picture Ar­ rangement is relatively reliable (rxx = . 74). It correlates more highly with Picture Completion (r = . 51) and Vocab­ ulary (r = . 51) than with the other subtests. It correlates moderately with the Full Scale IQ (r = . 61) and to a lesser degree with the Performance Scale IQ (r = . 56) and the Verbal Scale IQ (r = . 57 ) .

Administrative and interpretive considerations.

The administrative and interpretive considerations presented for the WISC-R Picture Arrangement subtest generally apply to the WAIS-R (see Chapter 7 ) . However, the bonus points awarded on the WISC-R subtest for speed are not awarded on the WAIS-R subtest.

Block Design The Block Design subtest contains nine items. The exam­ inee is shown two-dimensional, red-and-white pictures of abstract designs and then must assemble a design that is identical to each picture, using three-dimensional red and white plastic blocks . All examinees start with item 1 . On this item only, the examinee is required to reproduce a design from a model constructed by the examiner.

WECHSLER ADULT INTELLIGENCE SCALE-REVISED (WAIS-R)

The patterns are arranged in order of increasing diffi­ CUlty. Four blocks are used for the first five designs, and nine for the last four designs. All items are timed . The first five items have a time l imit of 60 seconds; the last four items have a time limit of 120 seconds. On items I and 2, the examinee is given 2 points for successful completion on the first trial or 1 point for successful completion on the second trial . On items 3 to 9 , 4 points are given for a correct completion, with u p t o 3 (items 3 and 4) or 4 (items 5 through 9) additional time­ bonus points awarded for quick execution. The subtest is discontinued after three consecutive failures.

Rationale. The rationale presented for the WISC-R Block Design subtest applies to the WAIS-R Block Design subtest (see Chapter 7 ) . Factor analytic findings.

The Block Design subtest i s the best measure o f g among the Performance Scale sub­ tests (53 percent of its variance may be attributed to g) . The subtest has either ample or adequate specificity across the entire age range to permit specific interpretation of its functions. The Block Design subtest contributes substan­ tially to the Perceptual Organization factor (Mdn load­ ing = . 74).

Reliability and correlational highlights.

Block Design is a reliable subtest (rxx = . 87 ) . It correlates more highly with Object Assembly (r = . 63) than with any other sub­ test. It correlates moderately with the Full Scale IQ (r = . 68), the Performance Scale IQ (r = . 70) , and the Verbal Scale IQ (r = .61).

Administrative and interpretive considerations.

The administrative and interpretive considerations described for the WISC-R Block Design subtest generally apply to the WAIS-R (see Chapter 7 ) . Timing and bonus points differ on the two subtests.

Object Assembly In the Object Assembly subtest, the examinee must put j igsaw pieces together to form common objects : a manikin (6 pieces), a profile of a face (7 pieces), a hand (7 pieces) , and an elephant (6 pieces) . The items are presented one at a time, with the pieces presented in a specified disarranged pattern . Examinees are administered all four items . All items are timed . The time limit is 120 seconds for the first two items and 180 seconds fOT the last two items. Up to 3 time-bonus points may 'be awarded on each item for quick execution. The manikin has a maximum score of 8 ;

239

INTERPRETING THE WAIS-R

the face , 12; the hand, 10; and the elephant , 1 1 . Pnints are also awarded for partially correct performances. Rationale. The rationale presented for the WISC-R Ob­ ject Assembly subtest applies to the WAlS-R Object As­ sembly subtest (see Chapter 7 ) . Factor analytic findings. The Object Assembly subtest is a fair measure of g (40 percent of its variance may be attributed to g). The subtest has ample specificity to permit specific interpretation of its functions only at ages 18 to 19 . At all other ages , subtest specificity is inadequate . The Object Assembly subtest has a high loading on the Percep­ tual Organization factor (Mdn loading = . 7 1 ) , which makes it inte rpretable as a measure of perceptual organization . Reliability and correlational highlights. Object As­ sembly is the least reliable of the WAIS-R subtests (rxx = .68). It correlates more highly with the Block De­ sign subtest (r = .63) than with any of the other subtests. Its correlation is relatively low with the Full Scale IQ (r = . 5 7 ) , moderate w ith the Performance Scale IQ (r = .62), and low with the Verbal Scale IQ (r = .49) . Administrative and interpretive considerations. The administrative and interpretive considerations described for the WISC-R Object Assembly subtest generally apply to the WAIS-R. Digit Symbol The Digit Symbol subtest is similar to Coding B on the WISC-R. The subtest requires the copying of symbols that are paired with numbers . The sample (or key) consists of nine boxes, each of which contains one of the numbers 1 through 9 and a symbol. Each test box contains a number in the upper portion and an empty space in the lower portion . In the empty space, the examinee must draw the symbol that was paired with the number in the key. There are seven practice boxes, followed by 93 boxes in the subtest proper. Both examiner and examinee should use pencils without erasers. It is important that the examinee have a smooth drawing surface . One point is allotted for each correct item . The time l imit is 90 seconds ; no time-bonus points are awarded. Rationale. The rationale for the WISC-R Coding sub­ test applies to the WAIS-R Digit Symbol subtest (see Chap­ ter 7).

Factor analytic findings. The Digit Symbol subtest is a fair measure of g (36 percent of its variance may be attributed to g). It has either ample or adequate specificity across the entire age range to permit specific interpretation of its functions . Digit Symbol has modest loadings on the Verbal Comprehension (Mdn loading = . 38), Perceptual Organization (Mdn loading = .41), and Freedom from Distractibility (Mdn loading = . 26) factors . Digit Symbol cannot be uniquely allocated to any one of the three fac­ tors , however. Reliability and correlational highlights. Digit Symbol is a reliable subtest (ru = . 82 ) . It correlates more highly w ith Block Design (r = .47) and Vocabulary (r = . 47) than with any other subtests. It has a relatively low correla­ tion with the Full Scale IQ (r = .57), the Performance Scale IQ (r = .52), and the Verbal Scale IQ (r = . 54) . Administrative and interpretive considerations. The administrative and i nterpreti ve considerations for the WISC-R Coding B subtest generally apply to the WAIS-R Digit Symbol subtest (see Chapter 7) .

I NTERPRETI NG T H E WAIS·R

Almost all of the material in Chapter 8 on interpreting the WISC-R pertains to the WAIS-R. For example, the suc­ cessive level approach to test interpretation , profile analy­ sis, Verbal-Performance Scale comparisons, factor score comparisons, and subtest comparisons are essentially the same for bOlh tests. The estimated percentile ranks for subtest scaled scores are shown in Table C-41 in Appendix C . This table also shows suggested qualitative descriptions associated with scaled score s . Table BC-2 on the inside back cover shows the classifications associated w ith WAIS-R IQs. In Appendix C, Table C-13 summarizes the functions associated with each subtest (substituting Digit Symbol for Coding) , Table C-42 presents information about the three scales and factor scores, and Table C-43 g ives suggested remediation activities for combinations of Wechsler subtests. Profile Analysis As noted above , approaches to profile analysis on the WAIS-R are basically the same as those on the WISC-R (see Chapter 8). The only difference is that different tables in Appendix C must be used. 1 . Comparing Verbal and Performance Scale IQs. Table C -26 in Appendix C provides the critical values for com-

240

CHAPTER

paring the Verbal and Performance IQs: U at the . 05 level and 16 at the .01 level at ages 16 to 1 7 , and 10 at the . 05 level and 13 at the .01 level for the average of the nine age groups. Verbal-Performance IQ differences are essentially similar in male and female groups in the standardization sample (Matarazzo, Bornstein, McDermott , & Noonan, 1986). This finding means that similar interpretations can be given to the Verbal-Performance profiles of both males and females . Probabilities associated w ith various Verbal­ Performance Scale differences are shown in Table C-32 of Appendix C . 2 . Comparing each Verbal subtest age-corrected scaled

score with the mean Verbal age-corrected scaled score. Table C-27 in Appendix C provides the critical values for comparing Verbal subtests with the mean of the Verbal subtests (all age-corrected scores) . They range from 1 . 8 to 3 . 0 at the . 05 level and from 2 . 1 to 3 . 5 at the . 01 level for the six Verbal subtests. 3 . Comparing each Performance subtest age-corrected

scaled score with the mean Performance age-corrected scaled score. Table C-27 in Appendix C provides the critical values for comparing Performance subtests with the mean of the Performance subtests (all age-corrected scores) . They range from 2 . 5 to 3 . 0 at the . 05 level and from 3 . 0 to 4 . 2 at the .01 level for the five Performance subtests . 4 . Comparing each subtest scaled score with the mean of the 11 age-corrected subtest scaled scores. Table C-27 in Appendix C provides the critical values for comparing subtests w ith the mean of all the subtests (all age-corrected scores) . They range from I . 9 to 3 . 5 at the . 05 level and from 2 . 2 to 4 . 8 at the .01 level for the 11 subtests . 5 . Comparing sets of individual age-corrected subtest scaled scores. Table C-26 in Appendix C provides the critical values for comparing sets of subtest scores (all age­ corrected scores) . They range between 2 and 5 at the .05 level and between 3 and 6 at the .01 level . The values in Table C-26 in Appendix C for subtest comparisons are overly l iberal (that is, they lead to too many significant differences) when more than one comparison is made . They are most accurate when a priori planned comparisons are made , such as Comprehension versus Picture Arrange­ ment or Digit Span versus Arithmetic . (See Chapter 8 for additional information that can guide interpretations of subtest comparisons.) Silverstein (l982c) advises that when you make multiple subtest comparisons you first determine the di fference between the highest and lowest age-corrected subtest scores. If this difference is 6 or more age-corrected points,

10

WECHSLER ADULT INTELLIGENCE SCALE-REVISED (WAIS-R)

it is significant at least at the . 05 level . Differences of 6 or more points between subtests can then be interpreted. If the difference between the highest and lowest subtest score is less than 6 points, multiple comparisons between indi­ vidual subtests should not be made . Other Approaches to Profile Analysis The supplementary approaches to WAIS-R profil e analysis described in this section are similar to those discussed for the WISC-R (see Chapter 8). Base rate differences between each subtest score and an average subtest score in the WAIS-R standardization sample. Table C-31 in Appendix C allows you to see how frequently a particular difference between a subtest score and an average WAIS-R Verbal , Performance, or overall score occurred in the standardization sample. D i fferences of approximately 3 to 4 points between each subtest score and the respective average Verbal or Performance Scale score were obtained by 5 percent of the standardization sample. This table should be used only to evaluate differ­ ences that have first been found to be reliable. (See num­ bers 2, 3, and 4 in the preceding section on profile analysis . ) Base rate Verbal-Performance differences (or proba­ bility-of-occurrence approach) . Table C-33 in Appen­ dix C presents the empirically observed percentage of individuals in the standardization sample who obtained a given discrepancy between Ve rbal and Per formance Scales. The percentages are shown for five IQ groups and for the total group. For example, 3 7 . 8 percent of the individuals in the standardization sample had a lO-point difference between the two IQs. Comparisons of factor scores. To compare factor scores, you must first convert them to Deviation IQs (see Table C-28 in Appendix C for the conversion table) . Table C-26 in Appendix C provides the differences required for significance at the .05 and .01 levels between the three factors (Verbal Comprehension , Perceptual Organization, and Freedom from Distractibility) for ages 16 to 17 and for the average of the nine age groups. Finally, Table C-29 i n Appendix C provides the difference scores needed to deter­ mine whether any of the individual age-corrected subtest scaled scores that comprise the three factors are signifi­ cantly different (at the . 05 and .01 levels) from the average of all the age-corrected scores that comprise each factor for ages 16 to 17 and for the average of the nine age groups .

24 1

ASSETS OF THE WAIS-R

As in the WISC-R, factor scores should be compared with other factor scores and not with the Verbal and Perfor­ mance Scale IQs, to guard against the use of overlapping subtest s . Use o f Norm Tables The norm tables allow each examinee to be compared with his or her age group. The transition point between two normative age groups may affect scores, however, and lead to interpreti ve problems in some test-retest situations. For example, whereas an examinee aged 19 years, 11 months , 29 days who earns a WAIS-R scaled score total of 100 points receives a Full Scale IQ of99, the same examinee at 20 years, 0 months receives a Full Scale IQ of 92 for the identical scaled score total . If the same examinee is tested at 19 and 20 years of age , an examiner might be inclined to interpret the drop of 7 points as suggesting a lowering in functioning, when in fact the change is an artifact of the test norms. Such considerations are important to keep in mind when one is evaluating practice effects or conducting psychoeducational reevaluations. ASS ETS O F THE WAIS-R

The WAIS-R is a well-standardized test, with good relia­ bility and validity. It divides the II subtests i nto two sections and provides three IQs - Verbal , Performance , and Full Scale . This procedure is helpful in cl inical and psycho­ educational work and in the assessment of brain-behav ior relationships . The fact that all examinees take a compara­ ble battery of subtests is a valuable featu re of the test . Parts of the test also can be administered to examinees limited by sensory impairments (for example , the Verbal Scale to b lind i ndividuals or those with motor handicaps; the Per­ formance Scale to deaf individuals) . 1 . Good validity. Concurrent validity data are uni­ formly positive . WAIS-R [Qs correlate significantly with scores from other intelligence tests, measures of academic achievement, and years of formal education . Factor ana­ lytic studies indicate that the Full Scale IQ is an excellent measure of g, and in most situations the division of the subtests into Verbal and Performance sections is appropri­ ate . Inspection of the age norms for the test indicates that if the Performance Scale subtests are considered as measures of fluid intelligence and the Verbal Scale subtests as mea­ sures of crystallized intelligence, then the WAIS-R norms are consistent with predictions based on the Cattell-Horn theory of intelligence . 2 . High ,eliabilities. The WAIS-R Full Scale IQ has

excellent reliability (average 'x.x = . 97 ) . The SE",'s of the IQs on the three scales are consistently less than 5 points. The WAIS-R manual provides reliability data , SEI/l's , and intercorrelations of subtest scores for the nine age groups as well as for the average of the nine age groups , permitting evaluation of the test's properties throughout the entire age range covered by the scale. Confidence intervals can be established for IQs for each of the nine age groups (as well as for the average of the nine groups) ; thus estimates can be made that are applicable to the examinee's specific chrono­ logical age group . 3 . Excel/em standardization. The standardization pro­ cedures were excellent, sampling four geographic regions, both sexes, white and nonwhite populations, urban and rural residents, and the entire range of socioeconomic classes. 4. Good administration procedures. The prescribed procedures for administering the WAIS-R are excellent. Examiners actively probe responses in order to evaluate the breadth of the examinee's knowledge and determine whether the examinee really knows the answer. On items that require two reasons for maximum credit , the examinee is asked for another reason when only one reason is give n . These procedures prevent examinees from being penalized for not fully understanding the demands of the question . The emphasis on probing questions and queries is ex­ tremely desirahle. 5. Good manual. The WAIS-R manual is easy to use; it provides clear directions and tables. The examiner's in­ structions are pri med in a different color to facil itate read­ ing of the directions. Helpful suggestions are provided about abbreviations to use in recording responses, such as "Q" for question, DK for don't know, "Inc . " for in­ complete , and " NR" for no response . The test materials are interesting to adolescents and adults . 6. Helpful scoring criteria. The criteria for scoring re­ plies have been carefully prepared. The Vocabulary, Com­ prehension, and Si milarities scoring guidel ines, for exam­ ple , give the rationale for the use of 2, I, and 0 scores and are accompanied by a number of examples that demon­ strate the application of the scoring principles . Many typ­ ical responses are scored, and those deemed to need fur­ ther inquiry are indicated by a "(Q) . " Studying the scoring sections will aid you not only in scoring responses but also in administering the test. "

"

LIM ITATIONS O F THE WAIS-R

Although the WAIS-R is an excellent instrument, some problems exist with the test and the manual .

242

CHAPTER 1 0

1 . Limited floor and ceiling. The range of Full Scale IQs (45 to 150) is insufficient both for moderately to se­ verely retarded persons and for extremely gifted persons . The test is designed s o that examinees receive I scaled score point on each subtest, even if they give no correct answers . Thus the WAIS-R may not be an appropriate measure for the exami nee who earns 0 or 1 raw score point on most subtests. 2. Nonuniformity ofsubtest scores. The range of scaled scores on all subtests is not uniform . At the reference age group (ages 20 to 34) used to obtain scaled scores to calculate IQs, a scaled score of 19 can be earned on only five subtests: Digit Span, Vocabulary, Comprehension , Block Design, and Digit Symbol . The maximum scaled score is 18 on Information , Similarities, and Object As­ sembly, and 17 on Arithmetic, Picture Completion , and Picture Arrangement . For purposes of profile analysis, the scaled score range is more uniform at ages 16 to 1 7 . At this age group, a scaled score range of 19 can be obtained on nine of the 11 subtests. On the two remaining subtests ­ Picture Completion and Picture Arrangement - 18 is the maximum scaled score. Nonuniformity is also a problem w ith the age-corrected subtest scores. The restriction of scores occurs for each of the nine age groups . For example, at ages 45 to 54, scores of 19 can be earned on only six subtests: Digit Span , Vocabulary, Comprehension , Block Design , Object As­ sembly, and Digit Symbol . At ages 65 to 69 , scores of 1 can be obtained on only six subtests: Information, Digit Span , Vocabulary, Comprehension, Object Assembly, and Digit Symbol. The lack of uniformity of available scaled scores and age-corrected scores makes profile analysis more diffi­ cult to apply, particularly to the profiles of retarded and gifted examinees. It may be misleading to apply profile techniques uniformly to all subtests for every examinee , because the same number of scaled score points, or age­ corrected score points, cannot be obtained on all subtests. 3. Difficulty of scoring responses. When responses on the Vocabulary, Comprehension, and Similarities subtests differ from those that appear in the WAIS-R manual , they may be difficult to score . Such difficulties may lead to halo effects in scoring and may contribute to other types of examiner bias . Although mastering the scoring criteria is important for all subtests, it is particularly crucial for these three . Close attention must also be given to the mechanics of scoring . 4. Lack of normative data for ra w scores. The WAIS-R manual fails to give means, standard deviations, and fre­ quency distributions for the raw scores. 5 . Failure ro describe the procedure for establishing

WECHSLER ADULT INTELLIGENCE SCALE-REVISED (WAIS-R)

discontinuance criteria. The WAlS-R manual states that discontinuance criteria were established using data from the standardization sample . The procedure is not de­ scribed further, however.

TEST Y O U R S K I L L

The WISC-R Test-Your-Skill Exercises in Chapter 8 also apply to the WAlS-R . You are encouraged to review these exercise s .

S U M MARY I . The revised edition of the Wechsler Adult Intelligence Scale, the WAIS-R, is similar to its 1955 predecessor, with 80 percent of the original items retained . The WAlS-R is applicable to individuals from ages 16-0 to 74-11. Standardization ofthe scale was excellent, including both white and nonwhite individuals. 2. Reliabilities for the IQs associated with the Verbal, Perfor­ mance, and Full Scales are extremely high for the standardization sample (average rn 0f . 97, .93, and .97, respectively), with SEm's for the Full Scale of about 2 . 5 IQ points. 3. Practice effects (after a two- to seven-week test-retest interval) for standardization subjects were about 3 . 3 points on the Verbal Scale, 8 . 4 points on the Performance Scale, and 6 . 2 points on the Full Scale. 4. Concurrent and construct validity studies of the WAIS-R have been uniformly positive. The scale correlates significantly with education and with other ability measures . Moreover, the distribution of the age norms is consistent with theory-based expectations, and factor analysis indicates that all II subtests measure g with a moderate to high degree of success. 5. The WAIS-R tends to provide lower IQs than the WAIS and slightly higher IQs than the WISC-R and Stanford-Binet: Fourth Edition, especially for low-functioning individuals . 6 . Higher intercorrelations are observed among the Verbal subtests than among the Performance subtests . Within their re­ spective scales, Vocabulary and Block Design show the highest correlations with the Full Scale IQ ( . 82 and .67, respectively, at ages 16 to 17) . 7. A study of WAIS-R IQs in relation to various stratification variables showed that mean differences between males and females were less than 3 points; the IQs of white subjects were higher than those of black subjects by about 15 points (101.4 vs. 86.8); the mean Full Scale IQ of persons in the highest occupa­ tional group was 22 points higher, on average, than that of persons in the lowest occupational group (1I1.0 vs. 89.0); urban­ rural differences and geographic region differences were small; and Full Scale IQs of persons with 16 or more years of education

243

SUMMARY

were , on average, 33 points higher than those of persons with less than 8 years of education . 8 . Factor analyses of the WAIS-R usually yield three fac­ tors - Verbal Comprehension, Perceptual Organization, and Freedom from Distractibility - but some also report two-factor solutions. 9. The subtests that provide the best measure of g are Vocab­ ulary, Information, Similarities, Comprehension . Arithmetic, and Block Design . The remaining subtests are fair mea ures of g . 1 0 . Only some of the subtests have ample subtest specificity at ages 16 to 17 and at other ages throughout the scale. At ages 16 to 17, seven of the subtests have ample or adequate subtest specif­ icity (Information, Digit Span, Vocabulary, Picture Completion, Picture Arrangement, Block Design, Digit Symbol) . and four ( Arithmetic , Comprehension, Similarities, Object A sembly) have inadequate subtest specificity. 1 1 . Somewhat purer factor scores can be obtained by use of Table C-28 in Appendix C, which converts age-corrected scores into Deviation Quotients as follows: (a) Information, Vocabulary, Comprehension , and Similarities into a Verbal Comprehension Deviation Quotient; (b) Block Design, Object Assembly, and Picture Completion (at every age level except 18 to 19, where the latter subtest should be omitted) into a Perceptual Organization Deviation Quotient; and (c) Digit Span and Arithmetic into a Freedom from Distractibility Deviation Quotient. 1 2 . Although the administrative procedures for the WAIS-R are generally similar to those for the WISC-R, there arc some differences. Be sure to use the appropriate procedures for each scale. 1 3 . Table C-30 in Appendix C presents extrapolated lQs for scaled scores that are above those shown in the WAIS-R manual . 14. Numerous WAIS-R short forms have been developed . Tables C-34 through C-40 in Appendix C provide infornl�lioll lO assist you in the selection and use of WAIS-R short forms . 1 5 . The rationale and administrative and interpreti ve consid­ erations for each WAIS-R subtest are similar to those for the respective WISC-R subtests . Consult Chapter 7 for a discussion of these issues. 1 6 . Information is a good measure of g and contributes to the Verbal Comprehension factor. Subtest specificity is adequate except at ages 45 to 54. I nformation is a reliable subtest and correlates highly with Vocabulary. 1 7 . Digit Span is a fair measure of g , and contribute to the Freedom from Distractibility factor. Subtest specificity is ample at all ages. Digit Span is a reliable subtest and correlates highly with Arithmetic . 1 8 . Vocabulary is the best measure of g in the scale and contributes to the Verbal Comprehension factor. Subte t specif­ icity is adequate at all ages. Vocabulary is the most reliable subtest in the scale and correlates highly with Information. 1 9 . Arithmetic is a good measure of g and contributes to the Freedom from Distractibility factor. It has ample or adequate subtest specificity except at ages 16 to 17. Arithmetic is a rel iable subtest and correlates best with Vocabulary and Information.

20. Comprehension is a good measure of g , and contributes to the Verbal Comprehension factor. Subtest specificity is ample or adequate except at ages 16 to 24 and 55 to 64. Comprehension is a reliable subtest and correlates highly with Vocabulary, Informa­ tion , and Similarities. 2 1 . Similarities is a good measure of g and contributes to the Verbal Comprehension factor. Subtest specificity is ample or adequate except at ages 16 to l7 and 20 to 24 . Similarities is a reliable subtest and correlates best with Vocabulary and Comprehension. 22. Picture Completion is a fair measure of g and contributes to the Perceptual Organization factor at all ages except 18 to 19 . Subtest specificity is ample at all ages . Picture Completion is a reliable subtest and correlates highly with Vocabulary. 2 3 . Picture Arrangement is a fair measure of g . Subtest spe­ cificity i ample except at ages 20 to 24 and 65 to 69. Picture Arrangement is a reasonably reliable subtest and correlates highly with Picture Completion and Vocabulary. 24 . Block Design is a good measure of g and contributes to the ' Perceptual Organization factor. Subtest specificity is ample or adequate at all ages. Block Design is a reliable subtest and correlates highly with Object Assembly. 25. Object Assembly is a fair measure of g and contributes to the Perceptual Organization factor. Subtest specificity is ample only at ages 18 to 19. Object Assembly is the least reliable subtest in the scale and correlates highly with Block Design. 26. Digit Symbol is a fair measure of g. Subtest specificity is either ample or adequate at all ages. Digit Symbol is a reliable subtest and correlates more highly with Block Design and Vocab­ ulary than with the other subtests. 27. The methods de cribed in Chapter 7 for interpreting the WISC-R also pertain, for the most part , to the WAIS-R. 28. Th e norm ta b les should be carefully studied in test-retest situat ions . 29. The assets of the WAIS-R include its good concurrent val idity, high reliabilities , excellent standardization , good admin­ istration procedu res. good manual, and helpful scoring criteria. 30. Limitations or the WAIS-R include limited floor and ceil­ ing, nonuniformity of subtest scores, difficulty of scoring re­ sponses, lack of normative data for raw scores. and failure to describe the procedure for establishing discontinuance criteria.

KEY TERMS, CONCE PTS, AND NAMES

Wechsler-Bellevue Intell igence Scale (p. 220) WAIS-R standardization (p. 220) WAIS-R Deviation IQ (p. 220) WAIS-R scaled scores (p. 220) Prorating procedure (p. 220) Scaled score equivalents for age groups (p. 220) Rel iabil ity of the WAIS-R (p. 220) Standard errors of measurement of the WAIS-R (p. 221) Stability of the WAIS-R (p. 222)

244

CHAPTER 1 0

Precision range of WAIS-R IQs (p. 222) Criterion validity of the WAIS-R (p. 223) Relationship between the WAIS-R and the WAIS, the WISC-R, and the Stanford-Binet : Fourth Edition (p. 223) Construct validity of the WAIS-R (p. 224) Cattell-Horn theory (p. 224) WAIS-R subtest intercorrelations (p. 225) WAIS-R IQs and stratification variables (p. 227) Premorbid estimation of WAIS-R IQ (p. 227) Factor analysis of the WAIS-R (p. 227) WAIS-R Verbal Comprehension factor (p. 228) WAIS-R Perceptual Organization factor (p. 228) WAIS-R Freedom from Distractibility factor (p. 228) WAIS-R subtests as measures of g (p. 228) Subtest specificity on the WAIS-R (p. 228) WAIS-R factor scores (p. 228) Starting rules on the WAIS-R (p. 231) Discontinuance rules on the WAIS-R (p. 231) Extrapolated IQs on the WAIS-R (p. 232) WAIS-R short forms (p. 234) Satz-Mogel abbreviated procedure for the WAIS-R (p. 234) WAIS-R Information (p. 235) WAIS-R Digit Span (p. 235) WAIS-R Vocabulary (p. 235) WAIS-R Arithmetic (p. 236) WAlS-R Comprehension (p. 236) WAIS-R Similarities (p. 237) WAIS-R Picture Completion (p. 237)

WECHSLER ADULT INTELLIGENCE SCALE-REVISED (WAIS-R)

WAIS-R Picture Arrangement (p. 237) WAIS-R Block Design (p. 238) WAIS-R Object Assembly (p. 238) WAlS-R Digit Symbol (p. 239) Profile analysis (p. 239) Base rate Verbal-Performance differences (probability-of­ occurrence approach) (p. 240)

STU DY Q U ESTIONS I . Discuss the following topics with respect to the WAIS-R: standardization, Deviation IQs, scaled score equivalents for age groups, reliability, and validity. 2 . Describe WAIS-R factor analytic findings. 3. Discuss some important factors involved in administering the WAIS-R. 4. Discuss WAIS-R short forms. Include a discussion of their values and limitations. 5 . Discuss the rationale, factor analytic findings, reliability and correlational highlights, and administrative considerations for each of the following WAIS-R subtests: Information, Digit Span, Vocabulary, Arithmetic, Comprehension, Similarities, Picture Completion, Picture Arrangement, Block Design, Object Assembly, and Digit Symbol . 6. Discuss the intent of profile analysis, methods of profile analysis, and approaches to profile analysis on the WAIS-R . 7 . Discuss the assets and limitations of the WAIS-R.

STAN FORD-B I N ET I NTELLI G E N C E SCALE : FO U RTH E DITION

A ll things

1937 and 1960 Stanford-Binet Intelligence Scales

are engaged i n writing their history. The planet, the pebble, goes attended by its shadow. The rolling rock leaves its scratches on the mountain; the river, its channel in the soil; the animal, its bones in the stratum; the fern and leaf, their modest epitaph in the coal. The falling drop makes its sculpture in the sand or the stone. Every act of the man inscribes itself ill memories, manners alldface. Every object is covered with hints which speak to the intelligent.

Introduction to the SB: FE Standardization Composite Scores, Standard Age Scores, and Jest-Age Equivalents Reliability Validity

-Emerson

Intercorrelations Between Subtests and Between Subtests and th Composite Score S B : FE Composite Scores and Stratification Variables Factor Analysis Administering the SB: FE SB: FE Subtests Interpreting the SB: FE A Successive Level Approach to Test Interpretation

Comparisons Between Factor Scores That Can Guide Interpretations Comparisons Between Subtests That Can Guide Interpretations Additional Suggestions for Using and Interpreting the SB: FE Assets of the SB: FE Limitations of the SB: FE Concluding Comments on the SB: FE Summary

245

CHAPTER I I

246

The Stanford-Binet Intelligence Scal e : Fourth Edition (SB : FE) (Thorndike, Hagen, & Sattler, 1986a) is a battery of15 subtests that covers an age range of2 through 23 years (see Figure 11-1 and Exhibit 11-1) . The SB: FE uses a point­ scale format similar to that of the Wechsler scales. Prior editions of the Stanford-Binet Intelligence Scales, as noted in Chapter 3 , used an age-scale format. Some of the major characteristics of the earlier versions of the Binet-Simon Scales and the Stanford-Binet Intelligence Scale are sum­ marized in Table 11-1 .

1 937 A N D 1960 STAN F ORD-B I N ET I NTELLI G E N C E SCALES

1937 Stanford-Binet Scales In 193 7 , after 21 years, the 1916 Stanford-Binet Intelligence Scale was revised . Two new forms were designed, Form L and Form M . The 1937 revision was recognized as a

,.

Figure I I - I .

A

STANFORD-BINET INTELLIGENCE SCALE: FOURTH E DITION

milestone in the progress of the individual testing of intel­ ligence. The scale was better standardized, two forms were made available, and there were more performance tests at the earlier levels . New types of tests were more prevalent at the preschool and adult levels, and more use was made of differential scoring of the same test. Improve­ ments were made in memory tests, in the wording of questions, in year-level assignments, and in the scoring of the Vocabulary test. The scale was extended downward to year level II, with tests appearing at half-year levels be­ tween years II and V and upward to the Superior Adult III level. Tests were also provided for year levels XI and XII . The scales had excellent reliability (ranging from . 98 for those with IQs below 70 to .90 for those with IQs above 129) and acceptable validity (r's of .40 to . 50 with school success) . Factor analytic studies indicated that most of the tests loaded heavily on a common factor, although group factors (for example, verbal , memory, visualization, spa­ tial , and reasoning) were also reported . The 1937 scales represented a significant improvement

_ .,,' ..I,.

1.*"::

Stanford-Binet I ntelligence Scale: Fourth E dition. Courtesy Riverside Publishing Company.

1 937 AND 1 960 STANFORD·BINET INTELLIGENCE SCALES

247

Table I I - I Some Characteristics of the Binet-Simon and Stanford-Binet Scales from

Scale year

Authors

Number of tests

Year levels covered

Very low grade idiots to upper elementary grades III to XIIl

1 905

Binet and Simon

30

1 908

Binet and S imon

59

191 1

Binet and Simon

54

III to Adult

1916

Terman

90

lIT to Superior Adult I

1 90 5

to

1 972

Modifications made in revisions

Limitations

Poorly standardized (N 50) Inadequate range Tests did not always discriminate No objective method for arriving at a total score =

ew tests added Some tests eliminated, especially those at the idiot level Tests grouped according to age commonly passed Mental-age concept introduced

New tests added, some eliminated Credit given for fraction of a year Tests shifted More detailed instructions Adult year level included New tests added Some tests revised Location changed for some tests Scoring and administrative procedures changed and better organized Alternate tests introduced IQ concept introduced Representative sampl ing attempted

1 937

Terman and Merrill

1 29

II to Superior Adult III

Better standardization (N 3,184) Two forms (L and M) More performance tests at earlier year levels

1 960

Terman and

142

II to Superior

One form (L-M) which incorporated best tests from Forms L

=

Inadequate standardization (N 203) No credits given for fractions of a year Lower year level tests too easy, higher year level tests too difficult Scoring and administrative procedures inadequate Unequal number of tests at different year levels Almost same limitations as those noted for 1 908 scale; there were no fundamental changes =

Poor standardization at extremes Only single form Inadequate standardization (N 1,000 native-born Californian children and 400 adults) Inadequate measure of adult mental capacity Too heavily weighted with verbal and abstract materials Inadequate scoring and administrative procedures at some points Some tests dated by the 1930s Some tests placed at wrong age level Too much credit for rote memory Equal variability at all ages not present Sample somewhat higher in socioeconomic level than general population, and more urban than rural subjects included Some tests difficult to score Low ceiling with above-average adolescents Too heavily weighted with verbal materials =

(Table continues next page)

248

CHAPTER I I

STANFORD-BINET INTELLIGENCE SCALE: FOURTH EDITION

Table I I-I (cont.)

Scale year

Authors

Merrill

1 973

Number o/ tests

Modifications made in revisions

Year levels covered

Adult III

Thorndike

and M New group of children used to check changes in test difficulty (N = 4,498) Some tests relocated , dropped, or rescored Substitution of Deviation IQ for ratio IQ- standard score with M = 100 and SD = 16 Use of age 18 years as ceiling level rather than 16 years Clarification of scoring principles Restandardized in 1 972 (N = 2 , 1(0)

over the 1916 scale and were greeted favorably by many reviewers . The scales were improved statistically and had many clinical applications. They were seen as efficient instruments for diagnosing mental retardation and for gaining insights into a child's temperament. Reviewers characterized the 1937 Stanford-Binet as combining the facets of the clinical interview with those of an objective assessment. Dissatisfaction was expressed, however, with the emphasis on verbal material , age-scale format, ceiling procedure , item placements, emphasis on rote memory, administration procedures, incomplete statistical data, use of one score only, inadequate measurement of g, inap­ propriateness of many items for adults, and applicability of the scale in clinical situations. Despite these criticisms, the 1937 scales were extremely popular, yielding acceptable validity coefficients and serving as the standard for the development of other tests. The scales, although by no means perfect instruments, served as important tools in clinical and educational settings.

1960 Stanford-Binet Scale: Form L-M In 1960 a new revision appeared. In many ways the 1960 edition was not a genuine revision. The revision was car­ ried out by selecting the best items from Form L and Form M and combining them into a new form. A new standard­ ization group was not obtained; instead, a sample of 4 ,498 subjects who had taken the scale between 1950 and 1954 was used to check on changes in item difficulty. New material was not introduced, nor were the essential fea­ tures of the scale changed. With the 1960 revision, only one form was available. Validity data were not presented with

Limitations

Originality and creative abilities not measured Inadequate for very superior students Abstract verbal tests appeared at too Iow a level and rote memory tests appeared at too high a level Restandardization procedures not appropriate

the revision; its validity rested on the fact that the same types of tests were used as in the 1937 scales. One of the most important developments in Form L-M was the replacement of the 1937 scale's IQ tables, which represented the conventional ratio IQ, with Deviation IQs for ages 2 through 18 years. The Deviation IQ is basically a normalized standard score with a mean of 100 and a stan­ dard deviation of 16. It expresses the deviation of the ratio IQ from the mean ratio IQ at each age level. The Deviation IQ controls for the variability in IQ distributions that was found to exist at various levels of the former revisions. A specific IQ at different ages in Form L-M indicates close to the same relative ability or standing regardless of the age of the examinee. Dissatisfaction was expressed with the manner in which the Deviation IQs were constructed and with the norming sample, however. Additionally, the 1960 scale was criti­ cized for being too heavily weighted with verbal materials, for not measuring creative abilities, and for improper placement of some items. The 1960 revision, however, still produced acceptable validity coefficients and remained one of the standard instruments for the assessment of children's intelligence until the 1980s.

1972 Norms Revised norms for Form L-M were published in 197 3 . Except for two minor changes in the test procedures (a more attractive female doll card was used and the word "charcoal" was substituted for "coal") , the tests in the scale and the directions for scoring and administration were the same . The standardization group for the revision consisted

1 93 7 AND 1 960 STANFORD-BINET INTELLIGENCE SCALE�

249

E xhibit 1 1 - 1 Stanford-Binet Intelligence Scale: Fourth Edition-Like Items

1 . Vocabulary (14 pictures, 32 words) The task on the first 14 items is to name the pictured object. Words 15 through 46 are both given orally and shown to the examinee; the examinee must define these words. train dime taut cryptography 2_ Bead Memory (42 items)

The task is to reproduce bead patterns by finding them in photographs or placing beads on a stick (see Figure 11-1)

3. Quantitative (40 questions)

The task is to solve quantitative problems. Match : : (see Figure 11-1) . Count the number of blocks in the picture. What is the smallest whole number that can be divided evenly by I, 2, and 3? How many 12-inch-by-12-inch tiles will be needed to cover a floor that is 6 feet by 6 feet?

4. Memory for Sentences (42 items) The task is to repeat successively longer sentences. Say : Small cow. Say : H igh clouds appeared on the horizon. Say : The field of science can aid mankind by discovering many useful things. 5 . Pattern Analysis (42 items)

For the first 6 items the task is to complete a form board (see Figure 11-1). For the remaining items the task is to reproduce stimulus designs using two, three, four, six, or nine blocks (see Figure 11-1). An example of a Pattern Analysis design is shown below.

9. Copying (28 items) The task is to reproduce designs with blocks (items I to 12) or to copy simple and complex geometric designs, such as l ines, rectangles, and partial circles, that are shown on cards.

10. Memory for Objects (14 items) The task is to recall pictured objects in the exact sequence i n which they were presented . The examinee points to the pic­ tured items o n a card that contains both the stimulus items and d istractor items. For example, the examinee is first shown a picture of a knife and a dog and then shown a card containing the pictures of a knife, dog , spoon, cat, and house. 1 1 . Matrices (26 items) The task is to select the object, design, or letter that best complet.es the matrix. An example of a Matrices item is shown below.

· Y A X ll A B C D E 1 2 . Number Series (26 items) The task is to predict the next two numbers in a series. To succeed, the examinee must discern the pattern in the series. 5 , 4 , 3 , _, _ 4 4 4 4 1 4 ' \3 ' If it _,

_

10, 2 , 1 1 , 2 , 1 2 , _, _

13. Paper Folding and Cutting (18 items) The task is to select the picture that shows how a folded and cut piece of paper would look unfolded. 6. Comprehension (42 questions) The task is to point to body parts (at earlier levels) and to answer questions dealing with social comprehension. Point to the doll's foot (see Figure 11-1). Why do we have nurses? What advantages does an airplane have over a car? Why do we have a Congress?

DO

7. Absurdities (32 items)

The task is to identify the incongruity in pictures. Writing with a spoon. Riding a bicycle that has one wheel missing. A map of the United States with Florida on the West coast.

8. Memory for Digits

The task is to recall digits correctly. Digits Forward contains 14 series of digits, 3 to 9 digits in length (Example: 3-9-5 ) . Digits Reversed contains 12 series o f digits, 2 t o 7 digits in length (Example: 5-8-9-4) .

A

B

c

14. Verbal Relations ( 1 8 items)

D

E

The task is to indicate how the first three items are alike but d ifferent from the fourth. (Exhibit continues next page)

250

CHAPTER I I

STANFORD-BINET INTELLIGENCE SCALE: FOURTH EDITION

Exhibit I I-I (cont.) How are a dog, cow, and pig alike but different from a bird? How are a car, truck, and motorcycle alike but different from a bicycle? 1 5 . Equation Building (18

items) The task is to arrange numbers and mathematical signs ( + , - , -7- , etc . ) into an equation.

of a representative sample of 2 , 100 children, with approxi­ mately 100 subjects at each Stanford-Binet year level . A special procedure, based on test scores from the group­ administered Cognitive Abilities Tests, was used for strat­ ifying each age sample to ensure proportionate representa­ tion of all ability levels . Unlike the 1960 norms, which did not include nonwhites in the standardization group, the 1972 norms contained nonwhites (including black and Spanish-surnamed individuals) and whites. Subjects were, however, excluded from the normative sample if English was not the primary language spoken in the home. The greatest differences in test scores between the 1972 and 1960 norms were found at the preschool level . For a similar performance, the 1972 norms yielded IQs that were about 10 points lower than those given by the 1960 norms . (See Sattler, 1982b, for a comprehensive review of adminis­ trative and interpretive procedures for the Stanford-Binet: Form L-M . )

4 2 6 + 'A 2 7 I Nore. These items are similar to those on the Stan rord-Binet: Fourth Edition, but they are not actually rrom the test.

ysis, Quantitative , Bead Memory, and Memory for Sen­ tences. A three-level hierarchical model was used to guide the construction of the SB: FE. The model postulates (a) g (a general intelligence factor) at the highest level of inter­ pretation ; (b) crystallized, fluid, and short-term memory factors at the second level ; and (c) more specific factors ­ such as verbal reasoning, quantitative reasoning, and ab­ stract visual reasoning - at the third level . The specific factors at the third level, plus short-term memory at the second level, form four area scores in the SB: FE. Within each area specific subtests are designated (see Table 11-2) ; however, the placement of subtests was not based on factor analytic findings. The Composite Score reflects the highest level and is considered to be the best estimate of g in the scale .

STA N DARDIZAT I O N I N TRODUCTI O N T O T H E S 8 : F E

Some continuity i s maintained between the SB: F E and prior editions. Item types the editions have in common are vocabulary (both pictures and words) , comprehension, picture absurdities, paper folding and cutting , copying, repeating digits , memory for sentences, copying a bead chain from memory, similarities, form-board items, and quantitative items. The S B : FE has some new items of these types and also some new item types (memory for objects, number series, and equation building). Every effort was made to design items with culture-fair content. A panel of ethnic minority reviewers evaluated the items for biased content. The 15 subtests are not used through all ages of the scale (see Table 11-2) . Some are administered only at the pre­ school and elementary school ages (for example, Absurdi­ ties and Copying) , whereas others are administered only at the upper year levels (for example, Number Series and Equation Building) . Of the 15 subtests, only 6 run through­ out the scale - Vocabulary, Comprehension , Pattern Anal-

The standardization sample consisted o f 5 ,013 individuals in 17 age groups. The number of individuals ranged from 194 in the 18-0 to 23-11 age group to 460 in the 5-0 to 5 -11 group. The sample was selected so as to be representative of the U . S. population according to 1980 census data. Stratification variables included geographic region, com­ munity size , ethnic group, age, gender, and socioeco­ nomic status. Because the final sample included too many children with high S ES backgrounds , weighting pro­ cedures were used to make the sample conform to the census data.

C O M POSITE SCORES, STA N DARD AGE SCORES, AND T EST-AGE E Q U I VALE NTS

Raw scores are converted into three types of standard scores: standard age scores (or scaled scores) for the subtests (M = 5 0 , SD = 8 ) , area scores (M = 100 , SD = 16), and a Composite Score (M = 100, SD = 16) . The Composite Score is similar to the Deviation IQ em­ ployed on the Wechsler scales . The Wechsler scales use a

25 1

RELIABILITY Table 1 1 ·2 Areas, Subtests, and Age Spans for Stanford·Binet: Fourt h Edition

Designated area

Subtes(

Age span

Verbal Reasoning

Vocabulary Comprehension Absurdities Verbal Relations

2 to 23 2 to 23 2 to 14 12 to 23

Abstract/Visual Reasoning

Pattern Analysis Copying Matrices Paper Folding and Cutting

2 to 23 2 to 1 3 7 to 2 3 1 2 to 2 3

Quantitative Reasoning

Quantitative Number Series Equation Building

2 to 2 3 7 to 2 3 1 2 to 2 3

Short-Term Memory

Bead Memory Memory for Sentences Memory for Digits Memory for Objects

2 to 2 3 2 to 23 7 to 2 3 7 to 2 3

standard deviation of l 5 , however, whereas the S B : F E uses a standard deviation of 16, which is consistent with pre­ vious editions of the Stanford-Binet. Raw scores obtained by the examinee on each subtest are first converted into standard age scores within the examinee's own age group through tables in the Guide for Administering and Scoring the Fourth Edition. Age groups are divided into 4-month

intervals from 2 to 6 years (the one exception is a 3 V2 month interval at age 2-0-0 to 2-3-15) , 6-month intervals from 6 to 12 years, I-year intervals from 12 to 18 years, and a 6-year interval from 18 to 23-11 years . The procedure used in the SB: FE allows for the com­ putation of area and Composite Scores when less than the entire battery is administered. Proration or special short form procedures are not needed to calculate area or Com­ posite Scores . After the subtest scaled scores have been obtained, they are entered in a table that provides area standard age scores (M = 100, SD = 16) . The area score can be obtained by use of one or more of the subtest scaled scores that comprise the area. The area standard scores then can be converted to a Composite Score through use of a table on pages 187 and 188 in the Guide. This table is divided into two sections ; one part is for ages 2-0-0 to 9-11-15 , the other for ages 9-11-16 to 23-11-15. As with the areas scores, the Composite Score can be obtained by use of 1, 2, 3 , or 4 area scores . As you w i l l read shortly, this text advocates the use of factor scores in place of area scores. Consequently, the

factor scores will be highlighted throughout the chapter. The area score tables in the Guide, however, still must be used to compute the Composite Score and the factor scores. The Stanford-Binet: Fourth Edition Technical Manual provides a table (Table G . 1) of test-age equivalents to facilitate interpretation of a child's performance. Chapters 2 and 4 in this text discuss the advantages and disadvan­ tages associated with the use of test-age equivalents .

RELIABI LITY

The Composite Score of the SB: FE has excellent reliabil­ ity. Internal consistency reliabilities for the Composite Score range from . 95 to . 99 over the 17 age groups. The median Composite Score reliability is .97. (Reliability of the factor scores will be discussed later in this chapter. ) As expected, the reliabilities for the subtests are less satisfactory than those for the Composite Score (see Table 1 1-3) . The median subtest reliabilities range from a low of . 7 3 for Memory for Objects to a high of .94 for Paper Folding and Cutting. Subtest reliabilities differ somewhat according to age group, ranging from a low of . 66 for Memory for Objects at age 10 years to a high of .96 for Pattern Analysis at ages 18-23 .

Standard Errors o f Measurement The median standard error of measurement (SEn) in scaled-score points (M = 100 , SD = 16) is 2 . 8 for the Composite Score (see Table 11-3). For the subtests, the smallest median SE,,/s , are associated with Paper Folding and Cutting and Pattern Analysis (2 . 0 and 2 . 3 scaled-score points, respectively), and the largest median SEm's are associated with Memory for Objects and Memory for Digits (4 .2 and 3 . 3 scaled-score points, respectively) .

Stability The stability of the SB: FE was assessed by retesting two groups (57 children with a M age of 5-2 and 55 children with a M age of 8-1) after an interval of two to eight months (Thorndike et al . , 1986b). For the 5-year-olds, the sta­ bility coefficient was excellent for the Composite Score (r.xx = .91). Stability .coefficients for the eight subtests ranged from .56 for Bead Memory to . 7 8 for Memory for Sentences. For the 8-year-olds, the stability coefficient was again excellent for the Composite Score (rxx = . 90) . Sta-

252

CHAPTER I I

Table 1 1 -3 Median Reliabi l ity Coefficients and Standard Errors of Measurement for Stanford-Binet Intelligence Scale: Fourth Edition Subtests, Factor Scores, and the Composite Score

Subtest, Factor Score, or Composite Score

Median standard Median reliability error of coefficient measurement

Subtest

Vocabulary Comprehension Absurdities Verbal Relations Pattern Analysis Copying Matrices Paper Folding and Cutting Quantitative Number Series Equation Building Bead Memory Memory for Sentences Memory for Digits Memory for Objects

.87 . 89 .87 .91 .92 . 87

. 90

. 94

. 88

. 90

.91

.87 . 89 . 83 .73

2.9 2.6 2.9 2.4 2.3 2.9 2.5 2.0 2.8 2.5 2.4 2.9

Precision Range Table CM in Appendix C shows the 68, 85 , 90 , 95 , and 99 percent confidence intervals for the three factor score s ­ Verbal Comprehension, Nonverbal MemorylVisualiza­ tion, and Memory - and for the Composite Score for all age groups in the standardization sample. Confidence in­ tervals are not provided for the average of the age groups because the composition of the scale changes at various age levels. Additionally, the confidence intervals change con­ siderably with age. For example , the confidence interval for the Composite Score at the 68 percent level is ±4 for 2year-olds and ± 2 for 10-year-olds . Confidence intervals therefore should be reported for the examinee's specific age group.

VALI DITY . 95 .96 .91

3.6 3.2 4.8

. 97

2.8

Composite

Composite score

individual subtest scores. The average change noted for the Composite Score was about half of a standard deviation . Considerable fluctuation in scores was evident on the indi­ vidual subtests, suggesting that they should not be relied on to provide stable measures of ability.

2.6 3.3 4.2

Factor

Verbal Comprehension Nonverbal Reasoning/visualization Memory

STANFORD-BINET INTELLIGENCE SCALE: FOURTH EDITION

Source: Adapted, i n part, from Thorndike et aI . ( l 986b).

bility coefficients for the 12 subtests ranged from .28 for Quantitative to . 86 for Comprehension. In both groups, children obtained higher scores on the second testing. The mean Composite Score change was 8 . 2 points for the 5-year-olds and 6.4 points for the 8-year­ olds. For the individual subtests in the 5 -year-old sample, mean changes ranged from a low of l . 5 scaled-score points on Vocabulary to a high of 4 . 3 scaled-score points on Copying. For the individual subtests in the 8-year-old sample, mean changes ranged from a low of 1 . 2 scaled­ score points on Memory for Objects to a high of 3 . 7 scaled-score points o n Pattern Analysis. The test-retest reliability coefficients indicate that the Composite Score is substantially more reliable than the

The Technical Manual for the SB: FE presents several studies investigating the Scale's criterion validity. Com­ parisons were made with various other tests, including the Stanford-Binet: Form lrM, WISC-R, WPPSI, WAIS-R, and K -ABC, for both normal and exceptional populations . In the 13 studies reported in the manual, correlations be­ tween the SB: FE and these criterion measures ranged from a low of .27 to a high of . 91 (Mdn r = . 80) . The median r of . 80 supports the concurrent validity of the SB: FE. The one study with a low correlation was with a gifted sample of 82 7 -4-year-old children who were admin­ istered both the 5B: FE and Form lr M. In 10 of the 1 3 studies the SB: FE yielded lower mean scores than did the criterion test. In most cases, differences were 5 points or less . In two studies, however, the SB: FE yielded substan­ tially lower scores. Scores of a gifted population were , on the average, 13 . 7 points lower than Form lrM scores, those of a mentally retarded popUlation were 9 . 3 points lower than WAlS-R scores. The reasons for these findings are unclear at this time . They could be associated with ceiling or floor effects, item difficulties, better differentiation of abilities, or some combination of these or other factors. Additional studies have investigated the concurrent va­ lidity of the SB: FE. In a sample of 30 normal children (M age 1 l . 3 ) , the S B : FE correlated highly with Form =

253

SB: FE COMPOSITE SCORES AND STRATIFICATION VARIABL.ES L- M ( r = . 72 ) and produced similar overall sc res (M = 114 .42 , SD = 12 . 3 1 on the S B : FE; M = 113 . 07 , SD = 15 . 18 o n Form L-M) (Hartwig, Sapp, & Clayton, 1 98 7 ) . W i t h a sample of 120 g i fted c h ildren (M age = 6 . 11 ) , the S B : FE yielded lower overall scores (M = 122 .46, SD = 9 . 17 ) than Form L-M (M = 130 .45 , SD = 1 1 . 20 , r = . 64) (Livesay, 1986) . With a population of 19 special education children aged 7 to 16 years the WISC-R and the SB: FE yielded scores that were essentially similar (M = 83 . 63 , SD = 14 . 82 on WISC-R and M = 83 .11, SD = 1 3 . 30 on SB: FE, r = . 88) (Hollinger & Baldwin, 1987) . With 32 normal second­ grade children (M age = 7 . 77), the SB: FE yielded a mean Composite Score that was 7 points lower than the WISC-R score (M = 105 . 5 3 , SD = 9 . 84 on SB : FE; M = 112 . 5 3 , SD = 9 .42 o n WISC-R) (Rothiisberg, 1987) . With 166 gifted children (M age = 10 . 3 3 ) , the SB : FE and the W I S C - R p r o d u c e d s i m i l a r s c o r e s (M = 12 1 . 4 7 , SD = 9 . 72 o n S B : F E , M = 123 . 3 3 , SD = 9 . 5 8 on WISC-R, r = . 55 ) (Livesay, 1986) . And with 32 normal college students (M age = 18.0), the WAlS-R and the SB: FE yielded similar scores (M = 103 . 5 , SD = 10.9 on WAlS-R, M = 100. 9 , SD = 9 . 1 on SB: FE, r = .91) (Car­ vajal, Gerber, Hewes, & Weaver, 1987). The results of the preceding studies suggest that for populations within the average intellectual range, the SB: FE is likely to yield Composite Scores that are similar to those provided by the WISC- R, WAlS-R, and Form L­ M . With gifted and mentally retarded populations, how­ ever, the SB: FE may yield lower scores than Form lrM and the WAlS-R. In studies with the WRAT-R, the SB: FE Composite Score has been found to have acceptable validity coeffi­ cients . With a sample of 47 emotionally disturbed child.en (ages -5 to 16-1), the correlations were . 5 1 with Reading , . 55 with Spelli n g , and . 58 with Arithmetic (Blakes­ lee, 1987) . With 49 emotionally disturbed children (M age = 13-4), the correlations were . 48 with Reading, .37 with Spelling, and . 70 with Arithmetic (Lewis-O'Donnell ' N�. The results of the preceding studies are based on the administration of the entire battery. It is not known to what extent results would be s imilar if the complete battery were not administered . Construct validity was established in a number of ways. . First, raw scores increase as a function of age. Second, factor analyses support a number of dimensions of the scale, such as adequate to high g loadings of the subtests and specific factors at various age levels of the scale. Third, all of the &ubtests correlate moderately to highly positively with the Composite Score.

�

I NTERCORRE LAT I O N S B ETWEE N SU BTESTS AN D BETWEEN SUBTESTS A N D THE COM PO S I T E SCORE

I ntercorrelations provide i n formation about the rela­ tionships of the SB: FE subtests to eacn other and to the Composite Score . Median i ntercorrelations between the 15 subtests range from a low of . 29 to a pigh of . 73 (Mdn r = .47) . The highest median subtest intercorrelations ( . 60 and above) are between Vocabulary and Comprehen­ sion ( . 73), Vocabulary and Verbal Relations ( . 71) , Quan­ titative and Number Series ( . 67 ) , Matrices and Number Series ( . 66), Number Series and Equation Building ( . 64) , Vocabulary and Memory for Sentences ( . 64) , Comprehen­ sion and Verbal Relations ( . 63 ) , Vocabulary and Absur­ dities ( .62) , Quantitative and Equ�tion Building ( . 61), Pa­ per Folding and Cutting and Number Series ( . 61), and Pattern Analysis and Paper Folding and Cutting ( . 60) . The lowest median subtest i ntercorrelations (less than .40) are between Memory for Objects and Paper Folding and Cutting ( . 29), Memory for Objects and Equation Building ( . 30) , Memory for Objects and Verbal Relations ( . 32) , Memory for Objects and Absurcjities ( . 33 ) , Mem­ ory for Digits and Copying (. 3 3 ) , Memory for Digits and Comprehension ( . 34), Memory for Digits and Absurdities ( . 34) , Memory for Digits and Pattern Analysis ( . 35) , Memory for Sentences and Paper Folding and Cutti�g ( . 35 ) , Memory for Objects and Pattern Analysis ( . 36), Memory for Digits and Verbal Relations ( . 36) , Memory for Di its and Paper Folding and Cutting ( . 37), Memory for Objects and Comprehension ( . 38), and Memory for Sentences and Equation Building ( . 3 8 ) . Median correlations between the subtests and the Com­ posite Score range from . 60 to . 82 (Mdn = . 74) . Quan­ titative and Number Series have the highest correlation with the Composite Score (both . 82) , followed by Vocabu­ lary ( . 81), Matrices ( . 78), and Comprehension and Verbal Relations (both .76) . Memory for Objects has the lowest correlation with the Composite Score ( . 60) , followed by Memory for Digits ( . 64) , Copying ( . 66) , Absurdities ( . 72), and Bead Memory ( . 72 ) . Thus reasoning and verbal comprehension subtests have the highest correlations with the Composite Score and memory subtests have the lowest correlations .

SB: FE COM POSI T E SCORES A N D STRATIFICAT I O N VAR I A BLES

The relationship between Composite Scores and the demo­ graphic characteristics of the standardization sample are

CHAPTER I I

254

shown for three age groups (Table 4.5 in the Technical Manual): (a) 2 through 6, (b) 7 through 11, and (c) 12 through 23 . Differences between males' and females' mean Composite Scores were less than 2 points. Thus sex differ­ ences are not large enough to assume any practical signifi­ cance on the scale. Mean Composite Scores of white examinees were about 10 to 17 points higher than those of black examinees: 14 points at ages 2 through 6 (104 . 7 vs. 91.0), 10 points at ages 7 through 11 (102 . 6 vs. 92 . 7 ) , and 17 points at ages 12 through 23 (103 . 5 vs. 86 . \) . White exam­ inees also obtained higher scores than other ethnic groups, except at ages 7 through 1 1 , where Asian examinees ob­ tained a mean Composite Score of 103 . 6 and the white examinees a mean Composite Score of 102 . 6 . This differ­ ence , however, is not very meaningful . Table 11-4 shows the mean Composite Scores based on the median scores of the three age groups for the various demographic charac­ teristics studied .

STANFORD-BINET INTELLIGENCE SCALE: FOURTH EDITION

Mean Composite Scores show a clear relationship to parental education and parental occupational groups : Composite Scores of children whose parents were college graduates were about 16 points higher, on the average, than those of children whose parents had less than a high school education (110 . 1 vs. 94 . 4 ) . Similarly, Composite Scores of children of managerial and professional workers were 14 points higher, on the average , than those of children of operators , fabricators, and others (108 . 0 vs. 94 . 0) . Com­ munity size differences varied depending on the age group in the standardization sample . For example , in the 2 through 6 age group the range was about 5 points (103 . 4 v s . 9 8 . 1) . I n the 7 through 11 age group, the range was about 12 points - a low of 91.4 for cities with 300,000 to 999,999 versus a high of l03 .0 for moderately sized cities (25 ,000 to 99,999) . Finally, in the 12 through 18-23 age group, the range was about 8 points (102 . 6 for moderately sized cities vs. 94 . 4 for small towns) .

Table 1 1 -4 Relationship of Stanford-Binet I ntelligence Scale: Fourth Edition Composite Scores to Sex, Race, Parental Education, Parental Occupation, and Community Size Demographic variables

M

SD

Sex

Males Females

99. 1 1 00.4

1 6 .0 15. 1

Race

Asian Black Hispanic Native American White

99.9 9 1 .0 94 .9 94. 7 103 . 5

15.4 1 3 .2 13.4 17.7 15.8

I . Less than high school 2 . High school graduate 3 . I t o 3 years college 4 . College graduate

94 .4 99.0 103.4 1 1 0. 1

14.8 14.3 15.2 14.4

108.0 1 0 1 .5 93.5 96.0 96. 8 94. 0

15.3 14.3 15.7 1 6. 7 16.6 15.0

98.6 98.5 100.4 1 03 . 0 98 . 1 1 0 1 .0

15.8 1 7 .9 14.4 14. 1 1 6. 6 14.8

Parental education

Parental occupational group

I . Managerial/professional

2. 3. 4. 5. 6. Community size

Technical/sales Service occupations Farming/forestry Precision production Operators, fabricators, other

I . 1 ,000,000 or more

2. 3. 4. 5. 6.

300,000 t o 999,999 1 00,000 to 299,999 25,000 to 99,999 2 ,500 to 24,999 less than 2,500

Note. These are mean Composite Scores based o n the median scores of the three age g roups listed i n Table 4 . 5 o f the Technical Manualfor the Stanford­ Binet: FOl/rth Edition . SOl/ree: Adapted from Thorndike et a l . ( l 986b ) .

FACTOR ANALYSIS

255

FACTOR ANALYSIS

are not continuous throughout the scale and because differ­ ent subtests are administered at different ages, the factor structure of the scale differs at different ages . The results of the principal components analysis are summarized in Table 11-5 . At ages 2 through 6, a two­ factor solution best characterizes the scale: Verbal Com­ prehension and Non verbal Reasoning/Visualization . At ages 7 through 23 , a three-factor solution is most appropri-

The findings in this section are based on the re ults ob­ tained from principal components analysis with varimax rotation of the 15 subtests of the SB : FE administered to the standardization sample. It differs from the one presented in the Technical Manual. A principal components analysis lends itself to the development of factors that may be useful in guiding interpretations needed for clinical and psycho­ educational evaluations . Because subtests in the SB : FE

Table 1 1 -5 Stanford-Binet I ntelligence Scale: Fourth Edition Subtest Loadings on Factor A (Verbal Comprehension). Factor B (Nonverbal ReasoningNisualization), and Factor C (Memory) for Seventeen Age Levels Following Varimax Rotation Age level Subtest

2

3

4

5

6

7

8

9

10

II

12

13

14

15

16

17

1 8a Mdn.

72 52 72 61 49 74 43 18 38 33 01 35 42 17

76 70 40 53 27 24 37 22 52 29 23 40 69 46 43

79 68 49 67 23

78 62

75 68

78 46

77 63

70 21

75 26

60 15

67 28

33 37 49 33 34 29 57 25 17

24 28 38 25 38 10 39 15 18

38 31 56 41 45 29 51 15 15

25 23 41 24 19 30 45 20 25

25 24 40 28 22 14 56 32 07

77 68 50 67 27 25 25 24 41 28 23 29 58 22 17

33 29 45 45 60 32 65 67 59 74 58 51 19 35 31

33 40

33 45

37 39

30 28

38 27

33 32

32 76

30 61

34 71

37 51

45 32

34 66 52

63 65 56 61 42 50 16 25 32

64

60 63 54 69 51 51 20 25 19

58 70 60 69 73 40 18 26 16

65 65 47 70 72 38 19 28 18

60

Factor A - Verbal Comprehension

Vocabulary Comprehension Absurdities Verbal Relations Pattern Analysis Copying Matrices Paper Folding and Cutting Quantitative Number Series Equation Building Bead Memory Memory for Sentences Memory for Digits Memory for Objects

75 66 53

76 78 72

78 76 60

73 82 59

79 74

44

71 71 41

77 79 50

79 75 49

70 66 46

81 68 51

28 13

34 28

29 34

32 28

24 18

17 16 10

26 27 20

28 19 13

18 30 18

27 29 28

33

32

42

40

36

40 18

44 27

34 23

41 26

50 32

11 67

20

36 56

40 59

26 63

31 60 26 20

28 58 20 16

26 51 21 08

31

42 63 22 28

64

Factor B

Vocabulary Comprehension Absurdities Verbal Relations Pattern Analysis Copying Matrices Paper Folding and Cutting Quantitative Number Series Equation Building Bead Memory Memory for Sentences Memory for Digits Memory for Objects

17 37 44 64

58

43

55 12

29 32 33 66 70

63

65 26

36 34 42 71 62

68

63 37

38 32 43 71 60

70

63 39

-

64

27 11

IS

Nonyerbal Reasoning/Visualization

27 25 43 70 53

64

63 63

35 29 48 55 48 23

30 30 49 72 52 57

32 26 49 69 52 63

45 45 69 77 49 57

48 14

40 57

43 59

54 61

56 20 19 19

57 26 24 37

42 15 20 22

48 16 14 30

35 33 58 64

36 47 27 03 39

46 -04 55 47 25 35 49 48 68 51 54 33 26 24 33 25 36 2 1

44

44

71 58 73 60 36 13 24 26

65 54 64

58 51 20 25 24

(Table continues next page)

CHAPTER I I

256

STANFORD-BINET INTELLIGENCE SCALE: FOURTH E DITION

Table 1 1 -5 (cont_)

Age level Subre:;'

2

3

4

5

6

7

8

9

10

11

Factor C - Memory

Vocabulary Comprehension Absurdities Verbal Relations Pattern Analysis Copying Matrices Paper Folding and Cutting Quantitative Number Series Equation Building Bead Memory Memory for Sentences Memory for Digits Memory for Objects

22 26 25

22 15 14

28 21 13

29 21

26 25 16

32 26 53

17 16 31

22 10 37

19 22 40

29 37 41

11 80

22 28

29 33

27 38

44

14 36 57 42

24 50 81

37 61 55 53

34 53 71 58

29 45 71 37

44

04

52

12

13

14

15

16

17

I /fa Mdn_

44

24 20 42 07 28 82 31 15 23 36 18 31 37 38 30

33 30 30 16 27

35 38

32 21

32 36

29 28

17 27

16 28

33 12

15 29

37 14 49 32 18 43 65 70 51

31 17 39 36 24 47 57 74 54

20

25 17 40 34 17 49

43 13 12 14 37 29 15 40 30 29 45 67 60 63

33 15 38 43 29 43 57 64

53

04

33 42 27 29 69 64

32

64

66 61

29 26 15 16 27 26 32 15 36 36 24 36 57 65 52

Note_ Decimal points omitted. Factor loadings are for a two-factor solution at ages 2 , 3 , 4 , 5 , and 6 ; three-factor solution at ages 7 , 8 , 9 , 1 0 , I I , 1 3 , 1 4 , 1 5 , and 1 6 ; and four-factor solution a t ages 1 2 , 1 7 , and 1 8-23 . Represents ages 1 8-23. •

ate : Verbal Comprehension, Nonverbal Reasoning/Visu­ alization, and Memory. The grouping of the subtests for factor scores at various age levels is shown in Table 11-8. The rationale is explained further in the section on Factor Scores later in this section_ The Verbal Comprehension factor score measures ver­ bal knowledge and understanding obtained by formal and informal education and reflects the ability to apply verbal skills to new situations. The Nonverbal Reasoning/visu­ alization factor score reflects the ability to interpret and organize visually perceived material , to perform basic arithmetical operations using visual cues or verbal cues , to visualize patterns, to demonstrate visual-motor skills, and to use reasoning to solve problems. Both reasoning and visualization are key components of this factor. The Mem­ ory factor measures the ability to attend or concentrate, or short-term memory, but also may involve sequencing skills.

SB: FE Subtests as Measures of g As shown in Table 11-6, all of the SB : FE subtests are either good or fair measures of g (the general intelligence factor) . Based on the median of the 17 age l evels , the subtests with the highest loadings are Vocabulary ( . 80) , Number Series

( . 79) , Quantitative ( 77), Comprehension ( . 75 ) , Matrices ( . 74) , Absurdities ( . 72) , Memory for Sentences ( . 7 1 ) , and Pattern Analysis ( . 70) . The subtests with fair g loadings are Paper Folding and Cutting ( . 69) , Bead Memory ( . 68), Verbal Relations ( . 68 ) , Equation Building ( . 67) , Copying ( . 61), Memory for Digits ( . 60) , and Memory for Objects ( . 54) . The subtests with high g loadings include both verbal and nonverbal subtests. _

Subtest Specificity Subtest specificity refers to the proportion of a subtest's variance that is both reliable (that is, not due to errors of measurement) and distinctive to the subtest. Although individual SB: FE subtests overlap in their measurement properties , many of them possess a relatively high degree of specificity, which justifies interpretation of specific sub­ test functions (for example, interpreting Verbal Relations as a measure of conceptual thinking) (see Table 11-7) . Of the 15 subtests in the scale, 11 have ample or adequate specificity throughout the entire age range covered by the subtests. These subtests are Absurdities, Verbal Relations, Pattern Analysis , Copying, Matrices, Paper Folding and Cutting, Number Series, Equation Building, Bead Mem­ ory, Memory for Sentences , and Memory for Digits . The

257

FACTOR ANALYSIS Table 1 1 -6 Stanfo rd-Binet I ntelligence Scale: Fourt h Edition Su btests as Measures of g Fair measure of g

Good measure of g

Subtest

Median loading of g

Vocabulary Number Series Quantitative Comprehension Matrices Absurdities Memory for Sentences Pattern Analysis

. 80 .79 .77 . 75 . 74 . 72 .71 . 70

PercenTage of variance attributable fa g

Subfesf

64

Paper Folding and Cutting Bead Memory Verbal Relations

62 60 56 55 52 50 49

Equation Building

Copy ing Memory for Digits Memory for Objects

four subtests with inadequate specificity at various ages are Vocabulary, Comprehension , Quantitative, and Memory for Obj ects. At ages at which subtests have inadequate specificity, scores should not be interpreted as measuring specific functions , and cautious interpretation is required for sub­ tests falling within the adequate specificity category. These subtests , however, can be interpreted as measuring g and the appropriate factor (Verbal Comprehension , Nonverbal ReasoninglVisualization, or Memory) .

Median loading ofg

Percel1lage of variance attributable to g

. 69 .68 .68 .67 .61 . 60 .54

48 46 46 45 37 36 29

Obtaining Factor Scores Of the 15 subtests in the battery, 12 have been selected as the primary subtests to form the factor scores. The preferred way to obtain the three factor scores is to use the following combination of subtests: Verbal Comprehension

=

(a) Sum of scaled scores on Vocabulary, Comprehension , Absurdities , and Memory for Sentences at ages 2 through 7

Table 1 1 -7 Amount of Specificity in Stanford-Binet: Fourth Edition Subtests Ample specificity

Inadequate specificity

Adequate specificity

Subtest

Ages

Subtest

Ages

Subtest

Ages

Vocabulary Comprehension Absurdities Verbal Relations Pattern Analysis Copying Matrices Paper Folding and Cutting Quantitative Number Series Equation Building Bead Memory Memory for Sentences Memory for Digits Memory for Objects

2 and 1 5 2 , 7, 10- 1 3 , 1 6- 1 8 2-9, 1 1 - 1 4 12-18 2- 1 3 , 1 5- 1 8 2- 1 1 , 1 3 7-18 1 2- 1 8 2-5, 7- 1 3 , 1 7 8- 1 1 , 1 4- 1 5 1 2- 1 8 all ages 2-17 7-9, 1 1 - 1 8 7 - 1 1 , \ 3- 1 8

Vocabulary Comprehension Absurdities Pattern Analysis Copying Quantitative Number Series Memory for Sentences Memory for Digits

1 0 , 1 2- 1 3 , 16- 1 8 3-4, 1 4- 1 5 10 14 12 1 4- 1 6 7, 1 2- 1 3 , 16- 1 8 18 10

Vocabulary Comprehension Quantitative Memory for Objects

3-9, 1 1 , 14 5-6, 8-9 6, 1 8 12

NOfe. Age 1 8 includes ages 1 8 through 2 3 . The procedure described by Kaufman ( 1 975a) was used to evaluate the specificity of the subtests.

258

CHAPTER I I (b) Sum of scaled scores on Vocabulary, Comprehension , and Absurdities at ages 8 through 1 4 (c) Sum o f scaled scores on Vocabulary, Comprehension, and Verbal Relations at ages 1 5 through 23

Nonverbal Reasoning/ Visualization

= (a) Sum of scaled scores on Pattern Analysis, Copying, Quantitative , and Bead Memory at ages 2 through 11 (b) Sum of scaled scores on Pattern Analysis , Matrices , Quantitative , and Bead Memory at ages 1 2 through 23

Memory (not calculated for ages 2-6) = (a) Sum of scaled scores on Memory for Digits and Memory for Objects at age 7 (b) Sum of scaled scores on Memory for Sentences, Memory for Digits, and Memory for Objects at ages 8 through 23 In addition to the subtests noted above , other subtests can be used to obtain factor scores. For the Verbal Com­ prehension factor, Verbal Relations can be used at ages 12 through 14 . For the Nonverbal Reasoning/visualization factor, Matrices and Number Series can be used at ages 9 through 1 1 , Number Series and Equation Building at ages 12 through 18-2 3 , and Paper Folding and Cutting at ages 13 through 18-2 3 . Table 11-8 l ists the factor scores by age level and subtests, and Exhibit 11-2 shows a worksheet for com­ puting factor scores, together with an example. The factor scores are highly reliable (r;u's range from .91 to .96, w ith SEm's of 3 . 2 to 4 . 8 ; see Table 11-3 ) .

STANFORD-BINET INTELLIGENCE SCALE: FOURTH EDITION

(1987) provides useful information for administering the SB: FE.

Adaptive Testing All examinees are first given the Vocabulary subtest, which is also referred to as the routing test. Performance on this subtest, together with the examinee's chronological age , is used to determine the entry level for all of the other subtests from a routing chart. The routing subtest and routing chart, however, may not be accurate for mentally retarded children and perhaps other groups. One study (Bissette , 1987) and several field reports have indicated that the entry levels were too high for trainable mentally retarded children . In fact, the entry levels suggested in the Guide tend to be above TMR children's ceiling levels on most subtests. Examiners will have to adjust the entry level to make it more applicable to the examinee's probable level of performance .

Basal level and ceiling levels. The two consecutive lev­ els at which all items are passed (4 items) are termed the basal level. The two consecutive levels at which three of the four items or all four items are failed are termed the ceiling level. The ceiling level is also the point at which a subtest is discontinued . When a double basal level is ob­ tained, the lower basal level is used in calculating the raw score . When a double ceiling level is obtained, the higher ceiling level is used in obtaining the raw score . These procedures take into account all of the child's actual failures and successes. A complete administration of the SB: FE requires the establishment of one basal level and one ceiling leve l . In cases where either a basal or a ceiling level is not reached, scores can still be reported, but with the notation "Estimate" following the scores. Chronological age. The child's chronological age is ob­ tained by subtracting date of birth from date of examina­ tion . Chronological age is recorded in years and months, with 16 or more days rounded to the next higher month . Timing of items.

A D M I N ISTE R I N G T H E S 8 : F E

In the S B : F E items in each subtest are arranged in order of increasing difficulty, with two items of approximately equal difficulty placed at each level. Levels are designated by letters. The levels are used to determine entry and discontinuance points . Figure 11-2 shows the front page of the record booklet . A guide by Delaney and Hopkins

Of the 15 subtests in the SB: FE, only one - Pattern Analysis- calls for timing of the examinee . (In several subtests the presentation of stimulus items is timed . ) In the other subtests, judgment must be used to decide when to proceed to the next item.

Scratch paper.

Scratch paper may be used on the Quan­ titative, Number Series, and Equation Building subtests . The scratch paper should be collected after the exami­ nation.

ADMINISTERING THE SB: FE

259

Table 1 1 -8 Suggested Subtest Combinations for Factor Scores at Various Age Ranges for the Stanford-Binet: Fourth Edition Age

Factor score Verbal Comprehension

2 through 7

8 through 1 4

15 through 1 8-23

Nonverbal Reasoning/visual ization

2 through

II

1 2 through 1 8-23

Memory

2 through 6 7

8 through 1 8-23

Subtest combination Vocabulary Comprehension Absurdities Memory for Sentences Vocabulary Comprehension Absurdities Vocabulary Comprehension Verbal Relations Pattern Analysis Copying Quantitative Bead Memory Pattern Analysis Matrices Quantitative Bead Memory None Memory Memory Memory Memory Memory

for Digits for Objects for Sentences for Digits for Objects

Other subtests may be added or substituted to obtain factor scores. For the Verbal Comprehension factor, Verbal Relations can be used at ages 1 2 . 1 3 , and 1 4 . For the on verbal ReasoninglVisualization factor, (a) Matrices and Number Series can be used at ages 9 , 1 0, and I I ; (b) Number Series and Equation Building at ages 1 2 through 1 8-23; and (c) Paper Fold i ng and Cutting at ages 1 3 through 1 8- 2 3 .

No/e.

Physical Abilities Necessary for the SB: FE Adequate hearing and language functions are required for verbal subtests, and adequate vision and/or visual-motor ability are required for the nonverbal subtests (see Table 1 1-9). As discussed in Chapter 6 for the WISC-R, for examinees with physical disabilities, you must attempt to find ways to give the test without, in the process, providing the child with cues to answers. Alternative ways of admin­ istering subtests to children younger than 6 or 7 years of age are restricted by the fact that they have limited writing and reading skills . Most of the discussion about adaptive administration of the WISC-R (see page 140) also pertains to the SB: FE.

Short Forms Various short forms are suggested in the Guidefor Admin­ istering and Scoring the Fourth Edition, including a four-

subtest short form composed of Vocabula ry, Bead Memory , Quantitative, and Pattern Analysis. Another is a six-subtest short form composed of Vocabulary, Bead Memory, Quantitative, Memory for Sentences, Pattern Analysis, and Comprehension . Both of these short forms contain subtests that are found at all age levels of the scale . Additionally, the six-subtest short form yields estimates of two factor scores: Verbal Comprehension (Vocabulary, Comprehension , and Memory for Sentences) and Nonver­ bal Reasoning/Visual ization (Pattern Analysis, Quan­ titative, and Bead Memory). When the complete battery is administered, Memory for Sentences is part of the Verbal Comprehension factor at ages 2 through 7 and part of the Memory factor after age 7. Because Memory for Sen­ tences also has high loadings on the Verbal Comprehen­ sion factor, it can be used to obtain a Verbal Com­ prehension factor score when the six-subtest short form is administered at any age . Short forms on the SB: FE have the same advantages and disadvantages as they do on the

CHAPTER I I

260

STANFORD-BINET INTELLIGENCE SCALE: FOURTH E DITION

Exhibit 1 1 -2 Worksheet for Computing Factor Scores J_ i-m______ Examinee's name: ______---'D 1 9:... c e . 1 , Date of birth: _____--=-=-=.'-.::.:'---.:. .: 8:..: .: 0__ M 1 1 __ Sex: 6-..:;. Age: _ _ __= - .::_

_

_ _ _ _

_ _ _

_ _

STANDARD SCORE

� � � �

STEPS

( I ) Sum of standard scores on Vocabulary

+ Comprehension + Absurdities (2) Verbal Reasoning (VR) Area SAS (p. 1 83 of Guide for 3 subtests) (3) Short-Term Memory (STM) Area SAS (multiply the standard score by 2) (4) Sum of ( 2 ) + (3) (5) Verbal Compre!lension Factor Score (p. 1 87 of Guide for 2 area scores)

+ Comprehension + Absurdities

STANDARD SCORE

5 Pattern Analysis (AIVR) 1 1 Matrices (AIVR)

(3) Verbal Comprehension Factor Score (p. 1 87 or 1 88 of Guide for I area

3 Quantitative (QR) 2 Bead Memory (STM)

score)

STEPS

Ages 15 through 18-23 STANDARD SCORE

+ Comprehension + Verbal Relations (2) Verbal Reasoning (VR) Area SAS (p. 1 83 of Guide for 3 subtests)

Area SAS (p. 1 84 of Guide for 2 subtests) (3) Quantitative Reasoning (QR) Area SAS (multiply the standard score by 2) (4) Short-Term Memory (STM) Area SAS (multiply the standard score by 2) (5) Sum of (2) + (3) + (4) (6) Nonverbal Reasoning/Visualization Factor Score (p. 1 87 or 1 8 8 of Guide for 3 area scores)

SUBTEST

(2) Verbal Reasoning (VR) Area SAS (p. 1 83 of Guide for 3 subtests)

STEPS

� � � �

5 Pattern Analysis (AIVR) 9 Copying (AIVR) 3 Quantitative (QR) 2 Bead Memory (STM)

Ages 12 through 18-23

STEPS

( I ) Sum of standard scores on Vocabulary

( I ) Sum of standard scores on Vocabulary

STANDARD SCORE

SUBTEST

(2) AbstractIVisual Reasoning (AIVR)

Vocabulary (VR) 6 Comprehension (VR) 7 Absurdities (VR)

Vocabulary (VR) 6 Comprehension (VR) 14 Verbal Relations (VR)

Ages 2 through 1 1

STEPS

STANDARD SCORE

SUBTEST

NONVERBAL REASONING/VISUALIZATION

( 1 ) Sum of standard scores on Pattern Analysis + Copying

Ages 8 through 1 4 SUBTEST

_

_ _ _ _ _

(3) Verbal Comprehension Factor Score (p. 1 88 of Guide for 1 area score)

Ages 2 through 7

Vocabulary (VR) 6 Comprehension (VR) 7 Absurdities (VR) 4 Memory for Sentences (STM)

I ,_--'N_ 1 9--'7_______ o_ v._-' 8_ _ = ____ S..:.: m.::.ith -=B:..:i1:.:I-.= --

_ _

_

_ _ _

VERBAL COMPREHENSION

SUBTEST

_ _

_ _ Examiner's name:

Date:

( I ) Sum of standard scores on Pattern Analysis + Matrices (2) AbstractIVisual Reasoning (AIVR) Area SAS (p. 1 84 of Guide for 2 subtests) (3) Quantitative Reasoning (QR) Area SAS (multiply the standard score by 2) (4) Short-Term Memory (STM) Area SAS (multiply the standard score by 2) (Exhibit continues next page)

26 1

5B: FE 5UBTE5T5

Exhibit 1 1 ·2 (cont.) (3) Memory Factor Score (p. 1 87 of Guide for 1 area score)

(5) Sum of (2) + (3) + (4) (6) Nonverbal Reasoning/Visualization Factor Score (p. 1 88 of Guide for 3

Ages 8 through 18-23

area scores)

SUBTEST

M EMORY Age 7 SUBTEST

STANDARD SCORE

8 Memory for Digits (STM) 10 Memory for Objects (STM) STEPS

( I ) Sum of standard scores on Memory for

Digits + Memory for Objects (2) Short-Term Memory (STM) Area SAS (p. 1 86 of Guide for 2 subtests)

Wechsler scales (see Chapter 6) . They should be used primarily for screening purposes.

Content Area Scores The SB: FE provides four content area scores: Verbal Reasoning, AbstractlVisual Reasoning, Quantitative Rea­ soning, and Short-Term Memory. Because these area scores are not supported by factor analysis, they should not be used for most interpretive purposes. The four areas must be used , however, to obtain the Composite Score and to obtain the factor scores. The factor scores are preferred for interpretative purposes .

STANDARD SCORE

4 Memory for Sentences (STM) 8 Memory for Digits (STM) 10 Memory for Objects (STM) STEPS

( 1 ) Sum of standard scores on Memory for Sentences + Memory for Digits + Memory for Objects (2) Short-Term Memory (STM) Area SAS (p. 1 86 of Guide for 3 subtests) (3) Memory Factor Score (p. 1 87 or 1 88 of Guide for 1 area score)

and psychoeducational usefulness . At ages 2 through 6, an 8-subtest battery is recommended: the six subtests that run throughout the battery plus Absurdities and Copying. At ages 7 through 11, the lO-subtest contains these eight sub­ tests plus Memory for Digits and Memory for Objects . At age 12, Matrices i s substituted for Copying, and at age 15 , Verbal Relations is substituted for Absurdities. The routine use of these batteries is recommended for all children . Administration time will be considerably shortened, and the use of a standard battery will facilitate interpretive and placement decisions . The remaining three subtests - Paper Folding and Cutting, Number Series, and Equation Build­ ing - can be used for special diagnostic purposes.

Profile Sheet After the scale has been scored and factor scores com­ puted, a profile sheet can be used to plot the examinee's scores (see Exhibit 11-3). This profile sheet has entries for the Composite Score , factor scores, and subtest scores. The profile sheet provides a visual picture of the exam­ inee's performance.

Recommended Subtest Battery Table 11-10 shows the recommended subtest battery for each age level of the S B : FE. The 8-subtest and lO-subtest combinations shown in Table 11-10 were selected on the basis of factor analysis, time considerations , and clinical

SS: FE SU STESTS

Vocabulary The Vocabulary subtest contains 46 items divided into a picture vocabulary section (items 1 to 14) and an oral vocabulary section (items 15 to 46) . For the picture items, the child is asked to name the picture or give the most pertinent detail of the picture . For the oral vocabulary section, the child is asked to explain orally the meaning of each word . The subtest is administered throughout all ages covered by the scale . Each word is scored l or O .

262

CHAPTER I I

STANFORD-BINET INTELLIGENCE SCALE: FOURTH EDITION 9-74539

*

S TANOAAO AGE SCORE

RAW SCORE

Verbal Reasoning

1 Vocabulary

Comprehension

RECORD BOOKLET

Absurdities

1 4 Verbal Relations

Stanford-Binet I ntelligence Scale: Fourth Edition

Sum 01 Subtest SAS's Verbal Reasoning SAS

Abstract/Visual Reasoning

5 Pattern AnalYSIS

Name

9 CopyIng "

Sex

Ethnicity

NA

H

B

WfNH

OfAA

PI

YEAR

Date 01 Testing

Other

Matnces

t3 Paper FoldIng & Cu tt I ng Sum of Subtest SAS's

_ _ _ _ _ _ _

MONTH

AbstractlVisual Reasoning SAS

DAY

Quantitative Reasoning

3 Quantitative

1 2 Number Senes

15 Equation Building

Birth Date

Sum of Subtest SAS's

Age

Quantitative Reasoning SAS

School

Short· Term Memory

2 Bead Memory

Grade

4 Memory For Sentences

E xamin er

8 Memory For DIgIts

to Memory For Obiecis

Father's Occupation:

Sum of Subtest SAS's

Mother's Oc cupation :

Short-Term Memory SAS Sum of Area SAS's

FACTORS AFFECTING TEST P E R FORMANCE

Overall Rating 01 Conditions

*

COMPOSITE SCORE

Test Composite Optimal

Good

Average

Detrimental

Seriously detrimental

Partial Composite based on

* Be

_ _ _ _ _ _ _ _ _ _ _ _ _ _

sure that all Standard Age Scores

(SAS's) are based on the tables In the Guide

AnenllOn a) Absorbed by lask

Easily OtslraCleO

with the number 9-74502 on the cover.

Reacllons DUring lesl Perlormance a) Normal aCllvily level

I _ L__ ,- ,-_ ,-- �

b) Inlllates activity c) QUick to respond

Abnormal activity level Walts to be tolO Urging needed

Emolional lndependence a) Socially confident

Insecure

b) Realistically self-confident

Distrusts own aOllity

c) Comfortable In adult company

III-ai-ease

d) Assured

AnxIOus

Problem-Solving Behavior a) Persistent

Gives up eaSily

b) Reacts to 'allure realistically

Reacts to lallure unrealistically

c) Eager 10 conllnue

Seeks

10

termlnale

d ) ChallengeCl by harCl tasks

Prelers only easy lasks

a) Needs minimum of commendation

encQuragemem

Independence 01 Examiner Support

eeds constanl praise and

Expressive Language a) Excellent articulation

Very poor anicuiaiion

Rober! L . Thorndike

Receptive Language a) Excellent sound diSCrimination Was II difficult

10

Very poor sound dlSCflmlnallon

Elizabelh P. Hagen Jerome M . Saltier

eSlaoltsh rappor1 With thiS person ? Easy

Figure 1 1 -2.

Th� Rlversld� Publishing Company

,-_ ,-'� _ -'_ -'-_ _ I

Cover page of Stanford-Binet I ntel ligence Scale: Fourth Edition record booklet. Reprinted with permission of

the Riverside Publishing Company from Stanford-Binet Intelligence Scale: Fourth Edition by R. L. Thorndike, E. RIVERSIDE PUBLISHING COMPANY, 8420 W Bryn Mawr Avenue, Chicago, IL 6063 1 . Copyright 1 986_

P.

Hagen, 1 . M. Sattler. THE

263

SB: FE SUBTESTS Table 1 1 -9 Physical Abi lities Necessary and Adaptable for S tanford-Binet I ntelligence Scale: Fourth Edition

Subtest

Vision

Hearing

Oral speech

Arm-hand use

Vocabulary Comprehension Absurdities Verbal Relations Pattern Analysis Copying Matrices Paper Folding and Cutting Quantitative Number Series Equation Building Bead Memory Memory for Sentences Memory for Digits Memory for Objects

Xr Xr X Xr X X X X X Xa Xa X

Xa Xa Xa Xa Xa Xa Xa Xa Xa Xa Xa Xa X X Xa

Xa Xa Xa Xa

Xw Xw Xw Xw X X Xa Xa Xw Xw Xw X Xw Xw Xw

X

XO Xo Xa Xa Xa Xa Xa Xa

Note. The code is as follows: X - This ability i s required. Adaptation i s not feasible if this function is absent or more than mildly impaired. Xa - This ability i s required for standard administration, but the subtest is adaptable. Xo - Examinees who are able to speak can say their answers. Xr - Examinees who are able to read can be shown the questions. If the examinee cannot read, hearing i s necessary. If neither the ability to read nor the ability to hear is present, the subtest should not be administered. X w - Examinees who are able to write can write their answers.

Rationale. The rationale presented for the WISC-R Vo­ cabulary subtest generally applies to the oral vocabulary section of the SB: FE (see page 151) . Formal education is less likely to be an influence for picture vocabulary than for oral vocabulary. Experience is likely to be the major contributing factor to the vocabulary development of pre­ school children. The purpose of the picture vocabulary section is to see whether the child correctly identifies a picture by any of its appropriate names. V isual perception is involved, as are verbal retrieval and word recall abilities . In complex pic­ tures, the child must appreciate the main element in the picture . Correct association of word with object, the abil­ ity to vocalize the word , and the ability to comprehend the spoken word are all important . These skills represent emergent aspects of language use. The emphasis is on comprehending what the spoken word stands for and not on articulation ski l l s . Factor analytic findings. The Vocabulary subtest is a good measure of g (64 percent of its variance may be attributed to g) . The subtest has ample or adequate specif­ icity at ages 2 , 10, 1 2 , 13, 1 6 , 1 7 , and 18-23 to permit spec ific interpretation of its functions . At other ages ,

where its specificity is inadequate . the Vocabulary subtest still can be interpreted as a measure of verbal comprehen­ sian . Vocabul a ry contributes substantially to the Verbal Comprehension factor at all ages (Mdn loading = . 77) . Reliability and correlational highlights. Vocabulary is a reliable subtest ( r.u = . 87 ) . It correlates more h ighly with Comprehension (r = . 7 3 ) than with any other sub­ test . It also correlates highly with the Composite Score

(r

=

.81).

Administrative a n d interpretive considerations. The administrative and interpretive considerations presented for the WISC-R Vocabulary subtest generally apply to the SB: FE Vocabulary subtest (see page 15 1). On the SB: FE, however, different administrative procedures are used for the picture vocabulary and oral vocabulary sections. Study carefully the "General Guidelines for Scoring the Vocabulary Test" (Appendix A) in the Stanford-Biner: Fourth Edition Guide so that you will know which re­ sponses need further inquiry. The examples indicate that many responses should be queried. Unlike the WISC-R, on which vocabulary responses may be scored 2 , 1 , or 0, the SB: FE uses a 1 or 0 system in an attempt to make

CHAPTER I I

264 •

STANFORD-BINET INTELLIGENCE SCALE: FOURTH EDITION

Exhibit I 1 -3 Profile Sheet for the Stanford-Binet I ntel ligence Scale: Fourth Edition

Name:

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

Date of birth:

_ _ _ _ _ _ _ _ _ _ _ _ _ _

Summary scores

Sex:

_ _ _ _

Date of examination:

CA:

_ _ _ _

Examiner's name: _

_ _ _ _ _ _ _ _

______

_ _

Nonverbal Reasoning! Visualization

Verbal Comprehension

Memory

s::

.2

.

""'"

s:: l2 � 'i:: '"

Nonverbal Reasoning/Visualization a . Verbal skills are better developed than performance skil l s . b . Auditory-vocal processing skills are better devel­ oped than visual discrimination skill s . c . Knowledge acquired through accumulated experi­ ence is better developed than immediate problem-solving ability. d . Practical tasks are more difficult than nonpractical tasks. e. Performance deficits may exist, including deficits in copying skills. f. Limitations in visual-motor integration may be influ­ encing performance. 2 . Illustrative Hypotheses for Verbal Comprehension

> Memory a. Verbal skills are better developed than short-term memory skil l s . b . Long-term memory is better developed than short­ term memory. c . Limitations in attention , possibly due to anxiety and distractibility, may be influencing performance . 3 . Illustrative Hypotheses for Nonverbal Reasoning/Vi­

sualization > Verbal Comprehension a. Performance skills are better developed than verbal skills. b . Visual-motor discrimination skills are better devel­ oped than auditory-vocal processing skill s . c . Immediate problem-solving ability is better devel­ oped than knowledge acquired as a result of accumulated experience . d . Reading skill and performance o n academic tasks are not satisfactory. e . A language deficit may exist . f. Limitations i n auditory conceptual skills and auditory processing skills may be influencing performance . 4 . Illustrative Hypotheses for Nonverbal Reasoning/Vi­

sualization > Memory a . Nonverbal reasoning skills are better developed than short-term memory skills. b. V isual-motor discrimination skills are better devel­ oped than memory skills. c . Immediate problem-solving ability is better devel­ oped than memory skill s . d . Limitations in attention, possibly due t o anxiety and distractibility, may be influencing performance.

28 1

b . Short-term memory is better developed than long­ term memory. c . Encoding ability (such as sequencing ability) is better developed than knowledge acquired as a result of accumu­ lated experience . d . Reading skill and performance on academic tasks are not satisfactory. e . A language deficit may exist. 6. Illustrative Hypotheses for Memory > Non verbal

Reasoning/Visualization a. Short-term memory skills are better deveioped than nonverbal reasoning and visualization skills . b . Short-term memory skills are better developed than visual-motor discrimination skills. c . Encoding ability (such as sequencing ability) is better developed than immediate problem-solving ability. d . Practical tasks are more difficult than nonpractical tasks. e. Performance deficits may exist, including deficits in copying skills. f. Limitations in visual-motor integration may be influ­ encing performance .

COMPARISONS BETWEEN SU BTESTS THAT CAN G U I D E I NTERPRETATIONS

Once differences between subtest scaled scores have been found to be significant (see Table C-45 in Appendix C) , the findings must be translated into meaningful descriptions. Interpret ing differences between subtest scores is not an easy matter. The material in this chapter will help you in making interpretations . Table 11-16 charts suggested abili­ ties and factors associated with the 15 SB: FE subtests . You should view the interpretations that follow as hypotheses that may prove to be useful in the evaluation of a child's performance . These hypotheses, however, need to be fur­ ther investigated through a study of the child's entire test performance and clinical history. The following list of subtest comparisons does not in­ clude all possible comparisons , nor does it reflect all possible interpretations. On the basis of these examples, you should be able to develop other comparisons and formulate other interpretations. All hypotheses should be

treated as tentative. Conclusions derivedfrom profile anal­ ysis should never be referred to as definitive findings. (See

5 . Illustrative Hypotheses for Memory > Verbal Com­ prehension

Table C-52 in Appendix C for a summary of interpretive rationales and suggested remediation activities for the SB: FE subtests . ) All hypotheses must be developed on the

a. Short-term memory skills are better developed than verbal skill s .

basis ofboth significant differences between subtest scores and the absolute values of the subtest scores. Thus scaled

282

CHAPTER I I

STANFORD-BINET INTELLIGENCE SCALE: FOURTH EDITIO N

Table 1 1 - 1 5 Short Forms, Factors, and Composite Score Ranges i n the Stanford-Binet I ntelligence Scale: Fourt h Edition

Short forms, factors, and composite scores

+4 (164-149)

+3 (148-133)

+2 (J32- 1 l 7)

2-0-0 to 1 2- 1 1 - 1 5 2-0-0 to 1 6- 1 1 - 1 5 2-0-0 to 1 5- 1 1 - 1 5

1 2- 1 1 - 1 6 to 1 7- 1 1 - 1 5 1 6- 1 1 - 1 6 to 23- 1 1 - 1 5 1 5- 1 1 - 1 6 to 23- 1 1 - 1 5

1 7 - 1 1 - 1 6 to 23- 1 1 - 1 5

2-0-0 to 1 5- 1 1 - 1 5 2 -0-0 to 1 6- 1 1 - 1 5 2-0-0 to 23- 1 1 - 1 5

1 5- 1 1 - 1 6 to 23- 1 1 - 1 5 1 6- 1 1 - 1 6 to 23- 1 1 - 1 5

2-0-0 to 1 7- 1 1 - 1 5 2-0-0 to 23- 1 1 - 1 5

1 7- 1 1 - 1 6 to 23- 1 1 - 1 5

+1 (1 16-101)

Short Form

2-Subtest short form 4-Subtest short form 6-Subtest short form

-

-

-

-

-

Factor

Verbal Comprehension factor Nonverbal Reasoning/visualization factor Memory factor

-

-

-

-

-

-

-

Composite

Composite Score (factor scores) Composite Score (full battery)

-

-

-

-

-

Note. Subtest short forms are as follows: 2 subtests: Vocabulary and Pattern Analysis; 4 sub!ests: Vocabulary, Bead Memory, Quantitative, and Pattern Analysis; 6 subtests: Vocabulary, Bead Memory, Quantitative, Memory for Sentences, Pattern Analysis, and Comprehension. See Table 1 1 -8 for sub tests that form the factor scores.

scores of 50 or higher never reflect absolute weaknesses, and scaled scores of 49 or lower never reflect absolute strengths . COMPARING VERBAL COMPREHENSION FACTOR SUBTESTS 1 . Vocabulary (V) and Comprehension (C). Both sub­ tests involve verbal processing , but in somewhat different contexts . V > C : High Vocabulary and low Comprehension may suggest inability to use verbal ability and general knowl­ edge fully in life situations and therefore may indicate impaired judgment. V < C: Low Vocabulary and high Comprehension may suggest limited concept formation, except when concep­ tualizing ability is applied to solving problems in the social world. 2 . Vocabulary (V) and Verbal Relations (VR). Both Vocabulary and Verbal Relations measure level of abstract thinking and ability to form concepts, but Verbal Relations is a better measure of these abilities. VR > V : High Verbal Relations and low Vocabulary may suggest good ability to do abstract thinking but lim­ ited ability to understand the meaning of word s . V R < V : Low Verbal Relations and high Vocabulary may suggest difficulty in forming abstract concepts but good ability to understand the meaning of individual words. 3. Verbal Relations (VR) and Comprehension (C). Ver-

bal Relations and Comprehension both involve, in part, conceptualizing skills. Verbal Relations usually requires a single word response, whereas Comprehension requires an extended response that interrelates a set of ideas (that is, propositional thinking) . VR > C : High Verbal Relations and low Comprehen­ sion may suggest good abstract thinking but difficulty in applying conceptualizing ability to problem solving in the social world . This pattern also may suggest a verbal­ expressive deficit involving propositional thinking. VR < C: Low Verbal Relations and high Comprehen­ sion may suggest a deficit in verbal concept formation relative to real-world conceptualizing ability. 4 . Comprehension (C) and Absurdities (A). Com­ prehension and Absurdities both reflect social intelligence to some extent. The comparison relates knowledge of social conventions (Comprehension) to the ability to dis­ cern incongruities in visually presented material (Absurd­ ities) . A > C : High Absurdities coupled with low Com­ prehension may suggest keen observational skills, but lim­ ited understanding of social conventions . A < C : Low Absurdities coupled with high Compre­ hension may suggest an understanding of social conven­ tions but difficulty in using observational skills. 5. Vocabulary (V) and Memory for Sentences (MS). Vocabulary and Memory for Sentences both involve verbal processing (particularly from 2 to 7 years of age). Com­ paring the two subtests may provide an index of the relative

283

COMPARISONS BETWEEN SUBTESTS THAT CAN GUIDE INTE RPRETATIONS

-] (1 00-84)

-2

(83-68)

-3 (67-52)

-4 (51-36)

2-0-0 to 2-3- 1 5 2-0-0 to 2-7- 1 5 2-0-0 t o 2-7- 1 5

2-3- 1 6 to 3-3- 1 5 2-7- 1 6 to 3-7 - 1 5 2-7- 1 6 to 3-3- 1 5

3-3- 1 6 to 4-7- 1 5 3-7- 1 6 to 4- 1 1 - 1 5 3-3- 1 6 to 4-3- 1 5

4-7- 1 6 to 23- 1 1 - 1 5 4- 1 1 - 1 6 to 23- 1 1 - 1 5 4-3- 1 6 to 23- 1 1 - 15

2-0-0 t o 2-3- 1 5 2-0-0 t o 2- 1 1 - 1 5

2-3 - 1 6 to 3-3- 1 5 2- 1 1 - 1 6 to 3-7 - 1 5 6- 1 1 - 1 6 to 7- 1 1 - 1 5

3-3- 1 6 to 4-3- 1 5 3-7- 1 6 to 4- 1 1 - 1 5 7 - 1 1 - 1 6 to 8- 1 1 - 1 5

4-3- 1 6 to 23- 1 1 - 1 5 4- 1 1 - 1 6 to 23- 1 1 - 1 5 8- 1 1 - 1 6 to 23- 1 1 - 1 5

2-0-0 t o 2-7- 1 5 2-0-0 t o 2-7- 1 5

2-7- 1 6 to 3-3- 1 5 2-7- 1 6 to 3-3- 1 5

3-3- 1 6 to 4-3- 1 5 3-3- 1 6 to 4-3- 1 5

4-3- 1 6 to 23- 1 1 - 1 5 4-3- 1 6 to 23- 1 1 - 1 5

ShorT forms, factors, and composiTe scores Short Form

2-Subtest short form 4-Subtest short form 6-Subtest short form Factor

-

Verbal Comprehension factor Nonverbal Reasoning/Visualization factor Memory factor Composite

balance between long-term memory (Vocabulary) and short-term memory (Memory for Sentences) . V > M S : High Vocabulary and low Memory for Sen­ tences may suggest that long-term memory is better devel­ oped than short-term memory. V < M S : Low Vocabulary and high Memory for Sen­ tences may suggest that short-term memory is better de­ veloped than long-term memory. 6 . Comprehension (C) and Memory for Sentences (MS). Both comprehension and Memory for Sentences involve verbal processing (particularly from 2 to 7 years of age) . This comparison provides an estimate of verbal rea­ soning versus short-term verbal memory. C > M S : High Comprehension and low Memory for Sentences may suggest that verbal reasoning is better de­ veloped than short-term verbal memory. C < M S : Low Comprehension and high Memory for Sentences may suggest that verbal reasoning is not as well developed as short-term verbal memory. COMPARING NONVERBAL REASONING/ VISUALIZATION FACTOR SUBTESTS 7 . Pattern Analysis (PA) and Copying (CP). This com­ parison relates visual-spatial ability to visual-motor ability. PA > C P : H igh Pattern Analysis and low Copying may suggest adequate visual-spatial ability but inadequate visual-motor ability. PA < CP: Low Pattern Analysis and high Copying may suggest inadequate visual-spatial abi lity but adequate visual-motor ability. 8 . Pattern Analysis (PA) and Matrices (M). Both Pat­ tern Analysis and Matrices involve nonverbal reasoning ability. In Pattern Analysis, however, the reasoning in-

Composite Score (factor scores) Composite Score (full battery)

volves analysis and synthesis, whereas in Matrices the reasoning is more analogic . PA > M : High Pattern Analysis and low Matrices may suggest adequate analysis and synthesis skills but inade­ quate analogic reasoning skill s . PA < M : Low Pattern Analysis and high Matrices may suggest inadequate analysis and synthesis skills but ade­ quate analogic reasoning sldlls. 9 . Pattern Analysis (PA) and Paper Folding and Cut­ ting (PF). Both Pattern Analysis and Paper Folding and Cutting involve visual-spatial abil ity, but Paper Folding and Cutting is a purer measure of this abil ity. In Pattern Analysis a concrete representation of the stimulus is pro­ vided, whereas in Paper Folding and Cutting the examinee must discern how the folded and cut paper would look if it were unfolded. PA > PF: High Pattern Analysis and low Paper Folding and Cutting may suggest that visual-spatial abil ity is more adequate when the stimulus is concrete than when it is less conc rete . PA < PF : Low Pattern Analysis and high Paper Folding and Cutting may suggest that visual-spatial ability is less adequate when the stimulus is concrete than when it is less concrete. 1 0 . Pattern Analysis (PA) and Quantitative (Q). This comparison relates nonverbal spatial reasoning to numer­ ical reasoning . PA > Q: High Pattern Analysis and low Quantitative may suggest adequate nonverbal spatial reasoning ability but inadequate numerical reasoning ability. PA < Q: Low Pattern Analysis and high Quantitative may suggest inadequate nonverbal spatial reasoning abil­ ity but adequate numerical reasoning ability.

STANFORD-BINET INTELLIGENCE SCALE: FOURTH EDITION

CHAPTER I I

284

Table 1 1 - 1 6 Suggested Abilities or Factors Associated with the 1 5 Subtests of the Stanford-Binet Intelligence Scale: Fourth Edition

�

c-

oS

::s �

8 �

V

Q ·

s::

-.. 3.-

9 '" s:: '"

.�

-s::

�

�

� a

::s

'"

�

�

C

['" s::

.g

oS

�

�

1:

�

G:' � 0.0 5 .t:

a ]13

-.. � '-'-

.� '" .Q

�

�

�

t::

�

ii:'

� ·

0.0

5 �

a

A

CP

A

CP

�

� '"' .r::

0.0 s::

:§

§ .�

� �

c;:j'

0.0

�

�

�

� '" '" '"' s::

�

!:S. c-

�

� � l �

�

�.:i

� .r::

c;:j'

�

.§

�

�

s::

... � c-

� ... '" 1} a:;

.§

M

PF

Q

NS

EB

BM

M

PF

Q

NS

EB

BM

BM

MS

�

ii:

& Q

til

1

:;E

�

�'-'" �t .�

.�

t::l ... � c-

�

�

.�

C5

... � c-

�

�

M

MS MD MO

Attention

MS MD

Auditory memory

MS MD MO

Concentration

Cultural

A

Q

A VR

V

PA

M

PF

PA

M

PF

MD MO

Q Q

Long-term memory NS

Numerical ability

EB

Perceptual organization

Perceptual planning ability

M

PA

PF

Perceptual synthesis

PF

Perception of abstract stimuli

MO

M

A

V(P)

Perceptual reproduction

BM

CP

PA

NS

Q

C V

C

A

V(P) V(O)

CP

C

MS MD MO

Sequencing

MS

Short-term memory

MD MO

Social judgment BM

Q

Spatial perception

Verbal comprehension Little verbal expression

NS

VR

Much �rbal expression

VR

Verbal concept formation

Q

VR

C

BM

VR

C V

EB

PF

A

V

Long verbal questions MD

Q

A PA

CP

Short verbal questions

Verbal reasoning

ED BM BM

PF

Visual-motor coordination

MO

=

Picture Vocabulary, V(O)

Visual memory Working under time pressure

PA Note. V(P)

Perception of meaningful stimuli Psychomotor speed

PA

PA

opportunities at home

Freedom from distractibility

Nonverbal reasoning

M PA

Suggested abilities or factors

=

Oral Vocabulary. M

=

mean of the subtest scaled score or factor.

285

COMPARISONS BETWEEN SUBTESTS THAT CAN GUIDE INTERPRETATIONS

1 1 . Pattern Analysis (PA) and Bead Memory (BM). Pat­ tern Analysis and Bead Memory both involve visual dis­ crimination and spatial relation s . In Bead Memory, however, these skills are measured in the context of a short-term visual memory task, whereas in Pattern Analy­ sis the skills are measured using stimuli that remain in view during the task. Additionally, Pattern Analysis has a strong reasoning component, whereas Bead Memory does not. PA > BM: High Pattern Analysis and low Bead Mem­ ory may suggest that visual discrimination and spatial relations skills are adequate when short-tenn memory is not involved but are inadequate when short-term memory is involved. PA < BM: Low Pattern Analysis and high Bead Mem­ ory may suggest that visual discrimination and spatial relations skills are adequate when short-term memory is involved but are inadequate when short-term memory is not involved. 1 2 . Quantitative (Q) and Number Series (NS). Both Quantitative and Number Series involve numerical reason­ ing ability, but Number Series is a purer measure of reason­ ing ability. Q > NS: High Quantitative and low Number Series may suggest adequate understanding of mathematical con­ cepts but inadequate numerical reasoning ability. Q < NS: Low Quantitative and high Number Series may suggest inadequate understanding of mathematical concepts but adequate numerical reasoning ability. 1 3 . Quantitative (Q) and Equation Building (EB). Both Quantitative and Equation Building involve numerical rea­

soning ability, but Equation Buildi�g requires more logical reasoning and flexibility in rearranging and manipulating mathematical symbols. Q > EB : High Quantitative and low Equation Building may suggest adequate understanding of mathematical con­ cepts but inadequate logical reasoning and flexibility in the use of mathematical symbols. Q < EB: Low Quantitative and high Equation Building may suggest inadequate understanding of mathematical concepts but adequate logical reasoning and flexibility in the use of mathematical symbols .

Auditory memory Visual memory

Figure 1 1 -3.

1 4 . Number Series (NS) and Equation Building (EB). Both Number Series and Equation Building involve nu­ merical reasoning. Equation Building , in addition, in­ volves knowledge of conventional arithmetical operations and flexibility in the use of mathematical symbols . N S > EB: High Number Series and low Equation Building may suggest adequate numerical and logical rea­ soning ability but inadequate knowledge of conventional operations and limited flexibility in the ljse of mathe­ matical symbols. NS < EB : Low Number Series and high Equation Building may suggest inadequate numerical and logical reasoning ability but adequate knowledge of conventional operations and flexibility in the use of mathematical symbols. COMPARING MEMORY FACTOR SUBTESTS 1 5 . Memory for Sentences (MS) , Memory for Digits (MD) , Memory for Objects (MO) , and Bead Memory (BM). The Memory for Sentences, Memory for Digits,

Memory for Objects, and Bead Memory subtests all in­ volve short-tern1 memory. Differences exist among the four subtests , however, in the stimuli used to elicit re­ sponses (meaningful or nonmeaningful) and the process­ ing modality (auditory or visual) (see Figure 1 1 -3). Mem­ ory for Sentences involves meaningful stimuli presented in the auditory modality. Memory for Digits involves non­ meaningful stimuli presented in the auditory modality. Memory for Objects involves meaningful stimuli prentw in th i u�1 modality. Bead Memory involves non­ meaningful stimul i presented in the visual modality. The subtests also load on different factors. Bead Mem­ ory loads on the Nonverbal ReasoninglVisualization fac­ tor, whereas the other three subtests load on the Memory factor. (Memory for Sentences also loads on the Verbal Comprehension factor throughout the scal e . ) On Bead Memory the stimuli are shown as a group (or presented as a gestalt), whereas on the other subtests the stimuli are presented sequentially. MS > MD: High Memory for Sentences and low Mem­ ory for Digits may suggest adequate short-term memory

Meaningful memory

Nonmeaningful memory

Memory for Sentences

Memory for Digits

Memory for Objects

Bead Memory

Classification of four memory subtests on the Stanford-Binet Intelligence Scale: Fourth Edition.

286

CHAPTER I I

for meaningful material but inadequate short-term mem­ ory for non meaningful material . MS < M D : Low Memory for Sentences and high Mem­ ory for Digits may suggest inadequate short-term memory for meaningful material but adequate short-term memory for nonmeaningful material . MS > MO : H igh Memory for Sentences and low Mem­ ory for Objects may suggest that, for meaningful stimuli, short-term auditory memory is adequate but short-term visual memory is inadequate . MS < MO: Low Memory for Sentences and high Mem­ ory for Objects may suggest that, for meaningful stimuli , short-term auditory memory i s inadequate but short-term visual memory is adequate . MD > B M : H igh Memory for Digits and low Bead Memory may suggest that, for nonmeaningful material , short-term auditory memory is adequate but short-term visual memory is inadequate . MD < B M : Low Memory for Digits and high Bead Memory may suggest that, for nonmeaningful material , short-term auditory memory is inadequate but short-term visual memory is adequate . BM > MO: H igh Bead Memory and low Memory for Objects may suggest that short-term visual memory is adequate for nonmeaningful material but inadequate for meaningful material . BM < MO : Low Bead Memory and high Memory for Objects may suggest that short-term visual memory is less adequate for non meaningful material than for meaningful material .

1 6 . Memoryfor Digits - Digits Forward (MD-F) versus Digits Reversed (MD-R). The two components of Memory for Digits - Digits Forward and Digits Reversed - both involve attention . Digits Reversed , however, involves more complex attentional processes . MD-F > MD-R : H igh Digits Forward and low Digits Reversed (differences of 3 or more raw-score points) may indicate that the child did not put forth the extra effort needed to master the more difficult task of recall ing digits in reversed sequence . Alternatively, it may indicate good auditory memory but poor short-term visual memory based on auditory information (a very tentative hypoth­ esis) . MD-F < MD-R: Low Digits Forward and high Digits Reversed may indicate that the child sees Digits Reversed as a challenge rather than as a task involving mere repeti­ tion of numbers.

1 7 . Memory for Sentences (MS) and Memory for Ob­

jects (MO) versus Memory for Digits (MD) and Bead Memory (BM). This comparison contrasts subtests that measure short-term memory for meaningful material

STANFORD-BINET INTELLIGENCE SCALE: F OURTH EDITION

( Memory for Sentences , Memory for Objects) versus short-term memory for nonmeaningful material (Memory for Digits, Bead Memory) (see Figure 11-3 ) . MS, MO > MD , B M : High Memory for Sentences and Memory for Objects coupled with low Memory for D igits and Bead Memory may suggest that the examinee has better short-term memory when the material is meaningful than when it is nonmeaningful . MS, MO < MD , B M : Low Memory for Sentences and Memory for Objects coupled with high Memory for Digits and Bead Memory may suggest that the examinee has better short-term memory when the material is non­ meaningful than when it is meaningful . 1 8 . Memory for Sentences (MS) and. Memory for Digits

(MD) versus Memory for Objects (MO) and Bead Memory (BM). This comparison contrasts short-term auditory memory with short-term visual memory (see Figure 11-3) . M S , MD > MO , B M : H igh Memory for Sentences and Memory for Digits and l ow Memory for Objects and Bead Memory may suggest that the examinee's short-term au­ ditory memory is better than his or her short-term v isual memory. MS, MD < MO , B M : Low Memory for Sentences and Memory for Digits and high Memory for Objects and Bead Memory may suggest that the examinee's short-term au­ ditory memory is poorer than his or her short-term visual memory. 1 9 . Bead Memory (BM) versus Memory for Sentences

(MS) , Memory for Digits (MD) , and Memory for Objects (MO). This comparison contrasts a subtest that uses simul­ taneously presented stimuli with those that use sequen­ tially presented stimu l i . BM > M S , M D , MO : High Bead Memory and low Memory for Sentences, Memory for Digits, and Memory for Objects may suggest that the examinee's short-term memory is better when the stimuli are presented simul­ taneously than when they are presented sequentially. BM < MS, M D , M O : Low Bead Memory and high Memory for Sentences, Memory for Digits, and Memory for Objects may suggest that the examinee's short-term memory is poorer when the stimuli are presented simul­ taneously than when they are presented sequentially. COMPARING VERBAL COMPREHENSION, NONVERBAL REASONINGIVISUALIZATION, AND MEMORY FACTOR SUB TESTS 20. Verbal Relations (VR) and Pattern Analysis (PA). Verbal Relations and Pattern Analysis both reflect abstract reasoning ability. They both require the abstraction of relations among stimulus items. VR > PA : High Verbal Relations and low Pattern Anal-

COMPARISONS BETWEEN SUBTESTS THAT CAN GUIDE I t'JTERPRETATIONS

ysis may suggest that abstract reasoning abil ity i. better with verbal stimuli than with nonverbal stimul i . VR < PA : Low Verbal Relations and high Pattern Anal­ ysis may suggest that abstract reasoning ability is better with nonverbal stimuli than with verbal stimuli .

2 1 . Comprehension (C) and QuantiTaTive (Q) versus

Vocabulary (V) , Verbal Relations (VR) , and Memory for Digits (MD). This comparison contrasts subtests that have relatively long verbal questions (C omprehension and Quantitative) with those that have relatively short verbal questions (Vocabulary, Verbal Relations , Memory for Digits ) . C , Q > V, V R , MD: High Comprehension and Quan­ titative coupled with low Vocabulary, Verbal Relations, and Memory for Digits may suggest that the child per­ forms better when the verbal stimuli are long than when they are short. The child may put forth more effort to attend to verbal material that is of a relatively long dura­ tion . The pattern may also reflect the child's ability to benefit from the contextual cues contained in the longer questions . C , Q < V, VR, MD: Low Comprehension and Quan­ titative coupled with high Vocabulary, Verbal Relations, and Memory for Digits may suggest that the child per­ forms better when verbal stimuli are short than when they are long. The child may put forth more effort to attend to verbal material that is of a relatively short durat ion . This pattern also may suggest an auditory processing deficit associated with deriving meaning from spoken language . 2 2 . Vocabulary (Oral) (V(O)) , Comprehension (C) , and

Verbal Relations (VR) versus Absurdities (A) , Quantitative (Q) , and Number Series (NS). This compari on contrasts

subtests that require a fair amount of verbal expression (Vocabulary [Oral ] , Comprehension and Verbal Relations) with those that require relatively little verbal expression (Absurdities , Quantitative , and Number Series) . V(O) , C , VR > A, Q, NS: High Vocabulary (Orai ) , Comprehension, and Verbal Relations coupled with low Absurditie s , Quantitative, and Number Series may suggest that the child performs better when the tasks require a fair amount of verbal expression than when they require rela­ tively little . One possibility is that the child may put forth an extra effort in situations that require verbal expression . V(O) , C , VR < A , Q , S: Low Vocabulary (Oral), Comprehension , and Verbal Relations coupled with high Absurdities , Quantitative , and Number Series suggest that the child performs better when the tasks require relatively little verbal expression. One possibil ity is that the child may put forth effort only when tasks require minimal verbal effort. Additionally, this pattern may be associated

287

with commun ication problems or shyness associated with speaking in relatively long sentences.

2 3 . AbsurdiTies (A) versus PaTTern Analysis (PA) , Ma­ Trices (M) , and Paper Folding and CUTTing (PF). This comparison contrasts a subtest that contains relatively meaningful perceptual stimul i (Absurdities) with those that have relatively abstract perceptual stimuli (Pattern Analysis , Matrices , and Paper Folding and Cutting) . (Ma­ trices contains some meaningful stimuli as wel l , but pri­ marily at the younger ages . ) A > PA , M , PF : High Absurdities coupled with low Pattern Analysis, Matrices , and Paper Folding and Cut­ ting may suggest that the child performs better when the visual stimuli are meaningful than when they are abstract or nonmeaningful. A < PA , M , PF : Low Absurdities coupled with high Pattern Analys i s , Matrices , and Paper Folding and Cut­ ting may suggest that the child performs better when the visual stimuli are abstract than when they are concrete .

24. Vocabulary (V) , Verbal Relations (VR) , and Quan­ titative (Q) versus Bead Memory (BM) , Memory for Sen­ tences (MS) , Memory for Digits (MD) , and Memory for Objects (MO). This comparison contrasts subtests that involve long-term memory (Vocabulary, Verbal Relations, and Quantitative) with those that involve primarily short­ term memory (Bead Memory, Memory for Sentences, Memory for Digits , and Memory for Objects) . V, VR, Q > B M , M S , MD, MO: High Vocabulary, Verbal Relations, and Quantitative coupled with low Bead Memory, Memory for Sentence , Memory for Digits, and Memory for Objects may suggest that the examinee per­ forms better on tasks requiring long-term memory than on those requiring short-term memory. Y, VR, Q < B M , M S , M D , MO : Low Vocabulary, Verbal Relations, and Quantitative coupled with high Bead Memory, Memory for Sentences , Memory for Digits, and Memory for Objects may suggest that the examinee per­ forms better on tasks requiring short-term memory than on those requiring long-term memory.

25 . PaTTern Analysis (PA) , Paper Folding and Cutting (PF) , and Bead Memory (BM) versus Quantitative (Q) and Matrices (M). This is a comparison that distinguishes subtests that may tap spatial visualization processes from subtests with perceptual tasks that do not involve these processes. PA , PF, BM > Q , M : H igh Pattern Analys i s , Paper Folding and Cutting, and Bead Memory coupled with low Quantitative and Matrices may suggest that the examinee performs better when perceptual tasks require spatial visu­ alization than when the tasks do not .

288

STANFORD-BINET I NTELLIGENCE SCALE : FOURTH EDITION

CHAPTER I I

PA , PF, BM < Q , M : Low Pattern Analysi s , Paper Folding and Cutting, and Bead Memory coupled with high Quantitative and Matrices may suggest that the examinee performs more poorly when perceptual tasks require spa­ tial visualization than when they do not . ADDITIONAL SUGGEST I O N S F O R

The Vocabulary, Comprehension , Memory for Sen­ tences, Memory for Digits, Verbal Relations , and Number Series subtests may be a useful screening tool for measur­ ing the cognitive skills of visually impaired childre n . The examinee's performance on the Pattern Analysis subtest may provide some clues about how he or she performs under time pressure. •

•

U S I N G A N D I NTERPRETING T H E S8: F E

The fol lowing additional suggestions for interpreting the SB: FE apply only in cases where other information is available to confirm the hypothesis . Because the SB: FE is essentially a new test, the suggestions below must be viewed in a tentative manner. They must be confirmed by research investigations as well as clinical experience. Some subtests may be less culturally loaded than oth­ ers . Candidates i nclude Paper Folding and Cutting, Matrices, Pattern Analysis, Number Series, and Bead Memory. Some subtests may be especially sensitive to brain damage that affects conceptual processes or visualization or flexibility in rearranging material s . Possible candidates are Pattern Analysis, Matrices, Paper Folding and Cut­ ting, Verbal Relations, and Equation Building. Other sub­ tests may be sensitive to brain damage that affects short­ term memory (Bead Memory, Memory for Sentences , Memory for Digits, Memory for Objects) , long-term memory (Vocabulary, Verbal Relations, Quantitative), or judgment and reasoning (Comprehension , Absurdities, Matrices ) . The four memory subtests - Bead Memory, Memory for Sentences, Memory for Digits , and Memory for Ob­ jects - may prove to be especially helpful in assessment of learning disabled children. These subtests permit com­ parisons of simultaneous and successive processing, au­ ditory and visual processing, and meaningful and non­ meaningful processing . The three numerical ability subtests - Quantitative, Number Series, and Equation Building - provide a power­ ful measure of quantitative skills. These subtests may be especially useful in counseling students about their mathe­ matical talents . Four subtests provide potentially useful information about skills in spatial perception - Pattern Analysis, Copy­ ing, Paper Folding and Cutting, and Bead Memory. These subtests may prove to be useful in identifying children with visual-spatial talents . The Nonverbal ReasoninglVisualization factor sub­ tests together with the Memory for Objects subtest may be a useful screening tool for measuring the cognitive skills of hearing-impaired children. •

•

•

•

•

•

ASSETS OF T H E S8: F E

The SB: FE i s a wel l standardized test, with excellent internal-consistency reliability and adequate concurrent validity. Of the 15 subtests in the scal e , 12 are recom­ mended for routine use. These 12 subtests form three factors : Verbal Comprehension (Vocabulary, Compre­ hension, Absurdities [at ages 2 through 14] , Memory for Sentences [at ages 2 through 7] , and Verbal Relations [at a g e s 1 5 t h r o u g h 1 8 -2 3 ] ) , Non verbal Reasoning/ Visualization ( Pattern Analysis , Copying [at ages 2 through 11] , Quantitative, Bead Memory, and Matrices [at ages 12 through 18-23] ) , and Memory (Memory for Sen­ tences [at ages 8 through 18-23] , Memory for Digits, and Memory for Objects) . The three remaining subtests can be considered as supplementary, to be used for special pur­ poses (Paper Folding and Cutting, Number Series, and Equation Building) . The division of the scale into two factors at lower age levels (2 through 6 years) and three factors at elementary and high school years is especially helpful in clinical and psychoeducational work and aids in the assessment of brain-behavior relationships. The fol­ lowing are the assets of the SB: FE :

1 . Good validity. The SB: FE has adequate concurrent validity, demonstrated by correlations with a variety of ability measures. 2 . High reliabilities. The internal consistency rel i­ abilities of the Composite Score and factor scores are extremely h igh (Mdn rxx's range from .91 to .97) , with standard errors of measurement of less than 5 points on the Composite and factor scores. Because the Technical Man­ ual provides reliability data, standard errors of measure­ ment, and intercorrelations of subtest scores by one-year age intervals as wel l as median reliabilities, the scale's properties can be evaluated throughout its entire age range . Confidence intervals can be established for the Composite Score for each of the 17 separate age groups , thereby providing estimates that are specifically applicable to each child's chronological age . 3 . Excellent standardization. The standardization pro­ cedures were excellent, with sampling done on the basis of geographical region, community size, ethnic group, age , gender, and socioeconomic status . Weighting procedures

LIMITATIONS OF THE SB: FE

were used to make the sample conform to the 1980 cen.Sus data. 4. Good administration procedures. The pre crilb ed procedures for administering the SB: FE are excellent. Borderl ine responses are probed , and sample items are provided for many of the subtest s . 5 . Adequate administrative guidelines and test mate­ rials. The administrative guidelines are adequate, al­ though somewhat cumbersome . The use of both letters and numbers for items complicates the use of the scale. The test directions , however, are easy to read . The art work and photos are clear, and the materials are wel l constructed. The easel-format for administering the subtests is a de­ cided advantage. The sample items on many subtests help the examinee understand the task requirements. Finally, the minimal time requirements are an advantage for those examinees who do not perform well under time pressure. 6. Helpful scoring criteria. The criteria for scoring the responses have been carefully prepared. The Vocabulary, Comprehension, and Verbal Relations scoring guidelines, for example, detail the rationale for 1 and 0 scores. A number of examples demonstrate the application of the scoring principles. Many typical responses are scored , and those deemed to require further inquiry are placed under a query section .

L I M I TATI O N S O F T H E S 8 : F E

I . Lack of a comparable battery throughout {he age ranges covered by the scale. The SB : FE fails to provide a

comparable battery of subtests throughout the age ranges covered by the scal e . This is a limitation because it means that scores obtained by children at different ages are based on different combinations of subtests . Only six subtests run throughout the entire scale (Vocabulary, Comprehen­ sion, Pattern Analysis, Quantitative , Bead Memory, and Memory for Sentences) . Two subtests begin at the lowest age levels but stop during the adolescent years (Absurdities and Copy ing) . Four subtests begin at age 7 and run to the upper age level (Matrices, Number Series, Memory for Digits, and Memory for Objects ) . One subtest begins at age 12 and goes to the upper age level (Equation Building) . Only at two ages can a child be given all 15 subtests (ages 12 and 13). The lack of continuity across ages and within subtests makes it difficult to monitor changes in perfor­ mance on individual subtests and may make it difficult to perform longitudinal studies . 2 . Variable range ofscores. The S B : F E fails to provide the same range of Composite Scores , factor scores, or subtest scores throughout the age levels covered by the

289 scal e . For example, Composite Scores of up to 164 can be obtained from ages 2-0-0 to 12-11-15 . After this age they begin to drop -at the age of 17-11-16 the highest Composite Score is 149 . This means that even ifan examinee answers every item correctly, his or her score will differ by 15 points

(about I standard deviation) ifhe or she is testedfirst at age 12 and again at age 18 years. Similarly, a Composite Score as low as 36 cannot be obtained at all age levels. For example, the lowest possible scores at the 2-year level are 95 (2-0-0), 87 (2-3-16), 80 (2-7 -16) , and 73 (2-11-1 6) . It is only at age level 3-3-16 that a child can obtain a score more than 2 SD's below the mean . This limited floor means that a diagnosis of mental retarda­ tion cannot be established through use of the SB: FE for children who are between 2 and 3 years of age . The nonuniformity of scores also holds for alJ compara­ ble combinations of subtest s , either in factors or otherwise established. Applying profile anaJysis uniformly to all subtests would be misleading for individual cases because the same number of scaled-score points cannot be obtained on all subtests . 3 . Limited support forfour area scores. Factor analysis does not support the four area scores throughout the age levels of the scale. Thus the routine use of area scores is not recommended. Factor scores are useful in describing the examinee's abilities, however. 4 . Difficulry in scoring responses. When responses on Verbal Relations, Comprehension, and Vocabulary sub­ tests differ from those that are in the Guide, they may be di fficult to score . The Copying subtest is also l ikely to be difficult to score. Such difficulties may lead to halo effects in scoring and contribute to examiner bias. 5. Difficult}' in interpreting norms for subtests that have estimated scaled-score values at some ages. Figure 2 in the Guide lists nine subtests that have estimated standard score values. For example, estimated standard score values are provided for ages 15 , 16, and 17 years for the Absurdities subtest . For these years, raw scores can be converted into standard scores. The Guide fails to provide any descriptive statistics for these age levels, however. Similarly, it fails to provide descriptive statistics for any of the ages at which the subtest scores are estimated. Therefore it is recom­ mended that subtests be used only for ages at which de­ scriptive statistics are provided. 6. Lack of description of procedure for establishing cutoffcriteria. Neither the Guide nor the Technical Manual provides information concerning how the cutoff criteria (the number of items that should bs administered before the test is discontinued) were determined (that i s , em­ pirically or intuitively) . 7 . Overly long administration time. Because the SB: FE

290

CHAPTER I I

has only one timed subtest, examinees can take an inordi­ nate amount of time to solve the problems . Even in cases where the child does not take an excessive amount of time to respond , it may take two hours or more to administer the complete battery to an adolescent. 8. Incorrect entry level points. The entry level points indicated in the Guide may not be appropriate for all children . Therefore , the test may be unnecessarily pro­ longed and children may be given items that are too diffi­ cult at the beginning of the test .

CONCLU D I NG C O M M ENT O N THE S8: FE

The S B : FE i s a potentially powerful tool for assessment of the cognitive ability of young childre n , adolescents , and young adults . It has some serious shortcomings, however, that must be recognized. The most serious of these are the lack of a comparable battery throughout the age levels covered by the scale and the nonuniforrnity of Composite Scores, factor scores, and scaled scores. Users of the scale must be extremely alert to these features of the scal e . Profiles and changes in performance cannot b e evaluated in a routine manner. In each case attention must be given to the range of scores provided by the scale for the examinee's particular age . Critical decisions must be made with these considerations in mind .

SUM MARY I . The S8: FE represents a significant departure from for­ mer editions, although some continuity is maintained with for­ mer editions. The 1937 and 1960 editions of the Stanford-Binet were generally well received, although there were criticisms associated with their emphasis on verbal material, age-scale format, ceiling procedure , item placements, emphasis on rote memory, administrative procedures, inappropriateness for use with adults, and use in clinical situations. The scales produced acceptable validity coefficients, however, and remained popular until the 1980s. 2. Only 6 of the 15 subtests run consecutively throughout the scale. 3. The standardization sample was selected to be represen­ tative of the U . S . population based on 1980 census data, but weighting procedures were needed because the sample contained too many children with high SES backgrounds. 4. The SB: FE continues to use M = 100 and SD = 16 for the Composite Score. For the subtests, M = 50 and SD = 8 are used. 5. Internal consistency reliabilities are excellent for the Composite Score (Mdn r.u = .97, Mdn SEm = 2 . 8 ) . Reliabil ities

STANFORD-BINET INTELLIGENCE SCALE: FOURTH EDITION

for the subtests are less satisfactory (r.",.'s range from . 73 to .94). Stability coefficients are satisfactory for the Composite Score (ra: = .90), but not for some of the subtests (r,,:s range between . 2 8 and . 86). 6 . Val idity studies comparing the SB: FE with other intel­ ligence tests indicate satisfactory concurrent validity (Mdn r = . 80) . S8: FE Composite Scores and IQs on other tests may not be comparable, however, especially in gifted or mentally retarded populations. In these populations the SB: FE has been found to yield lower scores than either the Stanford-Binet: Form L-M or the WAIS-R. 7 . Higher intercorrelations are observed among the verbal subtests and nonverbal subtests than among the memory subtests. The memory subtests have low correlations with most other subtests. The subtests with the highest correlations with the Composite Score are Quantitative, Number Series, Vocabulary, Matrices, Comprehension, and Verbal Relations (all above . 75) . 8 . The relationships between S B : FE Composite Scores and demographic characteristics indicate that black children tend to score 10 to 17 points lower than white children. There is about a 14-point difference between children in the highest and lowest socioeconomic status groups ( 108 vs. 94) . Differences associated with gender were minimal , but some meaningful differences were associated with community size. 9. A principal components analysis of the SB: FE supports two factors at ages 2 through 6 - Verbal Comprehension and Nonverbal Reasoning/Visualization - and three factors at ages 7 through 23 - Verbal Comprehension, Nonverbal ReasoninglVi­ sualization, and Memory. The subtests that comprise these fac­ tors differ at different ages. 10. All SB: FE subtests are either good or fair measures of g . Those with the highest g loadings are Vocabulary, Number Se­ ries, Quantitative , Comprehension, and Matrices. 1 1 . Of the 15 subtests in the SB: FE, only Vocabulary, Com­ prehension, Quantitative , and Memory for Objects have inade­ quate specificity at some ages. 1 2 . Factor scores are the preferred way to identify meaningful psychological dimensions on the SB: FE. The factor scores are highly reliable. 1 3 . The routing procedure used on the SB: FE may not be applicable to mentally retarded children. Entry level points for these children may have to be considerably lower than those indicated in the record bookJet. You should feel free to adjust the entry level when testing a mentally retarded child (and other children as well, if necessary) . 1 4 . A complete administration of the S B : F E requires estab­ lishing a basal level (two consecutive levels at which both items are passed) and a ceiling level (two consecutive levels at which three or four items are failed). 15. Specific physical abilities are required to take the subtests, but some modifications may be made for physically handicapped children. 1 6 . Short forms discussed in the chapter are available for screening purposes. 1 7 . A profile sheet is useful in plotting the examinee's scores .

29 1

SUMMARY

It can be used in conjunction with the recommended b�llery for each age group or with the entire battery. 1 8 . The interpretive rationale, factor analytic findin",s. relia­ bility and subtest correlations, and admini strative considerations for each of the 15 S B : FE subtests are presented in the chapter. 1 9 . The rationale for the WISC-R Vocabulary subtest gener­ ally applies to the oral section of the Vocabulary subtest i n the SB: FE. For the picture section of the Vocabulary subtest, formal education l ikely plays a minor role. The subtest is a good mea­ sure of g and contributes to the Verbal Comprehension factor. It is . 87) . Scoring requ i res considerable a rel iable subtest ( rc, j udgment. 20. The rationale for the WISC-R Comprehension subtest generally applies to the SB: FE Comprehension subtest. The subtest is a good measure of g and contributes to the Verbal Comprehension factor. It is a reliable subtest (ru 8 9) Scoring requires considerable j udgment . 2 1 . Absurdities measures the ability to isolate the incongrui­ ties and absurdities of visual material. Success depends on per­ ception of detail, alertness, concentration, and social under­ standing. The subtest is a good measure of g and contributes to the Verbal Comprehension factor. I t is a rel iable subtest (rxx . 87) and relatively easy to administer. 2 2 . Verbal Relations measures verbal concept formation and reasoning ability. It is a fair measure of g and contributes to the 9 1) Verbal Comprehension factor. It is a reliable subtest (rr..but may be somewhat difficult to score. 2 3 . The rationale for the WISC-R Block Design subtest gener­ ally applies to most of the items on the Pattern A nalysis subtest of the SB: FE. The form-board items, however, primarily measure visual-motor ability and recognition and manipulation of forms . The subtest is a good measure of g and contributes to the Nonver­ bal Reason i ng/Visual ization factor. It is a reliable subtest (r :x . .92) and is easy to score. 24. The Copying subtest measures visual-motor ability and eye-hand coordination. The subtest is a fair measure of g and contributes to the Nonverbal ReasoninglVisual ization factor. It is a reliable subtest (r.xx . 87 ) . The subtest may be the most difficult of all SB: FE subtests to score. 25 . M atrices measures analogic reasoning, attention to detail , and concentration. The subtest i s a good measure of g and contributes to the Nonverbal ReasoninglVisualization factor. It is a reliable subtest ( 1:« . 90) and is easy to administer. 26. Paper Folding and Cutting measures visualization , spatial ability, the integration of visual and spatial abil ities, and attention to visual cues. The subtest is a fair measure of g and contributes to the Nonverbal Reasoning/Visualization factor. It is a reliable subtest ( rrx .94) and is easy to score. 2 7 . The rationale presented for the WISC-R A rithmetic sub­ test generally applies to the Quantitative subtest, although formal education probably has less influence on performance on the early Quantitative items . The subtest is a good measure of g and contributes to the Nonverbal ReasoninglVisual ization factor. It is a reliable subtest ( r rx . 88) and is easy to score. 28. Number Series measures logical reasoning and concentra=

=

.

.

=

=

=

=

=

=

=

.

,

tion when using numbers. The subtest is a good measure of g and contributes to the Nonverbal Reasoning/Visualization factor. It is a reliable subtest (r = .90) and is easy to score. 29. Equation Building measures logi c , flexibility, and trial and error. The subtest is a fair measure of g and contributes to the Nonverbal Reasoning/Visualization factor. It i s a reliable subtest (r .91) and is relatively easy to score. 30. Bead Memory measures short-term visual memory and involves form perception and d iscrimination, spatial relations , and alertness to detail. The subtest is a fair measure of g and contributes to the Nonverbal Reasoning/Visualization factor. It is = a reliable subtest (rxx . 8 7) and is easy to administer. 3 1 . Memory for Sentences measures short-term auditory memory for meaningful material . The subtest is a good measure of g and contributes to the Verbal Comprehension factor and Memory factor. It i s included in the Verbal Comprehension factor at ages 2 through 7 and in the Memory factor at ages 8 through 18-23 . It is a reliable subtest (r . 89) and is easy to administer. 3 2 . The rationale for the WISC-R Digit Span subtest applies to the Memory for Digits subtest. The subtest is a fair measure of g and contributes to the Memory factor. It is a reliable subtest (r", . 83) and is easy to administer. 3 3 . Memory for Objects measures short-term visual memory. The subtest is a fai r measure of g and contributes to the Memory factor. It is a moderately reliable subtest (r", . 73) and is easy to administer. 34. Although the same considerations that apply to profile analysis on the WISC-R apply to the SB: FE, more care should be taken in using profile analysis with the S B : FE because it is essential ly a new instrument. The proposed interpretive ra­ tionales and possible implications of high and low scores are summarized in Table C-52 of Appendix C . 3 5 . The non-uniform scaled-score ranges for subtests, short forms , factor scores, and Composite Scores compl icate the use of profile analysis on the SB: FE. Profile analysis must be per­ formed by considering the available score ranges, as shown i n Table C-51 in Appendix C. 36. As with the Wechsler scales, a successive level approach to test interpretation is helpful with the SB: FE. 3 7 . The Verbal Comprehension factor is dependent on the examinee's accumulated experience. The Nonverbal Reasoning/ Visual ization factor is more dependent on the examinee's imme­ diate problem-solving ability. The Memory factor is dependent on the examinee's ability to sustain attention. A l l three factors in­ volve verbal and nonverbal strategies i n the solution of problems. 38. Significant differences between factor scores must be in­ terpreted within the context of the examinee's entire performance and clinical history and always in relationship to the examinee's absolute level of functioning. 39. When the Verbal Comprehension factor score is higher than Nonverbal Reasoning/Visual ization and Memory factor scores, auditory processing skills may be better developed than visual processing and memory skills. When the Nonverbal Rea­ soningIVisualization factor score is higher than Verbal Com­ prehension and Memory factor scores, nonverbal processing xx

xx

=

xx =

=

=

292

CHAPTER I I

skills may be better developed than verbal and memory process­ ing skills. When the Memory factor is higher than the Verbal Comprehension and Nonverbal Reasoning/Visualization factor scores, short-term memory skills may be better developed than verbal and nonverbal reasoning skills . 40 . Any hypotheses generated from subtest comparisons should be treated as tentative and formulated in relation to the examinee's absolute level of scaled scores. 4 1 . Special combinations of subtests in the SB: FE may prove useful in generating interpretive hypotheses. 42. The assets of the S8: FE include its good validity, high reliabiJities, excellent standardization , good administration pro­ cedures, well-constructed and well-designed materials, and help­ ful scoring criteria. 43. The limitations of the S8: FE include the lack of a com­ parable battery throughout the age ranges covered by the scale, variable range of scores, limited support for four area scores , difficulty in scoring some subtests, difficulty in interpreting norms for subtests that have estimated scaled scores at some ages, the manual's failure to describe procedure for establishing cutoff criteria, overly long administration time, and incorrect entry level points. 44. The S8: FE is a potentially powerful tool for the assess­ ment of cognitive ability of young children, adolescents, and young adults. Profile analysis, however, cannot be done in a routine manner- attention must be given to the range of scores available for each subtest and age.

STANFORD-BINET INTELLIGENCE SCALE: FOURTH EDITION

Ceiling level (p. 258) Physical abilities necessary for the SB: FE (p. 259) Short forms of the SB: FE (p. 259) Profile sheet for the SB: FE (p. 261) SB: FE Vocabulary (p. 261) SB: FE Comprehension (p. 266) SB: FE Absurdities (p. 267) S8: FE Verbal Relations (p. 267) SB: FE Pattern Analysis (p. 268) SB: FE Copying (p. 268) S8: FE Matrices (p. 269) S8: FE Paper Folding and Cutting (p. 269) S8: FE Quantitative (p. 271) S8: FE Number Series (p. 271) S8: FE Equation 8uilding (p. 272) S8: FE 8ead Memory (p. 273) S8: FE Memory for Sentences (p. 274) S8: FE Memory for Digits (p. 274) S8: FE Memory for Objects (p. 274) Profile analysis (p. 276) 8ase rate differences between factor scores (the probability-ofoccurrence approach) (p. 276) Standard age score (or scaled-score) ranges (p. 277) Successive level approach to i nterpreting the S8: FE (p. 278) Formulating hypotheses about factor score discrepancies on the SB: FE (p. 280) Comparisons between subtests on the SB: FE (p. 281)

KEY TERMS, C O N C E PTS, AND N A M ES

Three-level hierarchical model (p. 250) SB: FE standardization sample (p. 250) Standard age score (p. 250) Area score (p. 250) Composite Score (p. 250) S8: FE test-age equivalents (p. 251) Reliability of the S8: FE (p. 251) Standard errors of measurement of the SB: FE (p. 251) Stability of the SB: FE (p. 251) Validity of the SB: FE (p. 252) S8: FE subtest intercorrelations (p. 253) Stratification variables on the SB: FE (p. 253) SB: FE Verbal Comprehension factor (p. 255) SB: FE Nonverbal Reasoning/Visualization factor (p. 255) SB: FE Memory factor (p. 255) SB: FE subtests as measures of g (p. 256) Subtest specificity on the SB: FE (p. 257) Factor scores on the SB: FE (p. 257) Adaptive testing (p. 258) Routing subtest (p. 258) Basal level (p. 258)

STUDY Q U ESTIONS I . Discuss the S8: FE . Include in your discussion the follow­ ing issues: standardization, Composite Score, test-age equiv­ alents, reliability, and validity. 2 . Describe and interpret intercorrelations on the S8: FE (a) between subtests and (b) between subtests and the Composite Score. 3 . Describe and interpret the S8: FE Composite Scores with respect to the stratification variables used in the standardization sample. 4. Describe and interpret S8: FE factor analytic findings. 5. Discuss S8: FE administrative considerations. 6. Discuss S8: FE short forms. 7. Discuss the rationale, factor analytic findings, reliability and correlational highlights, and administrative considerations for each of the following S8: FE subtests: Vocabulary, Com­ prehension, Absurdities, Verbal Relations, Pattern Analysis, Copying, Matrices, Paper Folding and Cutting, Quantitative, Number Series, Equation 8uilding, 8ead Memory, Memory for Sentences, Memory for Digits, and Memory for Objects . 8 . Discuss the intent of profile analysis, methods of profile

STUDY QUESTIONS

analysis, and approaches to profile analysis on the SB: FE. Include in your discussion the effect of score range on profile analysis. 9. Discuss how to interpret differences between SB: FE factor scores. 1 0. Discuss how to interpret differences between SB: FE sub-

293 tests. Cite at least seven subtest comparisons in your presenta­ tion. I I . What are some general considerations in interpreting the SB: FE? 1 2 . Discuss the assets and limitations of the SB: FE.

ASSESSM E NT OF I NTE LLIG E N C E AN D I N FANT DEVELO PM E NT WITH SPECIALIZE D M EASU RES

McCarthy Scales of Children's Abilities

A n "intelligence quotient " may b e ofprovisional value as a first crude approximation when the melltal level of an individual is sought; bur whoever imagines that in determining this quantity he has summed up "the " intelligence of an individual once alld for all . . . leaves off where psychology should begin.

Kaufman Assessment Battery for Children Detroit Tests of Learning Aptitude - 2 Detroit Tests of Learning Aptitude - P Extended Merrill-Palmer Scale

-William Stern

Slosson Intell igence Test Raven's Progressive Matrices Goodenough-Harris Drawing Test Leiter International Performance Scale Pictorial Test of Intell igence Columbia Mental Maturity Scale Blind Learning Aptitude Test Hiskey-Nebraska Test of Learning Aptitude Piagetian Tests Bayley Scales of Infant Development Other Infant Assessment Measures Psychological Evaluation Summary

294

MCCARTHY SCALES OF CHILDREN'S ABILITIES

This chapter covers a selection of spec ial ized instruments for the assessment of cognitive abil ity. Some of the teo ·ts are special ized because they are l imited to a specific (a) age range , such as infancy or early childhood , (b) handicap­ ping condition, such as visual impairment or hearing im­ pai rment , or (c) area of cognitive abil ity, such as fi gural reasoning . Other tests make excellent fol low-up tools to more traditional measures. These tests allow you to ex­ plore or confirm clin ical hypotheses generated by tests that have been more rigorously standardized and cover a wide spectrum of cognitive functioning. The individually administered tests of intell igence sur­ veyed in this chapter make valuable additions to your fund of assessment techniques ; they are especially useful in situations in which it is not feasible or practical to adminis­ ter the WISC-R, WPPSI , WAIS-R, or Stanford-Binet In­ telligence Scal e : Fourth Edition . The infant tests discussed in this chapter are best seen as measures of development (or maturational growth) rather than of intell igence . The developmental scores obtained from infant tests, however, do have prognostic value for assessing the cognitive abili­ ties of handicapped infants (see Chapter 4). The tests described in this chapter can be used for screening , follow-up evaluations, and assessing handi­ capped as well as normal children. Unlike other standard measures of intelligence, some of these tests require only a simple response such as pointing, eye movements , or a "yes" or "no" sign symbolized by a prearranged signal . In some cases instructions may be pantomimed . The admin­ istration of some tests in this chapter (for example, the Siosson Intell igence Test) can be learned relatively quickly and easily, whereas that of other tests (for example, the Bayley Scales) requires considerable training. In any case, considerable ski l l is needed to establ ish rapport, recognize signs of psychopathology, and i nterpret assessment findings . Although most of the infant and early chi ldhood tests covered in this chapter provide lim ited material for qualitative analysis, especially those that require only a pointing response, possible reasons for the child's failures and successes should be considered. For example, do failures represent limited cognitive ability or inability to understand the directions? Are they associated with sen­ sory limitations? How reliable and valid are the results? Critical decision s , such as placement decisions or pre­ dictions about future academic or vocational success, nor­ mally should not be based solely on any of the tests de­ scribed in this chapter. Some of the tests covered in the chapter have l imited validity and reliabil ity, others have a limited or out-of-date norms, and still others measure only some limited subdomains of abil ity. Additionally, for indi-

295 vidual children large d iscrepancies have been reported between IQs obtained on some of these tests and those obtained on the WISC-R, even though mean IQs may not be significantly different . The WISC-R, WPPS I , WAIS-R, and Stanford-Binet Intelligence Scale: Fourth Edition are recommended for the assessment of intelligence in prefer­ ence to these other measures when a child has the neces­ sary physical capacities to respond to the test questions , when the child is of appropriate age to take the test, and when time is not at a premium . Admi nistration of one of the more traditional measures is particularly important when the IQ is to be used for critical decisions . When verbal responses cannot be elicited from the child, when sensory or motor handicaps limit the child's performance, or when time is at a premium and only a screening decision is being made, however, specialized intell igence tests, brief intelligence tests , or brief forms of longer intel­ l igence tests can be usefu l . (Appendi x F presents the highl ights of the tests covered in this and other chapters of the book . )

M CCARTHY SCALES O F C H I LDREN'S ABI LITIES

The McCarthy Scales of Children's Abil ities ( McCarthy, 1972) is a well standardized and psychometrically sound measure of the cognitive abilities of young children (see Figure 1 2 - 1 ) . It is individually administered , covers the age range from 2 Y2 to 8 1/2 years, and takes approximately 45 to 60 minutes to administer, depending on the age of the child . The McCarthy Scales has some unique features valuable for the assessment of young children with learn­ ing problems or other exceptionalities (Sattler, 1978) . It renders a general measure of i ntellectual functioning cal led the General Cognitive Index (GC I ) , as well as a profile of abilities that includes measures of verbal abil ity, nonverbal reasoning ability, number aptitude , short-term memory, and coordination . Several items also assess hand dominance.

Tests Contained in the McCarthy Scales The McCarthy Scales contains the following six scales: Verbal Scale, Perceptual-Performance Scale, Quantitative Scale, Memory Scale , Motor Scale, and General Cog­ nitive Scale. The General Cognitive Scale contains the 15 tests that comprise the Ve rbal Scale, the Perceptual­ Performance Scale, and the Quantitative Scale. The only tests not included in the General Cognitive Scale are three tests that measure gross motor coordination . (Tests in the

296

Figure 1 2- 1 .

CHAPTER 1 2 ASSESSMENT OF INTELLIGENCE AND INFANT DEVELOPMENT WITH SPECIALIZED MEASURES

McCarthy Scales of Chi ldren's Abilities. Courtesy of The Psychological Corporat ion .

Memory Scale , as wel l as the rest of those in the Motor Scale also appear in other scales . ) The abil ities thought to be measured by the scales and subtests are shown in Table 12- \ .

Scores The fi ve scale indices derived from the McCarthy Scales s ubtests represent standard score s , with M = 50 and SD = 10. The overall GCI has M = 100 and SD = 16. The GCI is an estimate of the child's abil ity to integrate his or her accumulated knowledge and adapt that knowledge in

order to perform the tasks on the scales . This functional definition of the GCI is strikingly similar to definitions associated with the Inte l l igence Quotient, a term McCar­ thy deliberately has avoide d . Further evidence suggesting that the two terms were intended to be comparable comes from (a) the descriptive c l assifications associated w ith the GCI , which are almost the same as those used for IQs on Wechsler's scales, and (b) the availability of mental ages for the GCI, which can serve as indications of mental competence . For all practical purposes, the definitions and uses proposed for the GCI and the IQ are the same ; however, psychometrica l l y these scores are not compara­ ble . This point will be emphasized later in this section.

297

M CCARTHY SCALES OF C H ILDREN'S ABILITIES

Table 1 2- 1 Abi lities Thought to B e Measured b y McCarthy Scales and Subtests Verbal Scale (Abiliry to understand and process verbal stimuli and to express thoughts) Word Knowledge

Pictorial Memory Short-term memory (auditory and visual) Early language development Attention

Verbal concept formation Early Language development Verbal expression (Part I I )

Verbal Fluency

Verbal Memory

Opposite Analogies

Verbal concept formation Logical classification Creativity (divergent thinking) Verbal expression

Short-term memory (auditory) Verbal comprehension Attention Concentration (Part I I ) Verbal expression (Part I I )

Verbal concept formation Early language development Verbal reasoning

Perceptual-Performance Scale (Visual-motor coordination and nonverbal reasoning through manipulation of concrete materials) Block Building

Puzzle Solving

Tapping Sequence

V isual-motor coordination Spatial relations

V isual perception Nonverbal reasoning V isual-motor coordination Spatial relations

Short-term memory (primarily visual) V isual-motor coordination Attention

Right-Left Orientationa Spatial relations Verbal concept formation Nonverbal reasoning Directional ity

Draw-ADesign Visual perception Visual-motor coordination Spatial relations

Draw-AChild Nonverbal concept formation V isual-motor coordination Body image

Conceptual Grouping Logical classification Nonverbal reasoning Verbal concept formation

Quanritative Scale (Faciliry in dealing with numbers (/lid understanding oj' quantitative concepts) Number Questions

Numerical Memory

Counring and Sorting

Numerical reasoning Computational skills Number facts and concepts Concentration Verbal comprehension

Short-term memory (auditory) Attention Reversibil ity (Part I I )

Rote counting Number concepts Numerical reasoning

Memory Scale (Short-term memory across a wide range of visual and auditory stimuli) Pictorial Memory Short-term memory (auditory and visual) Early language development Attention

Tapping Sequence Short-term memory ( primarily visual) Visual-motor coordination Attention

Verbal Memory Short-term memory (auditory) Verbal comprehension Attention Concentration (Part II) Verbal expression (Part II)

Numerical Memory Short-term memory (auditory) Attention Reversibility ( Part II)

(Table conrillues next page)

CHAPTER 1 2

298

ASSESSMENT OF INTELLIGENCE AND INFANT DEVELOPMENT WITH SPECIALIZED MEASURES

Table 1 2- 1 (cont.)

Motor Scale (Gross and fine motor coordination) Leg Coordination

Gross motor coordination Balance

Arm Coordination

Gross motor coordination Precision of movement

Imitative Action

Gross motor coordination Fine motor coordination

Draw-A-Design

Fine motor coordination

Draw-A-Child

Fine motor coordination

General Cognitive Scaleb (Reasoning, concept formation , and memory when solving verbal and numerical problems and when manipulating concrete materials) a

For ages 5 and above . The 1 5 separate tests included in the General Cognitive Scale are described i n the Verbal , Perceptual-Performance, and Quantitative Scales. Source: Adapted from Kaufman and Kaufman ( 1 977).

b

Standardization The standardization of the McCarthy Scales was excel lent, with the sample closely matching the 1970 census data. Stratification variables included age , sex , race , geographic . region, father's occupation, and urban-rural residence . A total of 1 ,032 children between the ages of 2 V2 and 8 1/2 years were tested ; 100 to 106 children , equally divided by sex, were tested for each of the 10 age levels included in the standardization sample . Sex differences in the standardiza­ tion sample are minimal (Kaufman & Kaufman, 1973) .

Reliability Reliability coefficients, standard errors of measurement, and intercorrelations between the scales are reported for the 10 different age levels in the standardization sample for each of the six scales . The average split-half rel iabil ity of the GCI is excellent (rxx = . 93 ) , and average split-half rel iabilities for the other five scales are satisfactory ( from . 79 to . 88 ) . The average standard error of measurement of the GCI is 4 points . The lowest intercorrel ations occur between the Motor Scale and the other five scales. Stability of the McCarthy Scales, measured over a retest interval of approximately 30 days for a sample of 125 children, is adequate, with coefficients of .90 for the GCI and of .69 to . 89 for the other scale indices. The GCI also has adequate stabi l ity ove r a one -year period for wh ite c h i ldren ( ru = . 85 ; Davis & Slettedahl , 1976) and for Mexican­ American children ( rxx = . 8 5 ; Valencia, 1983) .

Validity Concurrent validity of the McCarthy Scales is acceptable with the Stanford-Binet : Form L-M , WISC, WISC-R , W PPSI , K-ABC, and Slosson Inte l l igence Test used as c riteria; correlations range from .45 to .91 (Mdn r = . 75) ( Arinoldo, 1982 ; Bickett, Reuter, & Stancin, 1984; Bondy, Sheslow, Norcross , & Constantino , 1982 ; Davis, 1975 ; Davis & Rowland, 1974 ; Davis & Wal ker, 1977 ; Gerken, Hancock, & Wade , 197 8 ; Goh & Youngquist, 1979; Har­ rison & Wiebe , 1977; Hynd et aI . , 1 980; Krohn & Traxler, 1979 ; Levenson & Zino, 1979; McCarthy, 1972 ; Nagl ieri , 1980a ; Naglieri , 1985a; Reilly et aI . , 198 5 ) . Concurrent validity is also acceptable with achievement tests used as c riteria (Mdn r = . 5 8 ) ( Massoth & Leve nson , 1982 ; Naglieri, 1980b ; Nag l ieri & Harrison, 1982 ; Sturner, Funk, & Green, 1984). Although concurrent validity coefficients are satisfac­ tory, large differences have been reported between McCar­ thy GCls and Stanford-Binet : Form lrM and W ISC-R IQs for samples of preschool children (Gerken et aI . , 1978) , mentally retarded children ( Levenson & Zino, 1979) , and l e arning-d isabled childre n (Goh & Youngquist, 1979) . With gifted children, means differed by 10 points on the Stanford-Binet: Form L-M (M GCI = 105 , M IQ = liS), but individual discrepancies were as much as 20 to 30 points (with McCarthy GCls l ower than Stanford-Binet: Form L-M IQs) . With mentally retarded children, McCar­ thy GCls were , on the average , 20 points lower than Stanford-Binet : Form L-M IQs (44 v s . 64) . With learning-

299

MCCARTHY SCALES OF CHILDREN'S ABILITIES

disabled childre n , GCIs were lower than WISC-R IQs by 8 to I S points, on the average . These results uggei>t that GCI s are not interchangeable with either Sanford- Binet: Form L-M IQs or WISC-R IQs . Satisfactory predictive validity is indicated by correla­ tions with performances on various achievement tests, including the Metropolitan Achievement Tests , Peabody Individual Assessment Test, Wide Range Achievement Test, and California Achievement Test (Mdn r = . 66) (Fu n k , Sturner, & Green, 1986; Harrison , 198 1 ; Massoth , 1985 ; Reilly et al . , 1985 ; Taylor & Ivimey, 1980) . C onstruct validity appears to be questionabl e . Although the standardization data. yielded five factors generally throughout the scale - Verbal , Motor, General Cogni­ tive , Memory, and Perceptual-Performance (Kaufman , 1975b ) - later studies have not replicated these findings. I n a large sample ( N = 300) of school-aged children 6 to 8 '/2 years old, only three factors were found: General Cog­ nitive , Verbal , and Motor (Keith & Bolen, 1980) . With low-functioning childre n , no general factor was found (Nagl ieri , Kaufman, & Harrison, 198 1 ) . In a sample of 4year-old c h i ldre n , fou r factors were found : Genera l , Perceptual-Performance , Verbal , and Motor (Trueman, Lynch , & Branthwaite, 1984) . Finally, different factor structures were found for males and females (Purvis & Bol e n , 1984; Wiebe & Watkins, 1980). These f indings suggest that caution should be exercised in interpreting the McCarthy Scales in a similar manner for all groups of exceptional children . Additionally, be­ cause the fi ve scales are not factorially independent, they should not be i nterpreted as reflecting distinct abilities. This is especially true of the Quantitative and Memory scales. The failure to fi nd distinct factors in some samples means that profile analysis using all five scales may not always be appropriate . The scales may be better measures of general cognitive abilities than of the specific abil ities designated by the names of the scales. Factor analysis of the McCarthy Scales in samples of black and white children (Kaufman & DiCuio, 1975) and in samples of Mexican-American and white childre n ( M ishra, 1981) indicate that the factor structures are sim­ ilar in these groups .

Some Useful Administrative and Interpretive Considerations Confidence intervals and percentile ranks.

Confi­ dence intervals for the fi ve scales and the GCI by age level are shown in Table C-S4 in Appendix C. Tables BC-I and

BC-2 on the inside back cover can be used to obtain the percentile ranks for GCIs and the classifications associated with the GCls, respectively.

Short forms.

Various short forms have been proposed for the McCarthy Scale s . One consists of Puzzle Solving, Word Knowledge, Numerical Memory, Verbal Fluency, Counting and Sorting, and Conceptual Grouping (Kauf­ man , 1977) . Another consists of Counting and Sortin g , Pictorial Memory, Number Questions , Verbal Fluency, Numerical Memory, and Tapping Sequence (Taylor, Slocumb, & O'Neil l , 1979) . Still another consists of Right­ Left Orientation, Draw-A-Design, Numerical Memory, Leg Coordination, and Conceptual Grouping ( McCarthy, 197 8 ) . These short forms should be used only for screen­ ing purposes .

Extrapolated GCls. Extrapolated GCls above ISO and below SO are found in a publication by Harrison and Naglieri ( 1978) . These extrapolated GCls may be used to avoid the floor and ceiling effects, which l imit the scale's ability to provide GCI s for gifted or low-functioning men­ tally handicapped childre n . Significant differences for index scores.

Table C - 5 S in Appendix C provides the critical differences required i n order for significance t o be attributed t o a discrepancy between an index and the mean index for each of the five scales. The critical differences are 8 points for the Verbal Index , 9 points for the Perceptual-Performance Index, and 10 points for the Quantitative , Memory, and Motor In­ dices. One or mort! of the five scale indices were signifi­ cantly different from the mean score on 62 percent of the profiles from the standardization sample (Kaufman , 1976a) . These results suggest that one should exercise caution in interpreti ng variability on the McCarthy Scales. McCarthy suggested that about a IS-point difference is needed in order for a discrepancy between the Verbal and Memory, Perceptual-Performance and Motor, and Quan­ titative and Memory Scales to be considered "noteworthy. " Yet d i ffere nces as low as 10 points are s i g n i ficant (Ysseldyke & Samuel , 1973). Because significant differ­ ences between scale indices vary by as much as 4 scaled­ score points at various ages, it is best to use the table presented by Ysseldyke and Samuel (1973) to obtain the significant differences between scales needed at various age levels.

Conversion to scaled scores.

Reynolds (198Sb) pro­ vides tables for converting raw scores to scaled scores

300

CHAPTER 1 2

ASSESSMENT OF INTELLIGENCE AND I N FANT DEVELOPMENT WITH SPECIALIZED MEASURES

(M = 100, SD = 15) for the Draw-A-Design and Draw-A­ Child tests for children from 4 years , 4 months to 8 years, 7 months of age . Limitations of the McCarthy Scales Although the McCarthy Scales has many strengths, it also has limitations (cf. Kaufman & Kaufman, 1977 ) . 1 . Much clerical work is required t o transform scores on the 18 separate tests into indices on the six scales . 2 . The failure to include social comprehension and judgment tasks or many abstract problem-solving tasks limits the breadth of the scale . 3 . The scale may not b e suitable for school-age children because of the cumbersomeness of some of the pro­ cedures. 4 . The GCI floor of 50 l imits the scale's usefulness in assessing the abilities of severely mentally retarded chil­ dren and 2 1/2 -year-olds with below-average cognitive abili­ ties; the low ceilings on many tests limit the scale's useful­ ness in assessing the abilities of older gifted children . 5 . The absence of norms for older children and adoles­ cents limits the utility of the scales in follow-up evalua­ tions. 6 . The internal consistency rel iabilities for some of the indices are low (below . 80) . 7 . Some of the scales overlap in content, and factor analyses do not support the construct validity of indepen­ dent scales in all populations. 8. The scales makes no provision for prorating when tests are spoiled or not administered. Thus it is not possible to compute an index for any scale that includes a spoiled or omitted test . [Fortunately, Kaufman and Kaufman ( 1977) describe proration procedures that permit the estimation of scores in such situation s . ]

Comment on the McCarthy Scales of Children's Abilities Because the McCarthy Scales does not provide scores that are equivalent to those provided by the W ISC-R, caution should be exercised in using GCls for placement deci­ sions, especially in the assessment of mentall y retarded or gifted children. Is the McCarthy GCI more valid than the WISC-R IQ? Unfortunately, there is insufficient research to answer this question . According to McCarthy, the individual tests are not sufficiently reliable by themselves to permit meaningful evaluation. The validity of this statement cannot be ver­ ified, however, because individual test reliabilities are not

presente d . The lack of standard scores for each test by age level limits the diagnostic usefulness of the McCarthy Scale s . Additionall y, there is l imited information about the extent to which the scaled scores can assist in educational or clinical treatment decisions. In spite of these l imitations, the McCarthy Scales of Children's Abilities provides a profile of abilities that may be particularly u seful in evaluating young childre n . The manual is convenient to use , the general guidelines for testing are thorough, the materials are well constructed , and the tasks are l ikely to appeal to children. The test is u seful for assessing the cognitive ability and , to a lesser extent, the motor abilities of young children and therefore dese rves serious consideration . (Further information about the McCarthy Scales of Children's Abilities can be obtained in the text by Kaufman and Kaufman, 1977 . )

KAU F M A N ASSESS M E N T BATTE RY F O R C H I LD R E N

T h e Kaufman Assessment Battery for Children (K-ABC) ( Kaufman & Kaufman, 1983) is a measure of intell i gence and achievement designed for childre n ages 2-6 through 12-5 years . Four scales are included in this multi-subtest battery : Sequential Processing Scale , Simultaneous Pro­ cessing Scal e , Achievement Scale, and Nonverbal Scale . The K-ABC is intended for use in school and clinical settings , with administration time being approximately 45 minutes for preschool children and about 75 minutes for those of school age . Not all subtests are administered at every age . Only three subtests run throughout the ages covered by the battery : Hand Movements , Gestalt Closure , and Faces and Places . Thus composite scores are derived from different combinations of subtests, depending on the child's age . Of the 16 subtests in the K-ABC , no more than 13 art< adminis­ tered to any one child .

Description of Scales and Subtests The Sequential and Simultaneous Processing Scales are hypothesized to reflect the child's style of problem solving and information processing . Scores from these two scales are combined to form the Mental Processing Composite , which serves as the measure of intel ligence on the K -ABC . The Sequential and Simultaneous Processing Scales were designed to reduce the effects of verbal processing and gender and ethnic bias . In contrast, the Achievement Scale is more heavily loaded with verbal stimul i . The inter-

KAUFMAN ASSESSMENT BATT ERY FOR C H ILDR E N pretive rationales and implications of high and low K-ABC mental processing and achievement subtest scores are shown in Table C-56 in Appendix C .

Sequential Processing Scale.

The Sequential Process­ ing Scale was designed to measure children's abilities to solve problems that require the arrangement of stimuli in sequential or serial order. This scale contains three sub­ test s : Hand Movements , Number Recall , and Word Order. The Hand Movements subtest requires children to re­ produce raps on a table using the fist, pal m , or side of the hand . On the Number Recall subtest children repeat a series of numbers . On the Word Order subtest children touch or point to a series of silhouettes of objects in the order in which they are named by the examiner. For school-age children, some items also i nclude an inter­ ference task.

Simultaneous Processing Scale. The Simultaneous Processing Scale was designed to measure children's abili­ ties to solve spatial , analogic , or organizational problems that require the processing of many stimuli at once. This scale contains seven subtests: Magic Window, Face Recog­ nition , Gestalt C l osure , Triangles, Matrix Analogies, Spa­ tial Memory, and Photo Series. The Magic Wi!1dow 5ubtest requires the identification of a picture that is passed slowly behind a w indow so that only part of the picture is visible at any one time . The Face Recognition subtest requires the selection of a photograph of a face previously seen from a group of photograph s . On the Gestalt Closure subtest children identify an object or scene in a partiall y completed drawing (ink blot). On the Triangles subtest chil d re n reproduce a design by using several identical rubber triangles. The M atrix Analogies subtest requi res selection of the picture or design that best completes a visual analogy. On the Spatial Memory subtest children recall the location of pictures random!y arranged on a page. The Photo Series subtest requires the arrange­ ment of a series of photographs in a meaningful order.

30 1 The Achievement Scale contains six subtests : Ex­ pressive Vocabulary, Faces and Places, Arithmetic , R id­ dles, Reading/Decoding, and Reading/Understanding . On the Expressive Vocabulary subtest children name objects pictured in photographs . On the Faces and Places subtest children name fictional characters, famous persons, and wel l known places . On the Arithmetic subtest children identify numbers, count, compute, and demonstrate other mathematical skill s . On the Riddles subtest children name a concrete or abstract verbal concept after having been given several of its characteristics. The Reading/Decoding subtest requires that children identify l etters and read and pronounce word s . On the Reading/Understanding subtest children read sentences silently and then act out the com­ mands given in the sentences .

Nonverbal Scale.

The Nonverbal Scale i s not a new scale. It is composed of those subtests from the Sequential and Simultaneous Processing Scales (Face Recognition , Hand Movements , Triangles, Matrix Analogies, Spatial Memory, and Photo Series) that do not require words. The examiner conveys instructions through gestures, and the child responds with movements .

ScorC!s Raw scores for the global scales are transformed into standard scores with M = 100 and SD = 1 5 . Subtests, in turn , are converted into scaled scores with M = 10 and SD = 3. An index of mental ability, called the Mental Processing Composite, is derived only from scores on the Simultaneous and Sequential Processing Sca l e s ; the Achievement subtests are not included. The Mental Pro­ cessing Composite is more heavily weighted with Simul­ taneous Processing subtests than with Sequential Process­ ing subtests. Chatman, Reynolds , and W i l l son ( 19 84) provide tables that show leve l s of interscale and subtest variability and frequencies of significant interscale and intrascale differences on the K-ABC .

Achievement Scale.

The Achievement Scale was de­ signed to assess children's factual knowledge and skills. The subtests on the scale were kept separate from those in the Sequential and Simultaneous Scales in an effort to distinguish between knowledge acquired by exposure to environmental stimulation or educational opportunities and knowledge that results from an integration of sequen­ tial and simultaneous processing. This separation appears to be faulty, however, because it artificia l l y distinguishes the ways in which children acquire and process infor­ mation.

Standardization The standardization of the K-ABC was adequate , with the sample, for the most part, closely matching the 1980 cen­ sus data. Stratification variables included age , sex, geo­ graphic region , socioeconomic status ( parental educa­ tion) , race or ethnicity, and community size . A total of 2 ,000 children between the ages of 2 -6 and 12-5 years were tested, with 200 to 300 childre n , equall y divided by sex , at each of nine age leve l s .

302

CHAPTER 1 2

ASSESSMENT OF INTELLIGENCE AND INFANT DEVELOPMENT WITH SPECIALIZED MEASURES

Reliability

Comment on the K-ABC

Rel iabil ity coefficients, standard errors of measurement, and intercorrelations between the scales and between the subtests are reported. I n ternal consistency rel iabilities for the Mental Processing Composite and the Achievement Scale were, on the average, . 9 1 and . 93 , respectively, for preschool childre n and . 94 and . 9 7 , respective l y, for school-age childre n . Average internal consistency reli­ abilities for the other three scales are satisfactory, ranging from . 86 to .93 . The Mental Processing Composite has an average standard error of measurement of 4 . 6 points for preschool childre n and 3 . 5 points for school-age children. The highest average intercorrelation i s between the Simul­ taneous and Achievement Scales (r = . 66) for school-age children, whereas the lowest average intercorrelation is between the Sequential and Simultaneous Processing Scales for preschool children (r = .41) . Stability of the K-ABC , measured over a retest interval of two to four weeks for three samples of children (Ns of 70 to 92) , i s adequate, with median coefficients o f . 88 for the Mental Processing Composite and . 95 for the Achievement Scal e . Median gain scores were 4 . 9 points for the Mental Pro­ cessing Composite and 2 .0 points for the Achievement Scale .

Following are are some practical and theoretical issues that need to be considered in connection with the use of the K-ABC for the assessment of intell igence.

Validity Statistics on various forms of validity are presented in the K-ABC manual . Evidence of construct validity is pre­ sented in the form of i ncreases in subtest raw scores with age . Factor analysis supports the organization of the K-ABC into three scales. Evidence of concurrent validity i s presented i n the form of correlations of the K-ABC with various individual and group tests of intelligence and achievement. Median corre­ lations between the Mental Processing Composite and the WISC-R or WPPSI are . 50 with the Verbal Scale, .65 with the Performance Scal e , and . 70 w ith the Full Scal e . These medians are based on 13 to 18 samples . With the Stanford­ Binet: Form lrM , the median correlation was .63, based on six samples . Median correlations with tests of achieve­ ment were . 56 for the Mental Processing Composite (based on 12 correlations) and .68 for the Achievement score (based on 13 corre l ations) . Correlations of the K-ABC with various achievement tests admi nistered 6 to 12 months after the K-ABC indicate adequate predictive validity. Median correlations with the total score on the achievement tests were . 56 for the Mental Processing Composite and . 80 for the Achievement score (based on six samples) .

1 . The use of the term mental processing for some subtests and the term achievement for others i s potentially m isleading . Subtests i nvolving vocabu lary, arithmeti c , a n d riddles, for example, which are included o n the Achievement Scal e , require mental processing just as do memory and visual analogy tasks . Similarly, the tasks involving recall i ng numbers, solving puzzle problems, and recognizing faces measure achievement as well as mental processing . Many of the subtests on the Achieve­ ment Scale may measure verbal ability more than achieve­ ment (Keith , 1985 ) . In fact, Achievement scores correlate more highly with scores on the W ISC-R Verbal Scale than with scores on the Iowa Test of Basic S ki l l s . 2 . It is n o t c lear w h y the Mental Processing Com­ posite places more emphasis on S imultaneous Processing subtests than on Sequential Processing subtests. There is no theoretical j ustification for this differential weighting. 3. The l ack of verbal comprehen sion or reason i ng items on the Mental Processing Composite is a fundamen­ tal weakness of the K-ABC . In an attempt to reduce the cultural bias attributed to existing tests and to permit more accu rate assessment of exceptional children, authors of the K-ABC eliminated from its principal composite score one of the key components of inte l l ectual ability - name l y, tasks that measure verbal ski l l s . 4 . T h e heavy rel iance on short-term memory a n d at­ tention tasks may reduce the effectiveness of the K-ABC in providing a valid measure of cognition for children with attention and short-term recall difficulties. 5 . For children ages 2 V2 to 3 years, only five subtests form the Mental Processing Composite . The number of tests, and hence the types of processes measured, is too l imited to provide a complete picture of young childre n's i ntell ectual or problem-solving ability. 6 . There are potential difficulties i n using the term processing to characterize some of the K-ABC subtests and not others . Subtests on the Achievement Scale involve sequential and simultaneous processing, as do those on the Simultaneous and Sequential Processing Scales. Although the K-ABC authors recognize this fact, the separation is l ikely to cause some confusion. Additionall y, nothing i n the K-ABC a llows one t o determine whether a child's response is arrived at through sequential or simultaneous processing modalitie s . The responses required of a child are j ust l ike those he or she woul d give on other standard­ ized individually admi n istered i ntelligence tests - for ex-

303

DETROIT TESTS OF LEARNING APTITUDE-2

ample , reciting numbers given in a digit memory task or selecting a card to complete a matrix analogy. In no case does the scoring take into account how the child processes info rmation . Evaluation of the manner in which the child processes information must be infe rred from other sources- such as subtest score s , other test scores, obser­ vations, or case h istory i nformation. 7. The terms simultaneous processing and sequential processing are ambiguous . Do they refer to how informa­ tion is presented to the child? Do they refer to the form of the response given by the child? Do they refer to certain areas or hemispheres of the brain i nvolved in processing information? Do they refer to specific strategies used by the child - such as labeling, rehearsal , chunking , elabora­ tion , or imagery - in receiving information and in giving a response? Do they reflect executive functions involved in a child's knowing when to stop processing information or when to gi ve a response? Do they mean all of these things and other things as well? These terms are too vague to enable us to understand cognitive processes better. Keith (1985) suggested that the S i multaneous Processing Scale can be interpeted as a measure of nonverbal reasoning ability, whereas the Sequential Processing Scale can be thought of as a measure of verbal memory, particularly at the 5 -year-old level . Additionally, it is important to recog­ nize that each subtest combines dements of rnore tban one processing style . 8 . The K-ABC should not be used for the classification of mental retardation over the entire age range covered by the scale. A 2 1/2 -year-old child who failed every item on the Sequential and Simultaneous Processing Scale would obtain a Mental Processing Composite of 7 9 . For a 3-year­ old this same failure rate would y ield a Mental Processing Composite of 70. Even at 4 years of age complete failure would result in a composite score of 60. 9 . The K-ABC has a low ceiling that may limit its usefulness in evaluating gifted childre n . Over half of the subtests on the Simultaneous and Sequential Processing Scales provide maximum scores that are only 2 standard deviations or less above the mean. The Achievement Scale also has a restricted range . The highest scores range from 133 to 144 for 10 1/2 - to 12 V2 -year-olds. Any attempt to apply a discrepancy formula to differences between mental pro­ cessing and achievement of gifted children is likely to yield misleading results . Significant discrepancies result even when adolescent gifted children pass all items on the Mental Processing and Achievement Scales . The K-ABC yields scores that are as much as 13 to 22 points below the WISC-R and Stanford-Binet: Form L-M (for example, M = 124 . 5 v s . 144 . 3) ( McCallum et aI . , 1984; Van Melis­ Wright & Strein , 1986) .

1 0 . I t is difficult to evaluate the effectiveness of instruc­ tional strategies based on a child's K-ABC profile . The studies in the K-ABC manual are few, and their sample sizes were small. One study (Ayres & Cooley, 1986) found that first-grade children who performed better on a se­ quential learning task obtained higher scores on the Simul­ taneous Processing Scale than on the Sequential Process­ ing Scale. This result is contrary to what would be predicted by the K-ABC . This study failed to support the construct validity of the K-ABC and the proposed remedial instruction strategies offered in the K-ABC manual . Addi­ tionally, Goetz and Hall ( 1984) concluded that the K-ABC does not provide the information needed to develop appro­ priate educational i ntervention s . Further research is needed to evaluate the validity of the proposed instruc­ tional strategies based on K-ABC profiles. I I . There are maj o r sampling problems w ith the K-ABC standardization sampl e . H ispanic-Americans were underrepresented by 24 percent, and low-educa­ tional-level blacks were underrepresented by 10 percent . Bracken ( 1985) concluded that adj usting for sampling dif­ ferences at the uppermost age level produces a difference of approximately 12 points between blacks and whites, in favor of whites. This difference approaches the difference of 1 standard deviation found on other intelligence test s . Thus there is Iittie evidence that t h e K-ABC significantly reduces differences between blacks and whites. The K-ABC may prove to be useful in certain situations , particularl when information is needed about nonverbal cognitive abilities. In most case s , however, the K-ABC should not be used as the primary instrument for identify­ ing the intellectual abilities of normal or special childre n , including t h e mentally retarded , gifted, or learning dis­ abled . Neither should it be the primary instrument for measuring intelligence in clinical assessments .

DETROIT TESTS O F LEAR N I N G APTI T U D E-2

The Detroit Tests of Learning Aptitude ( DTLA-2) (Ham­ mill , 1985a) is a multidimensional battery designed to measure intellectual abilities. Separate scores are provided for subtests, composite s , and the General I ntelligence Quotient . The battery was fi rst published in 193 5 ; a revised manual , based on the 1935 norms, was published in 1967 . A need for updated norms and better standardization pro­ cedures led to the latest revision. The DTLA-2 is designed for children 6 through 18 years of age . It takes approximately 50 minutes to 2 hours to

CHAPTER 1 2

304

ASSESSMENT OF INTELLIGENCE AND INFANT DEVELOPMENT WITH SPECIALIZED MEASURES

administer. There are 11 subtests in the battery : Word Opposites, Sentence Imitation, Oral Directions, Word Se­ quences, Story Construction, Design Reproduction, Ob­ ject Sequences, Symbolic Relations, Conceptual Match ·· ing , Word Fragments , and Letter Sequences . Seven subtests were retained from the original DTLA , and four new ones were added (Story Construction , Symbolic Rel a­ tions, Conceptual Matching, and Word Fragments) . The 1 1 subtests are grouped into eight separate com­ posites , fou r domains, and one overa l l composite , as shown in Table 12-2 . All 1 1 subtests are used i n calculating the Overall Aptitude Composite. The separate composites i nclude varying numbers of tests . For example, the six subtests that form the Verbal Composite are Word Op­ posites, Sentence Imitation, Oral Directions , Word Se­ quences , Story Construction, and Word Fragment s . The five subtests that make up the Nonverbal Composite are Design Reproduction, Object Sequences, Symbolic Rela­ tions, Conceptual Matching , and Letter Sequences. The

Table 1 2-2 Domains, Composites, and Subtests of the DTLA-2

Linguistic Cognitive Attentional Motoric Domain Domain Domain Domain

.�'" 0

.�0

'"

.�0

'"

@-

t3

Subtest

Word Opposites Sequence Imitation Oral Directions Word Sequences Story Construction Design Reproduction Object Sequences Symbolic Relations Conceptual Matching Word Fragments Letter Sequences

� -e

�

@-

t3

� -e � 7

8

362

CHAPTER 14

ASSESSMENT O F VISUAL-MOTOR PERCEPTION . AUDITORY PERCEPTION. AND MOTOR PROFICIENCY

testing session - the task is innocuous, nonthreatening , interesting , and appealing to childre n .

Administrative Suggestions The child should be given a number 2 pencil with an eraser, and an unlined blank sheet of 8 1/2 x 11 inch paper should be placed vertically in front of the child on the tabl e . Extra sheets of paper, usually equal to the number of cards, should be available on the table, along with an extra pencil in case of breakage. Recommended instructions are as follows: Place the stack of nine Bender-Gestalt cards face down (card A on the top and card 8 on the bottom) on the table and say: "I have nine cards here with designs on them for you to copy. [Point to the stack of cards . ] Here is the first one . [Thrn over the first card . ] Now go ahead and make one just l ike it . " or "Now I would l i ke you to draw some designs for me . There are nine cards here and each card has a drawing on it. [Point to the stack of cards . ] I want you to copy the drawings . Make them the best you can . " [Thrn over the first card . ] I f the child raises any questions , give a noncommital reply such as "Make it look as much l ike the picture on the card as you can ," "Do it the way you think best ," or "Do the best job you can ." Present each card individually, beginning with card A and following with cards I through 8 in numerical order. (The cards are numbered sequentially i n approximate order of difficulty. ) Permit the child to erase, but do not allow the use of any mechanical aids such as rulers; the drawings must be done freehand. The nine designs take approximately 5 minutes to com­ plete . Note any large deviation in the time required to complete the designs. Large time deviations may be sug­ gestive of the child's approach to novel situations. For example , children who require 15 minutes to copy the designs might have a slow, methodical approach to situa­ tion s , compulsive tendencie s , or depressive features , whereas those who finish in less than 2 o r 3 minutes might have an impulsive style. A caret (1\) should be placed at the top of any paper containing a design that appears to be rotated. The caret will help you recall at a later time the extent of the child's rotations from the visual plane . Finally, note the child's approach to the task (for example, degree of impulsiveness or compulsiveness , ability to cope with frustration, or attitude toward the task) .

Variations in Administration Several variations are commonly used in administering the Bender-Gestalt. Two of the most popular are (a) the tachis­ toscopic procedure and (b) the memory phase procedure . In the tachistoscopic procedure , each card is shown for 5 seconds. It is then removed, and the child is asked to draw the design from memory. In the memory procedure , the child first completes the standard procedure . After all of the designs have been copied, the child is asked to draw as many of the figures as he or she can remember ("Now draw as many of the designs as you can from memory") . The Bender-Gestalt can also be administered as a group test. Four methods of group administration have been used: (a) presentation of enlarged Bender-Gestalt cards at the front of the room; (b) presentation of special Bender­ Gestalt test booklets in which the designs are drawn and the children copy the designs in blank spaces; (c) presenta­ tion of the designs by means of an opaque projector, overhead projector, or slide projector; and (d) presentation of individual decks of Bender-Gestalt cards . The most successful method with large numbers of children is the first method, the presentation of enlarged Bender-Gestalt cards. The projector methods have the d isadvantages of necessitating special equipment and requiring that the chil­ dren draw their designs in semi-darkness. Individual decks have been most successful with hyperactive or im­ mature children who require extra attention, although only two or three children should be tested at once . Overall , studies indicate that group administration of the Bender­ Gestalt yields reliable protocols comparable to those ob­ tained under individual administration (Koppitz, 1975 ) .

Koppitz Developmental Bender Scoring System The Koppitz Developmental Bender Scoring System (Koppitz , 1964, 1975) for evaluating the Bender-Gestalt drawings of young children is probably the most popular objective scoring system for the test. It is composed of two parts: (a) developmental scoring and (b) scoring of emo­ tional indicators. The first part has the most relevance for the evaluation of visual-motor perception. There are 30 developmental scoring items, with each item receiving I or o points, depending on whether an error occurs (see Table 14-1) . A four-category system is used to c lassify errors : (a) distortion of shape, (b) rotation, (c) integration difficul­ ties, and (d) perseveration. A scoring sheet can be used to record I point for each distortion made by the child. The points are summed to obtain a total error score, which is

363

BENDER VISUAL MOTOR GESTALT TEST

then compared to norms for the child's age . Percentile norms are available for children aged 5-0 through 11-11 years ( Koppitz , 1975 ) . Standard scores (M 100 , SD = 15) are also available in Table C-62 of Appendix C . These scores are most suitable for ages 5-0 through 8-0 . Standard scores (M = 50, SD = 15) are also available for the SOMPA (see Chapter 13) standardization sample (Mer­ cer & Lewis , 1978) . The Koppitz and SOMPA standard score equivalents are not interchangeable (Moore & Zarske, 1984; Sattler & Bowman, 1981) . Far the same raw score, the two systems may yield standard scores that are as much as I standard deviation apart . For example , a raw score of 9 (for an 8year-old child) is about 1 SD below the mean on the SOMPA norms, and about 2 SD below the mean on the Koppitz norm s . Until further validity evidence is available, caution is needed in making any placement decisions based on either set of norms. ==

Error classifications. The four types of errors in the Koppitz system are as follows: I . Distortion of shape involves destruction of the Ge­ stalt, such as misshapen figures, disproportion between the sizes of the component parts of the figure ; substitution of circles or dashes for dots; substitution of distinct angles for curves or total lack of curves where they should exist; extra angles; or missing angles . Distortion of shape is scored for Figures A, I, 3, 5, 6, 7, and 8, for a possible total error score of 10 points .

2 . Rotation i nvolves a rotation of the figure or any part thereof by 45 degrees or more . Rotation is scored when the Bender-Gestalt card is rotated, even if it is copied correctly as shown on the rotated card . This error is scored for Figures A , 1 , 2 , 3 , 4, 5 , 7 , and 8 , for a possible total error score of 8 points .

3 . Integration involves (a) failure to connect properly the two parts of a figure , either by leaving more than Ys i nch between the parts or causing them to overlap ; (b) fai lure to cross two lines or crossing them in an incor­ rect place ; or (c) omission or addition of rows of dots or

loss o f the overall shape in the case of figures composed o f dots o r circles . Integration difficulties may b e scored for Figures A, 2 , 3 , 4, 5 , 6 , and 7 , for a possible total error score of 9 points .

4 . Perseveration i nvolves i ncrease , continuation, or prolongation of the number of units in the design. It is scored for three of the designs: (a) when there are more

Table 1 4- 1 Classification of Errors i n the Koppitz Scoring System for the Bender-Gestalt

Figure

Error

Figure A

la. Distortion of shape l b . Distortion of shape (disproportion) 2 . Rotation 3 . Integration 4 . Distortion of shape (circles for dots) S . Rotation (j Pel'severation 7 . Rotation S . Integration (row added , omitted) 9. Perseveration 1 0 . Distortion of shape (circles for dots) 1 1 . Rotation 1 2a . I ntegration (shape lost) l 2b . I ntegration (lines for dots) 1 3 . Rotation 1 4 . Integration I S . Distortion of shape (circles for dots) 1 6 . Rotation 1 7a . Integration (shape lost) 1 7b . I ntegration (lines for dots) I Sa . Distortion of shape (angles in curves) 1 9b . Distortion of shape (straight line) 1 9 . Integration 20. Perseveration 2 1 a . Distortion o f shape (disproportion) 2 1 b . Distortion of shape (incorrect angles) 2 2 . Rotation 2 3 . Integration 24. Distortion of shape (incorrect angles) 2S. Rotation

Figure

1

Figure 2 Fi gure 3

Figure

4

Figure

S

Figure

6

Figure

7

Figure

S

NOTe. See Koppiu ( 1 964) for a description of the complete scoring system.

364

CHAPTER 1 4

ASSESSMENT OF VISUAL-MOTOR PERCEPTION. AUDITORY PERCEPTION. AND MOTOR PROFICIENCY

you SEe -

- AT THIS AG£ THEY CANT COpy A DI AMOND PROPERLY !

Cartoon by Tony Hal l from Of Children. An Introduction to Child Development, Third Edition by Guy R. Lefrancois © 1 980 by Wadsworth, Inc . , Belmont, California 94002. Reprinted by permission of the publisher.

than 15 dots in a row in Figure 1 ; (b) when there are more than 14 columns of circles in a row in Figure 2; and (c) when there are 6 or more complete curves in either direction in Figure 6. A total error score of 3 points is possible. This type of perseveration is known as within­ card perseveration. There is a second, much rarer type of perseveration called card-to-card perseveration, which oc­ curs when a preceding design or parts of It influence succeeding designs . This type of perseveration is not scored in the Koppitz system. - -

-

.. _

- - -

Emotional indicators. There are 12 emotional indica­ tors in the Koppitz scoring system. These indicators evalu­ ate the child's visual-motor performance qualitatively in regard to emotional stability. Little is known about the validity of these indicators, however. Standardization. The 1975 Koppitz norms are based on a sample of 975 elementary school children, aged 5-0 to 11-11 years, living in rural areas, small towns, suburbs, and large metropolitan centers in the West , South, and North­ east. The composition of the sample was 86 percent white , 8 . 5 percent black, 4 . 5 percent Mexican-American and Puerto Rican , and 1 percent Asian. The sample was not representati ve of the country, as its geographic distribution

was highly skewed in favor of the Northeast. The socioeco­ nomic characteristics of the sample were not reported .

Reliability. With samples of 19 to 193 children in kinder­ garten through sixth grade, test-retest reliabilities for the Developmental Scoring System total score range from . 50 to . 90 (Mdn rxx = . 77), with intervals ranging from the same day to 8 months (Koppitz, 197 5 ) . These reliabilities are not sufficiently high to warrant the making of diag­ nostic decision s . They do appear adequate for formulating hypotheses about visual-motor ability, however. Interscorer reliabilities are highly acceptable, ranging from . 79 to . 99 (Mdn r = . 91) (Koppitz, 1975 ) . These reliabilities were based on samples of normal and mentally retarded children . The test-retest reliabilities of the four separate error scores (distortion , rotation, integration, and persevera­ tion) are much lower than those ofthe total score (ru = . 83 for total score, and rxx's of . 29 to .62 for error scores) (Wallbrown & Fremont, 1980) . Therefore, the focus in interpreting the Bender-Gestalt should be on the total score and not on the individual sources of error. Validity.

The validity of the Developmental Scoring Sys­ tem depends on how the test is used . Various types of valid­ ity are now considered.

Perceptual-motor development. When used as a test of perceptual-motor development in children, the Bender-

365

BENDER VISUAL MOTOR G ESTALT TEST

Gestalt appears to have acceptable validity. Copying errors decrease steadily between ages 5 and 9 years, suggesting that the test is sensitive to maturational changes. For chil­ dren over 8 years of age, however, the Developmental Scoring System distinguishes only those with below­ average perceptual-motor development from those with normal development, because most children obtain near­ perfect performance after 8 years of age (Koppitz , 1964) .

Other measures of visual-motor perception. Concur­ rent validity for the Developmental Scoring System has been established by correlating it with various tests of v i sual-motor perception (Koppitz , 1975) . Correlations with the Frostig Developmental Test of Visual Perception range from . 39 to . 56 (Mdn r = .47 ) , and correlations with the Developmental Test of Visual-Motor Integration range from . 59 to .73 (Mdn r = . 65) (Breen, 1982; De­ Mers , Wright & Dappe n , 198 1 ; Krauft & Krauft, 1972 ; Lehman & Breen, 1982; Porter & Binder, 1981; Skeen, Strong , & Book , 1982; Spirito, 1980 ; Wright & DeMers, 1982 ) . Correlations with a variety of i ntelligence tests range from - . 19 to - .66, (Mdn r = - . 48) (Koppitz, 197 5 ) . (The negative correlations occur because the Developmental Scoring System yields error scores - higher error scores are associated with lower intelligence test score s . )

Measures of intelligence.

Studies indicate that the Bender-Gestalt has low to moderate (below "- .40) correla­ tions with measures of reading , arithmetic, and school grades for elementary school children (Blaha, Fawaz , & Wallbrown, 1979; Caskey & Larson, 1980; Fuller & Wall­ brown, 1983 ; Koppitz , 1975 ; Lesiak, 1984 ; Vance, Fuller, & Lester, 1986) . These findings suggest that the rela­ tionship between the Bender-Gestalt and academic skills is too weak to warrant the use of the Bender-Gestalt for predicting school achievement for any individual chi ld. Successful academic achievement depends on many fac­ tors, including cognitive ability, level of maturity, experi­ ence , school atmosphere , pedagogy, perceptual-motor de­ velopment, motivational variables, and familial factors; perceptual-motor skills, therefore, comprise only one part of the total picture .

Educational readiness.

The Bender-Gestalt as a Diagnostic Tool The Bender-Gestalt is frequently used to assess brain damage . Although signs have been proposed for the diag­ nosis of brain damage and even psychosis, there are no specific pathognomonic signs definitively associated with

brain damage or psychosi s . The Bender-Gestalt should

never be used alone to make a diagnosis of brain damage orpsychosis. It is a tool for evaluating visual-motor ability, and inadequate visual-motor ability may or may not be reflective of brain damage. For the assessment of brain damage, the Bender-Gestalt should be used in conjunction with a battery of neuropsychological tests (see Chapter 22) . The following sections emphasize further cautions that apply to the use of the Bender-Gestalt in evaluating brain damage, mental retardation, psychosis, and various other handicapping conditions .

Brain damage.

Some signs that may be associated with brain damage, particularly with children older than 1 1 years o f age, are a s follows (Marley, 1982 ) . 1 . Sequence confusion - changing directions three or mQre times (a directional change is noted when the order is different from the expected or logical progression ) . Example :




. .

Jrr . . . .

,

•

.. .

-

.

·

•

#I

•

. . , .. •

• •

•

•

•

,

0

•

. -:. . . . .

· "

·

.

..

..

-

2 . Collision - crowding the designs or allowing the end of one design to touch or overlap a part of another design. Example:

·

·

..

.. ~ ..

. .

'

3 . Superimposition of design - drawing one design (or more) directly on top of another design. Example :

CHAPTER 1 4

366

ASSESSMENT O F VISUAL-MOTOR PERCEPTION, AUDITORY PERCEPTION, AND MOTOR PROFICIENCY

4 , Workove r - reinforcing a line or lines of a part or whole design. Example :

design, or omitting some elements of a design , Example:

5 . Irregular line quality -drawing irregular lines, par­ ticularly when observable tremor is present during the drawing of the lines. Example:

1 2 . Retrogression - substituting solid lines or loops for circles; substituting dashes for dots, dots for c ircles, or circles for dots; filling in circles; or a mixture of the above . Example :

00

..

6. Angulation difficulty - drawing a design that shows an increase, decrease , distortion, or omission of the an­ gulation on any figure . Example :

7 . Perseveration - redrawing an entire design or a part thereof. Example : Q

0 0

o

0

C)

t they are white, then green , then red, red, red' I want to paint ' " He goes to the easel and quickly snatches up a smock. Sliding in beside Wayne, he whispers to him caressingly and persuasively, "Way ne, you want blue? I g ive it to you okay� You give me red because I'm going to make cherries, lots of red cherries '" After the boys e xchanged paint jars, Winky sits erecl . and with a sigh of contentment starts quickly but with clean strokes to ease his brush against the edge of the jar. He makes dots all around the outer part of the paper. His tongue l icks his upper l ip, his eyes shine, his body is quiet but intense. The red dots are big, wel l-rounded. ful l of color, and clearly separated. While working , W i nky sings to himself, "Red cherries , big, round red cherries ' " The first picture completed, he calls the teacher to hang it u p to dry. The next picture starts as the fi rst d id , dots at the outside edge, but soon filling the whole paper. He uses green too, but the colors do not overlap. Still singing his l ittie phrase , W inky paints a third and fourth picture, concentrating intently on his work. The other childre n pick up his song and Wayne starts to paint blue dots on his paper. Waving his brush, Winky asks, "Wayne , want to try my cherries?" Swiftly and jubilantly he w i shes his brush across Way ne's chin. Laughing, he paints dots on his own hands. "My hands are full of cherries , " he shouts. He runs into the adjoining room cal ling excitedly to the children, "My h a nds a re full of cherries ! " He strides into the bathroom emphatically to wash his hands. Susie fol lows him i n , cal ling, " Let's see , Winky." "Ha, I ate them all," he gloats as he shows his washed hands with a sweeping movement. He grabs a toy bottle from the shelf, fills it with water and asks the teacher to put the nipple on. He l ies down then on a mattress and contentedly sucks the bottle , his face softly smiling , his eyes big and gazing into space, his whole body limp with satisfaction. Source: Cohen and Stern ( 1 970, p. 34).

i nterval recordin g . In this case , lime sampling is used to refer to a p rocedure in which brief observations are made at specified times during the day or at random t i mes; it requires that the target behav ior have a moderate to high frequency of occ u rrence. Inlerval recording , in contrast , is used to refer to a procedure i n which a spec ific observa­ tional period - for example, 15 or 30 m i nutes - is divided i nto a speci fied n u mber of i nterval s . )

'Iypes of inter; al recording.

There are several di fferent

i nterval recordI ng procedure s .

\ . I n panial-il1lerval lime sampling, a behavior is re­ corded (scored) only once r egardless of how long it lasts or how many t imes it occurs in the interval . This is the most popular interval recording method (see Figure 1 7-2 ) . Par­ ticularly useful for behaviors that occu r fleetingly, it w i l l reveal the consi stency o f behavior. 2 . In whole-interval lime sampling , a behavior is scored only when it occurs at the beg i nning of the i nterval and lasts throughout the entire interval . This method is particu­ larly useful when you want to know which behaviors (such as out-of-seat behavior) are performed without interrup­ tion (or continuously) during an interval . 3 . In point-lime interval sampling, a behavior is scored olliv when it occurs at a specific time (or t i mes) during the interval . Recording a spec ific behavior if it occurs only during the fi rst 10 seconds of each hour, but not during the remaining 59 minutes and 50 seconds, i s an example of point-time sampl ing . This procedure allows you to ob­ serve behavior for brief periods at diffe rent times during the day. When the procedure i s used with g roups of chil­ dren, a rotat ional system can be set up i n which each child is observed in l u rn . 4 . In momentary time interval sampling, a behavior is scored only i f it Occurs at the moment the interval ends . For example. if the interval is 10 seconds , only behaviors that are observed at the end of the 10-second i nterval are scored . Th i .· proced u re can be used for observing g roups of children. For f i ve children, a 50-second observation cycle mi ght bc set up, with a diffe rent one of the five ch ildren be ing obscrved at the end of each IO-second interval within the 50-second cycle . This variant is useful with behaviors that occur at moderate but steady rates . Examples incl ude tics, hand move ments, thumbsuc king , frequent stereotypic behavior, and facial expressions . 5 . I n variable inleroccasion il1lerval lime sampling , only behaviors that occu r at preselected random t ime inter­ vals are scored (that i s , the time between occasions of observations is randomly varied) . I n stead of always re­ cording behaviors during the fi rst m i nute of each hour (a fixed interval ) , one might randomly designate a one­ mi nute observation period for each observation hou r. Thus for a six-hour observation period, a random one-m i n ute observation schedule might be as fol lows : first hou r - 30th minute to 3 1 st mi nute second hour - 12th minute to 13th m inute third hou r - 51st minute to 52nd m i n ute fourth hou r - 2nd mi nute to 3 rd m inute

482

CHAPTER 1 7

ASSESSMENT OF BEHAVIOR BY OBSERVATIONAL METHODS

"'" Exhibit 1 7-2 A Narrative Running Record

7 : 00 . Mrs. Birch said with pleasant casualness , "Raymond, wake u p . " With a little more urgency in her voice she spoke again: "Son, are you going to school today?" Raymond didn't respond immediately. He screwed up his face and whimpered a l ittle . H e lay sti l l . His mother repeated, "Raymond , wake up." This was said pleasantly ; the mother was apparently in good spirits and was w i l l ing to put up w ith her son's rei uctance . Raymond w h impered aga i n , and kic ked his feet rapidly in protest . He squirmed around and rol led over crossways on the bed . H i s mother removed the covers.

Raymond wore a T-shirt and pajama pallls. He again kicked his feet in protest. He sat up and rubbed his eyes. He glanced at me and smiled . I smiled in return as I continued making notes.

Mrs. Birch lOok some clothes from the bureau and laid them on the bed next to Raymond. n1ere were a clean pair ofsocks, a clean pair ofshorts, a white T­ shirt and a striped T-shirt. Raymond's blue-jean pallls were on a chair near the bed. Mrs. Birch colllinued to stand beside the bed. 7 : 0 I . Raymond picked up a sock and began tugging and pulling it on his left foot .

fifth hou r - 8th m inute to 9th mi nute s ixth hou r - 46th m i nute to 47th mi nute This method is useful when you want to obtain a sample of behavior over an extended period of t i me and rule out temporal effect s .

Major uses of interval recording. I nterval recording i s useful for observing behaviors that are overt, that d o not have a dearcut beginning and end, and that occur with moderate frequency, such as once every 10 to 15 seconds . Examples include readi ng, working , sitting, touching ob­ jects, roughhousing, conversing appropriately, shouting, screaming, h itting, playing w ith toys, making noise , s m i l ­ ing, ly ing down, and thumbsucking .

As his mother watched him she said kiddingly, "Can't you get your peepers open?" Raymond stopped putting on his sock long enough to rub his eyes again. He appeared to be very s leepy. He said plaintively, "Mommie ," and continued mum­ bling i n an unintell igible way something about his undershirt. 7 : 0 2 . His mother asked, "Do you want to put this u ndershirt on or do you want to wear the one you have on?' Raymond sleepily muttered something i n reply.

His mother left the room and went into the kitchen. Raymond struggled out of the T-shirt which he had on. He put on the clean striped T-shirt more efficiently. 7 :03 . He pul led on his right sock. He picked up his left tennis shoe and put it on. He laced his left shoe w ith slow deliberation , looking intently at the shoe as he worked steadily until he had it all laced. 7 :04 . He put on his right shoe . He laced up h is right shoe . Again he worked intently, looking at the shoe as he laced it. H is mother called, "Raymond, do you want an egg for breakfast?" in a pleasant, inquiring tone . Raymond responded very sleepil y, " No . " H i s voice showed no irritation or resentment; he j ust answered i n a matter-of-fact, sleepy way, "No . " NOTe. Material i n italics reflects details o f the situation . Source: Barker and Wright ( 1966, pp. 15-17) .

How to design an interval recording.

In designing an i nterval recording you must decide on (a) the number of times you w i l l observe the child, (b) the l e ngth of the observation period , (c) the t i me periods during which observations will be conducted , (d) the type of i nterval recording to be used , (e) the length of the observation i nterval , (f) the length of the recording i nterval , i f needed , (g) the target behaviors to be observed, and ( h ) the method of recording data.

Frequency, length , and time of observation period. The child's age , the setting , and the reason for the assess­ ment w i l l determine in part the number of t i mes you w i l l need to observe t h e c h i l d , t h e length of t h e observation

483

OBSERVATIONAL RECORDING METHODS

Exhibit 1 7-3 Narrative Recording Exercises Exercise I

With a co-observer, observe one child on a playground . Make an anecdotal recording for 5 minutes. If the playground is associated with a school , obtain permission from the adminis­ tration before engaging in this activity. Compare your record w ith that of your co-observer. How similar are the two recordings? What did your co-observer i nclude that you did not , and vice versa? Consider the follow­ i ng questions i n evaluating your narrative recording. How detailed i s your recording of the child's behavior? ( For example, does the recording provide a visual picture of what was happen ing?) What behaviors might you have missed? What led you to record some behaviors and not others? How d id the selling affect the child's behavior? To what extent was the child's behavior representative of that of other children of the same age level? What biases , if any, may have affected your observations? Were your observations primarily of specific details or of genera! behav iors? What hypotheses did you develop about potential prob­ lem behav iors? What specific behaviors would you like to observe more carefully at another time? How did your presence alter the child's behavior? What could you have done to avoid this influence? How d id your narrative record ing contribute to your understanding of the child? Which statements in your recording reflect high , me­ dium, and low inferential judgments? •

•

Table 1 7·A Evaluating the Inference Level of a Narrative Recording

Inference level (high , medium , low)

Statement i . Mrs. Birch said with pleasant

casualness, 2 . "Raymond, wake up." 3 . With a l ittle more urgency in her voice she spoke again: 4 . "Son. are you going t o school today?" 5 . Raymond d idn't respond immediately. 6. He screwed up his face 7 . and whimpered a little. 8 . He lay stil l .

Medium Low Medium Low Low Low Low Low

•

•

•

•

•

•

To answer the last question. construct a form with two column head ings : (a) statement and (b) inference level (high, medium, low ) . Select 30 statements from your report - I O from the beginning, 10 from the middle, and 1 0 from the end . Statements may be complete sentences or sentence frag­ ments. Thus one sentence in your recording may generate one or more statements. Number each statement in your descrip­ tion and place the numbered statement in the fi rst column. Then decide whether the statement reflects a high, medium, or low level of inference . Complete this same form for your co-observer's record ing, and have your co-observer do the same for your recording . (Photocopy i ng the two forms afrer the statements are typed but before the classifications are made w i l l facil itate this process . ) Table 1 7-A illustrates how the form would be completed for the first part of the narrative running record shown in Exhibit 1 7- 2 . Determine the level of interobserver agreement for each statement by calculating the percentage agreement ( number of agreements divided by the number of statements - that i s . 30) .

Write a one- or two-paragraph report describing your ob­ servation. Include information about (a) the child (age , sex , phys ical handicaps , a n d other relevant characte ri stics) , (b) the physical setting i n which the observation took place, (c) the length of time you observed the child , (d) what you observed, (e) the level of agreement with your co-observer, and (f) the im p li�:j!inns f lhe fi ndi ngs (for example , whether the behavior was appropriate or i nappropriate) . Exercise

2

With a co-observer, observe a group of children on a play­ ground. Make a narrative running record of the group's behav­ ior for a 5-minute time period . Follow the guidelines given i n Exercise I for eva luating a recording, but substitute group for child as the focus of your observation. Exercise 3

Compare the recordings obtained in Exercises I and 2 . What are the diflerences between observing one child and observ­ ing a group? What purposes are served by each type of recording? What information do you gain (or lose) in each type of record ing? Write up your analysis in a one- or two­ paragraph report. Exercise 4

Narrative recordings can also be used to observe specific types of behavior in various settings. Study the attachment behavior of 1- to 2-year-olds by observing their behavior

(Exhibit continues next page)

CHAPTER 1 7

484

ASSESSMENT OF BEHAVIOR BY OBSERVATIONAL METHODS

Exhibit 1 7-3 (cont.) when they are left at a day care center or at a nursery maintained by a church or synagogue during a religious service. Obtain approval from the center, church, or syn­ agogue admi nistration before engaging in this activity With a co-observer, make an anecdotal record of (a) one child's behavior at the time his o r her parent leaves, (b) the parent's reaction , (c) the caregiver's behavior, (d) the child's response to the caregiver, and (e) the child's behavior after the parent has gone. If your observation takes place in a church or

synagogue, also observe the child's and pare nt's reactions when the parent returns. ( You should also observe these reactions at a full-time day care center if you can return at the end of the day ) Conduct the observation for at least a 30- to 60-minute period. If time permits, observe other children, but only one child at a time. Be sure that you and your co-observer agree on the child to be observed . A rrive early. Use the guidelines in Exercise I for evaluating your recording .

Class: M rs . Jones Time: 1 1 :00 to 1 1 :03

Referred child: Bill Comparison child: Ted Date: March 2 , 1 986

Passive off-task Disruptive off-task On-task

2 3 4 5 6 7 8 9 10 I I 1 2

Tot .

Behavior

A.M.

5

R

X 0 0 0 0 0 X X X 0 0 X

I

C

0 0 0 0 0 0 X 0 0 0 0 0

1

R

0 0 0 X 0 0 0 0 0 0 0 0

0

C

0 0 0 0 0 0 0 0 0 0 0 0

6

R

0 X X 0 X X 0 0 0 X X 0

11

C

X X X X X X 0 X X X X X

-

Figure 1 7·2. Example of a three-minute partial-interval time sample recording. Abbreviations are as fol lows: R referred child, C comparison child, X behavior observed, 0 behavior not observed, Tot . Total . Each number reflects a lO-second observation period fol l owed by a 5-second pause for recording data. Three types of behavior were recorded: passive off-task behavior, disruptive off-task behav ior, and on­ task behavior. Bill engaged in off-task behavior in 6 of the 12 intervals: 5 of the off-task behaviors were passive . Thus in 50 percent of the intervals he s howed some kind of off-task behav ior. In contrast, Ted had only one interval with off-task behavior ( passive). =

=

=

F igure 1 7·3.

=

=

�

General recording protocol . Alessi ( 1 980, p. 36, with changes in notation) provided the following description of the general

recording protocol : The top hal f has spaces for identify ing information. The bottom hal f contains spaces down the left side for writing in the behaviors being observed. Across the page are numbers with columns of boxes underneath . Each number can refer to either successive (a) interval s , (b) behavior counts. (c) duration measures, or (d) latency measure s . For the interval and behavior measures, data would be recorded with "Xs" (target behavior occurred) or "Os· (target behavior did not occur) in each block. For the duration and latency measures. the actual elapsed time would be entered in the successive boxes ( for example, 1 4"/8"/22"/9") . Each space for writing in a behavior category has six rows of boxes after it: R referred student, C comparison student, T teacher's reaction, P peer reaction. G group reaction. and Sc Scan chec k . Two spaces are provided for summarizing the data recorded in the blocks across the first two rows. =

=

=

=

=

=

Note. The duration recording is entered in the same boxes as the interval data, but it does not refer to any specific intervals . I n this example there were four occasions during the session in which the referred child was out of seat and one occasion when the comparison child was out of seat. The duration recording was made independently of the interval recording. Courtesy of G . J. Alessi and © National Association of School Psychologists.

VVest�rn Michigan University C LASSROOM OBSERVATION R ECORD PROTOCOL Comparis()n ( C ) :

Chelsea

Referred ( R ) : Age:

8-6

Grade:

Third

Schoo l :

Pine Eiementar�

Teacher:

Mrs. Graves 10

Date:

month

Andrea

Ag e :

6

Paraprofessional

Other Observer:

87

year

day

Psychologist

Observer:

8-5

Class Size:

31

Class Type:

regu lar

Time Stop:

I I : 16

Time Start :

I I :09

Total Time

:7

(What quest ions do we want to answer"n: To observe whether Chel sea·s behavior during read ing differs from that of anothe; child. Reason for observation

C1assmom activity a n d explicit rules i n effect at lime of observation:

Activity : Read ing. Rules: I . Work quiet l y : 2 . sit at desks: 3. raise hand for help. (circle one) L large group S small group 0 = one to one =

=

Behaviors' Verbal Off-Task

I

Motor Off-Task

= anent ion to all = positive attention to pupil

Teacher (T)/peer

Grouping situation IG) :

0= independent act F = free time

t± TOI

-

R C T

30

I

45

60

0 0 X 0 X 0 0 X

=

=

=

Observation recording method:

reaction codes:

(a) (b) (c) (d) (e)

negative attention to pupil no attention to pupil neutral attention to pupil

15

X

X

Ao An

Ao Ao

(P)

30

2

45

60

15

30

3

45

60

interval size J.2:. t i me sample: size _ . event count du ration latency

15

30

4

45

0 0 X X 0 0 X X 0 X 0 0 0 0 0 0 0 0 0 0 Ao Ao

Ao An

60

15

30

5

45

60

Ao

P G

�c "

4

15

AA A+ AAo An

R r--X 0

X

Ao

Ao

I C T

0 0 0

0 0 0 0

0 0

0 0 0 0 0 0

X

X

0 0 0 0 X 0 0 0

0

0

0

0 0 0 0 0 0 0 0 0 0

Ao An Ao

P 2

Passive Off-Task

I

G Sc R

C I T

0

0

0 0 0 0 0 0 0 0

X

0

0 0 0 X 0 0

Ao An

0

P

3 On-Task

-

G Sc

3 R I I C T

f---

0 X

X

0

0 0 0 0 X 0 0 0 0 0 0 X X 0 X 0 X X 0 X X X

Ao An An An

Ao

Ao Ao An Ao

X X

0 X

Ao An Ao

P 4

Out of Seat (duration)

-

G Sc R

14"

6" C -

6"

53"

T p

8"

22"

9"

A - A - An A -

G

5

I Sc

Were reliability data collected? No. I f yes. interobserver % agreements = � % . • Inclurle specific behavior definitions on back, as well as comments (strengths . contextual observations, etc . ) .

@

••

S can chec k .

486

CHAPTER 1 7

period , and when the observations w i l l be conducted . An observation session may last from 10 to 30 minutes or longer. Try to time your observations so that you can observe a representative sample of the target behavior. If possible, observe the child on more than one occasion and at different times duri ng the day.

Length of the observation in terval. The length of the observation interval w i l l depend on the target behaviors . The interval length should be geared to the onset and termination of the behaviors under obser vation . An appro­ priate i nterval length w i l l minimize d istortion of the behav­ ioral sequences and frequencies. Short i ntervals are prefer­ able for behaviors that last a short time , such as making excessive noise , pushing other childre n , mouthing ob­ jects , and self-stimulation . Long intervals are useful for behaviors that last a long time, such as arguing excessively, watching an excessive amount of televi s ion , or sleeping in the classroom . A recording inter­ val (an interval in which no observations are made) should always be used whenever the scoring (or recordi ng) w i l l interfere with t h e ongoing observations . Whenever a re­ cording interval is used, some type of cuing device is needed to signal the onset and offset of the observation and recording i ntervals . A lthough various electronic devices may be used, a simple method is to record a tone s ignal ing recording intervals on an audio cassette tape and then listen to the tape via an earjack while observi ng . A lter­ natively, a voice on the tape can be used to signal observa­ t ion and recording i ntervals. For a lO-second observation interval and S-second recordi ng interval , the fol lowing sequence would be used, with the words "first ," "second ," and "third" referri ng to the interval numbe r : "First" (0 seconds) , "Record" (10 seconds ) , "Second" ( I S seconds), " Record" (25 seconds ) , "Third" (30 seconds) , "Record" (40 seconds) , and so forth . Similar cuing systems can be employed when only observation intervals are used . What­ ever the cuing system , it should not i nterfere with the ongoing observation s .

Length of the recording interval.

Target behaviors. Target behaviors should b e selected on the basis of prior narrative recordings, interview infor­ mation , referral questions , or test behav ior. When a pre­ selected coding system is used (see the section in this chapter on observational coding systems), the target be­ haviors will be speci fied in the coding system . Data c a n b e recorded with penc il and paper or w ith electronic recording devices,

Method of recording data.

ASSESSMENT OF BEHAVIOR BY OBSERVATIONAL METHODS

such as the Radio Shack PC-2 , Sharp PC-ISOO , Quasar H H C , Panasonic L i n k , Hewlett-Packard 7 S , or Epson HX-20. F igure 17-3 shows a standard recording form that can be used with various recordi n g methods and codi n g systems . A nother approach is to use a system i n which the data are automatically graphed as they are recorded . When you finish your observation s , you have a ready picture of the child's behavior that can be shared immediately w ith the referral source . Figure 17-4 shows a self-graphing data recording system in which minutes and days are used as the time i ntervals . The data coll ected can also be graphed across days, as shown in Figure 1 7- 5 . When you use i nterval recording, you may record the score for the behav ior(s) either during the i nterval or immediately afterward . If you record during the interval , there w i l l be no break between i ntervals; the observation i ntervals w i l l successive. For a lO-second observation in­ terva l , the observation intervals would be as fol lows : observation interval and recording (10 seconds) observation i nterval and recording (10 seconds ) observation interval and recording (10 seconds) (sequence continues) If you record after the i nterval , the observation intervals w i l l alternate w ith intervals for recording behavior. For example , the observation period might consist of a series of cycles in which a lO-second observation interval was followed by a 5 -second recording interval . A typical se­ quence would be as fol lows : observation i nterval (10 seconds) recording interval (5 seconds) observation i nterval (10 seconds) recording interval (S seconds) (cycle repeats) The second method is almost a necessity when you are recording a number of behaviors during an interval , for you must take your eyes away from the child in order to put notations in the appropriate categories. The length of the observation interval - as wel l as the length of the recording interval , i f used - should remain fi xed across all observa­ tions in order to ensure uniformity of the observation s . To observe t h e referred c h i l d , the teacher, a n d t h e class , you might use a sequential procedure i n which you observe fi rst the child, then the teacher, and then the class . A 60second observation period could be div ided in the fol low­ ing way : observe child ( 1 - 10 seconds) observe teacher ( 11-20 seconds) observe c lass (21-30 seconds)

487

OBSERVATIONAL RECORDING METHODS

5 4 3

r-�---+---+---r--+---+-�r--�--1---+-�

2 5

10

15

20

25

30

35

40

45

50

a . Graph paper with series of columns, each fi ve blocks high. Double heavy l i ne marks off 10 columns, for 5

0

0

0

4

0

0

0

3

0

0

2

X

X

X

X

X

5

10

15

20

25

30

35

40

45

a

50-minute period .

50

b. Chart after 1 3 minutes of monitoring pupiJ's behavior. First two columns are completed, and the third is partially completed. If the pupil behaves appropriately during the next ( 1 4th) minute, the observer will mark an " X" in the third column just above the other " X . " I f the pupil misbehaves, the observer will mark a n "0" in that column just under the other two "Os . " 5

0

0

0

0

0

0

0

0

0

0

4

0

0

0

0

0

0

X

0

X

X

3

0

0

0

X

0

X

X

X

X

X

2

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

5

10

15

20

25

30

35

40

45

50

c . Chart after observer has completed 50-minute period. Figure 1 7-4.

Self-graphing data recording system. Alessi ( 1 980, pp. 39-40, with changes in notation) provided the following description

of the graphing method and commentary on the data shown. To set u p a self-graphing data recording system, start with a piece of graph paper. Mark two heavy lines across the paper so that five blocks are between the l ines. You have now a series of columns, all fi ve blocks high. Each block will represent an interval (e.g . . minute) of observation time. Mark off the number of 5-block columns needed for the scheduled observation period : a 50-minute period would need 1 0 columns of 5 blocks, a 30-minute period would need 6 columns, a 45-minute period would need 9 columns, and a 5-minute period would need only one column of 5 blocks. For now let's assume you have scheduled a 50-minute period for your observations, as shown in (a) . You have marked off 10 columns on your paper, each five blocks high, for a total of 50 blocks: one block for each minute scheduled . For each interval (minute) in which the on-task behavior occurs, you will place an "X" in a box . For each interval in which the behavior does not occur, you will place an "0" in a box . Start with the left column and work toward the right. In each column, work from the bottom up with the "Xs," but from the top down with the "0" marks. When the "Xs" and "Os" meet in the middle, the column is filled. Move to the next column to the right and continue: "Xs" from the bottom, "Os" from the top down, until they meet. As you move across the row of 5-block columns, the data recorded will automatically form a graph without any extra effort on your part. With thi s method, trends in data across the session can be easily identified and shared with school personnel , referral sources, or parents. Focusing on the Xs in (c) shows that the amount of on-task behavior by the pupil is steadily increasing during the observation session ( i . e . , there are fewer "Xs" in the first columns , and more "Xs" in the later columns). Reprinted w ith permission of the publisher and author from G . J. Alessi, "Behavioral Observation for the School Psychologist : Responsive-Discrepancy Model," School Psychology Review, 1 980, 9, p. 40. © National Association of School Psychologists.

observe child (31-40 seconds) observe teacher (41-50 seconds) observe class (51-60 seconds) (cycle repeats) Recording i ntervals, if needed, can be i nterspersed with observation period s . The following sequence consists of7-

second obse rvation i ntervals and 3-second recording intervals: observe child ( 1-7 seconds) record behavior ( 8 - 10 seconds) observe teacher ( 1 1 - 1 7 seconds) record behavior ( 18-20 seconds)

CHAPTER 1 7

488

i9 0

0

0

0

0

17 0

0

0

0

0

15 0

0

18 0

16 0 14 0

0 0

0

0 0 0 X

0

X

X

X

X

9 X

X

X

X

13 X

12 X II

10 X 8 X 7 X

X X X

6 X

X

4 X

X

5 X 3 X 2 X X

X X X X

2

Figure 1 7-5.

X

0 0 0

0

0

0

0

0

0

0

0

0

0

0 0

0

0

0

0

0

0

0

X

0

0

0

0

X

0

0

0

X

0

0

0

0

0

X

X

X

X

X

X

X

X

X

X

X

X

X X

X

X

X

3

4

X

0

0

0

X

0 0

X

X

0

0

0

0

X

0

0

0

X

0

0

X

X

0

X

X

X

0

0 X

0 0

X

X

X

X

7

8

5

X

6

X

X

X

X

X

9

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

X

X

0

0

0

0

0

0

X

0

0

0

0

0

X

0

0

0

0

0

0

0

0

0

0

0

0

0

X

X

0

0

0

0

X X

0

0

0

0

0

0

X X

0

0

0

X

X

0

0 0 0

0

0 0

0

0

0

0

0

0

0

0

X

0

0 0

X

X

X

X

0 0

0

0 X

0

0

X

X

X

X

10 I I

X

0

0

X X

0

0

0

X

0

0

0

X

X

ASSESSMENT OF BEHAVIOR BY OBSERVATIONAL METHODS

0

0 0 0

0

0

X

X

0

0

0

0

0 0

0

0

0

0

0

0

X

0

0

0

0

0

0

0

0

0

0

0

0 0

0 0

0

0

0

0

0

0

0 0 0

0

X

X

0 0

0

0

0 0

0

0

0

X

X

0 0 0

0

0

X

X

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

X

X

0 0

0 0

0

0

0

X

0

0

0

0

X

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

X

0

0

0 0

0

0

0 0 0

0 0

0

0

0

0

0

0

0 0

0

0

0

0

0

0

0

0

0

0

0 0 0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

X

0

0

0

0 0

0

0

0

X

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0 0 0

0

0

0 0 0 0

0

0

0

0

0

X

0 X

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0 0

0

0

0 0

0

0 0

0

0

0

0

0

0

0

0

0

0

0

0

0 0 0 0

0

0

0

0 0

0

0 X

0 0 0

0

0 0

0

0

0

0

0

0

1 2 1 3 1 4 1 5 1 6 1 7 1 8 1 9 20 2 1 22 23 24 25 26 2 7 28 29 30

Automatic graphing data collection procedures, with days # I t h rough #30 across the abcissa and intervals

# I through # 1 9 up the ordinate. The heavy l i ne between days #4 and #5 indicates the beginning of an intervention plan . Intervals with talk-outs are scored with "Xs," and quiet intervals are scored with "Os . " By reading from left to right and focusing on the "Xs," you can see a clear downward trend in the number of i n tervals scored for talking out over the 30-day period ( more "Xs" in the earlier columns and fewer "Xs" in the later columns). There is a swift drop in the number of intervals scored for talking out just after the intervention was implemented (day #5) , suggesting the effectiveness of this procedure . Reprinted, with changes i n notation, with permission of the publisher and author from G . 1 . Alessi, "Behavioral Observation for the School Psychologist: Responsive-Discrepancy Model," School Psychology Review, 1 980, 9, p. 4 1 . © National Association of School Psychologists.

observe class (21-27 seconds) record behavior (28-30 seconds) (cycle repeats) Sequential observation procedures permit great flexibil­ ity in recording the behavior of individuals and groups. They also can be used with a variety of behavioral coding systems to fit particular assessment need s .

Quantitative data in interval recording.

The primary p iece of quantitative data obtained in interval recording is the frequency count of the number of intervals in which the target behaviors did or did not occur. Note that the fre­ quency count reflects the number of intervals in which the behavior occurred, not the number of time s the behavior occurred (see Event Recording below ) . I f information o n the intensity o f the behavior i s desired , an intensity dimension can be built into the behavioral

code . For exampl e , if you want to record the intensity of hyperactive behavior, you can include codes representing different degrees of i ntensity ( for example, mild , moder­ ate , and extreme hyperactivity ) .

Advantages of interval recording.

Interval recording

(Kazdin , 1 9 8 1 ; Nay, 1979) may help to define important time-behavior relations faci l itates checking for interobserver rel iabil ity helps to ensure that the predefined behaviors are ob­ served under the same conditions each time uses time efficiently focuses the observer's attention on the child's behavior by structuring the observations permits the recording of virtually any behavior • allows for the collection of a large number of observa­ tions in a short period of t i me • •

•

•

•

489

OBSERVATIONAL RECORDING METHODS

'" Exhibit 1 7·4 I l lustration of I nterval Record ing i n a Classroom

OBSERVING ON-TASK AND OFF-TA SK BEHAVIOR Students and Setting

Three students with attentionaJ problems from a class for the mentally retarded were the focus of the observation . The three children could fol low simple instructions and were achieving academically at a first grade leve l . When assigned an aca­ demic task to complete, they were generally off-task . fre­ quently glancing up from assigned work , turning to watch other children, and playing with objects on their desks. A fourth child was selected as a comparison student. Her teacher reported that she did not have attentional problems . The observation took place in the classroom during math and spell ing periods. There were ) 3 students in the class. The daily curriculum was aimed at developing basic math , spell­ ing, and phonics skills. Specially prepared materials were constructed for the observation period . These materials would again be used after special training procedures were begun and consisted of sheets of simple one-digit math prob­ lems and spelling exercises. During each math session, 1 40 addition and subtraction problems were given to the children, considerably more than any of them could complete. The spelling exercises required children to copy three- and four­ letter word s . During each session, the students were asked to copy two pages of spelling words, eaeh containing 1 6 words . Coding Categories

The on-task behavior had to meet the fol lowing requiremeJlis: (a) the children's buttocks had to be touching the seat bottom

Disad vantages of interval recording.

I n terval re­

cording provides a somewhat artificial view of the behavior sequence (because the time interval , not the behavioral episode , dictates the recording framework) may allow i mportant behaviors related to the problem to be overlooked • does not provide information about the qual ity of the behaviors (that i s , how the behavior is performed) or about the situation (that is, whether the child was agitated . pay­ i ng attention , or sleeping while in his or her seat) unless such information is specifically coded into the recording system may not reveal the actual frequency or duration of the behavior (for exampl e , one 60-second continuous period •

•

•

of the desk and (b) their eyes had to be oriented toward the task materials while (c) they interacted manually with the task material s . Two of the three referred children were observed simultaneously on an alternating basis - that is, every 5 min­ utes a d i fferent pair of children was observed . Thus. during each 30-minute session, each child was observed for a total of 20 minutes. The comparison child was observed separately, once a week for 20 minutes. /\. whole-interval rating system was used, with 1 0-second observation intervals . Any break in eye or manual contact with the task materials or incorrect posturing resulted in that interval's being scored as off-task. Other responses recorded included the percentage of 1 40 math problems completed , the percentage of attempted math problems correct, the percent­ age of 350 letters completed . and the percentage of completed words spelled correctly. Rel iability checks were performed twice during each ses­ sion by a second observer. Reliability for both occurrence and nonoccurrence of on-task behavior was computed o n an interval-by-interval basis. The number of agreements was divided by the number of agreements plus disagreements, and the result was multipl ied by 1 00 . The reliability of the perfor­ mance measures was checked by percentage agreement . Reprinted. with changes in notation. with permission of the publisher and authors from T. L. Whitman, J . W Scibak, K. M . Butler, R . Richter, and M . R . Johnson, "Improving Classroom Behav­ ior in Mentally Retarded Children Through Correspondence Train­ i n g , " Journal oj Applied Behavior Analysis, 1 982 , 15, pp. 557-558. © Society for the Experimental Analysis of Behavior, Inc. Source:

of off-task behavior would be recorded as four separate events in a IO-second observation-5-second recording in­ terval system) • may overestimate the frequency of low- rate behaviors or behaviors of short duration ( particularly during point­ time sampling) and underestimate the frequency of high­ rate behaviors

I llustration of interval recording.

Exhibit 17-4 il lus­ trates how interval recording was used to observe on- and off-task behavior of mentally retarded children in a class­ room . In addition to the data on on- and off-task behaviors , data were obtained on the number of problems completed and solved. The rel iabil ity checks referred to in the il­ lustration are disc ussed l ater in this chapter.

CHAPTER 1 7

490

ASSESSMENT OF BEHAVIOR BY OBSERVATIONAL METHODS

"" Exhibit 1 7·5 Interval Recording Exercises

Exercise I

With a co-observer, observe one child on a playground for 5 minutes . Select a child who appears to be engaged in play with another child. Use a partial- i nterval t i me sampl ing procedure, with 10 seconds for the observation interval and 5 seconds for the recording interval . Use a tape recorder, preferably with earphones, to signal the beginning and end of the observation and recording intervals. Use a two-category coding system: (a) aggressive behavior and (b) nonaggressive behavior. Ag­ gressive behavior is defined in Table 1 7- 1 1 . Mark an X for aggressive behavior and an 0 for nonaggressive behavior, using the recording protocol shown in Figure 1 7-3 . A fter you complete your recording, determine the level of interobserver agreement . Calcu l at e the fol lowing inter­ observer agreement indices: (a) percentage agreement for occu rrence of target behavior, (b) percentage agreement for nonoccurrence of target behavior, (c) percentage agreement for both categories, and (d) kappa . Formulas for obtaining these indices are covered toward the end of this chapter. Consider the following questions in evaluating your inter­ val recording. Which behaviors were observed most clearly (that is, most easily classified)? To what extent might the time of day have affected the child's behavior? To what extent did the behavior setting affect the child's behavior? To what extent were the observational categories repre­ sentative of the child's behavior in those categories? What could be done to improve the representativeness of the observations? To what extent was the child's behavior representative of that of other children of the same age leve l ? •

•

•

•

•

•

Exercises for developing interval recording skills. Some exercises for developing you r skills in interval re­ cording are given in Exhibit 17-5 .

Event Recording In event recording (also referred to as event sampling), each instance of a specific behavior or event is recorded as it occurs duri ng the observation period . Like interval rec­ ord i n g , event recording samples behavior. However, whereas the unit of measure in i nterval recording is the time interval imposed on the target behavior, the unit of measure in event recording is the behavior itself. The

Did the coding categories reveal i n fo r m a t i o n that might have been missed i f only a narrative record ing had been used" What biases. if any. may have affected your obse rvat ions') How did your presence alter the child's behavior" What could you have done to avo id this influence" How did your interval recording contribute to your un­ derstanding of the chi ld') •

Write a one- or two-paragraph report that describes your observation . Include information about ( a ) the child ( age . sex . physical hand icaps , and other re l evant characte ristics ) . (b) the physical setting in wh ich the observation took place . (c) the length of time you observed the child. ( d ) the number of intervals in which the target behav ior(s) occu rred , (e) the level of agreement with your co-observer. ( f ) any d i fficulty in determining when the target behav ior( s ) began and ended . (g) whether the defi nitions of the target behav iors were satis­ factory and suggestions for improving the defi nitions. and ( h ) the impl ications of the findings ( for example, whether the behavior was appropriate or inappropriate) . Exercise 2

Follow the steps described in Exercise I . Using a whole­ interval time sampl ing procedure, observe a different child on the playground . Again choose one who appears to be engaged in play with another child. Exercise 3

Compare the recordings obtained in Exercises I and 2 . What are the di ffe rences between using a partial-interval and a whole- i nterval time sampl ing procedure') Which one g ives you a more accurate picture of the child's behaviors? Why" Write up you r analysis in a one- or two-paragraph report .

observer waits for the preselected behavior (the event) to occur and then records it. Like interval recording, event recording is especially useful for controlled observations and laboratory studies .

Event re c o rd i n g pro­ v ides a continuous temporal record of the observed behav­ iors and thus is particularly appropriate for measuring discrete responses that have clearly defined beginnings and ends . Examples are spelling a word correctly, completing a problem, making a social response ( for exampl e , say ing "hello" or sharing a toy ) , pull ing clothing , acting ag­ gressively, getting out of a seat , using profane words, Major uses of event recording.

49 1

OBSERVATIONAL RECORDING METHODS

toileting , eating, asking questions, having seizures , mak­ ing a speech error, or coming late to class . Event recording is less suitable for high-rate behav iors or for behaviors that vary in duration . For example , hand clapping is a behavior that may occur so frequently that separating each occurrence becomes difficult. Other be­ haviors that may occur too frequently for event recording incl ude roc k i ng movements ; rapid jerks of the head , hand s , or leg s ; running; and tapping objects . An example of a behavior that varies in duration is aggressive interac­ tion w ith peers ; it may be di fficult to differentiate between a 5-second and a 5-minute i nteraction . Other responses that may extend over different periods of time incl ude thumbsuck i n g , reading, and l istening.

How to design an event recording. I n designing an event record i ng you must decide on (a) the number of times you w i l l observe the child, (b) the l ength of the observation period , (c) the time periods during which observations w i l l b e conducted , ( d ) t h e target behaviors to b e observed , and (e) the method of recording data . Frequency, length , and time of observation period. The chil d's age , the setting, and the reason for the assess­ ment w i l l in part determine the number of times you will need to observe the child, the length of the observation period, and when the observations w i l l be conducted . A n observation session may l ast from 10 t o 30 minutes or longer. Try to t ime your observations so that you can observe a representative sample of the target behavior. I f possible, observe the child o n more than one occasion and at different times during the day.

Target behaviors. As i n interval recording, the selec­ t ion of target behaviors is based on prior narrative record­ ings, interview i n formation , referral questions, or test behavior. If you use a predesigned coding system, the target behaviors w i l l be specified in the system (see the next section on Observational Coding Systems) . Re­ member to select behaviors that have an easi l y discernible beginning and end. Method ofrecording data. Responses can be recorded i n various ways , including using a checkl ist , wrist counter, hand counter, e lectromechanical counter, or Datamyte E vent Recorder or transferring small objects from one pocket to another. For paper-and-penci l recordings, vari­ ous methods can be used to make tal l ies. One is the traditional stroke method :

I II III IIII W

Table 1 7·3 Two Paper·and.Pencil M ethods for Recording Frequency of Behavior

Method Behavior

Dot and line

Stroke

0

Utt I I I I IIII II

Agg ression

•

•

Cooperation

•

•

Cry i ng

•

•

Frequency o/behavior 9 4 2

Note. In the dot-and-line method, each dot represents one count and each line represents one count.

Another is the dot-and-Iine method , which is often used when there is l imited space on the protocol : .1,

• •

2,

:

•

3,

: :4,1 :5,I I6, U7,US, t2l9, N10

Table 17-3 i l lustrates these tal l ies. The general recording protocol shown in Figure 1 7 -3 can also be used for event recording . For behaviors that occur frequently, a combination of event recording and a one­ minute interval recording may be the best choice (see Figure 17-6) (Alessi, 1980) . For recordi ng the duration of a behavior, you may use a stopwatch , time clock, wall clock, or some other timing device . Counters can also be used to record both the frequency of an event and its duration . For example, a counter panel that has several keys can be used , with one key assigned to each behavior. You hold the key down for the duration of the behavior, and the panel records the frequency and duration of the behavior on a continuous sheet of paper.

Quantitative data in event recording.

The primary piece of quantitative data obtained in event recording is the frequency count - the number ofoccurrences of a behavior in a given time period. For example, an event recording might y ield the i n formation that "Chris used 10 profane words during a 20-minute observation period . " In addition to the frequency of the behavior, several other behavioral dimensions can be measured in event recording, including the rate of the behav ior, the duration of the behavior, the intensity of the behavior, and the latency of the behavior. Let us consider each of these dimension s . 1 . Rate ofbehavior. T h e rate at w h i c h a behavior occurs during the session is obtained by dividing the number of behaviors by the length of the observation period .

492

Behaviors

Tot .

CHAPTER 1 7

ASSESSMENT OF BEHAVIOR BY OBSERVATIONAL METHODS

2

3

4

5

6

7

8

9

10

11

12

13

14

15

. . : : L: : . . . . . U � 0 0 U L�� �r--+� ����r--1---+--1-�

Talk-outs by event record

. . . .

T Tal k-outs by i n terval record

x

X

X

X

X

X

X

0

X

X

X

X

X

X

X

o

X

0

X

0

X

0

0

0

o

o

X

0

0

0

2 Figure 1 7-6. Comparison of event and interval records of observation conducted in I -m inute intervals. Top part shows event data for talking-out behavior within each interval. Bottom part shows same data as scored by the interval-only meth od . The comparison shows that interval scoring is not as sensitive to the dynamics of the high rate of behavior as is the event-within-interval record. With the event (top) record, one can see a sudden decrease in rate of talking out after minute 7. The interval record is insensitive to this change. Likewise, the d iscrepancy between the two pupils' data is greater as measured by the actual rate (event) measure; it is underestimated by the interval measure. R referred pupil , C control pupil , T teacher. Reprinted, with changes i n notation, with permission of the publisher and author from G . J . Alessi , "Behavioral Observation for the School Psychologist: Responsive-Discrepancy Model , " School Psychology Review, 1980, 9, p . 39. © National Association of School Psychologists. =

=

=

Rate of behavior where

= !2

the total duration of the behavior by the length of the observation period .

t

11 = number of behaviors t = length of observation period

Percentage duration rate of behavior

For example , if a g i rl was observed to be out of her seat 40 times during a 10-minute observation period , her rate of behavior would be as follows : Rate of behavior

=

n t

-

40 10 m i n .

--'--=-= -".

= 4 occurrences

4 I

where

=

t

d = total duration of behavior

1 00

(time spent responding)

t = length of the observation period An average duration rate per response is computed as fol lows :

per minute

Her rate of out-of-seat behav ior was observed to be four times per m i nute . Rate of behavior is a useful index for noting changes in the child's behavior, especially across observation sessions of different lengt h . 2 . Duration of behavior. Duration o f be Kavior refers to how long each occurrence of the behavior lasts - the period between the beginning and the end of the behavior. A duration count might be used, for example , to determine the duration of temper tantrums , crying episodes, argu­ ments, verbal t i rade s , sustained conversations, on-task behavior, out-of- seat behavior, cooperat i ve behavior, thumbsucking, off-task responding, or delays in returning home from school (Gelfand & Hartmann, 1984; Sulzer­ Azaroff & Reese , 1982 ) . Unless the behavior is discrete, a duration count w i l l be d i ffic u l t , i f not impossible, to obtain . In addition to the duration count, there are two other measures of the duration rate of behavior (Cone & Foster, 1982 ) . A percentage duration rate is computed by dividing

4. x

Average duration rate of behavior where

d

=

e

=

=Q e

total duration of behavior (time spent responding) number of episodes of the behavior

Example: Suppose a child has two 3-minute tantrums (two episodes) in a 30-minute observation session on Day I and six I -minute tantrums (six episodes) during a 60minute session on Day 2 . The total duration of the tan­ trums is 6 minutes for both days, but the response patterns d i ffer. Using the formula for the percentage duration rate of behavior, we have Day l : Duration rate of behavior

= 4. x t

1 00

=

6 min. x 1 00 30 mm.

= 2 0 % per session

493

OBSERVATIONAL RECORDING METHODS

Day

2:

Duration rate o f behavior

=

1c

=

10 %

t

x

1 00

=

6 min . 60 mID.

X

1 00

per session

Using the formula for the average duration rate o f behav­ ior, we have Day

1:

' · rate 0f be haVlOr Average d uratlOn

d =e =

Day

2:

Average duration rate o f behavior

= =

=

6 min. --2-

separate freq uency count would be recorded for each category. 4 . Latency o.fbehavior. Latency refers to the amount of time that elapses between the initiation of a request (or between an event known to produce or faci l itate the occur­ rence of a behavior) and the onset of the behavior; it tell s us how long it took the child to begin the behavior. Latency i s usually measured b y using a stopwatch t o determine the time from the initiation of the request to its execution. Latency measures are useful when you need to determine the time it takes a child to begill working after instructions have been given to start, to begi n complying w ith a re­ quest , or to get up after an alarm rings (Sulzer-Azaroff & Reese, 1982) .

3 m i n . per response

Advantages of event recording.

�

1

din, =

6 7n .

min . per response

The percentage method y ields a duration rate of behav­ ior of 20 percent for Day I and 10 percent for Day 2, whereas the average method y ields an average duration rate of behavior of 3 minutes per response for Day 1 and 1 m inute per response for Day 2 . The first method i s preferable when you are interested in how much time the child spends in a particular activity relative to other activities, but care l ittle about the duration of individual instances of the behavior. This method masks the duration per response. The second method is useful when you are i nterested in the average duration of a re­ sponse, such as when you are assessing the average dura­ tion of an appropriate behavior. This method ignores the l ength of the time interval over which the data are collected (Cone & Foster, 1982). If desired , both duration rates can be reported . 3 . In tensity of behavior. The i ntensity of behavior is obtained by dividing the behavior i nto various degrees of i ntens ity, as in interval recording. For example , if you want to record various intensities of aggressive behavior, you can categorize behaviors as slightly aggressive , moder­ ately aggressive , and severely aggressive . I f you are inter­ ested in observing a boy whose teacher has reported that he turns in all assignments but complains, you can create three categories , such as "(1) hands in assignment on time w ith no complaints; (2) hands in assignment late with no complaints; (3) hands in assignment late and complains" (Cone & Foster, 1982 , p. 316) . In both of these examples , a

1981; Nay, 1979)

Event recording (Kaz­

detects behaviors with l ow frequency rates, particu­ larly when observations are made by persons who are ordinarily in the setting • fac i litates the study of many different behaviors or events uses time and personnel efficiently (especially when observations are made by persons ordinarily i n the setting) can accommodate many different recording methods • provides information about changes in behavior over time and about the amount of behavior performed •

•

•

Disadvantages of event recording.

Event recording

provides a somewhat artificial v iew of the behavior sequence by separating the present event from conditions in the past that may have l ed up to it does not reveal sequences or temporal patterns unless the time of the response i s recorded • breaks up the continuity of behavior by using limited categories • is not suited to recording behaviors that are not clearly discrete presents difficulties in establishing rel iability across multiple observers • requires observers to maintain an optimal level of attention over long periods of time , because few cues are used and responses may be relatively infrequent • l imits quantification of the how and why associated with the event, unless thi s information is also recorded makes comparison across sessions difficult if the length of the observation period i s not constant •

•

•

•

lIIustrations of event and duration recording.

Figure 17-7 illustrates how event recording can be used to com-

C H APTER 1 7

494

ASSESSM ENT OF BEHAVIOR BY OBSERVATIONAL METHODS

Date: March 1 , 1 986 Class: M rs . Jones

Referred child: Bill Comparison child: Ted Time of day Day

9:00 to 9. 30

1 1 . 00 10 1 1 . 30

2:00 to 2 . 30

Total (Bill/Ted)

Monday Tuesday Wednesday Thursday Friday

41 1 3/0 41 1 2/0 I /O

3/0 2/0 411 2/0 1 /0

2/0 0/0 1 /0 1/1 0/ 1

91 1 5/0 9/2 511 21 1

1 4/2

1 21 1

4/2

Total

--

30/5

Figure 1 7-7. Example of event recording of inappropriate tal king to another child. This record presents a summary of the observational records for two children: Bil l , the referred child, and Ted , the comparison child. The numbers indicate the number of t i mes B i l l or Ted spoke with another child inappropriately. Numbers for Bill are to the left of the slash, and numbers for Ted are to the right of the slash. The record indicates that during the 7 '12 hours of observation. B i l l spoke to another child six times more frequently than did Ted . His inappropriate behavior occurred most frequently on Monday and Wednesday and at 9 : 00-9 : 30 and 1 1 :00-11:30 A . M . The inappropriate behavior seldom occurred on Friday or at 2 : 00-2 : 30 P . M . Further investigation would be needed to detennine what factors in the chi ld's environment lead to increases and decreases in the inappropriate behavior. I n addition. fu rther observations should be made to determine the stability of the observed behavior pattern.

pare two childre n's i nappropriate talking. Exhibit 1 7 -6 shows how event recording, coupled with a duration count , was used to observe one child's out -of-seat behavior. Notice that a rel iabil ity check was performed .

Exercises for developing event recording skills.

Some exercises for developing your skil l s in event recording are given in Exhibit 17-7.

Ratings Recording I n the rating method , behavior is rated on a scale or checkl ist , usually at the end of the observation period . Rat i ng scales usuall y i nvolve a greater degree of observer subj e c t i v it y than do the other behav i o ra l record i n g methods .

Major uses of ratings recording. Ratings are useful for evaluating the more global aspects of behavior and for quanti fying impressions , such as after a psychometric assessment has been completed. The Behavior and At­ titude Chec kl ist in Chapter 5 is one such rating procedure . Rating scales are also useful for assessing behaviors or products that are difficult to measure directly. For exam­ ple, a rating scale that runs from very poor (I) to excellent (7) can be u sed to rate the legibility of handwriting, quality of arts and crafts products, neatness of a room , or perfor­ mance style during physical exercises or other activities. Results based on ratings can be compared with results

obtained from more specific observational procedures , such as interval or event recording. Such comparisons reveal the consistency of the results across method s . Rat­ ings are valuable in some assessment situations because they are less costly than other methods in terms of t i me and personnel resource s . Ratings also allow you to consider more subtle and unique clues , to overcome some of the fragmentation associated with behavioral counts , and to evaluate a quality and unity in the child's behavior that is inaccessible to more molecular and objective coding sys­ tems . Hence ratings may reveal more subtle aspects of impression formation . The quantitative dimension associ­ ated with rati ngs is sometimes termed "behav ior as a whole . "

How t o design a ratings recording. I n designing a ratings recording you must decide on (a) the number of times you will observe the child, (b) the length of the observation period , (c) the time periods during which the observations w i l l be conducted , (d) the target behaviors to be observed , and (e) the method of recording data . Frequency, length , and time of observation period. As in the other recording method s , the child's age, the setting, and the reason for the assessment will determine in part the number of times you will need to observe the child, the length of the observation period , and when the observations will be conducted . An observation session may last from 10 to 30 minutes or longer. Try to time the

495

OBSERVATIONAL RECORDING METHODS •

Exhibit 1 7·6 I l lustration of Event Recording in a Classroom

EVENT AND D U R ATION RECORD IN G : OBSERVING OUT-Of-SEAT BEHAV IOR Student and Setting

Linda, a 9-year-old girl with an IQ of 70 , had excessive out-of-seat behavior. She attended a primary level special class for the educable mentall y retarded in a publ ic school syste m . The teacher said that Linda functioned educa­ t ional ly at approximately the fi rst grade level and spent most of the day out of her seat . This behavior interfered with the completion of classroom work and distracted other childre n . Observations were held Monday through friday from 9 : 00 A . M . to 9 : 20 A . M . during the math period in the child's classroom. The observation was designed so as not to interfere w ith regular classroom routines. Coding Category

The observer sat against a wall in the classroom, approxi­ mately 10 feet from the child, and observed Linda . One k i nd of inappropriate response - out-of-seat behavior ­ was recorded . This behavior was rated as occurring when the child d id not have her buttocks in contact with the chair seat and body oriented toward her work. A n event record­ ing system was used to count discrete out-of-seat behav­ iors. In addiiion, the total duration of each behavior was recorded using a stopwatch that was started and stopped contingent on each behavior. Rel iabil ity was assessed by having two observers use the rating system to record Linda's behavior simultaneously but independently. These checks were taken a minimum of two times per session. Observer ag reement was calcul ated by d i viding the number of out-of-seat responses scored by the observer w ith the lower number of responses by the number of out­ of-seat responses scored by the observer with the greater number of responses and then multiplying by 1 00 . Dura­ t ion was calcul ated in a similar manner. Source: Adapted from Whitman. Scibak. Butler, R ichter. and Johnson ( 1 982).

Method of recording data. Ratings are recorded on a rating scal e , usuall y with five or seven points. Fol lowing are some typical setups for rating scales:

Example i:

Circle the most appropriate number : 5 2 4 3 highly co- moderately neither co- moderately highly un­ operative cooperative operative unco- cooperative nor unoperative cooperative

Example 2:

Shares toys (circle one number) :

always

2 often

3 occasionally

4 seldom

5 never

Example 3:

Place an X on the l i ne that best reflects your rating. anxious

_

:_:_ : _:_

not anxious

Rating systems can be designed to measure selected antecedents and consequences associated w ith the target behavior. For example, you might ask "When situation Z occurs, how often does M i ke do X?" or "After M ike does X, how often does [other person] react by doing Y?" Rat i ng scal es for these questions are shown in Examples 4 and 5 below.

Example 4. When M ike is asked to read aloud , how often does he throw a temper tantrum? Circle one .

almost ��s

2 frequently

3 sometimes

4 infrequently

5 almost ��r

Example 5:

After Mike throws a temper tantru m , how often do his peers react by laughing? Circle one .

observation period so that you obtain a representative sample of behavior. I f possible , observe the child on more than one occasion and at different times during the day.

Target behaviors. A s in interval and event recording, your selection of target behaviors should be based on prior information from narrative record ings, interviews . refer­ ral questions, or test behavior.

almost always

2 frequently

3 sometimes

4 infrequently

Quantitative data in ratings recording.

5 almost never

The prime source of data i n ratings i s the scale value (or number or score) on the rating scale . A major difficulty associated

496 •

CHAPTER 1 7

ASSESSMENT OF BEHAVIOR BY OBSERVATIONAL METHODS

Exhibit 1 7-7 Event Recording Exercises

Exercise I

Exercise 2

With a co-observer, observe one child on a playground for 5 minutes. Using an event recording procedure, record each time the child engages in play (as a general category) with another child. Play with another child includes parallel , coop­ erative, and uncooperative play (see Table E-4 in Appendix E for definitions) , but not solitary play. Use the dot-and-line method to record the target behavior. After you have completed your recording, determine the level of interobserver agreement by calculating percentage agreement (see the section on procedures for assessing relia­ bil ity toward the end of this chapter) . Also calculate the rate of the target behavior. Consider the following questions in evaluating your event recording .

Follow the same general procedure as described i n Exercise I . Now, however, observe three target behaviors: (a) parallel play, (b) cooperative play, and (c) uncooperative play (see Table E-4 i n Appendix E for definitions ) . Calculate the level of interobserver percentage agreement separately for each of the three target behaviors. Calculate the rate of behavior separately for each target behavior. Follow the guide l i nes in Exercise I .

Which behaviors were observed most clearly (that i s , most easily classified as indicating play)? Did the target behavior occur with sufficient frequency to be observed? To what extent might the time of day have affected the child's behavior? To what extent did the setting affect the child's behavior? To what extent was the target behavior representative of the child's behavior during the observation? What could be done to improve the representativeness of the observations? To what extent was the child's behavior representative of that of other children of the same age level? What biases, if any, may have affected your observations? How did your presence alter the child's behavior? What could you have done to avoid this influence? How did your event recording contribute to your under­ standing of the child? •

•

•

•

•

•

•

Write a one- or two-paragraph report that describes your observation. Include information about (a) the child (age, sex , physical handicaps , and other relevant characterist ics) , (b) the physical setting in which the observation took place, (c) the length of time you observed the child, (d) the frequency of the target behavior, (c) the level of agreement with your co­ observer, (f) the difficulty of determining when the target behavior began and ended. (g) whether the definition of the target behavior was satisfactory and suggestions for improv­ ing the definition, and (h) the implications of the findings (for example , whether the behavior was appropriate or i nap­ propriate) .

Exercise 3

Compare the recordings obtained in Exercises I and 2. What are the differences between observing play as a general cate­ gory, as in Exercise I , and observing d i fferent types of play, as in Exercise 2? What purposes do each type of recording serve? What information do you gain (or lose) with each type of recording? Which recording is more reliable and why? Write up your analysis in a one- or two-paragraph report . Exercise 4

With a co-observer, observe a child in a preschool for a 30minute period. Obtain permission from the school adminis­ tration before beginning this activity. Select a child who appears to be engaging i n inappropriate behavior (a target behavior) , such as fighting, temper tantrums, disruptive be­ havior, or uncooperative behav ior. Record each time the inap­ propriate behavior occurs, using the dot-and-line method. Also observe whether the child's inappropriate behavior receives attention from an adult in the room (a target behav­ ior). This information w i l l provide some indication of the consequences of the behavior. Record each time the child receives attention , using the dot-and-line method. You r rec­ ording form should have spaces for recording the frequency of the child's inappropriate behavior and the frequency of the adult's attention . Calculate the level of interobserver percentage agreement separately for the two target behaviors . Calculate the rate of behavior separately for the two target behaviors . Follow the guidelines in Exercise I . Exercise 5

Follow the same general procedure as described in Exercise I . Now, however, record the duration of the child's play with another child. Use a stopwatch o r other device to record the elapsed time. Calculate the level of interobserver agreement, the duration rate of behavior, and the average duration rate of behavior.

497

OBSERVATIONAL RECORDING M ETHODS

with ratings is that the assumptions underlying the scale values are not always clear; therefore observers may differ in their interpretation of the scale positions. Providing detailed examples of behaviors associated with each scale point will help observers to apply consistent standards in interpreting scale values . Ratings should always be made shortly after the completion of the observation session . Establishing ratings from memory creates opportunities for distortion and omissions .

Advantages of ratings recording.

The ratings method

is s u i ted to recording many d i ffe rent kinds of behaviors can be used to rate the behaviors of many individuals or of a group as a whole records subtle aspects of behavior generates data in a form suitable for statistical analysis uses time efficiently •

•

•

•

Disadvantages of ratings recording .

The ratings

method uses scale values which may be based on unclear assumptions may have low interrater rel iability because of complex or ambiguous terms and scale positions that are interpreted differently by different observers is not suited to recording important quantitative infor­ mation , such as the frequency, duration, or latency of behavior is not suited to recording antecedent and consequent events, unless a method for doing so is built into the design of the ratings recording may be inaccurate if there is a time delay between the actual behavior and the observer's rating •

•

•

•

•

Illustration of ratings recording.

The behavioral rating scale shown in Table 1 7-4 was developed to assess the p resence of distress behaviors (pain and anxiety) in chil­ dren undergoing a painful medical procedure - bone mar­ row aspiration treatment for cancer. The scale can be completed at various times during the procedure. The observers do not participate in the treatment procedure; they position themselve s so that they are unobtrusive but still have a c lear v iew of the child as he or she l ies prone on the treatment table . This scale can also be used for rating children's reactions to other painful medical procedures.

Exercises for developing ratings recording skills. S ome exercises for developing you r ski l l s in ratings rec­ ording are given in Exhibit 1 7-8.

Table 1 7-4 Procedure Behavior Checklist Directions: Rate each behavior using the following S-point scale.

very mild

2 mild

3 neutral

5 extremely i ntense

4 intense

C i rcle one number for each behavior. Behavior I . Muscle tension 2 . Screaming 3. Crying 4 . Restraint used S . Pain verbalized 6 . Anxiety verbalized 7 . Verbal stall ing 8. Physical resistance

Ralillg 2 2 2 2 2 2 2 2

:;

3 3 3 3 3 3 3

4

4

4 4 4 4

4

4

5 5 5 5 5 5 5 5

NOIe. Behaviors are defined as fol lows: ( I ) Muscle tension - contraction of any observable body part ( for example, shuts eyes tight, clenches jaw, stiffens body. clenches fists, or grits teeth); (2) Screamil/g - raises voice or yel l s with sounds or words; (3) Crying - d isplays tears or sobs; (4) Restraifil used- is held down by someone or has heavy tape placed across leg onto table; (5) Pain verbalized- says ··ow." " ouch," or com­ ments about hurting ( ior example, "You're hurting me" ) ; (6) Anxiety verbali�ecl- says '"I'm scared" or "I'm afraid" : ( 7 ) Verbal stalling - ver­ . bally expresses desire to delay ("Stop." ''I'm not ready," . ( want to tell you something . " etc . ) ; ( 8 ) Physical resistance - moves around, will not stay in position. or tries to climb off table. See Katz. Kellerman , and Siegel ( 1 980) for additional items. Source: Adapted from LeBaron and Zeltzer ( 1 984).

Interval, Event, and Ratings Recording Methods: Advantages and Special Considerations Interva l , event , and ratings recording methods may not provide the richness of information that narrative record­ ings do, but they allow you to evaluate systematically specific behaviors of interest , sample a large number of children and a variety of situations , compare children and develop norms, and generalize findings, all within a rea­ sonable period of time. I n both interval and event recording it is relatively easy to tal ly behaviors , particularly when they are clearly de­ fined and observable . Both methods provide information about behavior d u ring one time period and about changes in the child's behavior over time. When you wish to obtain information about behavior across time intervals (or temporal patterns of behavioral occurrences) , interval recording procedures should be considered. Interval recording can answer a question such as "Did Tom's off-task behavior occur throughout the ob­ servation period or j ust during part of the observational

CHAPTER 1 7

498

ASSESSMENT OF BEHAVIOR BY OBSERVATIONAL METHODS

'" Exhibit 1 7-8 Ratings Recording Exercises

Exercise 1

RATING SCALES

Directions: Place an X i n the appropriate space. I . cooperative

2 . sad

3 . active 4 . coordinated 5 . aggressive

What could be done to improve the representativeness of the observations? To what extent was the child's behavior representative of that of other children of the same age level? Did the scales reveal information that might have been m issed with a narrative recording? What biases, if any, may have affected your observations? How did your presence alter the child's behavior? What could you have done to avoid this influence? How did your ratings recording contribute to your u nder­ standing of the child? •

With a co-observer, observe one child on a playground for a period of 5 m inutes. A fter you observe the child, complete the fol lowing rating scales.

-'-"-'_.-'-"-'_.- ' -' - ' _ . -'-'-'-'-'-'-'_.-

uncooperati ve happy inactive uncoordinated passive

A fter you complete the scales, convert each rating to a number, assigning the number I to ratings in the left-most column and the number 5 to ratings in the right-most column. Determine the level of interobserver agreement by calculating (a) the percentage agreement (that is, how many ratings were exactly the same across the five scales) and (b) the product­ moment correlation (see Chapter 2 ) . Consider the following questions in evaluating your ratings recording. How did the rating scales guide your observations? What additional scales would have been useful? To what extent might the time of day have affected the child's behavior? To what extent did the setting affect the child's behavior? To what extent were the dimensions covered in the scales representative of the child's general behavior? •

•

•

•

•

interval?" Interval recording can be enhanced by making the length of the interval as c lose as possible to the duration of the behavior. Generally, i nterval recording provides a sample of behavior adequate for many clinical and psycho­ educational purposes , particularly when the concern is whether a behavior occurs . Infrequent momentary behav­ iors have a reasonable chance of being scored with interval recording if observation sessions are long . Event recording is somewhat more useful than interval recording when you want a measure of the number of t i mes a behavior is performed . Such a measure is i m portant when either an increase or a decrease in certain behaviors is an intervention goal . Event recording is especi a l l y useful when t h e characteristics o f an event o r behavior are of interest , such as the frequency of the behavior itself. Event recording, however, is not as useful as i nterval

•

•

•

•

•

•

Write a one- or two-paragraph report that describes your observation . Include information about (a) the child (age, sex , physical handicaps , a n d other relevant characteristics ) , (b) the physical setting in which the observation took place, (c) the length of time you observed the child, (d) the ratings you made, (e) the level of agreement w ith your co-observe r, (f) the difficulties you had in using the rating scales and suggestions for improving the scales, and (g) the implications of the fi ndings ( for example , whether the behavior was appro­ priate or inappropriate ) . E xercise 2

Design your own ratings recording procedure for observing children in some setting. Develop five rating scales different from those used in Exercise I . With a co-observer, observe one child or a g roup of children, depending on the specific procedure developed, for a period of 5 m inutes. Follow the procedures in Exercise 1 for evaluating your recording.

recording for behaviors that do not have a discrete begin­ ning and endin g .

O BSE RVATI O N A L CODI N G SYST E M S

Observational coding systems are used to categorize be­ havioral observation s . They usually consist of two or more categories that cover a wide range of behaviors, although on occasion a s i ngle category may be appropriate . Even a one-category system, however, impl icitly contains another category - the nonoccurrence of the target behavior. Thus even though the focus is on one behavior, a one-category cod i ng system used in an interval or event recording method can be conceptualized as having two categories ­ the presence of the behavior and the absence of it .

499

OBSERVATIONAL CODING SYSTEMS

Before you use a coding system, you should carefully evaluate the following areas (Nay, 1979) : (a) its rati, nal e , (b) the setti ng(s) i n which i t i s applicable, ( c ) definiti ons of the coding categories, (d) description of how beha1llior is sampled , (e) rules governing the behavior of obse rvers, such as hierarchy of codes, (f) reliabil ity, including types of rel iabi lity, overall rel iability, and rei iability of each codi ng category, (g) val idity, and (h) positive and negative features (including potential problem areas of the coding system) .

How to Design an Observational Coding System I n designing an observational coding system, select cate­ gories that best meet the assessment , treatment , or re­ search goal s . Categories range from those that are primarily global, i ncorporating many different specific behaviors or requiring inferential j udgments , to those that are narrow or specific, focu s i ng on only one or a few behaviors . The two-category system i n Table 17-6 contains global behavioral categories : on-task behavior and off-task behav ior. In contras! the IO-category system in Table 17-6 ' employs narrow categories that relate to specific class­ room behaviors . I n selecting (or designing) an observational coding sys­ tem , consider the following: What questions do you hope to answer with your observational assessment? Are you interested in investigating global areas of behavior or j ust one or a few specific behaviors? How many behaviors do you want to observe? A re t h e behav i o r s you want to observe eas i l y discriminated? •

•

•

•

Both global and narrow target behaviors should be defined as completely as possible to help the observer detect behav­ iors and distinguish one behavior from another. Global c ategories subsume a host of i n d i vidual behav iors , whereas narrow categories focus on only one or a few behav iors . When global categories are used, each indi­ v idual behavior must be classified into one of the global categorie s . Use t h e simplest possible coding system that w i l l answer the questions posed. I f your purpose is to obtain a general description of behavior, use global categories. If you are i nterested in only one or a few behaviors related to the referral question, use a detailed analysis of these behav­ iors. If you want to examine the relationship between a behavior and its environmental determinants , use a multi­ d i mensional system that includes relevant antecedent and consequent events . Finally, if you want to record sequential

observational data , use the sequential observational pro­ cedures described by Bakeman and Gottman ( 1986) . Generally, when there are j ust a few behaviors to be measured , devising an adequate recording system is not difficul t . Coding systems that require the observer to make a number of decisions and use many different categories should be avoided because they are difficult to use . Try to memorize the coding system before the formal observation begins, but keep the code definitions handy in case you forget them . Nu merou s qual itative observational coding systems have been developed to meet both general and specific needs. i nc l uding the observation of individual childre n , groups , a n d classes i n a variety o f settings . Observational codes can also cover environmental responses to the child's behavior. Table 17-5 l ists observational coding systems that are useful for a variety of purposes. Tables 17-6, 17-7, and 1 7 - 8 i l lustrate coding systems for observing children , teachers, and c lassrooms. The systems illustrated in these tables require immediate, not retrospective . observation ; the observer must observe and record behavior as it oc­ curs, keeping inferences to a minimum.

Examples of Coding Systems for Observing Children's Behavior Table 17-6 shows several coding systems for observing children's behavior. The two-category coding system is one of the simplest systems for observing the behavior of a referred child or group . The information it provides about on-task and off-task behavior is useful particularly when the general cl imate of a c lassroom or other facil ity is the focus of assessment. Distinguishing between the two cate­ gories in the system may require some inferential j udg­ ments on the part of the observer. The three-category system is a refi nement of the two­ category system, with off-task behavior divided into pas­ sive and disruptive off-task behavior. This refinement is useful for the assessing the passive and active dimensions of i nappropriate behavior. These two dimensions are sim­ ilar to the internalizing (passive) and externalizing (dis­ ruptive) dimensions of child behavior found on behavioral checkl ists . (See Chapter 15 for coverage of behavioral checklists . ) This three-category system can be used for individuals as wel l as for an entire class. The Jour-category system breaks down the disruptive off-task behavior category into verbal and motor off-task component s . This system is useful when information i s desired about whether the off-task behavior is verbal , motor, or passive. It is often used for observing individual children i n a classroom.

CHAPTER

500

17

ASSESSMENT O F BEHAVIOR BY OBSERVATIONAL METHODS

Table 1 7-5 Some Observational Coding Systems

Observational system

Authors Achenbach & Edelbrock ( 1 98 1 ) Conger ( 1 984)

Dunn, Barker, & Wahler ( 1 98 1 )

Fagot ( 1 984) Furey & Forehand ( 1 983) Jay & Elliott ( 1 98 1 ) Kirschenbaum, Steffe n , & D'Orta ( 1 97 8 ) Mash, Terda l , & A nderson ( 1 973) Reid ( 1 978) Roberts , M i l ich, & Loney ( 1 984)

Child Behavior Checklist - Direct Observation Form : A 9 6- i t em observation form that parallels tpe parent and teacher versions of the Child Behavior Checklist . Social Interaction Scoring System: A group of seven different coding systems emphasiz­ ing different aspects of social interactions, including exchanges within fam il ies. Codes are for six types of interactions, seven types of emotional affects, and five types of persons. Standardized Observation Codes: A 29-category coding system (a revision o f the Wahler, House , & Stambaugh , 1 976, coding system) designed to sample interchanges between a child and the child's adult and peer associates . Interactive Behavior Code: A 5 1 -category behavioral observation code designed to assess a child's play preferences. Daily Child Behavior Checklist: A 65-item checklist of pleasing and displeasing behav­ iors that may have occurred in the preceding 24 hours. Parent completes checklist. Observation Scale ofBehavioral Distress: A n I I -category observational code for record­ ing anxiety or pain in children undergoing painful medical procedures. Social CompeTence Classroom Behavioral Observation System: An I I -category behav­ ioral observation system , including 5 categories of task-irrelevant behavior, 3 catego­ ries of task-relevant behavior, and 3 categories of prosocial behavior. Response-Class Matrix: An observational coding system for recording mother-child interactions, with 7 categories for the mother and 7 categories for the child. Behavioral Observation Code Used with Families: A 29-category observational code for recording family interactions. StrucTured Observation of Academic and Play Settings: A 7-category structured play­ room observation procedure for evaluating hyperactivity.

Note. The observational code systems for Furey and Forehand

( 1 983); Kirshenbaum, Steffen, and D'Orta ( 1 9 7 8 ) ; Mash, Terdal , and Anderson ( 1 973 ) ; and Reid ( 1 978) are available i n the publications (see the reference section a t t h e end o f t h i s text ) . The other observational systems c a n b e obtained directly from the authors. Their addresses are as follows: Thomas M. Achenbach, University of Vermont, Burlington, VT 05405; Rand D. Conger, College of Home Economics, Iowa State University, Ames, 10 500 I I ; Elizabeth S. Dunn, Child Behavior Institute, University of Tennessee, Knoxville, TN 37996; Beverly I . Fagot, Department of Psychology, University of Oregon, Eugene, OR 97405; Susan M. Jay, Psychosocial Program, Division of Hematology-Oncology, Children's Hospital of Los Angeles, 4650 Sunset Boulevard, Los Angeles, CA 90027 ; and Mary Ann Roberts, Department of Psychology, University of Iowa, Iowa City, 10 52242 .

The ten-category system is a more extensive system for observing classroom behavior, with 9 of the 10 categories referring to i nappropriate behavior. This and similar sys­ tems provide detailed information about a child's action s . T h e complete system with recording instructions can be found in Appendix E , Table E-1 ; it is called the Classroom Observation Code : A Modification of the Stony Brook Code . The system is especially usefu l for recording hyper­ active behavior. Other cod i ng systems are also shown in Appendix E . Table E-2 i s a 28-item behavioral observation system for observing the social competence of preschool children ; it classifies behavior according to such dimensions as inter­ est, apathy, cooperation , and anger. Table E-3 is a 42-item observation scale for rating autism in real-life settings.

The items are grouped i nto five areas : sensorimotor re­ sponses, social relationships , affectual responses , sensory responses, and language . Table E-4 is a 4-category system for coding children's p lay.

Coding Systems for Observing Teachers' Behavior Observational coding systems are also useful for studying the behavior of classroom teachers . The teacher's behavior should be evaluated , because it may affect the referred child's behavior and the classroom c li mate . The two-, three-, and six-category systems in Table 17-7 provide a record of the teacher's interactions with a specific child or the class as a whol e .

50 1

OBSE RVATIONAL CODING SYSTEMS

Table 1 7·6 I ll ustrations of Cod i ng Systems for Observing Children's Behavior

Examples

Coding system I . Two Categories I . On-Task Behavior (appropriate behavior for the situation)

2. Off-Task Behavior ( inappropriate behavior for the s ituation)

Putting hand up when he or she wants to say something, l istening while teacher is talking . working quietly at desk, asking teacher for permission to leave desk a . Passive inappropriate actions ( for example. staring into space. lack of interest , short attention span , poor concentra­ tion, lack of perseverance) b. Active inappropriate actions ( for example, making noise , hitting. fighting, banging, being out of seat without permission , physical destructiveness, steal ing, threatening others, setting fires)

II. Three Categories

I . On-Task Behavior 2 . Passive Off-Task Behavior ( passive behavior that is inappropriate but does not disrupt others) 3 . Disuptive Off-Task Behavior ( inappropriate disruptive behavior)

See examples under I. I . See example under 1 . 2 . a . See examples under 1 . 2 . b .

I I I . Four Categories

See examples under i . I . Tal king out, t eas i n g Being out of seat. hitting others, throwing objects, playing with objects Daydreaming, sleeping, sulk.ing

1 . On-Task Behavior 2 . Verbal Off-Task Behavior 3 . Motor Off-Task Behavior 4 . Passive Off-Task Behavior 1 . Interference 2 . Off-Task 3 . Noncompliance 4 . M inor Motor Movements 5 . Gross Motor Movements 6. Out-of-Chair Behavior 7 . Physical Aggression 8 . Threat or Verbal Aggression 9 . Solicitation of Teacher 1 0 . Absence of Behavior

IV. Ten Categories

Interrupting teacher or student Engaging in other than assigned work Fail ing to fol l ow teacher's instructions Moving buttocks, rocking Leaving seat, standing without permission Remaining out of chair for a period of time Kicking, hitting Making threatening gestures, bullying Raising hand , calling out to teacher Engaging in no inappropriate behavior as defined by the above categories

Note. The I O·category system is from Abikoff and Gittelman ( 1 985) and can be found in Table E- I in Appendix E.

Coding Systems for Observing Students, Teachers, and Class Any of the separate coding systems designed for students and teachers can be combined i nto one syste m . Categories also can be added for the entire class .

Table 17-8 i l lustrates one such combined coding system , which emphasizes appropriate as well as i nappropriate behaviors . The 11 categories of student behavior i nclude 6 on-task behaviors , 4 off-task behaviors , and 1 neutral behav ior. There are fou r teacher codes and two c lass codes .

502

CHAPTER 17

ASSESSMENT OF BEHAVIOR BY OBSERVAT I ONAL METHODS

Table 1 7·7 I llustrations of Coding Systems for Observing Teachers' Behavior

Examples

Coding sysrem I . Tho Categories

I . Verbal Approval Responses (comments that follow an on­ task behavior) 2. Verbal Disapproval Responses (comments that follow an off-task behav ior)

"Bob, your spel l ing has improved considerably." "Class, stop making noise . "

I I . Three Categories

I . Praise (verbalization indicating that the teacher was pleased with the student's behavior) 2. Prompts (verbalization conveying additional information or d i recting the student's attention to the task) 3. Criticism (verbalization indicating that the teacher was displeased with the student's behavior)

"John, your reading was excellent . " "The fi rst step i n solving t h e problem is to divide t h e sales price by the number of items purchased ." "Mary, do not talk during the reading assignment . "

I I I . Six Categories

I . Academic Approval 2 . Academic Disapproval 3 . Social Approval 4. Social Disapproval 5 . M istake ( inappropriate use of one of the above four behaviors) 6 . No Approval or Disapproval

"Your score w a s much improved ." "Your study habits are not satisfactory." "r am pleased w ith your ability to work with Helen . " "Your relationship with your teammates is poor. " Informing child that behavior was unsatisfactory when there was no evidence that it was Absence of behaviors that could be recorded as approval or disapproval

RECORDI N G M ETHOD A N D CODING

• Design or select a coding system that represents (he behaviors of concern. • Select the recording method that best fits the coding system. (The fou r major types of recording methods are

SYST E M COM B I N E D

The three examples in this section i l l ustrate how recording methods are combined with coding systems. Exhibit 17-9 shows how event recording is combined with a three­ category coding system to examine aggressive behavior on the playground . In Exhibit 17-10 i nterval recording is com­ bined with a staff-resident interaction coding system to study efforts at habi litation of mentally retarded indi­ vidua l s . The detailed ward coding system allows for an in­ depth analysis of staff-resident interactions , whereas the playground coding system focuses on aggressive behav­ iors only. The playground coding system , of course, can be expanded to include other behavioral categories. Exhibit 1 7 - 1 1 shows how event and ratings recordings can be used in a planned incident procedure to observe preschool chil­ dren's emotions.

G U I DE LI N ES FOR D E S IG N I N G AN O BS ERVATIONAL ASSESSM E NT

The fol lowing guidelines w i l l assist you in designing a behavioral observation assessmen t :

summarized in Table 1 7-9 . ) • Use categories sparingly. D o not overload the coding system with too many categories . • Use categories that are easily discriminable. Clearly define each category, and be sure that the categories can be readily discriminated from each other. • Select an appropriate interval length. Use an i nterval length that i s l i kely to reveal the duration of the observed behaviors. • Select an appropriate length of time for the observa­ tions. Use a duration period that is s ufficient to reveal the

most salient features of the behavior under observation without taxing your abil i ty to record accurately. Exces­ sively long observations contribute to observer drift and unreliab i lity. •

Select appropriate observation times and places.

Schedule the observation period so that it coincides w ith the times of day when the target behavior is most l i kely to occu r. Observational findings are most generalizable when

503

GUIDELINES FOR DESIGNING AN OBSERVATIONA L ASSESSME NT

Table 1 7-8 Cod ing System for Observing Students and Teachers in the C lassroom Student Code Summaries

Attending (AT) The student must be (a) looking at the teacher when the teacher is talking, (b) looking at materials in the classroom that have to do with the lesson, or (c) be engaged in other looking behavior appropriate to the academic situation . Working (WK) The student is working on academic material without any overt verbal components either in a group or in individual seatwork situations. Volun teering ( VO) By verbal or nonverbal means, the student responds to teacher requests by volunteering information of an academic nature .

that is not appropriate for the academic activity that is going on at the time.

Look Around (LA) The child is looking away from the appropriate academic task at hand . Inappropriate Behavior (IB) This is a second broad category used to code inappropriate behaviors not otherwise defined. Illustrations include situations where the child calls out an answer when a question is directed to another student. or interrupts the teacher or another student who is talking. Teacher Code Summaries

Reading Aloud (RA) The student is reading aloud either individually or as a part of a group recitation .

Approval (AP) The teacher gives a clear verbal , gestural , or physical approval to the student or to the group of which the student is a member.

Appropriate Behavior (AB) This is a broad category used to code appropriate behavior not otherwise specifically defined, including asking or answering questions , raising hand for help, acqu iring or passing out materials.

Disapproval (DI) The teacher gives a clear verbal , gestural , or physical disap­ proval of the child's behavior either individually or as part of a group.

Interaction with Peer about Academic Materials (IP + ) The student i s interacting with a peer o r peers about academic materials and is not v iolating classroom rules. Verbal commu­ nications between peers, e .g . , talking, handing materials , work­ ing together on academic materials , etc . , were coded I P + . Interaction with Peer about Nonacademic Materials (IP - ) The student is i nteracting with a peer about academic materials inappropriate for the period in which the observation occurs (unless this has been approved by the teacher) , or about non­ academic material. The interaction may be verbal or nonverbal . Don 't Know (DK) The child indicates, i n either a verbal or nonverbal manner, that he or she does not know the answer. Inappropriate Locale (IL) The child, without the teacher's approval , is in a classroom area

No Response (NR) The t.eacher does not respond to the student either as a part of the group or individually. Verbal Interaction (VI) Verbalizations directed at the child or his or her group which are not aprrovals or disapprovals. Verbalizations relating to instruc­ tion or management. Class Code Summaries

Appropriate Behavior (ABg) The entire class (all students) is engaged in activities that are considered appropriate to the situation as defined by the teacher's rules and the act ivity at hand . Inappropriate Behavior (IB) At least one student in the class is observed engaged i n behaviors not considered appropriate according to the teacher's rules and the acti vity at hand .

Source: Reprinted, with changes in notation, with permission of the publisher and authors from C. R. Greenwood, H . Hops, H . M . Wal ker, J . J . Guild. J . Stokes, K . R . Young, K . S . Keleman, and M . Willardson, "Standardized Classroom Management Program : Social Validation and Replication Studies in U tah and Oregon ," Journal of Applied Behavior Analysis, 1 979, 12, p. 240. © Society for the Experimental Analysis of Behavior, Inc .

observations are conducted across multiple environments and on multiple occasions . •

Design or select an appropriate recording sheet.

C l early label precoded categories and provide clear spaces for entrie s .

• Design a final assessment strategy that is likely to detect the target behaviors of interest, given their typical rate and duration. To ensure that you meet this goal , conduct extensive general observations prior to formulat­

ing your specific observational strategy.

504 •

CHAPTER 1 7

ASSESSMENT OF BEHAVIOR BY OBSERVATIONAL METHO DS

Exhibit 1 7-9 Naturalistic Observation Recording of Children on a Playground: Recordi ng Aggressiveness and Related Behaviors

Coding Categories

Three classes of problem behaviors were observed : I . Aggression: Striking, slappi ng, tripping, kicking . push­ ing, or pulling others; "karate" moves ending within I foot of another person; doing anything that ends with another child's fal l ing to the ground. 2. Property abuse: Taking another person's property with­ out permission; throwing school books, l u nches, or anyone else's property ; throwing any object at passing or parked cars; diggi ng holes in the ground with one's feet or hands: breaking pencils or pens. 3. Rule vio/alions: Resisting or talking back to an aide; climbing more than I foot off the ground on a playground structure not meant for cl imbing; suspending one's sel f on a playground structure in any position that results in the head's not being 1 80 degrees above the feet . The Playground Observation System

The playground was divided into three roughly equivalent "pie sl ices" based on the existing geography ( for example, building corners or edges of playground equipment), which were the responsibility of separate observers. These slices were then halved (again, defined by other permanent struc­ tures ) , and each half was monitored for alternate I S-second periods. Thus an observer attended to only one-si xth of the playground at a time, and only hal f the p l ayground was observed at any given moment . Three observers stood in the middle of the playground facing thei r areas. A tape recording instructed them to start

R E LIABI LITY OF BEHAVIORAL OBSE RVATIONS: GEN ERAL C O N S I D E RATIONS

Sources of Unreliability The data obtained from behavioral observations, l ike those obtained from any other assessment procedure, must be rel iable and valid. It is important to establish observer agreement in order to ensure that your observations are replicable and consistent ( reliable) , which in turn will aid in establishing accuracy (validity ) . In the observation of behavior, rel iability and validity are influenced by the observer; the setting, scales, and i nstruments; the c h i l d ; t h e sample (in observations o f a group o f children) ; a n d the

watching the left-hand portion of their slice , at which time they began recording incidents with the aid of hand counters. After I S seconds. the tape cued a "switch" to the remaining portion of the observers' area. This continued for 2 minutes, when a "stop" signaled that the cumu l ative frequency of inci­ dents observed was to be entered on the data sheets. The entire process occurred for 1 0 iterations (that is, observe left for I S seconds , observe right for I S seconds, and back left again, recording the totals every 2 minutes) from 8 : 20 to 8 :40 A . M . A particular inappropriate incident (for example, kicking) d irected at one child by another child was only counted once per I S-second interva l . However, more than one incident was scored if one child inflicted several types of aggression on another ( for example , one child's hitting and kicking another resulted in two incidents' being counted) . I f two children assaulted a third individual or one child assaulted two peers, two incidents were scored . The I S-second intervals were arbitrarily considered to be independent; thus, if two children were observed to be wrestling with one another for two intervals, fou r incidents were recorded . Reliability

Rel iability was determined by having i nterobserver checks on various days. Source: Reprinted, with changes i n notation, with permission of the publisher and authors from H. A . Murphy, J . M. Hutchison, and J . S. Bailey, "Behavioral School Psychology Goes Outdoors: The Effects of Organized Games in Playground Aggression," Journal of Applied Behavior Analysis, 1 983 , 16, pp. 30-3 1 . © Society for the Experi­ mental Analysis of Behavior, Inc.

interactions among all of these sources. Sources of error in the fou r basic areas are described i n Table 17-10. A lthough most of Table 17-10 requires no explanation , some observer errors and errors associated with the referred child (or group) deserve further comment.

Observer bias.

Many errors associated w ith the per­ sonal qualities of the observer are subsumed under the broad category of observer bias . The term observer bias is used to refer to any of a number of different errors associ­ ated w ith the personal qualities of the observer. Observer bias e ncompasses anything an observer does that distorts the recording of behav ior - such as form expectations, prefe r certain categories or certain scale positions , exer-

... Exhibit 1 7· 1 0 Naturalistic O bservation i n an Institution for the Mentally Retarded: Recordi ng Staff and Residents' Efforts at Habilitation

other way threatens to harm another resident or a staff member. 6 . Self-stimulatory - resident engages in solitary activity but actively manipulates some object(s), or is engaged i n sol itary. asocial, repetitive behavior ( for example , rocking, headweav ing) .

Coding Categories

Behaviors of both staff and residents were recorded . The following coding categories were used for staff and residem behaviors : STA F F BEHAV IORS

I . No interacrion - no physical or verbal imeraction be­ tween the staff member and any resident. 2. Verbal instruction - through standard language (that is, either vocal or manual communication), staff instructs the res idem to perform some activity and offers no physical assistance. 3. Non verbal instrucrion - through a gesture (not includ­ ing manual communicatio n ) , staff instructs the resident to perform some activity and offers no physical assistance. 4 . Verbal instruction lVith physical assistance - through standard language (that i s . either vocal or manual commu­ nication), staff instructs the resident to perform some activity and provides physical assistance (for example, guides resi­ dent through a self-dressing task with verbal aid ) . 5 . Non verbal instruction with physical assistance ­ through a gesture (not including manual commun ication), staff instructs the resident to perform some activity and pro­ vides physical assistance ( for example , points to the door and guides resident to move toward the door) . 6. Physical assistance - w i t hout prior verbal or nonverbal instruction, staff physically assists resident (for example, staff helps resident put on his or her shoes) . 7 . Social- staff claps or praises, hugs, etc . , resident. 8. Custodial g uidance - staff physically assists res idem in a custodial manner in a non-task situation (for example, ties shoes of resident i n order to allow resident to move along quickly with other residents) .

Recording Procedure

For 16 days, four observers recorded for 250 minutes per day. Each person observed in one of five locations for about 50 minutes . Then the observer walked to another location and recorded for another 50 minutes. This procedure was fol­ lowed from about 9:30 to 1 1 : 20 A . M . and 1:00 to 3 : 50 P . M . each day, until each observer had recorded in the five loca­ tions. Sites were rotated so that no observer was in one site more than once per day, and so that each site was observed by each person about the same amount of time. Data were rec­ orded at 6-second intervals. with the imervals being signaled through earplugs by a portable tape recorder. At the end of each 6-second interva l , the observer marked any response category lhat had occurred within the 6-second interval . There were three recording rules other than that of s imply marking what had just occurred: (a) A fter observation of a staft· member. something had to be marked. If none of the seven respon e categories had occurred, the observer marked the no interaction category. (b) After each observation of a resident, on-task, off-task, or no programming had to be marked (the categories of aggression and self-stimulation were to be marked only i f they had j ust occurred ) . (c) If more than one resident or staff response occurred in the same interva l , both could be marked ( for example, sel f-stimulatory and off-task respond ing) . Reliability

R ES I DENT BEHAV IORS

Interobserver agreement was assessed each day by having a second observer randomly assigned to the various recording sites . This produced about 40 hours of rel iability assessment. Observations were coordinated through a y-plug from the tape recorder that allowed each observer to hear the beginning of each successive i me rval . Because something was marked at the end of each 6-second interval and because the observers were 3 meters apart, the observations were quite independent . Interobserver agreement was calculated by divid i ng the number of imervals in which both observers agreed by the total number of i me rval s .

I . all-task - resident emits a verbal or motoric response to a question, command, instruction, or nonverbal cue (for example, a gesture by the staff), or compl ies without making an overt response when no overt response is necessary or appropriate (for example, looking at pictures in a book) . 2. Off-task - in the presence of a cue for responding, resi­ dem either does not respond, responds inappropriately, or does not look at relevam task stimul i . 3 . No prog ra mming - nothing i s being asked o f the resi­ dent, being demonstrated to the resident, or being provided for the resident to do. 4 . Self- agg re ss i ve - resident i ntentionally strikes. bites , slaps , hits, or k icks own body, or causes his or her body to com act with force other objects. 5 . Other aggressive - resident i ntentiona l l y strikes at, throws objects at. or verbally threatens others, or in some

Source: Reprinted, with changes in notation, with permission o f the publisher and authors from A . C. Repp and L. E . Barton, "Natu­ ralistic Observations of Institutionalized Retarded Persons: A Com­ parison of Licensure Decisions and Behavioral Observations ," Jour­ nal of Applied Behavior Analysis, 1980, 13, pp. 335-337. © Society for the Experimental A nalysis of Behavior. Inc.

505

Table 1 7·9 Observational Recording Methods Recording melhod Narra/ive recording

Behavior is compre­ hensively described.

Types Anecdolal recording

Anything that appears noteworthy is recorded . Running record

Observer makes an on· the-spot description of behaviors.

VI o a-.

Il1Iervai recording

Observational period is divided into brief seg­ ments or intervals; ob­ server notes whether a behavior occurs i n each interva l .

Parliai-inlervai lime sam­ pling

Behavior is scored only once during the inter­ val , regardless of dura­ tion or frequency of occu rrence . Whole-inlervai lime sam­ pling

Behavior is scored only when it lasts from the beginning to the end of the interval . Poinl-lime inlervai sam­ pling

Behavior is scored only when it occurs at a designated time during the interval . MOI'l'lenlary lime inlervai sampling

Behavior is scored only when it occurs at the end of the interval .

Advanlages

Disadvanlages

Appiicclfions

Dala

Is useful as a precursor to more specific and quantifiable observations Helps in the development of hypotheses about factors controll ing tar­ get behaviors Provides an in-depth pic­ ture of behavior

No speci fic quantitative data, although the rec­ ord can be analyzed for various occurrences of behavior

Provides a record of child's behavior and general impressions Maintains original se­ quence of events Facil itates discovering critical behaviors and noting of continuing d ifficulties Requires a minimum of equipment

[s not well su ited to ob­ taining quantifiable data [s costly in terms of time and person power [s d i fficult to val idate [s time consuming May be insensitive to critical behaviors Produces findings with limited generalizing

[s useful for behaviors that are overt or easily observable, that are not clearly discrete, and that occur with reasonable frequency (for example, reading, working , roughhous­ ing, smiling, playing with toys)

Number of intervals in which target behaviors did or did not occur

Defines important time­ behavior relationships Facilitates checking inter­ observer reliability Maintains standard obser­ vation conditions in an economical way Enhances attention to specific behaviors Allows for flexibil ity in recording large num­ bers of behaviors

Provides somewhat a r­ tificial view of behav­ ior sequence May lead observer to overlook important behaviors Usually tells l ittle about qual ity of behaviors or situation Provides numbers that are usually not related to frequency of behaviors [s not sensitive to very low-frequency behav­ iors and , in point-time sampling, behaviors of short duration

Variable interoccasion in­ terval time sampling

Behavior is scored only when it occurs at des­ ignated random time intervals. Event recording

Each instance of a spe­ cific behavior (event) is observed and recorded .

Event

Observer waits for pre­ selected behavior to occur and then records its occurrence . Duration

Observer determines the amount of time that elapses between the beginning and the end of the behavior.

Is useful for behaviors that have clearly de­ fined beginn ings and endings. such as spell­ ing words correctly, rocking movements, asking questions, and speech errors

Number of occurrences of the behavior - fre­ quency count A lso, in some cases, rate of behavior, duration of behavior (time), inten­ sity of behavior ( i f built into code ) , la­ tency of behavior (time)

Fac ilitates detection of low-frequency behaviors Facilitates study of many d i fferent behaviors i n an economical and . flexible manner Provides information about the frequency with which behavior is performed and changes in behavior over time

Provides artificial view of behavior sequence and breaks up continuity of behavior Is not suited to recording nondiscrete behaviors Presents difficulties in es­ tablishing reliabil ity Limits quantification of the hows and whys as­ sociitted wirh hehavior Makes comparison across sessions d i fficult if the length of the observa­ tion period is not constant

Is useful for evaluating more global aspects of behavior and for quan­ tify ing impressions

Scale value (or number or score) on rating scale

A l lows for the recording of many different be­ haviors in an eco­ nomical manner A llows for the rating of many individuals and the group as a whole Permits rating of subtle aspects of behavior Facil itates statistical analysis

Uses scale values which may be based on un­ clear assumptions May have low rel iability Does not allow for re­ cording of important quantitative dimensions Does not allow for re­ cording of antecedent and consequent events

Intensity

Behavior is d ivided into various degrees of intensity, and behavior of each degree is rec­ orded separately.

n ::> ..J

Latellcy

Observer determines the amount of time that elapses between the in­ itiation of the request and the onset of behavior. Ratings recording

Behavior is observed and then rated on vari­ ous scales.

5-point scales 7 -poi nt scales Other d imensional scales

508 I'"

CHAPTER 17

ASSESSME N T OF B EHAVIOR BY OBSERVATIONAL METHODS

Exhibit 1 7- 1 1 Observing Preschool Children's Reaction� to Specially Designed Situations

Zahn-Waxler, McKnew, Cummings , Davenport, and Radke­ Yarrow ( 1 984) designed a setting for observing preschool children's reactions to specially created incidents . The re­ ferred child, a famil iar same-age playmate, parents of the two children, and staff members interact in the setting under various conditions intended to induce conflict, distress, frus­ tration, and enjoyment . Aggressive, altruistic , and other emotions may be revealed. The room in which the observations are conducted, prefer­ ably a living room-kitchenette area, should contain a standard set of toys (for example, rocking horse, ball , pull toy, toy telephone ) . The fol lowing conditions should be established (Zahn-Waxler et al . , 1 984, p. 2 3 7 , with changes in notation ) : I . A novel environment. Initially the children play in the new room , with the mothers watching (5 minutes) . 2 . A background climate of affection and sharing. Two female adults enter the adjoining kitchen. They greet the mothers and children, then cooperate with each other in a warm and friendly fashion while getting coffee for the moth­ ers and juice for the children and straightening up the kitchen (5 m inutes) . 3 . A neutral cOlllext. There are n o experimental interven­ tions (5 minutes) . 4 . A background climate ofhostility, angel; and rejection. The two women return and have a verbal argument while washing the dishes. Each accuses the other of not doing her share of work around the building (5 minutes). 5. A second neutral context (5 minutes) . 6. A reconciliation. The adults return, greet each other with affection, and apologize for their unpleasant behavior (2 m inutes ) . 7 . A friend's separation experience. The mother of the referred child's friend is asked to leave the room (I m i nute ) . 8 . Separation from mother. The referred c h i l d 's own mother is called from the room as wel l (1 minute ) . 9 . Reullion \Vilh lhe mOThe!: Both mothers return t o the room (4 minutes) . Mothers should be asked not to initiate activities o r to inter­ rupt i nteractions between the children unless something makes them uncomfortable or appears to be dangerous. The above conditions can be modified to suit the specific room arrangements.

ci se len iency, exercise variable attent ion , or be influenced by extraneous cues . Some examples may help to illustrate observer bias. A n observer's ex pectation that t h e referred c h i l d will a c t ag-

Suggested event observational recording codes are as fol ­ lows (Zahn-Waxler e t ai . , 1984 ) : I . Aggression: actions that have potential for causing phys­ ical or psychological harm a. Interpersonal physical aggress io n - hitt i n g , kicking, pushing, or throwing things b . Object struggle - attempts to grab or take anothe r's possession c. Undirected aggression - acts agai nst the physical en­ vironment (for example, banging on wall s , throwing things on the floor, kicking toys) d . Intense aggression - acts that are violent or potentially dangerous 2 . Altruism or empathic intervention - acts of kindness and caring directed toward others a. Child helps , cooperates, provides comfort, or sym­ pathizes with other person ( for exampl e , by patting or hug­ ging a crying person, kissing a hurt , saying "It's OK" or "Be careful," providing a bottle, sharing toys) b. Child shares either objects or self ( for example, invites others to join in particular play activities) . A suggested scale for rating various forms of emotional expressiveness is as follow s : RATING SCALE

emotion absent

2

3

emotion expressed sl ightly

emotion expressed somewhat

4 emotion expressed moderately

5 emotion expressed frequently

Raring a . positive emotion (laughter, smiling, happy excitement expressed facially. vocally, or bodily) b. anger (angry yelling, screaming, or facial expressions, impassioned threats or complaints) c. distress (crying, crankiness, whining , concerned facial expressions) d . emotional ity (combined scores for positive emotion, anger, and distress)

2

3

4

5

2

3

4

5

2

3

4

5

2

3

4

5

gressively may influence him or her to record marginally aggressive act s as aggressive, when other observers with­ out this expectation would record the same acts as nonag­ gressive . An extraneous c ue in the form of a teacher's

509

RELIABILITY OF BEHAVIORAL OBSERVATIONS: GENERA_ CONSIDERATIONS

Table 1 7· 1 0 Sources of Error i n Observations of Behavior Tvpe of error

Source of error

Errors associated w ith personal qual ities of the observer

Central tendency - Observer

uses the middle category of a rating scale more frequently than the end categories and in the process tends to underestimate intense behaviors and overesti mate weak behaviors. Leniency or generosity - Obser v er makes judgments about the referred child that are too favorable . Primacy effect - Observer al lows first i mpressions to have a distorting effect on later impressions or judgments. Halo effect - Ob e rver makes judgments based on a general i mpression of the referred child or on his or her most salient characteristic . Personal theory - Observer fits the observations to his or her personal theoretical assumptions. Personal valu es - Observer fits the observations to his or her personal expectations, values. and interests. Overestimation of traits or behaviors that are barely self-acknowledged - Observer overestimates in the referred child t raits and behav iors that he or she barely acknowledges in himsel f or herself. Logical erro r - Observer makes similar judgments on traits that seem to be logically related . Contrast error - On specific traits, observer judges others to be more different from h i mself or herself than they actuall y a re . Proximity error - Observer judges specific traits a s similar because the format of the judgments places them close together i n time or space. Personal effects- Unbeknow n to the observer, his or her personal characteristics (such as age , sex, race, and status) affect the referred child's behavior. Observer drift - Over time, observer changes criteria (or thresholds) for judging the presence or absence of a behavior because of fatigue or learning or other variables. Omission - Observer fails to score a behavior that has occurred. Commission -- Observer m iscodes a behavior. Expectancy effects - Observer's expectations infl uence what he or she records. or observer expects some­ thing to happen and communicates these expectations to the child . Observer reactiviry - Observer changes recording of behavior when he or she is aware of being observed . Nonverbal cues - Observer unintentionally furnishes to referred child nonverbal cues that act as reinforcers of certain behav iors. (Table

praising the referred child for having completed a past assignment may lead an observer to record on-task behav­ ior in the interval , even though the referred child was not doing the current class assignment at the time of observa­ tion . Observer reactivity (changes in the observer's behav­ ior as a result of situat ional influences) may occur not only when observers are being observed by another person . but also when they are told that their records will be compared with those of another observer. This knowledge may make them more careful , vigilant, and attentive to details than they ordinarily would be .

colllillues next page)

Difficulties in coding behavior. Global categories , such as off-task behavior or inappropriate behavior, require a higher level of inference than do specific categories , such as hilling or out-oj-seat behavior. Reliabil ity may be more difficult to achieve when global categories are used . A l ­ though every attempt should be made t o defi ne target behaviors precisely, some behaviors may be difficult to categorize . For example , how does one distinguish be­ tween a child who is staring into space and one who is thinking about a problem? Thus observational codes re­ quire careful j udgments on the part of the observer.

Observer drift .

When observation continues over a long period of time, observers may show signs of forget­ ful ness, fatigue , and decreased motivation . For example, an observer may begin with one standard for scoring wh ispers or brief vocalizations, but over time change that standard . Observer drift may occur even when the observ­ ers have agreed on specific definit ions of behavior.

Timing of behavior.

The timing of event is not as simple as it appears . For example, when exactly does a chi ld's refusal to eat begin and end? The time unit selected by the observer may not reflect an exact mapping of the behavioral event.

510

CHAPTER 17

ASSESSMENT OF BEHAVIOR BY OBSERVATIONAL METH O DS

Table 1 7- 1 0 (cont.)

Type of error

Source of error Errors associated with setting , codes, scales, and instruments

Unrepresentative behavioral selling Observer selects only one setting or only one time period and thereby fails to sample adequately representative behaviors. Coding complexity Observer cannot use codes accurately because there are (a) too many categories i n the system, (b) too many categories scored on a given occasion, and/or (c) too many children observed on a given occasion. Influence of extraneous cues - Observer is influenced by certain acts in the environment to score the occurrence of a behavior when the behavior is not occurring . Rating scales Observer inappropriately uses broad-category rating scales to classify behaviors and thereby loses fi ne distinctions. Mechanical instruments Observer fails to check the accuracy of mechanical devices used for recording data (for example, stopwatch or counter). -

-

-

-

Errors associated with the referred child (or children)

Child reactivity - Referred child's behavior changes as a result of the knowledge that he or she is being observed . Role selection - Referred child adopts a particular role as a result of the knowledge that he or she is being observed . Measurement becomes an agent of change Referred child makes a significant change in his or her behavior or attitudes as a result of being measured and observed. Response set - Referred child responds in a manner that conforms to cues from the observer. Behavior drift Child's behavior continues, but in a form that drifts outside the range of definitions being used . -

-

Errors associated with the sample (usually large samples or groups) Source:

Unrepresentative sample Observer fails to obtain a representative sample of the population. Sample instability - Observer fails to recognize population changes over time, making it difficult to compare present sample's results with those of previous samples. Unrepresentative data - Observer fails to recognize geographical and regional differences between samples. -

Adapted in part from Fassnacht ( 1 982) .

PROC E D U RES FOR ASSESS I N G R E LIABI LITY

Three useful procedures for estimating the rel iabil ity of observational coding are interobserver reliability or agree­ ment, test-retest reliability, and internal consistency relia­ bility. I nterobserver rel iabil ity is the most i mportant form of rel iabil ity for behavioral observation s . W ithout interob­ server rel iabil ity, the other forms of rel iability have l ittle meaning.

I nterobserver Reliability or Interobserver Agreement Esti mates of i nterobserver rel iabil ity (or i nterobserver agreement) are usual ly based On scores of two or more observers who record the same information while simul­ taneously and independently observing the same child or group (Nay, 1979 ) . The data may be in the form of classifi­ cations (or nomi nal scale ratings or categorical judgments)

or i nterval scale ratings . Once these data have been ob­ tained, an appropriate statistical index of agreement must be selected and calculated . Several procedures are avail ­ able for measuring interobserver agreement, including correlational coefficients (such as the product-moment cor­ relation coefficient and the intraclass correlation coeffi­ cient) and percentage agreement indices (such as kappa and uncorrected percentage agreement) . These fou r pro­ cedures measure different aspects of interobserver agree­ ment and thus may result in different rel iabil ity estimates for the same set of data.

Product-moment correlation coefficient. I f you are inter­ ested in the pattern of agreement among the observers' ratings, irrespective of the level of agreement , and you have interval data , then the product-moment correlarion coefficient is satisfactory (see Chapter 2 ) . The product­ moment correlation coefficient is sufficient when you sim­ ply want to establ ish whether One measure is l i nearly related to some other measure . As an i ndex of agreement

51 1

PROCE DURES FOR ASSESSING RELIABILITY

between observers , the product-moment correlation co­ efficient usually is not the method of choice . except with rating scale data.

IntracIass correlation coefficient. W hen both the pat­ tern of agreement and level of agreement are important and you have an interval scale of measurement , the intraclass correlation coefficient may be used (Fleiss, 1975 ) . The i ntraclass correlation coefficient takes i nto account the extent to which all observers mean exactly the same thing by their j udgment s . Kappa.

W h e n t h e data form an ordinal scale and you are interested i n correcting for chance agreement, kappa is a u seful index of agreement (Cohen , 1960, 196 8 ) . Kappa takes into consideration both the occurrence and the non­ occurrence of behavior corrected for chance agreement among observers . It is used in s ituations in which there are no i ndependent criteria or bases for i ndependent expert evaluation . Kappa measures the degree of consensus among observers; it evaluates precision, but does not tell whether the observations are valid . One of the preferred procedures, kappa can be used for mUltiple observers and multiple categorie s . E xhibit 17-12 shows the procedure for comput ing kappa . A m icrocomputer program is available for computing kappa for multiple observers , multiple cate­ gories, and missing data (Oud & Sattler, 1984) .

Interval recording percen tage agreement estimates. In interval recording, a number of d i fferent percentage agreement methods are used for determi n i ng interobserver agreement . Three such methods are (a) overall agreement, (b) agreement on the occurrence of the behavior, and (c) agreement on the nonoccurrence of the behavior. The key difference among the three methods is the spec ific i ntervals used to determine the level of interobserver agreemen t . The data in Figure 17-8 are used to i l lustrate the three interobserver percentage agreement measure s . I . Agreement on total observations. T h i s method of obtain ing percentage agreement takes into account the total number of intervals and the occurrence or nonoccur­ rence of a behav ior i n each i nterval . Agreement is defined as both observers' scoring e ither the occurrence or the nonoccurrence of a behavior in a g i ven interval . The pro­ cedure is as fol l ow s : a . Considering all interva l s , make two counts, one of the number of intervals in w h ich the observers agreed on the occurrence or nonoccurrence of a behavior and one of the number of i ntervals in which they d isagreed. b . Divide the number of agreements by the total number of agreements plus d isagreements and multiply by 100 . The result is the perr.entage of interobserver agreement for the total number of i ntervals . The formula for interobserver percentage agreement for the total number of intervals is as fol lows :

Percentage agreement .

When you desi re a measure that shows the percentage agreement among two or more ob­ servers but are not concerned w ith correcting for chance agreement , an uncorrected percentage agreement index is usefu l . Uncorrected percentage agreement, w h ich is sim­ ply the percentage of agreement of two or more observers, is particularly susceptible to overestimating agreement w he n chance agreement is high. A lthough percentage agreement is not synonymous w ith rel iability, it is useful as a pre l iminary check of the adequacy of your observational recordings because of its ease of computation and inter­ pretation and its sensitivity to bias and systematic errors . I n the material that fol lows , uncorrected percentage agree­ ment w i l l be referred to as percentage agreement.

at lo A' R

where

101

=

A Wl x I 00 A IO! + D

= interval recording percentage agree­ ment for the total number of intervals Alol = n u mber of i nterva l s in which Ob­ server I and Observer 2 agreed on whether the behavior occurred or did not occu r D = n u mber of i nterva l s i n w h i c h Ob­ server I and Observer 2 disagreed on whether the behavior occurred or did not occur

%A 'R

101

Observer I Observer 2 X indicates occurrence of behavior, 0 indicates nonoccurrence of behavior.

Figure 1 7-8.

Raw data for three interobserver percentage agreement measures.

CHAPTER 1 7

512

ASSESSMENT OF BEHAVIOR BY OBSERVATIONAL METHODS

Example: Two observers agreed on whether the target behavior occurred or did not occur in intervals 1 , 2 , 3 , 5 , 6 , 7 , and 10 (seven agreements ) , b u t di sagreed in intervals 4 , 8 , and 9 (three disagreements) . Therefore there was a 70 percent rate of agreement in scoring the target behavior over the total number of intervals recorded :

01

_

lo A IR lot =

A ,o, + D x 1 00 A tot 7 10

x

1 00

=

=

7

7

+

3

x 1 00

70 %

Exhibit 1 7- 1 2 Procedures for Computing Kappa (K) Kappa (K) is a useful statistic for measuring interobserver reliabil ity (or interobserver agreement) for categorical data. Kappa indicates the proportion of agreements, corrected for chance agreements. Like correlational coefficients, kappa ranges from + 1 .00 to - 1 . 00 . When kappa is positive, the proportion of observed agreement is more than the proportion of chance agreement. When kappa is equal to zero, the pro­ portion of observed agreement equals the proportion of chance agreement. When kappa is negative, the proportion of observed agreement is less than the proportion of chance agreement. Let us examine the rationale behind kappa for a hypo­ thetical situation in which two observers scored one child over 1 00 intervals for the occurrence or nonoccurrence of a behav­ ior. Suppose Observer I scored the occurrence of the behavior in 90 intervals and Observer 2 scored the occurrence of the behavior i n 80 intervals. In this situation there must be some agreement because of the overlapping of distribution of scores. Figure l 7-A shows that for the two observers the lowest possible number of overlapping occurrence intervals

D (J) ...J a:: UJ f­ Z

1 00

�

90

To i ntroduce the general kappa formula for two observers and multiple categories, let us set up a 3 x 3 contingency table (Exhibit continues next page)

N o n occurrence intervals

Occurrence i n tervals

I

N o n overlapping i ntervals

( 1 0)

20 N onoverlapping int ervals (20)

o ���----���O bs e rv e r O b s e rv e r

1

Figure l 7-A.

Kappa for Tho Observers and Multiple Categories

Overla p p i n g intervals ( 7 0)

u..

o a:: UJ (l) � => Z

(that i s , intervals scored identicall y by the two observers) is 70. This minimum overlap of 70 intervals occurs when 10 of the occurrence intervals scored by Observer 2 correspond to the 1 0 nonoccurrence intervals scored by Observer I . In this case, the correction for chance agreement i n the kappa for­ mula is 72 percent. The procedure for obtaining the chance correction is discussed below. Kappa can be used for multiple categories and multiple raters. Formulas are presented below for computing kappa for (a) two observers and multiple categories and (b) the special case of two observers and two categories ( 2 x 2 contingency table). Formulas for computing kappa for multiple categories as wel l as for multiple raters are found in Conger ( 1 980) and Uebersax ( 1 982) . Uebersax presents a generalized kappa formula that is also appropriate for handling missing data.

2

Distribution of occurrence and nonoccurrence intervals scored by two observers.

513

PROCEDURES FOR ASSESSING RELIABILITY

Exhibit 1 7- 1 2 (cont.) representing two observers and three recording categories. The designations for the contingency table are as fol lows:

observation categories over 10 intervals . The three codes used by the two obsen'ers were verbal off-task (YO ) , motor off-task ( M O ) , and on-task (OT ) . The data were as follows :

Observer 2

Il1Iervals

Observer I

11 "

11 , 2

1/ 1 )

11 2 1

11 2 2

11 2 3

11 3 1

11 + 1

11 3 2

11 + 2

11 ))

11 + 3

where

I

YO YO MO OT OT YO MO MO MO YO

MO YO MO OT OT YO MO YO MO YO

2 3 4 5

113 � N

+ ,

Po - Pc I - Pc

Observer 2

I

Each cell is designated by two subscripts, one for the row and one for the column (for example , 11 , , - 11 2 2 , 11 ) 3 ) ' The fi rst subscript refers to the row, the second to the column. Thus 11 2 3 designates the cel l in the second row, third column. The rows and columns correspond to the three different observation categories ( e" e2 , el) . The marginal totals for Observer I are designated by 11 1 + , 11 2 and 11 ) + , and those for Observer 2 are designated by 11 + 1 , n + 2 , and 11 + 3 ' T h e general formula for kappa i s K =

Observer 1

6

7 8 9 10

Placing these cores in a 3 x 3 contingency table gives us the fol l owing: Observer 2

I

Po = the observed proportion of agreement Pc = t h e p roportion of agreement expected by

Observer

I

YO

MO

OT

YO

3

I

0

4

MO

I

3

0

4

0

0

2

2

4

4

2

10

OT

chance alone

The computational formulas for Po and Pc are

Pc

=

B(

l1 + i

=

+ -= 11;;) ( 11 '''" + " 2 2 + 11 j ) + -'--' - =...::.=. N

3 + 3 + 2 = .1. = . 80 10 10

)( n + ;)

_ �-,I_ .!. i_ _ _

N2

(111+ X n + l) + (112+ X I1 +2) + (I13 + x n +3) + N2

where

-

To calculate kappa, we fi rst obtain P o and Pc :

Po c

'---- --

nii

=

ni+

=

n +i

=

N

=

. . .

+ (l1i+ x n +i)

total number of agreements for the ith category (main diagonal) marginal total for Observer I on the ith category marginal total for Observer 2 on the ith category total number of observation periods (for ex­ ample , intervals)

Let us apply this formul a to some hypothetical data ob­ tained by two observers who scored the same child using three

- (4 x 4) + (4 x 4 ) + (2 x 2) 1 02 _

=

l2..

100

=

.3

_ -

16 + 1 6 + 4 1 00

6

Then we put the values ofPo and pc into the formula for kappa: K =

. 80 I

-

-

.36 .36

=

. 44

.64

=

.

69

(Exhibit continues next page)

CHAPTER 1 7

514

ASSESSMENT OF BEHAVIOR BY OBSERVATIONAL METHOID S

Exhibit 17-12 (cont.) If a straight percentage agreement had been used , the level would have been 80 percent (or what Po equals) . Kappa gives us a coefficient of . 69, a somewhat lower level of agreement . A kappa of . 70 is considered to indicate an acceptable level of agreement . Kappa for a 2 x 2 Contingency Table

We now consider kappa for binary ratings, with two observers and two observation categories ( for example , occurrencel nonoccurrence or agreement/disagreement in a 2 x 2 con­ t ingency table) . This is a special case of kappa. The general formul a for kappa, as we have seen , is

I

K

=

Po - Pc I - Pc

I

=

the number of intervals i n which Observer I scored the behavior as occurring and Observer 2 scored the behavior as not occurring = the numbe r of interva l s in w h ich Observer I e scored the behavior as not occurring and Observer 2 scored the behavior as occurring d = the number of intervals i n which Observer I and Observer 2 scored the behavior as not occurring

b

Example: The data for two observers who scored one child over 100 intervals are summarized as follow s : Observer 2 Occurrence

In a 2 x 2 contingency tabl e , Po is computed by dividing the two cells i n which both observers agree by the total number of observation periods or i ntervals (N) . A nd Pc is computed by adding the products of the marginal frequencies and then dividing this value by the total number of observation periods or intervals squared. Thus, for the contingency table

Observer I

K =

_

-

Observer 2 NonOccurrence occurrence

Observer I

a

b

a +b

Nonoccurrence

e

d

e+d

a + e

b + d

N

= -;::;_

(a + b) (a + e) + (e + d) (b + d)

N2

= Po the observed proportion of agreement Pc = the p roportion of agreement expected b y chance alone

N = total number of observation periods

A computationally more convenient formula for computing kappa i n a 2 x 2 contingency table is K

where a

=

= the

=

2 (ad - be) (a + bleb + d) + (a + e)(e + d)

number of intervals in which Observer I and Observer 2 scored the behavior as occurring

Row total

Occurrence

20

6

26

Nonoccurrence

2

72

74

Column Iota)

22

78

1 00

2 (ad - be) (a + bleb + d) + (a + e)(e + d)

2[(20 X 72) - (6 x 2)] 2(1440 - 1 2) 2028 + 1 62 8 (26 x 78) + (22 x 74) _

2(1428) 3656

=

2856 3656

=

. 78

For t h e above data , Po ' t h e observed proportion of agree­ ment, is

Po

a+d

Pc -

where

total

Occurrence

Column tola)

Po

Row

Nonoccurrence

=

a +d

=

N

20 + 72 1 00

=

92 '1c = 92 100 0

and Pc ' the proportion of agreement expected by chance alone, is _

Pc -

(a + b)(a + e) + (e + d) (b + d)

= (26

=

x

N2

22) + (74 x 78) 1 002

6344 10,000

=

010 63 (

Again a kappa of . 78 is a more conservative estimate of interobserver agreement than the 92 percent agreement rate , uncorrected for chance.

SIS

PROCEDURE S FOR ASSESSING RELIABILITY

Agreement on occurrence observations. This method of obtaining percentage agreement takes into consi era­ tion only those i ntervals in w h ich at least one of the two observers recorded the occurrence of a behavior. Agree­ ment is defi ned as both observers' scoring the occurrence of a behavior in a g iven interval . This procedure is similar to the one used for total observations , except that only a portion of the i ntervals is used . a. Considering only those i n tervals in which at least one of the two observers recorded the occurrence of a behavior, make two counts - one of the number of intervals in which the observers agreed on the occurrence of a behavior and one of the number of intervals in which the observers disagreed. b. D iv ide the n umber of agreements by the total number of agreements plus disagreements and multiply by 100. The result is the percentage of i nterobserver agreement for those interva l s i n w h i c h the behavior was scored as occurnng . The formu l a for interobserver percentage agreement for behavior occurrence is a variant of the one used for overall percentage agreement : 2.

I %A IR

�

3 . Agreement on n.onoccurrence observations. This method of obtaining; pe:rcentage agreement takes into con­ sideration only tllos,e intervals in which either one or both observers recorded t e nonoccurrence of a behavior. Agreement is de - ned a both observers' scoring the nonoc­ currence of a be:1av ior in a given interva l . This procedure is similar to the one described above . a . Considerin g only those i n tervals i n which at least one of the two observers recorded the nonoccurrence of a behavior, make two counts - one of the number of intervals in which the ob-ervers agreed on the nonoccurrence of a behavior and one of the number of i ntervals i n which observers disagreed . b . Divide the number of agreements b y the total number of agreements plus disagreements and multiply by 100. The result is the percentage of interobserver agreement for those intervals i n which the behavior was scored as not occurring. The formula for interobserver percentage agreement on nonoccurrence of behav ior i s another variant of the one used for overall percentage agreement: at A

----I

I

where

ace

_

-

A

ee

a

A ace

+

D

X l Oa

% A 1 R ace = interval recording percentage agree­ A aee

=

D

=

ment for intervals in which occur­ rence of behavior is scored number of i ntervals in which both observers agreed that the behavior occurred number of intervals in which the ob­ servers disagreed on whether the be­ havior occurred

Example: lWo observers agreed that the target behavior occurred in i ntervals 1 , 3 , 5 , and 10 (four agreements), but only one of the observers scored an occurrence of the behavior i n intervals 4, 8 , and 9 (three disagreements) . Thus there was a 57 percent level of agreement for scoring the target behavior as occurring:

%A I R ace =

A a� A ace D

= 7 X 1 00 4

X 1 00 =

(0

=

! 3 X l Oa 4

57 %

I f neither observer records an i nstance of an occurrence of a target behavior, a rel iability index cannot be calculated for this category.

where

- A na+n X 1 00 IR nan A nan D

% A 1R nOn

==

A non

==

D

=

interval recording percentage agree­ ' ment for intervaJs in which nonoc­ currence of behavior is scored number of intervals in which both ob servers agreed that the behavior did not occur number of intervals in which the ob­ servers disagreed on whether the be­ havior d id not occur

Example: Two observers agreed that the target behavior did not occur in i ntervals 2 , 6 , and 7 (three agreements) . In intervals 4 , 8, and 9, however, only one of the observers scored the nonoccurrence of the behavior (three d isagree­ ments). Thus the level of agreement was 50 percent for scoring the target behavior as not occurring: %A 1 R nan

=

=

An 3_ X 1 00 ,; X 1 00 = _ + A non D 3 3 3

- X 1 00 6

==

50 %

If neither observer records an instance of a nonoccur­ rence of a target behavior, an agreement index cannot be calculated for this category.

516

CHAPTER 1 7

Comment on interval recording percentage agreement estimates. When observers score a very small propor­ tion of the intervals for the occurrence of a behavior, the most appropriate method may be the second one, which takes into account only those intervals in which an occur­ rence of the behavior was scored . In these situations, the use of the first procedure (total i n tervals) might cause some di stortion of the rate of agreement . Example: I n a 100-interval observation period , one ob­ server scored the occurrence of the target behavior in three intervals and the other observer scored the occurrence of the behavior in one overlapping i nterval . Their rate of agreement in scoring the occurrence of the behavior is 3 3 percent . Use o f t h e total intervals would result in a n agree­ ment rate of 98 percent . The 33 percent agreement figure better represents the observers' abil ity to identify the target behavior when it occurs. When more than one category is used i n an observation system , as is usually the case, a decision must be made about whether to evaluate interobserver agreement for the total observations, for the separate categories, or for both . I t is highly recommended that i nterobserver agreement be computed for each category as well as for the total obser­ vations . This approach prov ides val uable information about where potential difficulties may lie in the observation system or between observers. A percentage agreement above 80 percent is considered satisfactory. When occurrence of a behavior is scored in a large proportion of the i ntervals , it may be of i nterest to study the rate of agreement in those intervals in whi h nonoccur­ rence of the behavior was scored . In this case the third method can be used . All three methods may be used to report interobserver agreement estimates . Generally, you should choose the i ndex that best answers your quest ion . I n Table 17-1 1 , the three types of percentage agreement measures and kappa are calculated for 10 sets of data . Also included is a column showing chance agreement . It is evident that in some cases the fou r measures differ greatly. In cases I , 2 , 4 , 8 , and 10, kappa gives a . 00 coefficient , whereas the total percentage agreement measure ranges from 50 to 100 percent . Kappa thus provides a much more conservative i ndex of i nterobserver agreement . The table also shows that the value of kappa is significantly affected by chance agreement . In example 3 kappa is .66 ; a change in a single agreement in one interval causes the value of K to change to . 00 in example 4. Therefore it may someti mes be difficult to interpret kappa.

Event recording percentage agreement estimate. I n event recording, interobserver percentage agreement can be estimated by dividing the number of occurrences of the

ASSESSMENT OF BEHAVIOR BY OBSERVATIONAL METHODS

event reported by the observer recording the lowest fre­ quency by the number of occurrences reported by the other observer. The percentage agreement formula for event recording is as follow s :

% A ER where

% A ER

=

h.

=

It =

=

�

)h

x 1 00

event recording percentage agreeme n t frequency o f observer with t h e lower event frequency frequency of observer with the higher event frequency

Example: Two observers recorded out-of-seat behavior, the target event . During a 20-minute observation period , one observer recorded 5 occurrences of the behavior and the other observer recorded 8 occurrences . Substitute i nto the formula as fol lows : % A ER

=

�

)h

x 1 00

=

i x 1 00

=

62 . 5 %

There was a 62 . 5 percent agreement between the two observers in recording the number of occurrences of the target behavior. This level of agreement does not mean that the observers recorded the same target behavior, however. It could be that there were 13 occurrences of the target behavior, 5 of which were recorded by one observer and 8 of which were recorded by the other. The level of agree­ ment simply i ndicates that the ratio of events reported in common was 62 . 5 percent. UnJess specific i ntervals or times were used i n the observational recording procedure , there i s n o way of knowing whether the two observers recorded the same events.

Duration recording percentage agreement estimates. The interobserver percentage agreement estimate for du­ ration recording i s similar to the one u sed for event record­ ing. The percentage agreement formula for duration re­ cording is as follow s :

where

%A DR

=

d u ration recording percentage agree­ ment t, = t i me recorded by observer w ith the lower t i me duration th = t i me recorded by observer w ith the h igher time duration

517

PROCEDURES FOR ASSESSING RELIABILITY Table 1 7- 1 1 Comparison of Fou r Different Observer Agreement Formu las for

0,

O2

0[±HJ 0

NO I.

°

d

c

i o

0

2. 49

I

10

Cases of Data

Chalice percelllclge fig reelllellt nil

Occurrence

occllrrence

percelllage agrf'elllellf

1 00

1 00

98

98

94

NO/1occllrrenc{' percelltage agreement

Total percentage agreell1ellt

Kappa

1 00

. 00

0

98

. 00

98

50

98

. 66

0

0

98

98

. 00

25

1 00

1 00

1 00

1 . 00

24

96

96

98

. 96

24

67

67

80

.62

18

26

39

50

. 00

25

2

2

4

- . 92

28

36

31

50

. 00

0 0

3 . 48 0 0

4.

0 49 5 . 25 0 0 25 6 . 24 0 25 7 . 20

10

-�-

0 20

8.

9 9

9.

16 16 24

24

I

10. 14

12

13

I I

NOle. See text for computational formulas for each agreement index and kappa. The formula for chance percentage agreement on occu rrence is

CO

=

(a

+

b)(a + c)

N'

In the 2

x

2 table (column one ) . 0,

=

Observer I. 0,

Example: Two observers recorded the target event of a child's staring out the w indow. One observer clocked an episode at 360 seconds ; the other, 3 6 5 seconds . Substitute into the formula as follows :

% A OR

=

0." x

1 00

=

360 3 65

x 1 00

=

=

Observer 2. 0

Ratings recording percentage agreement estimates, In ratings recording, i nterobserver percentage agreement

occurrence, NO

=

nonoccurrence.

can be estimated by determining the extent to which the two observers gave the same rating on each scale. The percentage agreement formula for rati ngs record ing is as fol lows :

98 %

Thus there was a 98 percent agreement between the two observers in recording the duration of the staring-out-the­ w i ndow behavior.

=

% A RR where

% ARR

=

=

A A rr � D

x 1 00

rati ngs recording percentage agreement for the total number of scales = number of scales in which both observ­ rr A ers agreed on the rating D = number of scales in which the observers disagreed on the rating

518

CHAPTER 1 7

Example: A fter a 30-minute observation period , two ob­ servers completed ten 5-point rating scal e s . They agreed on ratings for 8 of the 10 scales . Substitute i nto the formula as fol low s :

% ARR

= =

A rr x 1 00 Ar+ D r 8

8 + 2

x 1 00

= 10 8

ASSESSMENT OF BEHAVIOR BY OBSERVATIONAL METHODS

behaviors used by the observer, or i n the methods used for the observation . Which factor or combination of factors is responsible for the instab i l it y must be determined through a care fu l study of all sources of data and of the procedures used in the observational assessment.

I nternal Consistency Reliability x 1 00

=

80 %

Thus there was an 80 percent rate of agreement between the two observers in making their ratings . I n ratings recording , a 7-point scale requires fi ner j udg­ ments than a 3 -point scal e . Thus the more categories you have in a rating scale , the lower the interobserver percent­ age agreement is l i kely to be . Percentage agreement takes into account only the absolute level of agreement - agree­ ment is counted only when both observers give the exact same rating . This agreement procedure does not consider the pattern of ratings. If one observer is consistently one scale position above (or below) the other observer, a dis­ agreement still will be counted for each scale . Conse­ quently, i n ratings recording it is also useful to compute a product-moment correlation (or intracl ass correlation co­ efficient) to ascertain the pattern of agreement between the two observers .

Test-Retest Reliability The consistency of behavior over time and situations is another measure of the reliability of behavioral observa­ tions. Every attempt should be made to sample the target behaviors on more than one occasion and in more than one setting. The percentage agreement interval recording for­ mulas can be used to evaluate test-retest rel iability. Various correlational procedures can also be used, depending on the scaling of the data. For example, i ntraindividual stability can be assessed by correlating the frequency with which each targeted behav­ ior was observed on one occasion with the frequency with which each behavior was observed by the same observer on another occasion . The product-moment correlation co­ efficient so obtained does not allow you to evaluate which of the behaviors show more or less stabil ity because the correlatio n is computed across all categories . By scanning the changes from the fi rst to the second observation peri­ ods for each category, however, you can obtain some idea of which categories show the most change . When i nstabil ity occurs, it may be due to changes in the child, i n the setti ng , i n the observer, i n the defi nitions of the

Internal consistency reliability tell s us how consistent an assessment instrument is in measuri ng the same charac­ teristics. One way of obtaining internal consistency esti­ mates is to divide the observation measure into two equal parts ( for example , odd- and even-nu mbered items) . Chapter 2 describes various formu las for measuring i n ter­ nal consistency rel iability. Factor analysis, discriminant function analysis, and various correlational procedures (depending on the scali ng of the data) can also be used to evaluate the internal composition of items in observational coding systems .

VALI DITY O F B E H AV I O RAL O B S E RVAT I O N S

The validity of an observational assessment is determined by studyi ng the meani ngfulness and relevance of the be­ havioral measure s . Validity refers to the extent to which the procedures are measuring what they are supposed to measure (see Chapter 2 ) . The validity of observational measures, however, is more often assumed than estab­ lished (Hoge , 1985 ) . If you observed behavior that you wished to classify as hype ractive , for example, you m ight consider the fol lowing questions (Bernstei n & Nietze l , 1980) : In terms of construct validity, do the behaviors coded (for examp l e , fidgeti n g , out-of-chair movements) con­ s t i tu t e a s a t i s fa c t o r y and fu n c t i o n a l d e fi n i t io n of hyperactivity? I n terms of content validity, do the data reflect the nature and deg ree of hype ractivity shown during the observation? In terms of concurrent validity, does the child's behav­ ior under observation accurately reflect his or her reactions i n other situations? Finally, i n terms of predictive validity, do the behav­ iors coded predict other i mportant c riteria? •

•

•

•

Answering these questions w i l l not be a simple matter, but only if you can answer them affi rmatively can you be fairly certain that you are measuring hyperactivity.

519

PROCEDURES FOR REDUCING ERRORS IN OBSERW TION

Representativeness and Generalizability of Findings Two of the major factors affecti ng the validity of the obser­ vational assessment are the representativeness and g ener­ alizability of the fin d i ngs (Evans , 1983). To w h at degree does the behavior observed during the time-sampling pro­ cedure reflect behav ior in the total t i me period and behav­ ior in other situations to which one w ishes to generalize the findings? These questions must be considered in every observational assessment.

Reactivity Child ren's behavior under observation may be affected by the knowledge that they are being observed, prior interac­ tions w ith the observer, or the observer's personal charac­ teristics. This effect is referred to as reactivity or the guinea pig effect. To miru mize reactivity, you should try to conduct your observations as unobtrusively as possible . Ordinarily you should not assume that stable behavior is typical behavior or that the changes you observe would have occurred if you were not present. Children's reactivity to a fami l iar observer presents a particular threat to the validity of the observational assessment; if the observer is someone the child knows, an independent assessment of the referred child should be carried out by someone not fam i l iar to the child. You can judge whether reactive effects have occurred by looking for (a) systematic changes in behavior over l ime , some of which may be related to the presence of the observer, (b) increased variab i l ity in behavior, (c) reports from childre n that they changed their behavior as a result of being observed , and (d) discrepancies between different measures of the same behavior, when measures were ob­ tained under d i fferent observation conditions (Haynes & Horn , 1982). Although these i ndices do not prove con­ c lusively that reactive effects are present, they suggest the possibility that such effects are operating . Harris and Lahey (1982 , p. 536) believe that reactivity is often so powerful that it clouds the observational data: "Unless it has been well-documented that reactivity is not a factor i n a given situation, observational data may be taken as a demonstration that a particular behavior is in a [child's] repertoire , but not that it is performed i n the absence of observation . " Not everyone agrees , however, that reactivity is neces­ sarily detrimental for observational assessment. Cone and Foster (1982 , p. 343 ) , for example, noted that reactivity may pose problems for some research and c l i nical objec­ t ives but not for others .

The important is �ue seems to be not whether observed i nd i ­ viduals react d i ffe:renltly under conditions of known observation but rather whether dalta collected under such conditions are less useful than those colliected surreptitiously . . . I n this vein it is co'nceivable that reactive data may have even greater util ity or social val idity than nonreactive data in some circumstances . This could occur when [you wish) to general ize to situations involving similar levels of obtrusive observation . For example. in assessing the adequacy of vocal presentations before audiences, it i s probably the case that data obtained from conditions in which the c l ient is aware of being observed w i l l correlate more highly with subseqIJent real - l i fe presentations. Similarly, it may be the case that relatively reactive heterosocial skil l s assessment will evidence greater utility than less reactive assessment, given that most heterosocial interactions contain an evaluative component . Removing a significant portion of this component from the assessment process coul d lead to lower correlations with relevant criteria. As Barker and Wright ( 1955) pointed out long ago, interaction between an observer and a person observed is important in its own right, not j ust as a potential confounding element to be uniformly e l i minated or contro lled .

PROCEDU RES FOR REDUC I N G E RR O RS I N OBSERVA nON

Reducing Errors in Reliability A number of procedures can be used to reduce errors i n reliability. O ne method i s t o study the e rrors l i sted i n Table 17-10 and then try to avoid them (or l imit the i r occurrence) when you perform behavioral observations. You can re­ duce or elim inate many of these errors simply by practic­ ing. To eliminate others you may need to further refi ne your recording procedures and rating scales. Some errors , such as those associated w ith your personal characteristics (for example, height, weight, skin color, voice) or w ith the reactions of the referred child, may be d ifficult to e l i m i ­ nate, but you can minimize their effect simply by being aware of their potential influence . Reliabil ity can be increased by hav i ng c lear and precise definitions of behaviors, systematic and precise rules gov­ erning the observations, wel l-trained observers, and ob­ servation periods that are not excessively long . Observer drift , although difficult to control , can be reduced by making frequent checks of the recordings, thoroughly train i ng observers beforehand, and having the observers make periodic calibrations during the course of the obser­ vation session to check the consistency of their observa­ tions . These procedures also serve to reduce many other sources of observer bias.

520

CHAPTER 1 7

Errors may also be reduced by gaining an understanding of your decision criteria, such as the degree of certainty you must have in order to report that a behavior did or did not occur, and comparing your decision criteria with those of other observers. Signal detection approaches (deriva­ tive measures of which are often referred to as d') can be useful in achieving these goals ( Lord , 1985) . These ap­ proaches focus on an observer's ability to detect stimu l i , taking into account the observer's response style . I n some cases, using broad categories to classify behavior (such as on-task or off-task) may be unwise, as these categories do not permit fi ne distinctions between behaviors . Neverthe­ less , broad categories can be used when they meet the assessment goals and are methodologically sound .

Comparing Observation Results with a Criterion During your training, you should regularly compare your results with those of a highly trained observer or with standard criterion recordings. The agreement between an observer's recordings and standard criterion recordings is known as criterion-related agreement. Even trained ob­ servers should occasionally compare their results with those of another observer or standard criterion recordings to check reliability. Another method of checking the rel iabil ity of your rat­ ings involves videotaping the behavior of a child (or class) . After recording your observations in the setting in which the behaviors occur, you then use the videotape to make a second independent recording of the behaviors. The level of agreement between the two recordings is a measure of the rel iabil ity of your ratings . This method is generally used only for training purposes, however; because the second observation may be affected by your memory of the fi rst, rel iability may be overestimated. If possibl e , also compare your observation of the video­ tape with that of an expert and those of several col leagues. Thoroughly discuss any disagreements and compute esti­ mates of interobserver agreement. Low interobserver per­ centage agreement may mean that the categories are not c learly defi ned , that one (or more) of the observers does not understand them, or that other factors are operating to interfere with agreement. You may be able to increase interobserver rel iability by practicing observation assess­ ment in environments similar to those in which you will work - observing in the natural settings and/or via video­ tapes.

All acrep!able level of interobserver agreement does 110! rule OU! observer error, however. You can have a high level of interobserver agreement and still have observer bias and observer drift if both observers have a similar bias or make

ASSESSMENT OF BEHAVIOR BY OBSERVATIONAL METHODS

similar errors. The level of agreement is even more l i ke l y to mask errors w h e n y o u compare t w o of your o w n recordings.

Reducing Reactivity Although reactivity may be useful in some cases, often you will want to minimize it. To this end, Nay ( 1979) offers the fol lowing suggestions :

Become as neutral a stimulus as possible. Avoid mak­ i ng eye contact with or interacting with the referred child or any other children during the observation period . To provide a rationale for your presence i n a classroom, the teacher might say to the class : "Ms . A is here today to see what we do. " At the beginning of each school year, make a few short v isits to each class so that the children w i l l become used to seei ng y o u v isit . •

Position yourself so that you are away from the ordi­ nary paths ofmovement in the classroom and yet still have an unobstructed view of the child and setting. A good •

position in the classroom is often to the rear or side of the room . • Shift your attention from one child to another. By so doing, you w i l l avoid calling attention to any one indi­ vidual child. • Limit your stimulus value. Do not dress or act in a manner that attracts attention . • Follow all rules, regulations, and informal policies of the school, institution , or family. Before going into a

classroom, institution , or home , review your specific pro­ cedures with the teacher, administrative personnel , or parents.

Enter the setting when your entrance will be least disruptive of the ongoing behavior. For c lassroom obser­ •

vations, enter the classroom before class begins or at break time . If possible, try to be in the setting for a period of time before the formal observation begins. If the children or teacher or family have an opportunity to become used to your presence , their behavior may be more natural (that is, less reactive) when the formal observation begins . These suggestions will help you to become a more natural part of the setting and dimin ish the c h i ldren's awareness of your presence . Awareness by itself does not necessarily affect children's behavior (Kazdin , 1979) . It is only when the children's awareness of you leads to changes in their behavior that the validity of the observations is diminished . The reason for the assessment may determine how much i nfl uence your presence w i l l have on the chil­ dren's behavior. Childre n who are told that the results of the assessment may be used to determine their class p l acement

PROCEDURES FOR REDUCING ERRORS I

may be more affected by your presence than those w 0 are told that the results w i l l have l ittle or no bearing on their status.

Establishing Informal Norms Developing informal norms w i l l help you to place the child's behavior in a meaningful context. In a group set­ ting , you might observe the behavior of the referred child and that of one or more peers as wel l ( Alessi . 1980) . The behavior of the peer children can then serve as a norm for comparing the behavior of the referred child. The peers should be children of the same age and sex as the referred child who have not been identified as experiencing behav­ ior problems and who are as representati ve of the total peer group as possible . Another procedure is to establish local norms for the entire c l ass, using the scan check method . The scan check method i nvolves scanning the entire class for, say, 2 seconds every 2 minutes ( for a period of 8 to 10 minutes) and recording how many students are off-task. Still another procedure for establishing informal norms is to compare the child's present behavior with his or her past behavior, thereby using the child as his or her own norm . Obta i n i ng a peer or c lass rating permits measurement of the relative difference between the frequency with which the referred child engages in a particular behavior and the frequency with which the peer or class does so. A discrep­ ancy ratio can be used to summarize the results of peer or class comparison s . The discrepancy ratio is the relative difference between the median level of the refencJ child's behavior and the median level of the peer's (or class's) behavior. The referred child who is off-task six ti mes per minute while his or her peers are off-task three times per minute could be described as "off-task twice as often as his or her peers . " Here is a n example o f how informal norms can be ob­ tained ( Alessi , 1980) . A ssume that a teacher referred Rob­ ert for his talking-out and out-of-seat behaviors in class . You observe Robert along w ith Todd , who has not been identified as talking out excessively and is considered an "average" child in this regard . Every few minutes you also scan the class and note how many of the children are talk­ ing out or are out of their seats. Your results are as fol lows:

Robert Todd Class ( % ) (6 scans)

52 1

OBSERVATION

Talking-our behavior

Out-oj-seat behm'ior

20 2 3 % ( 1 130)

10 1 6 % ( 2 /30)

These data suggest that Robert engages in more inap­ propriate behavior (as defined by the teacher) than does either the comparison child or the c lass as a whole . If we accept these data as reflecting approximate norms for behavior in this class , we can then conclude that Robert's behavior deviates from the norm . In this example both a comparison child and local class norms provide standards for interpreting the behavior of the referred child. Without these standard " , it would be d i fficult to know the extent to which the behavior of the referred child was deviant .

Developing Sensitivity to Teachers' Needs and Behaviors Fol lowing are a number of problems assoc iated with class­ room observations and suggestions for avoiding them ( A l ­ essi, 1980) .

Problem 1. The observation occurs during a part of the day in which the child does not exhibit the problem behav­ ior or on a day when the child is unusual l y good . Suggestion : Confer w ith the teacher before you schedule your observation s . [Ask the teacher the fol lowing questions: I "When d o these behav­ iors occur most often? At what times? Which day or days of the week? During which subjects? When is it best to observe the child to ensure that the behavior can be seen? When is the behavior at its worst? When does it usually occur?" A fter this information has been gathered. arrange your schedule to observe the child at a prime time. Whenever possible, plan to spread your observation over three or fou r 10-mi nute sessions rather than one 30-to-40 mi nute session . Also spread these shorter sessions across a week or two . " ( pp . 41-42 , with changes in notation )

Problem 2, You cannot obtain a val id sample of behavior because the teacher has trouble stay ing on a schedule. (This is a variant of problem 1 . ) Suggestion: Frequent scheduling problems may indicate that the teacher has a management problem . which may be contributing to the chi ld's problem. Try to help the teacher develop better management skills. If management prob­ lems are not resolved , it probably w i l l be difficult to provide a successful intervention for the referred problem . Problem 3. The referred child fai l s to behave inap­ propriately when you are present in the classroom. (This also is a variant of problem I . ) Suggestion : This is a positive sign i n that i t ind icates that the referred chi ld's inappropriate behavior is under some voluntary control . I n form the teacher of this possibi l ity and work with her or h i m to establish procedures that will

522

CHAPTER 1 7 ASSESSMENT OF BEHAVIOR BY OBSERVATIONAL M ETHODS

help the child keep the behavior continuously under con­ trol , even when the child is not being observed.

Problem 4. The comparison child also has a behavioral problem or, conversely, never engages in the behavior under observation. Suggestion: This potential difficu l ty can be controlled by consulting with the teacher about selecting an "average" child or selecting a different comparison child for each 10minute observation session . Problem 5.

The norm group is inappropriate .

Suggestion : I nvestigate other c lasses to see whether the

norms i n these classrooms are similar to those i n the referred child's c lass. If you fi nd that other c l asses have more lenient and flexible standards and i f you believe that the referred child would benefit from these standards, consider arranging a transfer for the child.

Problem 6.

The critical behaviors that should be ob­ served are poorly defined. Suggestion : Further discussion with the teacher about the child's problems may eliminate this potential source of difficulty.

Problem 7.

used for the data collection before beginning the obser­ vations . • Train yoursel f to be a critical observer of behavior. • Draw samples of behavior from a variety of situations, times, and - particularly when you are observing groups of c h i ldren or developing norms - c h ildre n . • D iscover what biases, faults, and weaknesses you have that pertain to the observation of behavior. Develop self-understanding and critical self-evaluation ski l l s . • Develop a healthy skepticism toward previous reports about the referred child's behavior in order to be as objec­ tive as possibl e . • Suppress assumptions and speculations about the meani ng and implications of the child's behavior while you are recording data . • When you have fi n ished your observations , consider what factors p recipitate and maintain the child's behaviors and how other i ndividuals in the child's setting respond to the behaviors . Periodically compare your observations with those of an i ndependent observer w ho is using the same scorin g syste m . • Regul arly recalibrate your recordings by checking them against standard protocols. • Keep abreast of research and theory i n the field of observation of behavior. •

l'he child has been referred for reasons other than those gi ven - the teacher has a "hidden agenda" behind the referral . Suggestion : This may be the most difficult problem to solve . Its solution depends on the teacher's w i l l i ngness to discuss openly with you her or his feelings and attitudes about the referred child. Teachers may not be aware of their own hidden agendas. Good consultation and i nter­ viewing skills (see Chapter 16) are especially i mportant i n this situation . Determining t h e teacher's standards c a n also be valuable. Some teachers want behaviors to be close to perfection, whereas others allow behavior that borders on being deviant .

You cannot avoid having beliefs and expectat ion s ; what you must avoid is prefiguring everything you will find. Your observations will be affected by the child's behavior, the reasons for the referral , your decision c riteria (your w i l lingness to score a behavior as occurring or not occur­ ring ) , your familiarity with the behavioral observation coding system, the amount of time you spend obse rving, your experience with exceptional groups of children, your fam i l iarity with the referred c h i l d , and other factors. These same factors will affect your co-observer's record­ i ngs . An understanding of these factors w i l l help you to obtain more rel iable and val id recordings.

General Guidelines for Obtaining Reliable and Valid Observations

CAU T I O N S O N THE USE O F

The following general guidelines will also assist you i n obtai n ing reliable and valid behavioral observations : Understand thoroughly the recording techniques, rat­ ing scales, checklists, and mechanical instruments used for the observations. Be sure that the critical behaviors are highly specific and c learly defined . Check the accuracy of all mechanical instruments •

•

O B S E RVA T I O N S

Observational methods, l i ke any other assessment pro­ cedure , have their strengths and weaknesses ( see Table 17-1 2 ) . Although time-sampling procedures are extremely usefu l , there are some disadvantages associated with ob­ servi ng exclusively a single behavior or a few behaviors . Fi rst , you may be subtly encouraged to identify behaviors according to how easily they can be observed and re-

523

REPORTING BEHAVIORAL OBSERVATIONS

corded . Second, the observation of a l i m ited number of behav iors may complicate detection of other behav iors ­ either positive or negative - that may reveal important in­ formation about the child . Thus when you focus on a single behavior, it is important to remain cognizant of the chil d's other behaviors as well .

Ethical Considerations in Observing Behavior The observation of certain behaviors, such as intimate sexual acti v ities, may not be appropriate ; in such cases self-reports may be more appropriate. Additionall y, so­ c i al l y undesirable behaviors, such as child abuse or exces­ sive punishment, may be difficult to record when the parent or aide i s aware of being observed.

Direct observation of behavior is always contraindi­ cated when your presence could lead to undesirable side effects (Cone & Foster, 1982 ) . Going i nto a work setting to observe an adolescent who has been in a mental hospital could lead the adolescent to be identified, labeled, and soc ially ostracized , for example . Therefore , always con­ sider any possible unintended hannful consequences asso­ c i ated with your behavioral observations, and be sure to take appropriate steps to avoid such outcomes.

R E PORT I N G B E H AVI ORAL O BSERVAT I O N S

Following is a l ist o f items general ly included in a report of observational findings. Not all of the information w i l l be applicable to every situation , of course . Chapter 23 gives further information on including behavioral observations in a report .

I . Personal data - ch i ld's age , sex, physical hand­ icaps , and other relevant characteristics. 2 . Setting data - date , time, place, length of observa­ tion , setting ( including type of room , people, and signifi­ cant others), recording method , and coding system (or behav iors observed) . 3 . Reliabi lity - i nc l u d i n g re l ia b i l ity methods , i f appropriate . 4. Validity . 5 . Intensity- how much the problem behavior inter­ fered with the child's other activities. 6 . Se verity - the extent to which the problem behavior reflects psychopathology. 7 . Duration - the length of the problem behavior episode .

behavioral o bservation or behavioral shaping ?

��--

-

---

CHAPTER 1 7

5 24

ASSESSMENT OF BEHAVIOR BY OBSE RVATIONAL METHODS

Table 1 7- 1 2 Arguments For and Against the Observatio n of Behavior A rgumel1ls for t h e ohservation of behavior

A rgumellls against the observation of behavior

I.

2. 3. 4.

5.

6. 7.

8. 9. 10. II. 12.

The observer may see an unrepresentative sample of the child's behavioral repertoire . The observer may influence the child's behavior in ways that cannot be controlled . Observation rooms with one-way screens are an unnatural environment . S ituational constraints may prevent observations from taking place in situations that could contribute to an u nderstanding of the chi l d . T h e norms o f society impose constraints o n observations, placing great emphasis on the privacy of the child's and family's personal l ife . Children may feel uneasy when they are observed. In classroom observations, teachers may feel intruded upon and insecure, even though it is the child who is being observed. There are no generally accepted system s of observation. Certain observation systems are tied to narrowly focused theories. I nterobserver reliability is a problem. Observation of behavior is a costly method of data collection in terms of personnel , time, and materials. I n ferential statistics are d i fficult to apply, particularly because of the need to sample behaviors, times, and situations.

I.

2.

3. 4.

5.

6. 7.

8. 9. 10.

I I.

12.

Observation of behavior can y ield a representative picture of the natural behavior of a child . Knowledge of a child's behavioral repertoire allows for the formulation of hypotheses about important dimensions of temperament, personality, and interpersonal relations. Exact descriptions of behavior patterns aid in developing treatment and remediation programs . Observation of behavior provides information about the development and adj ustment of a child to his or her physical and social surroundings . Observed behavior may differ from that reported by teachers and parents and by childre n themselves. Observation of behavior provides i n formation that cannot be gained in other ways from uncooperative childre n . Observation of behavior c a n yield a more fi n e l y differenti­ ated picture of the child's reactions than can broader measures such as test scores or n umber of right and wrong responses. Observation of behavior may y ield valid data. Observation of behavior can be tailored to the specific concerns of the referral source. Observation of behavior is important in evaluating and monitoring the progress of treatment and remediation programs. I mportant psychological concepts, such as attitude, role , and motivation, make most sense when anchored a t a behavioral leve l . Behavior patterns are a basis for making decisions about a child - such as whether a child should be admitted into a treatment program, special home , hospital , or reform school - and for formulating rewards and punishments, and these patterns can best be evaluated through observa­ tional methods .

Source: Adapted from Fassnacht ( 1 982 ) .

8 . Frequency - how often the problem behavior oc­ curred . 9 . Generaliry - the number of situations i n which the problem behavior occurred . 1 0 . Norms - how often the comparison c h i ld and the referred child's peer group (or class) e ngaged in the same behavior. I I . Antecedents of the problem behavior.

1 2 . Consequences of the problem behavior- for exam­ ple, how much the problem behavior disrupted the ac­ tivities of the other childre n . 1 3 . Peer group acceptance - whether the problem be­ havior was accepted by other children.

1 4 . Adult acceptance - whether the problem behavior was accepted by adults. 1 5 . Additional problem behaviors - any other problems exhibited by the child. 1 6 . Positive behaviors - behaviors that may be useful in designing i ntervention s . 1 7 . Observation difficulties - difficulties encountered in conducting the observation ( for example, problems in de­ termining onset or termination of response , counting number of response s , defining target responses) . 1 8 . Implications of the findings - for example, whether the behavior was age appropriate or setting appropriate.

525

REPORTING BEHAVIORAL OBSERVATIONS

Exhibit 1 7- 1 3 Psychological Evaluation: Behavioral Observation A ndy Lopez June I . 1 969 Chronological age: 1 6- 1 1 Observer: Todd 10hnson

May I , 1 986 May 6. 1 986 Jill Cole

Name:

Date of observation:

Date of birth :

Date of report. Co-observer:

Reason for Referral

A ndy was observed at Path Services in order to eValuate his progress in a special training program for mentall y retarded adolescents. The observation was conducted to ful fi l l a re­ quirement for a graduate psychological assessment course at Blank State University. Description of Client Observed and Setting

Andy was observed for a I -hour period in a class at Path Services , a vocational training and placement center for men­ tally retarded individuals. The observation was conducted on May I , 1 986 from 8 : 30 A . M . 10 9 : 30 A . M . Andy i s a black-haired , ol ive-complexioned mal e . H e i s 1 6 years, I I months o f age , 5 feet 1 0 inches tal l , and weighs 200 pounds. He wore a blue work shirt, blue work pants, and black calf- h igh work boot s . He looked clean and neatly groomed . Andy was observed during a class lectu re on win­ dow washing and a window washing activity. Observational Methods

Narrative, running record . event, interva l , and rating obser­ vation techniques were used . For the purpose of assessing rel iabil ity, two observers simultaneously but independently observed Andy's behav ior. During the l O-minute narrative recording, the entire class was observed. For the 5-minute running record, A ndy's behavior was observed and recorded . During the 1 0-minute event recording, the observers re­ corded the number of times Andy engaged in the fol lowing behaviors: touching a boot with his hand , tapping his foot on the floor, touching his face or head with his hand , yawning, and raising his hand or motioning with it for the purpose of getting another person's attention. For the interval recording , Andy was observed for three time periods. The first, which l asted for 6 minutes and 40 seconds, consisted of sixteen 1 5-second observation intervals interspersed with 1 0-second recording intervals . During this time period, the observers recorded whether or not Andy engaged i n any of the following nine behaviors: attending, working, vol unteering, reading aloud, display ing other ap­ propriate behavior, interacting with a peer about academic materials, interacting with a peer about nonacademic mate­ rials, looking around, and displaying other inappropriate behavior. The second and third interval recordings each lasted for 4 minute s , and consisted of sixteen 1 0-second observation in­ tervals interspersed with 5-second recordi n intervals. The

g

target behaviors for the second interval recording involved the fol lowing be haviors on the part of the teacher with respect to Andy or the entire c lass : approval , disapproval , no response, and verbal interaction. The target behaviors for the third interval recording were appropriate behavior and inappropri­ ate behavior of the class as a whole. Andy was observed for approximately 45 m inutes while the observers made narrative, rulming record, event, and interval recordings . , An additional 10 minutes of observation time was used for narrative recording of the teacher's behavior in relation to Andy. After all of the observations were fi nished , the observers completed eight 7-point rating scales: on-task (always- never) , verbal off-task (al ways-neve r ) , motor off­ task (always-never) , passive off-task (always-never), ver­ balization (clear-unclear) , teacher verbal approval responses (frequent-absent) , teacher verbal disapproval responses ( fre­ quent - absent ) . and class-appropriate behavior (al ways­ never) . A variety of interobserver agreement indices were calcu­ lated from the event , interva l , and rating observation data. These included overall percentage of agreement, percentages of agreement on occurrence and nonoccurrence of the behav­ iors, and kappa. Six reliability indices of the occurrence of particular behaviors could not be calculated because both observers agreed that the behavior never occurred . Fifty-three of the 63 rel iability indices calculated were satisfactory (at least 90 percent agreement) . The 10 unsatisfactory indices were primarily in areas in which the behavior assessed was not discrete or the observers encountered difficulty with d i rect observation because their view was restricted . Overa l l , i nter­ rater reliability appears to be satisfactory. The results also appear to be valid. Observational Findings

Andy engaged in predominantly appropriate class behavior. He did, however, occasionally exhibit inappropriate verbal and passive off-task behaviors. During the first 1 5 m inutes of class, Andy appeared tired, distracted, and restless. He sat in a slouched position, with his buttocks on the edge of the chair, legs outstretched, and upper back against the top of the chairback. He frequently closed his eyes or rubbed them i n what l ooked l i ke an attempt to wake up. He often coughed and yawned . When Andy heard noises or voices outside the class­ room, or when someone entered or left , he always turned around 10 look and he often waved . A ndy occasional ly bent (Exhibit continues next page)

526

CHAPTER 1 7

ASSESSMENT OF BEHAVIOR BY OBSERVATIONAL METHO IDS

Exhibit 1 7- 1 3 (cont.) over to untie and retie his boot laces. He frequently rubbed his head or the skin around his mouth and nose ; during event record i ng , he did this at a rate of once every 40 seconds. A ndy demonstrated that his ful l attention was not on the teacher by asking "Do we have to go outside and clean up?" approximately 2 minutes after the teacher spoke to the class about that same issue. Andy failed to sign in when he entered the room, and the teacher had to tel l him to do so approxi­ mately 10 minutes into the class period . Upon approaching the sign-in sheet, Andy groped in his pockets for a pen . When he realized he did not have one, he turned toward his class­ mates and asked i n a fairly loud voice if anybody had a pen that he could borrow. On a few occasions, Andy spoke to another class member during the lecture . Despite Andy's inappropriate behavior, he was not disruptive in the class­ room . As the lecture progressed, A ndy's boredom and rest­ lessness decreased, and his attentiveness increased . Despite his seeming distracted ness, he answered many of the teacher's questions appropriately. First, he eagerly and quickly sat up straight, raised his hand, and then waited for the teacher to call on him. However, on rare occasions he shouted out an answer. After a majority of the class members unsuccessfully guessed answers to a question, Andy answered with convic­ tion, "I know what it is - water is m i nerals . " In response to the teacher's question about what supplies the window-washer needs, Andy quickly said , "squeegee , bucke t , spray bottle, and rags - folded . " When the teacher subsequently asked what the bucket was used for and why the rags were folded, Andy promptly supplied the correct reasons. When he was not called upon, Andy tended to look disappointed . A ndy exhibited the same eagerness during the window­ washing task. He pulled his chair close to the demonstration window, attended to the teacher, and enthusiastically volun­ teered to perform the task. Andy apparently understood the teacher's instructions, as he washed the window accordingly. He worked diligently and responded wel l to the teacher's questions. When Andy was done, for example, the teacher said, "Now what are you going to do?" Andy quickly and correctly replied, "Look for streaks . " When A ndy found spots on the window, he asked the teacher, "Want me to get paper towels?" When the teacher indicated that that was a good idea, Andy promptly retrieved paper towels from the front of the

PSYC HO LOG ICAL EVALUATI O N

Exhibit 17-13 presents a psychological evaluation o f a men­ tally retarded adolescent who was observed at a vocational training center. The evaluation is a case study involving systematic behavioral observation . The purpose of this case study is to demonstrate the various observational

classroom and w iped the spots off the window. Andy meticu­ lously inspected the window and enthusiastically asked , "How does it look, Tim?" The teacher responded, "Good , I think," and Andy smiled . Andy's verbalizations were always c lear and grammatically correct. Although he usual ly sat off to the side, rather than among his classmates, he appeared to be friendly with the m . T h e teacher interacted with the students throughout the observation period . I nteraction was predom i nantly verbal - it included lecturing, asking questions, explaining how to wash the window, and commenting on the students' demonstrations. Most of the interaction was instructional ; thus approval and disapproval were infrequent . When approval was given, it was more frequently nonverbal (e. g . , a smile or a nod of the head) than verbal . Andy appeared content when he was given either verbal or nonverbal approval . When the teacher used d isap­ proval , it was i n a fi rm but gentle manner. Andy did not receive disapproval for any of his actions. Overall , the teacher was patient and had good rapport with the students. Andy and his classmates for the most part were cooperative, respectful, and appropriately behaved. There was , however, a mild air of distractibility and restlessness in the classroo m . Summary and Impressions

This observation was conducted on the morning of M ay I , 1 986 at Path Services i n order to fu l fi l l a course requi rement. Andy is a 1 6-year, I I -month-old male. At the beginning of the class on window washing procedures, A ndy appeared to be restless, distracted , and t i red. When the teacher began to ask questions, however, Andy became more i nvolved. His atten­ tion increased and he eagerly, correctly, and appropriately answered many of the teacher's questions . Andy's verbaliza­ tions were a l ways clear and coherent, and he was eager and enthusiastic as he washed windows. He fol lowed instructions and worked d i l igently and meticulously. Andy's distractibility seemed to be due to an over-alertness to the things happening i n his environment. He appeared to be outgoing and amiable. Andy was attracted to both mental and physical stimulation . He seemed to enjoy the recognition he received through successful participation in the class activity. (Signature) Todd Johnson, B . A . , Observer

approaches discussed in this chapter. In practice, observa­ tional procedures would typically be chosen j udiciously on the basis of the reason for referral , previous psychometric assessment fi ndings , and the frequency, duration , and type of behavior under observation . In most cases only one or two behavioral observational techniques would be used. I n the case study fi ve observational techniques are presented

SUMMARY

to demonstrate the applicat ion of each technique and the kinds of results each render s . The report p resents a de­ scription of the adolescent, a detai led description of the observational methods used ( including target behaviors and rating scales) , and a statement about rel iabil i ty and validity. A l l of the observational fi ndings are then inte­ grated into the report . A short summary, which h ighl ights the observational fi ndings , concludes the eval uation.

S U M M A RY I . The observation of behavior is an important part of the assessment process . It provides information about the child's behavior in everyday settings and about the child's interpersonal behavior and learning style. It serves a variety of evaluation and intervention functions. 2 . The assumption underlying the systematic sampl ing of behavior is that the behaviors recorded over time are represen­ tative of the behaviors under observation. 3. Observational methods are u seful for studying h igh­ frequency behaviors and global behaviors and for evaluating a child's progress over time. Observational assessment may not be the method of choice for certain kinds of behaviors, particularly those that occur rarely or only u nder highly specific conditions. 4. The observation of behavior i s particularly valuable for an ecological assessment . Ecological assessment focuses on observ­ able aspects of the setting, how the setting works, and how the setting provides for the needs of the persons in it . 5 . An observational assessment combines an observational recordi ng method and an observational coding syste m . 6. Behaviors d i ffer in t h e i r attributes . The attributes of the target behaviors should be considered when a recording method is selected. 7. The fou r principal observation recording methods used in clinical and psychoeducational assessment are narrative, inter­ val , event, and ratings recording. 8 . I n desig n i ng any recording proced u re , consider the number of times you will observe the child, the length of the observation period. when the observations w i l l be conducted, the type of recording to be used , the target behaviors under observa­ tion ( i f any ) , and the method of recording data . 9. Narrative recording focuses on a description of the child's natural behavior. Two forms of narrative recording are anecdotal recordings and running records . 1 0 . Observations can include molar (broad) descriptions of behavior and molecular ( fine) descriptions of behavior. Observa­ tions also form a continuum from low inferential j udgments to high inferential j udgments . I I . Narrative recording is particularly valuable as a precursor to more specific observations and as a method for observing groups. 1 2 . In conducting a narrative recording, focus on the child's behavior rather than on making interpretations. Describe the

527 setting. Be objective, accurate, and thorough in your recording. Recognize your role i n the assessment process and how your i mpressions may influence your observations. The final narrative recording should integrate all sources of information obtained from the observation. 1 3 . Narrative recording maintains the original sequence of events and allows for the discovery of critical behaviors and the noting of continuing difficulties. However, it is time consuming , poses quantification problems, may be insensitive to some types of critical behaviors, and may not provide generalizable fi ndings. 1 4 . Interval recording focuses on selected aspects of behavior as they occur within specified time i ntervals. Interval recording procedures include partial-interval time sampling (behavior is scored when it occurs any t i me during any interva l ) , whole­ interval time sampling (behavior is scored only when it lasts from the beginning to the end of the interva l ) , point-time interval sampling (behavior is scored only when it occurs at a specific time during the interval ) , momentary time interval sampling (behavior is scored only when it occurs at the moment the interval ends) , and variable interoccasion interval t i me sampling (behav­ ior is scored only when it occurs at preselected random time intervals) . 1 5 . Interval recording is useful for exami ning easily observ­ able behaviors that are clearly d iscrete and occur with mode�ate frequency. A recordi ng interval should be included whenever scoring will interfere with the observations . Sequential recording procedures permit observations of a number of different i ndi­ viduals i n a group. 1 6. The number of intervals i n which the target behavior occurs is the primary piece of quantitative data obtained in interval record ing .

1 7 . Interval reco rd ing provides precise behavioral data within and across observation periods, facilitates checking for rel iabil ­ ity, uses standard observation conditions, uses time efficiently, and allows for flexible recordings . However, interval recording provides a somewhat artificial view of the behavior sequence, may overlook important behaviors, provides l im ited information about the reasons for or qual ity of the behaviors , may not show the actual frequency or duration of the behavior, and may fai l to detect l ow-rate behaviors . 1 8 . Event recording involves the counting o f the number of target behaviors observed during an observation period. It is especially appropriate for measuring d iscrete responses that have clearly defined beginn ings and ends . I t is less suitable for high­ rate behaviors or for behaviors that vary i n duration. 1 9 . The frequency count is the primary piece of quantitative data obtained in event recording. Other behavioral dimensions that can be recorded are the rate , duration, intensity, and latency of behavior. 20. Event recording detects low-rate behaviors, allows for the study of many different behaviors , uses time efficiently, allows for flexible recordings , and provides information about changes in behavior over time and amount of behavior performed . However, event recording provides a somewhat artificial view of the behav­ ior sequence, breaks up the continuity of behavior, provides

5 28

CHAPTER 1 7

l imited i n formation about behaviors that are not clearly discrete . presents difficulties in cSlabl ishing rel iabil ity, requires a h i g h level of observer attention. provides l imited i nformation aboui the how and why of behavior, and ll1akes comparisons across sessions d ifficult if the length of the observation period i s not constant. 2 1 . Ratings recording invol ves using a scale or checklist to rate behavior. It is useful for evaluating the more global aspects of behavior and for quantifying impressions. 2 2 . The scale values on the rating scales constitute the pri­ mary quantitative data in ratings recording. 23. Ratings recordi ng can be used for recording many differ­ ent behaviors of both individuals and groups as a whol e . It allows for the observation of subtle aspects of behavior, generates data in a form suitable for statistical analysis, and uses time efficiently. Some disadvantages of ratings recording are that it uses scale values which may be based on unclear assumptions, presents difficu lties in achieving satisfactory reliability, and l imits the extent to which i mportant quantitative i n formation and antece­ dent and consequent events are recorded . 24. Observational coding systems cover an entire spectrum of behaviors and include global categories as well as extremely narrow ones . A coding system may be selected from preexisting ones or developed to meet the assessment goals. 25. Guidelines for designing an observational assessment in­ clude the following: select or design a coding system that meets the assessment goals, use categories sparingly, use easily dis­ criminable categories, select an appropriate i nterval length and an appropriate length of time for the observations, select appro­ priate observation times and places, use an appropriate recording sheet, and design the overall assessment so that the target behav­ iors are l i kely to be observed . 26. It is important to obtain reliable and valid obscrvational findings . Sou rces of error in the observation of behavior include those associated with the personal qualities of the observer and those associated with the setting, codes, scales, and i nstruments, referred child, and sample. 2 7 . Observer bias i s a general term for a variety of errors associated with the personal qualities of the observer. 2 8 . Observer drift refers to the tendency of observers to become less vigilant during the course of the observation . 2 9 . Interobserver reliability can be measured by a variety of procedures that compare one observer's results with those of another observer who simultaneously and independently con­ ducts observations of the same subject . 30. Kappa is favored by many statisticians as an index of agreement that controls for chance agreement . Kappa and per­ centage agreement may differ for a particular set of data. 3 1 . Percentage agreement is the most popular type of i nterob­ server agreement index. Although it does not take into account chance agreement, it is usefu l , particularly as a prel iminary check on the level of interobserver agreemen t . 3 2 . I n i nterval recording, it i s useful t o obtain percentage agreement for the total intervals and for those intervals in which

ASSESSMENT OF BEHAVIOR BY OBSERVATIONAL METHODS

at least one observer recorded the occurrence or nonoccu rrence of the behavior. 3 3 . In event recording, i nterobserver agreement can be ob­ tained by percentage agreement. 34 . I n rati ngs recording, percentage agreement can be used to determi ne the extent to which two observers gave the exact same rating on each scale. 35 . Other useful forms of reliabil ity for the observation of behav ior are test- retest rel i ab i l ity and i nternal consistency rel iability. 36. The validity of observational findings is more often as­ sumed than establ ished . Every effort should be made to deter­ m i ne the validity of observational findings. 37. Reactivity - defined as changes i n the chil d's behavior as a result of being observed - can threaten the validity of observa­ tional fi nd ings . I n some situations, however, reactivity may not be detrimental to the findings. 38. Errors i n observation can be reduced by having clear and precise definitions of behavior, systematic rules governing the observations, wel l -trained observers, continuous checks on the observers' recordings, and periodic calibrations made by the observer during the course of the observation. 39. As an observer, you can reduce your reacti ve effects by being as neutral a stimulus in the environment as possible, by positioning yourself in a remote location, by l i m it i ng your stim­ ulus value , by fol l owing all rules and policies of the setting , and by conduct i ng the observation at the least disruptive times. 40. Additional ways of reducing errors are to establish i n for­ mal norm s , and to develop sensitivity to the teacher's needs when the observations are conducted in a classroom . 4 1 . To ensure that your observations are as reliable and valid as possible , you should understand thoroughly the observational procedures and i nstruments , check the accuracy of all mechan­ ical instruments, train yourse l f to become a critical observer of behavior, discover your own biases , develop a healthy skepticism toward previous reports about the referred child's behavior, sup­ press assumptions about the meaning of the child's behavior while recording your observations, consider what factors precipitate and maintain the child's behaviors, periodically compare your findings w ith those of another observer, regularly recalibrate your recordings by checking them against standard protocols, and keep abreast of research on the observation of behavior. 4 2 . Be aware of ethical factors that may arise when you observe behavior. 4 3 . In your report of the observational findings, i nclude all important i n formation, such as personal and environmental data; rel iability and validity of the observations; i ntensity, severity, duration, frequency, and generality of the behavior; norms ; ante­ cedents and consequences of the behavior; peer group and adult responses to the behavior; additional problem behaviors; positive behaviors ; observation difficulties; and i mplications of findings for the assessment and for i ntervention planning.

5 29

STUDY QUESTIONS

KEY TERMS, CONCE PTS, A N D NAMES

Systematic observation of behavior ( p . 473) Ecological assessment ( p . 474) Planned incident procedure ( p . 474) Setting appearance and contents ( p . 475) Setting operation ( p . 475) Setting opportunities ( p . 475) Observational recording method ( p . 475) Observational coding system ( p . 475) Narrative recording (p. 476) Anecdotal recording (p. 476) Running record ( p . 476) Molar description ( p . 476) Molecular description ( p . 476) I n ferential j udgment ( p . 476) Behavioral descriptive statement ( p . 476) Target behavior ( p . 478) Interval recording (p. 480) Time sampling ( p . 480) Interval sampli ng ( p . 480) Interval time sampli ng ( p . 480) Partial-interval time sampling ( p . 481) Whole-interval t i me sampling (p. 481) Point-time interval samp l i ng (p. 481) Momentary time interval samp l i ng (p. 481) Variable interoccasion interval time sampli ng (p. 481) Intensity dimension ( p . 488) Event recording (p. 490) Event sampling ( p . 490) Frequency count ( p . 491) Rate of behavior ( p . 491) Duration of behavior (p. 492) Intensity of behavior ( p . 493) Latency of behavior ( p . 493) Ratings recording ( p . 494) On-task behavior ( p . 499) Off-task behavior ( p . 499) Reliability ( p . 504) Observer bias ( p . 504) Central tendency error ( p . 509) Primacy effect error ( p . 509) Halo effect error ( p . 509) Personal theory error ( p . 509) Personal values error ( p . 509) Logical error ( p . 509) Contrast error ( p . 509) Proximity error ( p . 509) Personal effects ( p . 509) Observer drift ( p . 509) Omission ( p . 509) Commission ( p . 509) Observer reactivity ( p . 509) Unrepresentative behavioral setting ( p . 5 10) Coding complexity ( p . 510) Child reactivity ( p . 5 10)

Role selection ( p. 5 10) Response set ( p . 5 10) Unrepresentative sample (p. 5 10) Sample instabiliry ( p . 5 10) Unrepresentative data ( p . 510) Interobserver rel iabi l ity ( p . 5 10) Interobserver agreement ( p . 5 10) Product-moment correlation coefficient ( p . 5 10) Intraclass correlation coefficient ( p . 5 1 l) Kappa (K) ( p . 51l) Uncorrected percentage agreement (p. 5 1 1) Percentage agreement (p. 511) Agreement of total observations (p. 5 11) Agreement of occurrence observations (p. 515) Agreement of nonoccurrence observations (p. 515) Test-retest rel iability ( p . 518) Internal consistency reliability ( p . 5 1 8 ) Guinea p i g effect ( p . 519) Criterion related agreement (p. 520) Informal nonTIS ( p . 521) D iscrepancy ratio ( p . 521)

STUDY Q U ESTIONS I . What are some of the purposes served by the d i rect obser­ vation of behavior? 2. What is an ecological assessment? Include in your discus­ sion Hiltonsmith and Kel ler's ecological assessment framewor k . 3 . What arc some limitations associated w i t h ecological a sessment? 4. Discuss Ihe narrative recording method. Include in your discus ' ion types of narrative recordings, molar and molecular descriptions of behavior, inferential judgments, major uses, de­ sign considerations, advantages, and disadvantages. 5. Discuss the interval recording method . Include in your discussion the five different types of interval recording methods, major uses, design considerations, quantitative data, advantages , and disadvantages . 6 . Discuss the event recording method. Include i n your dis­ cussion a description of the method , major uses, design consid­ erations, quantitative data, advantages , and disadvantages. 7 . Discuss ratings recording. Include in your discussion a description of the method , major uses, quantitative data, advan­ tages. and disadvantages. 8. Indicate which type of recording method is preferred for observing each of the fol l owing and explain why: (a) use of slang words, (b) tics, (c) quality of a story, and (d) the antecedent event preceding an aggressive act. 9 . Compare and contrast narrative recording, interval re­ cording, event recording, and ratings recording. 1 0 . What factors should you consider in designing an obser­ vational coding system? I I . Discuss factors that may affect the rel iabil ity of behavioral observations.

530

CHAPTER 1 7

1 2 . Discuss the fol l ow i ng four procedures for measuring the level of interobserver agreement: product-moment correlation coefficient, intrac1ass correlation coefficient, kappa, and percent­ age agreement . 1 3 . What factors may affect the test-retest reliability of behav­ ioral observations? 1 4 . Discuss some of the factors affecting the validity of behav­ ioral observations. 1 5 . Discuss reactivity in observational recordings . Explain

ASSESSMENT OF BEHAVIOR BY OBSERVATIONAL METHODS

why reactivity may not necessarily be detrimental and discuss ways to reduce reactivity. 1 6 . How would you go about reducing the number of errors that occur in your behavioral observations? 1 7 . Present seven general guidelines for obtaining reliable and valid observations. 1 8 . Discuss the strengths and l imitations of observational methods. 1 9 . Present 10 guidelines for reporting behavioral obser­ vations.

-

18

TH E ASSESSM E NT PROC ESS

It is probably unwise to spend much time in attempts to

General Assessment Considerations

separate off sharply certain qualities of man , such as his

Steps in the Assessment Process

illlelligence, from such emotional and vocational qualities as his interest ill mental activity, carefulness,

Assessment in Schools

determination to respond effectively, persislence in his

Assessment of Speech and Language

efforts 10 do so; or from his amount of knowledge; or from his moral or eSlhetic {Qstes. - E dward

L.

Assessment of V isual-Motor A b i l ity

Thorndike

Learning Potential Assessment Classification and Labe l i ng

Assessment lechniques are used by a professional

The DSM-III-R and C l i n ical and School Psychology

deleclive . and Ihe information provided by Ihat deteclive musl be good enough to s{Qnd up in the court

of life.

Comment on the Assessment Process

Thai is why special techniques are needed, why

Summary

reliability must be assessed, and why the "ullraviolel Iighls" provided by your lesls and procedures are necessary

10

see Ihrough the clients' shields. BUI YOLl ,

again like the delective, are constrained by elhical

considerations , and YOLi a re nor allowed 10 use Ihe

psychological equivalent of Ihe rubber hose, which in part is why you muSI respeci the rights of Ihe client; but of course , for the psychologist I'l!Spect plays a dual role; you will be more effective as well as more ethical if vou give, and thus receive, respect. -William A. H illix

771e lesting by one individual of another human's intellectual, personality, and related characteristics is an in vasion ofprivacy

10

an eXlent no less intimate thall

Ihal in volved ill an examinGlion carried out on Ihat same individual's person or resources by a physician , attorney, or agel1l of Ihe Imernal Revenue Service. -Joseph D. M atarazzo

53 1

532 G E N E R A L ASSESSMENT CONSI D E RATIONS

To be an effective assessor, one must know a great deal about tests and about people; be capable of using creative ski l l , scientific rigor, and caution in deve l oping hypoth­ eses; be flexible enough to modi fy or rej ect hypotheses in the l ight of new data ; know the situations about which inferences must be made; and be aware of one's own characteristics as an interpreter of test performance and human behavior. The multimethod assessment approach advocated in this text synthesizes information obtai ned from past records and evaluations; i nterviews with parents, teachers , and the child; observations of the child; and the results of stan­ dardized tests , i n formal tests , and other special pro­ cedure s . I nterviews and behavioral problem checklists are most useful for obtaining personal information ; an initial description and classification of behavior; impressions of the child, the child's parents, and the child's teachers ; and the child's perception of the problem. Observational pro­ cedures are most useful for obtaining information about the child's behavior in environmental settings, peer-group rel ationships, family relationships , and relationships with teachers and for monitoring progress throughout treat­ ment . Standardized tests lire most useful for comparing the child to some norm group and for evaluating changes associated with the handicap, disease process, or remedia­ tion effort s . I n formal tests are most useful for obtaining information about special abilities or skills that cannot be measured with standardized test s . A l l of these assessment procedures can be repeated at the termination of the inter­ vention program and at fol low-up to evaluate progress and change .

Purposes of Assessment The purpose of assessment may be screening, diagnosis, counsel i ng and rehabilitation , or progress evaluation. In a screening assessment, a relatively brief examination is given to identify children who are el igible for certain programs or who have a disorder i n need of remediation or rehabil itation or to determine whether a more comprehen­ sive assessment is needed. In a diagnosTic assessment, a detailed eval uation of the child's strengths and weaknesses in a variety of areas is undertake n . Counseling and re­ habiliTaTion assessmenr is similar to diagnostic assessment except that emphasis is placed on the child's abilities to adj ust to and successfully ful fi l l daily responsibilities; pos­ sible responses to treatment and recovery potential are also considered. In a progress evaluaTion assessmenT, the focus

CHAPTER 18

THE ASSESSMENT PROCESS

is on charting the day-to-day or week-to-week progress o f t h e child. Typical questions addressed in a diagnostic assessment include the fol lowi n g : 1 . What is t h e level o f t h e child's cognitive functioning? 2. I f there is a handicap, how has it affected the child's language , motor skjl l s , self-concept , interpersonal rel a ­ tions , a n d related areas? 3 . How is the child performing academically? 4 . What is the most appropriate educational program and setting for the child? 5. What realistic goals can be set for the child? 6. How can the school and family deal more effectively with the c h ild's problems or handicaps?

Psychological Assessment and Psychological Testing Psychological assessment is not synonymous with psycho­ logical testing. Psychological assessment takes into con­ sideration the u nique characteristics and the relevant context of the child who is being assessed . Psychological TeSTing is a more l imited professional activity that involves the administration and interpretation of one or more psy­ chological tests . TeSTing produces findings , whereas as­

sessment gives meaning TO thefindings within the context of the child's life situation and clinical history. Psychological Assessment and Medical Evaluation A psychological evaluation assesses the c h i ld's perfor­ mance and capabil ities and guides the establishment of appropriate i ntervention programs . Even in cases of brain injury or physical disabil ity, psychological assessment serves valuable function s . A medical evaluation provides information about the etiology of the chil d's exceptionality, areas of deficit or dimin ished ski l l , the need for special environmental adaptations or prosthetic aids, and possible medical treatments . A psychological evaluation comple­ ments a medical evaluation by providing information that will help the child to function optimally in everyday l i fe situations .

STEPS I N T H E ASS E SS M E NT PROC ESS

The usual steps in the assessment process are as fol lows (see also Figure 18-1) : I . Review referral information . Check with the refer­ ral source - for example , physician, teacher, parents, or court - to clarify any vague i nformation .

533

STEPS IN THE ASSESSMENT PROCESS

2. Obtai n information relevant to the child's medlical , soc ial , psychological , l inguistic , educational , and phys­ ical deve lopment, including previous psychological eval­ uations. This i n formation can be obtained from parents , teachers, and others famil iar with the child's pro' lem , from medical report s , and from other agencies . 3 . Assess the behavior of relevant adults. 4 . Observe the child in various settings . 5 . Administer an appropriate test batte ry. selected on the basis of the referral question; the child's age , physical capabilities, l anguage proficiency, and prior test re�ults: and teacher, parental , and medical report s . 6 . Interpret t h e data. 7. Formulate hypotheses . 8 . Develop intervention strategies . 9 . W rite a report, i nc l uding recommendations. 1 0 . Meet with parents, child ( i f appropriate), and other concerned individuals to discuss the assessment results and recommendations. 1 1 . Follow up on recommendations and retest .

Working with Relevant Adults A comprehensive assessment of a child's behavior requires the assessment of relevant adults' behavior as wel l . S ignifi­ cant adults usua l l y play an important role in the develop-

3. 4. 5. 6.

2.

3.

�

4. 5. 6.

Figure 1 8· 1 .

During testing. observe the child's abil ity to participate actively. take initiative . and assume responsibility for re­ sponses. On those occasions when you ask a child for further information , observe how he or she compl ies w ith the request - for example, whether the original response is rephrased or elaborated. These observations may provide insight into the child's flexibility and w i l l i ngness to com­ municate clearly with others . Observations should also be made of how the child approaches the test materials, how the tasks are attempted, and what factors lead to success and fa ilure . Both verbal and nonverbal responses should be observed . ( C hapters 5 a nd 17 provide additional obser­ vational guidelines . )

Output Stage

1 . Formal and informal tests

1 . Referral received and

2.

Observing the Child's Behavior

I ntervening Stage

Input Stage

evaluated Case history reviewed , including medical evaluation Prior test resu lts reviewed Referral source contacted for clarificatio n , if needed Evaluation procedures selected Parental permission obtained to perform evaluation

men! of the child's behavior. Ideally one should observe how the behavior' of parents and teachers affects the child ( see Chapter 1 7 ) . Th e interview w ith the parents or teacher should elicit inform.ation about how they view the prob­ lem. what they ihave done to alleviate the problem, and their role in maintaining the problem (see Chapter 16) . When the test fi ndings do not agree with the parents' or teacher's account of what the c h i l d can do, the reasons for the disagreemen should be investigated .

administered and scored Interv iews conducted with parents, child , teachers, and community practitioners, and others with significant i nformation Observations conducted in school (and in home, if possible) A l l sources o f data evaluated. compared, and interpreted Eligibil ity criteria for special education programs evaluated Recommendations, interventions ( i ncluding IEPI ndividual Educational Plan) . and placement decisions formulated

Stages of the assessment process.

1 . Psychological report 2 . Parent conference 3 . Child conference 4 . Teacher conference 5 . Staff conference 6 . Informal discussion o f results

�

and recommendations with referral source 7 . Continued monitoring of interventions 8 . Changes i n interventions a s a result of monitoring

CHAPTER 1 8

534

Interpreting the Data

to do given this i n formation (for example, recommended interventions) . Psychologists working w ith children must know what is age-appropriate behavior, in order to deter­ mine to what extent a child's behavior deviates from the norm .

A psychological assessment is a complex acti vity, as it requires gathering, i ntegrating, organizing, and interpret­ ing the following kinds of data: (a) developmental (includ­ ing information about intelligence , achievement skills, perceptual abilities, soci al skills, adaptive behavior, and emotional and personality characteristics); (b) behavioral ( i ncluding observations of behavior during the test, at school , and, where feasible, at home); (c) medical (includ­ ing neurological work-up if needed) ; and (d) environmen­ tal ( i ncluding i n formation about the school and family) . The information is used to describe the problem, estimate its severity, identify factors related to it, suggest areas for i mprovement, develop treatment or i ntervention pro­ grams, and determine a prognosis. The process of interpreting the test results can be con­ ceptualized as a chain with three major links: (a) the test response s , (b) theories through which the responses are integrated and conceptualized, and (c) knowledge of what

3

7

Cullural 8ac k g l ounrl Learning

1 . G E N E RA L

4

I N H E RITED

A B I L ITY

�

•

BACK·

t

G RO U N D

2 . SPECIAL

5

I N H E R IT E D

ABILITIES

8

Formal

*

Interest I n Tesl

Ihe

1

f+

-

PERSON· AliTY

Guessing

Response

Req Uired

t

SITUATION -...

12

1 8.

•

Speed 0 1

16

•

'"""r.

TEST

DEMANDS

�

-

RANDOM VARIATION

'"""- T E S T SCORE

t

PhYSical

C o n d i t i o n 01

E x perience

E x a m inee al Time 01 Tesl

With

t

Morale 01

ReqUired

IndiVidual

S i m i l a r Tests

6

I I

*

SpeCIfic

Abilities

01 Ihe Tesl

Problems

T r a i n ing

15

Perceived

I m portance

Motivation

a nd I n f o r m " ,

Factors that influence test scores. The factors that in­ fluence the output (the test score) include the input (innate factors and factors in the child's background and environ­ ment) and intervening variables (the factors in the child's personality, situational test demands, and random varia­ tion) (see Figure 18-2) . The relative importance of the various factors w i l l vary depending on the examinee . An accurate assessment of intelligence and other abilities re­ quires consideration of the factors i l lustrated in Figure 1 8-2 plus other factors unique to the examinee . The i n fluence of i nnate factors on intelligence test scores has been a matter of concern since the i nception of the testi ng movement. Although we still have no defi nitive

10

Achievemenl

THE ASSESSMENT PROCESS

9

AnXiety

General

13

t

Interier-

ence from

Heallh

1 7.

Misleading

lIems

19.

Clellcal

E r ror s

EnVironment

I a n d SpeCial H a n d icapsi

14

t

I n f l u ence

ol lhe Tesler

I N N AT E FACTORS

BACKG R O U N D AND

PERSONALITY

SITUATION

T E S T D E M A N DS

E N VIRON M E N T I N P UT

I N T E RV E N I N G

RANDOM VARIATION

TEST SCORE

OUTPUT

Figure 1 8-2. A paradigm for the analysis of influencing variables. Reprinted by permission of the publisher from D. A. Gosl i n , The Search /or Abiliry: Standardized Testing in Social Perspective, p . 1 30 . Copyright 1 96 3 , Russell Sage Foundation .

535

STEPS IN THE ASSESSMENT PROCESS

perament domains , whereas interviews, observations, and checkl ists will be useful i n assessing experiential and en­ vironmental factors .

information about the exact contribution of heredity, there is strong evidence that it does make a signi ficant contribu­ tion ( see Chapter 4). The specific extent to which genetic l i m itations can be overcome by env ironmental conditions is not know n .

Problems of generalization.

A child's performance in a standard test situation may not reflect how the child would perform if he or she were more comfortable, stimulated or inspired, healthier, or less upset by family anxietie s . Tests give information about how the child performs at a specific moment under specific conditions. The overall IQ may not reveal the underlying dynamic trends that are develop i ng in the child. Some children may appear to be slow learners at the time they are assessed, yet they may be storing observations and reflections that may not be revealed until a much later period of their l i fe . They may also have i nsights that are revealed in ways not adequately measured by standardized test questions . Look for signs of insight or creativity masked or not directly assessed by the objective score s . Consider the distinction between achievement, i n the

Accounting for failures. Test failures may occur for various reasons . Take the case of a c h i ld who fails to stack blocks when asked to do so. Possible causes of this failure i nc l ude limited hearing, v ision, comprehension, or motor capacities ; i nabil ity to understand the instructions ; nega­ t i v i st i c behavior; or neurological i mpai rment. Careful analysis of the child's entire performance w i l l be needed to arrive at a l i kely explanation of the child's failure . Do not simply report the fai lure w ithout providing an explanation . Both i n di v idual factors ( from n e u ropsychological , physical , experiential , and temperament domains) and en­ v i ronmental factors (from school , home , and peer group) may i nteract to produce fai l u res (see Figure 18-3 ) . Formal testin g will help to evaluate neuropsychological and tcm-

I

Neuropsychological

I

Poor m emory

I

L i m ited attention span

II

I

....-

Physical

I

/I

----

Experiential

I

Physical

M i n i m a l exposure

limitations

to material

that im pede the

I

Temperament

J

I L i m ited ability to grasp concepts needed for solutions

I

J

Personality and temperament traits that interfere with

I

School

I

Home

I

Inadequate learning environment at home

I

-] I

---

Inadequate

Peer group

I

Negative attitudes on the part of peer group toward schooling and other cognitive

--

activities

I

temporary

I n adequate

stress

teaching

home study

materials

environment

on the part of peer

I

I

group toward child

Clash between

Clash between

child' s and

child' s and

teache ! s

parents'

temperaments

temperament

I

Psychopathology

I

Other neuro' psychological deficits

Figure 1 8·3.

J

I

learning

I

Transient or

the test

Environmental factors

--

at school

I

with language of

r'

environment

leaming

M i n im a l familiarity

L i m ited reasoning ability

I

Failures

I

acquisition of material

I

I

Individual factors

I

Major factors that may account for failures on tests or in school.

Inadequate

Negative attitudes

536

restricted sense of level of performance on a given test, and the cumulative achievement that occurs over a much longer span of time (Atkinson, 1974) . The child's cumulative achievement reflects the outcome of complex interactions i nvol v i n g a b i l ity, motivat i o n a l d i spos itions (such as motives, knowledge , bel iefs , and conceptions) , task re­ quirements , i ncentives , opportunities in the i mmediate environment, and other factors that influence the child to engage in various activities. During an i ndividual assess­ ment, these same variables are l i kely to be operating, although on a more l i m ited basis. The child's i ntell igence test score must be seen as reflecting not only a level of ability, but also motivation . Unfortunately, there is no simple way to separate out the influence of each compo­ nent on the final test score . To what extent is the child's level of motivation affecti ng his or her performance? If motivational differences among children or even i n one child over the course of an assessment session are not taken into account , too much emphasis may be placed on the ability component of the test results . Because motivation is not constant among childre n , a pure abil ity i nterpretation of the final test score is i nadequate .

Problems in the measurement of psychological pro­ cesses. The psychological processes measured by spe­ c i fic tests are not easy to identify. Every test measures a number of different function s . For example, the WISC-R Digit Span subtest, which requires i mmediate repetition of orally presented sequences of digits , may measure se­ quencing, short-term memory, and attent i on as we l l as motivation , auditory acuity, understanding of task de­ mands, willi ngness to try, and strategy usage . Which of these are i mportant factors in the child's performance is best determined through the use of relevant assessment devices (Torgesen , 1979). Speci fy i ng the nature of relationships between skills measured on diagnostic tests and c lassroom performance is difficult (Torgesen, 1979) . For example , even though reading-disabled children may perform poorly on tests of short-term sequential memory, it is h ighly unlikely that they have not learned how to read because they cannot repeat sequences of digits . A more plausible assumption is that the Digit Span subtest taps processes that are impor­ tant to both digit repetition and learning to read . This assumption , of course , must be corroborated. Unti l such evidence is available, it will be difficult to know how test failures are relevant to classroom performanc e . Processing difficulties may b e related t o organic, experi­ ential , learning, and other factors . A failure does not tell u s which factor is involved and hence which i nterventions may be needed . A careful study of the child's entire perfor-

CHAPTER 1 8

THE ASSESSMENT PROCESS

mance and clinical history must be made to account for any processing difficulties .

Other problems. Other problems that may compound the difficulty of performing a psychological evaluation incl ude the fol lowing: I . C lassification systems are not u n i form , and symp­ toms often fal l into more than one d iagnostic category. 2 . A behavior that is of concern to adults may not be of concern to the child . 3 . Because of rap id developmental changes , problems a child has at one age may not be evident at another. A developmental lag rather than psychopathology may be the basis for the probl e m . 4. Assessment c a n be time consuming, a n d sufficient time may not be made available by some parents or school officials.

Formulating Hypotheses The process of formulating hypotheses starts with a review and analysis of all data sources, i ncluding formal and i n formal assessment results, observations, i nterviews, clinical history, and prior assessment i n formation . O n the basis of this review hypotheses are formulated, and con­ firming evidence is then sought. Those hypotheses that are supported by only one piece of minor evidence are re­ garded as very tentative or dropped . Those that are sup­ ported by more than one piece of evidence - especially i f the supporting data come from more than one source (for example , test results and observations) - are retained for further consideration . Evidence that may disconfi rm each hypothesis is also sought from the various sources of information . When this process has been completed, those hypotheses that have received support can be advanced. These theories are still hypotheses - that is, tentative and unproved explanations of a complex set of data - but they are ones that can be offered with some degree of confi­ dence . The examples in Table 1 8 - 1 i llustrate how tentative hy­ potheses about temperament , personality, and level of abil­ ity can be developed from a careful analysis of a child's responses and behavior. The examples indicate that both physical and psychological factors should be considered i n attempting t o understand a child's performance. A s noted above , hypotheses developed from performance on a few items , or from a few test responses, should be verified via other data sources before they are accepted and reported. Judgment always must be used in deciding whether a response reflects the child's habitual style or is a temporary

STEPS IN THE ASSESSMENT PROCESS

and t ransient expression. A lthough you m u ·t be careful not to overinterpret every minuscule aspect of the child's per­ formance , there are occasions when hypothe es developed from unique response� may prove to be valuable.

Developing Intervention Strategies Assessment and treatment are inextri cabl y intertwined . Assessment i nvolves a careful del i neation of the child's strengths and weaknesses, a description of the child's tem­ perament and personality, the formulation of a diagnostic label in those cases where it is needed and appropriate, and the development of recommendations . Traditional forms of psychometric assessment usually do not provide information about the conditions that best facil itate the c h ild's abil ity to learn - for exampl e . how type of material , rate and modal ity of presentation , cues, and reinforcements diffe rentially affect learning. Information about these and related factors (see Table 18-2) could help us design more effective i ntervention programs that might improve the cognitive abi l ity of childre n . A weddi ng of experimental and c l i nical/psychoeducational approaches , which has been in the making for many years , has yet to occur. Until it occurs, we must use the presently available procedures ( inferences from performance on standard tests, informal tests, and special ized procedures) to deter­ mine which conditions best faci l i tate the chi ld's abil ity to learn .

Following Up on Recommendations and Retesting Effective delivery of services requires close monitoring of recommendations and interventions . Both short- and long­ term follow-up are important parts of the psychological assessment process.

Short-term follow-up.

Short-term fol l ow-up ( w it h i n two to s i x weeks) is strongly recommended, because inter­ ventions may prove to be ineffective , either because of changes in the situation or because the interventions were inadequate from the beginning . Other issues requiring additional assessment also may be discovered .

Retesting. Assessment is not a once-and-for-all matter. Children change as a result of development , l i fe experi­ ences , and treatme nt . An evaluation conducted when the child is 2 years old may have l ittle meani ng a year later, except as a basis for comparison. Retesting is an important means of monitoring the child's response to an interven­ tion , the progression of a disease process, or the course and rate of recovery. Repeated assessment is especially

537

important when a c l i n ical interve nt ion procedure - such as medical intervention ( for example , chemotherapy or sur­ gery ) or a behavioral inte rvention ( for exampl e , a cog­ nitive rehabilitation program) - is use d . Repeated assess­ ment is also required when children are pl aced in special education programs and when preschool children are c las­ sified as having a developmental disabil i ty. Unfortunately, the results of repeated evaluations may be difficult to i nterpret for the fol lowing reasons : I . We cannot easily d i fferentiate those changes that are the result of practice effects (or a retest effect alone) from those that are associated with actual c l inical improvement (or changes in adaptive and behavioral functioning) . Prac­ tice effects refer to changes in performance (usually gains) that result from prior exposure to a tas k . A h igher retest score may be the result of the intervention or learning, or it may be related to prior exposure . For example, is the improvement exhibited by a brain-injured child on retest­ ing related to practice effects or to the child's improved cerebral fu nctioning? Perhaps neurological ly impa i red children who improve on a retest are showing improved learning abil ity and not simply a practice effect . (When the change is greater than I standard e rror of measurement, one can be more confident that the score reflects a rel iable change . ) 2 . Practice effects may not occur to the same extent in all popul at ions. The practice effects typically seen among normal (or healthy) children may not occur among men­ ta l ly retarded children and chi ldren with known cerebral dysfunction . Practice effects may also differ as a fu nction of the child's age . 3 . Practice effects vary for different types of tasks. For example , practice effects are greater on the Wechsler Per­ formance Scale subtests than on the Verbal Scale subtests. Pract ice effects may be minimal on fi nger-tapping and strength-of-grip tests . (See Chapter 22 for a description of these two tests . ) Thus the amount of learning i ndicated by a particular difference between test and retest scores may vary for different tests. 4 . The length of time that elapsed between tests affe cts retest score s . Shorter i ntervals produce greater practice effects than do longer interva l s . 5 . The magnitude o f t h e practice effect may vary as a function of the score on the fi rst test . Children whose initial scores were low tend to show more improvement on the retest than do children whose initial scores were high. Such regression-to-the-mean effects occur, in part , be­ cause of random factors that vary between the initial test and retest performance . 6. Different tests purporting to measure the same abil -

C HAPTER 1 8

5 38

THE ASSESSMENT PROCESS

Table 1 8· 1 I l lustrations of How H ypotheses Are Developed from the Chi ld's Test Behavior A rea

Example

Behavioral cues

Inability to respond at the moment, rather than lack of knowledge, may be indicated when a child stares blankly ahead and does not respond to one or more questions. Distress, rather than joy, may be indicated by laughter in some emotionally labile childre n . Aggressiveness may b e indicated when a c h i l d smashes blocks . Failure to understand directions or feelings of dependency may be seen when the child builds a design d i rectly on the examiner's model . Loss of motivation and interest or a passive-aggressive expression of defiance may be indicated when increasing amounts of time are needed in order for the child to complete the trials on some subtests (such as on pegboards). The expectation is that performance should improve with practice , not get worse. Difficulty in finding a solution, depression, perfectionism , difficu l ty in making a decision, or a habitual coping style may be revealed by excessive slowness. Impulsiveness (particularly when the solution is not correct) or habitual coping style may be indicated by excessive speed . (Impulsiveness may reflect difficulty in postponing action until the solution has been thought ou! . ) Hypomanic qualities are revealed by a n expansive, outpouring , and excited response sty l e . Limited intelligence, shyness o r timidity, or deliberate intention to b e sure, though slow, m a y be revealed by failure to answer a question within the allotted period of time.

Quality of verbalization

Artistic talent may be shown by definitions of vocabulary words that are given i n h ighly sensory terms. Obsessive tendencies may be indicated by (a) fou r or five explanations of courses of action in reply to questions; (b) three, four, or more l ikenesses in response to the Similarities questions; (c) elaborate and quibbling definitions of Vocabulary words; and (d) overdetailed and doubt-laden responses. Depressive features may be indicated by s low, hesitant, and blocked responses interspersed with selfdeprecatory remarks. Paranoid qualities are suggested by querulous, d istrustfu l , and legalistic responses. Dependency may be revealed by answers such as "Ask my mommy" or "Ask my daddy." Hypochondriacal preoccupat ions may be suggested by constant reference to the body or by responses such ,. as "I should rest. Fearfulness may he revealed by responses such as "They all hu rt you . " In such cases, the child may be expressing his or her fear that the world is a dangerous place. Ex.hibitionism may be revealed by responses referring to dress, parties, and ornaments. Aggressiveness may be expressed by responses referring to weapons or battles.

"TYpe of verbalization

Hearing difficulty may be disclosed on memory items by distorting words or numbers, dropping consonants. or omitting prepositions and other connecting word s . M isleading results occur when a child w i t h severe speech defects substitutes o n e word for another simply because the second word is easier to say. For example , the child says knife instead of scissors to a picture vocabulary card simply because it is easier to say knife .

Item content

Boredom may be present when a child passes difficult items but fai l s easier ones . This type o f child may be bored by easy material and stimulated by hard materia l . Dependency or submissiveness may b e indicated b y ready and assured responses to routine material s, but anxious, hesitant, and tentative responses to questions that require judgment and evaluation. Compulsive tendencies may be revealed when a child passes tests requiring meticulousness but then fails others because he or she is either more meticulous than the test requires or too inflexible in his or her thinking . A pedantic urge to be accurate may enhance performance on memory tasks. I nhibition and a fear of being incorrect may be shown when a child fai l s tasks requiring insight and imagination but not other tasks. Difficulty with handling aggression may be indicated when a child becomes disorganized when reading a passage that has aggressive content.

539

STEPS IN THE ASSESSMENT PROCESS

Table 1 8-2 Some Areas to Consider in Assessi ng Learning Ability and Developing I ntervention Strategies A rea

Questions

Attention

I . Can the c hild attend to the task long enough to learn it? 2 . Are the essential components of the task attended to?

Motivation

3 . What motivates the child to learn? 4. Are both intrinsic and extrinsic rewards helpful?

Materials

5 . What materials faci litate learning? (For example , does color, shape, or size of materials

differentially affect learning?) Rate

6. Does the rate of presentation affect performance?

Duration

7. What is the optimal duration of presentation of the material?

Modality

8. Which modality best facilitates learning (visual, auditory, tactile, or combinations)?

Content

9 . What contents or topics are most meaningful for the child?

Procedures

10. What procedures best faci litate learning ( for example, reauditorization, stimulation, modeling,

analysis of errors)? Cueing

I I . Do cues facilitate learning? If so, which cues are successful in eliciting correct responses ( for example , first phoneme of a word, written presentation, first letter of a word , synonyms , opposites, rhymes, use of error responses)? 1 2 . Is there a hierarchy of cues? 1 3 . Can the c hild learn to produce his or her own cues? 1 4 . Which type of cue is most effective ( verba l , visual, gestural , or combinations)?

Reinforcement

1 5 . How does reinforcement affect performance?

1 6 . Is learning affected by schedule of reinforcement (continuous, intermittent, i mmediate, delayed ) , type of reinforcement (verbal, gestural o r token, positive or negative ) , o r dispenser of reinforcement (self, clinician , parent)? Practicelrehearsal

1 7 . Does the child practice or rehearse independently? 1 8 . If the child is distracted from rehearsing , is any part of the material retained?

Patterns of performance

1 9 . Is learning random or consistent? 20 . A re there any strategies evident? 2 1 . How rapidly does the child learn?

Level of learning

2 2 . What is the child's level of mastery in the various areas of learning? 23. If a specific behavior cannot be learned, are there alternatives that can be learned? 24. Can alternative communication systems be learned i f they are needed?

Retention and transfer

2 5 . Can the child learn new material , retain it, integrate the knowledge, and apply it to new

situations? Analysis of failure

26. What were the sources of difficulty in mastering the material? 27. Was failure caused by the difficulty level of materials; meaningfulness of materials; ineffectiveness of cues, instructions , or examples; or deficits of the child?

Source: Adapted from Chapey ( 1 98 1 , pp. 55-56).

CHAPTER I B

540

ity, or even the same test at different age levels, may yield scores that are not comparable . If intell igence was mea­ sured by Test A on the first occasion and Test B on the second , changes in IQ may be due to differences in the two tests and not to changes in the child. Similarly, a test that covers a w ide age range may tap different abil ities at different age s . A knowledge of the test instruments and their interrelationships is c ritical in evaluating the extent to which retest changes are meaningfu l . What does it mean t o say that there has been improve­ ment as a resulf of prior exposure? Conceivably, any im­ provement might reflect some degree of learning , given chance fluctuations in scores. But what about children who do not improve on the retest, in spite of the fact that retest changes usually have been found in the population? It may be that these children have a greater skill deficit than those who do s how improve men t . The distinc t ion between changes in performance associated w ith practice effects and those associated w ith actual c linical improvement, particularly in neurologically impaired children, is one that may be di fficult to make in practice . For the results of repeated evaluations to be maximally usefu l , data are needed on the d ifferential effects of prac­ tice in relation to such factors as item content, age , sex , abil ity leve l , and i l lness (type , location, and chronicity) . A data base that provided the normative retest changes on various tests with speci fic normal and c l inical populations would be extremely helpful in evaluating practice effect s . Any c l inical sign i ficance attributed t o changes in test scores should be corroborated by other assessment and cl inical data ; val idity data wou ld be especially important in this regard . Unt il such data become available , you shou ld take into account the factors discussed above in evaluating retest change s .

Guidelines for Designing a Clinical and Psychoeducational Assessment The fol low ing guidelines w i l l help you to increase diag­ nostic prec ision and develop better intervention plans: I . Use a broad spectrum approach fO clinical assess­ men!. Assessment procedures should focus on (a) psycho­

l og i c a l fac t o r s ( i n c l u d i ng cog n i t i ve and affe c t i v e characteristics) and (b) soc ial and interpersonal factors . Additionally, information about somatic factors should be obtained (see F igure 18-4 ) . The interaction of these factors w i l l have profound effects on a child. The more sources (with in l imits) that are used to sample the child's behav ior, the better position you w i l l be in to make firm generaliza­ tions about the child and to develop intervention strategies .

THE ASSESSMENT PROC ESS

A complete individual appraisal , culminating in the de­ velopment of recommendations , should i nc l ude considera­ tion of the foll ow i ng areas as they concern the child (see also Table 18-3 ) : sensory capab i l ities (inc luding v i sual , auditory, a nd tactile modal it ies) cognition (inc luding inte l l igenc e , reading, arithmeti c , spell ing, written language , oral language, learning abil i ty, and learning style) affect and temperament (including attention and ac­ tivity level ) motor s k i l l s (including both fi n e a n d g ross motor skil l s ) adaptive ski l l s (including sel f-help skil l s and knowl­ edge of geographical env i ronment) interpersonal s k i l l s work and study skills family, school , neighborhood , friend s , and relatives strengths, weaknesses , extent of disability or impair­ ment, and resources for change w ithin the c h i l d and fam i ly community resources •

•

•

•

•

• •

•

•

2.

Use a variety of assessmenf procedures fO evaluate fhe child. No single procedure should be used exclusively, although for some referral questions greater emphasis can be g iven to one procedure than to another. The use of multiple methods allows you to evaluate the "val idity of any one source of i nformation, since information from multi­ ple sources may be systematically compared" (Nay, 1979, p . 2 8 3 ) . Observing the c h ild function i n natural setti ngs lends ecological val id ity to the i nformation obtained in the test setting . 3 . Use more than one method to evaluate a particular domain , iffeasible. Errors in the evaluation of a particular domain can be reduced by using more than one assessment tool . For example , two d i fferent instruments may be used to measure the child's level of inte llectual functioning. Although this may not always be feasible because of time constraints , it should be done whenever possible . 4 . Evaluare the child on more fhan one occasion. As­ sessment across time helps one to determine the stab i l ity of the finding s . Repeated evaluations may not always be practica l , however.

ASSESS M E N T I N SCHOOLS

The objectives of the assessment process for children in schools can be stated i n the fol lowing way : (a) to determine the nature of the learning or behavior problem, (b) to determine the child's strengths and weaknesses i n abilities

54 1

ASSESSMENT IN SCHOOLS

I

Somatic factors

--

[

I

I �

I

The C h i ld

1

�

Psychological factors

I

1 :::::::--:-....

I

Social a n d interpersonal factors

�

Basic

Develop-

Acute events

Basic

Develop·

endowment

Acute events

Basic

mental

or changes

endowment

mental

or changes

endowment

processes

processes

1 ==--

I

Develop-

Acute events

mental

or changes

processes

Constitutional

Changes in

Acute events

Intel ligence,

Develop-

Acute events,

Social status,

factors;

cerebral,

Develop-

or diseases,

temperament.

mental

such as

prenatal,

educational

somatic, and

mental

such a s brain

and aHect

such as

changes in

trauma, and

background,

postnatal,

neural

changes

damage and

changes i n

intelligence,

changes i n life

and family

and perinatal

structures

related to

inf la mmatory

temperament,

circumstances

background

socialization,

conditions

and aHect

events

education, and family,

Acute events,

schools, birth

of sibling, an d parental

divorce

including expectations and attitudes of caregivers

Figure 1 8-4.

Factors that relate to the child's functioning. Adapted from Spiel,

related to learning, (c) to evaluate the behavior problem, (d) to develop an educational plan that takes i nto account the child's abilities and personal ity, the teacher. and the family, (e) to assess the child's response to intervention

j.jAVe,N'T

'(0'-1

C.OM PL-e'fED

..,...E.T ?

1 98 1 .

e fforts, and ( f ) to recommend modifications in programs and class placements .

Visiting the Classroom In school setti ngs, the formal assessment should be supple­ mented with v i sits to the child's c l assroom , The charac­ teristics of the particular classroom environment should be observed (see Chapter 1 7 ) , and intervention strategies tai­ lored to this environment . Classroom visits provide an added benefit - the opportun ity to establish rapport with and consu lt with the teacher. During such vi sits , you shou l d avoid i nterfering w it h regular c l assroom pro­ cedures and try to ensure that the child under study i s not stigmatized . Although every effort should be made to reduce the teache r's anx iety about your visit, teachers must u nderstand that their behav ior may be part of the problem and that changes in their behavior may be part of the solution . The child's and teacher's behaviors are usual ly so i ntertwi ned that it is al most i mpossible to examine one without the other.

Writing the Evaluation and Developing Intervention Strategies From Mainslreaming Series: Individualized Educmional Programming (IE?). © 1 977 by Judy A . Schrag, Thomas N . Fairchi l d . and Bart L M i l ler. published by Teaching Resources Corporat ion , H ingham, Mas­ sachusetts. Reproduced by permission. All rights rese rved.

I nformation gained during the evaluation should enable you to describe the positive qual it ies of the child's fu nction­ i ng , the difficulties he or she faces in mastering problems,

CHAPTER 1 8

542

THE ASSESSMENT PROCIESS

Table 1 8·3 Variables to Consider in a Clinical and Psychoeducational Assessment Suggested categories or descriptions

Variable Chronological age Sex Grade level Locale Race/ethniciry Family history Socioeconomic status (SES) Language Educational history Educational placement Physical and health status Intellectual ability Reading achievement Arithmetic achievement Behavioral and emotional adjustment Activity level Attention Auditory perception

Fine motor coordination

Gross motor coordination Memory

Oral language V isual perception

M al e , female Rural , small town, suburban, urban Asian-American, black, Caucasian (not Hispanic), Hispani c , native American, other Birth order, number of siblings, parents' marital statu s , behavioral and emotional adj u stment of parents and siblings , family's health history (as pertinent to referral problem), other Upper, upper-middle, middle, lower-middle, lower English, bilingual , non-English (state language) Age-appropriate grade placement, retained in grade, changed schools often , spent most of educational career in special classes Regular c l ass placement with no special education, regular class placement with special education services, self-contained special education class Sensory deficits , physical disabilities, neurological impairment, chronic illness , medications take n , other Individual's repertoire of skills, knowledge, and learning sets Ability to recognize and comprehend written symbols Ability to perform numerical computations and to know and reason with mathematical concepts Ability to build and maintain satisfactory interpersonal relationships and to display and feel emotions appropriate to the situation Kind and amount of movement displayed during a specific period of time and/or in a specific context (for example , hyperexcitability, acting out, off-tas k behavior) ' Ability to filter out extraneous stimuli and focus selectively on a task over a period of time (for example, sustained and selective attention, impulsivity, concentration, distractibility) Ability to identify, discriminate , interpret, and organize elements of acoustical stimuli (for example, discrimination, blending association, listening comprehension, auditory figure/ ground) Ability to integrate small muscles, usuaUy in conjunction with visual perception (for example, observe handwriting, finger opposition, kinesthetic and tactile discrimination, finger motor speed) Ability to organize movements of the large muscle groups (for example, observe running, throwing , jumping, hopping, strength, clumsiness) Ability to store and retrieve information, regardless of mode of input (for example, short- and long-term memory, visual and auditory memory, sequencing, free recall , incidental learning, rehearsal skills) Ability to comprehend and produce verbal material in the spoken mode (for example, observe semantics, syntax, vocabulary, ability to communicate) Ability to identify, discriminate , interpret , and organize physical elements of visual stimuli (for example, as indicated by visual acuity, matching, visual-motor integration, figure/ground perception, spatial relations)

Source: Adapted from Keogh, Major-Kingsley, Omori-Gordon , and Reid ( 1 982).

and the qual ity and style of his or her intellectual and social functionin g . It is unlikely, however, that the test results w i l l provide the precise causes o f a child's school fai lure , such as the etiology of speci fic learn i ng deficits. The assessment report should not merely enumerate the child's deficits or areas of fai lure ; i t should also spec i fy

abi l i t ies that the child m i ght be able to use to master task s . By identi fy ing compensatory abilities , you c a n establish a base for developing i ntervention strategies . Consideration should be given to what the child can do, as wel l as what the child cannot do. D i agnosing educational handicaps is a complex and

ASSESSMENT IN SCHOOLS

difficult tas k . It is often even more difficult. hO\l.ever, to spec i fy with some degree of certa inty the procec! urcs needed to amel iorate the chi ld's problems, to create appro­ priate conditions for learning, and to foster p ychological development, social adj ustment, and successful participa­ tion in the community. Concern w ith intervention should not tempt us to go beyond the l i m its of our present knowl­ edge . The results of an assessment do not furnish us with wel l -validated intervention techniques - developing i nter­ ventions currently is as m uch an art as a science . In spite of the problems associated with developing interventions. you should work w ith the child's teachers to establish a set of i nstructional objectives that they can use . These objectives can include determining the tasks that are to be taught , arranging a hierarchy of instructional objec­ tives, elimi nating behaviors incompatible w ith effective instruction, and teaching the tasks most needed by the child.

Assessment - A Continuous Activity The assessment recommendations are not meant to be the final solution to the child's difficultie s . They are starting PO; '1ts for you , as cl inician. and for those responsible for the mtervent ions . Assessment should be a continuous ac­ tiv ity. w ith mod i fications made in the initial intervention plans when the child's needs change or when the plans do not work e ffectively. A l l 100 often , tests are given and recommendations are made and then the psychologist van­ ishes until the reeval uation years later. Effective consulta­ t ion requires continuous monitori ng of short-term fol io\\'­ up contacts.

A Recommended Assessment Model for School-Based Referrals The National Academy of Sciences' Panel on the Selection and Placement of Students in Programs for the Mentally Retarded ( He l le r. Holtzman, & Messick. 1982 ) set out to evaluate the pl acement procedures and qual ity of services provided to children referred for special education pro­ grams. Although the Panel members only addressed prob­ lems associated w ith the placement of chi ldren in classes for the mental ly retarded , their recommendations are valu­ able for the assessment of any child who is experiencing academic d i ffi cul ties . The Panel recommended that the assessment of referred children be conducted in two phases, as outlined below ( from Mess ick . 1984 ) .

I . A systelllatic evaluation of the child's learning en-

543

vironment and (he IlatUre and qualiry of the regular in­ struction receil'ed shou ld be conducted. A . Deficiencies i n the learning env i ronment should be ruled out . ( I f any de ficiencies are present, steps should be taken to improve them before the child is referred for an individual assessment . ) I . I s there evidence that the school is using programs and curricula that are effective for children of various ethnic and socioeconomic groups? Standardized achieve­ ment test data are one source of evidence . 2 . Is there evidence that the students in question have been exposed to the curricul u m adequately (for example, few absences or discipl inary exclusions from class)? 3. Is there evidence that the teacher has i mplemented the curriculum effectively (for example, proper choice of instructional strategies and rewards)? Observational data provided by a school psychologist , consultant , or another teacher are sources of evidence . B . Documentation should be obtained to sho\\: that the child failed to learn under reasonable alternative i nstruc­ tional approaches. (The child should not be referred for individual assessment until reasonable instructional ap­ proaches have been tried and have proven fruitless . ) I . I s there objective evidence that the child has not learned what was taught? C riterion-referenced tests tai­ lored d i rectly to the curriculum are one source of evidence . 2 . Is there evidence that systematic efforts were used to identi fy the learning difficulty and to take appropriate corrective instructional actions (for exampl e , introducing remedial approaches. chang i ng curricula, trying a new teacher)? Docul1lemation is needed that alternative pro­ cedures were attempted.

II. A comprehensive assessment bauery should be ad­ ministered that in eludes (a) measures of intellective skills , cognitive skills, and adaptive behavior and (b) procedures 10 screen for biomedical disorders. A . Assessment procedures should focus on functional ski l l s for which there exist potentially effective interven­ tions. These skills include academically relevant skills ( for exampl e , read ing , mathemat ics) , cognitive processing skills ( for example, general izat ion , self-monitori n g ) , adaptive and motivational skills (for example, i mpulse control , soc ial skill ) , and physical skills related 10 learn­ ing ( for example. vision , hearing) . B . The comprehensive battery should include a mea­ sure of global intel ligence and measures of information­ processing skills involved in comprehension , visualiza­ tion , memory, reasoning , and j udgme nt. These latter measures may be provided by subtests derived from a multisubtest intell igence test or by specialized procedure s .

544

Procedures that measure responsiveness to learning or structured teaching may also prove to be valuable. C . The comprehensive battery should also include mea­ sures that will help to determine whether the child's failure to learn is associated w ith sensory, motor, or other physical i mpairments . Limited vision or hearing , psychomotor malfunctions, or hunger may impair a c h ild's classroom performance . Similarly, neurological or endocrine dys­ functions may i mpede intellectual functioning or lead to behavioral problems . Finally, physical trauma or depriva­ t ion may lead to deficits in functioning. In each of these cases corrective treatment may be possible, particularly i n the case of sensory, neurological , or endocrine mal func­ tions . D . The comprehensive battery should contain mea­ sures of adaptive behavior, particularly in the cases of children who are being considered for classes for the mentally retarded. An adaptive behavior measure may help to identify emotional and behavioral deviance , and nonintellective adaptive strengths. Such fi ndings may as­ sist in designing i ntervention programs aimed at improv­ ing practical ski l l s and reducing maladaptive behaviors. E. Once a child is placed in a special education class, the student's progress should be monitored annually. Messick ( 1984) bel ieves that the above procedures will go a long way to ensuring the val idity of referral and assessment procedures and the qual ity of instructional services. A c h i ld should be placed in a special education program when there is evidence that effective instruction is be ing offered in that program and that a regular education c lass cannot ach ieve the appropriate educational goal s . Retention i s j ustified only when the special education placement is deemed more beneficial than a regular educa­ tion placement. The Panel's recommendations stress con­ tinuous program evaluation and performance monitoring .

ASSESS M E N T O F SPEECH A N D LA N G U A G E

The observation o f speech and language is a n important part of the assessment process. The assessment of lan­ guage and speech competency requires many technical sk i l l s . Rather than focusing on technical skills, however, this section outli nes a basic approach to speech and l a nguage a s s e s s m e n t and prov ides some i m portant guideposts.

An Approach to Language and Speech Evaluation A lthough there is no s ingle preferred way to evaluate a

CHAPTER 1 8

THE ASSESSMENT PROC ESS

child's language and speech , the following approach i s useful : I . Interview the referral source to learn about the c h i ld's l anguag e . 2 . Tal k w ith t h e c h i l d , and compare your observations w ith those of the referral source . 3 . Obtain i n formal l anguage samples (see below ) . 4 . Administer a normative language test - not neces­ sarily to obtain a standard language score , but to document the child's impa i rments . Normative testing of language must be carefu l l y applied to children of ethnic groups who d i ffer from the norm group . If these children speak non­ standard varieties of English, their language skills may be assessed as being deficient when in fact they are only d i fferent. 5. Observe the child i nteracting w ith the referring per­ son in order to determine areas of fau lty communication . 6 . Develop etiological hypotheses after evaluating the degree of language impairment .

Obtaining informal language samples. I n formal lan­ guage samples can be collected by various means , such as the fol l owing : 1 . Read a short paragraph to the child and then ask the child to repeat the story. (The story shou ld be one that is readily comprehensible . ) 2 . Have the child read a short story (one o r two para­ graphs) and then ask the child to repeat the story. (The story should be one that the child can easily read . ) 3 . Show a picture to the child, and ask the child to describe the picture . (The picture should show themes that are familiar to the child . ) 4 . Show a sequence of pictures to the child and ask the c h i ld to describe the sequence. ( Pictures should show themes that are fami l iar to the chi l d . ) 5 . A s k the child t o tell you something o n a given topic that is wel l known to him or her. 6 . Ask the child to tell you someth ing on a given topic that is somewhat abstract .

These six tasks provide language samples based on oral presentation (task 1 ) , w ritten presentation (task 2 ) . visual nonsequential presentation (task 3 ) , visual sequential pre­ sentation (task 4 ) , sel f-generated concrete content (task 5 ) , a n d s e lf-generated abstract content (task 6 ) . The tasks are useful for evaluat i ng whether the child shows disturbance of spontaneous speech (or expressive verbal behavior) in the area of repetition , reading, naming objects, or com­ prehe nsion .

545

ASSESSMENT OF SPEECH AND LANGUAGE

What to Evaluate in a Language Assessment Language assessment should focus primarily on the child's unique language characteristics, although consideration also should be given to how the child's language ability compares w ith that of age peers . A thorough analysis of language ability should i nclude an evaluation of the child's (a) consistencies within his or her own language system , ( b ) strategies and processes used to produce language , (c) language structures, and (d) performance on formal language tests . The assessment of children's language competencies is complicated by our i nability to understand fully the stages of normal language acquisition . Because formal language tests fai l to measure the context appropri­ ateness of language , every effort shou l d be made to evalu­ ate children's language as it is used in natural contexts .

Language and Speech Deviations Language difficulties may be related to syndromes or dis­ eases (such as infantile autism , brain damage , or hearing i mpairment), processing difficulties (such as l i mited mem­ ory span), or social or culturalfaCTors (such as inadequate exposure to Engl ish ) . Exhibit 18-1 covers the defi nitions, incidence, and etiology of speech disorders . Following are some of the major deviations in l anguage and speech that occur during the preschool and early school years : I . During the preschool years ( 3 to 5 years of age ) , deviations i nclude lack o f speec h , unintel l igible speech, and fai lure to speak in sentences . 2 . At about 5 to 6 years of age , deviations include substitution of easy sounds for d i fficult ones, consistent dropping of word endings , faulty sentence structure . no­ ticeable nonfluency, and abnormal rhythm , rate, and in­ flection of speech . 3 . B y the age of 7 years , deviations can include distor­ tions, omissions, or substitutions of any sounds .

Other signs of d ifficulties that are not specifically age­ related inc l ude a delay of more than a year in the ap­ pearance of i ndividual speech sounds; use of vowel sounds to the exclusion of al most all other sound s ; being embar­ rassed or disturbed by speech; consistentl y monotonous, i naudible, or poor-qual ity voice ; consistent use of a pitch i nappropriate for the child's age ; noticeable hypernasality or l ack of nasal resonance ; and unusual confusions, rever­ sals, or telegraphic speech (a type of speech in which connectives, prepositions , mod ifiers, and refinements of language are omitte d ) . When any of the above deviations are observed , the child should be referred for speech eval uation .

A chi ld's utterances may also be evaluated for signs of disturbed content or processing. Indications of disturbed processing include confabulations (for example, bizaire stories loosel y connected with the narrative they were supposed to develop) , stereotype phrases ( repeating a "pet" phrase over and over), misnamings, digression (caused, for example by accidental assoc iations) , m ispronuncia­ tions, d i fficulties in the initiation of the narrative (possible articulatory defic it) , inabil ity to recognize objects pre­ sented in the picture ( possible v isual defic it) , and inabil ity to repeat the story ( possible memory deficit ) . Table 18-4 provides examples of major kinds of language distortions . Severe language distortions may be related to mental disor­ ders or brain damage . Disc repancies may be observed in a c h i ld's language comprehension and production . Some c h i ldren fai l to dis­ criminate the distinctive contrasts in speech reception and yet make the contrasts in production, whereas other chil­ dren do not produce the contrasts but readily discriminate the m . Some children cannot imitate or understand sen­ tences that they can produce, whereas other c h i ldren can understand what they cannot produce . Evaluate where the difficulties l ie - in comprehension or p roduction . Diffe rent factors may affect the ability of children to provide names for things or actions ( W i l l iams, 1983) . These include (a) l i nguistic factors associated w ith the target words (such as how frequently the words appear in written passages and their length) , (b) characteristics of the refe rent stimulus (such as whether it is a real object or a picture of the object), (c) the manner in which response is el icited (such as sentence completion or free recal l ) , and (d) the way in which the naming response is produced (such as single word naming or nam ing in the context of connected speech) .

Questions to Consider About a Child's Language Usage Language usage is a guide to personal ity style and to thought processe s . The tempo , qual ity, and content of a child's verbal responses are also important . A nswers to the following questions w i l l provide cues about the child's general language style and communication skill s . Does the child understand what is being said? Is gesture understood? A re gestures or mime used to communicate? Does the child attend to faces or watch l ips when people speak? Does the child unde rstand oral communications when his or her back is turned to the speaker? •

546

CHAPTER 1 8

THE ASSESSMENT PROCESS

... Exhibit 1 8· 1 Definitions, I ncidence, and Etiology of Speech Disorders

Definitions Language and speech represent two different but interrelated concepts in communication. Language , a system of symbols , is evaluated in terms o f its receptive and expressive compo­ nents. Receptive language refers to the symbolic information received and comprehended, primarily through auditory and visual channels . Expressive language refers to components such as morphology, syntax, and semantics that are used orally. Speech , a motor output system, is the spoken vehicle by which thoughts, expressed symbol ical l y through lan­ guage, are communicated. Sound is initiated by respiratory air pressures that are interrupted by valving processes in the vocal tract. The valves consist of the vocal folds, the velo­ pharyngeal mechanism, tongue contact w ith the palate or teeth, and the lips. The speech system has been divided into four major areas on the basis of a processing classification system: phonation, nasal resonance, articulation, and prosody. Phonation is char­ acterized by duration, loudness, pitch, and quality of the sound. Nasal resonance refers to vibrations in the nasal cavity that may be perceived as excessive (hypernasality) or insuffi­ cient (hyponasality ) . A rticulation is related to the production of consonant and vowel sounds in various sequences to form words. Prosody is related to the melody, rhythm, and rate of the speech flow.

Incidence It has been estimated that approximately 1 0 percent of chil­ dren exhibit hearing or speech-language disorders . The ap­ proximate incidence of all speech disorders is as follows: 5 to 10 percent for phonation, 5 to 1 0 percent for resonance, 60 to 70 percent for articulation, and 10 to 1 5 percent for prosody.

Etiology Five major etiolog ical categories have been suggested : psychosocial , structural , neurological , fam i l ial/hereditary, and learning . These categories are not disorder specific, and in many instances they overlap . A psychosocial etiology may account for each of the four major speech disorders. In the area of phonation, children with voice disorders may be hoarse from abuse through "yelling" (for example, screamer's nodules ) . A whiny, hypernasal resonance may be heard in the insecure and overprotected child. Articulation disorders (for example, saying wabbit for rabbit and thull for SUIl) may be caused by emotional problems . Dysfluencies (stuttering) in the speech of the preschool child can be caused by environ­ mental stress factors. Congenital laryngeal web is an example of a structural

change that affects the phonation (voice) ability in children. The structural abnormalities of cleft lip and palate can mark­ edly impair resonance and articulation. Prosodic changes are not generally believed to be caused by physical defects. A l l four of the speech categories can be affected by a neurological etiology. Impairment may vary from congenital minimal brain damage , severe cerebral palsy, and acquired head trauma to infectious diseases , such as meningitis or Reyes syndrome . Children who have sustained neurological damage may experience difficulty in controll ing duration, loudness , pitch , or quality. Neurological dysfunction of the soft palate may prevent adequate velopharyngeal closure, which would make a difference in the flow of air through the oral and nasal cavities . Articulatory skills may also be ad­ versely affected by neurological damage ( for example, in­ coordination of tongue , lip, or jaw movement may inhibit precise production of consonant and vowel sounds ) . Distur­ bances in the melody, rhythm, and rate of speech (prosody) may be the initial indication of neurological damage. A mo­ notonous, staccato, or highly variable speech flow may be associated with cerebellar or basal ganglia impairment. Hereditary and /a m ilia l factors may also account for speech disorders. M a ny cleft l ip and cleft palate disorders are famil­ ial. Stuttering has long been suspected of having a hereditary or familial origin. Evidence from twin studies suggests that familial articulatory d isorders may occur. Transmission of stuttering in familie s may be explainable by a genetic hypoth­ esis. Many genetic syndromes ( for example, Huntington's chorea) are accompanied by a variety of speech disorders involving the categories phonation, resonance, articulation, and prosody. Speech d isorders may resul t from learning from poor speech models in other family members. Speaking with a breathy or nasal voic e , for example , can occur if the child mimics an older sibling or parent. Delays in learning proper speech patterns have been noted in some children who come from bilingual home s . M isarticulation, the most common speech problem , may result from the learning of improper sound usage, even in families who use excellent speec h . A rticulation errors a l s o develop from t h e learning o f inap­ propriate phonological rules. The etiological basis for the development of misarticulation is largely unknown, and is generally described as a functional articulation disorder. Source: Repri nted , with changes in notation, with permission of the publisher and authors from W M . Diedrich and D. B. Carr, "Identifi­ cation of Speech Disorders," Journal of Developmental and Behav­ ioral PediaTrics, 1 984, 5 , pp. 38-39. © Williams & Wilkins.

547

ASSESSMENT OF SPEECH AND LANGUAGE

Table 1 8-4 A Variety of Spoken Language Difficulties or Distortions Type of difficulty or distorrion

DefiniTion

Agrammatism

Omission of syntactical (grammatical , functor, or filler) words.

Anomia

Difficulty in finding the right word (for example, "Well, he, uh , just hurried along" when the child meant to say "ran" or "The thing you put in your mouth" for "spoon").

Blocking

Intermption of a train of speech before a thought or idea has been completed.

Circumstantiality

Pattern of speech that is very indirect, reaching its goal idea belatedly. The child brings in many tedious details and sometimes makes parenthetical remarks.

Clanging

Pattern of speech in which sounds rather than meaningful relationships appear to govern word choice. The intell igibility of the speech is impaired, and redundant words are introduced.

Derailment

Pattern of spontaneous speech in which the speaker goes from the original idea to one that is clearly but obliquely related, or to one that is completely unrelated. Things presented in juxtaposition may lack a meaningful relationship, or the child may shift idiosyncratically from one frame of reference to another. At times, there may be a vague connection between the ideas; at others, none will be apparent.

Distortion of the sounds of language, either in perception or production

Unintelligible speech by a child, usually younger than 8 or 9 years of age .

Distractible speech

Stopping repeatedly in the middle of a sentence or idea and changing the subject in response to a nearby stimulus, such as an object on a desk or the interviewer's clothing or appearance.

Dysarthria

Poor articulation.

Dysprosody

Speech that is unmelodic and dysrhythmic .

Echolalia

Pattern of speech in which the child echoes words or phrases of the interviewer.

Illogical ity

Pattern of speech characterized by illogical conclusions.

Incoherence

Speech that is essentially incomprehensible at times.

Loss of goal

Failure to follow a chain of thought through to its natural conclusion . The child begins with a particular subject and then wanders away from it, never to return to it.

Neologisms

New or unconventional words or phrases whose derivation cannot be understood. Neologbms can be created through (a) structural changes, where syllables are scrambled (helicopter is produced as copiheTer) ; (b) semantic changes, where a word that has similar connotative meanings is used ( pull-ons for boots); and (c) phonological changes , where sounds are exchanged (brod for bread) .

Paragrammatism

Incorrect use o f verbs, clauses, or prepositional phrases.

Paraphasia

Substitution of incorrect words for other words (for example, cow instead of spoon or "The flower is on the garden" instead of "The flower is in the garden").

Perseveration

Persistent repetition of words, ideas, or subjects . The utterance or idea that is perseverated may be at first meaningfu l , but its continuation is inappropriate.

Phonemic paraphasia

Recognizable mispronunciation of a word in which sounds or syl l ables a re out of sequence (for example,

psgherri for spagherri) . (Table continues next page)

548

CHAPTER 1 8

THE ASSESSMENT PROCESS

Table ' 8-4 (cont.) Type of difficulry or distorrion

Definition

Poverty of amount of speech

Restriction in the amount of spontaneous speech . The replies to questions tend to be brief, concrete, and unelaborated. Unprompted additional information is rarely provided .

Poverty of content of speech

Restriction in the information conveyed by speech . A lthough long enough , replies convey l ittle information. Language tends to be vague, often overabstract or overconcrete , repetitive, and stereotyped.

Pressure of speech

More spontaneous speech than is considered ordinary or socially customary. Speech is rapid and d ifficult to interrupt. Some sentences may be left uncompleted because of eagerness to get on to a new idea.

Self-reference

Speech patterns in which the child repeatedly refers the subject under discussion back to h i mself or herself when someone else is talking .

Stilted speech

Speech that has an excessively stilted or formal quality. It may seem rather quaint or outdated, or it may appear pompous, distant, or overpolite .

Syntactic errors

Unconventional ordering of words (for example, "My house, well , I l i ve in, wel l , my house, u h , I live in" for "I l ive in my house") .

Tangentiality

Replying to questions or statements in an oblique, tangential , or even irrelevant way ( for exampl e , when Bill is asked to pick up the blue notebook, he picks up the blue car and says "car") .

Telegraphic speech

Utterances that contain a subject, main verb, and object, but no functor or other "small" words (for example, "I got tricycle" for "I've got a tricycle") .

Topical or referential identification problems

D ifficulties in selecting the appropriate referent (for example, identifying the appropriate topic) in a conversation (for example, when Ann is asked "Did you put the book on my desk?" she responds "My desk is clean").

Word approximations

Using old words in a new and unconventional way or developing new words by using conventional rules of word formation.

Is there any speech? Are vocalizations normal? Is articulation normal? What is the usual tempo of the responses? Are they quick or slow? What is the output (approximate number of words spoken per m i nute)? What amount of effort is needed to produce speech (for example, do you note any visible struggles, facial g r i mace s , body post u r i n g , deep breat h i n g , or hard gestures)? What is the phrase length (for examp l e , s ingle words or short phrases or complete sentences)? • Is the speech melodic? • Does the c h i l d take t i me to consider before re­ sponding? Is thinking done aloud or is only the final answer given? How competent is the child in the use of words and •

•

•

•

•

grammar? For example, does the child's language con­ tain objects , actions , and events in a variety of relation­ ships? Does the child echo what i s said or use words in­ appropriately? Are utterances appropriate for the context? Are words used for social communication? • Are ideas expressed c l early? Does the child distinguish between relevant and irrele­ vant i n formation? Does the child verbalize a variety of possibilities and perspectives? Are responses self-critical? Can the child assume the role or v iewpoint of another? Are imitation and symbolization used in play? Does the child, while playing, respond appropriately to the speaker? Do l anguage difficulties, i f present, cross reading, writing, and speaking situations? •

•

•

•

•

•

•

•

549

LEARNING POTENTIAL ASSESSMENT

Examples of a Qualitative Language Anal) sis

LEAR N I N G POT E NTIAL ASSESS M E N T

The two i l l ustrat ions bel ow ( Luca s . 1980. p. 125 and p. 130, with changes in notat ion ) show how a qualitative analysis of a chi ld's language may appear i n a report . The first description i s of a 5 -year-old girl , and the second is of a 7-year-old boy.

Learning potent ial as�essment procedures are designed t o measure the chi ld's abi l ity t o improve performance fol low­ ing a systematic learning experience ( Budoff & Hami l ton , 1976 ; Feuerste i n . 1979 : Guthke . 1982 ; Hegarty & Lucas , 197 8 ) . The child i taught principles that will help him or her to solve a particular problem. For example, the princi­ ples that underlie the construction of a particular block pattern on a block design task might be taught . By measur­ ing the child's abi l ity to solve additional block design problems, one obtains an esti mate of his or her abIl ity to learn from the teaching and to apply this learning to similar problems . Learning potential assessment focuses on what the child can learn , how readily the child acquires strat­ egies, and whether the strategies are transferable (gener­ al i zable ) . The procedures i ncorporate m i n i - learning situations that permit the systematic observation and evalu­ ation of the child's responses to structured teach ing. I n a learn ing potential procedure the child is given progressively more complex tasks, with cont inual teach­ ing and feedbac k. The aim is to evaluate the child's abil ity to acquire and use problem-solving skills. An enrichment program - which stresses the acquisition of learn ing strat­ egies rather than the acquisition of spec ific subject-area content - may also be t ied to the evaluation procedure .

A qualitative analysis of the language sample utterances revealed the presence of semantic word errors ( for example. \'Qse for cup . table for bed, reacher for mailman) ; neologisms ( for example. pardor for parror. 1(IIIIl for !amp - these could be auditory mis­ perceptions; an ear wirh for fork . and Johnny jump for jump rope) : space and time errors ( for example . O\'er for nexr lO . runs TO for runs away, jumps up for stands) : syntactic errors related to semantic problems in organization and sequencing ( for exampl e . "The um, girl , um, found . u m . t h e lamp found t h e glrr and ''The teacher, u h . erased. uh the teacher, um erased the teacher" ) . A qualitative analysis of t h e l anguage sample indicated several auditory mi perceptions ( for example . boin for join . but most of these were unintell igible even though articulation was without error on the word tes t ) ; some semantic word errors related to parts of pictu res or meanings of the story rather than to all of the available information ( for example, a chair was described as missing a leg because the boy sitting in the chair obscured the fourth leg , a boy was said to be drowning because he was splashing even though the other pictures related to the story showed the boys playing in the water ) ; and an apparent deficiency in spatio-temporal words and some errors in the terms used ( for example, on for in , above for in /ronT oj) . When Jerome showed signs of anx iety. the word fi nd ing and verbal perseveration d i ffi­ culties became evident .

Three-Step Procedure A three-step test-train-test sequence u ual ly is followed in a learning potential procedure .

ASSESSM E NT O F VISUAL-MOTOR A B I LI TY

V i sual-motor abil ity is assessed primaril y by means of special test s , such as the Bender-Gestalt or the Develop­ mental Test of V i s ual-Motor Integration (see Chapter 14) . Items o n intell igence tests and special abil ity tests may also require v isual -motor skills, however. When you adm inis­ ter these items, you should pay part icular attention to chi ldren's reaction time and to their trial-and-error meth­ ods . Signs of visual -motor difficulty include much trial and error, fu mbl ing, tremor, repetition of copy ing errors . per­ severation, rotation , and variable l i ne qual ity. V i sual­ motor di fficulties may be associated with defects in visual , auditory, or muscular functions, with poor physical condi­ tion , with anx iety, or w ith other conditions .

I . Initial testing. The child is tested in the standard fashion with test stimuli that are minimally dependent on cultural exposu re . The stimul i used may be unfami l iar to the child, particularly when the stimul i are incorporated into the teaching procedure . Nonverbal reasoning tasks frequently are used , such as those found on the Kohs Block Design or Raven's Progressive Matrices. 2 . Teaching ofprinciples. The child receives instruction in sol ving the task . Principles include attending to all choices and using analytic skil l s . Various techniques are used during the tra i n i n g period , including feedback, prompts, and extended teaching with practice in item­ solving strategies . Training varies depending on the task. The goal of training is to help the child develop appropriate problem-solving strategies . 3 . Retesting. The child is retested on the original task or an alternative form of it.

s so

Aims of Learning Potential Assessment Learn ing potential assessment aims to evaluate the child's abil ity to acquire new information and problem-solving strategie s . Scores reflect not only the number of correct solution s , but also the number of helping steps needed to solve the problem . In a learning potential assessment , it is necessary to consider such factors as I . The stimu l i used for teaching ( for exampl e , spatial­ visual and other concept formation task s , number seria­ tion , pictorial series, geometric series , paired-associate learning tasks , or auditory-rote learning tasks) ; 2 . The teaching methods used (for example , feedback l i n ked to each step , open-ended feedback , or social or monetary reinforcement) ; 3 . The extent to which the child knew the correct re­ sponses before teaching was started; 4 . The extent to which the tasks and activities i nvolved in the test were unfamiliar to the c h i l d .

Ideall y t h e c h i l d should neither know t h e correct responses before teaching starts nor be fami liar with the test stimuli . I f e ither of these conditions i s not met , it w i l l be difficult to evaluate the effectiveness of the assessment because scores w i l l be confounded (that i s , they w i l l reflect both present learning and prior learn i ng) . In practice , these two as­ sumptions may never be fully met. As a result of training, some children w i l l improve their performance but others w i l l not . Those children who do not learn under one type of teaching strategy may learn more efficiently under another strategy, however. Simi­ larly, those childre n who perform poorly on some tasks may learn more rapidly on others. Consequently, every learn i ng potential assessment should include a variety of teaching strategies and tasks.

CHAPTER 1 8

THE ASSESSMENT PROCESS

Table 1 8-5 Comparison of Trad itional I ntel ligence Testing with Learning Potential Assessment Intelligence testing

Learning potellliai assessment

1 . Focuses on the content of

1 . Focuses on the process of

2. 3.

4. 5.

learning Attempts t o determine what has been learned Attempts t o determine how much of the material has been mastered Focuses on description of the child's performance Attempts t o predict future performance

learning 2 . Attempts to determine

why learning takes place 3 . Attempts to determine how material is mastered 4 . Focuses on a n explanation of learning 5 . Attempts t o prescribe appropriate remedial

interventions SOUTce: Adapted from Hegarty and Lucas ( 1 978).

The two procedure s , however, overlap considerabl y. Both focus on cognitive processes and provide information useful in describing and explaining the c h ild's present level of functioni ng . A child enters both an intelligence assess­ ment and a learni ng potential assessment w ith the same l e a r n i n g h i sto ry. B o t h p rocedu re s m e a s u re c u rrent achievemen t , although i n different way s . And although there is not a one-to-one correspondence between i ntel­ l igence and learn i ng abi l ity (as defined by a learning po­ tential assessment) , inte l l igence is related to the ability to learn . Intell igence and learning ability are complex multi­ dimensional concepts , w ith the forme r more easily defined and measured than the l atter.

Pitfalls in Interpreting the Word "Potential" Learning Potential Assessment and Intelligence Assessment Learning potential assessment diffe rs from traditional i n­ tel ligence testing i n a number of ways (see Table 18-5) . Learni ng potential assessment attempts to measure the process of learn i ng rather than the content of learni ng (Hegarty & Lucas, 1978) . Traditional intelligence test i ng focuses on the child's present level of cognitive develop­ ment ( what and how much has been mastered or learned) ; learning potential assessment focuses on the child's present learni ng process ( why and how the material has been mastered) . One procedure emphasizes description (intel­ l igence testing ) ; the other, explanation (learning potential assessment) .

The word potential i n the term learning potential assess­ ment may be misleading. A child's potential for learning is based primarily on present (and past) learnings . Because intelligence tests are useful in maki ng j udgments about a child's learning potential , attempts to equate learning po­ tential procedures exclusively with the measurement of learning potential are wron g . For example , a child with an IQ of 130 has a much higher potential for learning i n school than does a child w ith an IQ of 70 . There is no body of evidence indicating that scores on learning potential tests are better i ndices of future school-related achievement than are scores on i ntelligence tests (see, for example , Bailey, 198 1 ; Bradley, 198 3 ) . Although learning potential procedures are still experi mental , their i nclusion as addi­ tional components of an assessment battery deserves con-

55 1

CLASSIFICATION AND LABELING

sideration, particularly when the results t hey [u rm h aie l i kely to i mprove understanding of the child and aid in the design of remediation programs .

CLASS I F I CATION A N D LA B E L I N G

Classification systems have two major features (Zigler, 1982) . First, they establish rules for placing an i ndividual into a specific c lass; these rules establish the system's rel iability. Second, they provide information about the correlates of class membership - that is, what we know about individuals given a class designation. The robust­ ness of these c l ass correlates establishes the system's val i dity. Every description in some sense i mpl ies a labe l . The term normal range of functioning, for example, labels some aspect of the child's functioning. Although labels provide a poin t of reference , they should not be the basis for c l inical decisions. Such decisions should be based on the entire assessment results and clinical h istory.

Arbitrariness of Classification Systems Classifications such as moderately mentall y retarded (for example, IQs of 40 to 54) and mildly mentally retarded (for example, IQs of 55 to 69) are arbitrary cutoffs on a continuum of intelligence test score s . A child with an IQ of 54 is very similar to one with an IQ of 5 5 , even thoug h their classifications are different. Simi larly, a child with an IQ of 68 is l i ke a child w ith an IQ of 70, yet the former child is labeled mentally retarded, whereas the l atter is not . Al­ though i n developing your assessment findings and recom­ mendations, you should be guided by the child's perfor­ mance and not by a c lassification system of arbitrary cut­ off scores on an intelligence test , you must adhere to a classification system's speci fic cutoff points and labels when reporting such results (see , for example, Chap­ ter 2 1 ) .

Self-Fulfilling Prophecy - How Accurate? It has been alleged that plac ing a label such as mentally retarded on a child i n itiates a self-fulfi ll i ng prophecy. That is, the label is said to induce i ndividuals who come i n contact with t h e l abeled child t o treat t h e child in accord with the label , thereby producing a potentially i rreversible c hange in the child's status. Although negative stereotypes are often associated with behavioral deviancy labels and with labe l s i ndicating low levels of intellectual function-

ing, research i ndicates that children's classroom perfor­ mances are a m uch more potent force in i nfl uencing teach­ ers' expectancies than are labels given to the children (Brophy & Good, 1970 ; Dusek & O'Connel l , 197 3 ; Good & Brophy, 197 2 ; Yoshida & Meyers, 1975) . I t seems reasonable that observation of a child's class­ room behavior over weeks and months should play the crucial role in the formation of expectancies . When a teacher is told that a new child in the class is mentall y retarded , his or her expectations are l ikely t o b e generated by the labe l . These expectations w i l l be modified, how­ ever, if the child performs at grade leve l . The initial im­ pression will be tempered by the child's actual classroom performance . Thus although labels often initiate expecta­ tions, they hold l ittle power once the observer obtains direct information about the child's functioning. (Teacher expectancies are also disc ussed in Chapter 19 . )

Value of Classification and Labeling Some cl inicians object to diagnosis and classification, maintaining that labels should be avoided because they have a medical , disease-oriented connotation, provide no explanation of the child's d i fficulties, tell us nothing about the steps necessary for remediation, and lead to a preoc­ cupation with finding the right label rather than focusing on treatment. Diagnosis and classification, however, play crucial roles in the assessment process . D i agno is and classification lend organization to the complex and heterogeneous area of exceptionality and aid in the study of etiology, in the development of program s , in the evaluation of the outcomes of intervention programs , in communication with professionals , in problem solving, and in obtaining needed funding. Diagnostic labe l s are helpful in record keeping, statistical reporting, and the administration of treatment programs and researc h , help to point out areas in which study is needed, and serve as a ral l y i ng point for groups. Labeling may also have beneficial consequences , such as increasing altruistic behavior and tolerance on the part of those who must deal with deviant behavior ( Fe rnald & Getty s , 1980 ) . For example, a label such as mental retar­ dation may el icit more protective , altruistic responses from people than does the label normal. For some parents, labels provide closure - a sense of relief that their child's problem is understood - enabl ing them to view their child in a more favorable light. W ithout labels, parents (and others) may develop u n realist ically high expectancies, which in turn can lead to fai l u re , frustration , and low self­ esteem on the part of the child.

552

Comment on Labeling We must not al low the above-mentioned advantages to make us complacent about the effects of labeling . Labels do set up expectancies, and such expectancies may influ­ ence the observer's behavior, especially when contact w ith the referred child is l i m ited. I n some situations, it is con­ ceivable that the expectations generated by labels could be so powerfu l as to lead to severe restrictions of the child's opportunities. We know l ittle about the frequency of such occurrences, but even the thought of such a possibility is a potent rem inder of the importance of avoiding inappropri­ ate labeling.

T H E DSM·I II·R AND CLI N ICAL A N D S C H O O L PSYCHOLOGY

The Diagnostic and Statistical Manual of Mental Disor­ ders (Third Edition - Revised) , or DSM-III-R (American Psychiatric Association , 1987 ) , represents the most recent diagnostic classification scheme in the field of mental disorders . Classification procedures have been revised in order to achieve greater clarity of diagnostic criteria, re­ define major conditions to be con s i st e n t w ith research fi ndings, establish new categories , provide a multiaxial classification system, and provide a defin ition of mental disorder. In the DSM-III-R a mental d isorder is defi ned in the follow i ng way : Each of the mental disorders is conceptualiz('d as a clinically significant behavioral or psychological sync\rorne or pattern that occurs in a person and that is associated with present distress (a painful symptom) or disabil ity ( impai rment in one or more

CHAPTER 1 8

THE ASSESSMENT PROCE SS

important areas of functioning) or with a significantly increased risk of suffering death, pain, disabil ity, or an important l oss of freedom. In addition, this syndrome o r pattern must not be merely an expectable response to a particular event, e .g . , the death of a loved one. Whatever its original cause, it must cur­ rently be considered a manifestation of a behaviora l , psychologi­ cal , or biological dysfunction in the person . Neither deviant behavior, e .g . , political , religious, or sexual , nor conflicts that are primarily between the individual and society are mental disorders unless the deviance or conflict is a symptom of a dysfunction in the person, as described above . ( p . xxii)

W ith regard to etiology, the DSM-III-R generaJly takes an atheoretical approach. That is, the DSM-III-R does not exclusive l y favor biological , psychosocial , or socio­ c u ltural factors as the primary cause of abnormal behavior. Rather, it emphasizes descriptions of mental disorders without resort to theoretical underpinnings . Although the DSM-III-R is considered to be the defini­ tive guide to psychiatric classification , its psychological and practical utility have been questioned . The reliability and val idity of some DSM-III-R categories have not been substantiated, and some syndromes in the DSM-III-R have not been clearly substantiated in empirical studies using m u ltivariate procedures . I n spite of these potential prob­ lems, the DSM-III-R constitutes a major contribution to psychiatric classi fication and nomenclature . It attempts to provide more explicit criteria for d iagnosing disorders than were found in previous editions. The m u ltiaxial classification system used i n the DSM­ I I I -R has the following fi ve axe s : • A x i s I - Cl inical Syndromes; Conditions Not Attrib­ utable to a Mental Disorder That Are a Focus of Attention or Treatment

Downstow n , Copyright 1 983 , Universal Press Syndicate. Reprinted with permission. All rights reserved.

553

THE DSM·III·R AND CLINICAL AND SCHOOL PSYCHOLOGY

• A x i s I I - Deve l op me n t a l D i sord e r s ; Person a l ity

Disorders Axis I I 1 - Physical Disorders and Conditions Axis IV - Severity of Psychosoc ial Stressors • Axis V - Global Assessment of Functioni ng

• •

Table 1 8-6 briefly describes the five axes. The entire classification of mental d isorders i s contai ned in Axes I and I I . Where app licab l e , each clas s ification (or code) in the DSM-III-R's Axes I and I I i s described w ith the follow­ ing : essential features ( features that are generally required to make the d iagnosi s and that are always present ) , associ­ ated features (features that are ofte n , but not invariably, present) , age at onset, course , impairment, complications, predisposing factors , prevalence , sex ratio, fami l ial pat­ tern , differential diagnosis (how to d i stinguish the disorder being d iscussed from similiar d i sorders), and diagnostic criteria (a l isting for each disorder of specific criteria that are needed to make the diagnosi s ) . A brief discussion of each of the five axes fol lows. You are urged to consult the DSM-III-R for a complete description of the multiaxia! classification system .

Axis I Axis I contains all of the major c l in ical syndromes, except for developmental and personality disorders. Two other types of codes are also i ncluded : one for conditions that are not attributable to a mental disorder but are, nonetheless, a focus of referral or treatment ( for example, uncompl icated bereavement or parent-child problem) and the other for unspecified diag noses , no diagnosis on Axis I or I I , or defe rred diagnoses . A ny classification in the DSM-lII-R can be used for child re n , although some di sorders may not be evident in childhood . (Chapter 20 discusses the essen­ tial features of the attention-deficit hyperactivity di sorder and conduct d i sorder. )

Axis I I Axis I I contains two major classification codes : person­ al ity d isorders and developmental d i sorders . The person­ a l ity disorders c lassification, which is not l isted in Table 18-6, is applied primarily to adults ; it is used when there are enduring, maladaptive patterns or traits . Personal ity d i sorders may sometimes emanate from a di ordered childhood . Three adult personality disorder classifications have C h ildhood counterparts that are represented on Axis 1: antisoc ia l personal ity d i sorder - conduct disorder ; avoidant personal ity d isorder - avoidant disorder of child-

hood o r adolescence; and borderl ine personality d isor­ der - identity di sorder. W ithin the developmental disorders c lassi fication fal l mental retardation , pervasive developmental disorders, and specific developmental disorders . Specific categories within the developmental d isorders classifications are briefly examined below. (Chapter 2 1 provides an i n-depth di scussion of mental retardation, and Chapter 20 discusses pervasive developmental disorders and specific develop­ mental di sorders . )

Mental retardation.

The diagnosis of mental retarda­ tion essentially fol lows one proposed by the American Association on Mental Deficiency. Three specific criteria are l isted. First, intellectual functioni ng must be signifi­ cantly subaverage - an I Q of 70 or below i s usual l y re­ quired . Second, there must be significant deficits or i m ­ pai rments i n adaptive functioning, with the child's age taken into account. Third , onset of the condition must be before 18 years of age . Although the DSM- I II-R recognizes that an individual inte l ligence test must be used to obtai n an IQ, the guide­ l i nes fail to note that an I Q of 68 is a more appropriate cutoff for tests that have a standard deviation of l 6 . Fu rther­ more , the DSM - I I I-R indicates that a band between IQs of 65 and 75 can be used rather than a specific number. This flexibil ity was added to the DSM-I I I-R to allow for a diagnosis of mental retardation when there are significant deficits in adapt i ve behavior. Conversely, IQs somewhat lower than 70 do not automatically result in a diagnosi s of mental retardation when there are no signi ficant defic its in adaptive behavior. The DSM· I II-R advocates that , ideal ly, an adaptive behavior scale be used in conjunction with clin ical judgment of general adaptation for the assessment of adaptive behavior. When an individual beyond 18 years of age develops mental retardation, the classificat ion of dementia, an organic mental disorder, is used. When mental retardation develops before the age of 18 years in a child who pre­ viously had normal inte l l i gence, both the mental retarda­ tion and the dementia c1assi fications are used .

Pervasive developmental disorders.

The two catego­ ries under pervasi ve developmental disorders are autistic di sorder and pervasive developmental d isorder not other­ wise speci fied . These classifications replace such diag­ nostic terms as atypical deve lopment, symbiotic psycho­ sis. childhood psychosis, and childhood schizophrenia. Aut istic disorder also is known as infantile autism.

CHAPTER 1 8

554

THE ASSESSM ENT PROCESS

Table 1 8-6 DSM-III-R C lassifications That Are Most D i rectly Applicable to C hi ld re n and Adolescents Axis I. Disorders Usually FirsT Evident in Infancy, Childhood, or Adolescence

Gender identity disorder of adolescence or adulthood, nontranssexual type Gender identity disorder not otherwise specified Tic Disorders Tou rette's d isorder Chronic motor or vocal tic disorder Transient t ic disorder Tic disorder not otherwise specified Elimination Disorders Functional encopresis Functional enuresis Speech Disorders Not Elsewhere Classified Cluttering Stuttering Other Disorders of Infancy, Childhood, or Adolescence Elective mutism Identity disorder Reactive attachment disorder of infancy or early childhood Stereotypy Ihabit disorder Undifferentiated attention-deficit disorder

Disruptive Behavior Disorders Attention-deficit hyperactivity disorder Conduct d isorder group rype soliTary aggressive rype undifferenTiated rype

Oppositional defiant disorder Anxiety Disorders of Childhood or Adolescence Separation anxiety disorder Avoidant disorder of childhood or adolescence Overanxious disorder Eating Disorders Anorexia nervosa Bulimia nervosa Pica Rumination disorder of infancy Eating disorder not otherwise specified Gender Identity Disorders Gender identity disorder of childhood Transsexualism

Axis II. Developmental Disorders

Mental Retardation Mild mental retardation Moderate mental retardation Severe mental retardation Profound mental retardation Unspecified mental retardation Pervasive Developmental Disorders Autistic disorder Pervasive developmental disorder not otherwise specified Specific Developmental Disorders Developmental arithmetic disorder

Developmental expressive writing disorder Developmental reading disorder Developmental articulation disorder Developmental expressive language d isorder Developmental receptive language disorder Developmental coordination disorder Specific developmental disorder not otherwise specified Other Developmental Disorders Developmental disorder not otherwise specified

Axis Ill. Physical Disorders and CondiTions

Axis I I I permits the clinician to indicate any current physical disorder or condition that is potentially relevant to the understanding or treatment of the individual . Axis IV Severiry of Psychosocial STressors Scale: Children and Adolescents Examples of stressors Term

Code

Acute events

Enduring cirCUli/stances

None

No acute events that may be relevant to the disorder

No enduring c i rcumstances that may be relevant to the d isorder

2

Mild

Broke up with boyfriend or girl friend; change of school

Overcrowded l iving quarters: family arguments

3

Moderate

Expelled from schoo l ; birth of sibl i ng

Chronic disabl ing i l lness in parent: chronic parental discord (Table continues next page)

THE DSM-III-R AND CLINICAL AND SCH O O L ps. YC HOLOGY

555

Table 1 8·6 (cont.) Examples of.Hre ssors

Term

Code 4

Enduring circumstances

A cute evellls

Severe

Divorce of parents; unwanted pregnancy; arrest

Harsh or rejecting parents; chronic l i fe­ threatening i l l ness in parent; multiple foster home placements

5

Extreme

Sexual or physical abuse; death of a parent

Recurrent sexual or physical abuse

6

Catastrophic

Death of both parents

Chronic l i fe-threatening i llness

o

I nadequate information, or no change i n condition

Axis V. Global Assessment of Functioning Scale (GAF Scale) Consider psychological , social , and occupational functioning on a hypothetical cont inuum of mental health-ill ness. Do not include impairment i n functioning due to physical (or environmental) l imitations.

Note: Use intermediate codes when appropriate, e . g . , 45 , 68, 72. Code 90

I

81 80

I

71

Absent or minimal symptoms (e . g . , mild anxiety before an exam ) . good functioning in all areas, interested and i nvolved in a wide range of activities, socially effective , generally satisfied with l ife, no more than everyday problems or concerns (e . g . , a n occasional argument with family members). I f symptoms are present . they are transient and expectable reactions to psychosocial stressors (e. g . , difficulty concentrating after family argument); no more than s light impairment in social . occupational , or school functioning ( e . g . , temporarily fal l ing behind in school work ) .

61

Some m i l d symptoms (e . g . , depressed mood and mild insomnia) O R some difficulty i n social , occupational , or school functioning ( e . g . , occasional truancy, or theft within the household ) . but generally functioning pretty well , has some meaningful interpersonal relationships.

60 ' 51

Moderate symptoms (e . g . . flat affect and circumstantial speec h , occas iona l pallic attacks) OR moderate difficulty i n social , occupational , or school functioning (e g . , few friends . conflicts with co-workers) .

50 41

Serious symptoms (e . g . , suicidal ideation, severe obsessional ritual s . frequent shoplifting) OR any serious impairment i n social , occupational , o r school functioning (e.g . . n o friends, unable t o keep a job) .

40

Some impairment in reality testing or communication ( e . g . . speech is at times illogical , obscure, or irrelevant) OR major impairment in several areas, such as work or school , family relations, judgment , thinking, or mood (e . g . , depressed man avoids friends, neglects family, and is unable to work: child frequently beats up younger children, is defiant at home, and is failing at school ) .

70

I

,

I

31 30

I

21 20

Behavior i s considerably influenced b y delusions o r hallucinations OR serious impairment i n communication o r judgment ( e .g . , sometimes incoherent, acts grossly inappropriately, suicidal preoccupation) OR inability to function in almost all areas ( e . g . , stays i n bed all day ; no job, home , or friends) .

II

Some danger of hurting self or others ( e .g . , suicide attempts without clear expectation of death , frequently violent, manic excitement) OR occasionally fails to maintain minimal personal hygiene ( e .g . , smears feces) OR gross impairment in communication (e . g . , largely incoherent or mute ) .

10 1

Persistent danger of severely hurting self o r others (e . g . , recurrent violence) OR persistent inabil ity t o maintain minimal personal hygiene OR serious suicidal act with clear expectation of death.

I

,

Source: Axes IV and V reprinted, with a change in notation. by permission of the publisher from the Diagnostic and Statistical Manual of Mental Disorders (Third Edition - Revised) , 1 987. pp. 1 1 - 1 2 . Copyright © American Psychiatric Association. Axes I through I I I adapted from DSM-III-R.

CHAPTER I B

556

Specific developmental disorders.

The disorders en­ coded in Axis I I under specific developmental disorders are associated w ith circumscribed areas of development or learning and behavior that are not caused by another d isor­ der. Consequently, it is necessary to rule out other disor­ ders due to physical or neurologic impairment, pervasive developmental disorders, and deficient educational oppor­ tunities before making specific developmental d i sorders diagnose s . Because children may have associated learn­ ing, speech , and motor problems, more than one of the classifications may be used . The diagnostic criteria for

THE ASSESSMENT PROCESS

each of the specific developmental disorders are shown i n Table 18-7. Children recei v i ng a diag nosis of a developmental arith­ met i c , writing, or readi ng disorder must demonstrate skills that are markedly below the expected leve l , g iven the child's schoo l i ng and level of inte l l igence. The s ki l ls must be measu red by individuall y administered achieve ment and intell igence tests . A specific developmental d isorder classi fication is not used for c h i l d re n who a re s l ow learners . A devel opmental articulation disorder diagnosis is used

Table 1 8-7 Diagnostic Criteria for Specific Developmental Disorders in DSM-II I-R Disorder

Criteria

Developmental Arithmetic Disorder

A . Arithmetic skills, as measured by a standardized, individually administered test, are markedly below the expected level , given the person's schooling and intellectual capacity (as determined by an individually administered IQ test). B . The disturbance in A significantly interferes with academic achievement or activities of daily l iving requiring arithmetic skills. C. Not due to a defect in visual or hearing acuity or a neurologic disorder.

Developmental Expressive Writing Disorder

A . Writing skil l s , as measured by a standardized, individually administered test, are markedly below the expected leve l , given the person's schooling and intellectual capacity (as determined by an individually administered IQ test). B . The disturbance in A significantly interferes with academic achievement or activities of daily living requiring the composition of written texts (spelling words and expressing thoughts in grammatically correct sentences and organized paragraphs) . C . Not due to a defect in visual or hearing acuity or a neurologic disorder.

Developmental Reading Disorder

A . Reading achievement, as measured by a standardized , individually administered test, is markedly below the expected leve l , given the person's schooling and intellectual capacity (as determined by an individually administered IQ test). B . The disturbance in A significantly interferes with academic achievement or activities of daily living requiring read ing ski l l s . C . N o t due t o a defect in visual or hearing acuity or a neurologic disorder.

Developmental A rticulation Disorder

A . Consistent failure to use developmentally expected speech sound s . For example , in a three-year-old, failure to articulate p, b, and t , and in a six-year-old. failure to articulate r, s h ,- th, f, z, and I . B . Not due t o a Pervasive Developmental Disorder, Mental Retardation, defect in hearing acuity, disorders of the oral speech mechanism, or a neurologic disorder.

Developmental Expressive Language Disorder

A . The score obtai ned from a standardized measure of expressive language is substantially below that obtained from a standardized measure of nonverbal intellectual capacity (as determined by an individually admin istered IQ test). B . The disturbance in A significantly interferes with academic achievement or activities of daily l iving requiring the expression of verbal (or sign) language. This may be evidenced in severe cases by use of a markedly l imited vocabulary, by speaking only in simple sentences, or by speaking only in the present tense. In less severe cases, there may be hesitations or errors in recalling certain words, or errors in the production of long or complex sentences. C. Not due to a Pervasive Developmental Disorder, defect in hearing acu ity, or a neurologic disorder (aphasia) . (Table continues next pagt)

THE DSM-III·R AND CLINICAL AND SCH OOL PSY C H OLOGY

557

Table 1 8-7 (cont.) Criteria

Disorder Developmental Receptive Language Disorder

A . The score obtained from a standardized measure of receptive language is substantially below that obtained

Developmental Coordination D isorder

A. The person's performance in daily activities requiring motor coordination is markedly below the expected leve l , given the person's chronological age and intellectual capacity. This may be manifested by marked delays in achieving motor milestones (walking, crawling. sitting ) , dropping things, "clumsiness, " poor performance in sports, or poor handwriting. B. The d isturbance in A significantly interferes with academic achievement or activities of daily living . C. Not due to a known physical disorder, such as cerebral palsy. hemiplegia, or muscular dystrophy.

from a standardized measure of nonverbal intellectual capacity ( as determined by an individually administered IQ test ) . B . T h e disturbance i n A significantly interferes with academic achi£vement or activities o f d a i l y living requiring the comprehension of verbal (or sign) language. This may be manifested in more severe cases by an inability to understand simple words or sentences. In less severe cases, there may be difficulty in understanding only certain types of words, such as spatial terms , or an inability to comprehend longer or more complex statements. C. Not due to a Pervasive Developmental Disorder, defect in hearing acuity, or a neurologic d isorder (aphasia) .

Source: Reprinted with permission of the publisher from the Diagnostic and Statistical Manual of Mental Disorders (771ird Editioll - Revised) . 1 987. pp. 42-45 , 47-49. Copyright © American Psychiatric Association.

when there is a failure to develop consistent articulation of the later-acquired speech sound s , such as ,. , sh , th , j, z, and I. A developmental expressive language disorder or recep­ tive language d isorder diagnosis is used when the child's score on an expressive or receptive language test is sub­ stantially below that obtained on a standardized nonverbal measure of intel l ectual abil ity. A diagnosis of developmen­ tal coordi nation disorder is made when the child's motor coordination is markedly below the expected level for his or her age and i ntellectual capacity. The impairment must s igni ficantly inte rfere w ith academic achievement or ac­ t ivities of daily l iving. The diagnosis of spec ific develop­ mental d i sorder not otherwise specified is used for disor­ ders that do not meet the other criteria, such as aphasia or spe l l i ng difficulties.

Axis III A x i s I I I i s used to indicate current physical disorders or conditions (such as diabetes or asthma) that may be rele­ vant to understanding or treating the individual . These conditions are not included in either Axis I or Axis I I .

Axis IV and Axis V Axes IV and V can be used i n special clin ical and research settings to provide information that supplements the pri-

mary D S M - I 1 I - R diagnoses contai ned in Axes I , I I , and I I I . These last two axes can be especially useful for plan­ ning treatments and predicting outcomes . Axis IV is used to identify and evaluate psychosocial stressors relevant to the assessment and t reatment process by means of a 6 - po i nt rati ng scale (see Table 1 8-6) . The rating scale also has a 0 category to allow for a "no information" or "not applicable" rating. Axis V rates the individual's overal l level of psychologi­ cal , social, and occupational function ing at the time of the eval uation and during the past year. The ratings are made on a 90-point scale (see Table 18-6) , which represents a hypothetical continuum of mental health-illness . ( H igher ratings indicate better mental health . ) Two rati ngs are made: (a) the level of functioning at the t i me of the evalua­ tion (cu rrent rating) and (b) the highest level of functioning attained in at least a few months during the past yea r - at least one month during the school year should be incl uded for children and adolescents ( past year rating ) .

Classification with DSM-III-R Axes Typically, all of the D S M - II I - R axes are used in classifying a child or adolescent with a significant d isorder. Children with only clinical ly related problems are generally de­ scribed using Axes J, I I , I V, and V ; children with learning and developmental problems are usual ly described using

CHAPTER 18

558 •

THE ASSESSMENT PROCE SS

Exhibit 1 8-2 Application of DSM-I I I-R Multiaxial Diagnosis to Child with a Developmental Disorder Case History

Multiaxial D iagnosis

Mark is a 1 0-year-old boy who was referred by Child Welfare authorities. Background information reveals that pregnancy and birth were uncomplicated . Developmental milestones were within normal limits. He has experienced the usual childhood illnesses, but has had no serious physical illness. A sister, four years older, is Mark's only sibling in his middle­ class family. He has. had playmates in the neighborhood and has participated in socialization experiences through his ex­ tended family and church. During the interview, Mark was unresponsive and moved somewhat aimlessly around the room. He made sounds but no identifiable speech. When presented with toy s , he seemed to be uninterested in them and did not engage in cooperative play. At times, he sat in his chair staring off in the distance and waved his hands in the air. However, he found it difficult to sit still, moved around the room a great deal , and had to be restrained by the interviewer. When Mark was 5 years old, his father was burned in an accident at a chemical plant. The father was hospitalized for several weeks, and eventually died from these injuries. When visiting his father, the child would often scream and pull at the bandages covering his father's burns. It was at this point that Mark's mother began to notice "a change" in him. A fter his father's deat h , M ark began exhibiting per­ severative speech . For no identifiable reason , he began having "outbursts" of crying and excitability during which he could not be comforted . At times, he would laugh or smile to himself. H is mother reported that Mark stopped playing with his toys and would become preoccupied "waving his hands" or "moving his fingers. " He began to exhibit head-banging, especially when a change was made in his routine. Currently, Mark has no language. He continues to have unexplained "temper outbursts" with self-mutilative behavior. He requires careful supervision as he lacks appropriate fear reactions and has also experienced a regression in self-help skills. A pediatric examination revealed no medical prob­ lems. Child Welfare authorities are investigating proper placement facilities as his mother reports she can no longer care for the child.

Axis I : Axis I I :

Axes I I (developmental disorders), I I I , I V, and V Exhibit 18-2 illustrates the application of a DSM-III-R multiaxial diagnosis to a child with a developmental disorder.

Comment on the DSM-III-R The multiaxial c lassification system used in the DSM-III­ R recognizes that children may have disorders that bridge

No diagnosis Autistic disorder, with childhood onset ; mental re­ tardation, unspecified Axis I I I : None Axis IV : Psychosocial stressors: father's accidental and trau­ matic death Severity : 5 - extreme Axis V: Current: 10

Discussion Axis I.

No diagnosis would be appropriate on Axis 1 . The principal diagnosis in this case is autistic disorder, with childhood onset . This child exhibits a gross and sustained impairment in many areas of functioning, especially social relationships, which began after the age of 5 years. I f delusions, hallucinations, or marked loosening of associa­ tions could be identified , then a diagnosis of schizophrenia would need to be considered . Although this child is at present untestable , an additional diagnosis of mental retardation would also be given to note the presence of significantly subaverage intellectual functioning and impairment in adaptive functioning. Axis III. A pediatric exam y ielded no relevant physical conditions. Axis IY. Even though the father's traumatic death occurred five years prior to the present assessment, this event was critical in the onset of the disorder and is mentioned on Axis IV. A coding of Extreme - 5 seems most appropriate. If the father's death were judged to be a catastrophic event, a rating of 6 could be used. Axis V Since Mark has functioned at the level demonstrated in the interview during the entire preceding year, a rating of 1 0 (persistent danger of hurting self o r others) seems most appropriate . Axis II.

Source: Reprinted, with changes in notation and in the multiaxial diagnosis, by permission of the publisher and authors from L . J . Webb, C . C . DiClemente, E . E . Johnstone, J . L. Sanders, and R . A . Perley ( Eds . ) , DSM-llI Training Guide, pp. 147- 1 48 : 1 5 5 . Copy­ right 1 98 1 . Brunner/Maze! .

behav iora l , cognItIve , phys ical , a n d developmental di­ mension s . For example , a child with autism may receive more than one diagnostic code , such as autistic disorder and mental retardation on Axis II and asthma on Axis I I I , if that condition is present. Even though DSM-II I -R categories are more behav­ iorally explicit than those in prior system s , judgment i s stil l required of t h e c l i nician . Because some of t h e diag-

559

SUMMARY

nostic c riteria requ ire that a child's symptoms be age­ inappropriate , judgment must be used to decide whether a child's behavior i s , i ndeed , deviant. The DSM-II I-R does not provide age norms , base rates . or treatment recom­ mendations for any of its classifications. The D S M -I l I-R explicitly states that standardized . indi­ vidually ad m i n i stered read i n g , arithmeti c . and intel­ I igence tests should be used as appropriate to d iagnose developmental reading d i sorder and developmental arith­ metic d isorder. I n addition, individual intel l igence test results are required for the diagnosis of mental retardation. Consequently, c l in ical and school psychologi sts are best equipped to establ ish these d iagnoses . The DSM-III-R recognizes that the i nclusion of the speci fic developmental disorders classification is contro­ versial because many children with these disabil ities have no "other" signs of psychopathology and the detection and treatment of academic s k i l l d isorders take place within the educational syste m , not within the mental health system. Nevertheless , the i nclusion of educationally related diffi­ cult ies i s deemed appropriate because "these cond i ­ t i o n s . . . conform to t h e DSM-III-R concept of mental disorder" ( p . 40) . Children receiving a DSM - I l I -R diag­ nosis for these conditions , however, may be stigmatized by a psych iatric label whe n , in fact , such a label is clearly i nappropriate . S i m i lar considerations hold for stuttering and enu res i s . Overal l , the DSM-I I I - R provides a valuable system for classi fy ing i nd i viduals who d i splay symptoms of mental disorder. Whether it w i l l lead to improved treatment or better research re mains to be see n . Some classifications in the DSM-III-R cannot be made w i thout the specific as­ s istance of psycholog ists tra i ned to administer and inter­ pret psychological and psychoeducational tests. Thus psy­ chologists and psychoeducational specialists should be thoroughly fam i l iar with the DSM-III-R.

CO M M E N T O N THE ASSESSM E N T PROCESS

Assessment of i ntel l igence and special abil ities must be carried out with sensitivity and concern for the child and h i s or her fam i l y. As you perform an assessment, consider the child's expectancies regarding the assessment ; how your dress, appearance , ethnicity, and other personal at­ tributes may affect the child (and the child's parents) ; and how your own e xpectations, biases, and cultural values w i l l affect the assessment . Be ready to modify your assess­ ment plan in l ight of new findings that emerge during the assessment .

Tests, i nterVIews" observations, and checkl ists are be­ havior sampling tools . Their use in a multidimensional assessment battery w i l l help you to (a) determine the c h i l d's strengths and weakne sse s , (b) u nderstand the nature, presence , and degree of any handicapping condi­ t ions, (c) determine the conditions that i nh ibit and support the acqui sition of appropriate s k i l l s , (d) provide basel i ne information prior to an i ntervention program, (e) develop useful instructional programs, (f) guide i ndividuals in selecti ng appropriate educational and vocational pro­ grams, (g) monitor changes i n the child, and (h) measure the impact of the i nstructional programs. The assessment results tell you what the child has done at a particular t i me and place. Know i ng why the child performed as he or she did requi res a careful study of the entire clinical h istory and assessment results . The results also do not tei l you what the child might be abie to do under a different set of testing conditions . If you want to know what the child is capable of doing under altered testing conditions, testing-of-li mi t s procedures must be employed (see Chapter 5 ) . Murphy's (1975) insightful suggestions are a fitting con­ clusion to this chapter on the assessment process. They alert us to the need to d i scover each child's unique coping strategies : . . we need to focus on and better understand the nature of ongoing current coping struggles; how to support them, how to help the child to extract the strength and insight that successive ex periences may make available to him [or her] . We need to understand the positive strategic values of withdrawal in certain situations, and be very cautious about talking about a "withdrawn child . " Similarly we need to respect and value children's protests, resistances, altempts to change or control situations, and all the other active coping efforts that can give us cues to what the child finds intolerable, unsuitable, boring, distasteful or threatening to his integrity . . I am pleading . . . that each cl inician, each teacher, use all of the available resources along with his [or her] own fresh look at the child in his [or her] situation in order to discover the meaning of the child's behavior from the child's own point of view. (p. 42)

SUMM ARY I . The multimethod assessment approach synthesizes infor­ mation from the clinical history, interviews, observations, stan­ dardized tests, informal tests, and other special procedures. 2. Psychological testing involves the administration and in­ terpretation of psychological tests , whereas psychological as-

560 sessment considers the child's unique characteristics. current life situation, and cl inical history as well as the results of testing . 3 . Various purposes are served by assessment, including screening , diagnosi s , counseling and rehabil itation, and progress evaluation. 4 . A psychological evaluation can complement a medical evaluation by providing guidelines that can help the child func­ tion better in everyday l i fe situations. 5. The steps in the assessment process include rev iewing referral information; obtaining information from parents, teach­ ers, and others familiar with the child's problems, as wel l as from other agencies and physicians; assessing the behavior of relevant adults; observing the child in various settings; selecting and administering a test battery; interpreting data; formulating hy­ potheses; developing intervention strategies; writing a report; meeting with parents and child; and following up on recommen­ dations and retesting . 6. The test score is influenced by innate, background, and environmental factors as wel l as by factors in the child's person­ al ity, situational test demands, and random variation. 7. In attempting to account for a child's failures, consider all possible factors, including those associated with neuropsycho­ logica l , physical, experiential , and temperament domains . 8 . Test results give information about how a child performed at a specific time under specific conditions . Generalization of findings to other situations may be difficult. Be especially aware of nuances in the child's performance that may suggest special talents. 9 . It is difficult to specify the exact function that underlies performance on any given test and the relationship between the skills measured on tests and classroom performance. 1 0 . The di fficulty of assessing children is compounded by the fact that (a) classification systems are not uniform , (b) behavioral disorders are not always clear, (c) children undergo rapid devel­ opmental changes, and (d) assessment is time consuming . 1 1 . The diagnostic examples presented in the chapter illus­ trate how hypotheses can be developed from an analysis of the child's responses, performance, and behavior. 1 2 . The assessment of the relationship to learning ability of various factors - such as type of materia l , rate and modality of presentation, cues, and reinforcements - can contribute to the design of intervention efforts. 1 3 . Assessment and intervention are intertwined, with inter­ vention results providing feedback on the assessment hypotheses . Re-evaluation and fol l ow-up are important parts of the assess­ ment process . 1 4 . The results of repeated evaluations are difficult to interpret because there is no easy way to separate practice effects from actual clinical improvement. Additionally, differential practice effects may occur as a function of the specific population, task, retest interva l , initial test score , and specific test.

1 5 . Keys to designing an effective clinical and psychoeduca­ tional assessment include (a) using a broad spectrum approach , ( b ) using a variety o f assessment procedures, (c) using multiple

CHAPTER 1 8

THE ASSESSMENT PROCESS

methods to evaluate a particular domain (if possible) , and (d) assessing the child on multiple occasion s . 1 6 . T h e assessment process in schools incorporates the gen­ eral principles of assessment, but focuses more d i rectly on learn­ ing and educational intervention. The evaluation should focus on the child's strengths and compensatory abilities as well as on areas of deficiency. 1 7 . A l though visits to the classroom may provide valuable information, it is necessary that the teacher approve and support such visits. 1 8 . Assessment should be a continuous activity and interven­ tions should be continually monitore d . 19. T h e Panel on the Selection a n d Placement of Students i n Programs for t h e Mentally Retarded of t h e National Academy o f Sciences made the fol l owing recommendations for t h e assess­ ment of referred children for special education programs: (a) the child's learning environment and the nature and the quality of the regular instruction received should be evaluated systematically and (b) a comprehensive battery that includes measures of intel­ ligence, cognition, adaptive behavior, and screening procedures to detect biomedical disorders should be administered . 20. A language assessment should include evaluation of the child's consistencies within his or her own language system, strategies and processes used to produce language, language structures, and language distortions. Samples of verbal and non­ verbal behavior in a variety of settings should be obtained. Both comprehension and production skills should be evaluated . Lan­ guage difficulties may be associated with syndromes or diseases, processing difficulties, or social and cultural factors. 2 1 . Some speech and language deviations are associated with developmental period s , whereas others may occur throughout the developmental period. Lack of speech, unintelligibl e speech, and failure to speak in sentences may be fi rst noticed at about 3 to 5 years of age ; substitution of easy sounds for more difficult ones , dropping of word endings, faulty sentence structure, noticeable nonfluency, and abnormal rhythm, rate, and speech inflection are of concern at about 5 to 6 years; and d istortions, omissions, and substitutions of any sounds are of concern at about 7 years. Late­ appearing speech sounds , being embarrassed by speech, poor voice quality, and telegraphic speech are examples of deviations that are not specifically age related . 2 2 . Visual-motor performance can be observed not only on visual-motor tests, but also on intelligence and special ability test s . 2 3 . In t h e learning potential assessment strategy, t h e child is first tested , then given specialized training, then tested again. Learning potential assessment attempts to measure the process of learning rather than the content of learning , including why and how the material was mastered . The word potential may be misleading because there is l ittle evidence that learning potential assessment measures are better at predicting school-related achievement than are intelligence tests. Learning potential pro­ cedures are still in an experimental stage and should not be substituted for intell igence tests in any decision-making activity.

56 1

STUDY QUESTIONS 24 . Although classification and labeling have potentia l nega­ tive consequences, they serve many useful functions, such as aiding in !he study of exceptionality, facilitating communication and record keeping, enhancing others' altruistic behavior and tolerance for deviant behavior, and providing a sense of closure for parents. 2 5 . The Diagnostic and Statistical Manual of Mental Disor­ ders , 3 rd edition - Revised (DSM-JII - R) is the standard system for diagnostic classification of mental disorders . It has a five-level multiaxial classification system: 1 - clinical syndromes, I I - de­ velopmental disorders and personality disorders, I I I - physical disorders and conditions, IV - severity of psychosocial stressors, and V - global assessment of functioning. Clinical and school psychologists are the individuals best equipped to establ ish the diagnoses of mental retardation and specific developmental disorders. 26. Clinical and psychoeducational assessment must be car­ ried out with concern for the child and his or her family and sensitivity to your role as an examiner in the assessment process. The key to effective assessment is to portray each child's unique ability pattern.

KEY T E R MS, CONCE PTS, AND NAMES Multimethod assessment (p. 532) Psychological assessment ( p . 532) Psychological testing (p. 532) Cumulative achievement ( p . 536) Psychological processes ( p . 536) Repeated assessments ( p . 537) Practice effects (p. 5 3 7 ) Regression-to-the-mean effects (p. 5 37) National Academy of Sciences' Panel on the Selection and Placement of Students in Programs for the Mentally Retarded (p. 543) Language assessment (p. 544) Receptive language ( p . 546) Expressive language (p. 546) Speech (p. 546) Phonation ( p . 546) Nasal resonance (p. 546) Articulation (p. 546) Prosody ( p . 546) Dysfluencies ( p . 546) Congenital laryngeal web (p. 546) Agrammatism (p. 547) Anomia (p. 547) Blocking (p. 547) Circumstantiality (p. 547) C l anging (p. 547) Derailment (p. 547) D i stortion of language sounds (p. 547)

Distractible speech (p. 547) Dysarthria (p. 547) Dysprosody (p. 547) Echolalia (p. 547) I llogicality (p. 547) Incoherence (p. 547) Loss of goal (p. 547) Neologisms (p. 547) Paragrammatism (p. 547) Paraphasia (p. 547) Perseveration (p. 547) Phonemic paraphasia (p. 547) Poverty of amount of speech (p. 548) Poverty of content of speech ( p . 548) Pressure of speech (p. 548) Self-reference (p. 548) Stilted speech ( p . 548) Syntactic errors (p. 548) Tangentiality (p. 548) Telegraphic speech (p. 548) Topical or referential identification problems (p. 548) Word approximations (p. 548) Learning potential assessment (p. 549) M ini-learning situations (p. 549) Classification and labeling ( p . 551) Self-fulfi lling prophecy ( p . 551) DSM-III-R ( p . 552)

STUDY Q U EST I O N S I . Desc ribe and eval uate the multimcthod assessment approach . 2 . How does psychological assessment di ffer from psychological testing? 3. Discuss the purposes of assessment. 4 . Describe the steps i n the assessment process. 5. What are some of the difficulties associated with the assessment of children? 6. Discuss a model that is useful for understanding factors that may influence test scores. 7. What factors should be considered in accounting for failures on a test or assessment procedure? 8. What are some problems in generalizing from a child's test performance to other situations? 9. Discuss the problems involved in the measurement of psychological processes. 1 0 . What difficulties are involved in evaluating the results of repeated assessments? I I . Present at least seven i l lustrations of how hypotheses may be formulated from a child's responses and performance. 1 2 . Discuss the role of assessment of learning abil ity in devel­ oping intervention strategies.

562

1 3 . Describe some guidel ines for designing a clinical and psychoeducational assessment . 1 4 . Discuss the assessment process in school s. 1 5 . Discuss the assessment model for school-based referrals recommended by the Panel on the Selection and Placement of Students in Programs for the Mentally Retarded. 1 6 . What are some developmental considerations in evaluat­ ing language and speech deviations?

CHAPTER 18

THE ASSESSMENT PROCESS

1 7 . Discuss some important guideposts for observing the child's language and speech . 1 8 . Discuss learning potential assessment. 1 9 . Discuss the pros and cons of classification and label ing . 20. Discuss the implications of the DSM-III-R for the practice of clinical and school psychology.

-

19 ASSESSM E NT O F ETH N I C M I N O RITY C H I LDREN

Introduction

IQ and achievemem tests are nOlhing bw updated versions of the old signs down South that read "For Whites Only. "

Arguments Against the Use of I ntel li gence Tests i n Assessing Ethnic Minority Children

-Robert L. Williams

Arguments for the Use of I ntelligence Tests in Assessing Ethnic Minority Children

Moreol'er, it is precisely the black studems who need IQ tests most of all, for it is precisely \\lith black students that alternative methods of spo[{ing intellectual ability have failed.

General Considerations in Ethnic Minority Testing Subcultural Considerations

-Thomas Sowell

Recommendations for Assessing Ethnic M i nority Ch ildren

Shared standards are like mortar, holding the muhicolored mosaics of civilization together. Good tests, like good laws, embody shared standards.

Comment on Assessment of Ethnic M i nority C h ildre n Psychological Evaluation

-Barbara Lerner

Summary

563

564 •

CHAPTER 1 9

ASSESSMENT OF ETH NI C MINORITY CHILDREN

Exhibit 1 9- 1 How to Establish Rapport with an Inner-City Child: The Education of a Psychologist

Some years ago, a birthday party for a member of the staff at a well-known psychological clinic played a novel role in the test performance of a black child. Prior to the party this boy, whom we shall call lames . had been described on the psycho­ logical record as "sullen, surly. slow, unresponsive . apathetic , unimaginative, lacking in i nner l i fe . " This description was based on his behavior in the cl inic interviews and on his performance on a number of psychological measures. includ­ ing an intelligence test and a personality test. His was not an unusual record; many [m inority] children are similarly portrayed . On the day of the birthday party, lames was seated in an adjoining room waiting to go into the cl inic ian's office . It was j ust after the lunch hour and lames had the first afternoon appointment . At the conclusion of the lunch break on this particular day, the staff presented a surprise birthday cake to one of the cl inicians, who was black. The beautifully deco­ rated cake was brought in and handed to the cl inician by lames's cl inician, who was white, as were all the other mem­ bers of the staff. The black cl inician was deeply moved by the cake - and the entire surprise . In a moment of great feeling, she warmly embraced the giver of the cake . James inadver­ tently perceived all this from his vantage point in the outer office. That afternoon he showed amazing alacrity in taking the tests and responding i n the interview. He was no longer sullen and dul l . On the contrary, he seemed alive and enthusi­ astic, and he answered questions readily. His psychologist was astonished at the change and in the course of (lie nex! few weeks repeated with lames the tests on which he had done so poorly. lames now showed marked improvement . and t he psychologist quickly revised 110t only lames's test appraisal on his cl inical record card , but her general personal ity descrip­ tion of him as wel l . The high point o f this new, positive relationship came some months later when lames confided to the psychologist that she had gotten off on the wrong foot with him on the first day in the first three minutes of contact. The psychologist was taken aback and said , "What do you mean? I was very friendl y ; I told you my name and asked you yours." He responded . "Yeh ,

and I said lames Watson and right away you called me l immy and you bin cal l i n' me l immy ever since. My name is James , 'cept to my very good friends maybe . Even my mother calls me lames . " Then he went on to tell her how he had changed his opinion of her on the day of the bi rthday party because of the close relationship he had seen between her and the black psychologist . This little story il lustrates a number of things. First, it shows that the test is a social situation . The testing situatio n , whether it be a psychological test or a ny other kind of test for that matter, reflects a relationship between people , a rela­ tionship that is often remarkably subtl e . And when anything hampers this relationship, the result is l i kely to show i n the test score itself. This can occur on an i ndividual test as wel l as a g roup test , an IQ test as well as a personality test , a subject­ matter examination as wel l as a psychological measure. The story also shows how the behavior evidenced in the clinical situation tends to be seen by the psychologist as indicative of the basic personality of the child. This is fre­ quently done with l ittle awareness of how much this behavior is a product of the particular relationship between the psychol­ ogist and the child , and of the testing situation as suc h . Children from different cultural backgrounds respond very differently to c l inical situations and to the idea of being tested or evaluated . The anecdote also points up the fact that a well-mean i ng . clinically trained , unprejudiced psychologist can have poor rapport with an inner-city child , not because of deficient psychological technique, but because of l imited knowledge about certain cultural attitudes. In this case, the attitude i n question is the feel ing held by many black people that the informality intended by nicknames signifies a lack of respect when it takes place across cultural lines. This does not suggest that the child himself was aware of this reasoni ng , but rather than he was simply reflecting his mother's wish that he be cal led by his full name . Source: Reprinted, with a change in notat ion , by permission of the publisher and author from F R iessman , The Inller-Ciry Child, pp. 53-55 . Copyright 1 97 6 , Harper & Row

I N TRODUCTI O N

T h e fi rst two sections of this chapter cover arguments for and against the testing of ethnic m inority children. General considerations in ethnic m inority testing are then dis­ cussed , including the effects of poverty, cultural diversity, response style, bilingual i s m , racial differences in intel-

l igence, and culture-fair tests. The chapter then focuses on cultural values and styles of blac k , H ispanic-American , American I ndian, and A si an-American ethnic groups. Fi­ nally, general recommendations for testing ethnic minority children are presented .

565

INTRODUCTION

Use of Appropriate Designations The label ethnic minority children is used to desi gnate children who belong to a recognized ethnic group and whose values, customs , patterns of thought . and/or lan­ guage are s ignificantly different from those of the majority of the society in which they l i ve . The groups from which ethnic minority children come incl ude blac k s . Mex ican­ Americans , American Indian , Puerto Ricans . and Asian­ American s . The use of such labe l s as "cu lturally handi ­ capped:' "cul tu ra l l y d isadvantaged , " and "cultural ly de­ prived" to designate ethnic m inority chi ldren has been u n fortunat e , because these terms have val ue impl ications. No one has the right to degrade a subculture because it does not conform to the patterns of the majority group. Different cu ltural mores and t rad itions expressed by mi­ nority groups may be both healthy and adaptive, for the l i festyles of these groups may differ markedly from those of the majority culture . The extent to which a group is handicapped may depend on the eyes of the beholder. This chapter therefore u ses the term ethnic minority childrel/ i nstead of alternatives that may have pejorative conno­ tations.

Adopting a Multifactor, Pluralistic Approach In order to understand cu ltural groups in our society, it is i m portant to adopt a multifactor, plural istic approach . C u ltural groups may vary with respect to cul tural values (stemming in part from cul t u ral shoc k . discontinuity, or conflict) ; language and nuances in language sty le: views of l i fe and deat h ; roles of fam i l y members ; problem-solving strategies ; attitudes toward educat ion , mental heal th, and mental il lnes : and stage of acculturation (the group may follow trad itional values, accept the dom inant group'S val­ ues , or be at some point between the two ) . You shou ld adopt a frame of reference that will enable you to under­ stand how particular behaviors make sense w ithin each cul tu re . "Racist thinking is revealed not in recognizing d i fferences as suc h . but in al igning them along an inferior­ s u perior axis and ascribing them to unchangeable deter­ m i nants" ( Korch i n , 1980. pp . 2 64-265 ) .

am iner. and the reasons for the assessment . Ethnic minor­ ities have developed coping patterns that are a response to 2n environment that has been less than hospitable . It is important to understand how these coping patterns faci l i ­ tate o r hinder children's adj ustments t o their own sub­ culture . as wel l as to the cul ture at large .

Guarding Against Inappropriate Generalizations Generalizations about a part icular m inority group may not apply to individual members of that group . Each indi­ vidual examinee must be evaluated on the basis of his or her own dynamic , family, and subculture . There is no one monol ith ic blac k , Hispanic-America n , American Indian , or Asian-American subcul t u re , so you should not assume that every minority person expe riences his or her ethnicity in the same manner. Research on lower-class members of an ethnic group may not general ize to other socioeconomic c lasses in that grou p .

Are

1ental Health Services Needed?

Ethnic mi norities in the United States face overwhel m ing problems associated with j obs , phys ical health, and hous­ ing. as well as general oppression and rac ism. The ir men­ tal health is l i kely to be directly related to these problems . One may ask whether it is even appropriate to look at mental health problems and related assessment questions when these othe r areas need attention . This is a valid q llc�t i o n . and i t is not easy to answer. But I bel ieve that , in spite of the inj ustices and rac ism rampant in our society. mental health professionals and psychoeducational spe­ cial ists can improve the lot of ethnic m i norities by helping them with their psycholog ical and educational concerns . Like white chi ldren, ethnic m i nority children have mental health problems and learn ing di fficu l t ies and experience t rauma associated brain damage . They may perform poorly in school ; have learn i ng problems ; need vocational gu idance : become depressed . suicidal , anxious . fearful . psychot ic. or alcohol i c : or experience family discord and self-doubts .

Assessment Issues as Part of the Fabric of Societ)1 Appreciating Cu ltural Values and Coping Patterns It is important to develop an appreciation for each ethnic m i nority'S culture , because one or more of the prev iously named factors may affect the assessment situation. As the inc ident described in Exhibit 19-1 show s . these factors are powerful force that shape a child's attitudes toward the assessment situation , including the test materials. the ex-

The issues involved in the assessment of ethnic m i nority chi ldren are complex, because they are woven into the very fabric of society. Assessment results have an impact on children's sel f-esteem and i n fl uence their chances of hav­ ing a successful l i fe . If tests are detrimental to ethnic mi nority chi ldre n , then they must be changed or e l i m i ­ nated . If tests are beneficial t o ethnic minority childre n ,

566

CHAPTER

however, then their elimination may be a disserv ice to countless childre n . Whenever tests are used, you must ensure that the results are employed for the good of the child . The purpose of this chapter is to demonstrate how this goal can be accomplished.

ARG U M E NTS AGAI NST T H E U S E OF I N T E LL I G E N C E TESTS IN ASSESSI N G E T H N I C M I N O RI TY C H I LD R E N

Many al legations have been made about t h e i nappropriate­ ness of using tests, and in particular intell igence tests, with ethnic minority children. The major arguments are l i sted below and then are considered in more detail in the fol low­ ing subsections. •

Inrelligence tests have a cultural bias. Standard i ntel­

l igence tests have a strong white , Anglo-Saxon , m iddle­ c lass bias . National norms are inappropriate for minorities . Na­ t ional norms based primarily on white, m iddle-class , Anglo-Saxon samples are i nappropriate for use with ethnic m i nority chi ldre n . Minorities are handicapped i n test-taking skills . Eth­ nic m i nority children are handicapped in takjng tests be­ cause of (a) deficienc ies in motivation , test practice, and reading: (b) failure to appreciate the achievement aspects of the test situation ; and (c) l imited exposure to the culture . •

•

Thefact that most examiners are white has the effect of depressing the scores of ethnic minority children. Rapport •

and communication problems exist between wh ite exam­ iners and ethnic m inority c h i ldren. These problems inter­ fere with the abil ity of ethnic minority chi ldren to respond to the test items . • Tests results lead to inadequate and inferior educa­ tion . Test results are the main reason why ethnic minority

children are segregated into special classes. These classes have i nadequate curric u l u m and provide inferior educa­ tion . Test results also create negat ive expectanc ies in teachers.

Cultural Bias Argument There are many definitions of test bias, each of which has some value in explaining the properties of tests and their uses . I n exami n i ng test bias , you shou ld first consider the meaning of a test score . Is a test score an indication of past ach ievement or of aptitude for future achievement,) Many arguments concerning test bias rest on which of these interpretations is considered correct. It is an important

19

ASSESSMENT OF ETHNIC MINORITY CHILDREN

distinction and one that is often made between achievement tests and aptitude tests . This distinction has never been c learly resolved in the case of intelligence tests . This textbook advocates that i ntel l igence tests and other special abil ity tests be used as measures of achievement, not as pure measures of aptitude or capacity. Inte l l i gence test scores, representing the interplay of biolog ical factors and environmental factors, reflect past l earning s . Low scores obtained by ethnic m inority childre n indicate a need to i mprove educational systems, not to abandon the tests .

Measures of test bias.

The various c riteria used to mea­ sure test bias are reviewed in the fol lowing paragraphs (obtained from Flaugher, 1978), as are findings of research studies based on these defi n it ions of test bias .

Mean differences. A test is considered by some to be biased when it y ields lower scores for one group than for another. This definition is not acceptable, howeve r, be­ cause mean d i fferences are not a legitimate standard for identifying test bias . Because of disparities among various groups in our nation with respect to socioeconomic status and other variables, it would be surpris i ng if intelligence and achievement tests did not show mean d i fferences in favor of some groups . When disparities in socioeconomic status between black and white groups are reduced, IQs become more s i m i lar. For example , the IQs of 7 -year-old black children were less than 5 points below those of white 7-year-olds when both groups came from the same socioeconomic level , l ived i n the same c ity, and had mothers who had gone to the same hospital for prenatal care ( N ichols & Anderson , 197 3 ) . This 5 -poi nt d i fference between black and wh ite c h i l d ren contrasts w ith the oft-reported 15-point difference ( i n favor of white childre n ) . Single-group o r differential validity. Another mea­ sure of val idity is whether a test is an equally good predic­ tor for two (or more ) ethnic groups w ithin the United States . Two related ways of comparing validity coefficients are used to determine whether this form of test bias is present . W ith the single-group val idity approach . test bias is considered to be present when a validity coefficient is significantly d i fferent from 0 for one ethnic group but not for another. With the d i ffe rential validity approac h . test bias is cons idered to be present when there is a significant d i fference beMeen two val idity coefficients. The majority of research w ith ethnic m inorities in our country ind icates that these forms of test bias are not commonly present. That i s , neither s ingle-group validity bias nor di fl'erential validity bias is general l y found . Although there are in-

567

ARGUMENTS AGAINST THE USE OF INTELLIGEN CE TES TS I N ASSESSIN G ETHNI C MINORITY CHILDREN

stances in which single-group or differential val .dity has been found, "the fact that they are so el usive . difficult to detect , and debatable is good evidence that they are not very poten t phenomena rel a t i ve to a l l other pos sible sources of problems in the interaction of minOrIties a n d testing" ( Flaugher, 1978, p . 674) . Studies w ith the W ISC-R , Stanford-Binet : Form L-M , PPVT, and Raven's Progressive Matrices indicate that the regression l i nes for black children , white children . and H i spanic-American c h i 1.dren a re s i m i l a r when W ide Range Achievement Test scores and other achievement i ndices are used as criteria (Bossard , Reynolds , & Gutkin , 1980; H al l , Huppenz, & Lev i , 197 7 ; Reschly & Sabers , 1979 ; Reynolds & Hartlage , 1979) . I n addition . studies have shown that the concurrent validity of the WISC-R, WISC, and Stanford-Binet : Form L-M with such c riteria as the Cali fornia Achievement Test, W ide Range Achieve­ ment Test, Metropolitan Achievement Tests , teacher rat­ ings, Stanford Achievement Test, and California Test of Mental Maturity, is excellent for blac k , white, Mex ican­ American , and American I ndian childre n , with median validity coe ffi c ie nts in the . 50s l see Table 19-1 ) . These findings support the conclusion that i nte l l i gence tests gen­ eral ly are equally good predictors for black, H ispanic­ American , and white childre n .

Differen tial construct validity. Another way t o evalu­ ate the possible bias of i ntell igence tests is to study the extent to which they measure s i m ilar abi l ities in various ethnic groups. Studies examining the factor struct ure or the WISC , WISC-R, W PPSI , and McCarthy Scales of C hildren's Abil ities for blac k , H ispan ic-American. and white children ind icate that these groups have comparable factor structu res (Dean , 1980; Greenberg , Stewart , & Hansche, 1986; Gutkin & Reyno l d s , 19 8 1 ; Guy. 197 7 ; Jensen & Reynold s , 1 9 8 2 ; Johnston & Bolen . 1984 ; Kau f­ man & DiCuio, 1975 ; Kaufman & Hollenbeck , 1974 ; Law­ l i s , Stedman, & Cortner, 1980 ; Miele, 1979; Reschly, 197 8 ; S e m l e r & I s c o e , 1 9 6 6 ; S i l ve r s t e i n , 1 9 7 3 ; Va nce , Huelsman , & W herry, 1976; Vance & Wal lbrow n , 1978 ) . These fi nd i ngs suggest that ( a ) intell igence tests measure the same abil ities in white , black , and H ispanic-American c h il dren and (b) the Verbal-Performance disti nction in some of these tests is appropriate for the three ethnic groups . The mean WISC-R subtest scores obtained by black and w h ite children in the standardizat ion sample are shown in Table 19-2 . These scores indicate that average scores are similar on both the Verbal and Performance sections of the scale for black chi ldre n (M Verbal scaled score = 8 . 25 vs. M Performance scaled score = 8 . 18) and for wh ite chil-

Table 1 9- 1 Concurrent Validity-Median Correlations Between I ntelligence Tests and Achievement Tests for Anglo­ American, Black-American, and H ispanic-American Chi ldren Achievemel1l area Reading

A rithmetic Number

Number of

Mdn

of

Mdn

Ethnic group

studies

r

studies

r

Anglo-American Black-American H ispanic-American

14 17 7

. 52 . 60 .51

12 11 7

. 54 .58 .53

Nore. These median correlations are based o n the following studies: Bossard. Reynolds, & Gutki n . 1 980; Dean , 1 977a, 1 979b: Henderson, Butler, & Goffeney, 1 969; Henderson. Fay, Lindman n . & Clarkson, 1 973: Kennedy. Van de Riet , & White, 1 963; Komm , 1 97 8 ; Oakland, 1 980. 1 983a: Oakland & Feigenbaum , 1 979: Reschly & Reschly, 1 979; Reynolds & Gutkin. 1 980a: Reynolds & Nigl, 1 98 1 ; Sewel l . 1 979; Sewell & Severson. 1 974, 1 975 ; Svanum & Bringl e , 1 98 2 : Weaver, 1 968; Weiner & Kau fman , 1 979.

dren (M Verbal scaled score = 10 . 3 3 vs. M Performance scaled score = 10 . 3 7 ) .

Content bias. Evaluation o f content bias focuses o n whether the content of particular test items i s unfair to some groups of the population . Inspection of standardized Table 1 9-2 Mean WISC-R Subtest Scores for Black and White Children in the Standardization Sample Croup While SlIb,est

I n formation Similarities Arithmetic Vocabulary Comprehension D igit Span Picture Completion Picture Arrangement Block Design Object Assembly Coding Mazes

(N

=

1868)

104 10.3 104 104 104 10. 1 104 104 104 104 10.2 104

SOI/rce: Adapted from Gutkin a n d Reynolds ( 1 98 1 ) .

Black (N

=

305)

8. 1 7.9 8.6 7.9 7.8 9.2 8. 1

8. 1 7.7 7.9 8.9 84

568

intelligence tests reveals few, i f any, items that appear to be biased systematicall y in favor of one group over another. Agreement among the members of a panel of experts as to which items are biased tends to be very low. A useful empirical approach for i nvestigating content bias is to examine item performance statistics group by group to determine the difficulty level of each ite m . "If a particular item i s extraordinarily difficult for m i nority-group mem­ bers relative to the d i fficulty of other items in the same test, then that item is a good candidate for suspic ion of this kind of bias" ( Fl augher, 197 8 , p. 675) . Studies c ited by Flaugher suggest that the elimination of such items from standard­ ized tests makes l ittle if any difference in test scores. Several d i fferent types of internal criteria are available for studying content bias. They include examining differ­ ences between groups i n (a) the rank order of the percent passing each item, (b) the percent passing adjacent items i n the test , ( c ) t h e number of persons passi ng each item when both groups are equated for total score , and (d) the types of item content that d iscriminate most and least between the two groups (Jensen, 1974a) . Applications of these criteria to the intelligence test and ability test performances of black and w hite children and adults have produced no evidence that differences between blacks and whites are related to cultural bias (Jensen , 1974a ; Meyer & Goldstei n , 197 1 ; Miele, 1979; N ichol s , 197 1 ; Olivier & Barclay, 1967 ; Sandoval , 1979; Sandoval , Zimmerman , & Woo-Sam, 1983) . Tests in these investigations included the Stanford­ Binet : Form L-M , W I SC-R, PPYT, and Raven's Progres­ s i ve Matrice s . M iele ( 1979) reported that the WISC Comprehension item 4 - "What should you do if a child smaller than you begins to fight with you?" - which was s i ngled out by Robert W i l liams in the CBS documentary "The IQ Myth" as a blatant manifestation of cultural bias, proved to be easier for black childre n than for white child re n . It was the 42nd easiest item for I I I black children (M age = 6-1) and the 47th easiest item for 163 white children (M age = 6-2 ) . Sandoval and Miille (1980) reported that neither black, H i spanic-American , nor Anglo-American j udges were able to determine accurately which WISC-R items were more difficult for minority stude nts . The ethnic back­ ground of the j udges made no difference in accuracy of item selection . Individual inte l l igence test items have been evaluated by federal judges for biased content (see C hapter 24) . Seven WISC and W ISC-R items j udged to be culturally biased by Judge Grady in the PASE case (Parents in Action on Special Education v. Joseph P Hannon) were empirically evalu­ ated w ith a sample of 180 black childre n and 1 80 white c h i ldren attending school s in Chicago. The average age of

CHAPTER 19

ASSESSMENT OF ETHNIC M I N ORITY CHILDREN

the children was 1 1 1j2 years . As Table 19-3 shows, although the largest differences between the two groups were a 9 percent difference on the rubies item and a 4 percent difference on the stomach item, both in favor of white childre n , none of the differences between the two ethnic groups was statistically significant . The notorious fight item disc ussed above was passed by 7 3 percent of the black children and by 71 percent of the white childre n . These results conclusively demonstrate that an "armchair" in­ spection of test items cannot reveal which items are more d i fficult for one ethnic g roup than for another.

Factors affecting the validity of test results. Measures of val idity can provide statistical evidence that a test does not have an i n herent bias against any ethnic m inority group. Whether use of a particular test in a particular situation results in discrimination, however, will depend on such factors as the purposes to which the results are put, how the results are i nterpreted , and how the test is administered . Uniqueness of the black experience. One of the main thrusts of the culture-bias argument has been that intel­ ligence tests are not relevant to the experiences of ethnic minority childre n . Black children, for example , are said to develop u nique verbal skills that are neither measured by conventional tests nor accepted b y the middle-class­ oriented classroom ( W i l liams , 1970) . There has been lit­ tle, i f any, research to support this contention . Further­ more , items on intell igence tests represent important as­ pects of competence in the common culture ; the items are Table 1 9-3 Percent of Black and White Children Passing Seven Items on the W I SC Comprehension Subtest Said by J udge Grady to Be Culturally Biased Group Item

(N = I SO)

White (N = ISO)

Rubies Stomach C . O. D . Loaf of bread Fight Pay bills by check Give money to charity

26% 7 I 74 73 21 12

35 % II 0 67 71 22 12

Black

NOle. Mean age o f black children was 1 1 . 30; mean age o f white child-en was 1 1 . 2 7 years. Source: Adapted from Koh, Abbatiello, and M cLoughlin ( 1 984).

ARGUMENTS AGAINST THE USE OF INTELLIGENCE TESTS IN ASSESSING ETHNIC f"IINORITY CHILDREN

not reflective of purely middle-class values. For a demo­ cratic society to endure , these common cultural forms and practices need to be maintained and extended to the culture as a whole "cultural apartheid ought not to be encour­ aged in this society" (Ebe l , 197 5 , p . 86, it alic s a dded ) . Ebel ( 1 97 5 , p . 8 7 ) also noted :

569

be given extra poi nts? All selection procedures are tied to ethical and oc ial values . Universal acceptance of one set of val ues will be d ifficult to ach ieve .

-

The bias which accounts for poor test performance by some minority persons is not in the tests so much as it is in the culture , and thus is another problem altogether. So long a s t h e tests under scrutiny truly measure the skills necessary to success in the prevail ing culture , minority interests are not well served by blaming "test bias" for poor performance . The tests we use in education ought to be as free of bias as we can make the m . But the extent and seriousness of bias in our current educational tests can be , and probably has been, exagge­ rated. The "well-known" bias of tests against minority group members seems to be more fanciful than factual .

The argument that i ntelligence tests are not valid be­ cause ethnic minorities have not had the same experiences as white middle-class children becomes difficult to accept when we consider the fact that a population quite far removed from w hite m iddle-class America actually does better on nonverbal tests than do American children them­ selves. The mean scores of children in Japan on many of the Wechsler Performance Scale subtests are higher than those of the American standardization samples ( Lynn, 1977 ) . Lynn bel ieves that these fi ndings i nd icate that tests such as the Wechsler Performance Scale may be much fai rer cultu rally than many critics have been willing to admit . Humphreys (1973 , p. 3 ) also believes that the cultural d i fferences between blacks and whites may not be as great as some propose : There is every reason to accept a single biological species for blacks and whites and a high degree of cultural similarity as wel l . W hile there are obvious environmental differences , these differ­ ences are not as profound as to require different principles in the explanation of black and white behavior. The two groups use a h ighly similar (if not identical) language, attend similar schools , are exposed t o similar curricula, l i sten to the same radio pro­ g rams, look at the same commodities, etc . Cultural differences a re a question of degree, not of kind.

Selection model. The seleCTion bias of a test refers to the e xtent to which the test has a differential effect on the n umber of exam i nees from various groups who enter cer­ tain programs ( such as spec ial classes, college , or voca­ tional training programs) or are selected for certain jobs . There is l ittle agreement about the best statistical pro­ cedure to use to reduce selection bias. Should the same cutoff scores be u sed for all groups, or should some groups

Validity criteria. A test can only be valid w ith respect to particular crite ria. Criteria vary in such characteristics as importance, rel iabil ity, and innovativeness . If the w rong criterion is used to establish validity, the test scores may in fact be biased . For example, a test composed only of verbal items may be valid for selecting people good at speaking or writing about music, but it may not be val id for selecting prom ising musicians who are talented in produc­ ing music . Although the criterion problem is a d i fficult one, attempts should be made to examine thoroughly the criteria Llsed to val idate tests. Atmosphere. If examinees feel out of place or un­ welcome when taking a test, they w i l l not give their best performance . If the examinees' real capacities are inh ibited when they are confronted by a test , then the test scores are biased. This type of bias appears to play only a l im ited rol e , if any, in most testing situations , however, because examine rs take much care to obtai n the child's best perfor­ mance i n the individual testing situation . When test u sers generalize from l imited domain of me:1surement to a broad range of abil ity, the i ssue of bias in the results is legitimately raised . For exampl e , it is a great leap to say that a child "lacks practical judgment" simply because he or she was unable to answer correctly a few problems on a test. :1

Overinterpretation.

Inappropriateness of National Norms The argument that national norms are i nappropriate for ethnic m inority children has led some writers ( for exam­ ple, Mercer, 1976) to advocate estab l i shing pluralistic norms. Those who favor pluralistic norms bel ieve that it is useful to know how a child's performance compares to that of others in his or her own ethnic g roup. Plurali stic norms are potentially dangerous , however, because they (a) pro­ vide a basis for invidious comparisons among different ethnic g roups , (b) may lower the expectations of ethnic minority children and reduce thei r level of aspiration to succeed, (c) may have l ittle relevance outside of the child's specific geographic area, and (d) furnish no i nformation about the complex reasons why some ethnic groups tend to score lower than others on intell igence tests (DeAvila & Havassy, 1974 ) . The renorming of tests to devise plural istic norms is i nappropriate because it does not i nvolve test modifications, nor does it take i nto account whether the

570

test should be used with ethnic minority children (Bernal , 1972 ) . And the use of pluralistic norms gives rise to new questions - what norms should be used for a child who has a Mexican father and a Hungarian mother? Some charge that norms for the major i ndividual i ntel­ l igence tests are based entirely on the performance of middle-class whites . This simply i s not t rue . The Stanford­ Bi net : Form L-M and Fourth Edition , WISC- R , W P PS I , WAIS-R, and McCarthy Scales have a l l used excel l e nt sampl i ng procedures to obtain their strat i fied samples for standardizat ion . Ethnic minority children are represented in each of the norm groups in proportion to the i r represen­ tation in the general population . National norms reflect the performance of the popula­ tion as a whole . Because they describe the typical perfor­ mance of our nation'S children, they are important as a frame of reference and as a guidepost for decision-making . This is not to say that other norms should not be used . Test results may be interpreted from several frames of refer­ ence . But it is important that test users and consumers of test information clearly recognize which norms are being used and why they were selected .

Deficiencies in Test -Taking Skills Ethnic minority children may be deficient in the abil ity to employ test-taking skj l l s , choose proper problem-solving strategies, and balance speed and power. They also may exhibit test anxiety. Western culture emphasizes achieve­ ment and problem sol ving, and by the time children begin schoo l , they are usual ly ready to accept intellectual chal­ lenge . Some ethnic mi nority children may fai l to com­ prehend or accept the achievement aspects of the test situation , howeve r. They may view it as an enjoyable child­ adul t encounter, rather than as a time to achieve; or, i f they recognize the problem-solving aspects of the situatio n , they may ignore the m . For examp l e , American Indian children may not work quickly on timed tests because of a desire no! to compete with others . Although ethnic minority children may have an adequate storage and retrieval system to answer questions correctly, they may fai l in practice because they have not been ex­ posed to the material (Zigler & Butterfield, 1968) . For example , they may respond incorrectly to the question "What is a gown?" because they have never heard the word gown . Motivational factors also may affect the perfor­ mance of some ethnic minority childre n ; they may know what a gown is, but respond w ith "I don't know" in order to terminate as quickly as possible the unpleasantness of i nteracting w ith a strange and demanding adult . Addi­ tionally, they may be more wary of adults, more motivated

CHAPTER 1 9

ASSESSMENT OF ETHN I C MINORITY CHILDREN

to secure adults' attention and praise, less motivated to e correct for the sake of correctness a lone, or more w i l l ing to settle for lower levels of achievement success. Therefore , low IQs may be associated w ith l i mited exposure to test content or w ith motivational factors . There is suffic ient evidence that some ethnic minority children have l imited test-taking s ki l l s . But this is true of some non-ethnic minority children as wel l . We do n ot know how pervasive this l i m itation is among ethnic mi nor­ ity children, or to what degree it l owers t he i r performance on tests . Further research is needed in this area.

Racial Examiner Effects With black children. The anx iety, i nsecurity, l atent prej udice, and other reactions to contemporary blac k­ white relations that are experienced by white c l i n ic ians i n their work w ith black c h i ldren may be t ransmitted t o the c h i ldren in a number of way s . Examiners may exhibit paternal ism , overidenti fication , ove rconcern , excessive sympathy, i ndulgence , reactive fear, or inhibition . Black children, in turn , may exhibit fear and suspicion, verbal constriction , strained and unnatural reaction s , or a facade of stupidity to avoid appearing "uppity. " Some may del iber­ ately score low to avoid personal thre at ; others may view the test as a means for whites, not blacks , to get ahead i n society. Although many o f these behav iors , patterns , and perceptions are l ikel y to exist and are important phe­ nomena in their own right , there is no way of knowing to what extent they affect black children's test scores ( Sattler, 1970, 1973a) . Some psychologists believe that white examiners impair the intell igence test performance of black c h i ldren . A careful study of the research l iterature refutes the myth of racial examiner effects, however. In 2 5 of the 29 publ ished studies deal i ng w ith racial examiner effects on individual i ntell igence tests or other cogn itive measures, no signifi­ cant relationship was found between the race of the exam­ i ner and the examinees' scores (Sattler & Gwynne , 1982b) . It is apparent that in the overwhelming majority of cases white exami ners have not impaired the intell igence test performance of black childre n . The result is all the more i mpressive when one considers the wide range of tests, grade leve l s , geographic areas , and dates of adm i n i stration encompassed by these studies. The tests included the WISC, WAI S , W PPSI , Stanford-Binet : Form L-M , PPVT, Draw-A-Man, Iowa Test of Preschool Development, and several other tests of cognitive abil ity. The grade levels of the children in these studies ranged from preschool through grade twelve . The geographic

ARGUMENTS AGAINST T H E USE OF INTELLIGENCE

-

ES TS Ir\J ASSESSING ETHNIC MINOR;ry CHIILDREN

locations, though largely urban , i ncl uded Eas;ter n . Mid­ western, Southern , and Western c ities . The ye ars of pub­ l ication ranged from 1964 to 197 7 . Communication difficulties may be present i n i nter­ ethnic testing situations (see Chapter 16) . Because misun­ derstandings may result i n mistrust and sustain ste eotypic j udgments, every attempt should be made to reduce the l ikelihood of any misunderstandings. Differe nces in di­ alect may be a source of d i fficulty. Some argue that black children do not clearly understand white examiners, and this causes their scores to be lower than those they obtain when tested by black examiners . Studies, however, have failed to provide any support for this position . Quay ( 1972 , 1974 ) , for example, reported that black children scored no higher when the Stanford-Binet : Form L-M was adminis­ tered in black dialect (by a black examiner) rather than in standard English . There is increasing evidence that black children are bidialectical in that they have the ability to comprehend black dialect and standard E nglish equally well (Genshaft & H i rt , 1974; Hal l , Turner, & Russel l , 1973 ; Harver, 1977 ; Levy & Cook , 1 97 3 ) . Although the race o f examiners has not been found to affect black children's performance on inte l ligence tests , examiners cannot be indifferent to the examinee's ethnicity. They must be alert to any nuances i n the test situation that suggest an i nval id performance . Testing chi ldre n from different cultures is a demanding task . At t i mes it may be d i ffi c u l t to u nderstand children's response s , and every effort must be used to enl ist their best efforts.

With Hispanic-American children.

Stereotypes held by the Anglo-American examiner about the Hispanic­ American examinee , or by the Hispanic-American exam­ inee about the Anglo-American examiner, may i nterfere w ith rapport . The two groups are keenly aware of the d i fferences that d iv ide them, and feel i ngs of resentment ­ stemming from a mutual lack of u nderstanding - may be present on both sides. Anglos generally do not know much about the customs and values of H ispanic-Americans, nor are they knowledgeable about the conditions that exist in the barrio ( section of town in which H ispanic-Americans l i ve ) . The H ispanic-American examinee's language may serve as a cue for group identification, and, l i ke skin color, it may i nfluence the examiner-examinee rel at ionsh i p . Every attempt must b e made b y examiners to overcome any stereotypes they may have . The assertion that Anglo-American examiners are not as effective as Hispanic-American examiners in testing H ispanic-American children has not received support . For examp l e , Gerken ( 1978) reported that neither the exam­ iners' ethnicity (Hispanic-American or Anglo-American)

57 1

nor their language f.aciUity (bili ngual or monol ingual) sig­ n ificantly affecte thle I s obtained by H ispan ic-American kindergarten chi ldren on the WPPSI or the Leiter Interna­ tional Performance Scale . Morales and George ( 1976) found that bil ingual Hi spanic-American fi rst, second, and third graders obtained h igher WISC-R Performance IQs when tested by monolingual non-Hispanic examiners than when tested by bilingual Hispanic examiners who gave the test directions in both Engl ish and Spanish . The children tested by non-Hispanic examiners also obtained signifi­ cantiy higher scores on the ITPA Grammatical Ciosure subtest , but not on the Screening Test of Spanish Grammar. These two studies. although limited , do show that Anglo­ American examiners do not necessarily impair the test performance of Hispani c-American elementary school c h ildren. Although the studies above provide no support for the assertion that AnglO-American examiners are less effec­ tive than Hispanic-American examiners , examiners of both ethnic groups must be aware of stereotyped attitudes toward Hispanic-American c h i ldren that may i nte rfere with their c l i n ical judgments . For exampl e , a study of teachers' attitudes toward Hispanic-American third and fourth graders who spoke minimally accented . moderately accented , or h ighly accented English showed that most favorable rat ings were given to the minimi\l l y accented speakers and least favorable ratings to the h ighly accented speakers (Carter, 1977 ) . Although H i s panic-American teachers had more favorable att i t u d e s than A n g l o ­ American teachers. both groups had more unfavorable attitudes toward the highly accented speakers . These re­ sults suggest that examiners may hold stereotypes about children who have accented speec h . Such stereotypes must not be allowed to affect test interpretations and recommen­ dations or to impair the examiner-examinee relat ionsh i p .

Placement i n Inadequate and Inferior Educational Programs Test results, it is c laimed , are used to place black chi ldren in special education c lasses (or tracks) , which are consid­ ered inadequate and infe rior. This argument is based on a number of premise s . One premise is that black children who are placed in special classes would achieve at a higher level if they were not removed from the regular class . Tests are held accountable because they are one of the means by which the schools place children in special classes. A second premise is that test resul ts produce negative expec­ tancies in teachers - when teachers learn that black chil­ dren have low scores, they then begi n to treat them as if they will perform at a below-average level .

572

The role of individual assessments.

A lthough test re­ sults may be one link in the educational chain that leads teachers and school administrators to assign children to special classes or programs, in most cases children are referred for indiv idual assessment only after they have performed poorly in school . Most black childre n , and other children for that matter, are never seen for a compre­ hensive individual assessment . Thus the school perfor­ mance of ethnic minority children can seldom be l inked to the results of individual psychological assessments.

Expectancy effects.

The premise that test results con­ tribute to the development of expectancy e ffects has some merit . The strength of such expectancy effects must be evaluated , however. The claim that initial negative expec­ tancies produce a self-fu l fi l l ing prophecy has l ittle merit. The sel f-fulfilling prophecy concept recei ved great i mpe­ tus from the work of Rosenthal and Jacobson ( 1968 ) , which supposedly demonstrated that children who were charac­ terized as "late bloome rs" performed better than other children on abil ity tests . This study, however, had so many pitfalls that its results cannot be accepted (Snow, 1969 ; Thornd i ke , 1968 ) . As Cronbach ( 1975, pp. 6-7) observed : In my view, Pygmalion il1 lhe Classroom merits no consideration as researc h . The " experimental manipulation" of teacher belief was unbel ievably casual - one sheet of paper added to the teacher's in-basket, which apparently moved within seconds to the wastebasket. The technical reviews indicate that the adver­ tised gains of the "magic" hildren were an artifact of crude experimental design and improper statistical analysis. ( No doubt there are expectancy efrects in the classroom. The question ought to be whether tests add to bias or instead bring expectations closer to the trut h . On that there is no direct evidence . )

Other studies have failed to document this type of self­ ful fi l l i ng prophecy (or expectancy) effect (Anttonen & Fleming, 1976; C laiborn, 1969 ; Dusek & O'Conne l l , 197 3 ; Fielder, Cohen , & Feeney, 1971 ; Fleming & Anttonen , 197 1 ; Ginsburg , 1970; Gozali & Meyen, 1970) . In a meta­ analysis of 47 teacher expectation studie s , the effect of teacher expectations on pupils' IQ was found to be very small (M effect size was . 16 of a poi nt for 22 effects) (Smith , 1980) . The i ntellectual growth of children does not appear to be hampered by teachers' knowing the pupil s' intell igence test scores. The study of self-fu l fi l l ing proph­ ecies in the classroom is quite complex , as it involves teache rs' com m u n i cation of expectations to stude n t s , teache rs' bel iefs about cu rriculum, the effectiveness of i nstruction , student motivation , the qual ity of the teacher­ student relationsh i p , and teacher and student individual d i fference variables (Brophy 1983 ) .

CHAPTER 1 9

ASSESSM ENT O F ETHN IC MINORITY CHILD E N

Are special education classes needed? We must also carefu l l y e x a m i n e t h e pre m i se that spec ial educat ion classes for the mentally retarded do not provide the ki nds of intervention programs needed by ethnic m inority c h i l ­ dren w h o are functioning in the mentally retarded ran ge . Do low-fu nctioning ethnic minority children need special programs emphasizing enrichment and verbal stimula­ tion? Do they benefit from the u sual type of program designed for the mental ly retarded? Definitive answers to these questions a re not available . We need to learn a great deal about the d i ffe rential effectiveness of special p ro­ grams - such as resource roo m s , tutoring , nong raded c lasses, and learning centers - and traditional programs designed for low-functioning childre n .

A R G U M ENTS F O R T H E U S E O F I NT E LL I G E N C E T ESTS I N ASSESS I N G E T H N I C M I N O R I TY C H I LD R E N

Although some o f t h e previous arguments against t h e use of intell igence tests in the assessment of ethnic m inority childre n deserve consideration , the c rucial ones related to culture bias have littl e , if any, merit . The arguments for the continued use of intell igence tests and other ability tests for assessing ethnic m i nority children w i l l now be considere d .

Useful in Evaluating Present Functioning I ntell igence test scores of ethnic m inority chi ldren are useful i ndices of their c urrent cognitive abil ity. The evi­ dence c ited earl ier indicates that i ntell igence tests have l ittle or no cultural bias . They generally have the same properties for ethnic mi nority children as they do for white chil d re n . Tests can provide val uable information about ethnic m inority children's cognitive strengths and weak­ nesses and can help to evaluate change and progre s s . For exampl e , in cases of brain damage , intelligence and spe­ c ial abil ity tests help to document the extent of damage (see C hapter 2 2 ) . Present test results can be compared with past results to evaluate changes as a result of the damage . I n cases o f psychopathology, tests serve similar purposes . Doing away w ith tests would deprive clinicians and educators of v ital information needed to assist c h ildren. A report sponsored by the American Psychological Associa­ tion's Board of Scientific Affa i rs (Cleary, H umphreys, Kendric k , & Wesman, 1975) stressed the i m portance of eval uation i n education : "Diagnosis, prognosis, prescrip­

tion , and measurement of outcomes are as important in education as in medicine " ( p . 1 8 , italics added) .

GENERAL CONSIDERATIONS IN ETHNIC

573

M I N ORI T Y TES TI N G

Useful i n Indicating Future Function i ng Standardized intell igence tests provide good indices of future leve l s of academic success and pe rformance as defined by the majority culture .

Useful in Obtaining Special Programs Tests can be useful in obtaining spec ial en richment pro­ grams and services for ethnic minority children as well as other childre n . Abandon ing formal assessment procedures may deprive handicapped ethnic m i nority children of the opportunity to obtai n appropriate attention and services to which they are legally entitle d . The problem of poor achievement of black inner-city children is real . It has not been caused by tests . Tests have been helpful in document­ ing the severity of the educat ional defi c i t s . As Green ( 1978 , p . 669) noted, "The tests are not b igoted v i l lains but color­ b l i nd measuring i nstru ments that have demonstrated a soc ial problem to be solved ."

Useful in Evaluating Programs Tests evaluate the outcomes of school and special pro­ grams. They can be used to determine whether children have learned to read or to perform arithmetical operations. Thus tests serve as a means of prov iding objective evidence of school accountabil ity. "As such , farfrom being a pari of

the problem , tests are an absolutely essential part of the solution" ( Flaugher, 1974 , p . 14, ital ics added ) . Those call i ng for the e l i m i nation of testing altogether WOUld, in fac t , allow the educational system to be released from any accountab i l i ty at a l l .

depend on the predi l ect ion s of teachers at a spe c i fi c school or on a student's int erpers nal relationship with teachers . Consequent ly. tem have helped to prevent educational mi placement of many ethnic mi nority chi ldre n . Ethnic m i nority groups should favor t h e use o f abil ity tests : these tests onstit ute a unive rsal and objective rather than a prejudicial standard of competence and potent ial . Other selection methods may decrease the opportun it ies of ethn ic min o r i ty children. Jensen ( 1975 . p. 67) observed that "objective means of reveal ing talents stand to benefit talented memhcr� of disadvantaged groups the most, since in their cultural circumstances certain talents arc more apt to go unrecogni zed and underdeveloped . " Educational tests encourage and rcward individual efforts to learn . If tests were abandoned. programs would be d i fficult to eval ­ uate and educational opportunities might be based more on ancestry and infl uence and less on aptitude and merit . Decisions on curriculum would be based less on evidence and more on prejudice and caprice.

Comment on the Value of Intelligence Tests Intell igence tests (and other abil ity tests) measure the abil i ­ ties that they we re designed t o measure w ith reasonable accuracy for ethnic mi nority chi ldren as well as for chil­ dren from the dominant cu lture . Tests have the potential to do much good in our society. Tests assess a child's cu rrent intel lectual functioning irrespective of race or soc ial sta­ tus . The consequences of not testing m ight be to increase

bias and d i scriminat ion .

G E N E RA L CONSI D E RATIONS I N ETH N I C M I NORITY TEST I N G

Useful in Revealing Inequalities By reveal ing the i nequal ities of opportunity available to various groups, tests may provide the stimulus for special i nt ervention to facilitate the maximum development of each child's potentialitie s . Perhaps those i n the field of testing now should be less concerned w ith unbiased pre­ dicti ve validity and more concerned with facil itating equal opportunity.

Useful in Providing an Objective Standard Tests serve as a corrective device by providing information that cannot be obtained easily or rel iably by other means. They give students an alternative means of demonstrating academic ability. Tests also provide a measure that is com­ parable across schools and across time. Test scores do not

The Culture of Poverty and Its Effects on the Test Performance of Ethnic Minority Children Many member of our nation's ethnic minorities suffer from economic deprivation . Poverty affects maternal heal th as wel l as the chi ld's own health and soc ial function­ ing; both may be related to school fai l u re . Poverty may also affect rate of learn ing, which in turn influences intel­ ligence and academic success . Figure 19-1 shows the vicious c i rcle uniting various environmental variables with poverty and educational fai lure . The conditions assaulting inner-city children'S central nervous systems include per­ inatal disease , malnutrition , i nfection , anemia , and lead poi oning . Blacks also have a considerably h igher rate of premature birth and a higher inc idence of abnormalities of pregnancy than do whites ( Wiener & M ilton , 1970) .

574

CHAPTER 1 9

ASSESSMENT OF ETHN I C M I NORITY CHILDREN

/

POVE RTY

Poor m a t e r n a l h e a l t h P o o r m a t e r n a l g ro w t h Poor m a t e r n a l n u t r i t i o n I n adeq u ate f a m i l y p l a n n i n g Poor o bs t e t r i c a l s u perv i s i o n

E l evated i n f a n t m o rt a l i t y E l evated i n fa n t m o r b i d i t y E l evated f a m i l y s ize

I \ j t

� ./

Malnutrition l l l ness A b S e n C of m e d i c a l ca

Soc i a l d e p r i v a t i o n Environmental i n adeq u a c y

I N C R EASED R I S K O F S C H O O L FA I LU R E

t

U N E M PLOY M E NT A N D U N D E R D E V E L O P M E N T

Figure 1 9- 1 . Environmental relationships between poverty and educational failure. Reprinted b y permission o f t h e publisher and authors from H . G . Birch and 1. D . Gussow, Disadvantaged Children: Health, Nutrition , and School Failure, p . 268. Copyright 1 970, Grune & Stratton .

Poverty, as suc h , is neither a necessary nor a sufficient condition to produce intel lectual deficits, especially if nu­ trition and the home cultural environment are adequate . It is only when these and related factors are i nadequate that there are l i kely to be learning deficits . Imagine, for exam­ ple , what effect poor nutrition and health care , substandard housing, low fam i l y income , p revalent fam i l y d isor­ ganizat ion , anarchic d i sc i p l i n e , d i m i n i shed personal worth, low expectations , frustrated aspirations, and other environmental handicaps may have on the development of adequate intellectual ski l l s . Table 19-4 summarizes some key ind ications of the culture of poverty.

l ikely to occu r when chi ldren have l ittle or no opport u nity to develop their mental abilities . A l ternate e laboration occurs when c h i l d ren develop mental abilities in ways that are d i ffe rent from those of most other childre n . Some childre n who l i ve in the i n ne r c ity acqui re a k i n d of cognitive language that may make it difficult for them to elaborate their abilities "or to learn academic material i n a traditional school setting" ( E l ki n d , 1 97 3 , p . 8 1 ) . T h e fol lowing case i l lustrates how a low­ income youth may appear to have l imited abil ity, as as­ sessed by standard tests , but function quite wel l in his community.

Elkind's premature structuring and alternate elabora­ tion hypothesis. Two intriguing concepts premature

One young man . . came to the court clinic because he was not going to school and was suspected of pushing dope and of having a couple of girl friends doing "tricks" for him. He was well dressed and smooth talking with a vocabulary rich in the argot of the street and the underworld. He was alternatively amused and annoyed at the test questions, and he did best in arithmetic and worst on the test of general information and vocabulary. H i s poor performance on information and vocabulary, how­ ever, does not reflect deficiency but rather alternate elaboration. This young man was far from lacking in general information. Indeed, his fund of information in many areas was much greater than mine. The same was true for his vocabulary, which was rich in words not found on intelligence tests . To call this young man deficient in ability would be a gross error. On the contrary, his at

-

structuring and alternate elaboration have been offered -

to account for the lower test scores of low-income and ethnic minority youth . Premature structuring occurs when chi ldren "are forced to apply thei r abi l ities to practical matters before these abi l ities are ful l y real ized or elabo­ rated . The effect is to stunt or l i m it growth" ( El kind , 197 3 , p . 79) . Low-income youth may have t o direct their i ntellec­ tual abil ities to practical issues, such as assuming responsi­ b i l ity for their sibl ings at an early age , and such early appl ication to real- l i fe matters may i n h i b it the future growth of cognitive ski l l s . Premature structuring is most

575

GENE RAL CONSIDERATIONS IN ETH N I C M INORIT'( TESTING

Table 1 9-4 Some Ind ications of the Cultu re of Poverty

Individuals

Comlllunity

Family

Feelings and beliefs

Arc unemployed or u nder­ employed in a III iscel lany of unskilled occupations Do not own properly Do not belong to labor unions Do not participate in Social Security Borrow from local moneylenders Lack food reserves in the home Lack savings Are chronically short of cash Are constantly strugg l i ng for survival Have a low level of education

Housing is poor and crowded . Organization is minimal be­ yond the level of the nu­ clear and extended fam ily.

Famil ies are often female- or mother-centered . Family sol idarity is an ideal rarely achieved. Childhood is not treated as a specially prolonged and protected stage in the l i fe cycl e . Structure is often authoritarian .

Strong feel ing of al ienation Feel ing that exist i ng institu­ tions do not serve interests and needs Strong feel i ngs of helplessness Sense of resignation and fatalism Strong feel i ng of dependence Fee l i ngs of powerlessness Belief in male superiority Martyr complex aIll o ng women

Source:

Adapted from Kutner ( 1 97 5 ) .

least average intellectual abil it ies were elaborated in a subcultural domain that is virtually unknown to the middle-class test maker and user. ( E l kind, 1 9 7 3 , p. 80)

Premature structuring and alternate elaboration are pos­ sible replacement models for other deficit model s that have been proposed to account for the poor test performance of low-income youth . "From a developmental standpoint the mental growth of some low income youth may simply be different in d i rection and elaboration than that of the off­ spring of more affluent parents" (Elkind, 1 97 3 , p . 8 1 ) .

Poverty and testing_

Formal testing procedures may identify those indi viduals who cannot compete in our tech­ nologically oriented culture. When improperly used, tests can become major instruments for casting out the phys­ ically sick, the uneducated , and those w ith special per­ sonal problems . The poor, as a g roup, are the most vulnerable, and it is po vert)' that is the main shared charac­ teristic of minority groups (Leland, 197 1 ) . The adaptive strategies of the poor are not always conducive to good test performance . They may cope poorly w ith external pres­ s u res and experience fai lure - even in areas where they have some cogn iti ve strengths - because they feel that things often happen to them in spite of themselves and w ithout their participatio n .

Testing situations may arouse tension and feel ings of suspicion in poor children . They may react with aggres­ s ion or w ith passivity, but simultaneously may feel that it is i m portan t to establish a friendly relationship with the ex­ aminer. If they become too preoccupied about their rela­ tionship with the exami ner, they may fail to give their undivided attention to the test questions . Such strategies leave the poor child i l l -equ ipped to cope with test s . Leland ( 1971) made the tel l ing observation that when children who perform adequately i n their own environment are given a label as a result of testing, the label makes the chi ldren visible and begins to c reate social problem:; where pre­ viously none had ex isted .

Cumulative deficit hypothesis. The term cumulative deficit refers to a progressive decline with age in the measured intel l igence or scholastic achievement scores of culturally or socially disadvantaged children relative to those of more advantaged children . The evidence is by no means c lear about the extent to which such defic its appear or the ethnic and subcultural groups involved (Cox, 1983). The theory behind cumulative deficit is that children who are deprived of e nriching cognitive experiences during their early years are less able to profit from environmental stimulation because of a m ismatch between their cognitive

576

schemata and the requirements of the new (or advanced) learning situations . The cumulative deficit theory accords a c rucial role to the child's early learning . It is not know n , however, t o what extent the quality o f a child's learni ng following early deprivation can improve cognitive func­ tioning .

Effects of Response Style on Test Scores Ethnic m inority children, as we have seen . may approach the test situation w ith attitudes and test strategies that differ from those of non-ethnic minority childre n . Those ethnic m inority children who have a tendency not to attempt a response when given a question or to terminate an initial response before completing it are more l ikely to obtain lower scores than are those who do try to solve the test questions . The scores of those who tend to terminate responses prematurely (giving up with l ittle effort) may be more an index of their "wil l i ngness to retrieve information or to problem solve on demand, and less an index of acc u m u l ated knowledge and a b i l ity" ( Moore , 1 9 8 6 , p . 322 ) .

Spontaneous elaboration response style.

Moore ( 1986) describes the importance of a spontaneous elaboration response sty l e : Children's tendency t o spontaneously elaborate o n their work responses may be a very important index of t.heir level of invol ve­ ment in task performance, strategies for prohlem solving, level of motivation to generate a correct response, and level of adjustment to the standardized test situation. For example. a child may not have learned the dictionary definition of the word "nuisance ," but may be famil iar with the word from some context. Once the child places the word in an experiential context, the meaning may be inferred and a definition constructed during testing ( e . g . , "Nui­ sance , um, that's what my teacher always calls me when I'm acting silly, um, it's something that bothers you, something that disturbs you , l ike fl ies; they're a nuisance when we go camping, right?") . The child's elaboration on the definition of a word can serve a number of purposes for the child and tester. First , it can provide feedback to the child on the efficiency with which the problem-solving strategy employed (in this case. association) can be used to generate correct responses to test items, and it can confirm that the tester understands why the child views the construction of the definition of nuisance as correct. The spon­ taneous extension may also serve as a tension release for the child . In providing the elaboration, especially when generating a response to a demand that challenges the child, the child verifies for him- or herself that based upon h is or her understanding of the problem. the response is reasonable and acceptabl e . T h e child's use o f spontaneous elaborations o n responses indi­ cates to the tester that the child is task focused by virtue of the

CHAPTER 1 9

ASSESSMENT OF ETHNIC MINO RITY CHILDREN

child's ability and willingness to overtly associate the problem with personal experiences. Such responses also indicate that the child is motivated to provide appropriate responses to the prob­ lems that the test poses and is sufficiently relaxed in the test situation to explore different strategies for generating solutions to the problem. ( p . 322)

Maternal behaviors and response style. There is some tentative evidence that maternal reactions to response style d i ffer in black and white groups. Black mothers in one study ( Moore, 1986) tended to be less supportive than white mothers of child-initiated strategies that did not contribute to problem solutions . Children raised in this envi ronment may believe that "rather than guess at the answer to a challenging question and risk d isapproval of adults (such as the tester) for being wrong, it is better to say '1 don't know' or request help from the tester" ( Moore , 1986, p . 325) . In contrast , the implicit message from white mothers was "It's okay to be wrong as long as you are trying" ( Moore , 1986, p . 3 2 5 ) . Perhaps the ethnicity o f t h e child's envi ronment , not just socioeconomic status and maternal education leve l , exerts a significant influence o n children's styles o f responding to standardized inte l ligence tests and their test achievement. The effects of the ethnic rearing environment appear to be mediated, in part, by mothers through their socialization of affective orien­ tations to impersonal problem-sol v i ng s ituations and the problem-solving strategies they encourage . (Moore, 1986, p. 325)

Bilingualism B i lingualism generall y involves the learning of a second language, although it may involve the learning of two languages simultaneously. Research on second-language learning in children provides no evidence that bilingualism affects general i ntel l ige nce , although in some individual cases verbal scores may be lowered and performance in English-speaking classrooms may be i mpeded because of ignorance of words, ideas, and grammatical structures (McLaughli n , 1977 ) . The effects o f bilingual ism depend in part o n whether children are adding a second language to a wel l-developed fi rst l anguage or whether a second language is gradually replac i ng the first language . H ispanic-American children, for example, usually are requi red to learn Engl ish as a second language and then to use this second language in their school work. They continue , however, to use Spanish at home and i n the community, in speaki ng but seldom in read i ng . Because of this form of bilingual ism, many chil-

GENERAL CONSIDERATIONS IN ETHNIC MINORITY TESilNIG •

Exhibit 1 9-2 Informal Assessment of Language Preference Questions for Teacher __

1 . What language does use in the c1assroom� 2. In what language can read? 3 . In what language does speak with his (or her) classmates? 4. What language does use on the playground� 5 . I n what language does write? 6 . Overal l , how competent is in Engl i,h? 7. Overall , how competent is in ( lan­ guage)? __

__

__

__

__

__

__

Questions for Parents 1 . In what language do you speak with ? 2 . In what language does your spouse speak with ? __

3 . In what languages do you speak with your spouse? 4. In what language does speak with you? 5 . In what language does speak with his (or her) father (mother)? 6. In what language does speak with his (her) sisters and brothers? 7. What language does prefer to speak at school? 8. In what language are the television programs watches? 9. In what language do you read stories to ? 1 0 . In what language does prefer to be tested? __

__

__

____

__

___

__

d ren fail to develop a suffic ient mastery of either language, and learning is more di fficult under such conditions . A number of overlapping informal methods are useful for determining the language preference of children who speak more than one language . F i rst , you may ask the child in which language he or she prefers to be tested . Second , you may observe which language the child uses i n the classroom and a t home . Third , you may a s k the teacher and parent to describe the child's language preference , using questions such as those shown in Exhibit 19-2 . C lassifying a chil d's degree of language proficiency is a d i fficult task. Although there are a number of formal i n struments availab l e for evaluating Spanish-speaking children's language proficiency, few, i f any, meet accept­ able psychometric standards . Additionally, norms on bi­ l ingual assessment i nstruments are not representative of all groups of H i spanic-Ame rican c h ildre n . B i l i ngual tests also may show a d isparity i n how they classify children as to their degree of English proficiency. For example, of a group of first-grade H ispanic-American children, 13 per­ cent were classified as fl uent English speakers by the

S77

Bil ingual Syntax Measure , 30 percent by the Language Assessment Scales , and 58 percent by the Bilingual I n ­ ventory of Natural Language ( Ul ibarri , Spencer, & Rivas , 1981 ) . Appendix F reviews the B i l ingual Syntax Measure , Bilingual Syntax Measure I I , Language Assessment Bat­ tery, and Language Assessment Scales. After making an informal and formal assessment of language proficiency, you should classify the child's degree of language proficiency. A useful 5-point classification scale i s as follows : I . Monol i ngual speaker of a language other than En­ glish (speaks the other language exclusively) . 2 . Predominantly speaks a language other than English (speaks mostly the other language, but also speaks some Engl i s h ) . 3 . B i l i ngual (speaks both the other language a n d En­ glish with equal ease ) . 4 . Predomi nantly speaks English ( speaks mostly En­ glish, but also speaks some other language ) . 5 . Monolingual speaker o f Eng l i s h ( speaks English exclusive l y ) .

Racial Differences in Intelligence In 1969 Arthur Jensen publ ished an article in the Harvard Educational Review called "How Much Can We Boost IQ and Scholastic AchievementT which raised a storm of controversy that has yet to be qu ieted . According to McGuire and Hi rsch (197 7 ) , this article was responsible for the resurgence of racist t h i n king in the United State s . Others maintain that Jensen's v iews must not b e dismissed lightly (Denniston , 197 5 ; Eysenck , 1 9 7 1 ; Herrnste i n , 197 3 ; Nichols , 1974 ) . In the ensuing years , Jensen published numerous books and articles defending and elaborating on his position, culminating in his text Bias in Mental Testing, published in 1980 . The major points of Jensen's 1969 article were as fol lows : 1 . I ntel ligence tests measure a highly relevant dimen­ sion of general abil ity. 2 . This di mension has a high degree of heritabi l ity (about 80 percent ) . 3 . Educational programs have not been effective i n bringing about significant changes in this abil ity. 4 . Genetic factors are strongly impl icated in the aver­ age black-white i ntell igence d i fference (about 1 5 points i n favor o f whites) , although environmental factors still may play a role in accounting for this difference . 5 . Teaching methods should be tailored to the child's particular learning ski l l s .

578

The heritabi l ity estimate of . 80 for inte l l igence is based on tudies of wh ite persons; consequently it is difficult to know how accurate this estimate is for other rac ial groups. As noted in Chapter 4 , recent advances in genetics suggest that a heritabil ity estimate of around . 50 may be more accurate . He ritabil ity estimates based on studies within one spec ific group may have l ittle relevance to an under­ standing of how genetic variation can account for differ­ ences in a trait between two or more groups . Consequently. the heritab i l ity of IQ differences between rac ial or ethnic groups cannot be estimated at present . The present consensus is that val id i n ferences cannot be drawn about genet ic differences among races as long as there are relevant systematic d i fferences in cultural pat­ terns and in the psychological environment. These differ­ ences influence the development of cognitive skills in com­ plex way s , and no one has succeeded in either estimating or e l i m i nating their effects . Centuries of discrimination have made meaningless the d i rect comparisons of biolog­ ical mental abil ity traits of blacks and whites. The data from which heritability is established come. in general . from studies of white children attending school or in families in which the value of schooling for a child's future is taken for granted . In general, that environment lacks at least one funda­ mental feature of the black child's: prejudice. with effects cumulative over generations. The black child is not only exposed to it himse l f; he is brought up also by parents who are al ready defeated by it. We have simply no way of estimating the extent of its effect on cognitive development . The direct effect of poverty is the lack of toys and books . and parents with time to talk to the child. The effect must be multiplied in an environment in which hope has been lost . There is no justification for general izing to this environment from that of the ordinary white child. ( Hebb. 197 1 , p. 736)

Even when social class and economic variables are equa­ ted , there are still important differences between ethnic groups i n l i festyle and experience . During the fi rst year or two of l ife , c h i ldren exhibit few d i fferences in intel lectual functions related to race or social class . At 3 or 4 years of age , however, race and social class differences arise that remain fairly stable during the school years . These fi ndings suggest that " the schools can mostly be exonerated from the charge of creating the black-white d i fference in average IQ-test performance or of increasing it d u r i ng the period of school attendance" ( Loeh l i n . Li ndzey, & Spuhler, 1975, p p . 156-15 7 ) . The lower scores of black children may be associated w ith the increased g loadi ng of inte l l igence test items between 2 and 5 years of age or with certain environmental diffe rences in the black home or community (Jensen . 1975 ) . The debate about the relative importance o f genetic and

CHAPTER 1 9

ASSESSMENT OF ETHNIC MINORITY CH ILD'REN

environmental infl uences on d i ffe rences between bLack and white groups in our country should not obscure the facts that (a) within both the black and the white population there are w ide individual variations in genetic potential and (b) there are w ide variations in educational opportu ity. Fu rthermore , a heritabil ity hypothesis should not be u.sed as a convenient rational ization for inadequate schooling or for our fai l ure to cope with the emergence of a growing and self-perpetuating lower c\ ss, disproportionately Afro- and Latin-American in its ethnic com­ position. excluded from the mainstream of American l i fe and al ienated from its values. isolated in rural areas and urban g et­ tos, and dependent for the means of bare survival on an in­ creasingly hostile and resentful majority. ( Layzer, 1972 , p. 2 67)

Ethnicity and Patterns of Mental Abilities An early study ( Lesser, Fifer, & Clark, 1965) i nvestigat i ng the organ ization of mental abi l ities i n various e t h n ic groups fou nd that patterns of i ntellectual ability differed in Chinese , Jewish , black, and Puerto Rican children . Chi­ nese children were found to have excellent spatial ability, but were considerabl y weaker than the other groups i n verbal abil ity. Jewish children's verbal s ki l l s were superior to those of the other three groups . Black c h i ldren were weak in spatial skills and average in verbal ability. The Puerto Rican children's verbal skills were weaker t han their other abi l ities . Other studies, however, have failed to replicate t hese findings . Sitkei and Meyers ( 1969) , using a battery of 22 tests, reported that the pattern of abilities of black and wh ite children from the lower and middle socioeconomic classes was essential ly simi lar. Flaugher and Rock ( 1972) found that a battery of nine cognitive tests given to j u n ior h i g h s c h o o l s t u d e n t s w i t h b l ac k , w h i t e , M e x i c a n ­ American , and Asian backgrounds y ielded similar pat­ terns of abilities, regardless of the children's ethnic back­ ground . Finally, after making an extensive study in which a battery of 10 mental abil ity and achievement tests was admini stered to 2 ,985 Afro-American , H ispanic, Jew ish , and Caucasian-gentile high school seniors, Hennessy and Merrifield ( 1976) reported that there are highly similar structures of mental abil ities in the four groups . These fi ndi ngs suggest that intell igence and achievement tests l i kely measure the same abi l ities in a variety of ethnic groups.

Development and Use of Culture-Fair Tests for Assessing Ethnic Minority Children Attempts to develop tests that are culture-fair have not been successfu l . W i l l iams ( 19 7 2 ) , for example, developed a

579

SUBCULTURAL CONSIDERATIONS

100-item m u l tiple-choice test te rmed the Blac k r ntel ­ l igence Test of Cultural Homogeneity ( BITCH ) , based on items drawn �rom the black culture . It is a culture-specific test measuring special information about the inner city. Items deal with black slang, which is itsel f not uni form throughout the country. Two examples are "Boot : (a) c tton farmer, (b) blac k , (c) I ndian, (d) V ietnamese citizen-' and "Clean : (a) just out of the bathtu b , (b) very well dressed , (c) very rel igious, and (d) has a great deal" (answers are b for both items ) . W i l l iams ( 1972) reported low correlations between the BITCH and the Cal i fornia Achievement Test ( Reading, r = . 39 ; Language , r = . 3 3 ; and Mathematics, r = . 1 8) in a sample of 28 blacks 16 to 1 8 years old . Other studies have indicated that the BITCH has ques­ t ionable val idity. One study ( Long & Anthony, 1 974) showed that a group of 30 black 16-year-old high school students , enrolled in educable mentall y retarded classes in Florida , did no better on the BITCH than on the W ISe . They scored below the 3rd percentile on the WISC and at the 1st percentile on the BITC H . In another study (Andre, 1976) of 150 seventh graders , both black and white , in a Southeastern urban school system , th� b!ack middle-class adolescents obtained higher scores on the BITCH than did the wh ite m iddle-c lass adolescents , whereas lower-class whites and blacks obtained the same score . In still another study (Kom m , 1978), mean scores obtained on the BITCH by 65 black 13 1/2 - to 16 V2 -year-olds in San Francisco were 25 poi nts lower than those obtained by the St . Louis standardization sample (M = 62 vs. 8 7 ) . Furthermore , the WI SC-R was found to be a better pred ictor of achievement as measured by the C a l i fornia Achievement Test than was the BITCH ( for exampl e , with Vocabulary, BIT C H r = .47, W ISC-R Full Scale r = 60) . The BITCH may be useful for bui lding black pride, but it is not a useful predictor of educational success . The test measures knowledge of black slang , not problem-sol ving or reason ing abilitie s . At present, the BITCH does not appear to be useful in the assessment of the cogn itive abi l ity of black children . Tests l i ke the Leiter International Performance Scale and Raven's Progressive Matrices tend to be thought of as c u lture-fair because they emphasize problems involving pictorial, spat ial , or figural content , wh ich many bel ieve may be answered on the basis of experiences that have been more nearly equal across ethnic and rac ial groups. These test s , however, have been no more successful than special culture-fair tests , such as the Davis-Eells Test of General Inte l l igence on Problem Solv ing Abil ity, the Culture Fair Intelligence Test, and the BITC H , in reducing measured abi l ity d ifferences among ethnic mi nority groups . Ethnic m i norities do not perform any better on supposedly

culture-fair tests than on the more conventional tests of inte l l igence (Arvey, 1 9 7 2 ) . I n fact, black children have as much difficulty on nonverbal tests and on culture-fair tests as they have on verbal tests . Ethnic m inorities may have even less chance of entering educational institutions or of being considered for a job or a t raining program when culture-fair tests , rather than more traditional tests of cognitive abil ities , are used as the selection instrument s . The avai lable evidence suggests that culture-fair tests do not show greater val id ity for ethnic minorities than do more verbally loaded tests , sLlch as the Stanford-Binet : Form L-M or the W ISC- R . Culture-fair tests are not the panacea that some believed they would be for the assessment of ethnic m i norities . Probably no test can be c reated that will enti rely e l i m i ­ nate t h e influence o f learni ng a n d cultural experiences . The test content and materials , the language i n which the questions are phrased, the test directions, the categories for c lassifying the responses, the scoring criteria, and the val idity criteria are al l culture bound . In fact , all human experience is affected by the culture , from prenatal devel­ opment on. As Scarr ( 1 978) observe d : "Intell igence tests are not tests of intelligence in some abstract . culture-free way. They are measures of the abil ity to function intel lec­ tual ly by virtue of knowledge and skills in the culture of which they [are a] sample" ( p . 3 3 9 ) . Every test is cullurally loaded t o some extent . Thus it is important to distinguish between a test that is eulrurally loaded and one that is eulrurally biased (Jensen, 1974a ) . A I [ h ough , a s discussed earl ier i n the chapte r, there is l ittle evidence that individual intel l igence tests are culturally biased according to most defi nit ions of bias , tests do vary in their degree of cultural loading. Picture vocabu lary tests , such as the PPVT-R , are highly cul turally loaded because they use pictorial stimuli that call for speci fi c information associated w ith a given culture , such a s famil­ iarity with the language of the culture and objects represen­ tative of the l inguistic terms . On the other hand , matrix te ts (such as the Raven's Progressive Matrices), digit span memory tests, and maze tests are eulrurally reduced test s , because they are less dependent o n exposure t o spec ific language symbols . Even these types of tests have some degree of cultural load i ng , howeve r - they are neither culture fair nor culture free .

S U B C U LTURAL C O N S I D ERATIONS

Ass i m i lation into the majority culture may be d i fficult for members of ethnic m i nority groups because they are phys­ ical l y distingu ishable, use a foreign language , and/or have

580

cultural practices that are not always compatible w ith those of the majority culture . This section briefly reviews mate­ rial on black American, Hispanic-American, American Indian, and Asian-American groups . Many of the issues discussed also pertain to other ethnic mi nority groups . Although generalizations about subcultural groups are al­ ways suspect and cannot be appl ied in a blanket way to individual members of a subculture , the information in this section should help you in you r work with ethnic minority children and their fami l ies.

Black American Children Blacks in the United States tend to be bicultural , incorpo­ rating aspects of mainstream culture and black culture . The roots of black culture i n the United States incl ude not only an Africa heritage, but also a survival strategy devel­ oped by people requi red to deal daily w ith institutional racism and personal discrimi nation . Black cultural pat­ terns are also a means of dealing with the bicultural situation of the group. which requires people to learn to l ive in two worlds . to coordinate the elements of the two traditions borne by the parents, to learn the conventions of two cultures. and to manage the contradictions between the m . ( Young. 1974, p. 4 1 1 )

To b e black in t h e United States is to b e more than simply a person of black color. The black experience incorporates the collective experiences un ique to black Americans , encompassing racism, language , child-rearing pract ices, role expectat ions, soc ioeconomic statu , and kin s h i p bond s . To survive in the United States , blacks must b e able to size up potential ly d i fficult situat ions and deal with persons who may be prejud iced against them and perform discriminatory actions . They must be able to interpret confl icting sets of messages and respond to the m . Black oriented music , rel igion , and speech patterns are means by which black children are soc ial ized into black American culture . Black culture has minimal organizat ion . particularly with regard to status and authority positions (You ng, 1974) . It is a pragmatic culture . Rather than util izing specific rule- and value-based behavior. it uses mode . equences . and styles o f behavior that have highly varied content and val ues . It is a remarkably adaptive culture . The various Afro-American cultures that exist in the United States attempt to retain their self-identity and interpersonal cohe­ sion while constantly respond ing to the forces exerted by the domi nant culture .

Portrait of black culture. Although it i s extremely diffi­ cult to arrive at a portrait of black culture . Boykin's ( 1983 )

CHAPTER 1 9

ASSESSMENT OF ETH N I C M I N O RITY CHILDREN

formulat ion captures some important d i mensions. Black Americans have cultural t raditions that are based in part on those found in traditional African societie s . The A frican perspective emphasizes (a) spiritualism in the u ni verse; (b) harmony with nature and other people ; (c) fee l i n gs , expressive ness, and spontaneity ; and ( d ) d uty t o the group and group property ( see E x h i b i t 1 9 - 3 ) . The E u ro­ American perspective , in contrast , emphasizes (a) mate­ rial i s m ; ( b ) mastery ove r nature and individua l i s m ; ( c ) control of impulses, self-disc i pl i n e , a n d dispassionate reason ; and (d) individual rights and private property. A fro-Americans are then faced with a triple quandary. They are l i kely to be incompletely socialized in the Euro-American cul­ tural ethos. They typically develop a styl istic repertoire that arises out of their African heritage but is at odds with mainstream ideology. And finally, they are victimized by racial and economic oppression. ( Boykin , 1983 , p. 350)

The withering effect of racism may lead to major psych i ­ atric problems among black individual s . Family patterns developed in the inner city, which may be adaptive for occupational and economic survival , are l i kely unsuitable for social izing c h i ldren to achieve in the middle-class mainstream culture . This quandary is part of the psycho­ logical experience of many, if not a l l , black Americans.

Distrust of the environment.

Some black children and their parents have learned to distru st their environment. Distrust develops when they encou nter events that place them in an inferior pos ition or that s i ngle them out simply because they are black . The tendency to blame crimi nal behavior on blacks, high unemploy ment rates, and mini­ mal soc ial or educational opportu nities are a few reasons why black famil ies often come to distrust their environ­ ment . When dist rust exists, it may be d i fficult to conduct an assessment . Cooperative behavior may be replaced by hostil ity, and silence substituted for openness . Although you are l i kely to feel frustrated in such situation s , every effort must be made to establ ish rapport . Changes will l i kely occur in the attitudes of black cl ients when they are better able to control the ir environment and influence the way information about them is used . '' I'd l i ke to think that I'd have my problems no matter what I was . and sometimes I bel ieve it . But sometimes I think it's all because I'm black . I can't be sure . I j ust don't know. Maybe you coul d tell me what it really means to be black" ( Karon , 197 5 . p. 165 ) .

Black English . Some black child ren speak a variant of English that is referred to variously as black English,

SUBCULTURAL CONSIDERATIONS •

58 1

Exhibit 1 9·3 A Portrait of the African Perspective

I . Spirituality. Spirituality entails approaching life as though

2.

3.

4.

5.

its primary essence were vitalistic rather than mechanistic. I t means conducting oneself in a manner consistent with the possibility that the nonobservable and nonmaterial have governing powers in the everyday affairs of people. Permeating all sectors of one's life space is the conviction that greater powers than man are continuously at play. One strives to remain i n touch w it h the greater spiritual essences. HQ/mony. Rather than seeing oneselfas distinct from one's environmental surroundings, one sees oneself as, and in turn acts as though one were, inextricably l inked to one's surroundings. Rather than striving to partition one's life space into discrete elements, the aim is to blend them together into some kind of organi c , harmonious whole. The conviction is that what will happen will happen, mainly because it is supposed to or because it is best that it does. Rather than attempting to maximize one's effort or attain excellence with reg
( ca t )

9.

( t ra i n )

( s hout )

( co r rect ) -

.bUl� ( heave n )

ed )' ���. uca te

'-'

16.

17. 18.

1Vlll&t•.qal} ( ma t e r ia l )

� ( ru l n )

�rv"

( fa s h l o n )

&2L061-{> ( be l ieve )

....,'ION. \0 t:;!,WW W\'e. Y'OWH .

From Washington Star Syndicate. May 9 . 1 974. Copyright 1 97 4 , Universal Press Syndicate. All rights reserved.

625

ASSESSMENT OF ATTENTION-DEFICIT HYPERACTIVITY DISORDER

performance of hyperactive children on laboratory tasks involving attention, memory, and learning, which require sustained attention and are regulated by an examiner. The findings indicate that broad-gauged cogniti ve abilities ­ such as reasoning and problem-sol ving skill s - do not ap­ pear to be affected directly by stimulants, whereas more refined skills - such as attentional skills - are enhanced by drugs. Drugs appear to help hyperactive children plan and control their responding . Approximately 70 percent of hyperactive children re­ spond positively to stimulant medication (Satterfield , Cantwell, & Satterfield, 1974). Methylphenidate does not appear to be particularly useful in the treatment of hyper­ activity in preschoolers , although more research is needed about the effects of stimulants on preschool children (Campbe l l , 1976) . When hyperactive children are not re­ sponsive to stimulants , alternative forms of therapy should be sought and the medication discontinued . Other treat­ ment approaches include psychoactive drugs (for exam­ ple , chlorpromazine and imipramine), lithium carbonate , behavior mod i fication , low-sugar diets, megavitamins, avoidance of artificial food additives, exercise , and op­ tometric treatment . The hyperactive children who respond best to stimulant medication appear to be those who have low central ner­ vous system arousal levels as measured by EEGs, low evoked cortical response, or low skin conductance levels (Satterfield et aI . , 1974). Stimulant medications may re­ store both central nervous system arousal levels and inhib­ itory lev�ls �o normal , thereby pr(lvicling hyperact i ve chd­ dren with better controls and permitting a wider range of behaviors . The use of chemotherapy does nO{ mean that special services are not needed. Remediation is required for all children who have serious academic deficiencies . Both appropriate educational management and counseling of parents should be used in conj unction with drug management . The principal aim in treating hyperactive children and other children with similar problems is to help them to focus and sustain their attention and keep impulsive re­ sponding under control . A useful method is to teach chil­ dren to verbalize to themselves effective problem-solving strategies , such as planning ahead, stopping to think , and being careful . Self-verbalizations of these kinds help hy­ peractive children to bring their behavior under their own control and also make it possible for them to reinforce themselves for employing appropriate strategies . A struc­ tured and predictable environment with clear, consistent expectations and i mmediate feedback can also help in the treatment of hyperactivity.

ASSESSME

T O F ATT E N TI O N - D E F I C I T

HYPERACTIVITY DISOR D E R

Many of the conSiderations involved in the assessment of learning-disabled children hold for those with an attention­ deficit hyperactivity disorder. Hyperactive children per­ form better in an individual test situation than in a group test situation ( M inde, Weiss, & Mendelson, 1972 ) . In the individual test situation, examiners can be responsive to lapses in attention, making sure that they have the child's attention before administering the te'st questions. Group intelligence tests may underestimate the abil ity level of hyperactive children: consequently, IQs obtained by hy­ peractive children on group-administered tests must be interpreted cautiously. The major difficulties of hyperactive children, as we have seen, lie in their inability to focus, sustain, and organize attention and to inhibit impulsive responding. These difficulties are l ikely to be reflected in their perfor­ mance on some but not all psychological tests. Their scores on individual intelligence tests may be more vari­ able than those of normal children . Lower scores may be obtained on the Bender-Gestalt, Bruininks-Oseretsky Test of Motor Proficiency, Developmental Test of Visual Per­ ception , and Draw-A-Man . No part icular patterns of scores ( for example , Verbal-Performance discrepancies or lower abstract re asoning scores) on the WISC-R ' or on other intell igence tests have been found to be associated with hyperactivity, however. Tn (lddi[io'1 to tandard intell igence and ability tests, three procedures have been found to be especial ly useful in discrim inating hyperactive from normal children. These are the (a) Porteus Maze Test (Porteus, 1959), a measure of planning and organization ; (b) Jumbled Numbers Game , a measure of sustained attention and sequencing (numbers from I to 18, printed on one of three colored backgrounds, are randomly arranged on a large black cardboard - the child's task is to say the numbers in order and, with each count, name the color of the background on which the number is printed) ; and (c) Bobo punching dol l , a measure of frustration tolerance and aggressive behav ior. (See Homatidis & Konstantareas, 1981, for administration and scoring guidelines . ) An i mportant cue in evaluat ing hyperkinesis is the child's performance on cognitive tasks that require concen­ trated effort over a period of time . Because attentional factors probably permeate these tasks, children with hy­ perkinesis may do more poorly on them. Useful behavioral rating scales that can be completed by teachers and parents are shown in Chapter 1 5 .

626

CHAPTER 20

ASSESSMENT OF LEARNING DISABILITIES. DISORDERS. AND IMPAIRMENTS

CONDUCT D I SO R D E R : AN OVE RV I E W

According ! O the DSM-I IJ-R (American Psychiatric Asso­ ciation, 1987 , p. 5 3 ) , the essential feature of the conduct disorder classification "is a persistent pattern of conduct in which the basic rights of others and major age-appropriate societal norms or rules are violated . The behavior pattern typically is present in the home, at school , w ith peers, and in the community." The diagnostic criteria for the conduct disorder classification are shown in E xh ibit 20-7 . IQs of children with a conduct disorder (or delinquents) average about 8 points lower than those of nondelinquents; this relationship appears to be independent of the effects of socioeconomic status (Moffitt, Gabriell i , Mednick, & Schulsinger, 198 1 ) . Among delinquents , there are more of low average and fewer of superior intelligence . Although the reasons for these fi ndings are not clear, it may be that bright delinquents are more apt to escape detection, com­ mit fewer del inquent acts , or are less likely to be prose­ cuted . Overal l , studies indicate that bright deli nquents share the same criminological , educational , and social charac­ teristics as the great majority of other delinquents. How­ ever, they are less frequently encountered in the juvenile court system, are treated more leniently by the courts, and are more often presented to the courts as emotionally disturbed (Gath & Te nnent , 1972 ) . Their comparative rarity may be a result of differential immunity given to them because of higher social class and higher intel­ ligence . Interestingly, the school performance of h ighly intell igent del i nquents is less satis factory than that of nondel inquents . Evidence is inconclusive about how the intelligence level of delinquents is related to their responsiveness to treatment . Rate of recidivism does not appear to be related to intell igence level (Tennent & Gath, 1975 ) . Delinquents may have less adequate neuropsychological abilities than nondelinquents, as shown by their poor per­ formance on neuropsychological tests ( Berman & Siegal , 1976; Yeudal l , 1979) . An intriguing hypothesis is that some indiv iduals may become del inquent as a conse­ quence of consistent failures caused by deficits in adaptive abilities that are needed for success in our society.

ASSESSM E N T O F CON D UCT D I S O R D E R

A battery of psychological tests should be used in the assessment of children who may have a conduct disorder. The battery should be similar to the one used in the assessment of learning disabilities, including measures of

I'"

Exhibit 20·7 D iagnostic Criteria for Conduct Disorder in DSM·I II·R

A . A disturbance of conduct lasting at least six month s , during which a t least three o f the fol l owing have been present: ( I ) has stolen without confrontation of a victim on

more than one occasion (including forgery) (2) has run away from home overnight at least twice while living in parental or parental surrogate home (or once without returning) (3) often lies (other than to avoid physical or sexual abuse) (4) has deliberately engaged in fire-setting (5) is often truant from school (for older person , absent from work) (6) has broken into someone else's house, building , or car ( 7 ) has del iberately destroyed others' property (other than by fire-setting) (8) has been physically cruel to animals (9) has forced someone into sexual activity with him or her ( J 0) has used a weapon in more than one fight ( I I ) often initiates physical fights ( 1 2) has stolen with confrontation of a victim (e.g . , mugging , purse-snatching , extortion, armed robbery) ( 1 3) has been physically cruel to people Note: The above items are l isted in descending order of discriminating power based on data from a national field trial of the DSM-III-R criteria for Disruptive Behavior Disorders . B. If 1 8 or older, does not meet criteria for Antisocial Personality Disorder. Criteria for severity of Conduct Disorder: Mild: Few if any conduct problems in excess of those required to make the diagnosis, and conduct problems cause only minor harm to others . Moderate: Number of conduct problems and effect on others intermediate between "mild" and "severe . " Severe: Many conduct problems i n excess of those re­ quired to make the diagnosis, or conduct problems cause considerable harm to others, e . g . , serious physical injury to victims, extensive vandalism or theft, prolonged ab­ sence from home . Source: Reprinted with permission of the publisher from the Diagnostic and Statistical Manual of Mental Disorders (Third Edition - Revised) , 1 987 . p. 5 5 . Copyright © American Psychi­ atric Association .

627

NONSPECIFIC BEHAVIOR DISORDERS AN OVERV IEW

intell igence, pe rsonal ity, v i sual -motor abi l ity. and aca­ demic achievement. Interviews, observations, and behav­ ioral checklists are also valuable . The evidence is strong that deli nquents , on the average , obtain h igher Wechsler Performance IQs than Ver al JQs (for example, Hays , Solway, & Schrei ner, 1978: Hubble & Groff, 1981b; and Saccuzzo & Lewandowski , 1976 ) . Con­ sequently, the perceptual organization abil ities of del in­ quents are likely to be better developed than their verbal comprehension abilities. But the fact That a Verbal­ Peljormanr:e discrepancy is likely to appear in delinquent children does not mean that this pattern can be used as a diagnostic sign of delinquency. Many normal and excep­ tional children also show this form of discrepancy. The d iscrepancy may have no diagnostic relevance, especially when it is not statistically significant. It may simply be a reflection of poor education, reading disabil ity, bilingual­ ism, cognitive style, or some other factor. Perhaps the Performance > Verbal pattern is a reflection of the learn­ ing handicaps that are a relatively frequent concomitant of delinquency, rather than of delinquency itself. Administering the WISC-R provides an opportunity to obtain many insights about the ch ild from the pattern of

subtest scores anld fro m the content of the responses. Chapter 8 pro1iide:s e'xamples of how interpretations can be developed fro m ;a care ful analysis of a child's WISC-R performance , patrticularly for a child w ith a conduct disorder.

NONSPEC I F I C B E H AVIOR DISO R D E RS: AN OVERVIEW

This section briefly covers nonspecific behavior disorders, surveying some findings related to gender differences and intelligence . There is no category in the DSM-III-R corre­ sponding to nonspecific behavior disorders . Rather, this discussion reflects surveys of children w ith a variety of behavior problems. This material also i s pertinent to chil­ dren with attention-deficit hyperactivity disorder or a con­ duct disorder. In a comprehensive review of research on sex differ­ ences in childhood behavior d isorders, maies were found to outnumber fe males in every major category (Eme, 1979). Consequently, the male child appears to be more at risk for maladjustment than the female child. Culturally

"His Freudian therapist says he has Oedipal conflicts, his Rogerian therapist says he has trouble with self-actualization, his Eriksonian therapist says he has identity­ diffusion, and Courtesy of Ford Button .

I say he is a brat! "

CHAPTER 20

6 28

ASSESSMENT OF LEARNING DISABILITIES, DISORDERS, AND IMPAIRMENTS

determined role expectations and biological differences are two possible explanations for these findings . Beginning with adolescence and continuing i nto adulthood, a differ­ ent pattern emerges: females outnumber males in neurotic disorders and affective psychotic disorders, while males continue to outnumber females in personality and gender identity d isorders, and no sex differences emerge in schizophrenic disorders. In a study (Stone, 1981) of a public elementary school population of approximately 25,000 children, teachers were asked to complete a modified form of the Behavior Problem Checklist. Children at lower IQ levels were rated as having more behavior problems than children at the average and upper IQ levels . As Table 20-8 shows, fou rth­ grade children with IQs below 70 were rated as having five times as many conduct problems and four times as many personality problems as those with iQs above 130. Chil­ dren with IQs below 70 were rated as having three times as many conduct problems and over twice as many person­ ality problems as children with average IQs . Similar trends were noted for sixth-grade childre n .

euphoria, and dysphoria. Language difficulties suggestive of a behavior disorder include speech di fficulty, rambling, blocking, circumstantiality, clang associations , circum­ locution , confabulation, overelaboration, and self-refer­ ence (see Chapter 18) , Behaviors suggestive of anxiety include restlessness; apprehensiveness; i mpaired attention and concentration; bodily expressions indicating discom­ fort (for example, tics, nailbiting , fidgeting, and cough­ ing) ; difficulty in finding words; impulsively blurting out unfinished, unchecked , or i nappropriate replies; and fum­ bling about for adequate formulations. The WISC-R is a reliable and stable instrument for eva luating c h i l dren w ith behavior d isorders (Dean , 1977d) . Unfortunately, there are no WISC-R patterns that can distinguish reliably between children with various behavior disorders, although children with some behavior disorders may exhibit greater variability of scores (see , for example , Clarizio & Veres, 1983 ; Hale & Landino, 1981; Hamm & Evans, 197 8 ; Morris, Evans, & Pearson , 1978 ) , Factor analytic studies o f the WISC-R indicate that there are no qualitative differences in the structure of intellectual abil ities of normal children and those with behavior disor­ ders (DeHorn & Kl inge , 1978 ; Petersen & Hart, 1979 ) .

ASSES S M E N T O F N O N S P EC I F I C B E HAVIOR D I S OR D E RS

During an examination children may exhibit attitudes sug­ gestive of a behavior disorder, including irritability and suspiciousness, restlessness, lack of spontaneity, variable mood, apathy, cynicism, regarding the test as "kid stuff, " Table 20-8 Mean N umber of Conduct and Personality Problems by I ntellectual Level

Fourth-graders IQ category Below 70 70-79 80-89 90-99 1 00- 1 09 1 1 0- 1 1 9 1 20- 1 29 1 30 and up

Sixth-graders

N conduct N personality N conduct N personality problems problems problems problems 10.6 1 1 .0 9.5 6.0 3.7 2.2 2.1 2.0

6.2 6.6 5.7 3.9 2.6 1 .7 1 .5 1 .4

13.9 1 2 .4 10.5 7.6 4.4 3.4 3.2 2.6

7.3 7.5 6.4 5.0 3.2 2.5 2.3 1.7

Source: Reprinted with permission of the publisher and author from F. B . Stone. "Behavior Problems o f Elementary-School Childre n . " in Journal of Abnormal Child Psychology, 9, p. 4 1 5 , copyright 1 9 8 1 by Plenum

Publishing Corporation .

P E RVASIVE D E V E L O P M ENTAL D I S O R D E RS : A N OVERVI E W

Pervasive developmental disorders is a broad classification of severe and pervas i ve di sorders affecting children's social skills and language , attention, perception , reality testing, and motor activ ity. Multiple areas of functioning are affected . The children so diagnosed display severe qual itative abnormalities that are not normal for any devel­ opmental stage . There are two classifications within this DSM-III-R category : autistic disorder and pervasive de­ velopmental disorder not otherwise specified . The diag­ nostic criteria for the autistic disorder classification are shown in Exhibit 20- 8 . (Schizophrenia , which is dis­ tingu ished from the autistic disorder by the presence of hallucinations and del usions, is not discussed in the text because it is a diagnosis given primarily to older adoles­ cents and adults . ) Autistic Disorder Children with an autistic disorder (formally classified as infantile autism) have a severe behavioral disorder that usually appears before 3 years of age . Autistic children may show poor response to sensory stimuli such as sound or light , may not recognize their parents, and may lack

'" Exhibit 20-8 Diagnostic Criteria for Autistic Disorder in DSM-I I I-R

At least eight of the following sixteen items are present, these to include at least two items from A, one from B, and one from C . Note: Consider a criterion t o be met only i f the behavior i s abnormal for the person's developmental leve l .

A . Qualitative impainnent in reciprocal social interaction as manifested by the following : (The examples within parentheses are arranged so that those first mentioned are more likely to apply to younger or more handicapped, and the later ones, to older or less handicapped , persons with this disorder. ) ( 1 ) marked lack of awareness of the existence or feel ings of others (e . g . , treats a person as i f he or she were a piece of furniture ; does not notice another person's distress; apparently has no concept of the need of others for privacy) (2) no or abnormal seeking of comfort at times of distress (e.g . , does not come for comfort even when ill. hurt, or tired; seeks comfort in a stereotyped way, e. g . , sa�s "cheese, cheese, cheese" whenever hun) (3) no or impaired i mitation (e . g . , does not wave bye­ bye; does not copy mother's domestic activities; me­ chanical imitation of others' actions out of context) (4) no or abnormal social play (e . g . , does not actively participate in simple games; prefers solitary play activities; involves other children in play only as "mechanical aids") (5) gross impairment in ability to make peer friendships (e. g . , no interest in making peer friendsh ipS ; despile interest in making friends, demonstrates lack of un­ derstanding of conventions of social interaction, for example, reads phone book to uninterested peer) B . Qualitative impairment in verbal and nonverbal commu­ nication, and in imaginative activity, as manifested by the following: (The numbered items are arranged so that those first listed are more likely to apply to younger or more handicapped , and the later ones, to older or less handicapped, persons with this disorder. ) ( 1) no mode of communication, such as communicative

babbling, facial expression, gesture, mime, or spo­ ken language (2) markedly abnormal nonverbal communication, as in the use of eye-to-eye gaze, facial expression, body posture, or gestures to initiate or modulate social interaction (e . g . , does not anticipate being held , stiff­ ens when held, does not look at the person or smile when making a social approach , does not greet par­ ents or visitors, has a fixed stare in social situations)

(3) absence of imagin .ative activity, such as playacting of adult roles , fantasy characters, or animals; lack of interest in stories about imaginary events (4) marked abnormal ities in the production of speech, including volume, pitch, stress, rate, rhythm , and intonation (e . g . , monotonous tone, questionlike mel ­ ody, or high pitch) (5) marked abnormalities in the form or content of speech , including stereotyped and repetitive use of speech (e.g . , immediate echolalia or mechanical rep­ etition of television commercial) ; use of "you" when "I" is meant (e. g . , using "You want cookie?" to mean "I want a cookie") ; idiosyncratic use of words or phrases (e. g . , "Go on green riding" to mean "r want to go on the swing") ; or frequent irrelevant remarks (e . g . , starts talking about train schedules during a conversation about sports) (6) marked impairment in the ability to initiate or sustain a conversation with others, despite adequate speech (e . g . , indulging in lengthy monologues on one sub­ ject regardless of interjections from others) C. Markedly restricted repertoire of activities and interests, as manifested by the following: ( l ) stereotyped body movements, e . g . , hand-flicking or -twisting, spinning, head-banging, complex whole­ body movements · (2) persistent preoccupation with parts of objects (e . g . , sniffing or smell ing objects, repetitive feeling of tex­ ture of materials , spinning wheels of toy cars) or at,achme;1t to ur.usual objects (e . g . , insi�ts on carry­ ing around a piece of string) (3) marked distress over changes in trivial aspects of environment, c . g . , when a vase is moved from usual position (4) unreasonable insistence on following routines in pre­ cise detail , e . g . , insisting that exactly the same route al ways be followed when shopping (5) markedly restricted range of interest and a preoc­ cupation with one narrow interest, e . g . , interested only in lining up objects, in amassing facts about meteorology, or in pretending to be a fantasy char­ acter. D. Onset during infancy or childhood.

Specify if childhood onset (after 36 months of age). SOl/rce. Reprinted with pe rmission o f the publisher from the Diag­ I/Ostic and Statistical Manual of Mental Disorders (Third Edition ­ Revised) , 1 987. pp. 38-39 . Copyright © American Psychiatric Association.

629

630

CHAPTER 20

ASSESSMENT OF LEARNING DISABILITIES, DISORDERS, AND IMPAIRME NTS

interest in the environment. Some children appear dis­ tressed for long periods of t i me and cry continuously ; others appear apathetic. Feeding and sleeping may be erratic and unpredictable. I n t i me , the child may show obsessional featu res, aloofness and lack of interest in people, language disturbances , and retarded i ntellectual functioning . Autistic children may have remarkable v isual-spatial ski l l s . They may notice minute details, react to very small changes in their surroundings, and spot missing objects. These skills, which enable them to locate things when others have given up searchi n g , may be present even in a u t i stic c h i ldren who are retarded i n other areas of functioning. The autistic disorder can be behav iorally defined as a specific syndrome that is manifested at b i rth or shortly thereafter. The various symptoms associated w ith an au­ tistic disorder appear to be expressive of an u nderlyi ng neuropathophysiological process that affects developmen­ tal rate, perception, language , cognition , intell igence, and social relationships, Prognosis is guarded, as almost all children manifest severe symptomatology throughout their l ives . In the foll owing case, w h ich traces the course of language , cognitive, and social development in an autistic c h i l d , we find the simultaneous presence of remarkable talents and profound deficit s . James was the third o f four children, born following a n uncompl i­ cated pregnancy and labor. His health during the fi rst 3 months of life was good, but shortly thereafter his mother expressed con­ cern because of his sensitivities to light and sound, his failure to make an anticipatory response to being picked up, his fluctuating moods between inconsolable crying and extreme passiveness, and his failure to look at her when she fed him . She reported that he preferred lying in his crib, staring at the mobile, to being held or played with . Because his motor m ilestones appeared at the appropriate times , James' pediatrician reassured his mother that his development was fine. However, by age 16 months, James had not begun to babble or say single words, and spent most of his time in a corner repetitively moving toy cars back and forth. At 20 months, other symptoms emerged: he developed unusual hand movements and body postures; his obl iviousness to people in­ creased; he reacted to even the most subtle interruption in his routine or other changes in the world with extreme disorganiza­ tion and panic; he developed a fascination with l ight switches and with studying tiny bits of paper and twigs. At 4 years, James had not yet begun to speak socially to others, but could identify by name many numbers and all of the letters of the alphabet. He was able to execute the most graceful maneu­ vers. spinning in circles about a room without touching a piece of furniture. At other times, he appeared clumsy and uncoordi-

nated. He persisted in lining up objects in the most complex patterns, but could never use objects appropriately. H is parents complained about how difficult it was to buy bi rthday and Ch rist­ mas gifts that could replace the tiny bits of paper and pieces of string he preferred . At about the age of 4 '/2 years, he began to echo long and compl icated sentences, some of which his mother reported he may have heard days or even weeks before. He was able to complete puzzles designed for 8- and 9-year-olds quickly, but was unable to reproduce a l i ne or circl e . At about age 5 , James made h i s fi rst spontaneous statement. His mother reported that he had been looking at the sky and said , "It looks like a flower." He did not speak again for 8 months, but then began talking in full sentences . Most often, the content was concerned w ith numbers - he would read encyclopedias and re­ port to whomever would l isten that a certain river was 1,000 miles long and had 200 tributaries. When he met strangers, he mechan­ ically introduced himself without ever establishing eye contact, and then rushed on to ask what the person's birthday, anniversary, and social security number were, often appearing not to pause long enough to get the answers. Years later, upon remeeting the person, he was able to recite back these facts . James was remarkably talented musically, and could sing the lyrics to popular tunes and TV commercials . While he could read well , he was unable to abstract from written information, to draw inferences or make conclusions. At age 9, he remained socially distant, although he had learned many of the rules for social situations. He received superior scores on the WISC Object Assembly, Picture Completion, and Block Design subtests, but very low scores on the Comprehension, Similarities, and Infor­ mation subtests. (Caparulo & Cohen, 1977 , pp. 623-624, with changes in notation)

Etiology of autistic disorder.

Various explanations have been offered to account for the etiology of autistic disorder. One v iew is that autistic disorder is a disorder of the central nervous system that manifests itself in impaired com­ prehension and use of language . The cause of the cognitive deficit , however, remains u nknow n . Bra i n damage may be present in some cases , whereas in others developmental or genetic factors may be i mpl icated . Autism may also in­ volve a dysfunction of the complex c i rcuitry providing the central connections of the vestibular system to the cere­ bellum and the brain ste m . This proposal attempts to account for the strange sensorimotor behavior observed in autistic children (for example , spontaneous spinning and fl icking of objects, flapping and osc i l lating of extremities , and whirling and rocking of the body) . There is substantial evidence that autism is probably not related to the psychological characteristics of autistic chil­ dren's parents ( Koege l , Schreibman , O'Neil l , & Burke , 1983 ) . When stress is found in parents of autistic children,

63 1

PERVASIVE DEVELOPMENTAL DISORD ERS: AN OVERVIEW

it is l i kely to be highly situational . Findi ngs such as these support a neurophysiological and biolog ical basis for the disorder.

Intelligence and autistic disorder. I n the past . I Qs ob­ tained from auti stic children were often considered to be invalid . The hope was that with the right treatment . intel­ lectual abil ity would develop to a normal level . Unfortu­ nately, h owever, therapy w ith auti stic chi ldren has not resulted in significantly improved levels of intellectual performance . Research on autistic children's intel lectual functioning points to several i mportant findings.

1 . As many as three-fourths of autistic children obtain IQs that are in the mentally retarded range of functioning ( D e Myer, 1 9 7 6 ; D e M ye r, Barton , & N o r t on , 1 972 : H ingtgen & Bryso n , 1 97 2 ; Rutter, 1 974 , 1 978; Rutter & Bartak, 1 97 1 ) . 2 . The IQs obtained by autistic c h ildren have the same properties as do those obtained by other ch ildren (DeMyer. Barton , Alpern , K imberlin, A l l e n , Yang, & Steele, 1 974; Lotter, 1 967 ; Rutter & Lockyer, 1 967) . Thus, for exam­ ple , (a) IQs show moderate stabil ity throughout childhood and adolescence (test-retest correlations for periods of 2 to 1 5 years ranged from . 63 to .90) , especially if the children are tested after 5 years of age , and (b) IQs are a reasonable predictor of later educational attainment s . 3 . T h e I Q s of autistic children fail t e change markedly even after their social responsiveness g reatl y inl p roves Poor mot i vation , consequently, does not appear to account for autistic children's below-average performance on intel ­ l igence tests . 4 . Initially untestable autistic children later have been found to perform in a manner similar to that of severely retarded children ( Bartak & Rutter, 1 97 1 , 1 91 3 ; DeMyer et a! . , 1 974; Gittelman & Birc h , 1 967 ; Lockyer & Rutter. 1 969 ; Rutter & Bartak , 1 97 3 ) . In add ition, those who appear to be u ntestable may be testable when given items representing sufficiently low mental age levels. 5. Autistic children who have adequate conversational speech or adequate social relationships obtain higher IQs than do other autistic children (DeMyer et a I . , 1 974) . . 6 . The fact that autistic children have relatively good Visual-spatial and memory abilities but poor sequencing and language s k i l l s suggests a specific cognitive defect involving the use of language. The above fi ndi ngs strongly suggest that autistic c h i l ­ dre n with l ow IQs function l i ke other children with low

IQs . Therefore . do nOlt di Sol iss the IQs obtained by autistic children as being a refSult of some easi ly reversible tempo­ rary impairment . and progulOsis of aut istic disorder. Autistic children can be h e l pe,d i f the fol l ow ing guide l i nes are kept in mind ( Gallagher &{ W iegeri n k , 1 976, p. 329):

1reatment

I.

Aut istic child rem are e:ducable .

2 . Their unique le:arn ing characteristics are related to basic cognitive deficins ilil ir:lformation processing . 3 . Such deficits cam �e compensated for, in part , by carefully structured edutor o rga­

n i zat ion . Lind n o n verbal seque n c i n g . Lcft h e m i sp h e rc processi ng h a s becn ch,mtct er ized a s an a l Y l i c . scque n t i a l . se r i a l . and d i llere n t i a l . w h c reas r i g h t hem i sp h c re proce s s i n g i s con s i d e red h o i i st i c . gestal t - I i ke . paral l e l . a n d i n t c g ra t i ve . For s i m p l e t a s k s . nonl'erbal s t i m ­ u l i c a n b e p roce s sed hol i s t i c a l l y b y e i t hc r h e m i . phere . The right h e m i s p h e rc i s c l e a r l y i n fe r i o r to t hc I c ft i n cxpre s s i ve

fu nct ions of speech a n d w r i t i n g . but is Icss d e f i c i e n t i n l a n g u age c o m p re h e n s ion . A m o re det a i l ed l i s t i n g of fu n c ­ t ions att r i buted t o t he l e ft and r i g h t c e re b ra l h e m i spheres appears in Ta b l c

22-2 .

Some of t hc s pe c i a l i ze d fu n c t i o n s of t he ccre b ral h e m i ­

sphcrcs i n c l uue t h e fo l l ow i n g . T h e .fi'OI//(/1

lobes are

asso­

c i ated w i t h p l a n il i n g . i n i t ia t i on . a n d m od u l a : ion of hchav­ ior and e x p re s s i ve verbal fl ue n c y : t h e rl!lIlpoI'(I1 IO/)l!s w i t h a u d i to ry pcrce p t i o n . a u d i t o r y c o m p rc h e n s ion . I ca rn i n gi

memor l'. a n d c ross-modal i n t e g r a t i o n : t hc poricwl lobcs

Table 22-2 Brai n F u nctions Attributed to the Left and Right Cerebral Hemispheres L,1; ( 'ereiJm/ helll isphere E xpressive speech Recept ive l anguage Language ( genera l ) Complex mOlOr fu nct ions V i g i lance Paired-associate l e a r n i n g Lia ison t o consciousness I deat ion Conceplllal s i m i l a ri t ies Temporal analysis Analysis of dcta i l A rithmctic W riting Calculat ion F i nger naming R ight-left orientat ion Sequential proces s i ng

SOllrce:

Adapted frolll Berent

(}cci,-)iw'l lohes

w i t h v i s u a l percept ion a n d t h e

sema n t i c L't \ll nol.a t i o n s o f v i s ual objects ( t hat i s . verbal I

e rbal encod i n gs g i ve n to v i sual object s ) .

De\,elop ment o f L a t e r a l i z a t ion

The devc l opmcnrr o f l ateral i z a t ion appears to be a g radual

procc s s . CI ili e nc eu in psychomotor devel opment by the

re p l accment o f h i l atera l move m e n t s with u n i lateral move­

ment s . The p roc ess beg i n s i n i n fa ncy and may cont i n ue

t h roughout t h e d ve l o p m e n t a l pe r iod . The deg ree of later­

a l i zat ion appcar' to be re l at e d to t h e s k i l l level of t h e task be i ng p ' rf orm c d . \.v i t h manual act i v it ies req u i r i ng h ig h

s k i l l bec o mi n g i n c re a s i n g l y l ateral ized and t hose requ i r i ng

l ow s k i l l rem a i n i ng l e s s l a t e r a l i zed ( M i l l er,

1982 ) . S i m i ­

l a r l y. l o \.\ e r le\c l s of l a n g u a g e deve l opme n t . such as i n vo l ­ u nt a ry pr i mit t " e speech response s . are p robab l y orga n i zed b i latera l l y : i n termcd iate l e ve l s o f l an g uage deve l o p m e n t . such as comprehens ion , are p robably pa rt i a l l y l atera l i zed ; and t he h i ghest l eve l s of l a nguag e , s u c h as propo s i t ional s peec h . arc l i ke l y s pec i a l i zed i n t h e l e ft hemisphere . A rev icll' o f resea rch ( H a h n ,

1987 ) cove r i n g stud ie s of

d i c h o t i c l i st l: n i l l g . t a c h i s t o s c o p i c v i ew i n g . e l ec t roe n ­ cephai llgraphy. hapt i c ide n t i ficat i o n . and somatosensory d i sc r i m in a t ion i l l normal c h i l dre n from i n fancy t h rough c h i ld hood i nd icated t h a t l i n g u i s t i c fu nctions a re l oc a l ized in thc l c i't hemisphere at b i rt h fo r c h i l d re n of both sexes .

Fu nctions l atl'ra l i zed i n t he r i g h t h e m i sphere . however, are

less st ra i g h t forward - ce r t a i n ab i l i t ie s a re l atera l i zed at b i rt h . where,!.., ( Ithers become lateral ized with age . For

Spat ial orientat ion S i m p l e language comprcllcn,ion Nonverbal ideat ion Pict u re and pallern sense P�rf0rnnnce- l i �e fu nct ions Spat i a l i ntegra t ion C reat i ve assoc iat i ve t h i n k i ng Facial recog n it ion Sound ( e n v i I'Onmental ) recogn i t ion Nonverbal p a i red-associate t h i n k ing

exam p l e . I he r i g l l t h e m i s ph e re appears to be spec i a l ized at

b i rt h i n pr( )ee ' s i n g no n l i n g u i s t i c s t i m u l i . w h e reas deve l op­ mental changes l i ke l y occ u r in the right he m i sphere's a b i l ­ i t y t o proces,> spat i a l i n fo rmation . Researc h on cerebral latera l i zatio n i n c h i l d h ood is an ongo i ng ac t i v ity.

C o m m ent on Latera l izat ion Lateral spec i a l i z a t ion c a n n o t a l ways be c l e a rl y e s t a b ­ l i shed . F o r e x a m p l e . t h e role of l ateral i zation i s l e s s c l e a r for v i s uoeons t ruct ional s k i l l s t han fo r l a n g u age s k i l l s . O n me l110ry ta�ks t h at a l l ow s t i m u l i t o b e e ncoded e it h e r or

Tac t i l e percept ion

verba l l y

Gestalt perccpt ion Logograph ic ( pictograph) proces s i n g I n t u i t i ve problem sol v i ng H umorous t h i nk i ng S i m u l t a neous processing

t h at arc ea s i l y recogn i z a b l e a n d have fa m i l i a r n a me s ) , the

( 1 98 1 ) .

perccplU a l l y ( s u c h as t hose i n vol v i n g fi g u res

s pec i fi c process used by a n i nd i v i d u a l often cann ot be spec i fied . and there fore i t i s d i ffi c u l t t o make i n ferences about II'hethe r the t a s k s i n v o l ve cerebral latera l i za t i o n . A l t h ough the ev idence for l a t e ral i zed cerebral spec i a l iza­ t ion i � � t rong . young c h i l d re n s h ow l a ng uage d i sorders even when they have r i g h t - s i d ed l e s i o n s ( H ecae n ,

1983 ) .

Perhaps there i s less c e rebral spec i a l i za t ion w it h respect to

692 language in young chi ldren than i n adults. Add itionally. even though language and speech appear to be lateral ized at birth , there are complex changes in the direction and strength of hem ispheric spec ial ization as development pro­ ceeds ( Lewkowicz & Tu rkew itz , 1982 ) . Once language develops , the left hemisphere appears to be domi nant for language for most individual s; only some potential for language functions is present in the right hemisphere . Language disorders i n c h i l d re n , as i n adults . a re assoc i ated more frequently w i t h l e ft-hem isphere damage than with right-hemisphere damage . Al though there may be some recovery of function mediated by the right hem isphere , deficits may continue to exist , part icu­ larly in syntactic , read ing, and writing skills ( Moscov itc h , 198 1 ) . It appears that a primary function of the right h e m i ­ sphere is t o act a s a concrete spatial synthesizer, al low i n g i n formation t o b e perceived a s a meaningful gestalt . " I t does not have t h e analytic s k i l l s that are required t o process l i nguistic input phonet ically or to decode complex syntax" ( Moscov itch , 198 1 , p. 47) . The right hem isphere doe s , however, support com mun ication . Impairments of the right hemisphere may interfere with communication , con­ ceptual ski l l s , memory, and other cogn itive functions , espec ially when a task requires the integration of multiple sources of i n formation or comprehension of nonliteral language , such as metaphor or sarcasm . Additionally, the right hemisphere is thought to be i m portant in prosody, or the affective aspect of language . I t gives emotional ex­ pression to speech and may be i m portant for evaluating the e motional content of the speech of others . APHASIA I N C H I LDREN

C h ildhood aphasia is a central nervous system dysfunction man i fested through disorders i n the percept ion, produc­ t ion, and symbol ic util ization of language. Three sub­ groups of aphasia are congenital aphasia , a language dys­ function marked by an almost complete fai lure to acq uire l a nguage ; developmental aphasia, a less pervasive cog­ n itive and developmental impairment i n which language is late i n onset and fails to develop ful l y ; and a cq u i red ap­ hasia, a language dysfunction resul t i ng from brain i njury fol lowing normal language developmen t . The subgroups differ from one another mainly in the severity and age of onset of the same basic language dysfunct ion .

Major Forms of Aphasia Aphasia may invol ve expressive components , receptive components , or bot h . I ndividuals w it h receptive aphasic

CHAPTER 2 2

ASSESSMENT OF BRAIN DAMAG E

distu rbances also have ex pre ssive deficits. but those with expressive distu rbances may not necessa rily have recep­ tive problems. Expressil'e aphasia is defi ned as impaired abil ity to use spoken or written languag e . A subcategory of expressive aphasia is oxraphia . the loss or impairment of the abil ity to express language i n written or printed form . Recepril'e aphasia i s defi ned as im pa ired abil ity to under­ stand spoken or written language . Audirur." aphasia refers to the loss or impai rment of the abil ity to comprehend the meaning of spoken word s , and alexia refers to the loss or impairment of the abil ity to comprehend written or printed language despite adequate vision and intell igence . Brain inju ry may also manifest itse l f in two other sym­ bolic deficits broadly classi fied as aphasic disturbances : agnosia, which i nvolves failure t o recogn ize o r understand the signi ficance of sensory st imul i . or defects of imagina­ tion, and ap raxia , which is a disturbance of the execution of learned movements that is not due to motor or sensory defect s . poor comprehension . or i ntel lectual deteriora­ tio n . Procedures used to evaluate aphasia. agnosia, and apraxia are illustrated in Table 22-3 and Table 2 2 -6 .

Differential Effects o f Aphasia in You nger and Older Children In children under the age of 10, aphasia is usually charac­ terized by a severe reduction in spontaneous speech (Satz & Bul lard-Bates, 198 1 ) . The pattern is generally in itial mu tism , fol l owed by l i m ited speech , hes itat ions, dys­ arthria ( i mperfect articulation of speech caused by distur­ bances of muscular control resulting from damage to the central or peripheral nervous syste m ) . and impoverished commun ications . Paraphasias (substitutions of incorrect and uni ntended words or sounds for correct ones chair for table , jail for hospiflll, lrab/e for rahle) are rare . Errors may i nclude omiss ion of crucial word s . errors in word choice , and disturbances of word order. C h i ldren may be unable to comprehend purely verbal commands unless they are accompanied by gesture s . Young aphasic children are usually alert , attent ive , and intent on communicating their thoughts and reaction s . After t h e age of 1 0 , a more mi xed pattern resembling that of adults emerges . with an increased frequency of di sor­ ders invol ving speech . The prognos is is more favorable for chi ldhood aphas ia than for adult aphas ia. with spon­ taneous recovery occu rring in the majority of cases . Re­ covery may be i ncomplete, however; serious cognitive and academic sequelae may be found . as we ll as some lan­ guage deficits . The fol lowing case describes the course of recovery i n -

PROGNOSIS FOR BRAIN-DAMAGED

CHILDRE

\J

693

Table 22-3 I l lustration of Procedu res Used in T est i ng for Aphasia, Agnosia, and Aprax i a! Pr, l('erilll"e'

A rea Aphasia

V i s u a l - verbal comprehension

Motor speech

A"k child to read a sentence from the nc�\'spapcr and explain its m ea n i n g .

I f child is

unable to ta l k . print i n s t ruct ions and dc:tcrmil1le whether the c h i l d c a n carry them

out . . A,k c h i l d to i m itate several d i fferent .. nunds and phrases : "Ia- I a . " "me-me , " . t h i s is a good book . " and others of i ncreasing d i liicul ty. Note abnormal word usage i n

con ve rsat iO!l .

Auto matic speech

Ask c h i l d to repeat one or two series that the ch Il d has l earned in the pas t , such as the

Vol it ional speech

Determ i n e whether c h i l d c a n answer questions r e l evant ly.

Writing

Have c h i l d w rite ( a ) his or her name and addres - ,

days of the week or the months of the .,·car.

(b)

a s i mple sent e nce , ( c ) one word ,

w it h eyes c l osed , and then ( d ) the name of an object t h a t is s h o w n to the c h i l d . with eyes opc n .

Agnosia Sound recogn it ion Aud itary-verbal comprehension Recog n i t ion of body parts a n d sidedness

Ask child

to

ide n t i fy fam i l iar s(lu nds, with cves c l osed .

Determine whether child can answc, q Licstions a n d ca rry Ollt i ns t ru c t ions .

Determine whether c h i l d k nows l e ft from right and recogn i zes body parts .

V isual object recog n it ion

A s k c h i l d to ide n t i fy fam i l iar objects such as a pe n or w ri stwatch .

Performance of s k i l led motor acts

Determ i n e whether c h i l d can com p lete mOlOr ac t s , such as d r i n k i n g from a c u p ,

Apraxia

c l o s i n g a safety p i n . o r u s i ng COllllllon 100 1 s .

an aphasic patient ( Lev i n , 1981 . p . 444 , with changes in notation ) : A 1 7-year-old student was i nvolved i n a motor ve h i c l e accident i n w h i c h s h e received a c l osed head i nj u ry. A l though s h e obeyed commands on the day of a d m i ssion , d e l ayed n e u ro l og ical dete r i ­ orat ion w a s reflected by t h e post-a d m i s s i o n development of a right hem i pare s i s and progressive aphas i a . T h ree days postinj u ry a part ial l e ft temporal lobectomy was performed w i t h evacuation of an i n t racerebral hematoma . The pat ient re m a i ned confused for a month after i nj u ry. but e x h i b ited gradual i mprovement of recep­

t i ve language. A d m i n istration of a n aphas i a battery 2 months

after i nj u ry d i sclosed fi n d i ngs cons istent with an a n o m i c apha s i a .

There were freq u e n t errors of c i rcu m l o c u t ion ( fo r e x a m p l e . an

islalld was

defined as

described as

a place where -,-au fish a n d a friallgle was fhe fhillg \"Ii/,{ use lI'h en -" ou pia\' pool ) . and selllantic

appro x i mat ion ( fo r exampl e . the t ru n k o f an e l ephant was de­

PROGNOSIS FOR BRA I N-DA M A G E D C H I LDREN

The prognosis for children who have experienced brain inju ry depe n d s on the age at wh ich the child incurs the injury, the location and extent of the brain damage , an,J the occurrence of post-traumatic compl ications, such as sei­ zures. Certa i n types of difficulties resulting from brain injury, such as hemiplegias ( paralysis on one side of the body ) and v isual field defect s , may show l ittle or no i m ­ provement after many years , whereas sensorimotor func­ tions tend to be recovered earl ier and more completely. Mental processes may he recovered gradual l y even years after the i nitial i nsult . These processes are of a more complex order than sensory or motor processes and are not as ci rcumsc ribed or anatom ica l l y restricted . Recoverabil­ ill" of illlelleClual funCTions is less certain wiTh se vere

scribed as a nose ) . Appro x i mately one year a ft e r the i nj u ry the pat ient 's recovery o f l a nguage was complete . with the exception

Trallma. earlr dall1age, jronto-Temporal area dall1age , and

of a subtle residual anomic d i s t u rb ance that was evident only

posT -Traumatic epilepsy.

u nd e r fo rmal detailed test i n g conditions .

Rate of recovery is most rapid i n the months immediately fol lowing injury. The extent of

CHAPTE R 2 2

694

ASSESSMENT O F BRAIN DAMAGE \\

i ntellectual impairment general l y is related to the severity of the head i nj u ry.

and nonverbal or v i suospatial d i sorders associated right hemi sphere lesion s .

Differential Effects of Brain Damage i n Young

Cerebral Plasticity

C h ildre n and Adu lts

Cerebral pl asticity refers to the abi I ity of one part or the brain to take over fu nctions Impaired by lesions in another part of the b ra i n . This process operates primarily between birth and 4 or 5 years of age. Changes in the functional organ ization of the central nervous system (or reorgan i za­ tion ) may e nable the child to recover fu nction s . The as­ sumption that the bra i n possesses such plastic ity has been questioned . however. A re the e ffe cts or early brain i n j u r ies more or less severe than t hose of later brain i n j u ries') Although there are no simple answe rs to this quest i o n . there is increasi n g ev idence that young chi ldren arc n o less , a n d perhaps even more . vul nerable a t ti mes than a re adults to the delete rious e ffects of diffuse bra i n injury on memory and cog nition ( Bruno. 1986: Lev i n . Eisenbe rg , W igg, & Kobayash i . 1982 ) One possible explanat ion is that it is during its phase of most rapid development that the i mmature b rain is most v u l nerable to damage from such conditions as encepha l i t i s , men i ngit i s . and therapeutic i r­ radiation of the brai n . Consequently, inj u ries susta ined after a long period of normal deve lopment may in terfere less with intellectual function i ng than do those that occur early in development ( Eiser. 1981 ) . Ove ra l l , stud ies o f pat ients with hem ispherectomies ( re­ section of a n entire cerebral hemisphere ) i n i n fancy, child­ hood , and adult hood , as well as re lated researc h . ind icate that there is insuffi c ient evidence to support the assumpt ion of plastic ity (St. lames-Roberts, 198 1 ) . In other words, there is l ittle reason to bel ieve that young children recover better than older childre n . Instead , recovery is variable with in each age g roup. Perhaps even more important than age or hemi sphere damaged are variables associated w ith the child's congenital makeup and experiential history and the type of bra i n damage . Research fi n d i ngs support nei­ ther a plastic ity hypothesis nor a c ritical pe riod hypothesis ( that certa i n functions can be developed only at spec ific times in developmen t ) . Changes in cogn itive development a fter bra i n damage may best be v iewed as invol v i ng cumu lative i nteractions among etiological . recovery pe­ riod , and experiential variables. Questions of compensation and pl astic ity make prog­ nosis much more precarious w ith c h i ldre n than with adults, as does the fact that many areas tested in adults­ such as language and speech - cannot be tested in young childre n who have developed only rud i mentary ski l l s . AcI-

The effects of brain damage in young ch ildren are different from those i n adults because the young chi ld's bra i n is st i l l developing . W he n an adult sustains bra i n i njury, there may be a l oss or d i ssolution of previously acqu ired functions , man i fested in impai rment of language , memory, social relations, or general i ntelligence . I n contrast , when a child susta i ns b rain inj u ry, there may be i nterfe re nce w ith devel­ opment rather than a striking l oss of fu nctio n . I f the inter­ fe rence is globa l , mental retardation may be the resu l t : i f it is region-spec i fi c , spec ific difficulties may resul t , such as w ith speech or with recognition of shapes. Acq u i red brain damage tends to produce less spec i fic dysfu nctions i n young children than in adults . Because cortical spec i a l i zation has not yet been completed , damage to the immature brain is I i kely to affect the development of the whole brain rather than produce localized abnor­ mal ities . I n young children b ra i n damage may have more than a simple depressing effect on the i r abil ity structures; it may alter the basic pattern of abil ity devel opme n t . Large unilateral i nj u ries in i n fants tend to produce a more wide­ spread deficit in intellectual functions than do similar inju­ ries i n adults . Brain damage i n ch ildren can be better understood by considering the princi ples of behav ior development and the relationship between neural structu res and behavior ( Shahee n , 1984 ) . The fi rst five years of l i fe constitute the period of g reatest cortical development . The progress of mye l i n ation in various anatom ical reg ions affects behav­ ioral development . Conceivably, the type of behavioral d i fficulties that occur could be re lated to the neurostruc­ tu ral components that are u ndergoing the most rapid devel­ opment at the time of the brain i nj ury. D i ffuse bra i n inju­ ries during the second year of l i fe , for example , may affect the development of speech and l anguage , whereas s im ila r i nj uries during the third year of l i fe m ay affect spatial­ symbol ic manipulation s . Older children who incur brain damage - particularl y those above the a g e of 1 1 o r 12 show a more mi xed pattern of deficits , w ith deficits being similar to those found in adu lt patients . This is true to some extent of all c h i ldre n above 5 years of a g e , espec ially i f speech h a s developed . C lassical neurological patterns may be show n , such as verbal disorders assoc iated w ith left hemisphere lesions -

ith

THE

CLINICAL

H ISTORY

d itional l y. study is needed not only of the degr 2e o f recov­ ery fol low i n g bra i n damage but a l so of t h e c h i l d\ abil ity 1 0 learn and re learn cog n i t i ve s k i l l s . AS P E CTS O F A N E U RO P SY C H O LO G I CAL ASSESS M E N T

A wide spectrum of psychological deficits. va rying I n nature a n d degree . accompany bra i n i njury i n children . A lthough a neuropsychological assessment may not estab­ l i sh the specific narure . site. and extent of 3n u nderl y i ng b ra i n lesion . it can accuratel y determine t he sensory, motor. and mental defi c i t s that may be present. Neuro­ psychological assessment batteries provide i n format ion about verbal a n d non verbal i n t e l l ec t u a l a b i l i t ies and perceptua l - Illotor a b i l i t i e s . i nc l ud i ng voc a bu la ry. com­ prehe nsion . concept formation . memory. percept ion . and motor s k i l l s . This i n formation complements the overall 1 0 . w h ich by i tsel f i s often i nsufficient for eval uat i ng the efrects of bra i n lesions on specific mental processes . In all case s . test scores should be based on age-appropriate norms. althuugh raw scores can be used to study the child's pe rformance ove r t i me . Repeated psyc holog ical test i ng ( o r serial test i ng) . where needed . is an excel lent means of mon itoring t he evol ving c l i nical picture for possible mental changes in the course of the d i sease process . Serial test i ng chart s t he rate of recov­ ery or deci i ne following su rgery. fol l ows the course of any degenerat ive process. and aids i n educat ional . vocat ion a l . and rehabil itational deci sions . The updated i n forll1;Jlion about the c h i l d's progress can help the fa m i l y. teachers . and others responsible for the ch i ld's we l fare determine what can real istically be expected of the child . A neuropsycho­ logical assessment Illay a l so help in estimating the prob;l­ ble course of deve lopment and progress in school and i n formulat i :lg i ntervent ion procedure s . A l l sources o f infor­ mation shou ld be considered in the neu ropsychological assessme n t . i n c l u d i ng the c l i nical h i story. behav ioral ob­ servations . c u rren t test results, prev ious test results ( where ava i l able ) , and the neu rological eva l uation . T H E C L I N I C A L H I STORY

C l i nical Assessment I n t erview with Pa re nts

The cl i n ical h i story should focus on the child's develop­ ment ( see Chapter 16 ) . The history should begi n with events surrounding the b i rth process ( pre natal . pe rinata l . and postnata l ) . Prematu rity, R h i ncompatib i l ity. d i fficult

695 labor. and a loll' Ap�ar score (a score derived from assess­ ment of the nenn3te··s heart rate , resp i ratory e ffort , muscle tone . cry. and color : see C hapter 12) a re relevant factors . Part icular empha,i" should be g iven to events that may have et iological i mportance for bra i n damage . These i n ­ c l ude prolonged high fevers ; i njuries to t h e head; use o f anesthet ics duri ng su rgery : poiso n i n g associated with foods. chem icah. 0 I ' medicat ions ; occasions of prolonged nausea and vOlll itin:g not re lated to comillon i l l nesses ; and changes in energy expenditure . I f the ch i ld �u tfc red any i nj u ry t o the head . you should inquire about the lC"cat ion of the i nj ury. whether the c h i l d lost consciousnes, a n d for how long, whether medical attent ion was m ailable. what the child's behavior was l i ke both i mmediately and for several days fol l ow i n g the i nj u ry, sleep d i stu rbance, . compl a i n t s of pai n . and whether there were any delayed personality change s . In evaluating a childs progress fall wing brain i n j u ry. it is useful to ask a parent or other informant the questions shown i n Exhibit 22-1 . They cover various areas . including motor, sensory­ k i nesthetic. a ffecti\'e . and cog n itive fu nct ion i ng ; conte n t , form . a n d pract ical u s e of language; changes i n c o n ­ sciousness: soc ial -personal it y characteristics: a n d t h e in­ formant's persollal react ions to the c h i ld's i nj u ry. With sl ight mod ification s . the questions can be u sed d i rectly with older brai n-inj ured c h i l d ren and adults who are able to communicate . Also inqu i re abollt n:cent changes in the child's l i festy l e , level of fu nct ion ing , alld emot ional stabil ity. For exampl e , i n s t a b i l i lY. i rrit;lh j l ilY. or lethargy m a y precede othe r symptoms of a bra i n tu mor. whe reas a progressive dec l i n e i n academic fu nctioning may b e an early s i g n o f degener­ ative brain di sease . Sudden and i nadequately expl a i ned changes ill be havior an! l i kely to be associated with acute , as opposed to chron ic . bra i n d i sorders . C l inical Assessment I nt e rview with t he Child

In i nte rview i ng the child . obtain i n formation about his or her general orientation to t he present t i me, place , and person : memory : changes in behavior; and other rel ated issues ( see Tables 16-9 and 16-10 in Chapter 16) . With older c h i ldre n ( or with parents acting as i n formants ) . beg i n with open-e nded questions and proceed to more specific ques­ t ions as the i nterview proceeds. Be sure to conduct the i nt e rview in a nonthreaten i ng and casual manner. El icit i n formation about the chil d's c h i e f compl a i n t ( What does the c h i l d see as t h e major problem? ) , a n d t ry to obtain a clear understand i ng of the onset and progression of symptoms . With older c h i l d re n . dete rmine such facts as

CHAPTER 22

696

ASSESSMENT OF BRAIN DAMAGE

Exhibit 22- 1 Sem i-Structured I nterview Format for Use with Parents of Brain-Injured Children

You might start the interv iew by say ing "I'd l i ke to talk w ith you about how your child is getti ng along now. " If the parent responds with a "yes" to any question, inqu i re fu rther. You might say, for example, "Tell me more about that" or "Please describe his [or her] behavior in more detail . " The questions can be rephrased or altered in any manner deemed appropri­ ate . In evaluating responses, take into consideration the child's age and the capabilities rhar were presel1l before rhe brain injury. See Chapter 1 6 for information about interview techniques. Motor

I . Is more hyperactive now than before the injury? 2. Is more awkward now? For example , does he (or she) seem clumsy or bump i nto things? 3 . Is his (or her) handwriting poorer now? 4. Printing? 5. Drawing? 6. Does have a rigid posture? 7. Does have more speech difficulties? 8 . Does seem to repeat movements? 9. Does complain of numbness or loss of sensation? 1 0 . Does have d ifficulty copying gestures? I I . Does have d ifficulty standing? 1 2 . Walking? 1 3 . Running? 14. Balancing?

30.

31. 32. 33. 34. 35. 36. 37. 38. 39.

__

__

40. 41.

Does experience ringing sounds? Is unusually sensitive to sound? Does hear music as noise? Has experienced any change in taste? I n smell? Have any of 's favorite odors become unpleasant and then pleasant again? Does report being dizzy? Has recently fainted? Does have to hold on to things to keep from fall ing? Does experience tingl ing sensations in the fin­ gertips or toes? Numbness? Has there been any disturbance i n 's abil ity to make turns or put his (or her) arms into the correct sleeve? __

__

__

__

__

__

__

__

__

__

Affective

__

__

Does easily lose control of his (or her) emotions? Does 's mood change easily? Does become angry without cause? Is irritable? Does talk too much or too l ittle? Is anxious? Is l istless and depressed? Has become more hostile? Aggressive? Uncooperative? Negative? Does l ie? Steal? How does feel about himself (or herself)? Have there been any changes in his (or her) personality? Is concerned about his (or her) body? 5 8 . Is compul sive? 59. Does deny that he (or she) has any problems? 60. Does feel i nsecure or weak or uncertain of himself (or herself)?

__

6 1 . Does have d i fficulty with memory? 62 . ( I f yes) In general? W ith specific events? With material that he (or she) just learned? With material that he (or she) learned a while ago? 63 . Does understand what is read to him (or her)? 64 . Has there been any change in 's understanding of what he (or she) reads? 65 . Has there been a general decline in 's overal l intel lectual abil ity? (bhibit COl1lillUeS next page)

__

__

__

__

__

Sensory IKinesthetic

1 5 . Does have a short attention span') For example, does he (or she) often request that you repeat infor­ mation? 1 6 . Does have d ifficulty comprehending or fol lowing directions? 1 7 . Is distractible? 1 8 . Does reverse or scramble what he (or she) hears? 1 9 . Does confu se words and sounds and t h e i r meanings? 20. Has there been a change in 's vision? 2 1 . Does reverse words, letters, or numbers when reading? 2 2 . Does have difficulty with depth perception? 2 3 . Does have diffi cu l ty i n recognizing objects? 24 . Does have diffic ulty in bui l ding or constructing things? 2 5 . Does report sme l l i ng odd odors? 26. Has experienced blurred or double vision? ( I f appropriate, add "even with glasses o n . ") 2 7 . Do l ights disturb ? 2 8 . Does experience l ight flashes? 29. Has there been a change in 's hearing? __

__

__

__

__

42 . 43 . 44 . 45 . 46. 47. 48. 49. 50. 51. 52. 53. 54 . 55. 56. 57 .

__

__

__

__

__

__

__

__

__

__

__

__

__

__

__

__

__

__

__

__

__

__

Cognitive __

__

__

__

697

THE CLINICAL HISTORY

Exhibit 22- 1 (cont.) 66. In judgment? 67. I n planning? 68. I n abil ity to combine things? 69. Does have problems w ith language? 70. Has there been any change in 's ability to recognize numbers, letters, or musical notations? 7 J . Has there been any change in 's writing? 7 2 . A r ithmetic skills? 7 3 . Does recognize n umbers w hen w ritten or spoken? 74 . Can do mental calculations? 75 . Does deal with problems more concretely? 76. Does he (or she) have difficu l ty i n changing tram one activity to another? __

__

__

__

__

__

( I f there have been changes i n the child's language, you can use the fol lowing questions to inquire about language content, language form , and practical use of language . ) Language Content 7 7 . Can __ define words? 7 8 . Can name objects that are shown to him (or her)? 79. Can complete an open-ended sentence such as "The sky is __ "'1 80. Can count? 8 1 . Name the days of the week? 8 2 . Can recognize an appropriate word when it is offered to him (or her)? For example , if he (or she) was holding an orange and could not name it and you said orange, would he (or she) recognize that you had said the correct name? 83 . Can repeat names? 84. Can follow changes in the topics of conversa­ tion? For instance, if you were talking about baseball and you switched to tal king about what you were having for d inner, could he (or she) fol low that? 8S . Can __ understand what you say? __

__

__

__

__

__

Language Form 86. Can construct phrases and sentences? 8 7 . Can use a variety of sentence types, such as simple sentences and questions? 88. Does use articles (such as a), prepositions (such as oj) , pronouns (such as me), and conj unctions (such as and)? 89. Does use word endings to make words pos­ sessive ( 's ) , plural (-s ) , and past tense (-ed or -I)? For e x a m p l e . wou l d he ( o r s h e ) say " Ma ry's parents wal ked"? 90 . Is aware of language errors when he (or she) makes them? __

__

__

__

__

Practical Use of Language 9 1 . Does 92 . Does

__ __

use the telephone? answer the telephone when it rings?

look up telephone numbers in the di rec93 . Does tory? 94. Is able to remember some telephone numbers? 95 . Does re:ad the newspaper? 96 . How has this c:hanged since the brain injury? 97. Does read magazines? 98. How has this changed since the brain injury? 99. Does I\·atch television? 1 00. How has this c hanged since the brain i nj u ry? 1 0 I . Does go shopping? 1 02 . How has this c hanged since the brain injury? 1 03 . Does handle money on his (or her) own? 1 04 . How has this c hanged since the brain injury') 1 05 . Does __ vtsit friends? 1 06. How has this changed since the brain injury? 1 07 . Does __ talk about the past? 1 08 . How has this changed since the brain injury? talk about the future? 1 09 . Does 1 1 0. How has this c hanged since the brain injury? I l l . When __ speaks. i s what he (or she) says appropri­ ate to each situation? That is, does what he (or she) says illatch the s itu at Ion ') I 1 2 . Can __ deal with ctifficulties in talking with others? For example. when others stop talking, can he (or she) respond and k('ep the conversation going? 'I 1 1 3 . How often do people I n your family talk to 1 1 4 . How has this Changed since the brain injury? I 1 5 . How often does talk to others? 1 1 6. How has this Changed since the brain inj ury? 1 1 7 . When woul d YIlU say that __ does the most talking? __

__

__

__

__

__

__

__

__

___

1 1 8.

Why')

to talk? 1 1 9 . Where i s the best place for 1 20 . Does __ have mal e or less opportunity to talk now than before the br;lin i nj u ry? 's desire to talk changed since the brain 1 2 1 . How has injury? 1 22 . Does ever argue? 1 2 3 . (If yes) About wh,II') 1 24. Is able to give directions? 1 25 . Does complain? 1 26. ( I f yes) About wh at? __

___

__

__

__

Changes in Consciousness

1 2 7 . Does appear to be disoriented? For example. does he (or she) get lost easily or have trouble finding his (or her) way around? 1 2 8 . Has experienced blackouts , during which he (or she) is unable to hear when someone talks to him (or her) or is unable to respond? 1 29 . Does experience feel ings that he (or she) is losing his (or her) body? 1 30. Has reponed strange fecl ings. such as a feel ing that some unknown danger is about to happen? __

__

__

__

(Exhibit continues next page)

CHAPTER 22

698

ASSESSMENT OF BRAIN DAMAGE

Exhibit 22- 1 (cont.) 1 3 1 . Has

been doing things that he (or she) i s unaware of? 1 32 . Has started out doing one thing and then found himself (or herself) doing something else? 1 3 3 . Has __ reported that he (or she) fee l s that the size of his (or her) hands or feet or head i s changing? __

__'s brain injury, do you fi nd that you are more protective of him (or her)?

14 1 . Since

__

Personal Reactions 1 34. Since 1 35 . 1 36 . 1 37 . 1 38. 1 39. 1 40 .

's brain i nj ury, do you lose patience with him (or her) more often? Since 's brain injury, have you l i m ited your social activities? Since 's brain injury, does he (or she) expect you to spend more time with him (or her)? Since 's brain injury, do friends who typically visited you at home still do so? Since 's brain injury, do you fi nd it more difficult to concentrate on your work? Has 's brain injury affected your physical health? Has 's brain i nj u ry affected your emotional health? __

General 1 4 2 . How do you aid 143.

146.

__

147.

__

__

__

__

1 44 . 145.

__

__

in understanding and producing speech? What could you change around your home so that would enjoy talking more? What sorts of things do you talk about with ? How has this changed since the brain injury? If a stranger were to visit , how would you pre­ pare the stranger for the visit? If I were coming to l ive in your home. what would you tell me it is l ike to live with __ since the brain inj u ry? __

__

Nore. The first 1 33 questions, with minor changes in wording ( for

example, "do you" for "does __ ") can be lIsed to interview children and adults directly. Source: Adapted , in part. from Chapey ( 1 98 1 ) and Small ( 1 980) .

__

the date the problem was fi rst noticed . how the symptoms developed , what aggravates and d i m i nishes the symptoms , what type of help h a s been received, and what t h e child's current level of functioning i s . The following symptoms m a y be of special neurological significance and are worthy of spec i fic i nquiry during the i nterview (Adams & Jenkins. 198 1 , pp. 2 72-2 7 3 , with changes in notat ion ) : I . headaches and times of day they occur 2 . alteration in consciousness 3 . abnormal movements (tremors , jerking. tics) 4 . seizures 5 . disturbances in vision 6. disturbances i n perception , espec i a l l y o l factory hal l ucinations 7 . pain and paresthesias (abnormal sensations ) 8 . muscle weakness . stiffness, or paralysis 9. vertigo w ith or without nausea or vom iting 1 0 . difficulty i n aud itory perception and/or the pres­ ence of tinnitus ( ringing in the ears) I I . difficulty in the use of language (expression and comprehension) 1 2 . bowel , bl adder. or genital dysfunction 1 3 . confusion , decreased i ntel lectual e ffi c iency, or memory di fficulties 1 4 . changes in mood or emotional regulation

During the i nterview be alert t o the chi ld's level o f consciousness, l anguage , memory, i ntellectual and cog­ nitive functioni n g , and sensorimotor functioning . The de­ velopmental h i story (see Tables 16-12 and 16-13 in Chapter 16) may provide etiological clues about the exami nee's symptomatology. If the child has not been seen for a neurological workup and you find symptoms suggestive of organic dysfunction - particularly complaints of memory deficits, seizure s , headaches, or unusual behavioral symp­ tom s , such as deterioration of academic performance , reversion to a n i mmature pattern of social and emotional behavior, and conversion reactions ( for example. loss of function of limbs or reduction i n visual acuity) - refer the child for a pediatric neurological examination .

B E HAVIORAL OBSERVATIONS

During the psychological evaluation , the child may exhibit behavior patterns suggestive of brain damage . Table 22-4 l ists a number of such patterns . You would be wel l advised to develop an awareness of these behavior patterns, if only to rule them out; the earlier brain i njury is detected , the better the possibilities for rehabi l itation . Also rev iew Table 1 8-4 in Chapter 1 8 , which l ists a variety of language diffi­ culties , many of which may be shown by brain-damaged patients.

699

BEHAVIORAL OBSERVATIONS

Table 22-4 Possible Signs and Symptoms of Brain Damage Observed During the Neurops ychlological Evaluation Motor

Sensory

Hyperkinesis (con­ stant movement. in­ abil ity to sit sti l l ; fi ngering, touch­ ing, and mouthing obj ects; vol uble and uninh ibited speech ) Awkwardness in loco­ motion (clum­ siness, bumping into things. atypical arm swing . incoor­ dination. t remors . invol untary move­ ment . asymmetry of facial mus­ culature and ex­ pressive gestures while tal k i ng) Awkwardness in s k i l led movement ( poor printing. writing, and d rawing) I mpaired copy ing of geometric designs Postural rigidity Speech d i fficul ties ( e . g . . slow speech ) M i xed left-right dom inance Repetitive movements Perseveration N umbness and loss of sensation D ifficulty in crossing l imbs across cente r of body D i fficulty in copy ing gestures

Short attention span ( e . g . , difficulty in retaining and re­ call ing i n formation or words) Poor concentration ( e . g . , difficulty in comprehending or fol lowing d i rec­ t ions, di fficul ty in performing a se­ quence of tasks in the order given) Dist ractibil ity Perceptual d ifficulty ( e . g . . closure d i ffi ­ culty. v isual-motor disturbances. use of fi ngers to trace v isual shapes, turn­ ing materials arou nd) D i fficulty in matching objects or symbols with verbal sti m u l i Reversal or scram­ bl ing of what is heard Confusing words and sounds and the ir meanings Difficulty in reading Word . letter. number reversals Confusing letters Difficulty with depth perception D i fficu l ty in recogniz­ ing objects Difficulty in build ing or constructing th ings Unusual episodic sen­ sory experiences ( e . g . . odd odors or vision of l ights)

Social/Personality Reduced frustration

threshold Emotional labil ity (e. g . , mood sh i fts. impulsivity, i rri­ tabil ity, aggressive­ ness . tearfu I ness . loss of control of emotions. inap­ propriate laughter ) Tal k i ng too much or too l ittle Anxiety ( occasional panic reactions ) Problems in sel f­ regulat ion ( e . g . . overact iv ity or underactivity) Depression or nat affect

Some i nlel lectual deficit Impaired j udgment Conceptual difficu l t ies (e.g . . i n abstract­ ing. planning. or­ ganizing. anticipating. ana­ lyzing and syn­ thesizing , and integrat i ng) Specific learning defi­ cit ( in reading. writing . spell ing . or arithmetic) Language d i fficulties (e.g . . mal­ apropisms. im­ precise synonyms. truncated sen­ tences. mispronun­ cialions . circumstantial i ty ) Att ract ion to m i nute details Impailed right-left orientat ion PCI ,elc:ral ion Conen·te . rigid . and Inflexible thinking Dilficulty in shift i ng from one activity to another Memory d i fficulty ( recent. remote . or both : visual . au1I itory. or bot h ) Di,orientation ( e . g . . gets lost eas i l y. i , disoriented in fa­ lIliliar settings)

i n te rpersonal difficul­

t ies (e.g . . hostile, aggressive . uncooperative) I m maturity ( e . g . . may regress to more childl ike forms of behavior) Negativism Antisocial behavior ( ly ing, steal ing. t ruancy, sexual offenses) D isturbed self­ concept Distu rbed body con­ cept or body image Changes in person­ a l ity ( e . g . , a pre­ viously fastidious child has become u n kempt and careless) Hypoc;hol)dri.a ca.1 prcoccupations COlllpul s i ve

tendencies Denial ( e . g . . a child lllay de ny that he or she h3S any problems) Ind ications of insecu­ rity ( c . g . . ex­ pressions of weak ness . uncer­ ta i nt y. and i nade­ quaey in dealing with test lllaterial s ) I mpaired social j udg lllent

CHAPTER 22

700 During the evaluation, consider the extent to which (a) special techniques are needed to secure the child's atten­ tion, (b) i nstructions have to be repeated before the child does what is asked, and (c) extraneous act ivity has to be prevented . Be especially alert for behaviors indicative of motor restlessness . Hyperactivity (or any other behav ioral sign), however, should never be taken by itse l f as indicati ve of cerebral damage or dysfunctio n .

N E U ROPSYC H O LOG I CA L TEST BATTERIES FOR C H I LD R E N

Halstead-Reitan Neuropsychological Test Battery for Older Children and Reitan-Indiana Neuropsychological Test Battery for Children Tho batteries for evaluating childre n suspected of having brai n damage are the Halstead-Reitan Neuropsychological Test Battery for Older Chil d ren, designed for children between 9 and 14 years of age , and the Reitan-Indiana Neuropsychological Test Battery for C h i l dren , designed for children between S and 8 years of age (Reitan & Dav ison , 1974; Reitan & Wol fson , 198 5 ; Selz , 198 1 ) . Both batteries contai n cognitive and perceptual-motor test s , some of which also appear in the adult battery a n d some of which were especially designed for young childre n . (See Table 22-S for a description of the batterie s . ) Table 2 2 -6 shows the Reitan-Indiana Aphasia Screeni ng Test, and Figure 22-1 shows the stimulus figures for the test. The complete Halstead battery also includes an intell igence test. Information about the reli a bi lity and vali d ity of the Halstead-Reitan and Reitan-Indiana children's batteries i s scarce , a n d norms are extremely l i m ited . Therefore , their usefu lness depends on the clinical sophist ication of the examiner. (Norms for the adult version are available for IS­ through l 7-year-old childre n , i n a publication by Fromm ­ Auch & Yeuda l , 1983 . ) I n one study (Selz & Reitan, 1979 ) , the Halstead-Reitan Neuropsychological Test Battery for Older Children, together w it h some additional tests , was found to classify children , ages 9 to 14 years , into one of three g roups (normal , learn ing-disable d , and bra i n ­ damaged) with 73 percent accuracy. A short screening battery drawn from the Halstead-Reitan battery was found to be useful in differentiating brain-damaged c h i ldren from controls (Reitan & Herring , 1985 ) . Performance on t h e Hal stead-Re itan Neuropsycho­ logical Test Battery for Older C h ildren i s not i ndependent of level of intellectual functioning. A study (Seidenberg ,

ASSESSMENT OF BRAIN DAMAGE

Giordani , Berent, & Bol l , 1983) of 1 2 1 9- to 14-year-ol d children w it h documented seizure d isorders i n dicated that those with h igher W ISC-R IQs had better Halstead-Reitan scores on 6 of 14 tests. Tests most related to i ntelligence were those dea l i ng w ith problem solving , language, and auditory-perceptual analysis (Category Test, TPT Loca­ tion , Speech Perception Test, Seashore Rhythm Test, Trai l Making Test B , and Aphasia Screening Test) . Tests least related to intelligence were those of simple motor func­ tions and right-left hand differences . Thus d i fferences i n intellectual ability were most evident i n tasks requiring problem solving and mental efficiency. The results empha­ size the i mportance of considering the child's IQ in i nter­ preting performance on the Halstead-Reitan Neuropsy­ chological Battery for O lder Children.

Luria-Nebraska Neuropsychological Battery: Children's Revision The Luria-Nebraska Neuropsychological Battery : Chil­ dren's Revi sion (LNNB-C) i s designed to assess a broad range of neuropsychological functions in children 8 to 12 years of age (Golden , 1987) . It can be considered a down­ ward extension of the adult version , although items are not necessarily i nterpreted in the same manner on the two tests . The LNNB-C is designed to assess cognitive deficits and aid in planning rehabilitation programs. The LNNB-C is individually administered. It contains 149 items grouped into 1 1 c l inical scales (see Table 22-7) and two optional scales . Additionall y, the items are re­ grouped into three summary scales and 1 1 factor scales. The clinical scales are designed to assess sensorimotor, perceptual , and cognitive abilities. The sumnwry scales provide i n formation for d iscrim i nating between brain­ injured and normal children. The factor scales are said to be helpful in assessing specific neuropsychological func­ tions , although they must be interpreted cautiously. The LNNB-C takes approximately 2 1/2 hours to adm i nister to a brain-damaged child. A l l items are scored 0, I, or 2. A score of ° i nd icates normal functioning, I indicates weak evidence of brain dysfunction , and 2 indicates strong evidence of brain dysfunctio n . Note that higher raw scores indicate a poorer response . Raw scores are transformed into T scores (M = SO, SD = 10) . H igher T scores indicate poorer per­ formance. A qualitative scoring system can be used to evaluate items on which errors occurred. The system has S7 individual categorie s , such as attention difficulties, fatigue , jargon , perseve ration , sequence e rrors , and tremors.

Scoring.

70 1

NEUROPSYCHOLOGICAL TEST BATTERIES FOR CH ILD REN

Table 22-5 Description of the Halstead-Reitan Neuropsychological Test Battery for Ol,de'" Children and the Reitan-Ind iana Neuropsychological Test Battery for C h i ldren Desc rip/ ion

Tes/

Measures concept formation : requires child to find a reason (or rule ) for comparing or sorti ng objects . Measures somatosensory and sensorimotor a bi l ity : requires child , while blindfolded , to place blocks in appropriate recess using dOlllinant hand alone. nondominant hand alone, and both hand s . Measures fi n e motor speed ; requires child t o press and release a lever l i ke a telegraph key as fast as possible . Measures expressive and receptive language functions and lateral ity : requires child to name common objects, spe l l . identify numbers and letters. read , write, calculate, understand spoken language . identify body parts , and d i fferentiate between right and left . Measures perceptual recogn ition; requires child to match figures a t t h e top o f a page with figures at the bottom of the page.

Category Testa Tactual Performance Testa

Finger Tapping Testa Aphasia Screening Testa

M atching Pictures Testb Individual Performance Testb Matching Figures Star Matching Vs Concentric Squares Marching Testb

Progressive F igures Testb Color Form Testb Target Testb Rhythm Teste Speech Sounds Perception Teste Trail Mak ing Test (Parts A and Sensory lmperceptione Tactile Finger Recognitione Fingertip Number W ritinge Tactile Form Recognitione Strength of Gripe

B)e

Measures perception; requi res child to match ditferent complex figures . Measures visual-motor ability : requires c h i ld to copy a staL Measures perception ; requires child to Illatch " V s . " Measures visual-motor ability : requires child t o copy a series o f concentric squares. Measures gross motor control ; requires child to (a) connect a series of circles w ith a crayon in a given order with right hand aillne and with left hand alone and (b) reproduce examiner's fi nger and ann movements. Measures flexibil ity and abstraction: requires child to connect several figures, each consisting of a small shape contai ned with in a large shape . Measures flexibil ity and abstraction : requ i res child t o c onnect color shapes ; first by c o lor and then by shape. Measures memory for figures: requi res c h i ld to reproduce a v isually presented pattern after a three-second delay. Measu res alertness, sustained attention . ;lI1d aUditory perception; requ i res child to indicate whether two rhythms are the same or d i fle renl . Measures auditory perception and auditory- v isu;t\ integration ; requ ires child to indicate . after l istening to a word on tape . whicll 01 fo u r spell ings represents the word . Measures appreciation of symbolic s i g n i f i cance of numbers and letters , scanning abil ity, flexibil ity. and speed ; requires child to connect circles that are numbered . Measures sensory-perceptual ability ; requ i res ch ild to perceive bilateral simultaneous sensory stimulation for tactile. aud itory. and v isual modalities in separate tests . Measu res sensory-perceptual ability: req u i res ch i l d . while blind folded , to recognize which finger is touched . Measures sensory-perceptual ability: requires child , while blindfolded . to recognize numbers written on fi ngertips. Measures sensory-perceptual ab i l ity : requ ires child to identify various coins through touch alone with each hand separately Measures motor strength of upper extremities: requ i res child to use Smedley Hand Dynamometer with preferred hand and nonpreferred hand

The W ISC-R (or WA IS-R) I S oflen admin istered as pan of the complete batlery " This test appears both on the Halstead-Reitan Neuropsychological Test Batlery fo r Older Childre n and on the Reita n - I ndiana Neuropsychological Test Battery for Children. b This test appears only on the Reitan-Indiana Neuropsychological Test Batlery for C h i ld re n . This test appears o n l y on the Halslead-Reilan Neuropsychological Test Batlcry for Older Children. Note.

c

Table 22-6 I nstructions for the Reitan-I ndiana Aphasia Screening Test Inslruclions

link

I . Copy Square

( point to the square) on your paper I wam you 10 do if wilhoul lijling .1'01/1' ( pointing to the square ) . Whal i s Ihal shape called ? ( o r ) Whal is Ihe name for Ihaljigure ?

Firsl, dralV Ihis

pencil froll1 Ihe paper: Make il abolll lhis same size

2 . Name sqllare 3 . Spell square

Would vall speli lhal 1V0rd for me ?

4 . Copy cross

( point to the cross) on YOllr paper: Co around Ihe olllside like Ihis (quickly draw a finger-line around the edge of the stimulus figure) wllil vou gel back 10 where you slar/ed. Make il aboril lhe smne size ( point to the cross) .

Drall' Ihis

5 . Name cross 6 . Spell cross 7 . Copy Iriangle 8 . Name Iriangle

Whm is Ihm shape called ? Would .mll speil lhe nall1e of il .? Now I wam you 10 draw Ihisjigure.

( Point to the triangle . )

Wlull 1V01lid vou cail lhal figure ?

9 . Spe l i lriallgie

Would vou speil lhe name of il for me ?

1 0 . Name baby I I . Write clock

Whal is Ihis .?

( Show item 1 0 . )

NOli' 1 011/ goillg

10

show you anOlher piClure bw do NOT lell m e Ihe lIame of il. I don i lI'an!

\'OU 10 say anHhing oUl loud. JUSI WRITE Ihe name of lhe picHlre on you r paper:

( Show

item I I . )

12. 13. 14. 15. 16. 1 7.

Whcll is Ihis ? ( Show item 1 2 . )

Name fork Read 7 six 2 Read M C W Reading I Reading I I Repeat Irial/gle

( Show item 1 3 . ) ( Show item 1 4 . ) NolV I wam )'ou 10 read Ihis. ( Show item Can \'OU read Ihis ? ( Show item 1 6 . ) I wam YOII 10 read Ihis.

Read Ihis.

NOli' I am going

10

15.)

say some words. I wam you 10 lislen carefully and say Ihem afler me as

carejiilly as vall can. Say Ihis word: "Iriangle. "

1 8 . Repeat Massachusells 1 9 . Repeat Melhodisl episcopal 20. Write sqllare

The neXI one is a lillie harder bUI do your besl. Say Ihis word: "Massachusells. " Now rep em Ihis one: "Melhodisl Episcopal. " DOli 'I say Ihis word 0111 loud. ( Point to the stimulus

word ·'square.") JUSI wrile il on your

paper

2 1 . Read sel'ell 2 1 A . Repea t sel'ell 2 2 . Repeat-explain He showed Write He showed Ihe Il'am illlj

24. Compute 85

I a/II going 10

SOl '

somelhillg Ihal f wan! you 10 say afler me, so lislen carejiJlly: "He showed

Ihe \\Iamil/g. " NoH' \'011 sav il. Tell ll1e ill your own I\lords whal Ihal lI1eans.

Ihe lI'omillg

23

( Show item 2 1 . ) Remove the stimulus card and say : Now, I wam you 10 say Ihis afler me: "se ven.

Would YOII read Ihis word ?

-

27

N(}\,' 1 11'01/1 you 10 lI'rile Ihal semence all Ihe paper

=

Here is an arilhmelic problem. Copy il down on your paper any way you like and Iry

25. C om p u te 1 7 x 3 26. Name keY 2 7 . Demonstrate u s e o f kev =

il OIl1.

( Show item 24 . )

work

Noll' do Ihis one in your head. Wrile down only Ihe answer: Whm is Ihis .? ( Show item 26 . ) (Still presenting the picture of the key : ) If you had one of Ihese in your hand, show me how .1'011 'mllld lise il.

NOli' I 11'(/111 YOII (0 drall' a picture (hal looks jusl like (his

2 8 . Draw kn

10

( point ing to the picture of the key ) .

7h' ( 0 make \'OlIr ke\' look enough like (his one (still pointing to the pict u re o f the key) so

11/(/( I 1I'Qllld kllOIl' i( was the same key frOIl1 your drawing. Make il about the same size.

( Show item 29 . ) B e sure t o note any false starts o r even mild expressions of

Read 30. Place le}i hand 10 righl ear

NOli'. lI'ould \'011 do lI'i1al il said ?

3 1 . Place lefl halld 10 lefl elboll'

NOli' I lI'ont \'011

29.

22- 1

WOllld \'011 read this ?

confusions.

10 pili

\'our left hand to your lejl eibolll,

Screening Test is pan of the Reitan-Ind iana europsychological Test B alle ry for admin ister and interpret the test as well as the ballery. Two books that can assist you i n i n terpre t i n g the ballcry a r e Aphasia and Sensorr-Perceplilol Deficils i n Adlills b y Reitan ( 1 984) a n d Aphasia a n d Sensory-Perceplllal Deficils in Children by R e i t a n ( 1 985 ) . A dd itio n a ll y Reitan and Wo l fson s ( 1 985 ) Halslead-Reitan Nell rops."chological TeSI Baller),: Theory and Ciinicol ll1lerpretalion i s an e x c e l l e n t sourcc for i nformation on how to i n tegrate the findings of the Reitan - I ndiana Aphasia Screen i ng Test with the rest of the results of the Hal stead­ Rei ta n Neuropsvchological Test Ballery for a complete neuropsychological assessment. Separate kits for adults and childre n . which i nc l ude the appropriate boo k . recording for m s . and test book lct . are available from the Ne u r o psychology Pres . 1 3 3 8 E. Edison Street, Tucson, AZ 857 1 9 . SOllree. R e p r i ntcd . w i t h changes i n notat ion . w it h permission o f the publ isher a n d authors from R . M Reitan a n d D . Wolfson. DJe Halslead-Reiran Nellrop.I \'('h"logiCilI 7(>,1'1 Blillen' ( Tucso n . AZ Ne u ro psy cho lo g y Press . 1 98 5 ) , pp. 75-78. Copyright 1 985 by Neuropsychology Press. NO/I'. See F i g u rc

for s t i m u l u s figures . The R e i ta n - I ndiana Aphasia

C hi ldre n . Considerable c l i n ical c x perience is needed to ,

'

N E U R O PS Y C H O L O G I C A L TEST BATTERIES

FOR CHIL D ".Ef\1

703

__ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __ � e � _ E� ____

D0~

7.

4.

1.

-

� C!J � � m � ~ :>. '

tI

' .... �

.

�

1 0.

7

SIX 2

1 3.

HE IS A rRJ[NDt.. Y """MAL, A f'AMOVS Wll
_ _

_ _

o INCREASE PERFORMANCE IN REAOINESS ACTIVITIES o Recognlle lollowlng COIOfS 0 Red 0 81u� D Grm 0 Y�lIo�' 0 OfJ�ge o Purpl, 0 Brown 0 Blae. 0 Prn, 0 Gray o Visually dlSCfllJllnale like and OINmnl symbols o foHow lefl lo lighl stQuenu o Copy lollawing �tJail� 0 Clrcl� 0 Plus Sl�n 0 SQuaJ� 0 llian.jle 0 Diamond o Draw·A·Person wllh detall

ol 19 body oaJls [Sl! 8I1Qanu fecord booklel for D lden" ly Spel:IIIC body Pp and i mplement a statewide, comprehensive, coordinated . multidis('iplinary, interagency program of early intervention SCI v icl(!s for handicapped infants and toddlers and -

their fam i l il's. ( 2 ) to fat i l iwtc t he (:oordination of payment for early inter­ vent ion SCI vices from Federal , State , local , and pri vate sources ( inc l ud mg ublic and private insu rance coverage) , and (3) to enilance its capacity to provide quality early interven­ tion services and r,xpand and improve existing early interven­ tion serv ices be ing provided to handicapped infants. toddlers, and their fam i l i e s . DEFI N ITIO S

Sec. 672 . As u�ed in this part ( I ) The term 'handicapped infants and toddlers' means indi­ viduals from birth to age 2 , inclusive, who need early interven­ tion services because they (A ) are experiencing developmental delays, as measured by appropriate diagnostic instruments and procedures in one or more of the following areas: Cognitive development, phys­ ical development. l anguage and speech development, psycho­ social development , or self-help skills, or (8) have a diagnosed physical or mental condition which has a high probabil ity of resulting in developmental delay. Such term may also include, at a State's discretion, individuals

CHAPTER 24

778 from birth to age 2 , inclusive, who are at risk of h;1Ving slubstan­ tial developmental delays if early intervention services are not provided . ( 2 ) 'Early intervention services' are developmental s<e rvices whic h ( A ) are provided under public supervision, (8) are provided at no cost except w here Federal r State law provides for a system of payment by families, including a schedule of sliding fees, (C) are designed to meet a handicapped infant's or toddler's developmental needs in any one or more of the following areas: (i) physical development, (ii) cognitive development, (iii) language and speech development, (iv) psycho-social development, or (v) self-help skills , (D) meet the standards of the State, including the require­ ments of this part, (E) include (i) family training, counseling, and home visits, (ii) special instruction, (iii) speech pathology and audiology, (iv) occupational therapy, (v) physical therapy, (vi) psychological services, (vii) case management services, (viii) medical services only for diagnostic or evaluation purposes, (ix) early identification, screening , and assessment ser­ vices, and (x) health services necessary to enable the infant or tod­ dler to benefit from the other early intervention services, (F) are provided by qualified personnel , including (i) special educators, (ii) speech and language pathologists, and audiologists, (iii) occupational therapists, (iv) physical therapists, (v) psychologists, (vi) social workers, (vii) nurses, and (viii) nutritionists, and (G) are provided in conformity with an individualized fam­ ily service plan. (100 STAT 1145-1147)

The act establishes a state grant program to provide early intervention services for handicapped infants, tod­ dlers , and preschoolers . The act w i l l provide handicapped i n fants and toddlers and their famil ies with an assessment of their u nique needs and an individualized fami l y service p l a n to meet these needs . The i ndividualized family ser­ v ice plan, which w i l l be evaluated once a year, w i l l i nclude a statement of the child's present level of development, the family's strengths and weaknesses, the major outcomes that can be expected, specific services to be provided, the dates of initiation and the duration of these services, and the name of the child's case manager.

ISSUES IN CONSULTATION

ASSESS M E N T PROC E D U RES A N D SPECIAL E DU CATION O N TRIAL

Over the past 20 years, the procedures used by schools to place children in speci a l education programs have been evaluated by the courts of our nation . Two principal issues have surfaced . (See Tabl e 24- 1 for a summary of the out­ comes of various lawsuits . ) One issue is the rights of c h i l dren , regardless of the degree of handicap , to receive a free and appropriate education . The courts have c learly ruled that handicapped c h i l dren are entitled to such rights . The second issue is the overrepresentation of ethnic m i nority children i n c lasses for the educable mentally retarded . Associated w ith this issue are the related matters of the assessment techniques used to certify placement i n special education programs a n d the value and role of special education . It has been argued that (a) ethnic minor­ ities are overrepresented in classes for the educable men­ tally retarded , (b) special education represents a dead end and provides substandard educational programs, (c) i ntel­ l i gence tests are biased , (d) pupils w ith l i mited fac i lity in English have been i nappropriately administered tests re­ quiring extensive faci l ity in Engl i sh , and (e) a ful l range of assessment techn iques has not been used i n arriving at placement decisions . Implied in these arguments is the contention that the c h i ldren were den ied equal protection under the Fourteenth Amendment of the Constitution. I n many o f t h e cases that have come before t h e judiciary, both parties have signed consent agreements, which have at­ tempted to rectify procedures that were unfair to ethnic m inorities . In addition, prov isions of PL 94-142 have sought to protect the rights of ethnic m inority chi ldre n .

Larry

P. v. Riles

I n Larry P. v. Riles , a federal court found the California State Department of Education to be in violation of Title V I of the C i v i l Rights Act of 1964, the Rehabilitation Act of 1973 , and PL 94-142 . I n October 1979 , Judge Robert Peck­ ham ruled that standardized inte l l igence tests "are racially and culturall y biased, have a d i scriminatory i mpact against black c h il d ren, and h ave not been validated for the purpose of essentially permanent placements of black c h i ldren into educationall y dead-end, isolate d , and stigmatizing classes for the so-called educable mentall y retarded . " I n January 1984 , the U . S . Ninth Circuit Court of Ap­ peals upheld, by a 2-1 margin, Judge Peckham's ruling in the Larry P v. Riles case . Judge W i ll iam B. Enright, the dissenter, noted that proper placement i n E M R (educable mentall y retarded) c l asses is a benefit, not a stigmatic dead-end assignment ; that before the tests in question can

ASSESSMENT PROCEDURES AND SPECIAL EDUCATION O�J TR.IAL

779

Table 24- 1 Representative Court Cases I nvolving Assessment and pmacement of Eth nic Milnority Children and Handicapped Children in Special Education Classes

Case Hobson v. Hansen - 269 F. Supp. 40 I ( D . D . C . 1 967) Diana v. State Board of Educatioll - C-70 3 7 RFT ( N . D . Cal . 1 970)

Pennsylvania Association for Retarded Children v. Commonwealth of Pennsyl­ vania - 343 F. Supp. 279 (E.D. PA 1 972) Wyatt v. Stickney- 344 F Supp. 387 ( M . D . Ala. 1 972) Guadalupe v. Tempe Elementary School District - Stipulation and Order (January 24, 1 972)

Mattie T. v. Holladay - No. DC-75-3 1 -S ( 1 979)

Larry P v. Riles - No . C-7 1 -2270 RFP ( 1 979) and No . 80-4027 DC No. CY 7 1 -2270 i n the United States Court of Appeals for the N inth Circuit ( 1 984)

Parents in Action on Special Education v. Joseph P. Hannon - No. 74 C 3586 ( N . D . I l l . 1 980)

Georgia State Conference of Branches of NAACP v. State of Georgia - Eleventh Circuit Court of Appeals, No. 84-877 1 (October 29, 1 985) and Marshall v. Georgia, U . S . D istrict Court for the Southern District of Georgia, CY 482-233 , June 2 8 , 1 984; Amended August 24, 1 984

Decision A U . S . district court invalidated the District f Columbia school system's educational tracking practices. Spec' ial :Ia ses, however, were permissible as long as testing procedures were rigor LIS and retesting was frequent. A U . S . federal court in California inval idated testing procedures that were used to evaluate Mexican-American children for placement in special education classes. The school system agreed that l i ng u istical l y different children would be tested both in thei r primary language and m Engl ish, that primarily nonverbal tests would be used for the assessment of these children's cognitive skills, and that an interpreter would be used if a bilingual examiner was not available. A U . S . federal court in Pennsyl vania ratified a consent agreement assuring that retarded children would have the right to publicly supported schooling appropriate to thei r needs. Extei1si�-e remedial education programs were then instituted for those retarded children previously denied an education. An Alabama federal court ruled that children in a state institution for the mentally retarded have a constitutional right to treatment . A U . S . district court in Arizona agreed to a stipulated agreement that children could not be placed in educable mentally retarded classes unless they scored lower than 2 standard deviations below the population mean on an approved IQ test administered in the child's own language. It was also stipulated that other assessment procedures must be used in addition to intelligence tests and that parental permission must be obtainC'd for such placements . A U . S . district court i n MissiSSippi approved a consent decree stipulating that (a) classification and placement procedures for special education must be ' evaluated by outs ide experts . (Ii a remedy must be devised to solve the problem l arge numbers of black children in classes for the mentally retarded , and (c) all of misclassificd children are to be identified and given compensatory education through tuloring or Y(lclllionul training, even beyond the age of 2 1 . A U . S . federnl d i strict court in Cal ifornia ruled that standardized inteliigence tests are culturally biased and cannot be used in the assessment of black children for possible placement in educable mentaliy retarded classes . The court fu rther stipulated that the proportion of black children i n classes for the educable mentally retarded must closely match their proportion i n the population. The Ninth Circuit Court of Appeals upheld the district court's verdict. A U.S. federal district court in Ill inois ruled that the WISC , WISC-R, and Stanford-Binet are not racially or culturally biased. Furthermore, when these tests are used in conjunction with other procedures, they do not discriminate against black children in the Chicago public schools. The court believed that the school system was complying with federally mandated criteria for determi ning an appropriate educational program for a child . The court, which heard from some of the same witnesses who testified for the plaintiffs in the Larry P. case , noted that in the wrry P. case the judge failed to u ndertake a detailed examination of the items on the intell igence tests. Without such an examination, the court believed, the issue of cultural bias could not be properly eva1uated. A U . S . court of appeals for the Eleventh District ruled that the State of Georgia did not discriminate against black children by placing them in disproportionate numbers in classes for low achievers or by using standard assessment procedures for the evaluation and placement of black children in special education programs for the educable mentally retarded. The Eleventh C ircuit Court of Appeals affirmed a lower U . S . d istrict court's ruling in favor of the State of Georgia.

780 be labeled as d iscriminatory, there must be evidence that i ntelligence tests resulted in improper placement in the E M R program and no such evidence was presente d ; that the widely recognized IQ tests employed by the defe ndants have long been hailed for their ability to correct the exact abuse complained of in this case - misevaluations and m is­ placement; and that educators have long recognized that subj ective evaluation, uncorroborated by obj ective c rite­ ria, carries enormous potential for abuse and m isplace­ ments based on the personal or cultural values of the eva luator. J u dge E n r i g h t a l so noted t hat the cou rt's decision was striking down the only obj ec tive c riteria for placeme n t . I n J u n e 1 9 8 6 , t h e U . S . Circuit Court o f Appeals issued an amended decision that reaffirmed the d istrict court's finding of violation of federal statutory law, but reversed the finding of violations of the equal protection clause of the Fourteenth Amendment of the Constitution . I n Sep­ tember 1986, the California State Department of Education issued a directive to implement further the federal court's decision . The directive stated that individuall y adminis­ tered intelligence tests were not to be used for the assess­ ment of any black child referred for special education services. A lthough the original case dealt w ith p l acement decisions for EMR c lasses only, the new stipulation covers all special education placements in the public schools .

Parents i n Action o n Special Education v. Joseph P. Hannon In Parents in Action on Special Education v. Joseph P Hannon , a federal court ruled that intell igence tests are not cultura l l y biased against black childre n . This court stipu­ lated that, when used with other criteria in the assessment process, i ntelligence tests comply with federal guidelines concerning the use of nondi scrim inatory procedures . Judge John Grady noted : "There is no evidence i n this record that such misassessments as do occur are the result of racial bias in test items or i n any aspect of the assessment process currently in use i n the C hicago public school system . " He found that the defendants were complying with federal guidel ines.

Georgia Conferences of NAACP v. Georgia I n Georgia Conferences of NAACP v. Georgia , t h e N AACP alleged that t h e State o f Georgia d iscriminated against black children by using evaluation procedures that resulted in their overrepresentation in classes for the edu­ cable mentally retarded. Their complaints para l l eled those

CHAPTER 24 ISSUES IN CONSULTATIOI\ol in the Larry P v. Riles case . Both the trial court and the Eleventh Circuit Court of Appeals rejected the c laims o f the N A ACP The courts noted that there was n o evidence 0 f differential treatment of black and white students . Over-­ representation of black childre n in c lasses for the mentally retarded by itse l f was not sufficient to prove d iscri m ­ ination .

Comment on Assessment Procedures and Special Education on 'frial Three federal courts and two federal appeals courts have now been involved in cases involving overrepresentation o f black children i n classes for the educable mentally re ­ tarded. I n all of these cases, individual intelligence tests , a key diagnostic tool in the placement process , have been characterized by the p laintiffs as being culturall y biased and therefore inappropriate for the assessment of black childre n . Courts in C a l i fornia agreed with this charac­ terization , whereas courts in I l l i nois and Georgia djd not . Obviously, further judicial rul ings w i l l be needed to re­ solve the thorny issues i nvolved in the placement of minor­ ity children in special education classes . As some j urists noted in Daniel Hoffman v. The Board of Education of the City ofNew York (see Chapter 1), the courts may not be the appropriate place to resol ve complex issues concerning the fai rness of tests and appropriate educational procedures .

C O U RT I N T E R P R E TATI O N S O F PL 94-142 A N D R E LATED LEGAL DECISIONS

State and federal courts, as wel l as the Supreme Court, have heard cases dealing w ith the provisions of PL 94-142 and related educational matters. This section reviews out­ comes of some i mportant cases.

Board of Education v. Rowley The U . S . Supreme Court i n June 1982 ruled i n Board of Education v. Rowley ( 80-1002) that school boards are not required to provide services that w i l l enable handicapped children to realize their full potential. By a vote of 6 to 3 , the justices decided that P L 94-142 only requires that handicapped children's education be sufficient to enable them to achieve passing marks and advance from g rade to grade . Amy Rowley, a lO-year-old deaf girl , wanted a sign­ language interpreter in her classroom , but school officials

78 1

CONSULTATION WITH PARENTS said it would be too costly and unnecessary. The majortty of the U . S . Supreme Court agreed with the school board , noting that Congress did not mean to give handicapped children "strict equal ity of opportunity" with other stu­ dents. A m inority of the justices disagreed. They said that the new law was designed to eliminate the effects of [aJ handicap. at least to the extent that the child will be given an equal opportunity to learn, if that is reasonably possible . Amy Rowley, without a sign language inter­ preter, comprehends less than half of what is said in the class­ room - less than half of what normal children comprehend . This is hardly an equal opportunity to learn, even if Amy makes passing grades .

Rettig v. Kent . SchlOo L District A federal appe;als ceo rt s upported a school dist rict's refusal to provide a 16-ye:ar--old autistic adolescent with summer classes and ye:ar-I ou nd therapy in Rettig v. Kent School Districl ('I< ,1 . 9'4-FDR-12 ) . Judge Robert Krupansky of the Sixth Circuit COUirt of Appeals based his decision in part on the Board of Education v. Rowley case . He wrote that states must use procedures that are "reasonably calculated" to enable a hantdic:apped child to receive educational bene­ fit s , and the sc hool was deemed to be u s i n g s u c h procedures . Doe

Bales

v.

Clark

In a similar case tried in V i rginia in 198 1 , Judge Dortch Warriner of the Eighth U . S . Circuit Court of Appeals ruled that a deaf child could be educated in local public schools even though she would get a better education in a state school for the deaf (Bales v. Clark, Oct. 16, 198 1 , Civ A. 80-0568-R, 523 Fed. Supp . p. 1366). Evelyn Ann Bales, a 13-year-old girl, sustained head injuries in an automobile accident in 1977. After two years of rehabilita­ tion , her parents wanted her to attend a private school for handicapped childre n that was located some distance from their home . The school , however, recommended a public special education center closer to her home school district. Judge Warriner ruled (hat neither they nor any other parents have the right under the law to write a prescription for an ideal education for their child and to have the prescription filled at public expense. The law requires an appropriate free education. Efforts to build this requirement into something more will threaten the substantial gains already made in the education of the handicapped .

The j udge concluded : "No language in state or fedaal law can properly be read as mandating that costs may not be considered in determining what education is appropri­ ate for a child - handicapped or nonhandicapped . " Factors such as the differences in travel costs to the two schools "must be considered in determining the appropriateness of the schools," he said. In the above case the parents had also requested reim­ bursement for summe r tutoring that they had obtained for their daughter. Judge Warriner ruled that the child is -not entitled to year-round schooling without showing an irrep­ arable loss of progress during summer months ," and no such loss had been demonstrated in this case.

v.

Boa.rd ()f Education of Montgomery County

I n a ca e s imi.lar to Hoffman v. Board of Education (see E xhibit \-\ in Chapter 1 ) , the Maryland Court of Appeals ruled that parents do not have the right to sue a board of education. In Dare v. Board of Education of Montgomery Counry (Md. Ct. App; 12122/82) ( 5 1 LW 2410 ) , the court ruled that c urts are not the proper place to test the validity of an educdtional decision to place a student in a special education c las ' . Permitting such suits would allow parents to file suits questioning the propriety of all procedures used in the education of c hildre n . The case involved a diagnostic dispute between a school psycholog.i st and a physician . On the basis of a psychologi� cal evaluat ion . the psychologist concluded that the child was brain damaged. functioning at the retarded or bor­ derline i nte-lle 'tual level , and in need of special education . The parents had the child examined by a physician who concluded that the child was dyslexic , rather than brain damaged , and recommended a regular class placement . When the chil d was not retested and was retained in the special education program, the parents sued for damages .

CONSULTAT I O N W I T H PARENTS

Earlier sections of this chapter covered parental rights u nder PL 94- 142 and The Family Education Rights and Privacy Act . This section briefly discusses some issues involved in working with parents of exceptional children and parents of minority c hildren. (Tech niques for i nter­ v iewing parents and presenting assessment information are discussed in Chapter 1 6 . ) I n working with parents who have a handicapped o r special child, always consider how t h e handicap affects the entire family. The family's , as well as the child's, attitudes, feelings , and reactions will have an impact on how the

CHAPTER 24 ISSUES IN CONSULTATION

782 child copes with his or her handicap . Help the parents see their c h i ld's assets as wel l as l i mitation s . Emphasize the active role that they and their c h i ld can play in coping with the disabi lity. Help them to understand that the disability is only one aspect of their child's l i fe , and that difficulties should be dealt with , not avoided . This positive , coping framework contrasts with a succumbing framework that focuses on the negative impact of disabil ity. Details of each framework are shown in Table 24-2 . Negative reactions of parents to their behaviorally dis­ turbed children may be a response to their child's behavior rather than a maj or cause of the behavior. Interaction conflicts are a two-way process: children affect parents and

parents affect chil d re n . Consequently, bidi rectional effects in fami l y i nteractions should always be examined.

Parents of Mentally Retarded Children Parents of mentall y retarded children do not simply "ad­ j ust" to their child's retardation . They experience periodic sorrow, especially at various important developmental milestones. Foll owing are some examples of parental reac­ tions to raising a mentally retarded child (Wikl er, Wasow, & Hatfield , 1 9 8 1 , p . 69) : "Perhaps disappointing would be a better word than sorrow. I fi rmly bel ieve we did have many peaks and •

Table 24-2 C haracteristics of the Coping-Succumbing Frameworks

Coping 1 . The emphasis is on what the person can do . 2 . Areas of l i fe in which the person can participate are seen as worthwhile . 3 . The person is perceived as playi ng an active role i n molding his o r h e r l i fe constructively. 4 . The accomplishments of the person are appreciated in terms of his or her benefits to the person and others (asset evaluation) and not devaluated because they fal l short of some irrelevant standard ." 5 . The negative aspects of the person's l ife , such as the pain that is suffered or difficulties that exist, are felt to be manageable . They are l imited because satisfactory aspects of the person's l i fe are recognized . 6. Managing difficulties means reducing limitations through changes in the social and physical environment as wel l as in the person. Examples are: a. e liminating barriers b. environmental accommodations c. medical procedures d. prostheses and other assistive devices e. learning new skills 7 . Managing difficulties also means living on satisfaclOry terms with one's limitations (although the disability may be regarded as a nuisance and sometimes a burden) . This involves important value changes . 8. The fact that individuals with disabilities can l i ve mean­ ingful l ives is indicated by their participation in valued activities and by their sharing in the satisfactions of l iving.

Succumbing I . The emphasis is on what the person cannot do. 2 . Little weight is given to the areas of l i fe i n which the person can participate. 3. The person is seen as passive, as a victim of misfortune. 4. The person's accomplishments are minimized by highlight­ ing his or her shortcomings (comparative-status evaluation, usually measured i n terms of "normal" standards) ." 5. The negative aspects of the person's l i fe , such as the pai n that is suffered o r d ifficulties that exist, are kept i n the forefront of attention. They are emphasized and exagge­ rated and even seen to usurp all of l i fe (spread). 6. Prevention and cure are the only valid solutions to the problem of disabil ity.

7 . The only way to l i ve w ith the disab i l ity is to resign oneself or to act as if the d isabi lity d id not exist .

8 . The person with a d isabil ity is pitied and his or her life essentially devaluated .

a The coping framework does not preclude instances where the nature of the situation requires that people be ranked on a given dimension (as in competitive activities such as sports and merit examinations). but it does preclude instances where comparative-status evaluation is actually i rrelevant to the situation. Source: Reprinted with permission of the publisher and author from B. A. Wright. Physical Disability - A Psychosocial Approach (2nd ed . ) (New York: Harper & Row, 1 983) . p . 1 95 .

78 3

C O N S U L T A T I O N WITH PARENTS

valley s . There is sadness, but many joyom: nd funn'y moments too . " "When 1 fi rst realized my son was lagging, I felt very guilty and wondered if I had somehow caused it. Then I began to tell mysel f that I'm too inept to cope with a retarded child , al l the while I was coping with him." "1 find each new crisis or situation has to be dealt with, and 1 always pray our family has enough strength and courage to work through the m . So far we have been able to keep the peace . but it isn't easy. " •

•

These and simi lar reactions suggest that families of re­ tarded c h i ldren may need continued services throughout the l i fe span of the retarded individual.

Parents of Autistic Children An autistic di sorder is one of the most baffling childhood behavior disorders (sec Chapter 20) . The parents and family of the autistic child must deal with problems and fee l i ngs unique to their situation . They need professional help in understanding the problem . (This section is based on Morga n , 1984 . ) First, g i ve parents a real istic and cautious interpretation of autism . presenting the child as a unique individual with a special set of problems. Parents may have misconcep­ tions about the disorder, perhaps stemming from a ster­ eotypic image of autism derived from television or maga­ zine articles . Help them recogni7.e that symptoms such as biza rre responses to the e n v i ron m e n t , i n � i I nee on sameness, attachments to objects , and deficient and Ull­ usual language are part of the syndrome of an autistic disorder. Second , help parents to understand their child's level of function ing in cognitive and adaptive (J reas and pos­ sib i l it ies for i�prove!11e nt. Often they want to bel ieve that cognitive impairment is only temporary, and that the child will return to normal when behavioral and emotional prob­ lems are resolved . Parents should be cautioned that iso­ lated abilities- such as early motor development or good rote memory - should not be equated with general intel­ l i gence . Convey to the parents their child's strengths and weaknesses, using age-equivalent scores where appropri­ ate . By interpreting the child's relative skills, you may help parents feel less threatened . and as a result they may become more receptive to your suggestions. Third , assure parents that they are not responsible for the child's refusal to interact with the world. Parents of autistic children often blame themselves for their child's condit ion , because autism often appears to be social and emotional in nature . A prime feature of autism , one that

distinguishes it from other disorders , i s the child's i nability to form affectionate relationships . This is most disturbing to parents . Their feelings of guilt can best be dealt w it h by presenting tbem wi th factual i n formation about the diag­ nosis and causes of a.utism . Knowing that they did not cause their c hild's disorder enables parents to move past thei r guilt and part cipate in treatment programs . Fourth . be prepared to deaJ with other parental reac­ tions, such as anger arnd denial , that may occur when they learn that the i r chil d is handicapped (see Chapter 16). Fifth , phrase statements about prognosis cautiously. Most cases of au tism are severe and long term . Parents can play an active role in - ntervention programs, however. Finally. be prepared to work w i th the autistic c h i ld's sibl ings i n order to hel p them understand thei r brother's or sister's disorder. F,tmil ies who gain an understanding of an autistic disorder wril l be in a better position to accept and help the autistic ch ild in their home .

Parents of Gifted Children Parents of children who are being evaluated for special programs for t he gifted need to be i n formed that the procedures for t he ide t ification of intelligence , creati vity, and related skills re:: fi ect the l i mitations of available assess­ ment instruments. Wh en a child is not accepted for a g i fted program. inform t t1!e p'arents that the child's performance at that particular t i mle d es not meet the specified entrance requirements and �ful l y explain the fi ndings . Understand the parents' dis re·ccmt court cases concerning assessment pro­ ced u res and special e,ducation. 7 . Discuss C(')urt interpretations of PL 94-142 and related legal decision�. 8 . Discuss consultation with parents. 9 . Discuss cons IItation i n courts and i n correctional set­ ting s . 1 0 . Discuss the ch,all e nges o f being an expert witness . I I . Discuss th:: wse of computers in psychological assess­ ment.

_ APPENDIX A ANSWE RS TO TEST-YO U R-SKI LL EXE RC ISES

C HAPTER 8: TEST·YOUR·SK I LL EXERCISES FOR THE WISC·R

Unless it serves to make some important point, do not include such information . 8 . This sentence provides u nnecessary technical information about test structure ("optional subtest") and test procedures ("not used in computing the IQ") . It should be deleted from the report . 9. Thi statement provides technical i nformation that com­ municates Ji((le to the average reader. Providing percentile equiv­ alents of scaled scores is a preferred way of reporting this informat ion . 1 0 . It is not necessary to report the total scaled score; rather, cite t he IQs obtained on the Verbal, Performance, and Full Scales. I I . StallJarJ Jeviation is a technical concept and should not be used in the report . The scaled scores should be interpreted rather than cited . Suggestion : "His range of knowledge is above average ." 1 2 . Most readers will not understand the reference to Ban­ natyne's recategorization syste m . A l so, citing a prorated IQ of 127 may cause confusion. The reader does not know how this IQ compares with the Full Scale IQ. Delete this sentence , substitut­ ing a more useful interpretation of the test scores. 1 3 . "Freedom from Distractibility" is a technical term that may not be understood by most readers of the report. Suggestion: " Frank had difficulties with attention and concentration . " 1 4 . This is a poorly written sentence. It presents technical information not needed in the report, but more importantly, it fails to communicate useful information. It tells neither which abil ity is better developed nor whether the abilities measured by these subtests are wel l or poorly developed. Assuming that her Comprehension score was significantly higher than her Vocabu­ lary score, a suggested sentence is "In the verbal area, Glenda's social reasoning abil ity i s better developed than her word knowledge . " 1 5 . This sentence fails to present useful information . The reader does not know whether these scores reflect strengths or weaknesses. It is not necessary to describe the exact relationships

Unnecessary Technical Information J . This information is too technical to present in a report . Furthermore, the sentence does not give the di rection of the d i fferenc e . The primary focus should be on differences between each subtest and the mean of the Verbal or Performance Scale . The intersubtest comparisons of interest should be made unly after comparisons with the mean of each scale have been made. Also, it is not necessary to present significance levels. 2. Stress what the child did do, not what he 0 1 she d iJ nol Jo . The l atter information serves no useful purpose in a report; it should be deleted. 3. It is not necessary to say in the report that 10 is average and 19 is the ceiling . This technical information about the test is not informati ve . Suggestion: "Bill's social reasoning and verbal com­ prehension are well developed." 4. Delete this sentence from the report unless there is some special significance to the pattern of missed items. If there is, discuss the significance of the pattern. 5. This sentence neither indicates direction nor gives inter­ pretation of the d i fference. I f the Object Assembly score were 12 and the mean Performance Scale score were 8 , you could say "Her attention to detail and perceptual organization skills are relatively strong in comparison with other performance skil ls." 6 . Reference to "5 points" is potentially misleading. The reader does not know whether the 5 points refers to a raw score or a standard score . This sentence should be written to convey the child's knowledge of what is required on the Similarities subtest. If his score is below average (scaled score of 7) and below his Verbal Scale mean, you could say "Bill's ability for conceptual thinking is less well developed (at the 16th percentile) than are his other verbal skill s . " 7 . This sentence fails to communicate useful information.

80 1

802 among the subtest scores. Suggestion: "Her sequencing and v isual memory abilities are weaker than her spatial and percep­ tual organization abil ities ." 1 6 . I t is not necessary to cite the probabil ity level (too tech­ nica l ) . It is preferable simply to say something like " John's Verbal IQ was significantly higher (or lower) than his Performance I Q , " Also, the example fails to indicate whether the Verbal IQ was higher or lower than the Performance IQ. 1 7 . It is not necessary to report the mean score. 1 8 . This sentence w i l l have l ittle meaning to the average reader. I f a l l scores are between 9 and II on the Verbal Scale subtests, you could say "On verbal tasks, his performance was consistently within the Average range . "

Poor Writing 1 9 . The IQ achieved by a child is a specific number. You do not have to say "approximately 9 8 . " The notion of" approximately" is handled by the confidence interval or precision range. Sug­ gestion : "He obtained a Full Scale IQ of 98 ± 6 on the WISC-R . T h i s score is i n the Average classificatio n . " 20. I t is redundant t o say "when compared t o other children o f her chronological age group." I ncluding this phrase every time a child's ability is discussed would urmecessarily lengthen the re­ port . The phrase should simply be deleted . 2 1 . This sentence is poorly constructed and redundant in places. Suggestion: "She has average attention and concentration ski l ls . " 2 2 . This sentence is awkward. One way of expressing the difference between the two scales is as fol lows: 'The 37-point difference between the Verbal and Performance Scales, in favor of the Performance Scale, indicates that v isual nonverbal skills are better developed than auditory processing skills." Other pos­ sible interpretations are discussed in Chapter 8 . 2 3 . This sentence is too colloquial . Suggestion: " A l l o f h i s Verbal scores were in t h e Average range . " 2 4 . This sentence i s confusing and vague. Recommendations should be as clear and concise as possible. Suggestion : "To increase his mathematics ski l l s , John's regular classroom teacher might make math more meaningful by incorporating it into daily activities. For example, he could be shown how to use mathemat­ ics to learn how to buy groceries and to make a bUdget . " 2 5 . T h e term "respectable" is not appropriate for describing a test score. It implies that some scores are "not respectable." Also, the reference to a score of 7, without some explanation, is not informative for the lay reader. Suggestion : "Al l of Mary's abilities appear to be developed at an average leve l , with the exception of short-term memory for digits, which is relatively weak ." 26. Both WISC-R Verbal and Performance Scales measure "mental" abil ities . The writer may have meant to say "verbal" instead of "mental . " 2 7 . The term "statistical factors" is too general , and the phrase "tenor of his performance" is vague. The recommended way to express the precision range is as follows : "Joe obtained an IQ of

APPENDIX A ANSWERS TO TEST-YO UR-SKILL EXERCIISES 123 ± 6. The chances that the range of scores from 117 to 129 includes his true IQ are about 95 out of 100 . " 2 8 . "Conception o f his environment" i s a vague phrase. Sug­ gestion: "Jim's nonverbal abilities are well developed . " 2 9 . This sentence is poorly written. It i s not clear why there i s a "thu s . " Did the child dismantle the design before allowing the examine r to record her performance? Suggestion: "She behaved impulsively when reproducing block designs, dismantling the designs, whether correct or not, as soon as they were completed ." 30. This sentence is too tentative, because the scores are definitely in the Average range. Suggestion : "Her verbal subtest scores are in the Average range ." 3 1 . The wording of these sentences is awkward and is l ikely to confuse most readers. Also, the fi rst sentence is grammatically incorrect because the subject of the sentence ("accuracy") takes a singular verb ("was," not "were"). Suggestion: "On many subtests he failed easy items but passed more difficult ones . " 3 2 . T h i s sentence is awkwardly written . Suggestion : "He showed outstanding abil ity in a task req u iring alertness to detail s . " 3 3 . This sentence is poorly writte n . The wording "appears to be a significance" i s incorrect; the sentence should say "a signifi­ cant difference at the . 05 level of confidence . " It is not necessary to state probab i l ity levels in the report, however. Assuming that the score on Block Design was significantly higher than the score on Object Assembly, the sentence could read: "On performance tasks, her deductive reasoning ability is stronger than her induc­ tive reasoning ability. " 34. The correct word is "peak," not "pee k . " Furthermore, this information, by itself, is not informative. Was the Comprehen­ sion score above average , average, or below average? Was it significantly different from the other scores? Assuming that the Comprehension score of 10 was significantly above the mean of the Verbal subtests, the sentence could read "The area in which she showed the strongest verbal abi lity is social reasoning and judgment (at the 50th percentile) ." 3 5 . The score itse l f is not retarded, although it may reflect a weakness or poorly developed skill . I f the score on Object As­ sembly is l ow, it is preferable to say "Her spatial v isualization skills are not wel l developed, as indicated by her weak perfor­ mance in completing puzzle items."

Technical Errors 36. Scaled scores of 9 or higher on any of the Wechsler subtests do not indicate "poor" ability. A scaled score of9 is only one-thi rd of a standard deviation below the mean scaled score of 10; it is within the Average range . 3 7 . Prorated IQs that refer to c lusters of subtest scores should not be cited in the report . These additional IQs are l ikely to confuse the reader, and l ittle is known about their validity. It is sufficient to report three IQs (Full Scale, Verbal Scale , and Performance Scale) . I n addition, do not refer to esoteric areas, such as Sequencing I and Sequencing I I . 3 8 . The WISC-R does not use mental ages, but i t provides

CHAPTER 8 TEST-YOUR-SKILL EXERCISES F O R THE NI:.C-R test-age equivalems for the 12 subtests. These test ages ,hou_d ble used cautiously. 39. This statement is not accurate because it focuses n omle of the subtests, rather than on the Ful l Scale, to characteiize thle child's overall level of ability. Furthermore, "standard deviati ons;" is a technical term and should not be included in the report . 40. It is not necessary to put in the report the technicail information contained in this sentence (that is. " . . . approache:s significance at the 5 % level . . . ") . Because the IO-point diflerencle is not significant, it is not appropriate to infer that verbal develop·­ ment is better than nonverbal development. This inference should only be made when there is a significant discrepancy between thte Verbal and Performance Scale IQs. 4 1 . A Full Scale I Q of 109 receives an Average classification., ' not an Above Average classification. Suggestion : "Her Full Scahe IQ of 109 ± 6 is classified in the Average range on this test." If there is reason to suspect a higher level of functioning than thle test scores indicate, discllss your concern s . 4 2 . T h i s range from 6 (9th percentile) t o 16 (98th percentile ) indicates both strengths and weaknesses. Do not say that thi s range indicates only good performance. 4 3 . It is preferable to report percentiles for individual sub·­ tests . Classification labe l s should be used only for the Full Scale and occasionally, if desired. for the Verbal and Performance Scales. Suggestion: "Her scores on the WISC-R ranged from the 5th to the 91st percent i l e . " 44. I t is not appropriate t o conclude that a subtest may be;;: "spoiled" or invalid because the score is lower than the score 011 another subtest. A subtest is spoiled when it is improperly admin­ istered or when the child does not attend to the task, not when about his i mmediate environment, asking the examiner several questions about objects in the examination room." 7. The examiner may be trying to convey that the child's abilities were equally developed, but the sentence i s difficult to fol low. Suggestion: "All of her M cCarthy scores were at an average level . " 8 . This sentence i.s awkward. Suggestion: " H e h a d difficulty distinguishing betwee essential and nonessential detail s . " 9 . The last part o f the sentence is awkward and does not provide useful information . Suggestion: "Her drawings were accurate ." 1 0 . This is not only an incomplete sentence - it is an unclear incomplete sentence . Suggestion: "Connie's parents should be informed of the results and helped to accept the findings. " I I . This sentence i s poorly written. We do not know what the word "between" refers to because three things are mentioned. The word "vision" is mentioned without any qualifier. The word "options" is not a possesive and requires no apostrophe. The entire sentence is awkward . Suggestion: "Vocational counseling is recommended to help Henry reevaluate his goal of becoming a corporate pilot, in view of his poor vision and high level of ., anxiety. 1 2 . This sentence has two problems. First, referring to "IQ" as a score is rrdundant . Second, the source of the IQ should be specified. A better way to express this finding would be to say "She obtained �I WISC-R IQ of III , and then present a confidence interval and other descriptive information. 1 3 . The second sentence is vague . The reader has no way of knowing why this illustration was included, or what was unusual about the response. The statement should be clarified. ] 4 . The two ideas in this sentence appear to be in conflict with each other. A l though the term "intell ectual maturity" i s im­ precise, average intelligence would seem to be suggestive of intel lectual maturity. 1 5 . Th i s sentence is poorly written . The colloquial ex­ pressions "j ust flew by in terms of time" and "without any hitch" are not appropriate. The expression "did the tasks quickly and directed" is not clear. Suggestio n : "lohn was cooperative, well motivated, and showed a quick response pattern . " 1 6 . This statement i s confusing because i t is not clear t o what "automatic level" refers. A possible restatement: "lohn's fi ne motor control is poorly devel o ped, as can be seen by his inability to write legibly." 1 7 . A term such as "rapidly" is preferable to the word "ner­ vously," which is too subjecti ve . 1 8 . This sentence is not c lear because we do not know what comparison the writer has in mind. 1 9 . This sentence is awkward . Suggestion: "The child is a small girl who has a pleasant disposition. Rapport was easily established . " 20. The final quotation marks should be placed after the period. "

Poor Writing 1 . These sentences would read more smoothly if they were expressed i n a more conversational manner. Suggestion: "Keith's academic weaknesses are in arithmetic complltation, spel ling, and word recognition ." 2 . These fi ndings should be presented in a more definitive and confident manner. Suggestion : "She scored in the 99th per­ centile overall on the Peabody I ndividual Achievement Test , i ndicating that she possesses strong academic skills." 3. Qualifiers such as "little" should not be used . Suggestion: "Linda appeared tired and became distracted toward the end of the testing ." If desired, the specific behaviors that led to this statement can then be l isted. 4. The examiner seems to be excusing the child's behavior. The information should be reported more objectively. Sug­ gestion: "Michell e became distracted about an hour after the testing began. She frequently stood up to look out the window and asked the examiner when she would be excused . " 5 . This sentence i s ambiguous because the subject o f the sentence is not clear. We do not know who was discussing the future - lohn or his parents - or what kind of expectations the parents have . Suggestion: "In discussing plans for his future, lohn said that his parents expect him to attend college and to pursue a professional career. " 6. The word "environment" should be clarified. Does it refer

807

808 2 1 . Attention and concentration are two separate processes. Consequently, the sentence needs the verb "were ." 2 2 . This sentence is awkward. Most of the last part of the sentence is not needed. Suggestion: "In view of the ideal testing conditions, the results are considered to be valid . " 2 3 . T h e word "which" is not the correct pronoun for "other students . " The proper pronoun is "whom" (with whom). Also, prepositions such as " with" generall y should not appear at the end of a sentence . 2 4 . The correct term is "standardization group," not standard­ ized group. 25. "Error" i s a noun and should not be used as a verb. Even i f "errored" were changed to "made errors," the sentence sti l l would not communicate adequately. Most children w i l l make errors as they reach the ceiling level on a test. The emphasis should be not simply o n the fact that a child makes errors, but on the overall performance, on specific noteworthy items, and on what the errors might reveal . 26. This sentence has many problems, including a spell ing error (extend for extent) and garbled writing. I f the test results are suspect, say so. Suggestion: "The results cannot be consid­ ered valid" or "The results are not valid." A n explanation of why the results are not valid should fol low. 2 7 . Observations of behavior should concentrate on what the child did, not on what was not done . Why does the writer choose to focus on the "egocentric babbling of a less mature child"? Why dots the sentence begin with "although"? Suggestio n : "Jane used mature language i n her conversation . " 2 8 . "Manipulating" has a number of connotations, o n e o f which is "to manage o r influence b y artful or devious ski l l . " Presumably the writer did not have this meaning i n mind. The sentence should be clarified . 29. This is a teasing sentence . First , i n what way does the child "appear" to have a good vocabulary? Second, why was it ironic that he performed below the examiner's expectation? Was the examiner's expectation based on objective data or just a hunch? Third, how far below the expected level did his perfor­ mance fal l ? 3 0 . This sentence is poorly written. Suggestion: "Every effort should be made to encourage Tom to work at a l evel commensu­ rate w ith his abi l ities . " 3 1 . The opening clause " i n l i ne with speed" is confusing and should be deleted from the sentence . Asking about how others performed may not necessarily indicate competitiveness. It could also indicate concern about one's own performance. Suggestion: "He appeared to be concerned with his speed and test perfor­ mance ; several t i mes he asked how his speed of performance compared with others'." 3 2 . The sentence has several problems. First, the word "im­ pacts" is not used properly. Second, the phrase "type of persevera­ tion" is vagu e . Third , the failure to cite specific behaviors leaves the reader with l ittle information . The writer should describe the child's reaction to criticism and how the child's reaction affected her performance . 3 3 . " Malingering," which means pretending to be i l l , is used

APPENDIX A ANSWERS TO TEST-YOUR-SKILL EXERCISES incorrectly i n this sentence . "Lingering" was probably i ntended. 34. This sentence i s poorly written. The style is awkward, and the logic is not c lear. What does the child's sister's hearing d i fficulties have to do with his hearing difficulties? Were there any indications during the test of hearing d ifficulties? If so, they should be described and the test results interpreted accordingly. 3 5 . The word "prioritizes" is jargon. The sentence should be rewritten, and the basis for the comment given. For examp l e : "Speed of performance appeared to b e important to her, a s she asked how fast she was working and seemed to enjoy being timed . "

Inaccurate, Incomplete, or Irrelevant Statements 3 6 . The interpretation of a rigid approach to solving a cog­ nitive task (for example, Block Design) as a possible indication of feel i ngs o f inadequacy is a precarious one . Obtain additional i n formation from other sources before making this type of interpretation . 3 7 . The interpretation may or may not have merit, depending on one's theoretical orientation . It is better to avoid making such interpretations, however, especially when they are based on the absence of data. 3 8 . There is no evidence i n the l iterature to back up these two interpretation s . This statement should be o mitted from the report . 3 9 . "Troublemaker" i mp lies a value judgment . Whenever such a label is used, the source of the label shoul d be cited, as well as the basis for it. Also, it is far from reasonable to assume that good social judgment and grasp of social conventionality might cause a child to become what might be described as a trouble­ maker. 40. "Scatter" is a technical concept and may be misunderstood by lay readers . It is better to use the term "differences" (for example, "contributed to the marked differences between his subtest scores") . 4 1 . The phrase "learning and performance sets" is not clear. A comment on learning skil l s should be based on observable perfor­ mance, and the observations should be presented. Standard psy­ chometric test scores do not allow definitive statements to be made about the rate of learning new sk ills. I n addition , "by the fact that" is a colloquial and awkward expression. 4 2 . This is an interesting statement. The examiner goes out on a l im b , making a statement about future performance on the basis of the IQ alone . The IQ should be used primarily to evaluate current abilities and performance, not to predict future perfor­ mance . There are many factors (for example, intell igence, moti­ vation, and study habits) that affect school performance , and intel ligence is only one of the m . However, you might state , "Her IQ of 134 indicates that she possesses the cognitive ability to excel in schoo l . " The failure of such a child to excel in school suggests the need to look for factors other than inte l ligence that could be inhibiting her academic achievement.

CHAPTER 23: TEST-YOUR-SKILL EXERCISES FOR GENERAL T EST IN1TER.PRETATION AND REPCOR'T WRITING 4 3 . This statement has two problems. First, rather t.han sin'1ply implying that the child should be able to write his name at thus age, it should specifically say so. Second , it should includlc positive statements about what the child can do. Sugge�tion: "Although he was not able to write his name when asked , Jeff was able to write all the letters of the al ph abet. " 44. This statement may be an overgeneralization about Joey" s attention in class . Suggestion: "Joey's abil ity to attend to his work i n class is l imited . For the two hours during which he was observed, Joey attended to his assignments 20 percent of tbe t i me . " 4 5 . There is much controversy about the relationship betwee n learning disabilities and food allerg ies. Therefore the seco d sentence should be deleted . The writer might recommend that Ted be referred to a physician, if he is not already under treatment. 46. It is difficult (if not i mpossible) to separate out items that reflect only natural endowment from those that reflect only edu ­ cational or experiential opportunities. Because all items require some degree of natural endowment together with exposure to the environment, it is misleading to write that some items reflect natural endowment. 47. This description is i ncomplete because it fails to indicate (a) the type of perceptual problem ( for example, visual or au­ ditory or both), (b) the type of verbal expression problem (for example, written word or spoken word or both) , and (c) how the problems were manifested in the test situation . A statement such as this one should be preceded or followed by a discussion of the examiner's observations that led to the conclusion. 48. Performing at the same level on all parts of a test does not necessarily mean that the child is well adjusted in school . Adjust­ ment in school depends on much more than minimal variability of test scores. Performance at the same level on the various areas of a test simply means that the child is not showing variabi l ity in the cognitive areas measured by the test. 49. All items on intelligence tests in one way or another reflect the child's abil ity to learn. Therefore it is misleading to imply that only specific items reflect learning abil ity. Items differ in the extent to which they tap old vs. new learning (or content vs. process , or crystall ized vs. fluid abilities) , but all items to some extent reflect learning ability. 50. This statement may or may not be true . Children with !earning disabilities or with emotional. physical , or sensory disabilities may have average (or above average) scores on an inte l ligence test . The regular classroom may not be suitable for these childre n . It is improper to conclude, solely on the basis of an IQ, that a regular classroom is the appropriate learning en­ vironment. Consider performance on other types of tests, class­ room behavior, and the case history before deciding on an appro­ priate classroom placement recommendation. 5 1 . This information is unl ikely to add to the understanding of the child or test results, and it is potentially damaging to the child. Omit it or replace it by a statement that may give some insight about the: child's feelings . For example, "James expressed resent­ ment about frequent male visitors to his house . "

809

5 2 . This long. 31Wk\waIfd sentence fai l s t o address the issue o f validity. Consequently" th� reader is left in doubt about the valid­ i ty of th e test results. Iff you bel ieve that the results are not valid , clearly ay so in thle rejpoll. For example: "Tim's level of anxiety was high during thle te;!st, which probably invalidated these test results . 5 3 . This conclmiorn nnay be faulty. M inority status of the examinee is not a suffficilent basis for concluding that the test results are biased. Wa!S thle child's native language English? Did the nonverbal scores mlatch the verbal scores? Were the achieve­ ment test scores sllmiliar lto scores on the intelligence test? A l l relevant informatiom mlUst b e considered in arriving a t a decision concerning the val idit:y 0 the test results, and this information should be discusse:d im the report. 54. The writer Deeals to specify the effect to which he or she i s referring . "Practice eflfect:s" i s a vague term. T h e sentence seems to imply that the resulus a-e not valid because they may overesti­ mate the examinee-'s albil i-ties . Suggestion: "Helen was familiar with the test items becmuse she has been tested on this instrument before . Therefore. thle results may overestimate her level of functioning . " �

Inappropriate Abbreviations 55 . The word "perc.:entile" should be written out. 56. Many readers will not know these abbreviations. They are best left out or, if they are needed , described fully. Suggestion: "Mark is in a classroom for children with learning problem s . "

Inappropriate Valllle Judgments 5 7 . This statem.ent reveal s the writer's prej udice. It has no placc in the ff'pOrt A de�cription of the child's appearance and hehavior, however, In.ay be appropriate. Suggestion: "Hector often says that he I S h ngry and asks his teacher for food . The teacher reports thal he sometimes seems to need a bath . " 5 8 . This type o f irlterpretation should only be made when there is sufficient e·vidence . Has the clinician observed the fam­ ily? If so, a statement of the observed behavior should be in­ cluded and then a pos�ible interpretation of the implications of this behavior offered . The report could say, for example , "In a meeting with the family, I observed that l ittle verbal interaction takes place between Joanne and her parents. Perhaps this pattern has hindered the development of Joanne's verbal skills." It is important to recognize that parents with l imited education and l imited verbal abilities may have children who excel on intel­ l igence tests. 59. The writer's prejudice or stereotypes may be showing i n t h i s sentence. N o such interpretation should b e made s imply o n the basis of family size and income . Large families w i t h low income may provide adequate cultural and social enrichment. To ascertain whether the child's family has or has not provided cultural and social enrichment, it is important to interview the parents or visit the horne.

810 60 . This statement is based on hearsay and should be verified. It may be best to omit the reference to the family's source of income. 6 1 . This statement represents a value j udgment, because the behavior in question may not necessarily be irritating to all adults. Suggestion: "Richard's mannerisms tended to irritate me" or "Richard's snorting, popping of gum, and wrigg ling in his chair were irritating" or "Richard's teachers report that he has unpleasant mannerisms that are irritatin g . "

Inappropriate Recommendations 62. The "auditory questions" raised by the evaluation should be described more fully. For example: "It i s recommended that she be evaluated further by a speech pathologist to obtain more information about her apparent difficulties in auditory attention and memory. "

APPENDIX A ANSWERS TO TEST-YOUR-SKILL EXERCISES 63 . Complex ideas are being expressed in these two sen­ tences. The writer's recommendation may be a good one , but it needs clarification. The type of remediation required by the child needs to be stated precisely. Also, the child's test performance must have involved factors that were not related to social percep­ tions , and these factors should be included in the discussion of the child's test performance. 64. I n addition to being awkwardly written, this recommenda­ tion for treatment is not backed by sufficient evidence that treat­ ment is needed. Always present such evidence when making a recommendation. For example : "He may benefit from counseling to help him accept the d ifficult family situation that he described during the test session." 65 . Further examination should not be recommended rou­ tinely without sufficient justification. If periodic examinations are recommended, present the justification for the recommendation.

_

APPEN DIX B

_ _ _ _ _ _ -

PU B LIS H E RS OF TESTS REV I EWED

T. M . Achenbac h , University Associates i n Psychiatry, One South Prospect St . , Burl ington , Vermont 05401. American Association on Mental Deficiency, 5201 Connecticut Ave . . N . W , Washington, D . C . 200 1 5 . American Guidance Service, I nc . , Publ ishers' Building, Circle Pines, Minnesota 5 5014. American Orthopsychiatric Association, I nc . , 1790 Broadway, New York, New York 10019. Bobbs-Merril l Co . , I nc . , 4300 West 62nd St . , Indianapolis. Indiana 46268 . W i l l iam C . Brown Co. , Publisher, 2460 K rp r B l v d , DUPuQIJI:, Iowa 52001. Cali fornia Test Bureau/McGraw-H i l l , Del Monte Research Park. Monterey, California 93940. Consulting Psychologists Press, I nc. , 577 College Ave . , Palo Alto, California 94306. Denver Developmental Materials , Inc . , P. O . Box 20037, Denver, Colorado 80220. The Devereux Foundation Press, 19 South Waterloo Rd . , Devon, Pennsylvania 193 3 3 . D L M Teaching Resources, One D L M Park, Allen, Texas 75002. Economy Company, Box 25308 , 1901 North Walnut, Oklahoma City, Oklahoma 73125. G rune and Stratton, I nc . , I I I Fifth Ave . , New York, New York 1000 3 . Marshall S . H iskey, 5 640 Baldwin. Lincoln , Nebraska 68507. lastak Associates, I nc . , 1526 G i lpin Ave . , Wilmington, Dela­ ware 19806.

Language Research Assoc iates, Inc. , 175 East Delaware PI . , Chicago, I l linois 60611 . Linguametrics G roup. P O . B o x 454, Corta Madera, Cali fornia 94925 . Charles E. Merrill Publ ishing Co. , 1300 A l u m Creek D r. , Co­ lumbus, Ohio 432 16. Modern Curriculum Press, 13900 Prospect Rd . , Cleveland, Ohio 44316. Neuropsychology Pl·es. , 1338 E. Edison St . , Thcson, A rizona 85719 Pro-Ed , 5 341 !r.� \ I Wi!!! Oaks Blvd . , Austin, Texas 7873 5 . The Psychological Corporatio n , P. O . Box 9954, San A ntonio, Texas 78204 . Psychological Test Special ists, Box 144 ! . M issoul a , Montana 59801. Herbert C. Quay. university of M iami , P. O . Box 248074, Coral Gables, Florida 3312d. . The Riverside Publishing C o. , 3 O'Hare Towers, 8420 Bryn Mawr Ave . , Chicago, I l l inois 60631. Science Research Asso iates, 155 North Wacker Dr. , C h icago, I l linois 60606. Siosson Educational Pu blications, I nc . , P. O . Box 28052 , East Aurora, New York 14052 . Stoelting Company, 1350 South Kostner Ave Chicago, Illinois 6062 3 . University o f IIIinoi Press, U rbana, I l l i nois 61801 . Western Psychological Services, 12031 Wilshire B l vd . , Los An­ geles, Cal ifornia 90025 . . •

81 1

_ APPENDIX C M ISCE LLAN E O US TAB LES

See the Contents, p . xv, for a complete listing o f tables in Appendix C .

Table C- I Confidence Intervals for WISC-R Scales

Age level Miz

(6-0-0 through 6- 1 1 -30) 7 112

(7-0-0 through 7- 1 1 -30)

8112

(8-0-0 through 8 - 1 1 -30)

9112

(9-0-0 through 9 - 1 1 -30)

Confidence level Scale

68 %

85 %

90%

95 %

99%

Verbal Scale IQ Performance Scale I Q Full Scale I Q

±4 ± S ± 3

±6 ± 7 ± S

±7 ± 8 ±6

± 8 ± 9 ± 7

± II ± 12 ± 9

Verbal Scale IQ Performance Scale I Q Full Scale IQ

±4 ±S ± 3

±6 ± 7 ± S

±7 ±8 ±6

± 8 ± 9 ± 7

± 10 ± 12 ± 9

Verbal Scale IQ Performance Scale I Q Full Scale IQ

±4 ±4 ± 3

±6 ±6 ± S

±6 ±7 ±S

± 8 ± 9 ± 6

± 10 ± 12 ± 8

Verbal Scale I Q Performance Scale I Q Full Scale IQ

±4 ±4 ± 3

±S ± t: ±S

±6 ±7 ±S

± 7 ± 9 ± 6

± 10 ± 12 ± 8

(Table continues next page)

812

MISCELLANEOUS TABLES

813

Table C·I (cont.)

Age level 1 0 '/2

( 1 0-0-0 through 1 0- 1 1 -30) I tt /2 ( 1 1 -0-0 through 1 1 - 1 1 -30) 1 2 '/2

( 1 2-0-0 through 1 2 - 1 1 -30) 1 3 '/2

( 1 3-0-0 through 1 3- 1 1 -30) 1 4 '/2

( 1 4-0-0 through 1 4- 1 1 -30) 1 5 Vl

( 1 5-0-0 through 1 5- 1 1 -30) 16'/2

( 1 6-0-0 through 1 6- 1 1 -30) Average

CO/lfidellce h'Zl'el Srale

168%

85 %

90 ��

95 %

99%

Verbal Scale IQ Performance Scale IQ Ful l Scale IQ

± 4 ± S ± 3

±5 ±7 ±5

± 6 ± 8 ±5

± 7 ± 9 ± 6

±

9

±

8

Verbal Scale IQ Performance Scale I Q Full Scale IQ

± 3 ± 4 ± 3

±S ± 6 ±4

±6 ±7 ±5

± 7 ± 9 ± 6

± 9 ± II

±

8

Verbal Scale IQ .'erformance Scale I Q Ful l Scale IQ

± 3 ± 5 ± 3

± 5 ±7 ±4

± 5 ±8 ±5

±

±

8

±

8

Verbal Scale IQ Performance Scale I Q Full Scale IQ

± 3 ± 5 ± 3

± 5 ±7 ± 5

±6 ±8 ±S

±

7 ± 10 ± 6

±

9

±

8

± 3 ± S

± 7

± 6 ±8

± 3

± S

± 7 ± 9 ± 6

± 9 ± 12

± 3 ± S

± S

Verbal Scale IQ Performance Scale IQ Full Scale IQ

Verbal Scale IQ Performance Scale I Q Full Scale IQ

± S

± 7

±S

±6 ±8

± 3

± S

Verbal Scale IQ Performance Scale IQ Full Scale IQ

±4 ± 5 ± 3

± 5 ± 7 ± 5

±6 ±8 ±5

Verbal Scale IQ Performance Scale I Q Full Scale IQ

±4 ± 5 ± 3

±S ± 7 ± 5

±6 ± 8

±5

±S

6 ± 9 ± 6

±

± ± ±

±

7

9

6 7

9

± 6

± 7 ± 9 ± 6

± 12

± 12

± 13

±

8

±

9

± 12 ± 8

± 9

± 12 ±

8

±

9

±

8

± 12

NOle. Chapter 2 describes the procedure for computing confidence intervals. For the WISC-R Full Scale IQ, the confidence intervals are obtained by the following procedure : (a) The appropriate SE", for the child's age is located in Table 10 of the WISC-R manual. For a 6-year-old child, SEm 3 .41. (b) =

This SE", is multiplied by the respective z value in order to obta in the confidence intervals for the desired level. At the 68 percent confidence level the SEm is multiplied by I ( I x 3 . 4 = 3 ) At the 99 percent level the SE", is multipl ied by 2.58 ( 2 . 58 x 3 . 41 = 9). The procedure used to arrive at the confidence intervals in this table has been questioned by some psychometrists (for example, Dudek, 1979: Knight, 1983 ) . They argue that confidence l imits should be based on the predicted value of the true score . nOl the obtained score , and that a different formula should be used 10 obtain the SE",. Table L-2 in Appendix L provides confidence intervals based on the predicted true score for the WISC-R and the other Wechsler scales.

814

APPENDIX C

Table C-2 Significant Differences Between W I SC-R Scaled Scores, IQs, and Factor Scores (.05/.0 1 significance levels)

S A V C DS PC PA BD OA CO M

I

S

4/5 4/5 3/4 4/5 4/5 4/5 4/5 3/4 4/5 4/5 4/5

4/5 3/5 4/5 4/5 4/5 4/5 3/5 4/6 4/5 4/6

V

A

C

DS

PC

PSIQ 4/5 4/5 4/5 4/5 4/5 4/5 4/6 4/6 4/6

4/5 4/5 4/5 4/5 3/4 4/5 4/5 4/5

4/5 4/5 4/5 4/5 4/6 4/6 4/6

4/5 4/5 4/5 4/6 4/6 4/6

4/6 4/5 4/6 4/6 4/6

PA

BD

VSIQ 12/15

OA

POIQ FDlQ

4/5 5/6 4/6 5/6

4/5 4/5 4/5

5/6 5/6

CO

VCIQ 12/16 13/18

POIQ 1 4/ 1 9

5/6

Note. Abbreviations: I = Information; S = Similarities; A = Arithmetic; V = Vocabulary; C = Comprehension; DS = Digit Span; PC = Picture Completion; PA = Picture Arrangement; B D = Block Design; OA = Object Assembl y ; CO = Coding; M = Mazes; V S IQ = Verbal Scale IQ; PSIQ = Performance Scale IQ; VCIQ = Verbal Comprehension IQ; POIQ = Perceptual Organization IQ; F D IQ = Freedom from Distractibility IQ. Sample reading: A difference of 4 points between scaled scores on the Information and Similarities subtests is significant at the 5 percent level; a difference of 5 points is significant at the I percent leve l . The first small box shows that a 12-point difference between the Verbal Scale IQ and Performance Scale IQ is needed for the 5 percent level, and a IS-point diffe rence is needed for the I percent level. The values in this table for the subtest comparisons are overly liberal when more than one comparison is made for a subtesl. They are more accurate when a priori planned comparisons are made, such as Information vs. Comprehension or Digit Span vs. Arithmetic. The values in this table are based on the average of the 11 age groups. See Chapter 8, Exhibit 8-1 for an explanation of the method used to arrive at magnitude of differences. See Table C-4 in the Appendix for the procedure used to obtain the Deviation IQs for the Verbal Comprehension, Perceptual Organization, and Freedom from Distractibility factors. Standard errors of measurement for WISC-R factor Deviation IQs obtained from Gutkin (1979) .

Table C-3 Differences Required for Significance When Each WISC-R Subtest Scaled Score Is Compared to the M ean Scalec. Score for Any I ndividual Child

Mean of 6 Verbal Scale subtests

Mean of 5 Verbal Scale subtests a Subtest

Infonnation Similarities Arithmetic Vocabulary Comprehension Digit Span Picture Completion Picture Arrangement Block Design Object Assembly Coding Mazes

Mean of 5 Performance Scale subtests b

. 05

. OJ

. 05

. 0]

2.81 3 . 07 3 . 14 2 . 74 3. 15

3 . 37 3 .68 3 . 76 3 . 29 3 . 78

2 . 94 3 . 22 3 . 30 2 . 86 3 . 32 3 .42

3 . 50 3 . 84 3 . 93 3.41 3 .95 4 .07

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

Mean of 6 Performance Scale subtests

. 05

. 01

. 05

. OJ

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

3.38 3 . 59 2 . 92 3 . 82 3 . 70

4 . 06 4.31 3 . 50 4.58 4 . 43

3 . 55 3 . 78 3 . 03 4 . 03 3 . 89 4 . 03

4 . 22 4 . 50 3.61 4 . 80 4 . 64 4 . 80

-

-

(Table continues next page)

815

M ISCELLANEOUS TABLES Table Col (cont.)

Mean of 1 F Subrestsb

Mean of

1 0 Subresrsc

Mean of 1 1 Subtests'

Mean of 12 Sub!ests

05

. 01

. 0' 5

. 01

. 05

. 01

. 05

. 01

Information Similarities Arithmetic Vocabulary Comprehension Digit Span

3 . 25 3 . 60 3 . 69 3.15 3.7 1

3 . 80 4.21 4 . 32 3 . 69 4 . 35

3.29 3.65 3.75 3. 1 9 3.77 3.89

3 . 85 4 . 27 4 . 38 3 . 73 4.41 4 . 55

3 . 30 3 . 66 3 .76 3 . 20 3 . 78

3 . 86 4.28 4 . 39 3 . 74 4 . 42

3 . 34 3.71 3.81 3 . 24 3 . 83 3 . 96

3 . 89 4 . 32 4.44 3 . 78 4 .47 4.61

Picture Completion Picture A rrangement Block Design Object Assembly Coding Mazes

3 . 86 4.14 3 . 20 4 . 45 4 . 29

4 . 52 4 . 85 3 . 75 5 . 22 5 .02

3 . 92 4. 2 1 3 . 24 4 53 4 . 36

4.58 4 .92 3 . 79 5 . 30 5. 10

3 . 93 4 . 22 3 .2 5 4 . 54 4 . 37 4 . 54

4 . 59 4.93 3 . 80 5.31 5 . 10 5.31

3 . 98 4.28 3 . 29 4.61 4 . 43 4.61

4 . 64 4 . 99 3.83 5 . 37 5.17 5 . 37

Subresr

.

-

I

-

-

-

-

-

-

-

NOle. Table C-3 shows the minimum deviations from an individual's average subtest scaled score that are significant at the .05 and . 0 1 levels. The fol lowing formula, obtained from Davis ( 1 959), was used to compute the deviations from average that l!-re significant at the desired significance levels: D = CR x SEm ( ( Tlm ) _ Z' ) ' where D is the deviation from average , CR is the critical ratio desired, and SEm((Tlm) _ Zd is the standard error of measureme�t of the difference between an average subtest scaled score and any one of the subtes! scaled scores that entered into the average. The standard error of measurement can be obtained by the foliowing formula:

SEm((T1m)

-

Z,)

=

�

SE;' T

-;,;-

+

( ) m

-

m

2 S _, E;"z,

-

where SE;', T is the sum of the squared standard errors of measurement of the m subtests, m is the number of subtests included in the average, Tim is the average of the subtest scaled scores, and SE;,z, is the squared standard error of measurement of any one of the subtest scaled scores. The critical ratio for the 5 percent level ranges from 2 . 58 to 2 . 8 7 , and that for the I percent level from 3 .09 to 3 . 34 , depending on the number of subtests. These critical ratios were obtained by use of the Bonferroni inequality, which controls the fanlilywise error rate at .05 (or . 0 1 ) by setting the error rate per comparison at .05/111 (or . O l lm ) . The fol low ing example illustrates the procedure. We w i l l determine th e minimum deviation required for a child's score on the WISC-R Information subtest to be significantly different from his or her average score on the five standard Verbal Scale subtests (Information, Similarities, Arithmetic, Vocabulary, and Comprehension) at the 95 percent level of confidence. We calculate SE;'r by first squaring and then summing the appropriate average standard errors of measurement for each of the five subtests . These standard errors of measurement appear in Table 10 of the WISC-R manual :

SE;'r

=

( 1 . 19)' + ( 1 3 4)2 + ( 1 3 8)2 + ( 1 . 15)2

+

( 1 . 39)2

=

8 . 37

We determine SE;"z, by squaring the average standard error of measurement of the subtest of interest, the Information subtest:

SE;"z, The number of subtests, Ill , equals 5 . Substituting these values into the formula yields (he following:

SEm((Tlm)

-

Z,J

=

=

( 1 . 19)2

=

1 .4161

�(s)l ( 2) 8 . 37

+

5

5 - 1 .4 1 6 1

-

-

=

1 . 087

The value, 1 . 08 7 , i s then multipl ied by the appropriate z value for the 95 percent confidence level to obtain the nlinimum significant deviation (D). The z value is 2 . 58 using the Bonferroni correction ( . 05/5 = .01).

D a

Digit Span excluded b Mazes excluded C Digit Span and Mazes excluded Source: The figures in the table courtesy of A. B. Silverstein.

=

2.58

x

1.087

=

2 . 81

APPENDIX C

816

Table C-4 Estimated WISC-R Deviation I Q s for Verbal Comprehension, Perceptual Organization, and Freedom from Distracti bi lity Factors

Sum of scaled scores 3 4 5 6 7 8 9 10 II

12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Note,

Estimated Deviation IQ Verbal Comprehension"

Perceptual Organization b

-

-

47 49 50 51 53 54 56 57 59 60 62 63 65 66 68 69 71 72 74 75 76 78 79 81 82 84 85 87 88 90 91 93 94 96 97 99

42 44 46 47 49 50 52 54 55 57 58 60 62 63 65 66 68 70 71 73 74 76 78 79 81 82 84 86 87 88 90 92 94 95 97 98

Freedom frolll Distractibilill'c 41 43 45 47 49 52 54 56 58 60 63 65 67 69 71 74 76 78 80 82 85 87 89 91 93 96 98 1 00 1 02 1 04 1 07 1 09

III

1 13 1 15 1 18 1 20

Estimated Del'imioll IQ

Sum of scaled scores

Verbal Comprehension"

Perceptual OrganizatiOiIO

Freedom frOIll Dist ractibi /ill'"

40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76

1 00 101 1 03 1 04 1 06 1 07 1 09 1 10 1 12 1 13 1 15 1 16 1 18 1 19 121 1 22 1 24 1 25 1 26 1 28 1 29 131 1 32 1 34 1 35 1 37 1 38 1 40 141 1 43 1 44 1 46 1 47 1 49 1 50 151 153

1 00 1 02 1 03 1 05 1 06 1 08 1 10

1 22 1 24 1 26 1 29 131 1 33 1 35 1 37 1 40 142 1 44 1 46 1 48 151 153 1 55 157 1 59

III

1 13 1 14 1 16 1 18 1 19 121 1 22 1 24 1 26 1 27 1 29 1 30 1 32 1 34 1 35 1 37 1 38 1 40 1 42 1 43 1 45 1 46 148 1 50 151 1 53 1 54 1 56 158

-

The formulas used to compute the Deviation Quotients, using subtest scaled scores, are as follows: =

Verbal Comprehension Deviation Quotient Perceptual Organization Deviation Quotient

=

Freedom from Distractibil ity Deviation Quotient

=

( I nformation + Sim i larities + Vocabulary + Comprehension ) + 4 1 . 2 . ( Picture Completion + Picture Arrangement + Block Design + Object Assembly) + 3 6 . 0 . 2.2 ( A rithmetic + Digit Span + Coding) + 3 4 . 1 . 47

1 . 60

, Verbal Comprehension subtests are Information, S i milarities. Vocabulary, a n d Comprehension. b Perceptual Organization subtests are Picture Completion, Picture Arrangement, Block Design. and Object Assembly. Freedom from Distractibility subtests are A rithmetic. Digit Span. and Coding. Source: First two colu mns adapted from Gutkin ( 1978).

C

-

817

MISCELLANEOUS TABLES

Table C·S Differences Required for Significance When Each WIS,C.R SLubtest Scaled Score Is Compared to the Respective Mean Factor Scaled Score for Any I ndividual C hi ld

Mean of

Verbal

Perceptual

Freedom from

Comprehellsion

Organization subresrsb

Disrractibility subreslsC

subre srs" Sublesl I nformation Similarities Arithmetic Vocabulary Comprehension Digit Span Picture Completion Picture Arrangement Block Design Object Assembly Coding Mazes

Mean

Meon of

. 05

. OJ

2 . 62 2 . 82

3 . 20 3 .44

-

2 . 58 2 . 93 --

-

3. J 2 3.54 -

-

-

-

-

-

-

-

--

-

-

-

-

I

. 05

of

. 05

. OJ

-

-

-

-

-

-

-

-

-

2 . 80

3 . 43

-

-

-

-

-

-

3.15 3 . 35 2.78 3 . 52

3.81 4.05 3 . 36 4 . 26

-

-

-

.-

-

.

OJ

-

-

-

-

2 . 85

3 . 50

-

-

-

-

-

-

-

-

3 . 04

3 . 72

-

-

Nme. Table C·5 shows the minimum deviations from an individual's mean factor scaled score that are significant at the .05 and . 0 1 levels. See Note in Table C·3 for an explanation of how differences were obtained . The following Bonferroni corrections were used: .05 = 2 . 500, .01 Comprehension and Perceptual Organization . . 05 = 2 . 39, .Oi = 2 . 93 for Freedom from Distractibility. a Verbal C omp re h ens i o n subtests are Information. Similarities, Vocabulary, and Comprehension. b Perceptual Organization subtests are Picture Completion, Picture Arrangement, Block Design. and Ohject Assembly. c Freedom from Distractibility subtests are Arithmetic, Digit Span, and Coding.

=

3 . 025 for Verbal

APPENDIX C

818 Table C·6 Extrapolated I Q Equivalents of Sums of Scaled Scores for WISC·R

Sum of scaled scores

IQ

Sum of scaled scores

IQ

5 6 7 8

39 41 42 44

44

5 .

.

94 95

Full Scale"

Perfonnance

Verbal

·

155 1 57

Sum of scaled scores

36 37 38 39 39

10 11 12 13 14 ·

·

.

·

·

91 92 93 94 95

IQ

156 158 1 59 1 60 1 62

·

.

1 85 1 86 1 87 1 88 1 89 1 90

161 1 62 1 63 1 63 1 64 1 65

Note. WISC-R regression equations: Verbal IQ Performance IQ Full Scale IQ

=

= =

3 7 . 35 1 3 + 1 . 2552 (scaled score) 3 2 . 6814 + 1 . 3 580 (scaled score) 29. 3456 + . 7U4 (scaled score)

a Wechsler recommends that a Full Scale IQ not be calculated unless raw scores greater than 0 are obtained on at least three Verbal and three Performance Scale subtests.

Table C·7 Estimates of the Differences Obtained by Various Percentages of the WISC·R Standardization Sample When Each WISC·R Subtest Scaled Score Is Compared to the Mean Scaled Score for Any I ndividual Child

Subresr I n formation Similarities Arithmetic Vocabulary Comprehension Digit Span Picture Completion Picture A rrangement Block Design Object Assembly Coding Mazes

Overall average

Performance average

Verbal average 10%

5%

2%

1%

2.9 3.0 3.3 2 .6 3 .2 4.0

3.4 3.6 4.0 3.2 3.9 4.8

4. 1 4.2 4.7 3.7 4.6 5.7

4 5 4.7 5.2 4. 1 5. 1 6.3

10%

3.5 3.6 3.0 3.3 4.3 3.8

5%

4.2 4.3 3.6 3.9 5.1 4.5

2%

5.0 5.1 4.2 4.7 6.0 5.4

1%

5.5 5.7 4.7 5.2 6.7 6.0

10%

5%

2%

1%

3.3 3.3 3.7 3.2 3.5 4.3

3.9 3.9 4.4 3.8 4.2 5. 1

4.7 4.7 5.2 4.5 5.0 6.0

5.2 5.2 5.7 4.9 5.5 6.7

3.8 3.9 3.3 3.8 4.4 4.2

4.5 4.6 :.9 4.5 5.2 5.1

5.4 5.5 4.7 5.4 6.2 6.0

5.9 6.0 5.2 5.9 6.9 6.6

Source: Reprinted with permission of t he publisher and author from A . B . Silverstein, "Pattern Analysis: The Question o f Abnormality Journal of Consulting and Clinical Psychology. 1 984. 52 . p . 93 8 . Copyright 1 984 by the American Psychological Association. . .·

819

MISCELLANEOUS TABLES Table C-8 Probability of Obtai ning Designated Differences Between Individual WISC-R V'erlbal and Performance IQs

Probability of obtaining given or greater discrepancy by chance

6 Vz

7Vz

8Vz

9 Vz

1 0 Vz

I l Vz

1 2 Vz

.50 . 25 . 20 . 10 . 05 .02 .01 .00 1

4.32 7 . 20 8.01 1 0.33 1 2 . 27 1 4 . 59 16. 16 20.66

4.32 7.20 8.01 10.33 1 2 . 27 1 4 . 59 16. 1 5 20.66

4 . 08 6 . 80 7.57 9 . 76 1 1 . 59 1 3 .78 1 5.26 1 9.5 1

3 . 99 6.66 741 9 . 55 1 13 5 1 3 .49 1 4 . 93 1 9. 10

4 .08 6 . 80 7.57 9 . 75 1 1 . 59 1 3 . 77 1 5 .25 1 9 .5 1

3.81 6 . 34 7 . 06 9. 1 0 10. 8 1 1 2 . 85 14.23 1 8 . 20

3 . 83 6.38 7. 1 0 9.15 1 0. 87 1 2 . 93 14.31 18.31

Age level 1 3 V:z

1 4 Vz

1 5 Vz

1 6 Vz

Av. a

6 6.93

4 . 02 6.71 7 47 9 . 62 1 1 . 43 1 3 . 59 1 5 .05 1 9 . 25

4 . 09 6.82 7.59 9 . 78 1 1 . 62 13.81 1 5 . 29 1 9 . 56

4 02 6 . 70 745 9.61 1 1 .4 1 1 3 . 57 1 5 .02 1 9.22

4 .06 6 . 77 7 . 54 9 . 72 1 1 .54 1 3 . 72 15. 19 1 9 43

4.

:.7!

9.94 1 1.81 14 04 1 5 . 54 1 9. 88

NOie. Table C-8 is entered in the column appropriate to the examinee's age. The discrepancy that isj ust l ess than the discrepancy obtained by the examinee is located. The entry in the same row. fi rst colum n . gives the probability of obtaining a given or greater d i sc repancy by chance. For example, the hypothesis that a 6 'I2 - year-old examinee obtained a Verbal-Performance discrepancy of 17 by chance can be rejected at the . 0 1 level of significance. Table C-8 is two-tailed. See Chapter 8 . Exhibit 8- 1 for an explanation of the method used to arrive at magnitude of differences. " Av. Average of J l age groups. =

Table C-9 Percentage of Popu lation Obtaining Discrepancies Between WISC-R Verbal and Performance IQs

Percel1lage obtaining given or greater discrepancy in either direction 50 25 20 10 5 2 I .1

6 Vz

7 Vz

8 Vz

9 Vz

I O Yl

l l Vz

1 2 Vz

1 3 Vz

1 4 Vz

1 5 Vz

1 6 Vz

A v. a

Percentage obtaining given or greater discrepancy in a specific direction

8 . 66 1 4 43 1 6 .06 20 . 7 1 2 4 . 60 29 . 24 32 . 38 4 1 .4 1

8.53 1 4 . 22 1 5 . 83 2041 2 4 . 24 2 8 . 82 3 1 .9 1 40 . 82

8 . 78 1 4 . 64 1 6 . 29 2 1 . 00 24.95 29 . 66 3 2 . 84 4 2 . 00

7 . 74 1 2.9 1 14.37 1 8 .52 2 2 . 00 26. 1 5 2 8 .96 3 7 .04

8 .4 1 14 01 1 5 . 60 20. 1 1 2 3 . 88 2 8 . 39 3 1 .44 40. 2 1

8 02 1 3 . 36 1 4 87 19. 1 7 22 . 77 27 . 07 29.97 3 8 . 34

7.61 1 2 . 68 14. 1 1 18. 19 2 1 . 00 25 .68 28 .44 36.37

8 78 1 4 64 1 6 . 29 2 1 .00 24.95 29 .66 32 84 42 .00

8.28 1 3 . 80 1 5 . 36 1 9. 80 23 .52 2 7 . 96 30.96 39.60

9 . 26 1 5 .43 17. 1 7 22 . 1 3 26 . 30 3 1 . 26 34 . 6 1 44 . 2 7

841 14.01 1 5 . 60 20. 1 1 23.88 2 8 . 39 3 1 .44 40 . 2 1

841 14.01 1 5 . 60 20. 1 1 2 3 . 88 2 8 . 39 3 1 44 40 . 2 1

25 1 2 .5 10 5 2.5 1 .5 .05

A8e level

Table C-9 is entered in the column appropriate to the examinee's age. The discrepancy that is just less than the discrepancy obtained by the examinee is located. The entry i n the same row. first col umn. gives the percentage of the standardization population obtaining discrepancies as large as or larger than the located discrepancy. For example. a 6 '/2 -year-old examinee with a Verba l-Performance discrepancy of 14 on the WISC - R w i l l be found in between 25 and 50 percent of the standardization population. The method used to compute the discrepancy between the Verbal and Performance Scale IQs that reflects the percentage of the population obtaining the d i screpancy is as follows: Discrepancy a ,�)2 2 ,,,, . The first term is the standard deviation of the test. the second is the selected z value, and the last ' is the correlation between the two scales. For exampl e . for a 6 '/2 -year-ol d child the discrepancy between the W ISC-R Verbal and Performance Scale IQs that represents 5 percent of the popu lation is 1 5( 1 .96) )2 - 2 ( .65) 24.60. , Av. Average of II age groups. Note.

=

=

-

=

APPENDIX C

820 Table C- I O Validity Coefficients of Proposed W ISC-R Short Forms

Dyad Short form

V I S C V V S I I I

BD BD BD BD OA PC V S PC V

Short form

r . 906 . 888 . 885 . 87 8 . 87 8 . 868 . 864 . 860 . 85 8 . 85 7

S I I I V S S V A V

V V C S C C V PA V PC

Short form

r

BD BD BD BD BD BD OA BD OA BD

Pentad

Tetrad

Triad

.93 1 . 929 .928 .925 .924 .92 1 .919 .919 .919 .919

I S I S I I I I S S

V V C A V S C S V V

C PA PC V PA C PA PA PC BD

Short form

r

BD BD BD OA BD BD BD BD BD OA

.947 .947 .945 . 944 . 944 . 944 . 944 . 943 . 943 . 943

S S S I

I S I I A A

A A A C V A S V V V

V V V PC PC C C C C C

r

PA PA BD BD BD PA PA PC BD PA

OA BD OA CO CO OA BD BD OA BD

.963 .962 .960 .960 .960 . 960 . 960 . 959 . 959 . 95 8

= Information; S = Similarities; A = Arithmetic; V = Vocabulary ; C = Comprehension; P C = Picture Completion; PA = I>icture Arrangement; BD = Block Design; OA = Object Assembly; CO = Coding. The following formula, obtained from Q. McNemar (Journal of Clinical and Consulting Psychology, 1974, 42, 145-146), was used to compute the part-whole correlations:

Note. Abbreviations: I

r

=

k + EErhj r=="""=�F�==== .In + 2I:r .J k + 2Er h ij g

where EErhj is the sum of the correlations between each of the k subtests and all the other subtests, n is the total number of subtests , Erij is the sum of the correlations between each of the n subtests, and Er h is the sum of the intercorrelations of the k subtests. The term .In + 2Erij becomes a constant g for all computations. The formula was applied to the average subtest intercorrelations for the I I age groups that composed the standardization sample (N = 22(0) to determine the correlations with the Full Scale of all possible short forms of two, three, four, and five subtests. The average subtest intercorre1ations are based on the 10 standard subtests (Digit Span and Mazes are omitted). The standard errors of estimate for the best dyad, triad, tetrad. and pentad are 6 . 34 , 5 . 4 8 , 4 . 82 , and 4 . 04 IQ points, respectively.

Table C- I I Yudin's Abbreviated Procedure for the WISC-R as Modified by Silverstein

Items administered

Multiply score by

Infonnation Similarities Arithmetic Vocabulary Comprehension

Every 3rd Odd only Odd only Every 3rd Odd only

3

Picture Completion Picture Arrangement Block Design Object Assembly Coding

Every 3rd Odd only Odd only Odd only All items

Subtest

Source: Adapted from Silverstein ( 1 968a) .

2 2

3 2

3 2 2 2

MISCELLANEOUS TABLES

82 1

Table C- 1 2 W ISC-R Structure of Intellect Classifications

Item

SOl abbreviation

SOl classification

Description

I nformation

Items I to 3

MMU

Memory for Semantic Units

Thle ab'ilit� to remember isolated ideas or

Items 4, 5

MMR

Memory for Semantic Relation�

Th,;: abil ity to remember meaningful

Items 6, 1 1 , 25 , 28

CMU

Cognition o f Semantic Units

Items

MMS

Memory for Semantic Systems

EMR

Evaluation of Semantic Relations

M MR

Memory for Semantic Relations

MMR

Memory for Semantic Relations

NMR

Convergent Production of Semantic Relations

Items 14, 1 8

MMR

Memory for Semantic Relations

Item 1 5

CMU

Cognition o f Semantic Units

MMS

Memory for Semantic Systems

Items 1 6 , 22

MMI

Memory for Semantic Impl ications

Item

MMR

Memory for Semantic Relations

NMR

Convergent Production of Semantic Relations

Item 1 9

MFS

Memory for Figural Systems

Item 20

MSS

Memory for Symbolic Systems

CMU

Cognition o f Semantic Units

7,

8, 1 0

Item 9

Items 1 2 , 1 3 , 1 7

17

\word meanings

connect ions between items of verbal i:nformat ion Th,� ability to comprehend the meaning of \.vord� or ideas Th,e abil ity to remember meaningfully ordered verbal information Thle ablity to make choices among semantic relationships based on the s i milarity and consistency of the meanings The ability to remember meaningful 'onnect ions between items of verbal lOformation The ability to remember meaningful connect ions between items of verbal I nformation The abil ity to produce a word or idea that conforms to speci fic relationship requirements The ability to remember meaningfu l . 'onnect ions between items of verbal mfonmltiol] The ahilitv to comprehend the meaning of ord s or ideas The ahil ity to remember meaningfully ordered verbal information The abil ity to remember arbitrary connections between pai rs of meaningful elements of information The ability to remember meaningful connect ion between items of verbal i nformation The ability to produce a word or idea that conforms to speci fic relationship requirements The abil ity to remember spatial order or placement of given visual information or to remember auditory complexes of rhythm or melody The ability to remember the order of symbolic information The ability to comprehend the meaning of words or ideas

(Table continues next page)

APPENDIX C

822 Table C· 1 2 (cont.)

Item

sal abbreviation

sal classification

Items 2 1 , 23, 29

NMU

Convergent Production of Semantic Units

Item 24

EMR

Evaluation of Semantic Relations

Item 26

MMR

Memory for Semantic Relations

NMI MMU

Convergent Production of Semantic Implications Memory for Semantic Units

NMU

Convergent Production of Semantic Units

CMU

Cognition of Semantic Units

MMR

Memory for Semantic Relations

CMR

Cognition of Semantic Relations

CMT

Cognition of Semantic Transformations

Item 1 6

CSR

Cognition of Symbolic Relations

Item 1 7

CMR

Cognition of Semantic Relations

CMT

Cognition of Semantic Transformations

Item 27

Item 30

Description The abil ity to converge on the appropriate name, or summarizing word, for any given information The abl ity to make choices among semantic relationships based on the similarity and consistency of the meanings The abil ity to remember meaningful connections between items of verbal information The abil ity to deduce mean ingful information implicit in the given information The ability to remember isolated ideas or word meanings The abil ity to converge on the appropriate name, or summarizi ng word, for any given information The abil ity to comprehend the meaning of words or ideas The ability to remember meaningful connections between items of verbal information

Similarities

Items I to 1 5

The abil ity to see relations between ideas or meanings of words The abil ity to see potential changes of i nterpretations of objects and situations The abil ity to see relations between items of symbol ic information The ability to see relations between ideas or meanings of words The abil ity to see potential changes of interpretations of objects and situations

Arithmetic

Every item

MSI

Memory for Symbolic I mpl ications

CMS

Cognition of Semantic Systems

CMU

Cognition of Semantic Units

The abil ity to remember arbitrary connections between symbols The abil ity to comprehend relatively complex ideas

Vocabulary

Every Item

The abil ity to comprehend the meaning of words or ideas

Comprehension

Every item

EMI

Evaluation of Semantic I mpl ications

Every item

MSS

Memory for Symbolic Systems

The ability to judge the adequacy of a meaningful deduction

Digit Span

The abil ity to remember the order of symbolic information

(Table conrinues next page)

823

MISCELLANEOUS TABLES Table C- 1 2 (cont.)

Item

Sal abbreviation

Sal classification

Description

Picture Completion

CFU

Cognition of Figural Units

EFS

Evaluation of Figural Systems

Item 1 4 (also)

MSS

Memory for Symbolic Syslems

Every item

EMR

Evaluation of Semantic Relations

NMS

Convergent Production of Semantic Systems

CFR

Cognition of Figural Relations

EFR

Evaluation of Figural Relations

Every item

The ability to perceive or recognize figural entities The ability to evaluate a system of figural units that have been grouped in some manner The ability to remember the order of symbolic information

Picture Arrangement

The ablity to make choices among semantic relationships based on the similarity and consistency of the meanings The ability to order information into a verbally meaningful sequence

Block Design

Every item

The ability to recognize figural relations between forms The ability to choose a for m based on the evaluation of what the relations are between the figures or forms in the sequence

Object Assembly

Every item

CFS

Cognition of Figural Systems

C FT

Cognition of Figural Transformations

EFR

Evaluation of Figural Relations

NFU EFU

Convergent Production of Figural Units Evaluation of Figural Units

The ability to comprehend arrangements and positions of visual objects in space The ability to visualize how a given figure or object will appear after given changes, such as unfolding or rotation The ability to choose a form based on the evaluation of what the relations are between the figures or forms i n the sequence

Coding A

Entire subtest

The ability to reproduce correctly a form The ability to judge units of figural information as being similar or different

Coding B

Entire subtest

NSU ESU

Convergent Production of Symbolic Units Evaluation of Symbolic Units

Factorial meaning has not been described The ability to make rapid decisions regarding the identification of letter or number sets

Mazes

Every item

CFI

Cognition of Figural I mplications

The ability to foresee the consequences involved in figural problems

Note. Templates and 501 profile forms as well as five workbooks to accompany prescriptions based on 501 templates are available from the 501 Institute. 343 Richmond, EI Segundo, CA 90245. Source: WISC-R 501 designations obtained from M . Meeker at the 501 Institute.

APPENDIX C

824

Table C · 1 3 Interpretive Rationales, Implications of High and Low Scores, and I nstructional I m p lications for Wechsler Subtests

Ability-

Background facTOrs

Possible implicaTions of high scores

Possible implicaTions of low scores

insTrucTional implications

Information

Verbal comprehension Range of knowledge Long-term memory

Natural endowment R ichness of early environment Extent of school i ng Cultural predilections Interests

Good range of factual knowledge Good range of information Possession of knowledge associated with the cultural and educational environment Good memory Enriched background Alertness and interest in the environment Intellectual ambitiousness Intellectual curiosity U rge to collect knowledge

Verbal comprehension Verbal concept formation Abstract and concrete reason i ng abil ities Capacity for associative thinking Ability to separate essential from nonessential details Long-term memory

A minimum of cultural opportunities Interests and reading patterns

Good conceptual thinking Good ability to see relationships Good ability to use logical and abstract thinking Good ability to discriminate fundamental from superficial relationships Good abi lity to select and verbalize appropriate relationships between two objects or concepts Flexibility of thought processes

Poor range of factual knowledge Poor range of information Poor memory Hostility to a schooltype task Tendency to give up easi l y Foreign background Low achievement orientation Impoverished background

Stress factual material by having child read newspaper articles, discuss current events, and do memory exercises Use other enrichment activities, includi ng calendar activities, science and social studies information, and projects involving animals and their function in society

Poor conceptual think i ng D i ffic ulty in seeing relationships D i fficulty in selecting and verbalizing appropriate relationships between two objects or concepts Overly concrete mode o f thinking Rigidity o f thought processes Negativism

Focus on recognition of differences and l i kenesses in shapes, textures, and daily surroundings Stress language development, synonyms and antonyms, and exercises involving abstract words, classifications, and generalizations

I nadequate abil ity in mental arithmetic Poor concentration D istractibility

Develop arithmetical skills Develop concentration skills

Similarities

Arithmetic

Freedom from distractibilityb and verbal comprehension

Opportunity to acqui re fundamental arithmetical

Facility in mental arithmetic Good ability to apply reasoning skills i n

(Table continues next page)

825

MISCELLANEOUS TABLES Table C· l l (cant.)

Abilirya Numerical reasoning abil ity Mental computation Appl ication of basic arithmetical pmcesses Concentration Attention Short-term memory Long-term memory

Background factors processes

Possible implications of high scores

Possible implications of 1011' scores

the solution of mathematical problems Good ability to apply arithmetical skills ir. personal and social problemsolving situations Good concentration Good abil ity to focus attention Good abil ity to engage in complex thought patterns ( for upper-level items, particularly ) Teacher-oriented student

Anxiety over a school -like task Blocking toward mathematical tasks Poor school achievement ( perhaps associated with rebel lion against authority or with cultural background) Anx iety (e . g . , worry over personal problems)

Use concrete objects to introduce concepts Dri l l in basic skills Develop interesting and "real" problems to solve

Poor word knowledge Poor verbal comprehension Poor verbal sk ills and language development Limited educational or family background DiHiculty in verbalizatio n Foreign language background Verbalization not encouraged in culture

Develop a working vocabulary Encourage child to discuss experiences, ask questions, and make a d ictionary Use other verbal enrichment exercises, including Scrabble , analogy, and other word games

Poor social judgment Failure to take personal responsibility (e . g . , overdependency, immaturity, limited involvement with others) Overly concrete thinking

Help child u nderstand social mores , customs, and societal activities, such as how other children react to things, how the government works, and how banks operate

Instructional implications

Vocabulary

Verbal comprehension Language development Learning ability Fund of information R ichness of ideas Memory Concept formation Long-term memory

Education Cultural opportunities

Good word knowledge Good verbal comprehension Good verbal skills and language development Good family or cultural background Good schooling Good abil ity to conceptual ize I ntellectual striving

Comprehension

Verbal comprehension Social judgment Common sense Use of practical knowledge and j udgment in social situations Knowledge of conventional standards of

Extensiveness of cultural opportunities Ability to evaluate and use past experience Development of conscience or moral sense

Good social judgment and common sense Good abil ity to recognize social demands when practical judgment and common sense are necessary Knowledge of rules of conventional

(Table conrinues next page)

826

APPENDIX C

Table C- 1 3 (cont.)

Abilitya

Background factors

behavior Ability to evaluate past experience Moral and ethical judgment

Possible implications of high scores behavior Good ability to organize knowledge Social maturity Ability to verbalize well Wide experience

Possible implications of low scores

Instructional implications

Difficulty i n expressing ideas verba l l y Doubt-laden individual Creative individual looking for unusual solutions Foreign-born with differing l i fe experiences

Discuss t h e actions of others to help children develop awareness of social relationships and what is expected of them i n terms of the behavior of others Role-play situations. such as reporting fi res, calling police, and obtaining help for plumbing problems

A n x iety Inattention D istractibility A possible learning deficit Difficulty i n auditory sequencing

Emphasize l istening skill s by using sequencing activities, reading a short story and asking child to recall detail s , and seeing whether child can fol low directions Use short and simple directions and repeat when necessary Use other memory exercises and memory games

A n x iety affecting concentration and attention Preoccupation with i rrelevant details Negativism ("nothing is missing") Limited interest in the environment Depression Reality distortion

Focus on visual learning techniques stressing individuc.J parts that make up the whole Use perceptual activities that focui on recognizing objects, describinf objects, and attention to details

Digit Span

Freedom from distractibilityb Short-term memory Rote memory Immediate auditory memory Attention Concentration Auditory sequencing

Ability to passively receive stimu l i

Good rote memory Good immediate recall ability Ability to attend wel l in a testing situation Good ability to attend to auditory stim u l i

Picture Completion

Perceptual organization Ability to differentiate essential from nonessential detai l s Identification of familiar objects (visual recognition) Concentration on visually perceived material

Experiences Alertness to environment

Good perception and concentration Good alertness to details Ability to establish a learning set quickly Good ability to differentiate between essential and nonessential details

(Table continues next pa&e)

MISCELLANEOUS TABLES

827

Table C- 1 3 (cont.)

Ability'

Background factors

Possible implications of high scores

Possible implications of lolV scores

Interest in the environment

Reason ing V isual organization V isual perception (closure) Visual long-term memory

Instructional implications (e.g .. maps and art work) Improve scanning techniques aimed at identifyi ng missing clements in pictures

Picture Arrangement

Perceptual organization Planning abil ity Interpretation of social situations Nonverbal reasoning ability Anention to details Alertness Visual sequencing Common sense Anticipation of consequences Recognition of plot

A minimum of cultural opportunities

Planning ability Ability to anticipate in a meaningful way what results might be expected from various acts of behavior A lertness to detail Forethought Good abil ity to detect sequences Good ability to synthesize parts into inte l ligible wholes

Perceptual organization V isual-motor coordination Spatial visualization Abstract conceptualizing ability Analysis and synthesis Nonverbal reasoning

Rate of motor activity Color vision

Good visual-motor­ spatial integration Good conceptualizing abil ity Good spatial orientation in conjunction with speed. accuracy, and persistence Good analyzing and synthesizing abil ity Speed and accuracy in sizing up a problem Good hand-eye coordination Good nonverbal reasoning ability Good trial-and-error methods

Difficulty with visual organization (sequencing) Difficulty in anticipating events and their consequences Difficulty in seeing cause-and-effect relationships I nattent i veness Anxiety Failure to use cues

Focus on cause·and­ effect relationships. logical sequential presentations . and part-whole relationships Use story completion exercises Discuss alternative behaviors and endings in stories and events

Poor visual-motor­ spatial integration Visual-perceptual problems Poor spatial orientation Poor nonverbal reasoning

Use puzzles, blocks, spatial-visual tasks, perceptual tasks involving breaking down an object and building it up again, and art work with geometric forms and flannel board Focus on part-to­ whole relationships and working with a model or key

Block Design

(Table continues next page)

828

APPENDIX C

Table C· 1 3 (cont.)

Ability-

Background factors

Possible implications of high scores

Possible implications of low scores

Instructional implications

Object Assembly

Perceptual organization V isual-motor coordination Abil ity t o synthesize concrete parts into meaningful wholes Spatial relations

Rate of motor activ ity Fam i l iarity with figures Capacity to persist at a task Experience w ith partwhole relationships Working for an unknown goal

Good visual-motor coordination Good ability to visualize a whole from its parts Abil ity to perceive a whole, with critical understanding of the relationships of the individual parts Successful trial and error Experience in assembling puzzles Persistence

Freedom from distractibil ityb Visual-motor coordination or dexterity Speed of mental operation Psychomotor speed Short-term memory Visual recall Attentional skills Symbol-associative skills

Rate of motor activity Motivation

Visual -motor dexterity Good concentration Sustained energy or persistence Abil ity to learn new material associatively and reproduce it with speed and accuracy Good motivation o r desire for achievement

V isual-motor d i fficulties V isual-perceptual problems Poor planning abil ity Difficulty in perceiving a whole M inimal experience with construction tasks Lim ited i nterest in assembly tasks L imited persistence

Develop perceptual and psychomotor skills through guided practice i n assembl ing parts into familiar configurations Encourage trial-anderror activities Re inforce persistence Work with puzzles and activities centering on recognition of missing body parts Employ construction, cutting, and pasting activities Focus on interpretation of wholes from minimal cues

V isual -motor coordination d i fficulties Distractibil ity V isual defects Poor pencil control Disinterest in a school -l ike task Excessive concern for detail in reproducing symbols exactly Lethargy

Use visual-motor learning exercises. such as having child develop a code for matching geometric figures and numbers, learn Morse Code, and work on tracing activities

Poor visual-motor organization Poor planning efficiency Di fficulty in delay ing action I m pulsivity Poor sustained

Focus on plann ing ski l l s . directional ity, visual discrimination, and other paper-andpencil activities emphasizing

Coding

Mazes

Perceptual organization Planning abil ity Foresight Visual-motor control Eye-hand coordination Attention and

V isual-motor organization Ability to delay action

Good perceptual organization Planning efficiency Speed and accuracy Good ability to fol low instructions Nonimpulsi vity Good sustained

" (Table continues next page)

829

MISCELLANEOUS TABLES Table C- 1 3 (cont.)

Possible implications of high scores

Background factors

Abilitya

POSS'ible I Iliplica ions of low scores

attention

concentration

Instructional implications planning and anticipation Help child evaluate responses prior to emitting them

attention

NOlI'. Select the appropriate impl ication(s) based on the entire test proiOcol and background information. Also see Table C-24 for the WPPSI-R . .. The first entry under "Abil ity" is based on faclOr analytic findings. Other entries are derived from clinical and psychoeducational hypotheses. Freedom from Distract ibil ity for WISC-R Arithmetic. Digit Span. and Coding: Ve rbal Comprehension for W I SC-III Arithmetic and Digit Span: Processing Speed for W I SC- 1 I 1 Coding . Source: Adapted. in pan . from 81all and A l l ison ( 1 96 8 ) . Freeman ( 1 962 ) . Glasser and Zimmerman ( 1 96 7 ) . Kaufman ( 1 975a). Rapaport. G i l l , and Schafer ( 1 968). and Searls ( 1 9 7 5 ) . h

Table C- 1 4 Confidence I ntervals for WPPSI Scales

Age level 4

(3- 1 0- 1 6 through 4-2-29) 4 1/2

(4-3-0 through 4-8 -29) 5

(4-9-0 through 5-2-29) 5 1/2

( 5-3-0 through 5-8-29)

6 (5-9-0 through 6-2-29) 6 1/2

(6-3-0 through 6-7- 1 5 )

Confidence level Scale

68 %

85 %

90%

95 %

99 %

Verbal Scale I Q Performance Scale I Q Full Scale IQ

±4 ±4 ± 3

± S ± 6 ±4

± 6 ±7 ±S

± 7 ± 9 ±6

± 9 ± I I ± 8

Verbal Scale IQ Performance Scale IQ Full Scale IQ

±4 ±4 ± 3

±S ± 6 ±4

±6 ± 7 ±S

± 7 ± 8 ± 6

± 9 ± 10 ± 8

Verbal Scale IQ Performance Scale IQ Full Scale IQ

±4 ±4 ± 3

±S ±S ±4

±6 ±6 ±S

± 7 ±7 ± 6

± 9 ± 10 ± 7

Verbal Scale I Q Performance Scale IQ Ful l Scale lQ

± 3 ± 3 ±3

± S ± S ±4

±6 ±6 ±4

± 7 ± 7 ± S

± 9 ± 9 ± 7

Verbal Scale IQ Performance Scale I Q Full Scale IQ

±3 ±3 ±3

±S ± S ±4

±6 ± 6 ±4

± 7 ± 7 ±S

± 9 ± 9 ± 7

Verbal Scale IQ Performance Scale IQ Full Scale I Q

±4 ±4 ± 3

± S ±6 ± 4

±6 ±7 ±S

± 7 ± 8 ± 6

± 10 ± 10 ± 8

NOle. See Table C- I for an explanation of method used to obtain confidence intervals.

830

APPENDIX C

Table C· 1 5 Significant Differences Between WPPSI Scaled Scores and Between I Q s (.05/.0 I significance levels)

V V A S C Se AH PC M GD BD

3/5 4/5 4/5 4/5 3/5 4/5 4/5 3/4 4/5 4/5

Nofl'. Abbreviations: I

A

3/5 3/4 3/5 3/4 4/5 3/4 3/4 3/5 3/5 =

3/5 4/5 3/4 4/5 3/5 3/4 4/5 4/5

I n formation: V

House: PC = Picture Completion: M

=

=

=

=

3/5 4/5 4/5 3/4 4/5 4/5

Se

AH

4/5 3 /4 3 /4 3/4 3/4

PC

4/5 4/5 4/5 4/5

Arithmetic: S = S imilarities: C

=

3/4 3/5 3/5

M

CD

PSIQ

VSIQ 1 1114

3/4 3/4

4/5

Comprehension: Se = Sentences; A H

Geometric Desi g n : B D = Block Desi g n ; VSIQ

=

Verbal Scale I Q ; PSIQ

=

=

A n i mal

Performance

3 points between scaled scores o n t h e I n formation a n d Vocabulary subtests is s i g nificant a t t h e 5 percent leve l : a

d i fference of 5 points is significant at the Scale IQ i s needed for the

C

4/5 3/4 4/5 3/5 3/4 3/5 3/5

Vocabulary: A

Mazes: G O

Scale I Q .

Sample reading: A d i fference of

S

I

percent level. The small box shows that an I I -point d ifference between the Verbal Scale IQ and Performance

5 percent level . and a 14-point d i fference is needed for the I percent level. The Sentences subtest was nO! included i n

computation o f significant differences for t h e Verbal Scale. The values i n this table for the subtest comparisons are overly l iberal when more than one comparison i s made. They are more accurate when a priori planned comparisons are made . such as I n formation vs. Vocabulary or S i m ilarities v s . Comprehension . The values i n this table are based on the average of the s i x age groups. See Chapter 8. Exhibit 8-1 for an explanation of the method used to obtain magnitude of d i fferences.

Table C· 1 6 Differences Required for Significance When Each WPPSI Subtest Scaled Score I s Compared to the Mean Scaled Score for Any I nd ividual Child

Mean oj 6 Verbal scale sl/bresrs

Mean oj Verbal scale subresr:;" 5

Subresr Information Vocabulary Arithmetic Similarit ies Comprehension Sentences An imal House Picture Completion Mazes Geometric Design Block Design

Mean oJ5 PerJormance scale subresrs

Mean oj

Mean oj Subresrs

JO Subresrs"

II

. 05

. 01

. 05

. OJ

. 05

. OJ

. 05

. OJ

. 05

. OJ

3 . 05 2 . 83 2 . 86 2 . 90 3 . 02

3 . 66 3 . 39 3 . 43 3 .48 3 62

-

3.19 2 . 94 2 . 98 3 .02 3. 1 5 2 . 89

3 . 80 3 . 50 3 . 55 3 . 60 3 . 76 3 . 44

-

-

3 . 55 3 . 25 3 . 29 3 . 34 3.51

4. 1 6 3 . 80 3 . 86 3.91 4. I I

-

3.61 3 . 29 3 . 34 3 . 39 3 . 56 3 . 22

4 . 22 3 . 85 3.91 3 . 96 4. 1 6 3 . 76

-

-

-

-

3 . 26 2.81 2.61 2 . 96 2.91

3.91 3 . 37 3.13 3.55 3 . 49

3 . 84 3 . 22 2 . 94 3 . 43 3 . 36

4 . 50 3 . 78 3 .45 4.02 3 . 94

3 . 90 3.27 2 .98 3 .49 3.4 1

4 . 56 3 . 82 3 .48 4.08 3 . 99

-

-

Note. Table C - 1 6 shows the m i n i m u m deviations from an individuai"s average subtest scaled score that are s i g n i ficant at the .05 and .0 I levels. See the

Note i n Table C-3 for an explanation of how the deviations were obtained. a

Scntcnces subtest excluded

SOl/ree: The figures i n the table counesy of A . B . Silverstein.

83 1

MISCELLANEOUS TABLES Table C- 1 7 Extrapolated I Q Equivalents of Scaled Scores for WPPSI

Verbal Sum of scaled scores

NOTe.

WPPSI

Full Scale

Performance IQ

5

45

95

1 56

Sum of scaled scores

IQ

5 6 7 8

39 40 41 43

92 93 94 95

157 158 1 60 161

Sum of scaled scores 10 II 12 13 14 15 16 17 18 19 20 21 22

IQ 35 36 37 38 38 39 40 40 41 42 43 43

44

Sum of scaled scores

IQ

1 78 1 79 1 80 181 1 82 1 83 1 84 1 85 1 86 1 87 1 88 1 89 1 90

1 56 1 57 157 1 58 159 1 60 1 60 161 1 62 1 62 1 63 1 64 1 65

regression equations: Verbal I Q Performance IQ Full Scale IQ

=

=

=

3 8 . 3946 + 1 . 2366 (sca!ed score) 3 1 . 7940 + 1 . 3624 (scaled score) 2 8 . 1 7 + . 7 2 (scaled score)

IQs of 156 and above reprinted by permission of the publishers and author from A. B. Silverstein, "WISC and WPPSI IQs for the Gifted," PsycilOiogical Reports , 1968. 22, p. 1 I68. Copyright 1968 Psychological Reports. IQs of 45 and below reprinted by permjssion of the publishers and author from A. B . S i l verstein, "WPPSI IQs for the Mentally Retarded ," American lournal ofMenral Deficiency, 73, p . 446. Copyright 1968, American Association on Mental Defic iency. Source:

Table C- I S Estimates of the Differences Obtained by Various Percentages of the WPPSI Standardization Sample When Each WPPSI Subtest Scaled Score I s Compared to the Mean Scaled Score for Any I ndividual Child

Verbal average SUbTeS( I n formation Vocabulary A rithmetic S i m ilarities Comprehension Sentences Animal House Picture Completion M azes Geometric Design Block Design SOl/rce:

Overall average

Performance average

10%

5%

2%

J%

2.9 3.2 3.2 3.3 3.0 3.3

3 .4 3.8 3 .9 3.9 3.5 3.9

4. 1 4.5 4.6 4.6 4.2 4.6

4.5 5.0 5. 1 5. 1 4.6 5. 1

10%

3.6 3 .4 3.3 3.2 3.3

5%

4.3 4. 1 4.0 3.9 3.9

2%

5. 1 4.8 4.7 4.6 4.7

1%

10%

5%

2%

J%

5.6 5.3 5.2 5. 1 5.2

3.2 3.5 3.3 3.7 3 .4 3.6 3.9 3.7 3 .9 3.8 3.6

3.8 4. 1 3.9 4.4 4.0 4.3 4.7 4.4 4.6 4.5 4.3

4.5 4.9 4.6 5.2 4.8 5.1 5.5 5.2 5.5 5.3 5.2

5.0 5.4 5. 1 5.8 5.3 5.7 6. 1 5.7 6. 1 5.9 5.7

Reprinted with permission of the publisher and author from A . B . Silverstei n , "Pattern Analysis: The Question of Abnormal ity." lournal of 1 984, 52, p . 9 3 7 . Copyright 1 984 by the American Psychological Association.

ConslIlTing and Clinical Psychology,

APPENDIX C

832

Table C· 1 9 Probability of Obtaining Designated D ifferences Between I nd ividual WPPSI Verbal and Performance IQs

Probabiliry of obtaining given or greater discrepancy by chance

4

41;2

5

5 1;2

6

61;2

. 50 . 25 . 20 . 10 . 05 .02 .01 . 00 1

3.8 6.6 7.3 9.4 1 1 .2 13.3 14.7 18.8

3.7 6.2 6.9 8 .9 10.6 12.6 13.9 17.9

3.5 6.0 6.7 8.6 10.3 12.2 13.5 17.3

3.3 5.6 6.2 8.0 9.5 1 1 .3 12.5 16. 1

3.3 5.6 6.2 8.0 9.5 1 1 .3 12.5 16.0

3.7 6.2 7.0 8.9 10.7 12.6 14.0 17.9

Age level

Note. Table C- 1 9 i s entered i n the column appropriate to the examinee's age. The discrepancy that i s just less than the discrepancy obtained b y the examinee is located. The entry in the same row, first column, gives the probability of obtaining a given or greater discrepancy by chance. For example, the hypothesis that a 4-year-old examinee obtained a Verbal-Performance d iscrepancy of 17 by chance can be rejected at the . 0 1 level of significance. Tabl e C- 1 9 is two-tailed. See Chapter 8, Exhibit 8-1 for an explanation of the method used to arrive at magnitude of differences.

Table C·20 Percentage of Population Obtaining Discrepancies Between WPPSI Verbal and Performance IQs

Percentage obtaining given or greater discrepancy in either direction 50 25 20 10 5 2 1

.1

4

41;2

5

5 1;2

6

61;2

Percentage obtaining given or greater discrepancy in a specific direction

8. 1 13.8 15.4 19.7 23.5 27 . 9 30.9 39.6

8.6 14.7 16.3 2 1 .0 25.0 29.6 32.8 42 . 0

8.7 14.8 16.5 2 1 .2 25 . 3 30.0 33.2 42.6

8.2 14.0 15.6 20 . 0 23.9 28.3 3 1 .4 40 . 2

7.7 13.1 14.7 1 8.8 22.4 26.6 29.4 37.7

8.8 15 0 16.8 2 1 .5 25.6 30.4 33.7 43 . 2

25 1 2 .5 10 5 2.5 1 .5 . 05

Age level

Note. Tabl e C-20 is entered in the column appropriate to the examinee's age. The discrepancy that is just less than the one obtained by the examinee is

located. The entry in the same row, first column, gives the percentage of the standardization population obtaining discrepancies as large or larger than the located discrepancy. For example, a 4-year-old examinee with a Verbal-Performance discrepancy of I S will be found in between 20 and 2 5 percent of the standardization population. See Table C-9 for an explanation of the method used to arrive at magnitude of differences.

833

MISCELLANEOUS TABLES Table C-2 1 Validity Coefficients of Proposed WPPSI Short Forms

Triad

Dyad Shorl form BD GD A A MA PC PC C BD GD

I I I V I A I A V V

Tetrad

Short form

r . 83 5 . 83 3 . 830 . 82 8 . 82 8 . 82 8 . 82 3 . 82 2 . 82 2 . 820

I I A I V V V A I I

A A C V A A A C V C

Short form

r PC MA PC GD PC GD MA MA BD GD

. 87 8 . 878 . 877 . 877 . 876 . 876 . 875 . 875 . 875 . 875

Pentad

V C A C C A V A C V

I I V

A

I I I I A I

Shorl form

r

GD MA PC PC GD PC A C PC MA

BD BD GD GD BD GD MA MA MA SD

V I V I I I I I V V

. 906 . 904 . 904 . 904 . 904 . 904 .903 . 903 . 903 . 903

A A A V A A

V

V A A

S C C A C S A A C C

r PC PC PC PC PC PC MA MA PC MA

GD MA GD GD GD GD BD GD MA SD

.92 3 . 923 .923 .922 .922 .922 .92 1 .92 1 .92 1 .92 1

Note. A bbrev i at ions : I Information ; V Vocabulary ; A Ari thmet i c ; S S i m i la r itie s ; C Comprehension; PC Picture Completion; Mazes; GD Geometric Design; BD = Block Design. MA Source: Reprinted by permission of the publisher and auth or from A. B. Silverstein, "Reappraisal of the Validity of the WAI S , W ISC, and WPPSI Short Forms," lournal of Consulting and Clinical Psychology, 1970, 34, p. 1 3 . Copyright 1970 by the American Psychological Association. =

=

=

=

=

=

=

Table C-22 Yudin's Abbreviated Procedure for the WPPSI as Modified by Silverstein

Subtest

Source: Adapted from Silverstein ( l 968a).

Item used

Multiply score by

I n formation Vocabulary Arithmetic Similarities Comprehension

Every 3rd Every 3 rd Odd only Odd only Odd only

3 3 2 2 2

A nimal House Picture Completion M azes Geometric Design Block Design

Unchanged Every 3rd Odd only Odd only Odd only

1 3 2 2 2

=

834 Table C-l] WPPSI Structure of I ntellect Classifications

Item

SOl abbreviation

SOl classification

Description

I nformation

Items I , 2 , 3 , 6

MMU

Memory for Semantic Units

Items 4 , 5 , 8 , 1 4

MMR

Memory for Semantic Relations

EMR

Evaluation of Semantic Relations

Items 7 , 1 2 , 20, 2 1

CMU

Cognition of Semantic Units

Item 9

MMR

Memory for Semantic Relations

CMU

Cognition of Semantic Units

Item 1 0

MFU CFS

Memory for Figural Units Cognition of Figural Systems

Item I I

EFU

Evaluation of Figural Units

Item 1 3

CFR

Cognition of Figural Relations

MMU

Memory for Semantic Units

Items I S , 1 7 , 1 9

EMR

Evaluation of Semantic Relations

Items 1 6 , 1 8 , 22

MSS

Memory for Symbolic Systems

Item 23

MMR

Memory for Semantic Relations

Every item

CMU

Cognition of Semantic Units

The abi lity to remember isolated ideas or word meanings The ability to remember meaningful connections between items of verbal i n formation The ability to make choices among semantic relationships based on the similarity and consistency of the meanings The ability to comprehend the meaning of words or ideas The ability to remember meaningful connections between items of verbal information The abi lity to comprehend the meaning of words or ideas The ability to remember given figural objects The ability to comprehend arrangements and positions of visual objects in space The ability to judge units of figural information as being similar or d i fferent The ability to recognize figural relations between forms The abi lity to remember isolated ideas or word meanings The ability to make choices among semantic relationships based on the similarity and consistency of the meanings The abi lity to remember the order of symbolic information The abi lity to remember meaningful connections between items of verbal information

Vocabulary

The ability to comprehend the meaning of words or ideas

Arithmetic

Items I to 4

EFR

Evaluation of Figural Relations

Items 5 to 8

MSU

Memory for Symbolic Units

Items 9 to 20

MSI

Memory for Symbolic Implications

CMS

Cognition of Semantic Systems

The ability to choose a form based on the evaluation of what the relations are between the figures or forms in the sequence The ability to remember isolated items of symbolic information, such as syll ables and words The abi lity to remember arbitrary connections between symbols The abi lity to comprehend relatively complex ideas

(Table continues next page)

835

MISCELLANEOUS TABLES Table C-2) (cont.)

hem

501 abbreviation

501 classification

Description

Similarities

EMR

Evaluation of Semantic Relations

NMS

Convergent Production of Semantic Systems

EMR

Evaluation of Semantic Relations

CMT

Cognition of Semantic Transformations

NMS

Convergent Production of Semantic Systems

EMR

Evaluation of Semantic Relations

Items I I to 1 6

CMT

Cognition of Semantic Transformations

Every item

EM!

Evaluation of Semantic Impl ications

Items I to 5

Items 6, 8, 1 0

Items 7 , 9

The abi l ity to make choices among semantic relatilonsh ips based on the similarity and consi stency of the meanings The abi lity to order information into a verbal ly meani ngful sequence The abi Jity to make choices among semantic relationships based on the s i m ilarity and consistency of the meanings The abi l ity to see potential changes of interpretations of objects and situations The abi l ity to order information into a verbal ly meaningful sequence The abil ity to make choices among semantic relationships based on the s i m ilarity and consistency of the meanings The abi l ity to see potential changes of interpretations of objects and situations

Comprehension

The abi l ity to judge the adequacy of a meaningful deduction

Sentences

Every item

MMS

Memory for Semantic Systems

Every Item

NFU EFU

Convergent Production of Figural Units Evaluation of Figural Units

Every item

CFU

Cognition of Figural Units

EFS

Evaluation of Figural Systems

CFI

Cognition of F igural Impl ications

The abi l ity to remember ordered verbal information

Animal House

The abil ity to reproduce correctly a form ' rhe abI l ity to j udge units 0 1 figural information as being similar or different

Picture Completion

The abil ity to perceive or recognize figural entities The abil ity to evaluate a system of figural units that have been grouped i n some manner

Mazes

Every item

The abil ity to foresee the consequences involved in figural problems

Geometric Design

Every item

NFU

Convergent Production of Figural Units

Every item

CFR

Cognition of Figural Relations

The abil ity to reproduce correctly a form

Block Design

The ability to recognize figural relations between forms

(Table continues next page)

836

APPENDIX C

Table C-23 (cont.)

helll

501

501

abbreviation

classification

EFR

FR

Description

Evaluation of Figural Relations

Convergent Production of F i gu ral Relations

The abil ity to choose a form based on the evaluation of w hat the relations a re between the figures or forms in the sequence The ability to reproduce correct relationships between forms

No/l'. Templates and SO l profile forms as well as five workbooks and modules for training with accompanying prescriptions based on SOl templates are

SOl Institute. 343 Richmond. EI Segundo. CA 90245 . SOl designations obtai ned from M . Meeker at the SOl Institute .

available from the

Source: WPPSI

Table C-24 I n terpretive Rationales and Implications of H igh and Low Scores for WPPSI-R and W PPSI Subtests

Possible implications of high scores

Background factors

Abilin"

Possible implications of low scores

Information

Verbal comprehension Range of k now l e dge Long-term memory

Good range of information Good memory Enriched background A lertness and interest in the environment Intellectual ambitiousness I ntellectual curiosity U rge to col lect knowledge

Natural endowment Richness of early environment Extent of pre-schooling Cultural predilections I nterests

Poor range of information Poor memory Tendency to give up easily Foreign background Low achievement orientation

Vocabulary Verbal comprehension

Language development Learning a b il i ty Fund of in format ion

Education Cultural opportunities

Good verbal comprehension Good family or cultural background Good pre-school i ng Good abil ity to conceptual ize Intel lectual striving

R ichness of ideas Memory Concept formation E x p ressive a b i l ity Verbal fl u e ncy Word know ledge

Poor verbal comprehension Limited educational or family background Difficulty in verbalization Foreign language background Verbal i zation not encouraged in culture

Arithmetic Verbal comprehension N u mcrical reason i ng a bi l ity Mental computat ion Appl icat ion of basi c arithilletical processes Concentration

Opportu n ity to acqu i re fu n dam e nt al ari t hm e t i c a l p rocesses

Facility in mental arithmetic Good concentration

Good abil ity to focus attention

Inadequate abil ity in mental arithmetic Poor concentration Distractibil ity Blocking toward mathematical tasks

( Table continues next page)

837

MISCELLANEOUS TABLES Table C-24 (cont.)

AbilitY'

Pos5ible implications () high scores

Possible implications oj low scores

Good conceptual thinking (e . g . , ability to select nd verbal ize appropriate relationships between twO objects or concepts) Flexibility of thought processes

Poor conceptual thinking (e . g . , difficulty in selecting and verbal izing appropriate relationships between two objects or concepts) Overly concrete mode of thinking Rigidity of thought processes Negativism

Background Jactor5

Attention Short-term memory Long-term memory Nonverbal reasoning abi l ity Quantitative concepts Similarities

Verbal comprehension Logical thinking (items 1 - 1 0) Verbal concept formation (items 1 1 - 1 6) Abstract and concrete reasoning abilities Capacity for associative thinking (items 1 1 - 1 6) Ability 10 separate essential from nonessential details Long-term memory Reasoning by analogy

A minimum of cultural opportunities Interests

Com prehension

Verbal comprehension Social judgment Linguistic skill Logical reasoning Common sense Use of practical k nowledge and judgment in social situations Knowledge of conventional standards of behavior Ability to evaluate past experience Moral and ethical judgment

Extensiveness of cultural opportunities Ability to evaluate and ust: past experience Development of conscience or moral sense

Verbal comprehension Short-term memory Rote memory Immediate auditory memory Attention Concentration Auditory sequencing Verbal facility

Ability to passively receive stimul i

Good social judgment Knowledge of rules of conventional behav ior Good ability to organize knowledge

Social maturity Ability to verbalize well Wide experience

Poor social judgment Failure to take personal responsibility (e . g . , overdependency, immaturity, l imited involvement with others) Overly concrete thinking Difficulty in expressing ideas verball y

Sentences

Good rote memory Good immediate recall ability Ability to attend well in a testing situation

Anxiety Inattention Distractibility Difficulty in auditory sequencing

(Table continues next page)

838

APPENDIX C

Table C-24 (cont.)

A bility"

Possible implications of high scores

Background faclOrs

Possible implications of low scores

Animal House V isual-motor dexterity Good concentration Sustained energy or persistence Ability to learn new material associatively and reproduce it with speed and accuracy Good motivation or desire for achievement

Perceptual organization Attention Goal awareness Concentration Finger and manual dexterity Learning ability

Rate of motor activity

Perceptual organization Ability to differentiate essential from nonessential details Identification of fami liar objects (visual recognition) Concentration o n visually perceived material Alertness to detail Reasoning Visual organization V isual perception (closure) V isual long-term memory

Experiences Alertness to environment

Perceptual organization Planning abi lity Foresight V i sual-motor control

Visual-motor organization

Perceptual organization V i sual-motor coordination Spatial visualization Abstract conceptualizing ability Analysis and synthesis

Rate of motor activity Color vision

V i sual -motor coordination difficulties D istractibil ity V i sual defects Lethargy

Picture Completion

Good perception and concentration Good alertness to details Ability to establish a learning set quickly

A nxiety affect i ng concentration and attention Preoccupation with irrelevant details Negativism ("nothi ng is missing")

Mazes

Good perceptual organization Plann i ng e fficiency Speed and accuracy Good ability to fol low instructions

Poor visual-motor organization Poor planning efficiency D ifficulty in delaying action

Block Design

Good visual-motor-spatial integration Good conceptualizing ability Good analyzing and synthesizing ability Speed and accuracy in sizing up a problem Good hand-eye coordination Good nonverbal reasoni ng ability Good trial-and-error methods

Poor visual-motor-spatial integration V isual-perceptual problems

(Table continues next page)

MISCELLANEOUS TABLES

839

Table C-24 (cont.)

Abiliry'

Background

Poss/b"! implica!/olls oj

implications of

facTOrs

h/gh scores

low scores

Possible

Geometric Design

Perceptual organization Perceptual-motor abil ity V isual - motor organization

MOlOr abil ity

Perceptual organization V isual-molOr coordination Ability to synthesize concrete pans into meaningful wholes Spatial relations

Rate of mOlOr activity Fam i l iarity with figures Capacity to persist at a task Experience with pan-whole relationships Working for an unknown goal

Good perceptual-motor abil ity Good eye-hand coordination

Poor perceptual-motor ability Poor eye-hand coordination Developmental i m maturity

Object Assembly

Good visual-motor coordination Good abil ity to visualize a whole from its parts Abil ity to perceiv� a whole, with critical understanding of the relationships of the individual parts Successful trial and error Experience in assembling puzzles Persistence

V isual-motor d i fficult ies Visual-perceptual problems Poor planning abil ity Difficulty in perceiving a whole M inimal experience with construction tasks Limited i nterest in assembly tasks Limited persistence

NOle. For each indi vidual examinee, select the appropriate implication (or impi ications) listed in the columns only after careful consideration of the e n ti re test protocol and background informat ion . " The first entry under "Ability" is based on faclOr analytic fi nd in g s . The other entries in thi� column are derived from clinical and educational . . . . . i nterpretations of the subtest functions . SOl/ree: Adapted , in pa r t , from Blal! and A l l ison ( 1 968 ) , Freeman ( 1 96 2 ) . G l asser and Zimmerman ( 1 967 ) . Herman ( 1 96 8 ) . Kaufman ( 1 975a ) . Rapaport . G i l l , a n d Schafer ( t 968), Searls ( 1 97 5 ) , and Wechsler ( 1 967).

840

APPENDIX C

Table C·25 Confidence I ntervals for WAIS·R Scales

Age level

Confidence level Scale

68%

85 %

90 %

95 %

99%

1 6- 1 7

Verbal Scale I Q Performance Scale IQ F u l l Scale I Q

±3 ± 5 ± 3

± 5 ± 8 ±4

± 5 ± 9 ±5

± 7 ± 10 ± 6

± 9 ± 13 ± 8

1 8- 1 9

Verbal Scale I Q Performance Scale IQ Full Scale I Q

± 3 ± 5 ± 3

±4 ± 7 ± 4

± 5 ± 8 ± 5

± 6 ± 9 ± 6

± 7 ± 12 ± 8

20-24

Verbal Scale IQ Performance Scale I Q Full Scale IQ

± 3 ±4 ± 3

±4 ±6 ±4

± 5 ±7 ± 4

± 6 ± 8 ± 5

± 8 ± 11 ± 7

25-34

Verbal Scale I Q Performance Scale IQ Full Scale IQ

± 3 ±4 ±2

± 4 ±6 ±3

±4 ± 6 ±4

± 5 ± 8 ± 4

± 7 ± 10 ± 6

3 5 -44

Verbal Scale IQ Performance Scale IQ Full Scale I Q

± 3 ±4 ±2

± 4 ±5 ± 3

± 4 ± 6 ± 4

± 5 ± 7 ± 4

± 7 ± 10 ± 6

45-54

Verbal Scale IQ Performance Scale IQ F u l l Scale I Q

± 3 ±4 ± 3

±4 ± 5 ± 4

±4 ± 6 ± 4

± 5 ± 7 ± 5

± 7 ± 10 ± 7

55-64

Verbal Scale I Q Performance Scale IQ Full Scale IQ

± 3 ±4 ± 3

±4 ±6 ± 4

± 4 ± 6 ±4

± 5 ± 8 ± 5

± 7 ± 10 ± 7

65-69

Verbal Scale I Q Performance Scale IQ Full Scale I Q

± 3 ±4 ±2

±4 ± 5 ± 3

±4 ± 6 ±4

± 5 ± 7 ± 4

± 7 ± 10 ± 6

70-74

Verbal Scale IQ Performance Scale IQ F u l l Scale I Q

± 3 ±4 ± 2

±4 ±6 ±4

±4 ± 7 ± 4

± 5 ± 8 ± 5

± 6 ± 10 ± 6

Average

Verbal Scale I Q Performance Scale IQ Full Scale IQ

± 3 ±4 ± 3

±4 ± 6 ±4

± 4 ± 7 ±4

± 5 ± 8 ± 5

± 7 ± 10 ± 7

Note. See the Note in Table C- l in Appendix C for an explanation of the method used to arrive at the confidence intervals.

MISCELLANEOUS TABLES

84 1

Table C-26 Significant Differences Between WA IS-R Scaled Scores, IQs, and Factor Scores at Ages 1 6 to 1 7 and for the Average of the Nine Age Groups (.05/.0 1 significance levels) D

I

v

A

C

PC

s

Ages 1 6- 1 7 o

3/4 4/5 4/5 4/5 4/5 4/5 3/4 5/6 4/5

3/3 2/3

3/4 3/5 4/5 4/5 3/4 4/6 4/5

3/4 3/4 3 /4 2/3 4/5 3 /4

3/5 4/5

4/5 4/5 3/4

4/5 3/4 4/6 4/5

o

5/6 4/5

3/5 5/6 4/5

5/6 4/5

OA VCIQ

POIQ

POIQ FDIQ

13/17 14/18

1 7/22

4/6 4/5

5/6

VIQ PIQ

3/4 213 3/4

12/16

4/5 3/4

Average of 9 age groups

V A C S PC PA BO OA OS

BD

V IQ PIQ

3/4 2/3 3/4 3/4 3/4 3 /4 3/4 3/3 4/5 3/5

V A C S PC PA BO OA OS

PA

101 13

POIQ FDIQ

3/4 3/4 3/4 3/5 3 /5 4/5

3/4 3/4 3/4 3/4 313 4/5

3/4 4/5 4/5

3/4 =

3/3 3/3 3/4 3/4 3/4 2/3 3/4 3/4

3/4 3/4 3/4 4/5 3/4 4/5

3/4 3/4 4/5 3 /4 4/5 3/5

3/4

=

3/5 4/5 3/4 4/5

4/5 3/4 4/5

3/4 4/5

3/5

3/5

4/5

=

=

4/5

POIQ'

11 114 1 11 1 5

13/ 1 7

4/5

3/4

=

VCIQa

=

=

Information ; D Digit Span; V Vocabulary; A Arithmetic ; C Comprehension ; S Nore. Abbreviations : I Similarities; P C Picture Completion; PA Picture Arrangement; BD Block Design; OA Object Assembl y ; DS Digit jy mbol; VIQ Verbal IQ; PIQ Performance Verbal Comprehension IQ; POIQ Perceptual Organization IQ; FDIQ Freedom fro,n Distractibility IQ. IQ; VCIQ Sample reading: A u i fference or 3 !Joinis between ,cakd scores 011 the Inf0rmation and Digit Spa ll sui>te,ts i, significant at tlie 5 perceni leve l ; a d i fference of 4 points is signi ficant at the I percent leve l . The values in this table for the subtest comparisons are overly liberal when more than one compari"m is made for a subtest. They are more accurate when a priori planned comparisons are made, such as Information vs. Comprehension or Digit Span vs. Arithmet ic . See Chapter 8. Exhibit 8 · 1 for an explanation of the method used to arrive at magnitude of di fference s . See Table C·28 for a description of the factor scores . For ages 1 8 - 1 9 . significant differences between the factor scores are a s follows: =

=

=

=

=

=

a

VCIQ POIQ FDIQ

12/16 12/16

POIQ

15/20

=

=

=

APPENDIX C

842

Table C-27 D ifferences Required for Significance When Each WAIS-R Age-Corrected Subtest Scaled Score Is Compared to the Mean Age-Corrected Scaled Score for Any I ndividual Examinee

Mean of5 Performance Scale subtests

Mean of 6 Verbal Scale subtests Subtest Information Digit Span Vocabulary Arithmetic Comprehension Similarities Picture Completion Picture A rrangement Block Design Object Assembly Digit Symbol

. 05

. 0/

2 . 33 2.91 1 . 77 2 . 73 2 . 85 2 . 93

2 . 78 3 . 47 2. 1 1 3 . 26 3 . 40 3 .49

-

-

-

-

-

-

-

-

-

-

Mean of / / Full Scale subtests

. 0/

. 05 -

-

-

-

. 05

. 0/ 3 07 3 . 92 2.2 1 3 . 66 3 . 84 3 . 95 3 . 98 4 .44 3.2 1 4 . 83 4 . 04

-

-

2 . 59 3.31 1 . 87 3 .09 3 . 24 3 . 34

2 . 92 3. 19 2.47 3.43 2 . 94

3 . 49 3 . 83 2 . 96 4. 10 3.53

3 . 36 3 . 75 2.71 4 .08 3.41

-

-

-

-

-

Note. Table C-27 shows the min i mu m deviations from an ind i v idual's mean subtest scaled score that are significant at the .05 and .0 I leve l s . S e e t h e Note i n Table C-3 for an explanation of h o w t h e deviations were obta ined.

Table C-28 Estimated WAIS-R Deviation IQs for Verbal Comprehension, Perceptual Organization, and F reedom from Distractibility at Selected Ages and for the Average of Nine Age Groups Using Age-Corrected Scores

Slim of agedscaled scores 2 3 4 5 6 7 8 9 10 II

12 13 14 15

/ 6I7

A v.

-

-

-

-

51 52 54 55 57 58 59 61 62 64 65

-

-

51 52 54 55 57 58 59 61 62 64 65

Freedom

Perceptuell Organization

Verbal Comprehension / 6/7

/ 8/9

-

-

46 48 51 53 55 57 59 61 63 65 67 69

52 55 58 61 64 66 69 72 75 78 80 83 86

-

from Distractibiliry Av. -

-

49 51 53 55 57 59 61 63 65 67 69 70

/ 6/7 49 52 55 58 60 63 66 69 72 75 78 81 82 86

AI! 49 51 54 57 60 63 65 68 71 74 77 79 82 85

SlI/n of agedscaled scores 40 41 42 43 44 45 46 47 48 49 50 51 52 53

Perceptllrd Organiwtion

Verbal Comp rehension / 6I7

A I�

1 00 10 1 1 03 1 04 1 06 1 07 1 08 1 10

1 00 101 1 03 1 04 1 06 1 07 1 08 1 10

I I I

I I I

1 1 1 1 1

1 1 1 1 1

13 14 15 17 18

13 14 15 17 18

/ 6I7 1 22 1 24 1 26 1 28 1 29 1 32 1 35 1 37 1 39 141 143 1 45 1 47 148

/ 8/9 -

-

-

-

-

-

-

-

-

-

Freedoll/

ji-Oll/ Distracti bility A I�

121 1 23 1 25 1 27 1 29 131 1 33 1 35 1 37 1 39 141 1 43 1 45 147

/ 6/7

Av.

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

(Table continues next pa.{e)

843

MISCELLANEOUS TABLES Table C-28 (cont.) 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39

66 68 69 71 72 73 75 76 78 79 80 82 83 85 86 87 89 90 92 93 94 96 97 99

66 68 69 71 72 73 75 76 78 79 80 82 83 85 86 87 89 90 92 93 94 96 97 99

I

72 74 76 78 80 82 84 86 88 90 93 95 97 99 101 1 03 1 05 1 07 1 09 III

1 14 1 16 1 18 1 20

89 92 94 97 1 00 1 03 1 06 1 08

73 75 77 79 81 83 85 87 89 90 93 95 97 99 101 103 105 107 1 09 1 10 1 13 ! 15 1 17 1 19

!II

1 14 1 17 1 20 1 22 1 25 128 131 1 34 1 36 1 39 1 42 1 45 1 48 1 50 -

88 91 93 96 99 1 02 1 05 1 07 1 10 1 13 1 16 1 19 121 1 24 127 1 30 1 33 1 35 1 38 141 1 44 147 1 49

89 92 95 98 101 1 04 1 07 1 10 1 13 1 15 1 18 121 1 24 1 27 1 30 1 33 1 36 1 37 1 42 1 44 1 47 1 50 153 -

54 55 56 57 58 59 60 61 62 63 64

65 66 67 68 69 70 71 72 73 74 75 76

-

=

1 20 121 1 22 1 24 1 25 1 27 1 28 1 29 131 1 32 1 34 1 35 1 36 1 38 1 39 141 1 42 1 43 1 45 1 46 148 1 49 1 50

1 20 121 1 22 1 24 1 25 1 27 1 28 1 29 131 [ 32 1 34 1 35 1 36 1 38 1 39 141 1 42 143 1 45 1 46 1 48 1 49 1 50

151 153 1 56

-

1 49 151 1 53

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

=

Nore. Verbal Comprehension Information, Vocabulary. Comprehension, Similarities: Perceptual Organization for all ages except 18-19 Picture Completion, Block Design. Object Assembly: Perceptual Organization for ages 18-19 Block Design and Object Assembly; Freedom from Distractibil ity Digit Span. Arithmetic; Av. Ave rage of the n i ne age groups. The formulas used to compute the Deviation Quotients, using age-corrected subtest scores, are a� follows for ages 16- 1 7 : =

=

=

Verbal Comprehension Deviation Quotient = 1 . 4 x ( Information, Vocabulary, Comprehension, Similarities) + 44 Perceptual Organization Deviation Quotient 2 . 1 x (Block Design, Objec( Assembly, Pict u re Completion) + 38 Freedom from Distractibi lity Deviation Quotient 2 . 9 x (Digit Span. Arithmetic) + 43 =

=

The formula used to compute the Deviation Quotient. using age-corrected subtest scores. is as follows for ages 18-19: Perceptual Organization Deviat ion quotient

=

2.8

x

( Block Design. Object Assembly)

+

44

The formulas used to compute the Deviation Quotients, using age-corrected subtest scores. are as fol lows for the average of the nine age groups: Verbal Comprehension Deviation Quotient Perceptual Organization Deviation Quotient Freedom from Distractibility Deviation Quotient

=

=

=

1.4 2.0 2.8

x

x

x

( Information , Vocabulary, Comprehension , Similarities) + 44 (Block Design. Object Assembly, Picture Completion) + 4 1 (Digit Span, Arithmetic) + 43

The SE,,,s for the three factors are as follows: Verbal Comprehension : SE", Perceptual Organization: SE", Freedom from Distractibil ity : SE",

= =

=

3.00 for ages 16-17 and SE", = 3 . 00 for the average of the nine age groups. 5 . 8 1 for ages 16- 17 and SE", = 4 . 74 for the average of the nine age groups. 6 . 54 for ages 16-17 and SE", = 4 . 97 for the average of the nine age groups.

Table C-29 D ifferences Required for Significance When Each WAIS-R Subtest Age-Corrected Scaled Score Is Compared to the Respective Mean Factor Age­ C orrected Scaled Score for Any I ndividual Examinee Mean of Verbal Comprehension slIbtests" Ages 1 6- 1 7

00 � �

Ages 1 8- J 9

A verage

Ages 1 6- 1 7

Subtest

. 05

. 01

. 05

. 01

. 05

01

I nformation Digit Span Vocabu lary Arithmetic Comprehension Similarities Picture Completion Picture Arrangement Block Design Object Assembly Digit Symbol

1 . 94

2 . 34

1 . 92

2.31

2 . 09

2.52

-

1 . 52 -

2 .40 2 . 60 -

-

1 . 83 -

2 . 90 3. 14 -

-

1 . 50 -

2.37 2 . 32 -

-

1 .8 1 -

2 . 86 2.81 -

Mean of Freedom from Distractibility subtestsC

Mean of Perceptual Organization subtestsb

-

1 . 68 -

2 .48 2 . 54 -

-

2 . 03 -

3 . 00 3 . 07 -

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

. 05

. 01

Ages 18- 19 . 05

. 01

Average . 05

. OJ

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

2 . 85

3 . 50

-

-

-

-

2 . 47

3 . 03

-

2 . 45 3 . 34 -

-

3.01 4. 1 1 -

2 .05 2 .05 -

2.58 2.58 -

-

2 .22 2 76 -

-

2 . 73 3 . 40 -

Ages 1 6- 1 7 . 05

. 01

Ages 1 8- 1 9 . 05

. 01

Average . 05

. 01

-

-

-

-

-

-

2 . 10

2 . 63

2 . 05

2.57

1 . 88

2 . 36

2. 10

2 . 63

2 .05

2 . 57

1 . 88

2 . 36

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

Note. Table C-29 shows the minimum deviations from an individual's mean factor scaled score that are significant at the .05 and .0 I levels. See the Note in Table C - 3 for an explanation of how d i fferences were obtained. Bonferroni corrections were used. " Verbal Comprehension subtests are Information. Vocabulary, Comprehension, and Similarities . b Perceptual Organization subtests are Picture Completion. Block Design, and Object Assembly for ages 1 6- 1 7 and average; subtests are Block Design and Object Assembly for ages 1 8 - 1 9 . Freedom from Distract ibility subtests are Arithmetic and Digit Span.

c

MISCELLANEOUS TABLES

845

Table C-30 Extrapolated WAIS-R IQ Equivalents of Sums of Scaled Scores (Not Age-Corrected) for Verbal Scale, Perlormance Scale, and Full Scale

Verbal Scale Sum of scaled scores

1 6- / 7

1 8- 1 9

20-24

25-34

35-44

45-54

55-64

65-69

70- 74

1 10 1 09 1 08 1 07 1 06 1 05 1 04 103 102 101 1 00 99 98 97 96 95 94 93 92 91

1 72 171 1 70 1 69 1 68 1 67 1 66 1 64 1 63 1 62 161 1 60 1 59 1 58 1 56 1 55 1 54 1 53 1 52 151

1 69 1 68 1 66 1 65 1 64 1 63 1 62 161 1 60 1 59 1 57 1 56 1 55 1 54 1 53 1 52 151

1 63 1 62 161 1 60 1 59 1 58 1 57 1 55 1 54 153 1 52 151

161 1 60 1 59 1 58 1 57 1 56 1 54 1 53 1 52 151

1 62 161 1 60 1 59 1 58 1 57 1 55 1 54 1 53 1 52 151

1 63 1 62 161 1 60 159 1 58 157 1 55 1 54 1 53 1 52 151

1 63 1 62 161 1 60 1 59 158 1 57 1 56 1 55 1 54 1 53 1 52 151

1 65 1 63 1 62 161 1 60 1 59 158 157 1 56 1 55 1 54 153 1 52 151

1 66 1 65 ! 64 1 62 161 1 60 1 59 158 1 57 1 56 1 55 1 54 1 53 1 52 151

Age

Performance Scale Sum of

Age

scaled scores

90 89 88 87 86 85 84 83 82 81 80 79 78 77 76 75 74 73 72

1 6- 1 7

1 8- 1 9

20-24

25-34

35-44

45-54

55-64

65-69

70- 74

1 67 1 65 1 64 1 63 161 1 60 158 1 57 1 56 1 54 1 53 1 52

1 63 1 62 161 1 59 1 58 1 56 1 55 1 54 1 52 151

1 57 1 55 1 54 1 53 151

1 59 1 58 1 57 155 1 54 1 53 151

1 62 1 60 1 59 1 58 1 56 155 1 54 153 151

1 65 1 64 1 62 161 1 60 1 59 1 57 1 56 1 55 1 53 1 52 151

1 73 1 72 1 70 1 69 1 68 1 66 1 65 1 64 1 62 161 1 60 1 58 1 57 1 56 1 54 1 53 1 52

1 84 1 82 181 1 79 1 78 1 76 1 75 1 74 1 72 171 1 69 1 68 1 66 1 65 1 63 1 62 1 60 1 59 158

1 92 1 90 1 89 1 87 1 86 1 84 1 83 181 1 80 1 78 1 77 1 75 1 73 1 72 1 70 1 69 1 67 1 66 1 64

(Table continues next page)

APPENDIX C

846 Table C-30 (cont.)

Performance Scale Sum of scaled scores

Age

1 6-1 7

1 8- 1 9

20-24

25-34

35-44

45-54

55-64

71 70 69 68 67 66 65 64

65-69

70- 74

1 56 155 153 152

163 161 1 60 1 58 1 56 155 153 152

Full Scale Sum of scaled scores

1 6-1 7

1 8- 1 9

20-24

25-34

35-44

45-54

55-64

65-69

70- 74

200 1 99 198 1 97 1 96 1 95 1 94 193 192 191 1 90 1 89 1 88 1 87 1 86 1 85 1 84 1 83 1 82 181 1 80 1 79 178 1 77 1 76 1 75 1 74 1 73 1 72 171 1 70 1 69

1 74 1 74 1 73 172 172 171 1 70 1 69 1 69 1 68 1 67 1 67 1 66 1 65 1 65 1 64 1 63 1 63 1 62 161 161 1 60 1 59 1 59 158 157 157 1 56 155 1 55 1 54 1 53

1 72 171 171 1 70 1 69 1 69 1 68 1 67 1 67 1 66 1 65 1 65 1 64 1 63 1 63 1 62 161 161 1 60 1 59 1 59 1 58 157 157 1 56 1 55 1 55 1 54 1 53 1 53 1 52 151

1 66 1 65 1 65 1 64 1 63 1 63 1 62 1 62 161 1 60 1 60 1 59 158 1 58 1 57 1 56 1 56 1 55 1 54 1 54 1 53 1 52 1 52 151 1 50

1 65 1 65 1 64 1 63 1 63 1 62 161 161 1 60 1 60 1 59 158 1 58 1 57 1 56 1 56 1 55 1 54 1 54 153 152 152 151 1 50

1 67 1 66 1 66 1 65 1 64 1 64 1 63 1 63 1 62 161 1 61 1 60 1 59 1 59 1 58 1 57 1 57 1 56 1 56 1 55 1 54 1 54 1 53 1 52 1 52 151 1 50

1 69 1 68 1 68 1 67 1 67 1 66 1 65 1 65 1 64 1 63 1 63 1 62 161 161 1 60 1 59 1 59 158 1 58 1 57 1 56 1 56 1 55 1 54 1 54 1 53 1 52 1 52 151 1 50

1 73 172 171 171 1 70 1 69 1 69 1 68 1 67 1 67 1 66 1 66 1 65 1 64 1 64 1 63 1 62 1 62 161 1 60 1 60 1 59 158 1 58 1 57 1 56 1 56 1 55 1 54 1 54 1 53 152

178 1 78 1 77 1 76 1 76 175 1 74 1 74 1 73 1 72 1 72 171 1 70 1 70 1 69 1 68 1 67 1 67 1 66 1 65 1 65 1 64 1 63 1 63 1 62 161 161 1 60 1 59 1 59 1 58 1 57

1 83 1 83 1 82 181 181 1 80 1 79 1 79 1 78 1 77 1 76 1 76 175 1 74 1 74 1 73 1 72 1 72 171 1 70 1 69 1 69 1 68 1 67 1 67 1 66 1 65 1 65 1 64 1 63 1 62 1 62

Age

(Table cOlllinues next page)

847

MISCELLANEOUS TABLES Table

ColO

(cont.) Full Scale

of scaled

Sum

scores

1 68

1 67 1 66 1 65

Age

1 6- 1 7

1 8- / 9

1 53 1 52 151 151

151

20-24

35-44

25-34

45-54

1 64 1 63 1 62 161 1 60

55-64

65-69

70- 74

1 52 151 1 50

1 57 1 56 1 55 1 55 1 54

161 1 60 1 60 1 59

153 153 1 52 151 1 50

1 59 158 1 57

1 56

1 58 1 58 157 1 56 1 55 1 55 1 54 153 1 53 1 52 151 151

1 55

1 54

1 53

NOie. WA IS-R reg�ess;on equations for the Verbal Scale are as follows: 1 6- 1 7 : 1 8- 1 9 : 20-24: 25-34: 35-44: 45-54: 55-64: 65-69 70- 74:

Verbal Verbal Verbal Ve rbal Verbal Ve rbal Ve rbal Verbal Ve rbal

Scale Scale Scale Scale Scale Scale Scale Scale Scale

IQ IQ IQ IQ IQ IQ IQ IQ

IQ

=

=

= =

= =

=

=

=

47.0969 + I 1 386 (scaled 45. 2039 + I . 1 22 1 ( sca l ed 43.429 1 + 1 . 0879 (scaled 3 9 . 8336 + 1 . 1 02 2 (scaled 42.9675 + 1 . 08 1 6 (scaled 44 . 5330 + 1 . 07 1 5 (scaled 45.5278 + 1 . 072 3 (scaled 46 . 5225 + 1 . 073 1 (scaled

4 7 .5 1 7 1

+

score) score) score) core) score) score) score) score) 1 0739 ( scaleLi score)

WA I S- R regression equations for the Performance Scale are as follows: 1 6- 1 7 : 1 8- 1 9 : 2 0- 2 4 : 25-34: 35-44: 45-54: 55-64 : 65-69 : 70-74:

Performance Scale Performance Scale Performance Scale Performance Scale Performance Scale Performance Scale Performance Scale Performance Scale Performance Scale

IQ IQ IQ IQ IQ IQ IQ IQ IQ

=

=

=

=

=

= =

=

=

41 . 8642 + 1 . 3 888 (scaled score) 4 1 .0293 + 1 . 3 5 8 1 (scaled score) 40.4008 + 1 . 29 1 1 (scaled score) 4 3 . 4307 1 . 2857 (scaled score) 4 7 . 5 354 + 1 . 2666 (scaled score) 50. 5086 + 1 . 2709 (scaled score) 52 . 1 1 49 + 1 . 3432 (scaled score) 5 2 . 4798 + 1 . 4589 (scaled score) 5 3 . 0 1 45 + 1 . 5437 (scaled score)

WA IS-R regression equations for the Full Scale are as follows: 1 6- 1 7 : 1 8- 1 9 : 20-24: 25-34: 3 5 -44: 45-54: 55-64 : 65-69 : 70- 74:

Full Full Full Full Full Full Full Full Full

Scale Scale Scale Scale Scale Scale Scale Scale Scale

=

IQ 3 8 . 7060 IQ 3 7 . 2474 IQ 36.0737 IQ = 36. 1 3 22 IQ 39.679 1 IQ 40. 6 1 29 IQ 4 1 .9824 IQ 42. 7008 IQ 4 3 . 4488 =

=

=

=

=

=

=

+

.6776 .6742 + .6500 + . 6460 + . 6366 + . 6424 + . 6536 + . 67 8 1 + . 7000 +

(scaled (scaled (scaled (scaled (scaled (scaled (scaled (scaled (scaled

score) score) score) score) score) score) score) score) score)

APPENDIX C

848

Table C-3 1 Estimates of the Differences Obtained by Various Percentages of the WAIS-R Standardization Sample When Each WAIS-R Age-Corrected Subtest Score Is Compared to the Mean Age-Corrected Scaled Score for Any I ndividual Examinee

Subtest

10%

5%

2%

1%

Information Digit Span Vocabulary Arithmetic Comprehension S imilarities Picture Completion Picture Arrangement Block Design Object Assembly Digit Symbol

2.6 3.6 2.2 3.0 2.7 2.8

3.1 4.3

3.7 5. 1

4.0 5.7 3.5 4.7 4.3 4.4

2.7 3.6 3.3 3.4

3.2 4.3 3.9 4.0

Overall average

Performance average

Verbal average

10%

3. 1 3.4 2.9 3.2 3.6

2%

5%

3.7 4.1 3.5 3.8 4.3

4.4 4.8 4. 1 4.5 5.1

1%

4.8 5.4 4.5 5 .0 5.6

10%

5%

2%

J%

3.0 3.8 2.7 3.2 3. 1 3.0 3.4

3.5 4.5

4.2 5.3 3.9 4.5 4.3 4.3 4.8 5. 1 4.7 5.4 5 .4

4.7 5.9 4.3 5.0 4.8 4.7 5.3 5.7 5.2 6.0 5.9

3.6 3.3 3.8 3.8

3.3 3.8 3.7 3.6 4. 1 4.3 3.9 4.5 4.5

Source: Reprinted with permission o f the publisher and author from A . B . Silverstein, "Pattern Analysis: The Question of Abnormality." Journal of Consuiling and Clinical Psychology, 1 984, 52, p . 938. Copyright 1 984 by the American Psychological Association.

Table C-32 Probability of Obtaining Designated D ifferences Between I ndividual WAIS-R Verbal and Perlormance IQs

Probability of obtaining given or greater discrepancy by chance . 50 .25 . 20 . 10 . 05 . 02 .01 . 00 1

Age level 1 6- 1 7

1 8- 1 9

20-24

25-34

35-44

45-54

55-64

65-69

70- 74

A v. a

4. 1 8 7 . 06 7 . 92 10. 1 3

3 . 80 6.43 7.21 9.23

3.12 5 . 28 5 . 92 7 . 57 9 . 00

3.16 5 . 35 6 . 00 7 . 68 9. 12 1 0 . 84 1 2 .00 1 5 . 35

3 . 14 5.31 5 . 96 7 . 62 9 . 05

3 . 22 5 .45 6. 1 2 7.83 9.30

3 . 38 5.71 6.40 8. 19

1 0 . 96 1 3 .03 1 4 . 43 1 8 . 46

3 . 20 5 . 42 6 .08 7 . 77 9.23 1 0 . 97 12. 15 1 5 .54

3 .06 5.18 5.81 7 . 43

1 2 . 04 14. 3 1 1 5 . 85 20.27

3 .49 5 . 89 6.61 8 .46 1 0 .05

1 0 .76 1 1 .9 1 1 5 . 24

1 1 . 05 1 2 . 24 1 5 . 65

1 1 . 94 1 3 .23 1 6 . 92

1 0.69 1 1 . 84 15.15

8 . 82 1 0.49 1 1 . 62 1 4 . 86

9 . 73 1 1 . 57 12.81 1 6. 3 8

Note. The table is entered i n the column appropriate t o the examinee's age. The discrepancy that i s just less than the discrepancy obtained by the examinee is located. The entry in the same row, first column, gives the probability of obtaining a given or greater discrepancy by chance. For example, the hypothesis that a 1 6-year-old examinee obtained a Verbal-Performance d iscrepancy of 1 6 by chance can be rejected at the .0 1 level of significance. This table is two-tailed . See Chapter 8, Exhibit 8- 1 for an explanation of the method used to arrive at magnitude of d i fferences . Av. Average of nine age groups. a

=

MISCELLANEOUS TABLES

849

Table C-33 Percentage of Cases in the WAIS-R Stand. ardization Sample with Differences Between Verbal (V) and Performance (P) IQs of Different Magnitudes 10 + Points

WAIS-R Full Scale IQ

22 + Points

N

V> P

P > V

70tal

V > P

P > V

Total

v>p

p> V

Total

V > P

p > V

Total

1 77 312 924 302 1 65

26.0 24.4 19.5

2 1 .5 20 . 2 2 1 .4

20.3

34 . 5 3 1 .4 25.4

9.6

16.9 6. 7

1.8

25.4 24. 7 19.2 9.6 2.4

0.7 0.0

3.4 2.6 3.7

7.6 2.4

15 8 14. 1 8.3 4.3 0. 6

6.2 4.2 3.6

1 6.2 7.3

14. 1 1 3 .5 14. 1 9.3 4.2

9.6 10.6 10.8

1 1 .6 8.5

47.5 44 . 6 40. 9 27.8 15.8

0.7 0.0

6.7 7.3 1 .3 0.0

1 880

1 8.7

19. 1

37.8

1 1 .9

12.3

24 . 3

8.7

9.0

1 7 .7

3.1

2.7

5.8

1 20 and above 1 1 0- 1 1 9 90- 1 09 80-89 79 and below

Total group

15 + Points

13 + Points

1 7 .9 1 1 .4

5.3

Source: Reprinted with permission o f the publisher and authors from J . D . Matarrazo and D . O . Herman. "Clinical Uses of the WAIS-R: Base Rates of Differences Between VIQ and PIQ i n the WA IS-R Standardization Sample," in Handbook of Intelligence: TJleories. Measurements, and Applications, edited by B. B. Wolman . New York: John Wiley, 1 98 5 , p. 9 2 3 .

Table C-34 Validity Coefficients of Proposed WAIS-R Short Forms

Short form V V I V C V S A V V

BD A BD PC BD S SO S OA C

Tetrad

Triad

Dyad

Short form

r . 90 .88 .88 .88 . 87

V V V I I V V I V V

. 87 . 87 . 87 . 86 . 86 =

S C A V S A PC C PA A

Short form

r BD BD BD BD BD PC BD BD BD S =

. 92 . 92 . 92 .92 . 92 . 92 . 92 .91 .91 .91

Pentad

V V V V V I 0 I V V =

A A A S C V V A A C

Short form

r

S PC C PC PC PC S S S BD

.94 .93 .93 .93 .93 93

BD BD BO BD BO BO BO

.93

BO

. 93

PC OS

.93 .93

=

=

V V D D

I

I

[ r [ [ V V

A A V V 0 V 0 0 A A

Note. Abbreviations: I Information; D Digit Span; V Vocabulary; A Arithmetic; C Compre hension ; S Completion; PA Picture Arrangement; BD Block Design; OA Object Assembly; DS Digit Symbol . The fol lowing formula was used to compute the part-whole correlations: =

=

r = c

=

+

=

j r= ���� = .In ==: + 2Er .Jk + 2Er h E rhh

S C S C V A S C S S =

r PC PC PC PC PC PC PC PC BD PA

BD BD · BD BD BD BD BD BD OS BD

Similarities; PC

=

. 95 .95 .95 . 95

. 94

.94 .94 .94 . 94 . 94 Picture

EE'h

ij , w here E'hh is the sum of the reliabilities of the k subtests, Erhj is the sum of the correlations between each of the k subtests and all the other subtests , n is the total number of the subtests, Erij is the sum of the correlations between each of the n subtests, k is the number of subtests in the short form, and Ergh is the sum of the intercorrelations of the k subtests. The term .In + 2Erij becomes a constant for all computations . The formula was applied to the average subtest intercorrelations for the nine age groups that composed the standardization sample (N 1880) to determine the correlations with the Full Scale of a l l possible short forms of two, three, four, and five subtests. The average subtest intercorrelations are based on the II standard subtests. =

850

APPENDIX C

Table C-35 Satz and Mogel'slYudin's Abbreviated Procedure for the WAIS·R

Subtest

Items used

Information Digit Span Vocabulary Arithmetic Comprehension Similarities

Every 3rd All items Every 3 rd Odd only Odd only Odd only

Picture Completion Picture Arrangement Block Design Object Assembly Digit Symbol

Every 3 rd Odd only Odd only Odd only All items

Multiply score by 3 I

3 2 2 2 3 2 2 2

Table C-36 Constants for Converting Wechsler Composite Scores into Deviation Quotients

'f, rjk

a

. 78 - . 9 2 . 66-. 7 7 . 54- . 65 .44- . 53 .35- .43 . 26- . 34 . 1 9- . 25

2.6 2.7 2.8 2.9 3.0 3. 1 3.2

4 Subtests

3 Subtests

2 Subtests

b

'f, rjk

48 46 44

2 . 1 6-2 .58 1 . 79-2 . 1 5 1 .48- 1 .78 1 . 2 1 - 1 .47 .97- 1 . 20 . 77- .96 . 59- . 76

42 40 38 36

a

1 .8 1 .9 2.0 2. 1 2.2 2.3 2.4

5 Subtests

b

'f, rjk

a

b

'f, rjk

46 43

3 . 95-4 . 85 3 . 2 1 -3 . 94 2 . 60-3 . 20 2 .09-2 . 59 1 . 66-2 .08 1 . 29- 1 . 65 . 98 - 1 . 2 8

1 .4 1 .5 1 .6 1 .7 1 .8 1 .9 2.0

44 40 36 32 28 24 20

6 . 96-8 . 83 5 . 50-6.95 4 . 36-5 .49 3 .45-4 .35 2 . 7 1 -3 .44 2 . 1 0- 2 . 70 1 . 59-2 . 09

40 37 34 31 28

a

1.1 1 .2 1 .3 1 .4 1 .5 1 .6 1 .7

b 45 40 35 30 25 20 15

Source: Reprinted b y permission o f the publisher and authors from A . Tellegen and P F Briggs, "Old Wine i n New Skins: Grouping Wechsler Subtests into New Scales," Journal o/ Consulting Psychology, 1 967 , 3 1 , p. 504. Copyright 1 967 by the American Psychological Association.

85 1

MISCELLANEOUS TABLES Table C-37 Estimated WISC-R. W PPSI. and WAIS-R Full Scale IQ Equivalents for Sum of Scaled Scores on Vocabulary and Block Design

ESlI:ma ted WAIS-R Full Scale IQ Age group Vocabulary plus Block Design scaled score"

ESfimared WISC- R Full Scale IQ 45

2 3 4 5 6 7 8 9 10 II 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

48 51 54 56 59 62 65

ESfim£J{ed WPPSI Full Scale IQ 43 46 49 52 55 58 61

1 6- / 7

25-44 65- 74 50 52 55 58 61 64

64

66

68 71 74 77

67

69 72 75 78

80 83 85 88 91 94

73 76 79 82 85 88 91 94

97 1 00

97 1 00

70

80 83 86 89 92 94 97 100

1 03

1 03

103

1 17 1 20 1 23 1 26 1 29 1 32 1 35 138

1 18 121 123

1 17 1 20 1 22

1 26 1 29 1 32 1 35

1 25

1 06 1 09 1 12 1 15

141 1 44 146 149 1 52 1 55 1 58

1 06 1 09 1 12 1 15

1 38 141 1 43 147 1 50 1 53 1 56 1 59

1 06 1 08 III 1 14

128 131 1 34

1 8-24

45-64

46 49 52

48 51 54 57

55 58

62

67 70 73 76

68 74 77 80 83 86 88

91 94

91 94 97 1 00

97 1 00

103

1 03 106 109 112

1 06 1 09 1 12 1 15

1 15 1 17 1 20 123 1 26 1 29 1 32 1 35 138

1 18 121 1 24 127 1 30

1 36 1 39 1 42 1 45 148

1 42 145 148 151

14) 144 1 46 149 1 52

1 54

" Use age-corrected scaled scores for the WAIS- R .

Source: The estimated WAIS-R

65 71

79 82 85 88

1 33 1 36 1 39

1 50

59

61 64

H . Brooker and 1 . 1 . Cyr, 983. Copyright 1 986 by the Clinical

Full Scale IQ equ i valents are reprinted with permission o f the publisher and authors from B .

"Tables for C l i nicians to Use to Convert WAI S - R Short Forms:' Psychology Publishing Company.

Journal a/ Clinical Psychology. 1 986. 42,

p.

APPENDIX C

852

Table C-38 Estimated WAIS-R Full Scale I Q Equivalents of Sum of Age-Corrected Scores on Vocabulary, B lock Design, and I nformation

Age group Sum of age-scaled scores

1 6- 1 7 25-54 65- 74

3 4 5

49

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

51 53 54 56 58 60 62

64 66 68 70 72 73 75 77

18-24 55-64 46 48 50 52 54 56 58 60 62

64

79

66 68 70 72 74 76 78

81 83

80 82

Age group

Age group Slim of age - scaled scores

/6- 1 7 25-54 65- 74

22 23 24 25 26 27

85 87 89 91 92 94 96 98 1 00 1 02 1 04 1 06 1 08 1 10 111 1 13 1 15 1 17

28 29 30 31 32 33 34 35 36 37 38 39

1 8-24 55-64 84 86 88 90 92 94 96 98 1 00 1 02 1 04 1 06 108 1 10 1 12 1 14 1 16 1 18

Sum of age- scaled scores

1 6- 1 7 25-54 65- 74

/ 8-24 55-64

40 41 42 43

1 19 121 1 23 1 25 127 1 29

1 20 1 22 1 24 1 26 128 1 30

1 30 1 32 1 34

1 32 1 34 1 36 138 1 40

44

45 46 47 48

49 50 51 52 53 54 55 56 57

1 36 138 1 40 142 1 44 1 46 148 1 49 151

142 1 44 1 46 148 1 50 152 1 54

Source: Reprinted with permission o f the publisher and authors from B . H . Brooker and J . J . Cyr. "Tables for Cli nicians t o Use t o Convert WA IS-R Short Forms , " Journal a/ Clinical Psychology, 1 986, 42. pp. 983-984. Copy right 1 986 by the Clinical Psychology Publishing Com pany.

853

MISCELLANEOUS TABLES Table C-39 Estimated WAIS-R F u l l Scale I Q Equivalents of Sum of Age-Corrected Scores o n Ifocabu lary, Block Design, Arithmetic, and Similarities Sum of age-scaled scores

Age group

------

16 -1 9 25- 74

4 5

46 48

6 7

49 51 52 54 55

8 9 10 II 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

57 58 60 61 63 64 66 67 69 70 72 73 75 76 78 79 81 82

Sum of age-scaled scores

20-24 42 44 46 47 49

I

29 30 31 32 33 34 35

SO 52 54 55 57 58 59

36 37

81

16-19 25- 74 84 85 87 88 90 91 93 94 96

38 39

97 99

40

1 00 1 02

41 42 43 44 45

62 63 65 66 68 70 71 73 74 76 78 79

Age g roup -

-

46 47 48 49 50 51 52

1 03 1 05 1 06 1 08 i 09 III i 12 1 14 1 15 1 17 118

----

Slim of age-scaled

20-24

scores

82 84

53 54 55

86 87 89 90 92 94 95 97 98 1 00 102 103 105 1 06 1 09 1 10 III 1 13 1 14 1 16 1 18 ! 19

56 57 58 59 60 61 62 63 64

65

66 67 68 69 70 71 72 73

74 75 76

Age group

------

1 6- 1 9 25- 74 1 20 121 1 23 1 24 1 26 1 27 1 29 1 30 1 32 1 33 1 35 1 36 1 38 1 39 141 1 42 1 44 145 1 47 1 48 1 50 151 1 53 1 54

20-24 121 1 22 1 24 1 26 1 27 1 29 i 30 1 32 1 34 1 35 1 37 1 38 1 40 1 42 1 43 145 i 46 1 48 I SO 151 1 53 1 54 1 56 1 58

-

H . Brooker a n d J . J . Cy r, "Tabl(;s for C l i n icians t o Use t o Convert WA I S - R Short Forms." JOllrllal o/ Clillical Psychology. 1 986. 42 . pp. 984-98 5 . Copyright 1 986 by the C l i nical Psychology Publish i ng Company.

Source: Reprinted with permission o f the publisher a n d authors from B .

APPENDIX C

854

Table C-40 Estimated WAIS-R Full Scale IQ Equivalents of Sum of Age-Corrected Scores on I nformation, Arithmetic, Picture Completion, and Block Design

Source:

Sum af four scaled scores

Estimated Ful! Scale IQ

Sum af fOllr scaled scores

Estimated Fill! Scale IQ

Sum af four scaled scares

Estimated Full Scale IQ

4 5 6 7 8 9 10

51 52 54

29 30 31

85 86 88

54 55

55 56 58 59

32 33

1 18 1 19 121

34 35

89 90 92

II 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

61 62 63 65 66 67

36 37 38 39 40 41

69 70 71 73 74 76 77 78 80 81 82 84

42 43 44 45 46 47 48 49 50 51 52 53

Clinical Neuropsychology, 1 98 3 , 5 .

58

93

59 60

95 96 97

61 62 63

1 22 123 1 25 1 26

1 02 1 03 1 04 1 06 1 07

66

1 28 1 29 1 30 1 32 1 33 1 34

67 68 69 70 71

108 1 10 III 1 13 1 14 1 15

72 73 74 75 76

1 36 137 1 38 1 40 141

98 99 1 00

64 65

1 42 1 44 1 45 1 47 1 48

117

Reprinted w i th permiss ion of the publisher and authors from C. R. Reynolds.

WA I S - R for C l i n ical Scree n i n g , "

56 57

V. L. W i l l son , and P L. Clark, " A Four-Test Short Form of the

p . 1 1 4 . Copyright 1 983 by Meln ic Press.

8SS

MISCELLANEOUS TABLES Table C-4 1 Percentile Ranks and Suggested Qualitative Descriptions for Scaled Scores on the Wechsler Scales

Scaled score

Percellfile rank

19

99

18

99

17

99

16

98

15

95

14

91

Qualitaril'e descriptions Exceptional mength or Very well developed or

Strength or

A bove

average

Superior or Excellent Strength or Wel l developed or

average or

Above

13

84

10

I:l

9

37

8

25

12 I I

j,l

63

50

9

5

5

3 2

Average

2

Average

Weak ne:;,

16

6

4

Good

or Poorly develop e d or

Below aver:lge 0 1'

Weakness or Below average

Poor Except ional weakness or Very poorly developed or Far below average or Very poor

856

APPENDIX C

Table C-42 Interpretive Rationales, I mplications of High and Low Scores, and I nstructional Implications for Wechsler Scales and Factor Scores

Ability

Background factors

Possible implications of high-scores

Possible implications of low scores

Instructional implications

Full Scale

General intelligence Scholastic aptitude Academic aptitude Readiness to master a school curriculum

Natural endowment Richness of early environment Extent of schooling Cultural opportunities Interests Rate of motor activity Persistence Visual-motor organization Alertness

Verbal comprehension Application of verbal skills and information to the solution of new problems Verbal ability Ability to process verbal information Abil ity to think with words

Natural endowment Richness of early environment Extent of schooling Cultural opportunities Interests

Good general intell igence Good scholastic aptitude Readiness to master a school curriculum

Poor general intelligence Poor scholastic aptitude Not ready to master school curriculum

Focus on language development activities Focus on visual learning activities Develop concept formation skills Reinforce persistence

Verbal Scale or Verbal Comprehension Factor

Good verbal comprehension Good scholastic aptitude Possession of knowledge of the cultural milieu Good concept formation Readiness to master school curriculum Achievement orientation

Poor verbal comprehension Poor scholastic aptitude Inadequate understanding of the cultural milieu Poor concept formation Bilingual background Foreign background Not ready to master school curriculum Poor achievement orientation

Stress language development activities Use verbal enrichment activities Focus on current events Use exercises involving concept formation

(Table continues next page)

857

MISCELLANEOUS TABLES Table C-42 (cont.)

Ability

Background factors

Possible implications of high-scores

Possible implications of low scores

Instructional implications

Performance Scale or Perceptual Organization Factor

Perceptual organization Abil ity to think in terms of visual images and manipu late them with fluency. flexibil ity, and relative speed Abil ity to interpret or organize visua l l y perceived material against a time l i mit Nonverbal abil ity Abil ity to form relatively abstract concepts and relationships without the use of words

Natural endowment Rate of motor activity Persistence V isual-motor organization Alertness Cultural opportunities Interests

Abil ity to sustain attention Short-term memory Numerical abil ity Encoding abil ity Abil ity to use rehearsal strategies Abil ity to sh ift mental operations on symbolic material Abi lity to se l f­ monitor

Natural endowment Abil ity to passively receive st imuli

Good perceptual organization Good alertness to detail Good nonverbal reason i ng abil ity Good persistence Good abil ity to work quickly and efficiently Good spatial abil ity

Poor perceptual organ izat ion Poor alertness to detail Poor nonverbal reasoning abil ity Lim ited persistence Poor abil ity to work quickly and efficiently Poor spatial abil ity

Focus on visual learning activities Focus on part ·whole relationships Use spatial-vi sual tasks Encou rage trial -and­ error activities Reinforce persistence Focus on visual planning activities I mprove scanning techniques

Di fficulty in sustaining attention Distractibil it)' Anx iety Short-term retention deficits Encoding di fficu lties Poor rehearsal strategies Di fficulty in rapidly shifting mental operations on symbol ic material Inadequate self­ monitoring skills

Develop attention skills Develop concen­ t ration skills Focus on smal l . meaningful units of i nstruct ion

Freedom from Distractibility

Good ability to sustain attention Good short-term memory Good numerical abil ity Good encoding abi l ity Good use of rehearsal strategies Good abil ity to shift mental operations on symbolic material Good abil ity to sel f­ monitor

Source. The fi rst column adapted . i n part. from Cohen ( 1 959) and Wal l brown ( 1 979 ) .

APPENDIX C

858 Table C-4l Suggested Remediation Activities for Combinations of Wechsler Subtests

Subtests

Ability

Information, Vocabulary, and Comprehension

General knowledge and verbal fluency

Similarities and Vocabulary

Verbal conceptual

Digit Span, Arithmeti c , Picture Completion, and Picture Arrangement

Attention and concentration

Block Design and Object Assembly

Spatial-visual

Coding, Digit Symbol , Block Design, Object Assembly, Animal House, and Mazes

Visual-motor

Activities ( I ) Review basic concepts, such as days of the week , months, time,

distances, and directions; (2) have children report major cu rrent events by referring to pictures and articles from magazines and newspapers ; ( 3 ) teach similarities and differences of designs, topography, transportation, etc . ; (4) have children make a scrapbook of pictures of animal s , buildings. etc . ; (5) introduce words, dictionary work, abstract words; (6) have children repeat simple stories; (7) have children explain how story characters are feeling and thinking. ( I ) Use show-and-tell games; (2) have children make a scrapbook of classifications, such as of animals , vehicles, and utensils; ( 3 ) have children match abstract concepts; (4) have children find commonality in dissimilar objects; (5) review basic concepts such as days of the week, months, time, directions, and distances. ( I ) Have children arrange cards in a meaningful sequence; (2) have children learn telephone number, address , etc . ; ( 3 ) use spel l i ng word games; (4) use memory games; (5) have children learn days of week, months of year; (6) use mathematical word problems; (7) use dot-to-dot exercises; (8) have children describe details in pictures; (9) use tracing activities; ( 1 0) use Tinker Toys . ( I ) Have children identify common objects and discuss details; (2) use guessing games involving description of a person , place, or thing; (3) have children match letters, shapes, numbers, etc . ; (4) use j igsaw puzzles; ( 5 ) use block-building activities. ( I ) Use paper-folding activities; (2) use finger-painting activities; (3) use dot-to-dot exercises; (4) use scissor-cutting exercises; (5) use sky-writing exercises; (6) have children string beads in patterns; (7) use pegboard designs; (8) use puzzles (large j igsaw pieces); (9) have children solve a maze; ( 1 0) have children follow a moving object with coordinated eye movements; ( I I ) use tracing exercises (e . g . , trace hand, geometric forms , and letters) ; ( 1 2 ) have children make large circles and lines on chalkboard ; ( 1 3) have children copy from patterns; ( 1 4) have children draw from memory.

Table C-44 Confidence I ntervals for the Stanford-Binet Intelligence Scale: Fourth Ed ition

Confidence level Age level 2 (2-0-0

through 2-1 1 - 1 5 ) 3

(2- 1 1 - 1 6

through 3- 1 1 - 1 5 )

4 (3- 1 1 - 1 6

through 4- 1 1 - 1 5 )

5 (4- 1 1 - 1 6

through 5 - 1 1 - 1 5)

6 (5- 1 1 - 1 6

through 6- 1 1 - 1 5 )

7 (6- 1 1 - 1 6

through 7- 1 1 - 1 5) 8 ( 7- I I - i 6

through 8- 1 1 - 1 5)

9 (8- 1 1 - 1 6

through 9- 1 1 - 1 5)

10

(9- 1 1 - 1 6

through 1 0- 1 1 - 1 5)

11

( 1 0- 1 1 - 1 6

through I I- I I - IS)

12

(1 1-1 1-16

through 1 2- 1 1 - 1 5 )

Factor and Composite Score

68%

Verbal Comprehension FaclOr Score Nonverbal ReasoninglVisualization Factor Score Memory Factor Score Composite Score

±4

85 %

90 % ±

95 %

99 %

±

6 9

± 8 ± II

± 10 ± 14

±

6

±

7

±

I

6 6

± ±

7 8

± 9 ± 10

±5

-

5

±

6

±

8

± 4 ±4

±5 ±5

-

6 6

± ±

7 7

± ±

9

± 3

±4

5

±

5

±

7

Verbal Comprehension Factor Score Nonverbal Reasoning/Visualization Factor Score Memory Factor Score Composite Score

± 4 ± 3

±5 ± 5

I

6 5

± ±

7 6

± ±

9

-"-

± 3

±4

±

5

±

5

±

7

Verbal Comprehension Factor Score Nonverbal Reasoning/Visualization Factor Score Memory Factor Score Composite Score

± 5 ± 4

±7 ± 6

±

7 6

± ±

9 8

± 12 ± 10

± 3

± 5

± 5

±

6

±

Verbal Comprehension Factor Score Nonverbal ReasoninglV isualization Factor Score Memory Factor Score Composite Score

± 5 ± 4

± 7

± 7

± 9 ± 8 ± 12

Verbal Comprehension Factor Score Nonverbal Reasoning/Visualization Factor Score Memory Factor Score Composite Score

± ± ± ±

Verbal Comprehension Factor Score Nonverbal Reasoning/Visualization Factor Score Memory Factor Score Composite Score

± 4 ± 4 ± 5

Verbal Comprehension Factor Score Nonverbal ReasoninglVisualization Factor Score Memory Factor Score Composite Score

± 4 ± 3 ± 5

Verbal Comprehension Factor Score Nonverbal ReasoninglVisualization FaclOr Score Memory Factor Score Composite Score Verbal Comprehension Factor Score Nonverbal Reasoning/Visualization Factor Score Memory Factor Score Composite Score

±6

±6 ±8

±4

±5

Verbal Comprehension Factor Score Nonverbal Reasoning/Visualization FaclOr Score Memory Factor Score Composite Score

± 4 :1: 4

±6 ±6

± 3

Verbal Comprehension FaclOr Score Nonverbal Reasoning/Visualization Factor Score Memory Factor Score Composite Score

859

± 6 ± 3 5 4 5 3

± 6 ± 9 i 4

± ± ± ±

7 6 7 4

I

7 ± 10 ± 5 i

± 7

±

± ±

6 8 5

8

± 12 ± II ± IS

±

5

±

± ± ±

9 8 9

±

5

± 12 ± 1O ± 12 ± 7

±

7

± 9 ± 10 ± 14

± 7 ± 6 ± 10 ± 5

± 9 ± 8 ± 13 ± 6

± 6 ± 5 ± 6 ± 4

± ±

7 6

± ±

7 5

± ± ± ±

± 5

±

± 7 ± 3

±

± 6 ± 6 ± 10 ± 5

±5 ±5 ± 7 ± 3

±

6

±

4

± 4 ± 3 ± 4 ± 2

i s

±5 ±5 ± 3

± 3 ± 3 ± 5 ± 2

8

±

± 6

± 2

9

7 ± 8 ± 10 ± 5

±5 ±6 ±8 ±4

± 3

9

i s

± ±

6

9

± 5 ± 5 ± 8

±

5 5 8

± 4

7

9 8 9 6

± 8 ± 8 ± 13 ± 6

(Table continues next page)

Table C-44 (cont.)

Confidence level Factor and Composite Score

68 %

85 %

Verbal Comprehension Factor Score Nonverbal Reasoning/Visualization Factor Score Memory Factor Score Composite Score

± 3 ± 3 ± 5

± 5 ± 4 ± 7

Verbal Comprehension Factor Score Nonverbal Reasoning/Visualization Factor Score Memory Factor Score Composite Score

± ± ± ±

Verbal Comprehension Factor Score Nonverbal Reasoning/visualization Factor Score Memory Factor Score Composite Score

± 3 ± 3 ± 5

± 2

± ± ± ±

Verbal Comprehension Factor Score Nonverbal Reasoning/visualization Factor Score Memory Factor Score Composite Score

± ± ± ±

3 3 5

± ± ± ±

Verbal Comprehension Factor Score Nonverbal Reasoning/Visualization Factor Score Memory Factor Score Composite Score

± ± ± ±

3 3 4

Verbal Comprehension Factor Score Nonverbal Reasoning/visualization Factor Score Merriory Factor Score Composite Score

± ± ± ±

3 2 4

Age level

13 ( 1 2- 1 1 - 1 6

through 1 3- 1 1 - 1 5 )

14 ( 1 3-1 1 - 16

through 1 4- 1 1 - 1 5 ) IS

( 1 4- 1 1 - 1 6

through IS-I I- IS) 16 ( 1 5- 1 1 - 1 6

through 1 6- 1 1 - 1 5)

17 ( 1 6- 1 1 - 1 6

through 1 7- 1 1 - 1 5)

18-23 ( 1 7- 1 1 - 1 6

through 2 3- 1 1 - 1 5)

± 2 3 3 5

2

2

± ±

3

5 5

± ±

5 5

± ± ± ±

7 2

± ±

8 4

5 4

± 5 ± 5 ± 8 ± 4

± ± ± ±

± 5 ± 5 ± 8 ± 4

± ± ±

7 2 5 4 7 2

±

4 3 6 2

± ± ± ±

99 %

95 %

± ± ± ±

± 4 ± 4 ± 6 ± 2 ± ± ± ±

2

5 5 7

± ±

± 2 ± ± ± ±

2

90%

6 5 9 3

± 8 ± 7 ± 12 ± 4

6 6 9

± 8 ± 8 ± 12 ± 6

5

± 8 ± 7 ± 12 ± 6

6 5 9 5

±

9 5

± 8 ± 7 ± 12 ± 6

± ± ± ±

5 5 8 3

± 7 ± 7 ± II ± 4

6 3

± 5 ± 4 ± 8 ± .3

± 7 ± 6 ± 1O ± 4

EB

BM

MS

MD

8/ 1 0 9/1 1 1 0/ 1 3

8/ 1 1 1 0/ 1 3

1 0/ 1 4

5 5

± ± ±

7 3 5 4

6 5

Table C-4S Significant Differences Between Stanford-Binet I ntelligence Scale: Fourth Edition Subtest Scaled Scores (.05/.0 1 significance levels)

C A VR PA CP M PF Q NS EB BM MS MD MO

V

C

A

81 1 0 81 1 1 71 1 0

81 1 0 7/9

6/8

71 1 0 8/ 1 1 8/10 7/9 8/ 1 0 8/10 7/8 8/ 1 1 8/ i o 9/ 1 1 1 0/ 1 3

7/9

VR

PA

CP

M

8/10 8/ 1 0 7/9 8/ 1 0 8/ 1 0 7/ 1 0 8/ 1 1

7/ 1 0 8/ 1 1 8/ 1 0 7/9 8/ 1 0 8/ 1 0 7/ 1 0 8/ 1 1

7/9 6/8 7/9 6/8 7/ 1 0 7/9 7/9 7/ 1 0

7/ 1 0 7/9 6/8 7/9 7/9 7/9 7/ 1 0

8/10 7/9 8/10 8/ 1 0 6/8 8/ 1 1

6/8 71 1 0 7/9 7/9 81 1 0

7/9 6/8 6/8 7/9

7/9 8/ 1 1 1 0/ 1 3

8/ 1 0 8/ 1 1 1 01 1 3

7/9 8/ 1 1 9/ 1 2

7/9 8/ 1 0 9/ 1 2

8/ 1 0 9/ 1 1 1 0/ 1 3

7/9 8/ 1 1 1 0/ 1 3

6/8 8/10 9/ 1 2

=

=

=

Q

PF

=

NS

7/ 1 0 7/ 1 0 8/ 1 0

7/9 8/ 1 0

7/ 1 0 8/ 1 1 1 01 1 3

7/9 8/ 1 1 1 0/ 1 3

71 1 0 7/9 81 1 1 91 1 2 =

=

NOle. Abbreviations: V Vocabulary; C Comprehension; A Absurdities; V R Verbal Relations; PA Pattern Analysis; C P Copying; M Matrices; PF Paper Folding and CUlling; Q Quantitative; NS Number Series; EB Equation Building; BM Bead Memory; MS Memory for Sentences; M D Memory for Digits; MO Memory for Objects . =

=

=

=

=

=

=

=

=

APPENDIX C

862 Table C-47 Differences Requ ired for Significance When Each Stanford-Binet I n telligence Scale: Fourt h Edition Subtest Scaled Score Is Compared t o the Respective Mean Factor Scaled Score for Any I ndividual Child

Verbal Comprehension 1 5 through 1 8-23

2 through 7

8 through 1 4

Subtest

. 05

. 01

. 05

. 01

. 05

. 01

Vocabulary Comprehension Absurdities Memory for Sentences Verbal Relations

6. 1 8 5 . 74 6. 1 8 5 . 74

7.5 I 6.99 7.5 I 6. 99

5 .57 5.18 5 . 57

6 . 84 6.37 6 . 84

5 . 57 5. 1 8

6 . 84 6.37

-

-

-

-

-

5 . 26

6 . 46

-

-

-

Nonverbal Reasoning/Visualization 2 through 1 1

Subtest Pattern Analysis Copying Quantitative Bead Memory Matrices

1 2 through 1 8-23

. 05

. 01

. 05

. 01

5 . 30 6. 1 5 6.02 6. 1 5

6.44 7.4 1 7 . 33 7.41

5 . 22

6.35

5 . 95

7 . 24

6. 1 0 5 . 50

7 . 42 6 . 69

-

-

-

-

Memory 7

Subtest Memory for Sentences Memory for Digits Memory for Objects

8 through 1 8-23

. 05

. 01

-

5 .98 5 . 98

. 05

. 01

-

5 . 98

7 . 50 7 . 50

6 . 60 7.51

7 . 34 8. 1 1 9.73

Note. See the Note in Table C-3 for a n explanation o f how d ifferences were obtained.

-

-

MISCELLANEOUS TABLES

863

Table C-48 Estimates of the D ifferences Obtained by Various Percentages of the Stanford-·Binet I ntelligence Scale: Fourth Edition Standardization Sample Between Each Subtest Score and an Average S ubtest Score

Verbal Comprehension average 15 through 1 8-23

8 through 14

2 th rough 7

SlIbtest

10%

5%

2%

1%

10%

5%

2%

1%

10%

5%

2%

1%

Vocabulary Comprehension Absurdities Verbal Relations Memory for Sentences

5 . 90 6.47 7 . 92

7 .02 7.68 9.4 1

8 . 34 9. 1 3 1 1.18

9 . 24 1 0. 1 1 1 2.38

6 . 24 6. 1 9 6.98

7.41 7 . 35 8.29

8.81 8 . 74 9 . 86

9.75 9.68 10.9 1

4.75 5.12

5 . 64 6.08

6.71 7 .22

7 .43 8 . 00

4 . 36 -

5 . 17

6. 1 5

6.81

-

7 . 92

-

9.41

-

1 1 . 18

-

-

-

-

1 2 .38

-

-

-

--

-

-

-

-

-

--

-

--

Nonverbal Reasoning/Visualization average 1 2 through 1 8-23

2 through 1 1

Subtest Pattern Analysis Copying Quantitative Bead Memory M atrices

10%

5%

2%

1%

10%

5%

2%

1%

7 . 92 8.55 8. 13 8.13

9.4 1 10. 15 9 . 66 9 . 66

1 1 . 18 1 2 .07 1 1 .49 1 1 .49

1 2 .38 1 3 . 36 1 2.72 1 2 .72

7 . 80

9 . 27

1 1 .02

1 2 . 20

7 . 59 7 . 80 7 . 80

9.02 9 . 27 9 . 27

1 0 . 72 1 1 .02 1 1 .02

1 1 . 87 1 2 . 20 1 2 . 20

-

-

-

-

-

-

-

-

Memory average 7

2 through 6

Subtest Memory for Digits Memory for Obj ects M emory for Sentences

2%

J%

-

-

-

-

-

-

--

-

10%

5%

-

-

8 through 18-23

10%

5%

2%

1%

10%

5%

2%

1%

6.98 6 . 98 -

8 . 29 8 . 29

9 . R6 9 . 86

1 0. 9 1 10.91

7 . 34 8.45 7.52

8 . 72 1 0.04 8 . 94

10.37 1 1 .93 10.62

1 1 .48 13.21 1 1 . 76

-

-

-

-

Nore. The following formula was used to compute the estimated differences (Silverstein, 1984) : SDOa

=

) 8, 1

+

G

-

2 T.

where SD Oa i s the standard deviation of the difference for subtest a; 8 is the standard deviation of the scaled scores on each of the subtests; G is the mean of all the elements in the matrix ( i ncluding the diagonal); and Ta is the mean of the elements in row or column a of the matrix (again including the diagonal). This formula is applied to the matrix of subtest intercorrelations, with Is in the main diagonal. SDoa is multiplied by the respective z value to estimate how large a difference was obtained by 10, 5 , 2, or I percent of the standardization sample.

Table C-49 Probability of Obtaining Designated Differences Between Stanford-Binet Intelligence Scale: Fourth Edition Factor Scores

Probabiliry of obtaining given or greater discrepancy by chance

2

3

. 50

4 . 62

3.61

3 . 44

3 . 26

4 . 07

.25 . 20

7.81

6. 1 1

8 . 76

6 . 85

5 . 82 6.53

5 . 52 6. 1 9

6 . 89 7 . 73

. 10

1 1 . 20 1 3. 3 1

8 . 76 1 0. 4 1

8 . 35 9 . 92

7 . 92

9 . 88

9.41

1 5 . 82 1 7 . 52

1 2.37 1 3 . 70

1 1 . 79 1 3 .05

22 . 4 1

1 7 .52

1 6 . 70

Ages

4

5

7

6

8

9

10

1/

12

13

14

15

16

17

18-23

3 . 26

Verbal Comprehension vs. Nonverbal Reasoning/Visualization

. 05 . 02 .01 . 00 I CD '" �

4 . 07

3.61

3 . 26

3 . 26

3 . 08

2.88

3 . 08

2 . 88

2 . 88

2 . 67

6. 1 1 6 . 85

5 . 52

5 . 52

5.21

4 . 86

5.21

4 . 86

5 . 52

6. 1 9

6. 1 9

5 . 84

5 . 46

5 . 84

5 . 46

4 . 86 5 . 46

4.5 1

8 . 00

6 . 89 7 .73

5 . 06

6. 19

1 0 . 23 1 2. 1 5

9.88 1 1 . 74

8 .76

7 . 92 9.41

7 .47 8 . 88

7 . 47

6 . 98

6 . 98

6.47

10.41

7 . 92 9.41

6 . 98

1 1 . 74

8 . 29

8 . 88

8 . 29

I U8 1 2 . 38

1 3 .96 1 5 . 45

1 4 .45 1 6 .00

1 3 .96

I I . 18

1 0.55 1 1 . 69

1 0.55 1 1 . 69

9 . 86

1 2.38

I I . 18 1 2 . 38

9 . 86

1 5 .45

1 2 . 37 1 3 . 70

8.29 9 . 86 10.91

10.91

7 . 92 9.41 1 1.18

1 5 . 84

19 .77

20 .t46

1 9 . 77

1 7 . 52

1 5 . 84

1 5 . 84

1 4 . 95

1 2 .94

1 5 . 84

4 . 07

3 . 77

6 . 89 7 . 73

6 . 38 7. 1 6

3 . 26

4 . 22 7. 1 3

10.91

1 3 . 96

1 4 .95

1 3 . 96

3 . 92 6 . 64

3 . 92 6 . 64

7 .44

7 . 44

Verbal Comprehension vs. Memory 5. 1 1

4 .49

3 . 92

3.61 6. 1 1 6 . 85

9 . 52

8 . 76

7 . 92

1 l .3 1

1 l.31

1 0 .4 1

9.4 1

1 3 .44

1 3 .44

1 3 .44

1 2 . 37

1 1 . 18

1 4 . 89

1 4 . 89

1 4 . 89

1 3 . 70

1 2 . 38

1 8 . 32

1 9 .04

1 9 .04

1 9 .04

1 7 .52

1 5 . 84

7 . 59

4 . 35 7 . 36

4 . 22 7. 1 3

3 . 44 5 . 82

8.51

8 . 26

1 0 . 89

1 0. 56

8 . 00 1 0.23

6.53

. 10

9 . 69 1 2 . 39

8 . 35

9 . 88

9. 1 6

9 . 52

.05

1 4 . 72

1 2 . 94

1 2 . 54

12. 15

9 . 92

1 1 . 74

1 0.88

1 l .3 1

. 02

1 7 . 50

1 5 .38

1 4 .9 1

1 4 . 45

1 1 . 79

1 3 . 96

1 2 .93

.01

1 9. 38

1 7 .03

1 6 . 50

1 6 . 00

1 3 . 05

1 5 .45

1 4 . 32

.00 1

24.78

2 1 .78

21.12

20.46

1 6 . 70

1 9 .77

. 25 .20

1 2.38

6 . 64 7 . 44 9.52

8 . 64

. 50

1 3 . 96

7 . 68 9. 1 3 10. 1 1

5 . 52 6. 1 9

Nonverbal Reasoning/Visualization vs. Memory . 50

4.98

4.22

4 .49

4 . 07

3 . 26

4.01

3.61

3 . 92

3 . 73

3 . 77

3 .44

3 . 77

.25 . 20

8 .43 9 . 46

7. 1 3 8 . 00

7 . 59 8.51

6 . 89 7 . 73

5 . 52 6. 19

6 . 79 7.61

6. 1 1 6 . 85

6 . 64 7 .44

6.3 1 7 . 08

6.37 7. 1 5

5 . 82 6 . 53

6.37 7. 1 5

. 10

1 2 .09 1 4 . 37

1 0.23

1 0 . 89

9 . 88

7 . 92

9 . 74

8 . 76

9. 1 4

8 . 35

9. 1 4

12. 15

1 2 . 94

1 1 . 74

9.4 1

1 1 . 56

10.41

9.52 1 l.31

9 .06

.05

1 0 . 76

1 0.86

9 . 92

1 0 . 86 12.91

.02

1 7 . 08

1 4 .45

1 5 .38

1 3 . 96

1 1 . 18

1 3 .75

1 2 . 37

1 3 .44

1 2 .79

1 2.9 1

1 1 . 79

.01

1 8.91

1 6 . 00

1 7 . 03

1 5 .45

1 2.38

1 5 .22

1 3 . 70

1 4 . 89

14. 1 6

1 4 . 29

1 3 . 05

1 4 . 29

.00 I

24 . 1 9

20.46

2 1 .78

1 9 . 77

1 5 . 84

1 9 .47

1 7.52

1 9 .04

18.12

1 8.28

1 6 . 70

1 8.28

Table CoSO Percentage of Population Obtaining Discrepancies Between Stanford-Binet Intelligence Scale: Fourth Edition Factor Scores

Percen/age obtaining given or greater discrepancy in either direction

Age level 2

3

4

5

6

1 0 . 66 1 8 .03

9.73 1 6.46

8 .29 1 4 . 02

8. 1 4 1 3 . 77

9.73 1 6. 46

7

8

9

10

11

13

12

14

15

16

I7

18-23

Mdn

8 . 70 1 4 . 72

7 . 54 1 2 . 75

7 . 85 1 3 . 27

7 . 38 1 2 .48

8. 14 1 3 . 77

Percen/age obtaining given or greater discrepancy ill a specific directioll

Verbal Comprehension vs. Nonverbal Reasoning/Visualization 50 25 20 10 5 2 .1 00 '" V1

9. 1 1 1 5 . 40

8 . 70 1 4 . 72

9.23 1 5 . 62

7 . 54 1 2 . 75

7 . 69 1 3 .0 1

7 . 85 1 3 .27

7 .54 1 2 .75

7 . 38 1 2 .48

25 12.5

20.23

1 8.46

1 5 .73

1 5 .44

1 8 . 46

1 7 . 27

16. 5 1

1 7 . 52

1 4. 3 1

1 4 . 59

1 4 . 89

14.31

1 4. 00

16.5 1

14. 3 1

1 4 . 89

1 4 . 00

1 5 . 44

25 . 87 30.73 36.53 40.45

23 . 6 1 2 8 . 05 3 3 . 34 36.92

20. 1 1 23 . 89 2 8 .40 3 1 .45

1 9.75 23 .46 2 7 . 89 30.88

23.61 2 8 . 05 3 3 . 34 3 6 . 92

2 2 . 09 2 6 . 24 3 1 . 20 34 . 55

2 1 . 12 25 . 09 29.82 3 3 . 02

22.41 2 6 . 62 3 1 . 64 3 5 . 04

1 8 . 30 2 1 . 74 25 . 84 28.61

1 8 .66 22 . 1 7 26.35 29. 1 8

1 9. 04 2 2 . 62 26.89 29.77

1 8 . 30 2 1 . 74 25 . 84 28.61

1 7 . 90 2 1 .27 25.28 2 7 . 99

2 1 . 12 2 5 . 09 2 9 . 82 3 3 . 02

1 8 . 30 2 1 . 74 25 . 84 28.61

1 9 .04 2 2 . 62 2 6 . 89 29.77

1 7 .90 2 1 .27 25 . 28 2 7 . 99

1 9 . 75 2 3 . 46 2 7 . 89 30. 88

5 1 . 74

47.22

40. 23

3 9 . 50

4 7 . 22

44 . 1 9

4 2 . 24

44 . 8 1

3 6 . 60

3 7 . 32

3 8 . 08

3 6 . 60

35 . 8 1

4 2 . 24

36.60

3 8 . 08

35 . 8 1

3 9 . 50

8 . 70 1 4 . 72 1 6. 5 1 2 1 . 12 25 . 09 29 . 82 3 3 .02

1 0. 09 1 7 . 07 19. 14 24 .49 2 9 . 09 34.58 3 8 . 29 48.97

8 . 00 1 3 . 52 15. 17 1 9 . 40 2 3 . 05 2 7 . 40 3 0 . 34 38 . 8 1

8 . 84 1 4 . 95 1 6 . 77 2 1 .45 25 . 48 30.29 3 3 . 54 4 2 . 90

9 . 42 1 5 . 93 1 7 . 87 2 2 . 85 27. 1 5 3 2 . 27 3 5 . 74 45 . 7 1

25 12.5 10 5 2.5

42 . 24

9.73 1 6.46 1 8 .46 23.61 2 8 . 05 3 3 . 34 3 6 . 92 4 7 . 22

8 . 70 1 4 . 72 1 6. 5 1 2 1 . 12 25 09 29 . 82 3 3 . 02 42 . 24

9.23 1 5 . 62 1 7 .52 22.4 1 26.62 3 1 .64 3 5 . 04 44 . 8 1

8 . 97 15. 17 1 7 . 02 2 1 . 76 25 . 85 30. 73 34.03 43.53

8 . 84 1 4 . 95 1 6. 77 2 1 . 45 25 .48 30.29 3 3 . 54 42 . 90

8. 14 1 3 . 77 1 5 .44 1 9 . 75 2 3 . 46 27 . 89 30 . 88 3 9 . 50

8.91 1 5 . 06 1 6 . 90 2 1 .61 2 5 . 67 30 . 5 1 3 3 . 79 43 . 22

25 1 2.5 10 5 2.5 1 .5 .05

10 5 2.5 I

.5 . 05

Verbal Comprehension vs. Memory 50 25 20 10 5 2

.1

1 0 . 88 1 8 .40 2 0 . 64 26.40 3 1 . 36 3 7 . 28 4 1 .28 5 2 . 80

9 . 73 1 6 . 46 1 8 .46 23.61 2 8 . 05 3 3 . 34 36. 92

47 . 2 2

1 0 . 44 1 7 . 65 1 9 . 80 25. 33 30 _09 35 7 7 39 _ 60

50 _ 66

9 . 85 1 6. 66 1 8 . 69 23 .9 1 28 .40 3 3 . 76 37 . 38

- H . 82

8 . 70 1 4 . 72 1 6. 5 1 2 1 . 12 25 09 2 9 . 82 3 3 .02

42 . 24

9. 1 1 1 5 .40 1 7 .27 2 2 . 09 2 6 . 24 3 l . 20 34 . 55

44 . 1 9

8.43 1 4 . 25 1 5 . 98 2 0 . 44 24 . 2 8 2 8 . 87 3 1 . 97

40 . 89

I

.5

.05

Nonverbal Reasoning vs. Memory 50 25 20 10 5 2 .1

1 1 .4 1 1 9 . 30 2 1 . 65 2 7 . 69 3 2 . 89 39 . 1 0 4 3 . 29 55.37

9.73 1 6 . 46 1 8 .46 23.61 2 8 . 05 3 3 . 34 3 6 . 92 4 7 . 22

1 0.55 1 7 . 84 20. 0 1 25 . 59 30.40 36. 1 4 40. 0 1 5U8

9 .67 1 6 86 1 8.91 24. 1 9 28.73 34 . 1 6 3 7 . 82 48 . 3 8

8. 14 1 3 . 77 1 5 .44 1 9 . 75 23 .46 27 . 89 30.88 39 . 50

8 . 70 14.72 1 6. 5 1 2U2 2 5 . 09 2 9 . 82 3 3 . 02 42 . 24

8 . 43 1 4 . 25 1 5 .98 20.44 24 . 28 2 8 . 87 3 1 . 97 40 . 89

Table C - 5 0 is enlered in Ihe column appropriale 10 Ihe examinees age . The discrepancy Ihal is jusl lcss Ihan Ihe discrepancy obtained by the examinee is located . The entry in the snme row. first column. gives the percentage of the standardization population obtaining discrepancies as large as or larger than the located discrepancy. For example, a 2-year-old exami nee with a Verbal Comprehension-Nonverbal ReasoninglVisual ization discrepancy of 30 in either d i rection in the Stanford-Binet Intell igence Scale: Fourth Edition will be found in between 5 and 10 percent of the standardization population. Note.

Table C-S I Ranges of Standard Scores on the Stanford-Binet I ntelligence Scale: Fourth Edition for Subtests, Short Forms, Factor Scores, and Composite Score Age Subtests , Short Forms, Factor Scores, and Composite Score

00 '" '"

Vocabulary Comprehension Absurdities Verbal Relations Pattern Analysis Copying Matrices Paper Folding & Cutting Quantitative Number Series Equation Building Bead Memory Memory for Sentences Memory for Digits Memory for Objects Two Subtest Short Form Four Subtest Short Form Six Subtest Short Form Verbal Comprehension Nonverbal Reasoningl Visualization Memory Composite - Factor Composite - Full

2-0-0

2-3 - / 6

2 - 7- / 6

2-11-/6

3-3-16

3 - 7- 1 6

3-11-16

4-3 - 1 6

4- 7- 1 6

4-11-16

5-3 - / 6

5- 7- / 6

5-11-16

6-5-/6

6-/1-/6

L

H

L

H

L

H

L

H

L

H

L

H

L

H

L

H

L

H

L

H

L

H

L

H

L

H

L

H

L

H

42

82

40

82

38

82

35

82

33

82

32

82

30

82

29

82

28

82

28

82

27

82

26

82

25

82

24

82

23

82

40 49

82 82

38 46

82 82

36 43

82 82

34 40

82 82

32 37

82 82

31 36

82 82

30 34

82 82

29 33

82 82

29 32

82 82

28 31

82 82

28 30

82 82

27 29

82 82

26 27

82 82

25 25

82 82

24 22

82 82

44 48

82 82

41 45

82 82

38 42

82 82

36 40

82 82

33 37

82 82

31 35

82 82

30 33

82 82

28 31

82 82

27 30

82 82

26 29

82 82

25 28

82 82

24 27

82 82

23 26

82 82

23 24

82 80

22 23 45

82 77 82

52

82

50

82

47

82

45

82

42

82

40

82

38

82

36

82

34

82

32

82

30

82

29

82

28

82

27

82

25 46

82 82

53 48

82 82

50 45

82 82

46 42

82 82

43 40

82 82

39 37

82 82

36 34

82 82

34 32

82 82

31 29

82 82

30 28

82 82

29 28

82 82

28 27

82 82

27 26

82 82

26 25

82 82

25 24

82 82

24 23 37 38

82 82 82 82

84 1 64

79 1 64

73 1 64

68 164

62 1 64

59 164

5 5 164

52 164

50 164

49 164

46 1 64

44 164

42 1 64

41

164

39 1 64

94 1 64

8 9 1 64

8 1 1 64

75 1 64

6 8 1 64

64 1 64

5 9 1 64

55 1 64

5 2 1 64

4 9 1 64

4 6 1 64

44 1 64

4 1 1 64

4 0 1 64

37 1 64

9 2 1 64

86 1 64

7 8 1 64

7 2 1 64

6 5 1 64

60 1 64

56 1 64

5 1 1 64

4 8 1 64

4 6 1 64

4 3 1 64

40 1 64

3 8 1 64

3 6 1 64

3 6 1 64

90 1 64

8 3 1 64

77 1 64

7 1 1 64

6 5 1 64

60 1 64

56 1 64

5 1 1 64

5 0 1 64

4 9 1 64

4 7 1 64

44 1 64

4 1 1 64

3 9 1 64

3 6 1 64

100 1 64

94 1 64

86 1 64

80 1 64

72 1 64

66 1 64

6 1 1 64

56 1 64

53 1 64

49 1 64

46 1 64

44 1 64

4 1 1 64

39 1 64

36 1 64 7 1 1 64

9 5 1 64 9 5 1 64

8 7 1 64 8 7 1 64

8 0 1 64 80 1 64

73 1 64 7 3 1 64

6 6 1 64 66 1 64

6 0 1 64 60 1 64

5 5 1 64 5 5 1 64

5 0 1 64 5 0 1 64

4 7 1 64 4 7 1 64

44 1 64 44 1 64

4 1 1 64 4 1 1 64

3 9 1 64 3 9 1 64

3 6 1 64 3 6 1 64

3 6 1 64 3 6 1 64

3 6 1 64 4 3 1 64

(Table continues next page)

Table

C-S I

(cont.) Age

7-5- / 6

7-JJ-/6

8-5- / 6

8-11-/6

9-5- / 6

9-/J-/6

/0-5-/6

/0-JJ-/6

JJ-JJ-/6

/2-/J-/6

/3-JJ-/6

/4-JJ-/6

/5-/J-/6

/6-JJ-/6

/ 7-JJ-/6

Composite Score

L

H

L

H

L

H

L

H

L

H

L

H

L

H

L

H

L

H

L

H

L

H

L

H

L

H

L

H

L

H

Vocabulary Comprehension Absurdities Verbal Relations Pattern A nalysis Copying Mat rice� Paper Folding & Cutting Quant itat ive Number Series Equation Build ing Bead Memory Memory for Sentences Memory for Digits Memory for Objects

22 23 20

82 82 82

21 22 18

82 82 80

21 21 18

82 82 79

20 20 18

82 82 77

20 19 18

82 82 75

19 18 18

82 82 74

19 18 18

82 8! 72

18 18 18

82 79 69

18 18 18 45 18 18 30

82 75 66 78 63

18 18 18 44 18

82 73 63 76

18 18 18

82 72

18 18 18 39

18 18 18 38 18

80 65 55 68 57

18

77

18 18 38 18

63 52 67 57

18 18

72 60

18

82 69 58 70 57

57 78 78

Subtests . Short Forms. Factor Scores. and

IX> '" oo",j

Two Subtest Short Form Four Subtest Short Form Six Subtest Short Form

21

19 22 39

77 71 82

19 21 37

74 69 82

18 2! 36

71 67

18 20

41

79 73 82

82

34

82

22

82

82

20

82

44

82

82 82

20

82

21 42

40

82

39

82

19 37

43

81 74 82

24 45

22

23 22 36 38

82 82 82 82

20 22

22 21 35 37

82 82 82 82

21 21 34 37

82 82 82 82

20 20 34 35

82 82 82 82

19 20 33 35

82 82 82 81

18 20 32 33

68 64 82

18 19 33

66 62 82

18 18 32

64 59 80

82 82

18 36

82 81

18 35

82 79

82 82 82 80

18 20 32 33

82 82 82 78

18 18 31 32

82 82 82 74

47 18 33 47 18 18 30 31

78 75 82 82 82 82 72

61

41 18

60 73 59

18 29

55 76

28

75

27

74

26

73

25

45 18 32

43 18 31

70 75

42 18

66 71

82 80 72

31 43 18 18 27 28

40 18 29

18 29 30

68 80 82 82 79 71

67 74 67 80 82 82 77 70

41 18 30

46 18

75 76 70 82 82

44 18 18 28 29

41 18 18 26 28

66 78 80 82 75 70

41 18 18 26 28

38

66

18

57

71

24

68

65

40 18 28 39

65

69 65 78 70

80 72 69

18

18

25 27

66

64 76 16

76

71 69

3 6 1 64

36 1 64

3 6 1 64

36 1 63

36 1 5 9

36 1 54

36 1 52

3 6 1 50

3 6 14 9

36 1 4 7

3 6 1 44

36 1 4 2

3 6 1 40

3 6 1 37

36 1 3 1

36 1 64

3 6 1 64

36 1 64

3 6 1 64

36 1 64

3 6 1 64

3 6 1 64

3 6 1 62

3 6 1 60

36 1 5 7

36 1 56

36 1 54

36 1 50

36 145

36 1 40

36 1 64

3 6 1 64

36 1 64

3 6 1 64

36 1 64

3 6 1 64

36 1 64

3 6 1 62

36 ! 5 9

3 6 1 56

36 1 54

36 1 5 2

36 1 4 8

36 1 45

36 1 42

Verbal Comprehension Nonverbal Reasoning/ V isual ization Memory

36 1 64

3 6 1 64

36 1 64

3 6 1 64

3 6 1 64

3 6 1 64

36 1 64

36 1 6 1

36 1 55

36 1 5 2

36 1 49

43 1 54

42 1 4 8

42 1 4 3

42 1 3 6

3 6 1 64 70 1 64

36 1 64 5 3 1 64

36 1 64

36 1 64 49 1 64

36 1 64 4 8 1 64

36 1 64 4 6 1 64

36 1 60

5 2 1 64

46 1 64

36 1 64 4 3 1 64

36 1 62 4 1 1 64

36 1 59 39 1 64

36 1 5 7 38 1 64

36 1 55 36 1 64

36 1 5 1 36 1 64

36 1 45 36 1 5 9

36 1 4 2 36 1 55

Composite - Factor Composite - Full

36 1 64 39 1 64

3 6 1 64 37 1 64

3 6 1 64 36 1 64

3 6 1 64 36 1 64

3 6 1 64 36 1 64

36 1 64 36 1 64

3 6 1 64 3 6 1 64

3 6 1 64 3 6 1 64

3 6 1 63 38 1 64

3 6 1 60 37 1 63

36 1 5 7 36 1 5 9

36 1 5 7 38 1 60

36 1 5 3 37 1 5 7

36 1 50 3 6 1 54

36 1 45 36 1 4 9

Note. Standard scor�' for subtests are based on M = 50. SD = 8 . Standard scores for short forms and composites are based on M = 100. SD = 1 6 . Short forms a r e composed of t h e following subtests: (a) 2 subtcst short form: Vocabulary and Paltern Analysis ( b) 4 subtcst short form: Vocabulary, Pattern Analysis. Quantitative, and Bead Memory (c) 6 subte s t short form: Vocabulary. Comprehension. Pattern Analysis . Quantitative. Bead Memory, and Memory for Sentences Composite - Factor refers to composite based on faclOr scores: Composite - Full refers 10 composite score based on fu i l balte ry, using nonextrapolated scores. See Table 1 1-8 f o r the subtcsts that comprise the factor scores.

APPENDIX C

868

Table C-52 I nterpretive Rationales, Implications of High and Low Scores, and Instructional I mplications for Stanford-Binet I ntelligence Scale: Fourth Edition Su btests

A bility

Background factors

Possible illlplications of high scores

Possible implications of low scores

Instructional implications

Poor word knowledge Poor verbal comprehension Poor verbal skills and l anguage development Limited educational or family background Difficulty in verbalization Foreign language background Verbalization not encouraged in culture

Develop a working vocabulary Encourage child to discuss expe riences, ask questions, and make a dictionary Use other verbal enrichment exercises, includ ing Scrabble, analogy, and other word games

Poor social judgment Failure to take per­ sonal responsibil ity ( e . g . , overdepen­ dency, immaturity, l i mited invol vement w ith others) Overly concrete thinking Di fficulty in express­ ing ideas verbally C reative ind i v idual looking for unusual solutions

Help child understand social mores , customs, and soci­ etal activities, such as how other children react to things, how the government works, and how banks operate Discuss the actions of others to help chil­ d ren develop an awareness of social relationships and what is expected of them in terms of the behavior of others Role-play situat ions. such as reporting fires, call ing po­ l ice . and obtaining help for plumbing problems

Vocabulary

Verbal retrieval Word-finding abil ity Recall and verbal ident i fication by recognition of fa­ miliar objects ( P Y ) Perceptions (PY) Yerbal comprehension Language development Learning abil ity Fund of information Richness of ideas Memory Concept formation Long-term memory

Language stimu lation Environmental stimulation Alertness to environment Educational experiences Interests

Good word knowledge Good verbal comprehension Good verbal skills and language development Good family or cultural background Good schooling Good ability to conceptual ize Intellectual striving

Verbal comprehension Social judgment Common sense Use of pract ical knowledge and judgment in social situations K nowledge of con­ ventional standards of behavior Abil ity to evaluate past experience Moral and ethical judgment Reasoning Abil ity to evaluate sit­ uations and give a pertinent response

Extensiveness of cul­ tural opportunities Abil ity to evaluate and use past experience Development of con­ sc ience or moral sense

Good social judgment and common sense Good abil ity to recog­ nize social de­ mands when practical j udgment and common sense are necessary Knowledge of rules of conventional behav ior Good abil ity to orga­ nize knowledge Social maturity Good abil ity to ver­ balize wel l W ide experience

Comprehension

(Table continues next page)

869

MISCELLANEOUS TABLES Table C-S2 (cont.) Possible implicafions of high scores

Abilify

Backgrollnd facTOrs

Verbal comprehension Social intell igence Ability to isolate incongruities and ab­ surd ities in visual material Attention Concentration

Richness of early env i ronment Interests Cu ltural opportun ities

Good verbal comprehension Good abil ity to isolate i ncongruities and absurdities in visual material Good attention Good social i nte l l igence

Verbal comprehension Verbal concept formation Conceptual thi n k i ng Abstract and concrete reasoning abil ities Capacity for asso­ ciative thinking Abil ity to separate es­ sential from non­ essential detai l s Long-term memory Verbal reason ing C lassification Flexible thinking Abil ity to view facts from various angles at the same time and coordinate the multiple relation­ ships involved Ability to test and d is­ card hypothetical situations Flexibil ity and factual knowledge Abil ity to discrimi­ nate on an abstract or ideational level Ve rbally mediated ab­ stract thinking Receptive and ex­ pressive vocabulary General fund of i n formation Verbal retrieval and word-finding abil ity

Cogn itive response style ( reflective/ impulsive) Concentration Cognit ive flexibil ity Tolerance for frustration Cu ltura l opportunities Extent of reading Language stimulation Env i ronmental stimulation Educational experiences I nterests

Good conceptual thinking Good ability to see relationships Good abil ity to use l ogical and abstract t h i n k i ng Good abil ity to dis­ crim inate funda­ mental from superficial relationships Good abil ity to select and verbal ize ap­ propriate rela­ t ionsh ips between two objects or concepts Flexibil ity of thought processes Good verbal reasoning

Possible implicarions of low scores

Insfruclional implicarions

Poor verbal comprehension Poor abil ity to isolate i ncongruities and absurdities in v isual material Distractibil ity Poor concentration Poor social intel l igence

Emphasize how things are organized Focus on visual learn­ ing techniques stressing cause­ and-effect relationships

Poor conceptual thinking Difficulty i n seeing relat ionships Difficu lty in selecting and verbal izing ap­ propriate rela­ tionships between two objects or concepts Overly concrete mode of thinking Rigid ity of thought processes Negativism Poor verbal reasoning

Focus on recognition of di fferences and l i kenesses in shapes, textu res. and daily surroundings Stress l anguage devel­ opment. synonyms and antonyms, and exercises involving abstract words, classifications, and general izat ions

Absurdities

Verbal Relations

(Tclble cominues neXf page)

APPENDIX C

870 Table C-S2 (cont.)

A bility

Background factors

Possible implications of high scores

Possible implications of low scores

Instructional implications

Pattern Analysis

Nonverbal reasoning V isual-motor coordination Visual-spatial ability Abstract conceptualization ability Analysis and synthesis Visual discrimination V isual imagery Visual processing Nonverbal conceptual ability

Attention Concentration Cognitive style Problem-solving strategies Anxiety

Good visual-motorspatial integration Good conceptualizing ability Good spatial orientation in conjunction with speed, accuracy, and persistence Good analyzing and synthesizing ability Speed and accuracy in sizing up a problem Good eye-hand coordination Good nonverbal reasoning ability Good trial-and-error methods

Poor visual-motorspatial integration V isual-perceptual problems Poor spatial orientation

Use puzzles, block s , spatial-visual tasks, perceptual tasks involving breaking down an object and building it up again, and art work with geometric forms and flannel board Focus on part-towhole relationships and working with a model or key

Poor visual-motor ability Poor eye-hand coordination Poor fine motor coordination Poor perceptual discrmination Poor integration of perceptual and motor processes Maturational delay

Use pencils and crayons to develop fine motor ski l l s Use tracing activities with a variety of shapes and designs

Poor perceptual ability D ifficulty with analogic reasoning Poor attention to visual detail Poor concentration

Focus on part-whole relations Develop understanding of sequencing and ordering of events Use patterns to demonstrate how component parts are related

Copying

Visual-motor ability Eye-hand coordination Fine motor coordination Perceptual discrimination Integration of perceptual and motor processes

Motor ability Experiences

Good v isual-motor ability Good eye-hand coordination Good fine motor coordination Good perceptual discrimination Good integration of perceptual and motor processes Good maturation

Perceptual reasoning ability Analogic reasoning Attention to visual detail Concentration Simultaneous processing

Experience with partwhole relationships Willingness to respond when uncertain

Good perceptual ability Good analogic reasoning ability Good attention to visual detail Good concentration

Matrices

(Table col1linues next page)

87 1

MISCELLANEOUS TABLES Table

C-52

(cont.)

Abiliry

Background factors

Possible implications of high scores

Possible implications of 101>' scores

InSiructiollal implications

Poor visualization Poor spatial orientation Poor conceptualizing ability

Develop visualization skills through use of blocks , geometric figures, and other forms Focus on analyzing how objects are v iewed from various perspectives Work with paper cuttings, analyzing how forms look when unfolded

Inadequate ability i n mental arithmetic Poor concentration Distractibil ity Anxiety over a school-like task Blocking toward mathematical tasks Poor school achievement (perhaps as§ocillted with rcbellion against authoriiy Of with cultural background) Anxiety (e. g . , worry over personal problems)

Develop arithmetical skills Develop concentration skills Use concrete objects to introduce concepts Drill in basic arithmetical skills Develop interesting "real" problems to solve

Poor numerical reasoning ability Poor logical reasoning ability Poor concentration

Develop logical reasoning skills

Paper Folding and Cutting

Perceptual organization V isualization Spatial ability Visual-spatial ability Anention to visual cues

Familiarity with figures

Good visualization Good spatial orientation Good conceptualizing ability

Numerical reasoning ability Mental computation Application of basic arithmetical processes Concentration Anention Short-term memory Long ·term memory Perception Understanding of mathematical concepts, symbols , and vocabulary Acquired knowledge General fund of information

Educational experiences Environmental stimulation Alertness to environment Interests Oppornmities to acqui re fundamental arithmetical processes

Facility in mental arithmetic Good ability to apply reasoning skills in the solution of mathematical problems Good ability to apply arithmetical skiils in personal and social problemsolving situations Good concentration Good ability to focus anention Good ability to engage in complex thought panerns (for upper-level items, particularly) Teacher-oriented student

Numerical reasoning Logical reasoning Concentration

Education Cultural opportunities

Good numerical reasoning ability Good logical reasoning ability Good concentration

Quantitative

Number Series

(Table continues next page)

APPENDIX C

872 Table C-S2 (cont.)

Abiliry

Background fact rs

Knowledge of numbers Knowledge of con­ ventional arith­ metical operations Flexibility in rear­ ranging and manip­ ulating materials Numerical reasoning Logical reasoning Application of basic arithmetical processes Concentration Willingness to engage in trial and error

Education Cultural opportunities

Poss:ible implicarions o{high scores

Possible implicarions of low scores

insrructional implicarions

Poor knowledge of numbers Poor knowledge of conventional arithmetical operations Poor ability to rearrange and manipulate arithmetical materials Poor numerical reasoning ability Poor logical reasoning ability Poor concentration

Develop number skills Help child understand basic arithmetical operations Stress relationship between numbers Use a weighing scale to show how numbers can be equated or balanced

Anxiety Inattention Distractibility Poor visual memory

Emphasize looking skills by showing pictures and asking child to recall de­ tails of pictures Help child learn to discriminate fea­ tures of objects Ask child to arrange different geometric designs in order and then recall the order Use other memory exercises and games

Anxiety Inattention Distractibility A possible learning deficit Difficulty in auditory sequencing

Emphasize listening skills by reading sentences of in­ creasing length and asking child to re­ call details Give directions of in­ creasing complex­ ity and have child follow them Use other memory exercises and games

Eql!lation Building Good knowledge of numbers Good knowledge of conventional arithmetical operations Good ability to rearrange and manipulate arithmetical materials Good numerical reasoning ability Good logical reasoning ability Good concentration

Bead Memory

Short-term visual memory Rote memory Form perception and d iscrimination Alertness to detail Ability to perceive spatial relations Attention Conceniration Simuitaneolls processing Fluid ability Eye-hand coordination

Ability to recei ve stimuli passively

Short-term auditory memory Immediate recall of verbal information Rote memory Familiarity with words Auditory acuity Attention Concentration Listening comprehension Sequential processing Auditory processing Fluid ability

Ability to receive stimuli passively Cultural opportunities

Good rote memory Good immediate re­ call ability Ability to attend well in a testing situation Good ability to attend to visual stimuli

Memory for Sentences

Good rote memory Good immediate recall ability Good ability to attend well in a testing situation

(Table conrinues n exr me)

873

MISCELLANEOUS TABLES Table

C-S2

(cont.)

Ability

Background jactors

Possible implications oj high scores

Possible implications of 10\\' scores

Instructional implications

Anxiety Inattention Distractibility A possible learning deficit Difficulty i n auditory sequencing Poor learning strat­ egies (rehearsal, chunking/grouping)

Emphasize listening skills by using se­ quencing activities , reading a short story and asking child to recall de­ tails, and having child follow directions Use short and simple directions and re­ peat when necessary Use other memory exercises and mem­ ory games

Anxiety Inattention Distractibility A possible learning deficit Difficulty in v isual sequencing

Emphasize looking skills by showing pictures and asking child to recall details Help child learn to discriminate fea­ tures of objects Ask child to arrange pictures in order and then recall order Use other memory exercises and games

Memory for Digits

Freedom from distractibility Short-term memory Rate memory Immediate auditory memory Attention Concentration Auditory sequencing V isualization Perceptual reorganization ability Fluid ability

Ability to receive stimuli passively

Good rote memory Good immediate re­ call ability Ability to attend well in a testing situation Good ability to attend to auditory stimuli Good learning strat­ egies (rehearsal , chunking/grouping)

Short-term visual memory Rote memory Immediate v isual memory Alertness to detail Attention Concentration Successive processing

Ability to receive stimuli passively

Good rote memory Good immediate recall ability Good ability to attend to visual stimuli Ability to attend well in a testing situation

Memory for Objects

Note. PV = Picture Vocabulary.

874

APPENDIX C

Table C-S3 Percentile Ranks for Standard Scores for Subtests on the Stanford-Binet I ntelligence Scale: Fourth Edition

Standard score

Percentile rank

Standard score

Percentile rank

Standard score

82 81 80 79

99. 99 99 . 99 99. 99 99 .98 99 .98 99.97 99.94 99. 9 1 99. 87 99.79 99.70 99.56 99. 3 8 99 . 1 1

60 59 58 57 56 55

89 87 84 81

38 37

78 77 76 75 74 73 72 71 70 69 68 67 66 65 64 63 62 61 Note. M

=

50, SD

=

99 98 98 97 96 95 93 91 8.

54 53 52 51 50 49 48 47 46 45 44 43 42 41 40 39

77 73 69 64 60 55 50 45 40 36 31 27 23 19 16 13 1 1 9

36 35 34 33 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18

Percentile rank 7

5 4 3 2 2 1

. 89 . 62 .44

. 30 .21 . 13 .09 . 06 .03 . 02 .02 .01 .01 .01

MISCELLANEOUS TABLES

875

Table C-S4 Confidence I ntervals for the M cCarthy Scales

Age level 2 1/2

( 2 -4- 1 6

through 2 - 8 -2 9 )

3

( 2- 9 -0

through 3 -2-2 9 )

3 1/2

( 3 - 3 -0

through 3 - 8 - 29 )

4

( 3 - 9 -0

through 4-2-2 9 )

4 1/2

(4- 3 - 0

through 4- 8 -2 9 )

5

(4-9-0

through 5-2-2 9 )

5 1/2

( 5 - 3 -0

through 5 - 1 1 -2 9 )

Confidenc. e level Scale

68 %

85 %

Verbal Perceptual-Performance Quantitative General Cognitive Memory Motor

± 3 ± 5 ± 5 ± 4 ± 5 ± 4

± 5 ± 7

Verbal Perceptual-Performance Quantitative General Cognitive Memory Motor

± 3 ± 4 ± 4 ± 4

± 5 ± 5 ± 6 ± 5 ± 7

Verbal Perceptual-Performance Quantitative General Cognitive Memory Motor

± 3 ± 3 ± 4 ± 3 ± 4

Verbal Perceptual -Performance Quantitative General Cognitive Memory Motor

± ± ± ± ±

Verbal Perceptual-Performa nce Quantitative General Cognitive Memory Motor

± ± ± ± ± ±

Verbal Perceptual-Performance Quantitative General Cognitive Memory Motor

± ± ± ± ± ±

Verbal Perceptual-Performance Quantitative General Cognitive Memory Motor

± 5 ± 4

± 7 ± 6 ± 7 ± 6

± 6

90%

± 7 ± 8 ± 8 ± 7

± 8 ± 7 ± ± ± ± ± ±

5 6 7 6 8 7

±

95 %

99%

± 6 ± 9 ± 10 ± 8 ± 9 ± 8

± ± ± ± ± ±

± ± ±

8 ± 7 ± 10 ± 8

i ll

± 6 ± 7 ± 10 ± 9 ± 8 ± 9

± 8 ± 9 ± 14 ± 12 ± 10 ± 12

5 4 ± S

± 5 ± 6 ± 9 ± 8 ± 7

± 7

± g

4

± 5

3 5 4

± 5 ± 7

± ± ± ± ± ±

3 4 5

5 4

± 5 ± 7 ± 6

6

±

5 6

I

....

5 8 6 8 7

± 6 ± 9 ± 7 ± 10 ± 8 ± ±

±

6 6 6 7

± ±

8 7

± 8 ± 10 ± 8 ± ± ± ±

± 10 ± 9

5 4

± ± ± ± ± ±

5 5 5 6 7 6

± 4 ± 4 ± 4 ± 4

± ± ± ±

5 6

± ±

5 6

± ±

6 6 6 7

± 5 ± 4

± 8 ± 6

± ±

9 7

4 4 4 4

± ±

±

i ll

8 8

± 5 ± 5 ± 8 ± 7 ± 6

± 4

± 8 ± 9 i ll ± 10 ± 13

8 7

5 6 6

± ± ± ± ± ±

4 5 6

II 12 10

± 7 ± 8 i ll ± 9

5 ± 5 ± 7 ± 6 ± 7 ± 7

± ± ± ± ± ±

6 7

8 12 13

±

7 7 7

7 8 7 8

± II

±

9

± 9 ± 12 ± 10 ± 13

± II

± 9 ± 10 ± 10 ± 10 ± 13

i ll ± ± ± i

9 10 10

ll ± 14 ± 12

(Table continues next page)

APPENDIX C

876 Table C-S4 (cont.) Confidence level

Age level

6 '/2 ( 6 -0-0

t h rough 6 - 1 1 ·29)

7 '/2

( 7-0-0

t h rough 7 - 1 1 ·29)

8 '/2 (8-0-0

through 8 - 1 1 -29)

Average

Scale

Verbal Perceptual - Performance Quantitative General Cognitive Memory Motor Verbal Perceptual- Performance Quantitative General Cognitive Memory Motor

68 %

85 %

± ± ± ±

± ± ± ± ±

4 S 4 S

95 %

99%

± ± ± ±

± 8 ± 9 ± 8 ± 10 ± 8 ± II

± ± ± ± ± ±

10 12 II 13 10 14

± ± ± ± ± ±

9 I I II 10 10 13

± ± ± ± ±

6 8 7 8 7

± 8

± ±

4 S

± ± ± ± ± ±

S 6 6 6 6 7

± ± ± ± ± ±

7 8

± 6 ± 8 ± 8 ± 8 ± 8 ± 10

S 7 6 6

± 7 ± 10 ± 8 ± 9 ± 8 ± 12 ± ± ± ± ± ±

± 4 ± 6 ± 3

± ± ± ± ±

6 7 6 7 6

90 %

4 4 4

9 S 7 7 6

Verbal Perceptual -Performance Quantitative General Cognitive Memory Motor

± 4 ± S ± 4 ± 4

± ± ± ±

± 4 ± 6

± 6 ± 9

± 6 ± 8 ± 7 ± 7 ± 7 ± 10

Verbal Perceptual- Performance Quantitative General Cognitive Memory Motor

± ± ± ± ± ±

± ± ± ± ± ±

± ± ± ± ± ±

3 4 4 4 4 S

S 6 6 6 6 7

6 7 7 7 7 8

7 8 8 8 9 9

f lO

±

± ± ± ± ±

13 II 12 II 16 9 10 II II 12 12

NOfe. See Table C · I for an explanation of the method lIsed to arrive at confidence intervals.

Table C-55 Differences Required for Significance When Each McCarthy Scale Index Any I ndividual Child

Note.

15

Compared to the Mean Scale I ndex for

Index

. 05

. 01

Verbal Perceptual Performance Quantitative Memory Motor

9 . 34 9 . 87 10.22 10.57

8 . 36

9 . 97 1 1 . 16

See the Note in Table C·3 for an explanation of how the deviations were obtained.

1 1 .78 1 2 .20 1 2 . 62

877

MISCELLANEOUS TABLES Table C-S6 I nterpretive Rationales and I mpl ications of H igh and Low Scores for K-ABC M ental Processing and Achievement Subtests

Ages

Ability

Possible implications of high scores

Backg round factors

Possible implications of low scores

Hand Movements 2 -6

through 1 2 -5

years

Sequential processing and simultaneous processing Fluid ability Perceptual organization Short-term visual memory Spatial abil ity Visual-motor coordination

Experience with visual­ motor integration tasks

Good attention span Good concentration

Inattention Poor concentration Distractibility Perseveration Anxiety Visual-motor coordina­ tion difficulty

Good rote memory Good immediate recall ability Ability to attend well in a testing situation Good ability to attend to auditory stimuli

Anxiety Inattention Distractibility A possible learning deficit Difficulty in auditory sequencing

Good attention to visual detail Good concentration Good perceptual organi­ zation and closure Good reasoning

Inattention Distractibil ity Poor perceptual organiza­ tion and closure Poor reasoning

Good visual-motor-spatial integration Good conceptual izing ability Good spatial orientation in conjunction with speed, accuracy, and persistence Good analyzing and syn­ thesizing ability

Poor visual-motor-spatial integration Visual-perceptual problems Poor spatial orientation Poor nonverbal reasoning ability

Good concentration Good attention span

Distractibility Inattention Anxiety

Number Recall 2 -6

through 1 2 -5

years

Sequential processing Fluid ability Short-term auditory memory

Ability to receive stimuli passively

Simultaneous process ing Attention to visual detail Perceptual organization and closure Ability to convert ab­ stract stimuli into a concrete object

Alertness to environment Willingness to respond when uncertain

Simultaneous processing Nonverbal concept formation Spatial visualization Perceptual organization Analysis and synthesis V isual-motor coordination

Rate of motor activity

Sequential processing Fluid ability Short-term auditory memory

Cultural opportunities

Gestalt Closure 2 -6

through 1 2 -5

years

Triangles 4-0

through 1 2 -5

years

Word Order 4-0

through 1 2 -5

years

(Table cOl1linues next page)

878

APPENDIX C

Table C-S6 (cont.)

Ages

AbiliTy

Possible implicaTions of high scores

Background aCTors

Possible implicaTions of low scores

Magic Window

2-6 through 4- 1 1 years

Simultaneous processing Attention to visual detail Short-term visual memory

Cultural opportunities

Good attention span Good concentration

Inattention Distractibility Impulsivity

Face Recognition

2-6 through 4- 1 1 years

Simultaneous processing Attention to visual detail Short-term visual memory

Ability to receive stimuli passively

5-0 through 1 2-5 years

Simultaneous processing Analogic thinking Attention to visual detail Perceptual reasoning ability

Experience with partwhole relationships Willingness to respond when uncertain

5-0 through 1 2-5 years

Simultaneous processing Immediate recall in a spatial context

Experience with developing strategies for sol ving problems

Good attention span Good concentration

Inattention Distractibility Impulsivity

Matrix Analogies

Good analogic thinking Good attention to visual detail Good perceptual reasoning ability

Poor analogic thinking Poor attention to visual detail Poor perceptual reasoning abil ity

Good concentration Good attention span

Poor concentration Inattention Distractibility Anxiety

Good planning ability Alertness to detail Good ability to synthesize parts into an intelligi­ ble whole Good attention span Concentration

Difficulty with visual or­ ganization (sequencing) Difficulty in anticipating events Inattention Poor concentration Impulsivity

Spatial Memory

Photo Series

6-0 through 1 2-5 years

Simultaneous processing Nonverbal reasoning ability Planning ability Alertness Visual sequencing Attention to details Perceptual organization

A minimum of cultural opportunities

Expressive Vocabulary

2-6 through 4- 1 1 years

Language ability Word knowledge Long-term memory Recall and verbal identification by recognition of familiar objects

Cultural opportunities

Good knowledge of words Good verbal ski l l s Good family o r cultural background

Poor knowledge of words Poor verbal skills Limited educational or family background

(Table cOl1linues neXT page)

879

M ISCELLAN E OUS TABLES Table C-56 (cont.) Possihle imp/iicQ{iolls Ages

AiJilirv

Background facrors

Faces and Places

2-6 through 1 2 -5 years

Range of knowledge Long-term memory

Cultural opportunities

of

high sc:ores

Good range of" specific knowledge Possession of Iknowledge associated '"" i th cultural and educational environmen t Good memory

Possible implicarions of low scores

Poor range of specific knowledge Poor memory Foreign background

Arithmetic 3 -0

through 1 2 -5

years

Numerical reasoning abil ity Mental computation Application of basic arithmetical processes Concentration Attention Memory

Opportunity to acquire fundamental arithmetic processes

Faci l ity in mental arithmetic Good reasoning skill in mathematical problems Good concentration Good ability to focus attention Teacher-oriented student

Inadequate abil ity in mental arithmetic Poor concentration Distractibil ity Anxiety over a sehool­ l ike task Blocking toward mathe­ matical tasks Poor school achievement Anxiety

Good verbal reasoning abil ity Good ability tu integrate sequentially presented auditory stimuli Good concentration Good attention span

Poor verbal reasoning abil ity Difficulty in integrating sequentially presented auditory stimu l i Poor concentration Inattention

Riddles 3-0

through 1 2 -5

years

Ve rbal reasoning Integration of sequen­ tially presented au­ d itory st imuli

Cu ltural opportunities

Reading/Decoding 5 -0

th rough 1 2 -5 years

Reading skills Long-term memory

Cu ltu ral opportunities Education

Reading comprehension A b i l ity to integrate au­ d itory information with a motor response

R ichness of early envi ronment Cultu ral opportunities Education Motor abil ity

Good reading skills Good long-term memory

Poor read ing skills Poor l ong-term memory

Reading/Understanding 7 -0

through 1 2 -5 years

Good reading comprehension Good auditory-motor in­ tegration of meaningful information Good attention span

Poor reading comprehension Poor auditory-motor inte­ gration of meaningful information Inattention

No il'. This table i s based i n part on Kaufman and Kaufman ( 1 98 3 ) . See Kaufman and Kaufman for a more complete discussion of each of the K - A B C subtests.

880

APPENDIX C

Table C-S7 Norms for the Coloured Progressive Matrices Age

Total score

5� 5. 03 to 5. 08

6 5. 09 to 6. 02

6� 6. 03 to 6. 08

35 34 33 32

7 6. 09 to 7 02

7� 7 03 to 7 08

31 30

99

97 95

98 96 94 92

29 28

97 95

93 91

90 87

99

7. 09 to

8. 02

90

83

75

87 84

79 75

69 64

92

87 84

80 75

70 65 60 55 50

59

22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6

83 79

75 70 66 61 56

75 70 66 62 56 48 31 16 10 6 2 1

80 75 70

70 65 60

65 55 50

55 50 45 39

44

33

24

36 25

24 15

10

15 10 7

10 7 5

16 10

6 3

5

3

3

1

50 39 26 17

60

45 41 37 31

7 5 3

11 1 0. 09 to 1 1 . 02

110 1 1 . 03 to l J . 09

99

97 95 93 90 85 76

95 93

95

90

94 90 83 75

76 67

65 60

66 57

94 89 77 64 56 50

88 75

85

93 89 78

67 60

60 55

50

44

50

55 50 45

50

45

45 39

40 35

34 30 25 21 17

30 25 21 17 14

76 68 63 58 54

93

90 87 83 79

10� 10. 03 to 1 0. 08

86 81

92 90

96

10 9. 09 to 10. 02

90 85

98

94 93 90 87

9� 9. 03 to 9. 08

92 90

96 94

23

9 8. 09 to 9. 02

97 95 93

26 25 24

8� 8. 03 to 8. 08

98 96 94

27 98

(in years)

55

50 45 41 36 30

39 34

36 31 25 18 13

24

14 12

12 10

10

17 14 12 10 8

10 8 6

8 6 5

10 7 5 3

8 6 5 4

6 5 4 2

5 4

4 2

54 50 45 41

1

20 15 12

29 25 21

44

39

40

34

35

30 25

30 25 21 17 14 12 10 8 6 5 4

21 17 14 12 10 8 6 5 4 2

65 57 50

44

39 34 29 25 21 17

14 12 10 8 6 5 4 2

2

2

2

1

1

Note. These are smoothed norms for U . S . children. Source: Reprinted with permission of the authors and publisher from J . C. Raven and B. Summers, Manual for Ravens Progressive Matrices and Vocabulary Scales - Research Supplement No. 3, 1 986, p. 36. London: H. K. Lewis & Co.

Table C-58 Norms for the Standard Progressive Matrices Age (in years)

Total score

6 V2 6. 03 to 6. 08

7 6. 09 to 7. 02

7V2 7. 03 to 7. 08

8 7. 09 to 8. 02

8 V2 8. 03 to 8. 08

9 8. 09 to 9. 02

9 V2 9. 03 to 9 08

10 9. 09 to 10. 02

1 0 V2 10. 03 to 10. 08

If

10. 09 to 1 1 . 02

f l V2 1 / . 03 to 1 1 . 08

12

J 1 . 09

to 12. 02

1 2 V2 12. 03 to 12. 08

13 1 2. 09 /0

1 3. 02

1 3 V2 13. 03 to 13. 08

14 1 3. 09 to / 4. 02

1 4 V2 1 4. 03 to 1 4. 08

15 1 4. 09 to 15. 02

15 V2 15. 03

16 15. 09

1 6 V2 1 6. 03

15. 08

16. 02

1 6. 08

99

99

/0

60 59

58

99

57 56 55 54

99

53

52 51 50

Q) Q)

49

99

48

99

47 99

46

98

45 99

44 43

42

99

96

98

95

97

9�

35

97

95

93

90

83

67

68

62

94

92

90

86

81

75

70

65

92

90

81

75

70

65

6!

58

53

90

86

75

70

65

87

81

70

65

61

8:)

,"

75

70

86 81 75

70

65

50

46

50

46

43

38

46

42

39

34

35

34

24

30

30

31

25

29

25

25

. 20

21 i8

39

33

32

57

50

40

81

71

63

53

46

68

60

50

43

31

96

78

91

85

75

65

56

46

89

82

72

61

53

43

40 37

43

37

34 31

46

35 28

25

61 57

57

53

53 50

53

53

34

61

57

57

57

37

46

65

61

61

61

6]

42

53

87

----

75

70

38

89

-

75

75

42

94

95

81

81

42

95

30

86

86

46

97

59

90

70

98

63

92 90

75

33

59

92

50

67

65

64

94

SO

75

69

68

94

54

84

75

74

96

57

91

82

97

61

61

87

92

70

70

87

90

� 10. 03

11 10. 09

l l Y> 1 1 . 03

12 1 1 . 09

10. 08

1 1 . 02

1 1 . 08

12. 02

28 25 22 19

22 19

31

34 31 28 25

15 14 12 I I

12 II 10 9

28

22

17

35 32

25 22

19 17

29 25

19 17

15 13

10 9 8 7

8 7

20

15

II

10 8

6

41

39 36 32

15 13 II

13 II 10 9 8 7

5 4

38 34 30

29 25 20

17 14

10 8 7

7

6

24

15 13 10 8

13 II 10 8 7

9 Y> 9. 03

58 56 53 50 47 44 42

50 47 44

40 37

41 38

8 7. 09

8 Y> 8. 03

8. 02

8. 08

87 85 82 78

78

69

75 71

65 62 59

75 69

65 62 59

66 63

53

50 47 44

72

60

50

19 18 17

69 66 62

16 15 14 13 12

59 55 49 39 26 18 14

57 53 49 45 40

46 43 39 35 30

33 25

24 18 13

18 13 10 8

10 7

6 5

5 4 3 2

4 3 2

Total score 29 28 27 26 25 24 23 22 21 20

CD CD tv

10 9. 09 to 10. 02

9 8. 09 to 9. 02

7V> 7. 03 to 7. 08

II 10 9 8 7 6 5 4 3

6Y> 6. 03

7 6. 09

6. 08

7. 02

93 91 89 87 85 83 81 78 75

(0

10 7 5 4 3 2

(0

72

18 13 10 7 5 4 3 2

68

56

(0

56 53

(0

6 5 4 3 2

II 10 8 7

(0

9. 08

34

6 5 4 3

6 5 4 3 2

2

I

6 5 4 3 2

(0

6 5 4 3 2

(0

16 14

(0

6 5

5 4

4 3 2

2

I

3

(0

6 6

3

2

1 2 Y> 1 2 . 03 to 12. 08

13 12. 09 to 13. 02

II 10 9

9 8 7

8 7

6

13 Y> 13. 03 (0

13. 08

14 1 3 . 09 to 14. 02 7

14. 08

15 14. 09 to 15. 02 5 4 3 2 I

14Y> 1 4. 03 (0

5 4

6

6 5 4

6 5 4 3

3

2

6 6

5 4

2

I

5 4

3

I

3

I

8 7

3

6

2

1 5 Y> 15. 03

16 15. 09

1 6 Y> 1 6. 03

15. 08

1 6. 02

1 6. 08

(0

5 4 2

(0

4 3

(0

2 I

I

I

I

2

2 I

I

I

I

I

I

I

I

I

I

I

Note. These are smoothed norms for U . S . children. Source: Reprinted with permission of the authors and publisher from J. C. Raven and B . Summers, Manualfor Raven s Progressive Matrices and Vocabulary Scales - Research Supplement No. 3. 1 986, p . 1 5 . London: H. K . Lewis & Co.

MISCELLANEOUS TABLES

883

Table C-S9 Norms for the Advanced Progressive Matrices

Percentile rank

Total score 13 15 17 18 19 20 21 22

I 3 4 6 7 II 14 18 24

23 24 25

29 37

Total score 26 27 28 29 30 31 32 33 34 35 36

Percentile rank 43 52 57 65 74 81 86 89 93 98 100

Note. B a sed on 300 students a, the University of Cal i fornia, Berkeley. Source: Adapted from Paul ( 1 985).

Table C-60 Landmarks of N ormal Behavior Development La ng u ag e

Personal and social behavior

Age

MOlOr behavior

Adaptive behm·ior

Under 4 weeks

Makes alternat i ng c rawl i ng movements Moves head l aterally when placed in prone position Tonic neck reflex posi­ tions predominate Hands fisted Head sags but can hold head erect for a few seconds Symmetrical postures predominate Holds head balanced Head l ifted 90 degrees when prone on forearm Sits steadily, leaning forward on hands Bounces actively when placed i n standing position

Responds to sound of rattle and bell Regards moving objects momentarily

Small , throaty, u n­ difl-crentiated noises

Quiets when picked up I mpassive face

Follows moving objects to thi'! midl ine Shows no interest and drops objects immediately

Beginning vocalization, �uch as coning, gur­ gli n g , and grunting

Regards face and di­ mini�hes activity Responds to speech

Follows a slowly mov­ ing object wel l A rms activate o n sight of dangl ing object

Laughs aloud Sustained cooing and gurgling

Spontaneous social smile Aware of strange situations

One-hand approach and grasp of toy Bangs and shakes ratt Ie Transfers toy

Vocal izes "m-m-m" when crying Makes vowel sounds, such as "ah, ah"

Takes feet to mouth Pats m i rror image

4 weeks

1 6 weeks

28 weeks

(Table continues next page)

APPENDIX C

884 Table C-60 (cont.)

Age 40 weeks

5 2 weeks

15 months

1 8 months

2

years

3 years

4 years

5 years

Adaptil'e behal'ior

Morar behavior Sits alone with good coordination Creeps Pulls sel f to standing position Wal ks with one hand held. Stands alone briefly Toddles C reeps upstairs

Language

Matches t\V o objects at midl ine Attempts to imitate scribble

Says "da-da" or equivalent Responds to name or n ickname

Releases cube in cup Tries tower of 2 cubes

Uses expressive jargon Gives a toy on request

Wal k s , seldom fal l s Hurls bal l Walks upstairs w ith one hand held

Builds a tower o f 3 o r 4 cubes Scribbles spontaneously and i m itates a w riting stroke

Runs well , no fal l ing Kicks large ball Goes upstairs and downstairs alone

Builds a tower of 6 or 7 cubes A l igns cubes . imitating train I mitates vertical and

Rides tricycle Jumps from bottom steps A l ternates feet going upstairs Walks downstairs one step per t read Stands on one foot for 4 to 8 seconds

Builds a tower of 9 or 10 cubes Imitates a 3-cube bridge Copies a circle

S kips. using feet alternatively Usually has complete sphincter control

Copies a square Draws a recognizable man with a head . body, l i mbs Counts 1 0 objects accurately

-

circular

st rokes

Copies a cross Repeats 4 digits Counts 3 objects with correct poi nting

Personal and social behavior Responds to social play. such as "pat-a-cake" and " peek-a-boo" Feeds sel f cracker and holds own bottle Cooperates in d ressing "Plays" ball

Says 3 to 5 words meaningfully Pats pictu res in books Shows shoes on request Says 10 words, including name Identifies one common object on picture card Names ball and carries out two d i rections , for example "put on table" and "give to mother" Uses 3-word sentences Carries out fou r simple d i rections

Points or vocal izes wants Th rows objects in play or refusal Feeds self in part. spi l l s Pulls toy on string Carries or hugs a special toy, such as a doll

Gives sex and ful l name Uses plurals Describes what is happening i n a picture book Names colors. at least one correctly Understands five prepositional d i rectives " on," " under," "in," " in back of" or "in front of, " and "beside" Names the primary colors Names coins: pennies. nickels, d imes Asks meanings of words

Puts on shoes Unbuttons buttons Feeds self wel l Understands taking tu rns Washes and d ries own face Brushes teeth Plays cooperatively with other children

Pulls on simple garment Domestic mimicry Refers to self by name

Dresses and undresses sel f Prints a few letters Plays competitive exercise games

Source: Reprinted, with a change in notation , with permission of the publisher and author from S. Chess, "Health Responses, Developmental Disturbances, and Stress or Reactive Disorders: I: Infancy and Childhood . " In A . M . Freedman and H . L Kaplan (Eds . ) . Comprehensive Textbook of Psychiatry, p . 1 362, Copyright 1 967, Williams & Wilkins, 1 967 .

885

MISCELLANEOUS TABLES Table C-6 1 Differences Requi red for Significance When Each Peabody I ndividual Achievement Test (PlAT) Su btest Is Compared to the M ean Score for Any I ndividual Child

Sublest M athematics Reading Recogn ition Reading Comprehension Spel l i ng Genera! I nformation

. 05

01

1 7 . 66

2 1 . 19

1 3 . 32

1 5 . 98

20 04

24.04

1 9. 8 1

2 3 . 76

17.15

20.58

Note. See

the Note i n Table C-3 for a n explanation of how the deviations were obtained. Source: Adapted from Silverstein ( 198 1 ) .

Table C-62 Standard Scores for the Koppitz Developmental Scoring System Chronological age

5-0 to

5-6

9- 0

9-6

10-0

1 0-6

{()

fa

10

8-5

8- 1 /

9-5

10

9- 1 /

1 0-5

/ 0- 1 1

7- 6

8-0

to

fa

to

6- 1 /

7-5

7- 1 1

6-6

to

8-6

7-0

6-0

to

tu

to

1 / -0

to

1 /- 1 /

Errors

5-5

5- 1 /

6-5

0

1 60

1 43

1 39

131

1 26

1 25

1 25

1 18

1 19

1 16

1 15

1 15

1 15

1 55

1 38

135

1 27

1 22

1 19

1 19

I I :?

1 12

1 09

1 07

1 07

1 04

2

1 50

1 34

1 30

1 22

1 17

1 14

1 13

1 06

94

3

1 46

1 30

1 25

1 18

1 13

1 09

1 07

1 00

4

141

1 25

121

1 14

108

1 03

10i

1 05

1 02

99

98

99

95

91

90

83

94

92

88

83

82

72

73

61

5

1 37

121

1 16

1 09

1 04

98

95

88

85

81

76

6

1 32

1 16

1 12

1 05

99

92

89

82

78

74

68

7

1 28

1 12

1 07

101

95

87

83

76

71

66

60

8

1 23

1 08

1 03

97

82

77

52

1 19

98

92

85

76

71

58

59

9

52

I OJ

90

10

1 14

99

94

88

81

71

65

II

1 10

94

89

84

76

66

59

12

1 05

90

85

79

72

60

53

13

1 00

85

80

75

67

55

47

14

96

81

75

71

63

50

15

91

77

71

67

58

16

87

72

66

62

54

17

82

68

62

58

49

18

78

63

57

54

45

19

73

59

53

49

20

69

55

48

45

21

64

22

60

23

55

7U

65

58

51

64

65

51

57

45

52

46

50 46

NOTe. These standard scores (M

=

100. SD

=

1 5 ) are based on a l i near transformation of the data obtained from E . M . Koppitz's ( 19 7 5 ) 1974 normative

sample. Standard scores are useful primarily

fro m 5 to 8 years of age . A fter the age of

developmental scores make standard scores not too mea n i ng fu l .

8 years. the low c e i l i n g and the skewed distribution of

APPENDI:X C

886 Table C-63 Norms for Conners Parent Rating Scale (48-ltem FOlrm)

Total score

3-5

6-8

Male

Female

Age (in years)

Age (in years)

9-1 1

12-14

/5- / 7

3-5

6-8

9-11

35 40

39 42 46 49 53 56

12-14

15- 1 7

40

39 43 46 50 54 58

Conduct Problem 0 I 2 3 4 5 6 7 8 9 10 I I 12 13 14 15 16 17 18 19 20 21 22 23 24

36 40 43 46 49 52 56 59 62 65 68 72 75 78 81 84 88 91 94 97 101 104 107 1 10 1 13

38 41 44 47 50 53 56 59 63 66 69 72 75 78 81 84 88 91 94 97 1 00 1 03 1 06 1 09 1 13

36 39 43 46 49 53 56 59 62 66 69 72 76 79 82 85 89 92 95 99 1 02 1 05 1 08 1 12 1 15

38 41 44 47 50 53 56 59 62 65 69 72 75

39 42 45 48 51 54 56 59 62 65 68 71 73

99 1 02 1 05

76 79 82 85 88 90 93 96 99 1 02

108 III

105 1 08

78 81 84 87 90 93 96

36 40 43 47 50 54 57 61 65 68 72 75 79 82 86 90 93 97 1 00 1 04 1 07 III 1 15 1 18 1 22

44

49 53 58 62 67 71 76 80 84

60 63 67 70 74 77

89 93 98 1 02 1 07 III 1 16 1 20 1 25 1 29 1 34 1 38 1 43

81 84 88 91 94 98 101 1 05 1 08 1 12 1 15 1 19 122

38 45 51 58 64

39 45 52 58 65 72

43 47 50 53 56 59 62 65 68 72 75 78 81 84 87 90 93 97 1 00 1 03 1 06 1 09 1 12 1 15

62 65 69 73 77 80 84 88 92 96 99 1 03 1 07 III 1 15 1 18 1 22 1 26 1 30

Learning Problem 0 I 2 3 4 5

35 42 50 58 65 73

36 41 47 52 58 64

6 7 8 9 10 II 12

80 88

69 75 80 86 91 97 1 02

95 103 III 1 18 1 26

38 43 47 52 57 62 67 71 76 81 86 91 95

38 43 47 52 56 60 65 69 74 78 82 87 91

39 43 48 52 57 61 66 71 75 80 84 89 93

39

44 48 52 57

61 65 70 74 79 83 87 92

71 78 84 91 97 1 04 III 1 17

78 85 91 98 1 04 III 1 18

40 46 51 57 62 68 74 79 85 90 96 101 1 07

41 47 54 61 67 74 80 87 93 1 00 1 07 1 13 1 20

(Table continues next page)

MISCELLANEOUS TABLES

887

Table C-63 (cont.)

Female

Male Age (in years)

Tora! score

3-5

6-8

0

45

1 2 3 4

62 79

44 55 66 77

9-1 1

-

12-14

15-1 7

A8e

(in years)

6 -8

9- 1 1

44

43

59 74

44

52

53 62 71 80 89

3-5

12-14

15-/ 7

42 51 60 69 78 86

42 52

Psychosomatic

5

95 1 12 1 29

6 7 8 9 10 11 12

145 1 62 1 79 1 95 212 229 245

43 53 62 72 82 91 101 1 10 1 20

88 99 1 10 1 20 131 1 42 1 53 1 64 175

1 30 1 39 1 49 158

45 51 56 62 68 73 79 85 90 96 1 02 108 1 13

45 55 64 74 83 93 1 03 1 12 1 22 1 32 141 151 1 60

88 1 03 118 1 32 1 47 1 62 1 76 191 206 221

61 71 80 89 99 108 117 1 26 1 36 1 45 1 54

98 1 06 1 15 1 24 133 1 42 151

95 1 04 1 13 1 22 131 1 40 1 49

62 72 82 92 1 02 1 12 1 22 1 32 142 1 52 1 62

Impulsive- Hyperactive 0 2 3 4 5 6 7 8 9 10 II 12

34 38 42 46 50 54 58 61 65 69 73 77

35 39 43 47 51 55 60 64

81

85

35

39 43 47 51 56 60 64 68 72 76 81 85

68 72 76 80

35 39 44 49 53 58 63 67

36 41 46 51 56 61 66 71

72

75

76 81 86 90

80 85 90 95

35 38 42 45 48 51 55 58

34 38 42 47 51

64 68 71 74

72 76 81 85

61

55 59

64

68

36 41 45 49 53 58 62 66 70 75 79

83 87

37 41 46 51 55 60 64 69 73 78 82 87 91

39 44 48 53 57 62 66 71 75 80 85 89 94

Anxiety 0 1 2 3 4 5 6 7 8 9 10 11 12

40 45 50 55

39 43 47 51 55 60 64 68 72 76 80 84

41 46 52 57 62 68 73 78 84 89 94

60 65 69 74 79 84 89 94

1 00 1 05

99

88

40 44 49 53 57 61 66 70

40 44 48 53 57

74 78 83 87

61 66 70 74 79 83 87

91

92

41 46 50 54 58 63 67 71 75 79 84 88 92

41 45 49 53 57

41 46 50 55 59

60

83

63 68 72 76 81 85 90

87

94

64 68 72 7S 79

40 45 49 54 59 63 68 73 78 82 87 92 96

40 45 50 55 59 64 69 73 78 83 88 92 97

Note. Norms are in Tscores (M 50, SD 10) . Item numbers fOr lhe five factor scores are as follows: Conduct Problem (2, 8 . 14 , 19, 20, 27, 35. 39); Learning Problem (10, 25, 31, 37); Psychosomatic ( 3 2 , 4 1 , 4 3 , 44): Impulsive-Hyperactive (4. 5 , I I . 1 3 ) : and Anxiety ( 1 2 . 16. 24. 47) . The Hyperactivity Index can be obtained from items 4, 7, II, 13, 14, 2 5 , 31, 3 3 , 37, and 38 and norms on page 888 used. Source: Courtesy C. Keith Conners. =

=

APPENDIX C

888 Table C·64 Norms for Conners Abbreviated Parent Questionnaire

Toral score

Male

Female

Age (il1 -"ears)

Age (ill years)

3-5

6-8

0

32

1 2

35 37

35 37

35 37

3

40 42

39 42

40 42

44 46 48

44 46 49

50 52 55 57

51 53 55 58

59 61 63 65 68 70

60 62

4 5 ·6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

45 47 50 52 55 57 60 62 65 67 70 72 75 77

9- 1 1

65 67 69 71 74 76 78 80 83 85 87 90 92 94

72 74 76 78

80 82

81

85 87

83 85 87 89 92 94 96 98

90 92 95 97 1 00 1 02 1 05 1 07

1 00 =

Note. Norms are in T scores (M 50. SD Source: Courtesy C . Keith Conners.

=

/2-/4

/5- / 7

3-5

6-8

36 38 41

38 40 42

36 38

43

45 47

33 36 39 42

45 47 50 52 54 56 58 61 63 E5

67 70 72 74 76 78 81 83 85 87 90 92

96 99

94 96 98

10 1 1 03

101 1 03

10).

50 52 55 57 60 62 64 67

40 41 43 45 47 49 50 52 54 56 58 59

69 72 74

61 63

77 79 81 84 86 89 91 94

65 66 68 70 72 74 75 77

96 99 10 1 1 03 1 06 1 08

79

86 88

III

90

81 83 84

45 47 50 53 56 59 62 65 67 70 73 76 79 82 85 87 90 93 96 99 1 02 1 05 1 07 1 10 1 13 1 16 1 19

9- 1 1

/2-/4

35 38

36 39

41 44 46 49 52

41 44 47

55 58 61 64 67 70 73 76 79 82 85 88 91 94 96 99 1 02 1 05 1 08 III l id

1 17 1 20 1 23

50 53 56 59 62 65 68 71 74 77 80 83 86 89 91 94 97 1 00 1 03 1 06 1 09 1 12 1 15 1 18 121 1 24

/ 5- / 7 38 41 44 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 96 99 1 02 1 05 1 08 III 1 14 1 17 1 20 1 23 1 26

889

MISCELLANEOUS TABLES Table C-65 Norms for Conners Teacher Rating Scale (39-ltem Form)

Tara I score

4

5

6

Male

Female

Age (in years)

Age (in years)

7

8

9

10

lJ

12

4

5

6

7

8

9

10

JJ

12

42 43 44 46

42 43

44

42

41 43

42 43 44 45

42 43 45

43

41 42 43

Hyperactivity 0 I 2 3 4 5 6 7 8 9 10 II 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78

40 41 42 43 44 45

39 40 41 42 43 43

46 47 48 49

44 45 46 47

50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74

79 80 81 83 84 85

75 76 77 78 79 80 81 82 83

86

84

48 49 50 51 51 52 53 54 55 56 57 58 59 59 60 61 62 63 64 65 66 67 67 68 69 70 71 72 73 74 75 75 76 77 78

39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79

39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55

56 57 58 59 60

60 61 62 63 64 65 66 67 68 69 70 71 72

80 81

73 74 75 76 77 78 79 80 81

82

82

39 40 41 42 43 44 45 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 67 68 69 70 71 72

39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

62 63

64

65 66 67 68 69 70 71 72 73 74 75

40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61

61

62 63 64 65 66 67

38 39 40 41 42 42 43 44 45 46 46 47 48 49 50 51 51 52 53 54 55 55

56 57

58 59 59 60

73 74 75 76 77 78 79 80

78 79 80 81 82 83

80 81 82

61 62 63 63 64 65 66 67 67 68 69 70 71 71 72 73

81

84

83

74

76 77

68 69 70 71 72 73 74 75 76 77 78 79

41 43 44 46 47 48 50 51 52 54 55 56 58 59 60 62

47 48 50 51 52 54 55 56 57 59 60 61

63 64 66 67 69 70

63 64 65 67 68

73 74 75 77

72 73 74 76 77

71

78 79 81 82 83 85 86 87 89 90 91 93 94 96 97 98 1 00 101

69

71

78 80 81 82 84 85 86 88 89 90 92 93 94 95 97 98 99

44

45 46 47 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64

65

66 67 68 69 70 71

72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87

45

46 47 49 50 51 52 54 55 56 57 59 60 61 62 64 65 66 67 69 70 71 72 74 75 76 77 79 80 81 82 84 85 86 87 89 90 91 92 94 95 96 97

44 45 46 47 48 50 51 52 53 54 56 57 58 59 60 62

47 48 49 50 51 52 54 55 56 57

63 64

58 60 61 62 63 64

65 66

66 67

67

68 69 70 72

69 70 71 72 73 75 76 77 78 79 80 82 83 84 85 86 88 89 90 91 92 94

73 74 75 76 77 79 80 81 82 83 85 86 87 88 89 91 92 93 94

46 47 48 50 51 52 53 55 56 57

44

45 46 48 49 50 51

44 45 46 47 48

53 54 55

50 51 52

56 57

53 54

58 60 61 62 63 65 66 67 68

59 60 61 62 64 65 66 67 69

55 56

70 71 72 73 75

71 72 73 75

76 77 78 80 81 82 83 85 86 87 88 90 91 92 93 95 96 97

70

76 77 78 80 81 82 83 85 86 87 88 89 91 92 93 94 96 97

57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86

(Table cOlllinlles nex{ page)

APPENDIX C

890 Table C-65 (cont.)

Total score 45 46 47 48 49 50 51

Male

Female

Age (in years)

Age (in years)

4

5

6

7

8

9

10

11

12

4

5

6

7

8

9

10

11

12

87 88 89 90 91 92

85 86 87 88 89 90 91

79 80 81 82 83 83 84

83 84

83 84 84

82 83 84 85 86 87 88

85 86 87 88 89

84 85 86 87 88 89 90

75 76 76

1 02 1 04 1 05 1 06 1 08 1 09 1 10

101 1 02 1 03 1 05 1 06 1 07 1 09

88 89 90 91 92 93

99 1 00 101 1 02 1 04 1 05 1 06

95 96 97 98 99 101 102

95 97 98

98 1 00 101 1 02 1 03 1 05 1 06

98 99 101 1 02 1 03 1 04 1 05

87 88 89 90 91 92 93

44 45

44 45

43 45

47 49 50 52 54

47 48 50 51

43 45 46 48

43 45 46 48 49

45 48

44 47 49 51 53 55 57 60 62

45 47 49 51 52 54 56 58 60 62

45 47 50 52 54

44 46 47 49

64

66

45 47 49 51 53 55 56 58 60 62

93

85 86 87 87 88

85 86 87 88

90

91

77 78 79 80

94

99 1 00 101 1 02

Conduct Disorder o I 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39

43 45 46 48 49 51 52 54 55 57 59

60 62 63 65 66 68 69 71 72 74 76 77 79 80 82 83 85 86 88 89 91 93 94 96 97 99 1 00 1 02 1 03

56 57 59 61 62

64 66 67 69 71 72 74 76 77 79 81 82 84 86 87 89 91 92 94 96 97 99 101 1 02 1 04 1 06 1 07 1 09

53 54 56 57 59 60 62 63 65 66 68 69 71 72 74 75 77 78 80 81 83 84 86 87 89 90 92 93 95 96 98 99 1 00 1 02

50 51

53 55 56 58 59 61 63 64 66 67 69 71 72 74 76 77 79 80 82 84 85 87 88 90

92 93 95 97 98 1 00 101 1 03 1 05 1 06

43 45 46 48 50 51 53 55 56 58

60 61 63 65 66 68

70

71 73 75 76 78 80 81 83 85 86 88 90 91 93 95 96 98 1 00 101 1 03 1 05 1 06 1 08

46 48 49 51

52 54 55 57 58 60 61 63 65 66

68

69 71 72 74 75 77 78 80 81 83 84

86 88 89 91 92 94 95 97 98 1 00 101 1 03

51 53 54 56 57 59 60 62 64 65 67

68

70 71 73 75 76 78 79 81 82 84 86 87 89 90 92 93 95 97 98 1 00 101 1 03 1 04

43 45 46 48 49 51 52 54 55 57 58

60

42 43

44

45 47 48 49 50 51 52 53 55

44

65

66 69 70

61 63

64

77 79 81 83 85

72 73 74

87 89 90 92 94 96 98 100 1 02 1 04 1 06

67

72 73 75 76 77 79 80 82 83 85 86 88 89 91 92 94 95 97 98 1 00 101

60

65 66 67 68 70 71

75 76 78 79 80 81 82 83 85 86 87

56 59 61 63 66 68 70 72 75 77 79 82

56 58 60 62 64 67 69

64

52 54

52 54

56 57 58 59

61 63

45 47 49

46 48 50

71 73 75

1 08 1 10 ! 12 1 14 1 16 1 17 1 19

84 86 89 91 93 96 98 1 00 1 02 1 05 1 07 1 09 1 1 1 1

12 14 16 19

121 1 23 1 26 128 1 30 1 32 1 35

45 47 49 51 53 55 57 59 61 63 65 67 69 71 73 75 77 79 81 83 85 87 89 91 93

50 52 54 57 59 61

64 66 68 70 73

1 05 1 07 1 09

75 77 79 82 84 86 89 91 93 95 98 1 00 1 02 1 05 1 07 1 09 1II 1 14 1 16 1 18

III 1 13 1 15 1 17 1 19 121 1 23

121 123 1 25 1 27 1 30 1 32 1 34

95 97 99 101 1 03

64 66 68 71 73 75 77 79 81 84 86 88 90 92 95 97 99

101 103 1 05 108 1 10 1 12 1 14 1 16 1 19 121 1 23 1 25 127 129

66 68 70 72 74 76 78 80 82 83 85 87 89 91 93 95 97 99 101 1 03 1 05 1 07 1 09 111 1 13 1 14

1 16 1 18 1 20

56 58 60 62

64 69 71 73 75 77 79 81 83 85 88 90 92 94 96 98 1 00 1 02 1 04 1 07 1 09 111 1 13 1 15 I 17 1 19 121 1 23 1 26 1 28

64 66 68 69 71 73 75 77 79 81 83 84

50 51 53 54 55 57 58 59 61 62 63 65 66 68 69 70 72 73

86 88 90 92 94 96 97 99 101 1 03 1 05

74 76

1 07 1 09 1 10 1 12 1 14 1 16 1 18

89 91 92 93 95

77 78 80 81 82 84 85 86 88

96 97

(Table coil/iI/lies I/exr page)

MISCELLANEOUS TABLES

89 1

Table C-65 (cont.)

Male

Female

Age (in years)

Age (in years)

------------------------------------- -----------------------------------

Total score

4

5

6

7

8

9

10

11

12

4

5

6

7

8

9

10

11

12

43 45 48 51

43 46 49 53 56 59

43 46 48 51 54

43 45 48 51 53 56

43 46

43 46

42

49 52

50 53 56 59 62 65

46 49 51 53

Emotional Overindulgent 0 1 2 3 4 5 6 7 8 9

41 44 47 49 52 54 57 59 62 65

55 57 60 62

43 45

42 45

47 50 52 55

48 50

47 50

53 55

53 55

58 60 63

58 61 63 66 68 71

58

65

98 101 1 03

98 1 00

10 1 1 04

67

13 14 15 16

75 77

20 21 22 23 24

50 52

42 45

64 67 69 72 74 76 79 81 84 86 88 91 93 96

10 11 12

17 18 19

43 45 48

70 72

RO 83 85 88 90 93 96

68 70 73 76 78 81 83 86 88 91 94 96 99

73 76

60 63 66 68 71 74 76 79 81 84

78 81 83 86 88 91 93 96 98 101

1 00 1 02

1 03

1 05

87 89 92 94 97

43 45 48

42 45

50 52 55

47 50 52 55

57 59 61

57 60 62

64 66

65 67

68

70 72 75

71 73 75 78 80 82 85 87 89 92 94 96 98

43 45 48

41 43

50 53 55 58

47 49 52 54 56 58

60 63 65 68

98

70 73 75 78 80 83 85 87 90 92 95 97

1 00 1 03

1 00 1 02

77 80 82 85 88 90 93 95

42 45 47 49 52 54 56 59 61

45

64 66 68

60 62 64 66 68

71 73 76 78 80 83 85 87 90 92

70 72 74 76 78 81 83 85

43 45 48 50 53 55 58 60 63 65 68 70 73 75 78 80 83 85 88 91

57 59 62 65 68 71 73 76 79 82 84 87 90 93 96

62 65 68 71 74 77 80 83 86 89 92 95 98

101

10 1 1 04 1 07

101 1 04 1 07 1 10 1 14

1 03

1 09

1 17

38

40

41 44 47 49 52 55

43 46 49 51 54

93 96

87 89

95

98

97

91

99

-L

54

98

57 60 63 66 69 72 75 77 80 83 86 89 92 95 98 101 1 04 1 06

58 61 64 66 69 72 74 77 79 82 85 87 90 93 95 98

1 09 1 12

1 00 1 03 1 06

39 42 45

38 41 43

48 51 54

46 49 52

57 60 63 66

57 60 63 66 69

57 60

55 58

63 66 69

72 75 78

60 63 66 69 72 75 77

55 59 62 65 68 71 74 77 80 83 86 89 92 95 98

68 71 74 77 80 83 86 90 93 96 99

1 16

1 02 1 05 1 08 III 1 14 1 17

37 40 43 46 49 52 55

39 42 46 49 52 55 58

58 61

61

101 1 04 1 07 1 10 1 13

44

55 57 59 61 63 65 67 69 71 73 76 78 80 82 84 86 88 90 92

Anxious - Passive 0 I

39 42

39 42 45 48

38 41 43

51 54 56

49 52 54 57 60 62 65 68

2 3 4 5 6

45 48

7 8 9

59 62 65

10 II 12

67

59 62 65 68

70 73

71 74

13 14 15

76 79 81 84

16 17 18

51 53 56

87 90

77 80 83 85 88 91

46

70 73 76 79 81 84 87

39 42 45 48 5I 54

39 42 45

38 41

48 51 54

47

57 60 63 66 69 72

57 61 64 67

44 50 53 56

50 54

59 62

60 63 66

65 68

81 84

70 73 76 79 82 85

87

88

85

90 93

91 94

88 91

75 78

37 41 44 47

70 73 76 79 82

57

70 73 76 79 82 86 89 92 95

39 42 45 49 52 55 59 62 65 69 72 75 79 82 85 88 92 95 98

37 40 43 46 49 53 56 59 62 65 68 71 74 77 80 83 86 89 92

38 41 44 47 49 52 55 57 60 63 66 68 71 74

39 42 45 47 50 53 55 58 61 64 66 69 72 74

68 71 74

76 79 82

77 80 82

77 79 82

81 84

72 75 78 81 84

87

87

85 87

85 88

85 87

90 93

90 93

80 83 86 89

64 67 71 74

64 67 70 73 76

77 80 83 86

79 82 85 88

89 92

91 94

38 40 43 45 48 50 53 55 58 61 63 66 68 71 73 76 78 81 84

(Table COl1fillUeS lIex{ page)

892

APPENDI X C

Table C-65 (cont.)

Total score

Male

Female

Age (in years)

Age (in years)

4

5

6

7

8

9

10

1J

12

43 47 50

44 49 53 58

44 48 53 57

44 49

43 47 51 55 59

42 45 49 52

61 65 69 73

43 48 52 56 60

44 48

62 67 71 75 80 84

43 47 51 55 59

63

89 93

86 90

87 92 97

63 67 71 75 79

94 98 1 02 1 06

1 02 1 06 III 1 16

5

6

7

44

44

48 52 56 60

50 55 60

45 50

45 50

54 59 63 68

55 60

4

8

9

10

11

44

44 49 53 58 62

44 49 53 57 62

45 49 53 58 62

67 72 76 81

66 71

12

Asocial 0 I

2 3 4 5 6 7 8 9 10

54 58 62 65 69 73 77 81

II 12

84 88

98

13 14 15

92 96 99

1 02 1 07 III

77 81

54 59 63 68 73 78 82

63 67 70

64

68

86

72 76 80 84 89

90 94 98 10 1

93 97 101 1 05

74 78 82

67 72 76 80 84 88 92 96 1 00 1 04

51 55 59

83 87 91 95 99 1 03

55 59 62 65 68 72 75

64

68 72

65 70 75 81

78 82

93

86 91 96 101 1 06

85 88 91

97 101 1 05

III 1 17 1 22

77 81 85 89

72 77 82 86 91 95 1 00 1 05 1 09 1 14

66 71 76 81 86 91 97 1 02

48 52 56 59 63 67 71 75 78 82

98

66 71 75 79 84 88 92 97

1 02 1 06 III

10 1 1 05 1 09

42 48 55 61 68 74

42 48

43 48

54

54 60 65 71

80 87 93 99

77 83 89 95 101

86 90 94 97

86 90 95 99 1 04 1 09

101

1 13

43 49 55 61 67 73

42 47 53 58

79 35

75 80 86

1 07 1 12 1 17 1 22

75 80 84 89 93

38 43 49 54 60 65 71 76 82 87 93 98 1 04 1 09 1 15 1 20

Daydream - Attendance Problem 0 I 2 3 4 5 6 7 8 9 10 11 12

41 47 53 58

41 47 52 58

70

63 69

64

76 82 88 94

74 80 85 91

1 00 1 05 III

1 02 1 07

96

40 45 49 54 59 63 68 72 77 81 86 91 95

41 46

41 46

51 56 61 66 71 76 81

51 56

60 65

87 92 97

70 75 80 85 90 95

102

1 00

40 45 50 55 60 64

69 74 79 84 89 93 98

41 46 51 56 61 66

41

39

46 51 56

43 47

61 66

71

71

76 81 86 91

75 80 85 90

96 101

95 1 00

51 56 60

64

68 72

76 80 85 89

43 48 54

43 49 54

60

60

66 72

66 72

78 84 90 96 1 02

77 83 89 94

1 08 1 14

1 00 1 06 1 12

43 48 52 57 62 67 72 77 82 87 92 97 1 02

91 97 1 02 1 08 1 14

64

69

91 97 1 02 108

1 06 1 12 1 19

60 66 71

1 06 1 12

77 83 88 94 1 00 1 05 III

41 45 49 53 58 62 66 70 74 78 82 86 90

= 50, SD = 10) Item numbers for the six factor scores are as follows: H y peractivity ( 1 . 2 , 3 , 4 , 5 , 6, 7 , 8 , 1 1 , 14, 1 5 , 1 7 , 24, 2 9 , 3 2 , 3 5 , 3 8 ) ; Conduct D isorder ( 1 5 , 1 6 , 1 7 , 1 8 , 19, 20, 2 1 , 2 5 , 29, 3 1 . 3 2 , 3 6 , 3 8 ) ; Emotional Overindulgent (3, 9 , 10, 1 2 , 1 3 , 16, 2 1 , 36); AnxiousPassive (24. 26. 30. 3 3 . 34. 3 7 ) : Asocial (22, 23, 2 5 . 27, 28): and Daydream-Attendance Problem (8. 1 1 , 22, 39) . The Hyperactivity Index can be obtained from items I, 3, 5 , 6, 7 , 8 , 1 3 , 14, 16, and 21 and norms on page 893 used.

Note. Norms are in T scores (M

Source: Courtesy C. Keith Conners.

893

MISCELLANEOUS TABLES Table C·66 Norms for Conners Abbreviated Teacher Questionnaire

Total score 0 1 2 3

4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Male

Female

Age (in years)

Age (in years)

3-5

6-8

9- l /

12- 14

15- 1 7

3-5

6-8

42 43

40 42

44

44

40 41 43

45 46 47 48 49 50 51 52 53 54 55 56 57 58 59

45 47 49 50 52 54 55 57 59 60 62 63 65 67 68 70 72 73 75 77 78 80 81 )3 85 86 88 90

40 42 44 47 49 51 54 56 58 61 63 65 68 70 72 75 77 79 82 84 86 89 91 93 96 98 1 00 1 03 1 05 1 07 1 10

41 43 45 48 50 52 54 56 59 61 63 65 68 70 72 74 76 79 81 83 85 88 90 92 94 96 99 101 1 03 1 05 1 08

39 40 42 43 45 46 48 49 51 52 54 55 57 58 60 61 63

42 44 46 49 51 53 55 58 60 62

60 61

62 63 64

66 67 68 69 70 71 72 73 =

Note. Norms are in T scores (M 50, SD Source: Courtesy C. Keith Conners.

44

46 47 49 50 52 54 55 57 58 60 61 63 64

66 67

69 70 72 74 75 77 78 80 81 83 84 86 =

10).

II

64

66 67 69 70 72 73 75 76 78 79 81 82 84

64

66 69 71 73 75 78 80 82 84 86 89 91 93 95 98 1 00 1 02 1 04 1 06 1 09

9-l / 42 44 46 43 50 53 55 57 59 61 63 65 67 69 71 73 75 78 80 82 . 84 86 88 90 92 94 96 98 1 00 1 03 1 05

12-14

1 5- / 7

43 47 51 55 59 63 68 72 76 79 84 88 93 97 101 1 05 1 09 1 13 1 17 1 22 1 26 1 30 1 34 1 38 143 1 47 151 1 55 1 59 1 63 1 68

44 46 47 49 51 52 54 55 57 59 60 62 64

65 67 68 70 72 73 75 76 79 80 81 83 85 86 88 89 90 93

APPENDIX C

894 Table C-67 Defi nitions of Categories in the Structure of Intellect Operations

C M D

N E

Major kinds of intellectual activIties or process,es ; things that the organism does with the raw materials of information, information being defined as "that which the organism discriminates , " Cognition. Immediate discovery, awareness, rediscovery, or recognition of information in various forms ; comprehension or understanding . Memory. Retention or storage , with some degree of availability, of information in the same form it was committed to storage and in response to the same cues in connection with which it was learned . Divergent Production. Generation of information from given information, where the emphasis is on variety and quantity of output from the same source. Likely to involve what has been called transfer. This operation is most clearly involved in aptitudes of creative potential . coNvergent Production. Generation of information from given information, where the emphasis is on achieving unique or conventionally accepted best outcomes. It is likely that the given (cue) information fully determines the response . Evaluation. Reaching decisions or making judgments concerning criterion satisfaction (correctness, suitability, adequacy, desirabil ity, etc . ) of information. Contents

F

S M B

Broad classes or types of information discriminable by the organism. Figural. Information in concrete form, as perceived or as recalled; possibly in the form of images. The term "figural" minimally implies figure-ground perceptual organization . Visual spatial information is figural . Different sense modalities may be involved; e . g . , visual kinesthetic. Symbolic. Information in the form of denotative signs, having no significance in and of themselves, such as letters, numbers, musical notations, codes, and words, when meanings and form are not considered . seMantic. I nformation in the form of meanings to which words commonly become attached, hence most notable in verbal communication but not identical with words. Meaningful pictures also often convey semantic informatio n . Behavioral. Information, essentially non-verbal , involved in human interactions where the attitudes, needs, desires, moods, intentions, perceptions, thoughts, etc . , of other people and of ourselves are involved. Products

U C R S T I

The organization that information takes in the organism's processing of it . Units. Relatively segregated or circumscribed itellls of information having "thing" character. May be close to Gestalt psychology's "figure on a ground . " Classes, Conceptions underlying sets of items of information grouped by virtue of their common properties , Relations. Connections between items of information based on variables or points of contact that apply to them. Relational connections are more meaningful and definable than implications. Systems. Organized or structured aggregates of items of information; complexes of interrelated or interacting parts . Transformations. Changes of various kinds (redefinition, shifts, or modification) of existing information or in its function. Implications. E xtrapolations of information , in the form of expectancies, predictions, known or suspected antecedents, concomitants, or consequences. The connection between the given information and that extrapolated is more general and less definable than a relational connection,

Source: Reprinted by permission of the publisher and author from M. N. Meeker, The Stmcture of Intellect, pp. 1 95- 1 96 , Copyright 1 969, Charles E. Merri l l .

895

MISCELLANEOUS TABLES Table C-68 Percenti le Ranks for Standard Scores with a Mean of SO and Standard Deviation of

Stalldard score 95 94 93 92 91 90 89 88 87 86 85 84 83 82 81 80 79 78

77 76 75 74

73

SD 10

15

99.99 9 9 . 99 99 . 99 99. 99 99 . 99 99. 99 99 . 99 99 . 99 99 . 99 99.98 99.98 99.97 99.95 99 . 9 3 99.90 99 . 8 7 99 . 8 1 99 . 74 99 . 65 99 . 5 3 99 99 99

99 . 87 99 . 83 99 .79 99 . 74 99. 69 99 . 62 99 . 5 3 99 99 99 99 99 99 98 98 98 97 97 96 96 95 95 94

Standard score 72 7i 70 69 68 67 66 65 64

63 62 61 60 59 58 57 56 55 54 53 52 51 50

SD 10

15

Standard score

99 98 98 97 96 96 95 93 92 90 88 86 84

93 92 91 90 88 87 86 84 82 81 79 77 75 73 70 68 66 63 61 58 55 53 50

49 48 47 46 45 44 43 42 41 40 39 38 37 36 35 34 33 32 31 30 29 28 27

82

79 76 73 69 66 62 58 54 50

SD

10

or I S

10

15

Standard score

46

47 45 42 39 37 34 32 30 27 25 23 21 19 18 16

26 25 24 23 22 21 20 19 18 17 i6 15 14 13 12

42

38 34 31

27

24 21 18 16 14 12

10 8 7 5 4 4

3 2 2 I

1

14

13 12 10 9 8 7 6

11

10 9 8 7 6 5

SD 10 I

1 .47 .35 .26 . 19 .13 . 10 .07 . 05 . 03 .02 . 02 .01 .01 -

15 5 5 4 4 3 3 2 2 2 1 1 1 1 1 1 .47 .38 .31 .26 .21

. 17 . 13

_ APPEN DIX D M O D I F I E D I NSTRUCTI O NS F O R ADM I N ISTE RI N G TH E WISC-R A N D WISC-I I I P E RFORMAN CE SCALE SU BTESTS TO D EAF C H I LD RE N

PANTO M I M E I NSTRUCT I O N S

Present age-appropriate initial item from WISC-R Picture Arrangement subtest . Point to series in general sweeping motion, and then point to child . I f child does not respond or does not arrange the cards in the correct order for Item 1 (FIGHT) and/or Item 2 (PICNIC ) , arrange the cards in the correct sequence . Then point to each card, designating first, second, and third . Allow child to look at correct sequence for about 10 seconds, and then put the cards i n their original numerical order. Point to series in general sweeping motion, and then point to child. If child fail s to arrange Item 3 (FIRE) and/or Item 4 (PLANK) in the correct sequence, put the cards in their original numerical order. Take Card "F" or "W" from the array and place it below the other three cards. Point to it and hold up one finger. Then point to remaining cards in general sweeping motion , and point to child. In Items 5 to 12, arrange cards i n numerical sequence, point to cards in general sweeping motion, and point to child. Follow instructions i n WISC-R manual regarding timing and discontinuation procedure s .

Picture Completion Show child Card 2 from Wechsler Intelligence Scale for Children (WISC) (Wechsler, 1949 ) . This item depicts a table with three legs. Count legs by pointing to each with index finger. Hold up three fingers, then point to missing leg by tracing its designated outline with index finger. Hold up four fingers. Summarize directions by counting each leg with index finger, holding up three fingers, pointing to missing area, and holding up four fingers. Present age-appropriate initial item from WISC-R Picture Completion booklet. Point to child, point to picture (not to missing part ) , and point to child again. Follow instructions in WISC-R manual regarding failure and discontinuation procedures. If child fails to indicate missing part on Cards 1 and 2 within 20 seconds, point to missing part .

Picture Arrangement

Block Design

Place sample item (SCALE) in front of child in numerical order indicated in WISC-R manual . Point to pictures in general . Ar­ range pictures in correct order. Point to Picture A and hold up one finger, point to Picture B and hold up two fingers, and point to Picture C and hold up three fingers, thereby designating first, second, and third . Rearrange pictures i n original administration order. Point to series in general sweeping motion, and then point to child.

Design 1. Place four blocks in front of child and turn each block to show the different sides. Point to each different side of each block during this demonstration. Arrange the four blocks into Design 1. Give child four other blocks. Point to child, point to child's blocks, and point to model . If child fails, assemble child's blocks to match modeled design . Point to child's blocks and point to model . Scramble child's blocks. Point to child, point to child's blocks, and point to model . Design 2 . Assemble Design 2 behind a screen . Present model to child in completed form . Point to child, point to child's blocks, and point to model . If child fails on first trial , follow procedure given for Design I .

Reprinted, with changes in notation, with permission of the author from P M . Sull ivan, A Comparison of Administration Modifications on the

W/SC-R Performance Scale lVith Different Categories of Deaf Children, unpublished doctoral dissertation, University of Iowa, 1978 .

896

MURPH Y -NEUHAUS-REED-SULLIVAN INSTRUCTIONS

Design 3. Display card depicting Design 3 . Assemble blocks into Design 3 in full view of child . Point to card and point to blocks. Scramble the blocks . Point to child, point to ch ild's blocks, and point to card . If child fails on first tria l , repeat above administration procedure . If child is 8 or older and subtest begins with this item, fol low procedure for designating color patterns and similarities of blocks given at the beginning of section on Design 1 . Designs 4 to ll. Display card showing each respective design. Point to child_ point to blocks, and point to card . Follow instructions in WISC-R manual regarding timing and d iscontinuation procedure s .

Object Assembly Arrange the pieces of sample item (APPLE) behind shield. Expose array and assemble pieces together. Rearrange pieces in original presentation sequence. Point to child and poini to array. If child does not respond , repeat above procedure. Correct any errors. Items 1 to 4. Arrange pieces behind shiel d . Expose array, point to child , and point to pieces. Correct errors on item I only. Follow instructions in WISC-R manual for timing and scoring.

Coding Coding A. Point to star. circle, triangle, cross . and box i n general sweeping motion. Point to each mark in each geometric form individually. Point to blank sample items and fill in the first circle. G ive child pencil, point to child, and point to remaining sample items. Stop child after completion of last sample item. Point to subtest items in general sweeping motions for each row. Point to child and point to first item . If child discontinues work after completing fi rst row, point to next row. Follow timing instructions in the WISC-R manual . Coding B. Point to numbered boxes in general sweeping mo­ tion. Then point to each number and its respective symbol indi­ vidually in the entire array. Point to blank sample items and fill in the first one. Give child pencil, point to child, and point to remaining sample items. Stop child after completion of last sample item. Point to subtest items in general sweeping motion for each row. Point to child and point to fi rst item. If child discontinues work after completing first row, point to next row. Follow timing instructions in WISC-R manual .

Mazes Give child booklet . Point to sample item. Point to figure in center of maze and to opening that leads to exit. Demonstrate sample item. On reaching opening to outside of center, pause, and, without l ifting the pencil, point to bl ind alley. Then point to correct route and fi nish the tracing . Items 1 10 9. Point to center of maze, to exit opening, and to child . Follow discontinuation procedure in WISC-R manual . If

897 child does not beg in work in center of maze, stop work, place pencil in center of maze, and point for child to continue. Follow timing and scoring instructions in WISC-R manual .

M U R P H Y- N E U H A U S-R E E D-SU LLIVAN I NSTRUCTIONS

Picture Completion This modification is based on Neuhaus (1967 ) . Materials. Three pictures are drawn on 3 '12" x 3 112 " white cardboard squares. Each item shows a picture with a missing detail on one side accompanied by the same picture with the missing detail filled in on the other side. The three constructed sample items include a picture of an arrow with the tip missing on one side and drawn in on the other, a picture of an elephant with the trunk missing on one side and drawn in on the other, and a picture of a doll with an arm missing on one side and drawn in on the other side. Procedure. Present Sample Item I (ARROW ) . Show side with missing detail fi rst, turn picture over and show side with com­ pleted detail . Point to completed deta i l . Repeat procedure above. Present Sample Item 2 (ELEPHANT) and Sample Item 3 (DOLL) in the manner given above. Present age-appropriate initial i.tem from WISeR .Picture Completion booklet. Follow instructions in WISC-R manual regarding failure and discontinuance procedures. If child fails to indicate missing part on Cards I and 2 within 20 seconds, point to missing detail .

Picture Arrangement This modification is based on Reed ( 1970) . Materials. One set of three 3 '12" x 3 '12" white cardboard square cards with the numeral 1 printed on one card, the numeral 2 on one card, and the numeral 3 on one card. One set of three 3 '12" x 3 '12" white cardboard square cards with the letter A printed on one card , the letter B on one card, and the letter C on one card. Use the sample item (SCALE) from the WISC-R Picture Arrangement subtest. Procedure. Present Sample Item I (NUMBERS) in order 2, 3 , I from child's left to right. Arrange cards i n correct numerical sequence. Allow child to view for 10 seconds. Rearrange cards i n original administration order. Motion child t o arrange cards by pointing to them in general sweeping motion. If child does not respond or arranges cards incorrectly, arrange cards in their correct sequence. Then rearrange cards i n original administra­ tion order and motion child to arrange cards. Follow same

898

APPENDIX D

A D M I N ISTERING T H E WISC-R P E R F O R M A N C E SCALE SU BTESTS TO DEAF C H I L D IR E N

procedure for Sample Item 2 ( LETTERS) and sample item (SCALE) from WISC-R Picture Arrangement subtest. !rems 1 10 4. Present age-appropriate initial item from WISC-R Picture Arrangement subtest. If child does not respond or does not arrange cards i n the correct order for Item I ( FIG HT) and/or Item 2 (PICNIC ) . arrange the cards in the correct sequence. Al low child to look at correct sequence for about 10 seconds and then rearrange cards in their original numerical order. Motion to child to rearrange cards. If child fails to arrange Item 3 ( FIRE) and/or Item 4 ( PLANK) in correct sequence, put cards in their original numerical order. Take Card "F" or "W" from the array and place it below the other three cards. Motion for child to complete arrangement by pointing to remaining cards in general sweeping motion . !rems 5 10 12. A rrange cards in numerical sequence and motion for child to complete arrangement by pointing to sequence in general sweeping motion. Follow instructions i n WISC-R manual regarding discontinu­ ance and timing procedures.

Object Assembly This modification is based on Neuhaus ( 1967 ) . Materials. Three pictures drawn o n 3 '12" x 3 '12" white card­ board squares. These pictures consist of an apple, a girl , a nd a horse. Procedure. Arrange the pieces of sample item (APPLE) be­ hind shield. Expose array and place Picture I ( A PPLE) beside it. Point to picture . Assemble pieces together. !rems 1 Gild 2. Arrange pieces behind shield. Expose array and place picture of GIRL or HORSE beside it . Point to picture. Motion child to assemble pieces together by pointing to pieces in general sweeping motion . Correct errors on !rem 1 ollly. T ime and score as in WISC-R manual . Ilems 3 Gild 4. A rrange pieces behind shield. Expose array. Motion child to assemble pieces together by pointing to pieces i n general sweeping motion. Time a n d score a s shown i n WISC-R manual .

Coding Block Design This modification is based on Murphy (1957). Materials. One white strip of cardboard 4" x I" on which the instructions "Make one l ike this" are printed . Procedure. Design 1. Set out four blocks in single l i ne with red color on top . Point to each block in sequence. Repeat this pro­ cedure with white and red/white sides al igned in top position of sequence. Arrange the four blocks into Design 1 , taking care to use both hands in the process. Give child four other blocks. Give child written instructions. Motion for child to make pattern by pointing to model and to child's blocks. If child fai l s , assemble child's blocks to match modeled design. Scramble child's blocks and point to instructions, to model , and to child's blocks. Design 2. Assemble Design 2 behind a screen . Present model to child in completed form. Give child written instructions. Motion for child to complete pattern by pointing to model and to child's blocks . If child fails on Design 2 , fol low procedure given for Design I . Design 3 . Display card depicting Design 3 i n full view of child, taking care to use both hands i n the process. Scramble the blocks. Give child written instructions. Motion for child to construct pattern by pointing to card and to child's blocks. If child fails on first trial , repeat above administration procedure. If child i s 8 or older and subtest begins with this item , follow procedure for designating color patterns and similarities of blocks g i ven at the beginning of section on Design I . Designs 4 10 11. D isplay card showing each respective design . Point to written instructions. Motion for child t o construct pattern by pointing to card and to child's blocks. Follow timing, scoring, and discontinuance instructions in WISC-R manual .

This modification is based on Murphy (1957) . Materials. Coding protocol from Wechsler Intelligence Scale for Children ( WISC) (Wechsler, 1949) . One white strip of card­ board 4" x 1" on which the instructions "Do this quickly" are printed . Procedure. The same procedure will be fol lowed for Coding A and B . S it next to child and fi l l out sample items on WISC Coding protocol . Point to each geometric form or number and its respec­ tive symbol with pencil before fi l l i ng in the first three sample items. Work quickly on remaining four sample items and do not point to components of cod e . Give child WISC-R Coding pro­ tocol and point to each geometric form or number and its respec­ tive symbol with pencil . G ive child pencil and point to sample items. When child has completed sample items , point to subtest items with pencil i n general sweeping motions for each row. Place written instructions beside protocol and point to them . If child ceases work after completion of Row I, redi rect attention to Row 2 by pointing to first item in row. Time and score as i n WISC-R manual.

Mazes This modi fication is based on Sul l ivan ( 1978) . Materials. Mazes protocol from Wechsler Intelligence Scale for Children (WISC) (Wechsler, 1949) . One white strip of card­ board 4" x I" on which the instructions "Start in the middle. Find your way out ." are written . Procedure. S it next t o child and point t o sample maze on WISC protocol . Place written d i rections beside protocol and

899

MURPHY -NEUHAUS-REED-SU LLIVAN INSTRUCTIONS p o i n t to the lll . Delllollstrate sample it ; ;:: ::;;; i�ma. American Psychologisl, 4 1 . 1 4-24 . Matarazzo, 1 . D . , Bornstein, R. A . , McDermott. P A , & I\oonan, 1 . V ( 1 986) . Verbal IQ vs. Performance IQ difference scores in males and females from the WAIS-R standardization sample . Journal of Clinical Psych ology. 42, 965-974. Matarazzo, 1. D . , & Herman, D . O. ( 1 984a) . Base rate data for the WAIS-R: Test-retest stabil ity and VIQ-PIQ differences. Journal of Clinical Neuropsychology. 6, 35 1 -366. Matarazzo, 1 . D . , & Herman. D . O. ( 1 984b) . Relationship of cducation and IQ in the standardization samplc . Journal of Consulling alld Clinical Psychologv, 52, 63 i -634 . Matarazzo, 1 . D . , & Herman, D. O. ( 1 985 ) . Cl inical uses of the WAIS-R: Base rates of differences between V IQ and PIQ in the WAIS-R standardization sample . In B. B. Wolman (Ed . ) . Handbook of intelligence: 77Jeories, measuremenrs and ap­ plications (pp. 899-932) . New York: Wiley. Matheny, A . P , 1r. ( 1 983). A longitudinal twin study of stability of components from Bayley's Infant Behavior Record . Child De velopmenr , 54, 356-360. Matheny. A . P , 1r. , Dolan. A. B . . & Wilson, R. S. ( 1 974) Bayley's Infant Behavior Record : Relations between behaviors and mental test score s . Developmenral Psychology. 10, 696-702 . Mather, N . , & Burch, M . ( 1 986) . An examination of Woodcock­ Johnson suppressor effects with learni ng-disabled and gifted subjects . Journal ofPsychoeducational Assessmelll. 4, 45-5 1 . Mattes, L . J . , & Omark, D . R . ( 1 984) . Speech and language assessmenr for the bilingual handicapp·ed. San Diego : Col­ lege-Hili Press. Maxwe l U . K . , & Wise, F. ( 1 984). PPVT IQ validity in adults: A measure of vocabulary, nO] of i ntf!l l igcncr , JUI/mal a/Cliniral Psychology, 40, 1 048- 1 05 3 . McCa l l , R . B . ( : 977). Childhvod IQ's a s pcedictol s v f aduli educational and occupational status . Science, 1 97, 482-483 . McCall , R . B. ( 1 979) . The development of intellectual function­ ing in infancy and the pred iction of later I . Q . In 1 . D . Osofsky ( Ed . ) , Handbook of infaIII development ( pp. 707-74 1 ) . New York: Wiley. McCa l l , R . B . , Appelbaum, M . I . , & Hogarty, P S . ( 1 973). Developmental changes in mental performance. Monographs of the Societyfor Research in Child Developmenr, 38(3) (Serial No. 1 50), 1 -83 . McCal lum, R. S . , Karnes, F. A . . & Edwards, R . P. ( 1 984). The test of choice for assessment of gifted childre n : A comparison of the K-ABC , WISC-R, and Stanford-Binet. Journal of Psy­ choeducational Assessment, 2, 57-63 . McCarthy, D . A . ( 1 972). Manual for the McCarthy Scales of Children 's Abilities . San Antonio: The Psychological Corpora­ tion. McCarthy, D. A. ( 1 978). McCarthv Screening Test (MST) . San Antonio: The Psychological Corporation . McConne l l , R. E . ( 1 930). The origin of mental tests . Education , 50, 464-473. McGee, M. G. ( 1 982) . Spatial abilities: The influence of genetic

955 factors . I n M . Pot,teg:al ( Ed . ) , Spalial abilities: Development and physiologj,caijfolllndations ( pp. 1 99-222 ) . New York: Aca­ demic Pre;s. M(c:Grew, K . S. ( 1 98;3) . Comparison of the WISC-R and Wood­ cock-10hnson rtest§ of cognitive abil ity. Journal of School Psy­ chology, 2 1 . 2 7 1 --276. M Grew, K . S. ( 1 9185) . I nvestigation of the VerballNonverbal structure of th,e Woodcock-Johnson: Implications for subtest interpretation and comparisons with the Wechsler Scales. Journal of PSI cho,educational Assessment, 3, 65-7 1 . McGrew, K . S . ( 1 98,6 ) . Clinical interpretation of the Woodcock­ Johnson Tem of Cognitive Ability. Orlando, F L : Grune & Stratton. McGue , M . . Shinn, M . , & Ysseldyke, 1 . ( 1 982). Use of cluster scores on the Woodcock-Johnson Psycho-Educational Battery with learning dis.abled students. Learning Disability Quar­ terly. 5 . 274-:�87. McGuire. T. R , & H irsch , 1 . ( 1 97 7 ) . General intell igence (g) and heritability (H2, h2). I n I. C. Uzgi ris & F Weizmann (Eds . ) , The struct �ring of experience (pp. 25-72) . New York : Plenum Press. McLaughlin. B. ( 1 977). Second-language learning in childre n . Psychological Bul/etin , 84, 438-459 . McLoughlin, C. S . . & Gullo, D . F ( 1 984). Comparison of three formal methods of preschool language assessment. wnguage, Speech . & Hearing Services in Schools, 1 5 , 1 46- 1 5 3 . McMahon , R . C . ( 1 98 1 ) . Biological factors in childhood hyper­ kinesis: A review of genetic and biochemical. hypotheses . . . . ' Journal 0/ Clinical Psychology, 3 7, 1 2-2 1 . McMahon . R. C . , Kunce, J . T. ( 1 98 1 ) . A comparison of the factor structure of the WISC and WISC-R in normal and xC'pliOnf J I grollps. Journal of Clinical Psychology, 3 7, 408 - 4 1 0 . McNamara, J . R . , Porterfield , C . L . , &. Miller, L . E . ( i 969) . The relationship of the Wechsler Preschool and Primary Scale of Intel! igence with the Coloured Progressive Matrices ( 1 956) and the Bender Gestalt Test. Journal of Clinical Psychology, 25, 65-68 . McNemar, Q . ( 1 942) . The revision of the Stanford-Binet Scale. Boston : Houghton M i ffl i n . McNemar, Q . ( 1 974) . Correction t o a correction. Journal of Consulting and Clinical Psychology, 42, 1 45 - 1 46. McShane, D . A , & Plas, 1 . M . ( 1 984) . The cognitive function­ ing of American I ndian children: Moving from the WISC to the WISC-R. School Psychology Review, 13, 6 1 -73 . Meacham, F R . ( 1 984) . A comparative study of the WISC-R and WAIS-R performance IQ scores of 1 6-year-old hearing im­ paired students in a residential program. Unpublished doc.­ IOral dissertation, University of Alabama, Tuscaloosa, A L . Meadow, K . P. ( 1 968) . Early manual communication in relation to the deaf child's intellectual, social , and communicative functioning . American Annals of the Deaf, 1 1 3, 29-4 1 . Meadow, K . P ( 1 97 5 ) . The development of deaf children. I n E . M . Hetherington ( Ed . ) , Review of child development research (Vol . 5 , pp . 44 1 -508 ) . C hicago: University of Chicago Press.

956

Meeke r, M. N. ( 1 969) . The STrUCTure ofintellecT. Columbus .. OH : Charles E. Merril l . Melear, J . , & Boyle, J . ( 1 974) . The relationship of the Coillmbia Mental Maturity Scale to the WISC with reference to low achieving Anglo and Spanish bilingual children. Colorado Journal of EducaTional Research , 1 4 , 8- 1 0 . Mercer, 1 . R . ( 1 976) . Pluralistic diagnosis in the evaluati()n of Black and Chicano childre n : A procedure for taking so­ ciocul tu ral variables into account in clinical assessment. In C . A . Hernandez, M . 1 . Haug , & N . N . Wagner (Eds . ) , Chi­ canos: Social alld psychological perspecTives (2nd ed . , pp. 1 83 - 1 95 ) . St. Lou i s : Mosby. Mercer, J . R. ( 1 979) . SYSTem ofMulticultural PluraliSTic Assess­ ment technical manual. San Antonio: The Psychological Cor­ poration. Mercer, 1. R . , & Lewis, 1. F. ( 1 978) . System of Multicultural PluraliSTic Assessment. San Antonio: The Psychological Cor­ poration. Mercy, J. A . , & Steelman, L. C. ( 1 982). Famil ial influence on the intellectual attainment of children. American Sociological Review, 4 7, 532-542 . Messick, S. ( 1 980) . Test validity and the ethics of assessment. American Psychologist, 35, 1 0 1 2- 1 027 . Messick, S. ( 1 984) . Assessment in context: Appraising student perfom1ance in relation to instructional quality. Educational Researcher, 13(3), 3-8. Meyer, W 1 . , & Goldstein, D . ( 1 97 1 ) . Performance charac­ teristics of middle-class and lower-class preschool children on the Stanford-Binet, 1 960 Revision . (ERIC Document Re­ production Service No. ED 044 429) Meyers, e. E. ( 1 969) . What the ITPA measures: A synthesis of factor studies of the 1 96 1 edition. Educational {/nd PsychrJ/of!, ­ ical Measurement, 2 9, 867-876. Meyers, C. E . ( 1 978) . Review of the Balthazar Scales of Adap­ tive Behavior. In O. K. Buros (Ed . ) , The eighth mental mea­ surements yearbook (pp. 696-698 ) . High land Park, N J : Gryphon Press. M iele, F. ( 1 979) . Cultural bias in the WISe . Intelligence, 3, 149- 1 64. M i l ler, C . A . ( 1 982). Degree of lateralization as a hierarchy of manual and cognitive levels . Neuropsychologia, 20, 1 55 - 1 62 . M i ller, C . K . , & Chansky, N . M . ( 1 972) . Psychologists' scoring of WISC protocol s . Psychology in the Schools, 9, 1 44- 1 52 . M il ler, C . K . , Chansley, N . M . , & Gredler, G. R . ( 1 970) . Rater agreement on WISC protocol s . Psychology in The Schools , 7, 1 90- 1 9 3 . M i ller, G . , Dubowitz, L . M . S . , & Palmer, P ( 1 984) . Follow-up of pre-tenn infants: Is correction of the developmental quo­ tient for prematurity helpful? Early Human Development, 9, 1 37- 1 44 . M i ller, J . F. ( 1 9 8 1 ) . Assessing language production in children . Austin, TX: Pro-Ed. M il ler, M . , Stoneburger, R . L . , & Brecht, R . D. ( 1 978) . WISC subtest patterns as discriminators of perceptual disability. Journal of Learning Disabilities, 1 1 , 449-452 .

REFERENCES M i ller, N. B . ( 1 979) . Parents of chi ldren with neurological disorders: Concerns and counseling . Journal ofPediarric Psy­ chology, 4, 297-306. Miller, R . A. ( 1 974) . Social milieu and the effects of reinforce­ ment on I . Q . tests. DisserraTioll AbSTraCTS il1lernational, 35, 5 1 7B-5 1 8B . (University Microfilms 0. 74- 1 5 ,373) Millham, J . , Chilcutt, J . , & Atkinson, B . L . ( 1 978). Com­ parability of naturalistic and comrolled observation assess­ ment of adaptive behavior. American Journal of Mental Defi­ ciency, 83, 52-59. Milne, N . D . M . ( 1 975) . Relationships among scores obtained on the Wechsler Intell igence Scale for C hildren, Columbia Mental Maturity Scale and Leiter International Performance Scale by Mexican-American children. Dissertation Abstracts international, 35, 65 1 6A . (University M icrofilms No. 759073) Minde, K . , Weiss, G . , & Mendelson , N . ( 1 972). A 5-year follow-up study of 9 1 hyperactive school children. Journal of The American Academy of Child Psychiatry, 1 1 , 595-6 1 0 . Mishra, S. P ( 1 98 1 ) . Factor analysis of the McCarthy Scales for groups of white and Mexican-American children. Journal of School Psychology, 1 9, 1 78- 1 8 2 . Mishra, S . P ( 1 983). Effects o f examiners' prior knowledge of subjects' ethnicity and intelligence on the scoring of responses to the Stanford-Binet Scale. Psychology in The Schools, 20, 1 33- 1 3 6 . M ishra, S . P , & Brown, K . H . ( 1 983). The comparabi lity of WAIS and WAIS-R IQ� and subtest scores. Journal of Clinical Psychology, 39, 754-75 7 . Mitche l l , J . V , Jr. ( Ed . ) . ( 1 985) . The ninth mental measurements yearbook ( Vols. 1 -2 ) . Lincoln, NE: The Buros Institute of Mental Me�su rements of the Uni versity of Nebraska ­ Lincoln . Mitchel l , R . E . , Grandy, T. G . , & Lupo, J . V ( 1 986) . Com­ parison of the WAIS and the WAIS-R in the upper ranges ofIQ. Professional Psychology: Research and PraCTice , 1 7, 82-83 . Moffitt, T. E . , Gabriel l i , W F , Mednick, S . A . , & Schulsinger, F. ( 1 98 1 ) . Socioeconomic status, IQ, and delinquency. Journal of Abnomwl Psychology, 90, 1 52 - 1 56 . Molyneaux , D . , & Lane, V W ( 1 982). EffeCTive il1lerviewing: Techniques and analysis. Boston: Allyn and Bacon. Moore, e. H . , Boblitt, W E . , & W ildman , R . W ( 1 968) . Psychiatric impressions of psychological reports. Journal of Clinical Psychology, 24, 373-376. Moore, e. L. , & Zarske, J. A. ( 1 984) . Comparison of Native American Navajo Bender-Gestalt performance w ith Koppitz and SOMPA norms. Psychology in the Schools , 2 1 , 1 48- 1 5 3 . Moore, E . G . 1 . ( 1 986) . Family socialization and the IQ test performance of traditionally and transracially adopted black children. Developmental Psychology. 22 , 3 1 7-326. Moore, M. V ( 1 969) . Pathological writing . Asha, 1 1 , 535-538. Morales, E. S . , & George, e . ( 1 976, September) . Examiner effects in the TeSTing of Mexican-American children . Paper presented at the meeting of the American Psychological Asso­ ciation, Washington, De .

957

REFERENCES

Moran, J. D . I I I , M cCullers, J. C , & Fabcs. R . A . ( 1 9 84) Developmental analysis of the e ffects of reward on selected Wechsler subscales . American JOllrnal of Psvcholofjv. 9 7. 205-2 1 4 . Morgan , S . B . ( 1 9 84 ) . Helping parents understand the diagnosis of autism . Developmental and Behavioral Pediatrics . 5 . 78-85 . Morris, J . D . , Evans , 1 . G . . & Pearson. D . R . ( 1 978). The WISC-R subtest profile of a sample of severely emotionally disturbed children. Psychological Reports, 42 . 3 1 9-325 . Morris, J . D . , M artin, R . A . , Johnson. E . , Birch. M . C . & Thompson, D. ( 1 97 8 ) . Subtest order and W1SC-R scores of a sample of educable mentally retarded subjects. Psychological Reports, 43, 383-386. Morsbach , G . , McGoldrick, G .. & Younger, 1 . ( 1 978) . Inter­ scorer reliabil ity of the Geometric Design subtest of the WPPS I . Journal of Behavioural Science, 2, 279-284 . Moscov itch , M . ( 1 9 8 1 ) . Right-hemisphere language. Topics in Language Disorders, 1 (4) , 4 1 -6 1 . Mueller, H . H . , Das h . U . N . , Matheson, D . W , & Short, R . H . ( 1 984) . WISC-R subtest patterning of below average. average . and above average IQ children: A meta-analysis. Alberra JOllr­ nal of Educational Research , 30, 68-85 . Mueller, H . H . , Matheson, D . W , & Short , R. H. ( 1 983) . Bannatyne-recategorized WISC-R patterns of mentally re­ tarded, learning disabled , normal, and intellectually superior children : A meta-analysis . Mental Retardation and Learning Disability Bulletin , 1 1 , 60-7 8 . M u l l i n s , J . B . ( 1 9 83 ) . T h e uses o f bibliotherapy in counseling fam i l ies confronted with handicaps. In M . Seligman (Ed . ) . 77lefamily with a handicapped child: Understanding and treat­ ment (pp. 235-259) . New York: Grune & Stratton. Muma, J . R . , & Pierce, S. ( 1 98 1 ) . Language intervention: Data or evidence? Topics in Learning and Learning Disabilities, 1(2), I - I I . Murphy, H . A . , Hutchison, J . M . , & Bailey, J . S. ( 1 98 3 ) . Behavioral school psychology goes outdoors: The effects of organized games on playground aggression . Journal of Ap­ plied Behavior Analysis, 1 6, 29-35 . Murphy, K . P. ( 1 95 7 ) . Tests of abi l ities and attainments. In A . W G . Ewing ( Ed . ) , Educational guidance and the deafchild (pp. 2 1 3-25 1 ) . Manchester: Manchester University Press . Murphy, L. B. ( 1 956) . Personality in young children (Vol . I ) . New York: Basic Books . Murphy, L. B. ( 1 97 5 ) . The stranglehold of norms on the indi­ vidual child . In M . D. Cohen ( Ed . ) , Testing and evaluation: New views ( pp . 39-44 ) . Washington , DC : Association for C h i ldhood Education I nternational . ( ERIC Document Re­ production Service No. ED 1 09 1 43 ) M yers, B . , & Goldstein, D . ( 1 979) . Cognitive development i n b i l ingual and monoli ngual lower-cJass children. Psychology in the Schools, 1 6 , 1 37- 1 42 . Nagl ieri, J . A . ( l 980a) . Comparison o f McCarthy General Cog­ nit ive Index and WISC-R IQ for educable mentally retarded , .

learning d is3lbled. a nd normal children. Psychological Re­ ports, 47. 59 1 -5 96 . iagl ieri , J . A. I I YSOb ) . McCarthy and WISC-R correlations with WRAT Ac hievement Score s . Perceptual and Motor Skills , 5 1 . 392-394 . lagl ieri , 1. A. ( 1 98 1 a) . Concurrent validity of the Revised Pea­ body Picture Vocabulary Test . Psychology in the Schools, 1 8 , 286-289. l':agl ieri , J. A. ( 1 98 I b) . Extrapolated developmental indices for the Bayley Scales of Infant Development. American Journal of Mental Deficiency. 85, 548-550. t\aglieri, J. A. ( l 98 I c) . Factor structure of the WISC-R for children identified as learning disabled. Psychological Re­ pons, 49, 89 1 -895 t\aglieri . J . A. ( 1 982) . Use of the WISC-R and PPVT- R with mentally retarded children. Journal of Clinical Psychology, 38, 635 -637. t\agl ieri . 1. A . ( 1 984) . Concurrent and predictive validity of the Kaufman Assessment Battery for childre n with a Navajo sam­ ple . Journal of School Psychology, 22, 373-379 . Naglieri. J . A . ( l 985a) . Normal children's performance o n the McCarthy Scales, Kaufman Assessment Battery, and Peabody Individual Achievement Test . Journal of Psychoeducational Assessment, 3, 1 23- 1 29. agl ieri . J . A . ( 1 985b) . Use of the WISC-R and K-ABC w i th learning disabled, borderline mentally retarded, and normal chi ldren. Psychology in the Schools, 22, 1 33 - 14 1 . NagIieri. 1 . A . , & Anderson, D . F. ( 1 98 5 ) , Comparison of the W ISC-R and K-ABC with gifted students. Journal of Psycho­ edl/cational Assessment, 3, 1 75 - 1 79 . Naglieri. J . A . , & Haddad , F A . ( 1 984) . Learning disabled children's performance on the Kaufman Assessment Battery for Children: A conrurrcnt validity st udy. Journal (If Psycho­ educational Assessment, 2, 49-56 . Nagl ieri , J . A . , & Harrison, P. L . ( 1 982). McCarthy scales, McCarthy Screening Test, and Kaufman's McCarthy short form correlations with the Peabody Individual Achievement Test. Psychology in the Schools, 1 9, 1 49- 1 5 5 . Nagli{;ri, J . A . , Kaufman, A . S . , & Harrison , P. L . ( 1 98 \ ) . Factor structure of the M cCarthy Scales for school-age children with low GCls. Journal of School Psychology, 1 9, 226-2 3 2 . Naglieri, J . A . , & Pfeiffer, S . I . ( I 983a) . Rel iability and stability of the WISC-R for children w ith below average IQs. Educa­ tional & Psychological Research , 3, 203-20 8 . Naglieri, J . A . , & Pfeiffer, S . I . ( 1 983b) . Stability a n d concurrent validity of the Peabody Individual Achievement Test. Psycho­ logical Reports, 52, 672-674 . Naglieri, J . A . , & Pfeiffer, S. I . ( l 983c) . Stability, concurrent and predictive validity of the PPVT- R . Journal of Clinical Psychol­ ogy, 39, 965-967. Naglieri, J. A . , & Yazzie, C. ( 1 98 3 ) . Comparison of the WISC-R and PPVT-R with avajo c hildren. Journal of Clinical Psy­ chology, 39, 598-600.

J

958 Nay, W. R. ( 1 979). Mulrimerhod clinical assessmenr . New York: Gardner Press. Nettelbeck, T. ( 1 985). I nspection time and mild mental retarda­ tion. In N. R. Ellis & N. W. Bray (Eds . ) . lnrernarional re\'ielV of research in menral rerardarion ( Vol . 1 3 , pp . 1 09 - 1 4 1 ) . Orlando, F L : Academic Press . Neuhaus, M . ( 1 967) . Modifications in the administration of the WISC Performance subtests for children with profound hear­ ing losses. Exceprional Children , 33, 5 73-574 . Newborg, J . , Stock, J . R . , & Wnek, L. ( 1 984) . Barrelle Develop­ menral Invenrory. Allen, TX : DLM Teachi ng Resources . Newcomer, P. L . , & Hammil l , D . D . ( 1 97 5 ) . ITPA and academic achievement: A survey. Reading Teacher, 28, 73 1 -74 1 . Newland, T. E . ( 1 97 1 ) . Blind Learning Aptitude Test. Cham­ paign, I L : University of Illinois Press. New York City Board of Education ( 1 977) . Language Assessmenr Barrery: Chicago: Riverside Publ ishing. N ichol s , P. L. ( 1 97 1 ) . The effects of heredity and environment on intelligence test performance in 4 and 7 year white and Negro s i b l i n g p a i r s . Disserration Abstracrs Internationa l , 32 , 1 0 I B- I 02 B . ( University M icrofilms No . 7 1 -8 1 , 874) N ichol s , P. L. ( 1 984) . Fami l ial mental retardation. Behavior Genetics, 14, 1 6 1 - 1 70 . N ichols, P. L . , & Anderson, V E . ( 1 97 3 ) . Intellectual perfor­ mance, race, and socioeconomic statu s . Social Biology, 20, 367-374. N ichols, P. L . , & Che n , T. C . ( 1 9 8 1 ) . Minimal brain dysfunction : A prospective study. H i llsdale, NJ : Lawrence Erlbaum. N ichol s , R. C. ( 1 959) . The effect of ego invol vement and success experience on intell igence test results . Journal of Consulring Psychology, 23, 9 2 . N ichols , R . C . ( 1 974). Review of Genetics and education b y A . R . Jense n . Educarional Srudies, 5 , 35-38. Nihira, K . , Foster, R., Shellhaas, M . , & Leland, H . ( 1 974). MMD Adaptive Behavior Scale ( rev. ed.). Washington, DC : American Association on Mental Deficiency. N itko, A . J . ( 1 980) . Dist i ng u i s h i ng the many varieties of criterion-referenced tests. Review of Educational Research , 50, 46 1 -485 . Oakland , T. D . ( 1 979a) . Research on the ABIC and ELP: A revisit to an old topic . School Psychology Digest, 8, 209-2 1 3 . Oakland, T. D . ( 1 979b ) . Research o n the Adaptive Behavior I nventory for Children and the Estimated Learning Potential . School Psychology Digest, 8, 63-70. Oakland , T. D . ( 1 980) . An evaluation of the ABIC , pluralistic norm s , and estimated learning potential . Journal of School Psychology, 1 8, 3 - 1 1 . Oakland , T. D . ( 1 983a) . Concurrent and predictive validity esti­ mates for the WISC-R IQs and ELPs by racial-ethnic and SES groups. School Psychology Review, 12, 57-6 1 . Oakland, T. D . ( 1 983b) . Joint use of adaptive behavior and IQ to predict achievement. Journal of Consulting and Clinical Psy­ chology, 5 1 , 298-30 I . Oakland, T. D . , & Dowl ing, L . ( 1 98 3 ) . The Draw-a-Person test:

REFERENCES Val idity properties for nonbiased assessment. Learning Dis­ ability Quarterly. 6, 526-534. Oakland, T. D . . & Feigenbaum , D. ( 1 979) . Multiple sources of test bias on the WISC-R and Bender-Gestalt Test. Journa l of Consulting and Clinical Psychology, 4 7, 968-974. Oakland , T. D . , & Feigenbaum , D. ( 1 980) . Comparisons of the psychometric characteristics of the Adaptive Behavior In­ ventory for C h ildren for different subgroups of children. Jour­ nal of School Psychology, 18, 307-3 1 6 . Oakland, T. D . , & Houchins, S . ( 1 98 5 ) . Testing the test: A review of the V i neland Adaptive Behavior Scales, Survey Form. Journal of Counseling and De velopmenr, 63, 585-5 86. Oakland, T. D . , King, 1. D . , White, L . A . , & Eckman, R. ( 1 9 7 1 ) . A comparison o f performance on the WPPSI, WISC, and SB w ith preschool child ren: Companion studies. Journal of School Psychology, 9, 1 44- 1 49 . Oakland, T. D . , Lee , S . w. , & Axelrad , K . M . ( 1 97 5 ) . Examiner differences on actual W ISC protocol s . Journal of School Psy­ chology, 1 3 , 227-2 3 3 . Obrzut, A . , Obrzut, J . E . , & Shaw, D . ( 1 984). Construct validity of the Kau fman Assessment Battery for Children with learning disabled and mentally retarded. Psychology in the Schools, 2 1 , 4 1 7-424. O'Grady, K. E. ( 1 98 3 ) . A confirmatory maximum l ikel ihood factor analysis of the WAIS-R . Journal of Consulting and Clinical Psychology, 5 1 , 826-83 1 . Olive , H . ( 1 972). Psychoanalysts' opinions of psychologists' re­ ports: 1 95 2 and 1 970. Journal of Clinical Psychology, 28, 50-54 . Olivier, K . , & Barclay, A . G . ( 1 967) . Stanford-Binet and Goodenough-Harris Test performances of Head Start childre n . Psychological Reports, 2 0 , 1 1 75 - 1 1 79 . Ollendick, D . G . , Murphy, M . J . , & Ollendick, T. H . ( 1 97 5 ) . Peabody Individual Achievement Test: Concurrent validity w i t h j u ve n i l e d e l i nquents . Psychological Reporrs , 3 7, 935-93 8 . Ollendick, T. H . , Finch, A . J , J r. , & G i n n , F. W. ( 1 974) . Comparison of Peabody, Leiter, WISC, and academic achieve­ ment scores among emotionally disturbed children. Journal of Abnonnal Child Psychology, 2 , 47-5 1 . O'Ne i l l , A . M . ( 1 98 1 ) . " . . . O n the other foot . " Journal ofSchool Psychology, 1 9, 7 1 -7 2 . Orpet, R . E . , Yoshida, R . K . , & Meyers, C . E . ( 1 976) . The psychometric nature of Piaget's conservation of l iquid for ages six and seve n . Journal of Generic Psychology, 129, 1 5 1 - 1 60 . Osgood, C . E. ( 1 95 7 ) . Motivational dynamics o f language be­ havior. In Nebraska symposium on motivation ( Vo l . 5, pp . 348-424 ) . Lincoln: University of Nebraska Press. Oud, J. H . , & Sattler, J. M . ( 1 984) . Generalized kappa coeffi­ cient: A microsoft BASIC program . Behavior Research Meth­ ods, Instrumenrs, & Computers, 1 6, 48 1 . Overcast, T. D . , Sales, B . D . , & Kesler, J . A . ( 1 98 3 ) . Psycholog­ ical evaluation of children at the request of noncustodial par­ ents. Psychorherapy in Private Practice, 1 , 65-74. Ownby, R. L . , & Matthews, C . G. ( 1 985). On the meaning of the

959

REFERENCES

WISC-R third factor: Relations to selccte neulr Jpsychr !og.· cal measures . Journal oj Consllifing and Clinical Psvcholog\ . 53. 5 3 1 -5 3 4 .

Ownby, R . L . . & Wall brown, F H . ( 1 983 ) . Evaluating school psychological reports. Part J : A procedure for systematic feedback . Psvchology in fhe Schools, 20, 4 1 -45 . Padula. W. Y. ( 1 979). A point of discrimination- Public Law 94- 1 42 . lOllrnal oJLearning Disabilifies, 1 2 , 682-683 . Palmer. J . O. ( 1 98 3 ) . The psychological assessmenl oj children ( 2 nd ed . ) . New York: Wiley. Palmer. M . , & Gaffney. P D . ( 1 9 7 2 ) . Effects of administration of the WISC i n Spanish and English and relationship of social class to performance. Psychology in fhe Schools, 9. 6 1 -64 . Paramesh, C . R . ( 1 982) Relationship between Quick Test and WISC-R and reading abil ity as used in a juvenile setting. PercepfUal and MOlOr Skills. 55, 88 1 -882 . Para. kevopoulos. J . . & Kirk, S . A . ( 1 969). 771e developmelll and psrchollleiric characferisfics oj Ihe revised Illinois Tesf oj Psvcholinguisfic A b ilifies. U rbana : U n i versity of I l l i nois Press . Parker, K . ( 1 98 3 ) . Factor analysis of the WAIS-R at nine age levels between 1 6 and 74 years. lournal oj Consulfing and Clinical Psychology, 5 1 , 3 0 2- 3 08 . Parne s . S . ( 1 966) . Workshop Jor crecllive problem solving in­ sfifllles and courses . B u ffalo, N Y : Creat i ve Educat ional Fou ndat ion . Pasewark, R. A . , Rard i n . M . w. , & Grice . J . E . . Jr. ( 1 97 1 ) . Relationshi p of the Wechsler Preschool and Primary. Scale of I ntell igence and the Stanford-Binet ( L-M) in lower class chil­ d re n . Journal oj School Psychology, 9, 43-50. Pasewark. R . A . , Sawyer, R . N . , S m ith, E . A .. Wasscrberger, M . , Del l , D . , Brito, H . , & Lee. R. ( 1 967 ) . Concurrent val idity of the French Pictorial Test of Intell igence . Journal of Educo­ lional Research , 6 1 . 1 79- 1 83 . Pasewark, R . A . , Scherr, S . S . , & Sawyer, R . N . ( 1 974) . Corre­ lations of scores on the Vane K i ndergarten, Wechsler Pre­ school and Primary Scale of Intell igence and Metropolitan Reading Readiness Tests . PerceplUal and MOlOr Skills, 38, 5 1 8. Pau l , S . M . ( 1 98 5 ) . The Advanced Raven's Progressive Ma­ trices: Normative data for an American university population and an examination of the relationship with Spearman's g. Journal oj Experimel1fal Education , 54. 95- 1 00 . Pedersen, D . M . , Shinedling, M . M . , & Johnson, D . L . ( 1 968) . E ffects of sex of examiner and subject on children's quan­ t i tative test performance . Journal oj Personality and Social Psychology, 10, 2 5 1 -254 . Perry, N . w. , J r. , M cCoy, 1 . G . , Cunni ngham , W. R . , Falgout. J . C . . & Street, W. J . ( 1 976) . Multivariate visual evoked response c orrelates of inte l ligence . Psychophysiology, 13, 323-329. Petersen, C. R . , & Hart, D. H. ( 1 979). Factor structure of the WISC-R for a c l i nic-referred population and specific sub­ g roups. Journal oj Consulting and Clinical Psvchology. 4 7, 643-645

eeterson, J . I 1 '- � 5 ) . Early concepfions and fests oj illlelligence . Yonkers-on-:..tudlson, N Y : World Boo k . Phares, E . 1 . ( 1 98.4 ) . Clinical psychology: Concepfs, mefhods , and proJession ( rev. ed . ) . Homewood , I L : Dorsey Press. Phelps. L . . & Ensor, A. ( 1 986) . Concurrent validity of the WISC-R using deaf norms and the H iskey-Nebraska. Psychol­ ogy in fhe Schools , 23, 1 38- 1 4 1 . Phelps, L . . Rosso. M . , & Falasco, S. L. ( 1 98 5 ) . Multiple regres­ sion data using the WISC-R and the Woodcock-Johnson tests of cognitive abil ity. Psychology in the Schools, 22, 46-49. Piersc l . W. C .. Brody, G . , & Kratochwil l T. R . ( 1 97 7 ) . A further ' examination of motivational influences on d isadvantaged mi­ nority group children's intelligence test performance. Child Developmel1f, 48, 1 1 42- 1 1 45 . Piersel. W C . . & McAndrews, T. ( 1 98 2 ) . Concept acquisition and school progress : A n examination of the Boehm Test of Basic Concepts. Psychological Reports, 50, 783-786. Piersel. W. c . , & Reynolds, C . R . ( 1 98 1 ) . Factorial validity of item classification on the Boehm Test of Basic Concepts (BTBC), Forms A and B. Educational and Psychological Measuremel1l. 4 1 , 5 79-583 . Plant, W. T. ( 1 967). Cited by D . Wechsler, Manual Jor the Wechsler Preschool and Primary Scale oj il1lelligence ( p . 34) . San Antonio: The Psychological Corporation. Plant, W. T., & Southern, M . L. ( 1 96 8 ) . First grade reading achievement predicted from WPPSI and other scores obtained 1 8 months earl ier. Proceedillgs oJfhe 76th Annual COil veil/ion oJthe American Psychological Association, 3. 593-594 . Plomin. R . , & DeFries , J . C . ( 1 980) . Genetics and intel l igence: Recent data. imelligence, 4 , 1 5-24. Plumb. G . R. , & Charles, D . C . ( 1 955) . Scoring d i fficulty of Wechsler Comprehension responses. Journal oj Educa/ional Psychology. 46, 1 79- 1 83 . Polubin,ki , J . , Melamed, L . E . , & Prinzo, O . Y. ( 1 986) . Factor structure evidence for developmental levels of perceptual pro­ cessing on the Developmental Test of V isual-Motor Integra­ tion . Psychology in the Schools. 2 3 , 337-34 1 . Pommer, L . T. ( 1 986) . Seriously emotionally d isturbed children's performance on the Kaufman A ssessment Battery for Chil­ dren: A concurrent validity study. loumal oj Psychoeduca­ fional Assessmel1l, 4, 1 55- 1 62 . Popoff-Walker, L . E . ( 1 98 2 ) . I Q , SES, adaptive behavior, and performance on a learning potential measure . loumal oj School Psychology, 20, 222-23 1 . Popovich, D . , & Laham, S . L . (Eds . ) . . ( 1 98 2 ) . The adaptive behavior curriculum: Prescripfive behavior analyses Jor mod­ era/ely, severely, and proJoundly handicapped studel1fs ( Vol . I ) . Baltimore : Paul H . Brookes. Portenier, L. G . ( 1 94 2 ) . Psychological factors i n testing and training the cerebral palsied . Physiotherapy Review, 22, 30 1 -303 .

Porter. G. L . , & Binder, D . M . ( 1 9 8 1 ) . A pilot study of visual­ motor development inter-test rel iability : The Beery Develop­ mental Test of V isual- Motor Integration and the Bender V isual

960

Motor Gesalt Test. Journal of Learning Disabiliries . 1 4. 1 24- 1 27 . Porteus. S . D . ( 1 9 5 9) . The Maze Tesr and clinical psychology. Palo Alto, CA: Pacific Books. Post, J . M. ( 1 970) . The effects of vocal ization on the ability of third g radc studcnts to complete selected performance subtesls from the Wechsler Intelligence Scale for C hildren. Disserta­ rion Absrracrs In!ernarional, 3 1 , l 579- A . ( University M icro­ films No . 70- 1 9 ,602) Poteet , J. A . ( 1 980) . I nformal assessment of written expression. Learning Disability Quarterly, 3(4), 88-9 8 . Prasse , D . P , & Bracken, B . A . ( 1 98 1 ) . Comparison o f the PPYT-R and WISC-R with urban educable mentally retarded students. Psychology in the Schools, 1 8, 1 74- 1 77 . Price , T. L . ( 1 976) . Sioux children's Koppitz scores on the Bender-Gestalt given by white or native American examiners. Perceprual and Moror Skills, 43, 1 223- 1 226. Price-Williams, D. R . , & Ramirez, M . , III. ( 1 977) . Divergent thinking, cultural differences, and bilingualism. Journal of Social Psychology, 1 03, 3 - 1 1 . Prichard, C . L . , Tekie l i , M . E . , & Kozup, J . M . ( 1 979) . Devel­ opmental apraxia: Diagnostic considerations . Journal of Com­ municmion Disorders, 1 2 , 3 37-348 . Prifitera, A . , & Ryan, J . 1. ( 1 98 3 ) . WAIS-R/WA I S comparisons in a clinical sample. Clinical Neuropsychology, 5, 97-99. Prizan t , B. M . , & Duchan, J. F ( 1 98 1 ) . The functions of imme­ diate echolalia in autistic children. Journal of Speech and Hearing Disorders, 46, 24 1 -249 . Proger, B . B . ( 1 973) . Review of the Balthazar Scales of Adaptive Behavior. Journal of Special Educarion , 7, 95- 1 0 I . Prosser, N . S . , & Crawford , V B . ( 1 97 1 ) . Relationship of scores on the Wechsler Preschool and Primary Scale of Intelligence and the Stanford-Binet Intell igence Scale Form LM . Journal of School Psychology, 9, 278-283 . The Psychological Corporation . ( 1 983 ) . Basic Achievemenr Skills Individual Screener. San A ntonio : Author. Puig-Antich, J . , & Chambers, W J . ( 1 98 3 ) . Schedule for Affec­ rive Disorders and Schizophrenia for School-Aged Children . Unpublished interview schedule, Western Psychiatric Institute and Clinic, Pittsburg h . Purvis, M . A . , & Bolen, L . M . ( 1 984 ) . Factor structure o f the McCarthy Scales for males and females . Journal of Clinical Psychology, 40, 1 08- 1 1 4 . Quay, H . c . , & Peterson, D . R . ( 1 983) . Revised Behavior Prob­ lem Checklis!. Coral Gables, F L : Author. (University of M iami) Quay, H. c . , & Peterson, D . R. ( 1 987) . Manual for rhe Revised Behavior Problem Checklis!. Coral Gables, F L : Author. (Uni­ versity of Miami) Quay, L . C . ( 1 97 1 ) . Language dialec t , reinforcement, and the intell igence-test performance of Negro childre n . Child Devel­ opmen!, 42, 5 - 1 5 . Quay, L . C . ( 1 972) . Negro dialect and Binet performance i n severely disadvantaged black fou r-year-olds . Child Develop­ men!, 43, 245-250.

REFERE CES Quay, L. C. ( 1 974) . Language dialect, age , and intell igence-test performance in disadvantaged black childre n . Child Develop­ melli, 45, 463-46 8 . Quereshi , M . Y ( 1 968). Intelligence test scores a s a function of sex of experimenter and sex of subject . Journal ofPsychology, 69, 277-284. Quereshi, M. Y , & Mcintire, D. H. ( 1 984) . The comparabi l ity of the WISC, WISC-R and WPPSI . Journal of Clinical Psy­ chology, 40, 1 036- 1 04 3 . Rabourn, R . E . ( 1 98 3 ) . The Wechsler Adult I ntelligence Scale (WAIS) and the WAIS - Revised: A comparison and a caution. Professional Psychology: Research and Pracrice, 1 4 , 357-36 1 . Ramanaiah, N . V , & Adams , M . L . ( 1 979) . Confirmatory factor analysis of the WAI S and the WPPS I . Psychological Reports, 45, 3 5 1 -355 . Ramanaiah, N . V , O'Donne l l , J. P , & Adams, M . L . ( 1 97 8 ) . A test of the theoretical model of the Revised I l l inois Test of Psycholinguistic Abilities. Applied Psychological Measure­ men!, 2, 5 1 9-525 . Ramey, C . T. , Bryant, D. M . , & Suarez, T. M . ( 1 985) . Preschool compensatory education and the modifiability of intelligence: A critical review. In D. K . Detterman (Ed. ) , Curren! ropics in human inrelligence ( Vol. 1) - research merhodology ( pp . 247-296 ) . Norwood , N J : Ablex Publishing Corporation . Ramirez, D. N . ( 1 978) . College of rhe Deserr guide: Educmion of handicapped adulrs. Palm Desert, CA : College of the Desert . Rapaport, D . , Gill , M . M . , & Schafer, R . ( 1 968). Diagnosric psychological resring ( rev. ed . ) . New York: International Uni­ versities Press. Rappaport, N. B . , & McAnulty, D. P ( 1 9 85 ) . The effect of accented speech on the scoring of ambiguous WISC-R re­ sponses by prej udiced and nonprej udiced raters. Journal of Psychoeducarional Assessmenr, 3 , 275-283. Rasbury, W C . , Falgout, J . c . , & Perry, N. W , J r. ( 1 97 8 ) . A Yudin-type short form of the WISC-R : Two aspects of valida­ tion. Journal of Clinical Psychology, 34, 1 20- 1 26 . Rasbury, W C . , McCoy, J . G . , & Perry, N . W , J r. ( 1 977) . Relations of scores on WPPSI and WISC-R at a one-year interval . Perceprual and Moror Skills , 44, 695-69 8 . Ratcliffe , K. J . , & Ratcl i ffe , M . W ( 1 979) . T h e Leiter scales: A review of validity fi ndings . American Annals ofrhe Deaf, 1 2 4 , 38-44 . Ratusnik, D. L . , & Koenigsknecht, R . A . ( 1 976) . Cross-cultural item analysis of the Columbia Mental Maturity Scale : Poten­ tial application by the l anguage clinician . Language, Speech , and Hearing Services in Schools, 7, \ 86- \ 90 . Raven , J . C . ( 1 93 8 ) . Progressive Marrices . London: Lewis. Raven , J . C . ( 1 960) . Guide ro using rhe Srandard Progressive Marrices. London : Lewis. Raven , J . C . ( 1 965) . The Coloured Progressive Marrices Tesr. London: Lewis. Raven , J . c . , Court, 1 . H . , & Raven, 1 . ( 1 983) . Manual for Raven's Progressive Mmrices and Vocabulary Scales (Secrion

REFERENCES 3) - Standard Progressive Matrices (1 983 edition) Londcn . Lewis. Raven, J. c . , Court, J. H . , & Raven , J. ( 1 986) . Manual for Raven's Progressive Matrices and Vocabulary Scales (Section 2 - Colollred Progressive Matrices (/ 986 edition . with U. S. norms) . London : Lewis. Raven , 1 . C . , & Summers, B . ( 1 986) . Manual for Raven 's Pro­ gressive Matrices and Vocabulary Scales - research supple­ ment no. 3. London: Lewis. Reed, M. ( 1 970). Deaf and partially hearing children. In P M ittler (Ed . ) , The psychological assessment of melllal and physical handicaps (pp. 403-44 1 ) . London: Methucn. Reed, M. ( 1 985). Books for parents and children on learning d i sa b i l it ie s . lournal of Clinical Child Psychologv. 1 4 . 257-263 . Reeve , R . R . , French, J . L . , & Hunter. M . ( 1 983 ) . A validation of the Leiter International Performance Scale with kinder­ garten childre n . loumal of Consulting and Clinical Psychol­ ogy, 5 1 , 458-459 . Reid, 1 . B . ( 1 97 8 ) . A social learning approach tofamily interven­ tion: Observations in the home setting ( Vo l . 2 ) . Eugene . OR: Castalia Publ ishing . Reid , W. B . , Moore , D . , & Alexander, D . ( 1 968) . Abbreviated form of the WISC for use with brain-damag.::d and mciltally retarded childre n . 101lrna! of Consulting and Clinical PSI'chol­ ogy. 32, 236. Reilly, T. P , D rudge , O . W. Rosen , J. C o o Loew. D . E . , & Fischer, M . ( 1 985). Concurrent and predictive validity of the WISC-R, M cCarthy Scales. Woodcock -JohnSo n . and a'c a­ demic achicvement . Psychology in rhe Schools , 22. 380-3 82. Reisman, F. K . ( 1 985). Sequential Assessment of Mathematics Inventories . San A nton io: The Ps chological Corporation . Reisman, J . M . ( 1 973). Principles of psychotherapy with chil­ dren . New York: Wiley. Reitan , R. M . , & Davison, L. A . (Eds . ) . ( 1 974). Clinical neu­ ropsychology: Current status and applications. Washington, DC : V. H. W i nston & Sons. Reitan, R . M . , & Herring, S . ( 1 985 ) . A short screening device for identification of cerebral dysfunction in children. loumal of Clinical Psychology, 41 . 643-650 . Reitan, R. M . , & Wol fson, D . ( 1 985). n,e Halstead-Reitan Neuropsychological Test BatterY. Tucso n : Neurop ychology Press. Rellas, A. J. ( 1 969) . The use of the Wechsler Preschool and Pri mary Scale (WPPSI) in the early identification of gifted students . California lournal of Educarional Research . 20. I I - 1 19. Renzull i . J . S . , Smith. L . H . , White. A . J . . Callahan. C . M . . & Hartman, R . K . ( 1 976). Scales for Raring rhe Behavioral Characreristics of Superior Srudents. Wethersfield, CT: Cre­ ative Learni ng Press. Repp. A. C . . & Barton, L. E. ( 1 980) . Natural istic observations of i nstitutionalized retarded persons: A comparison of l icen­ sure decisions and behavioral observations . 10ll/'lla l ofApplied Behavior Analysis, 13, 333-34 1 .

96 1 Reschly, D. J . ( 1 978 ) . WISC-R factor structu res among Anglos, �Iacks. Ch icano) . and native-American Papagos. lournal of Consulting and Clinical Psychology, 46, 4 1 7-42 2 . Re schly, D . J . & Reschly, J . E . ( 1 979) . Val idity of WISC-R factor scores in predicting achievement and attention for four sociocultural groups. lournal of School Psychology, 1 7, 355-36 1 . Re chly, D . J . . & Sabers, D . L. ( 1 979) . A nalysis of test bias in four groups with the regression definition. lournal of Educa­ tional Measurement, 16. 1 -9. Resnick. L. B . ( 1 979) . The future of IQ testing in education. Inrelligence , 3, 24 1 -253 . Retzlaff. P . Slicner, N . , & Gibertini, M . ( 1 986) . Predicting WAIS-R scores from the Shipley Institute of Living Scale in a homogeneous sample. lournal of Clinical Psychology, 42 , 3 57-3:;9. Reynolds . C . R . ( 1 979). Factor structure of the Peabody Indi­ vidual Achievement Test at five g rade levels between grades one and 1 2 . lournal of School Psychology, 1 7, 270-274. Reynolds. C . R . ( 1 984) . Critical measurement issues i n learning disabilities. Journal of Special Educorion , 1 8 , 4S 1 -476. Reynolds, C . R. ( 1 985a). Evaluating subscale performance on the Adaptive Behavior Inventory for Childre n . PsvcholoE!Y ill rhe Schools . 22, 14- 1 8 . Reynolds. C . R . ( 1 985b). Standard score tables for the McCarthy Drawing Tests . Psychology in the Schools , 2 2 . 1 1 7- 1 2 1 . Reynolds, C. R . , & Gutkin , T. B. ( 1 980a) . A regression analysis of rest bias on the WISC-R for Anglos and Chicanos referred for psychological services. lournal of Abnormal Child Psy­ chology, 8. 237-24 3 . Reynolds, C . R . & Gutki n . T. B . ( I 980b). Stability of t h e WISC­ R factor structure across sex at two age levels. louf'lla l of Clinical Psychology. 36. 775-777. Reynolds, C . R . . & Gutkin, T. B. ( 1 98 1 ) . A multivariate com­ parison of the intellectual performance of black and white children matched on fou r demographic variables . Personaliry (lnd Individual Differences, 2 . 1 75- 1 80 . Reynolds . C . R . , & Hartlage, L . ( 1 979). Comparison of WISC and WISC-R regression l ines for academic prediction with black and with white referred children. lournal of Conslliring (lnd Clinical Psychology, 4 7, 589-59 I . Reynolds, C . R . , & Nigl , A . J . ( 1 98 1 ) . A regression analysis of differential validity in i ntellectual asse sment for black and for white inner city childre n . lournal of Clinical Child Psychol­ ogy. 10. 1 76- 1 79 . Reynolds. C . R . . & Pierse l . W C . ( 1 983 ) . Multiple a pects of bias on the Boehm Test of Basic Concepts (Forms A & B) for white and for Mexican-American childre n . 10llf'lla l ofPsych0educarional Assessmelll, I , 1 35- 1 42 . Reynolds, C . R . , Willson, V. L . , & Chatman. S . P ( 1 984) . Item bias on the 1 98 1 rev ision of the Peabody Picture Vocabulary Test using a new method of detecting bias. 10uf'IIa i of Psych 0educaTional Assessmelll . 2. 2 1 9-224. Reynolds, C. R . . Willson. V. L . . & Clark, P L . ( 1 98 3 ) . A four.

REFEREINCES

962 test short form of the WA IS-R for clinical screening. Clinical Neuropsychology, 5, 1 1 1 - 1 1 6 . Reynolds, C . R . , W right, D . , & Dappen, L . ( 1 98 1 ) . A com­ parison of the criterion-related validity (academic achieve­ ment) of the WPPSI and the WISC- R . Psychology in the Schools , 18, 20-23 . Reynolds, W. M . ( 1 987) . Auditory Discrimination Test ( 2nd ed . ) . Los Angeles: Western Psychological Services. Rhodes, L. E . , Dustma n , R. E . , & Beck , E. C. ( 1 969) . The visual evoked response: A comparison of bright and dull c h i l d re n . Electroencephalog raphy and Clin ical Neu­ rophysiology, 2 7, 364-372 . Rice, J . A . ( 1 972 , September). Benton's Visual Retention Test: New age, scale scores, and percentile nonns for children . Paper presented at the meeting of the A merican Psychological Association, Honolu l u . Richards, H . c . , Fowler, P. c . , Berent, S . , & Bol l , T. 1. ( 1 980) . Comparison of WISC-R factor patterns for younger and older epileptic children. Journal of Clinical Neuropsychology, 2 , 333-34 1 . Richards, J . T. ( 1 968) . The effectiveness of the Wechsler Pre­ school and Primary Scale of Intelligence in the identification of mentally retarded c h i l d re n . Dissertation A bstra cts, 2 9, 3880- A . (University M icrofilms No. 60-40 1 9) Richards, J . T. ( 1 970) . Internal consistency of the WPPSI with the mentally retarde d . American Journal ofMenta I Deficiency, 74, 5 8 1 -582 . Richman, L. c . , & Kitchell , M . M . ( 1 98 1 ) . Hyperlexia as a variant of developmental language disorder. Brain and Lan­ guage, 1 2 , 203-2 1 2 . Ricklefs, R . ( 1 986, October 2 1 ) . Faced with shortages of un­ skilled labor, employers hire more retarded workers . Wall Street Journal, p. 39. Rie, H. E . , Rie , E . D . , Stewart , S . , & Ambuel , J. P. ( 1 976 ) . Effects o f methylphenidate o n u nderachieving childre n . Jour­ nal of Consulting and Clinical Psychology, 44, 250-260. Riessman , F ( 1 976) . The inner-city child. New York: Harper & Row. Rimoldi, H . J . ( 1 94 8 ) . A note on Raven's Progressive Matrices Test . Educational and Psychological Measurement, 8 . 347-3 5 2 . Riscalla, L . M . ( 1 97 1 , September) . The captive psychologist and the captive patient: The d ilemma and alternatives. In S. L . Brodsky (Chair), Shared results and open files with the client: Professional irresponsibiliry or effective involvement ? Sym­ posium presented at the meeting of the American Psychologi­ cal Association, Washington, D C . Ritter, D . , Duffey, J . , & Fischman, R . ( 1 974). Comparability of Columbia Mental Maturity Scale and Stanford-Binet, Form L-M , estimates of intelligence . Psychological Reports, 34, 1 74 . Roach, E . G . , & Kephart, N . C . ( 1 966) . The Purdue Perceptual­ Motor Survey. San A ntonio: The Psychological Corporation. Robb. G . P. . Bernardoni, L . c . , & Johnson . R . W. ( 1 972).

Assessment of individual memal ability. Scranton, PA : I . ntext Educationa l . Roberts, M . A . , M il ic h , R . , & Loney, J . ( 1 984) . Struc:tured Observation of Academic and Play Settings (SOAPS) . Un­ published manuscript, University of Iowa, Iowa C ity. Roberts, M . C . ( 1 986) . Pediatric psychology: Psycholo·gical ill/erventions and strategies for pediatric problems. New York: Pergamon. Robinson, H. B . ( 1 980, November) . A casefor radical accelera­ tion: Programs of the Johns Hopkins Universiry and the Uni­ versiry of Washington . Paper presented at the meeting of the 1 980 Symposium of the Study of Mathematically Preco,cious Youth , Baltimore . Robinson, H . B . ( 1 9 8 1 ) . The uncommonly bright child. I.n M . Lewis & L . A . Rosenblum ( Eds . ) , The uncommon child: Genesis of behavior ( Vol . 3 , pp. 57-8 1 ) . New York: Plenum Press. Robinson, N . , & Harris, S. R. ( 1 980) . Tricks ofthe trade: Testing infants and preschoolers . Unpublished manuscript, University of Washington, Child Development and Mental Retard.a tion Center. Roe, A . ( 1 95 3 ) . The making of a scientist. New York : Dodd, Mead. Roedel l , W. C. ( 1 980a ) . Characteristics of gifted young children. In W. C. Roede l l , N. E . Jackson, & H . B . Robinson (Eds . ) , Gifted young children (pp. 7-26) . New York: Teachers College Press. Roedell , W. C. ( 1 980b ) . Programs for g i fted young childre n . I n W. C . Roede l l , N . E . Jackson, & H . B . Robinson (Eds . ) , Gifted young children (pp. 66-89). New York : Teachers Col­ lege Press . Rogers, D . L . , & Osborne, D. ( 1 984 ) . Comparison of the WAIS and WA IS .. R at different ages in a clinical population. Psycho­ logical Reports , 54. 9 5 1 -956. Rogers , S. J. ( 1 977) . Characteristics of the cognitive develop­ ment of profoundly retarded childre n . Child Development , 48, 837-843 . Rosenthal , R. ( 1 966) . Experilllenter effects in behavioral re­ search . New York: Appleton-Century-Crofts . Rosenthal , R . , & Jacobson , L. ( 1 968). Pygmalion in the class­ room. New York: Holt, Rinehart and W inston. Ross, R . T. ( 1 972) . Behavioral correlates of levels of intel­ l igence. American Journal ofMental Deficiency, 76, 545-549. Rossman. B. B . & Horn , J. L. ( 1 972 ) . Cognitive, motivational and temperamental i ndicants of creativity and intell igence . Journal of Educational Measurement, 9, 265-286. Rosso, M . , Falasco, S. L . & Koller. J. R . ( 1 984) . Investigations into the relationship of the PPVT-R and the WISC-R with incarcerated delinquents . Journal of Clinical Psychology, 40, 588-59 1 . Roszkowski , M . J . ( 1 980) . Concurrent validity of the Adaptive Behavior Scale as assessed by the V ineland Social Maturity Scale . American Journal of Memal DeficienCl', 85. 86-89 . Roszkowski . M . J . & Bean. A . G. ( 1 980) . The Adaptive Behav.

.

963

REFERENCES ior Scale (ABS) and IQ: How much un hared variance i� the r;:') Psvchology in the Schools, 1 7, 452-459 _ Roszkowski , M . 1 . , Snelbecker, G . E . , & Sack , R . ( 1 9 8 1 ) . Children's, adolescents', and adults' reports of hand prefer­ ence : Homogeneity and d iscri m i nating power of selected tasks. lournal of Clinical Neuropsychology, 3, 1 99- 2 1 3 Roth, D. L . , Hughes, C . w. , Monkowsk i , P G . , & Crosson, B. ( 1 984) . Investigation of valid ity of WAIS-R short forms for patients suspected to have brain impairment . Journal of Con­ sulting and Clinical Psychology, 52 , 722-723. Rothlisberg, B. A. ( 1 9 8 7 ) . Comparing the Stanford-Binet: Fourth Edition to the WISC-R: A concurrent validity study. Journal of School Psychology, 25, 1 93 - 1 96 . Rothman, C . ( 1 974) . D ifferential vulnerability of WISC subtests to tester effects. Psychology in the Schools, 1 1 , 300-302 . Rubin, H . H . , Goldman, 1 . 1 . , & Rosenfeld, 1 . G . ( 1985 ) . A comparison of WISC-R and WAIS-R IQs in a mentally re­ tarded residential population. Psychology in the Schools, 22. 3 9 2- 3 97 . Ruchalla, E . , Schalt, E . , & Vogel, F ( 1 9 8 5 ) . Relations between mental perfomlance and reaction time : New aspects of an old proble m . Intelligence, 9, 1 89-20 5 . Rucker, C . N . ( 1 967a ) . Report writing i n school psychology : A c riti:.:al i n vestigation. lournal of School Psychology. 5 , .

1 0 1 ·- 1 08 .

Rucker, C . N . ( 1 967b ) . Technical language i n the school psychol­ ogist's report. Psychology in the Schools, 4 , 1 46- 1 50. Rupley, W. H . , & Blair, T. R. ( 1 979) . Reading diagnosis and remediation: A primer for classroom and clinic. Chicago: Rand M c Nally. Ruschiva l , M . L . , & Way, 1 . G . ( 1 97 1 ) . The WPPSI and the Stanford-Binet: A validity and rel iability study using g i fted preschool childre n . Journal of Consulting and Clinical Psy­ chology, 3 7, 1 63 . Rust, 1 . 0 . , & Lose , B . D. ( 1 980). Screening for gi ftedness with the Slosson and the Scale for Rating Behavioral Characteristics of Superior Students. Psychology in the Schools, 1 7, 446-45 1 . Rutter, M . ( 1 9 70) . Autistic childre n : I nfancy to adulthood. Semi­ nars in Psychiatry, 2 , 4 3 5 -450. Rutter, M . ( 1 974) . The development of i nfantile autism . Psycho­ logical Medicine, 4 , 1 47 - 1 63 . Rutter, M . ( 1 97 7 ) . Infantile autism and other child psychoses. In M. Rutter & L . Hersov ( Eds . ) , Child psychiatry: Modern approaches (pp. 7 I 7 - 747 ) . Oxford : Blackwell Scientific. Rutter, M. ( 1 97 8 ) . Diagnosis and definition . I n M. Rutter & E . Schopler ( Eds . ) , Autism: A reappraisal of concepts and treat­ mel1l ( pp. 1 39- 1 6 1 ) . New York: Plenum. Rutter. M . , & Bartak, L . ( 1 97 1 ) . Causes of infantile autism : Some considerations from recent research . Journal of AlIIism and Childhood Schizophrenia, I . 2 0- 3 2 . Rutter, M . , & Bartak . L . ( 1 97 3 ) . Special educational treatment of autistic childre n : A comparative study : I I . Follow-up find­ ings and i mpl ications for services. Journal of Child Psychol­ ogy alld Psychiatry and Allied Disciplines, 1 4 . 24 1 -270. Rutter. M . , Greenfeld. D . , & Lockyer, L. ( 1 967 ) . A five to

fifteen year bllowv-upl study of i nfantile psychosis: I: ' Social and behavioural Qlutcolme . British lournal of Psychiatry, 1 13, 1 1 8 3- 1 1 99 .

Rutter, M . , & Locky<er, L. ( 1 967 ) . A five to fifteen year fol l ow-up study of infantile IPsychosi s : I . Description of sample. British Journal of PS) ciliatry, 1 13, 1 1 69- 1 1 82 . Ryan, 1 . 1 . ( 1 985). Application of a WAIS-R short form with neurological pati,ents : Val idity and correlational findings. Journal of Ps}Cilo-educationai Assessment, 3, 6 1 -64 . Ryan, 1 . J . . Georgemiller, R . 1 . , Geisser, M . E . , & Randall , D . M . ( 1 985) . Tes t retest stability o f the WAIS-R in a clinical sample. Journal of Clinical Psychology, 41 , 552-556. Ryan, J. 1 . , Larsen . 1 . , & Prifitera, A. ( 1 98 3 ) . Val id ity of two­ and fou r-subtest short forms of the WAIS-R in a psychiatric sample . Jourrwl of Consulting and Clinical Psychology, 5 1 , -

460.

Ryan, 1 . 1 . , Prifitera, A. , & Larsen, 1 . ( 1 982) . Reliabil ity of the WAIS-R with a mixed patient sample. Perceptual and Motor Skills, 55, 1 277- 1 278. Ryan, 1 . 1 . , Prifitera. A . , & Powers, L . ( 1 98 3 ) . Scoring rel iabil­ ity on the WA IS-R. lournal of Consulting and Clinical Psy­ chology, 51 , 1 49- 1 5 0 . Ryan, 1. 1 . , Prifitera, A . , & Rosenberg, S . 1. ( 1 98 3) . Interre­ lationships between and factor structures of the WAIS-R and WAIS in a neuropsychological battery. International Journal of Neuroscience, 2 1 . 1 9 1 - 1 9 6 . Ryan . 1 . J . , & Rosenberg, S . 1 . ( 1 98 3 ) . Relationship between the WAIS-R and Wide Range Achievement Test in a sample of mixed patients . Perceptual and Motor Skills , 56, 623-626. Ryan, j . J . , Rosenberg, S. J . , & DeWolfe, A. S . ( 1 984) . Gener­ alization of the WAIS-R factor structure with a vocational r 1mbilitat i on ,ample Journal of Consulling and Clinical Psy­ chology, 52, 3 1 1 - 3 1 2 . Ryan. 1 . J . , R()�enberg, S . j . , & Heilbronner, R . L . ( i 984) . Comparative relation,hips of the Wechsler Adult Intell igence Scale - Revised (WAIS-R) and the Wechsler Adult Intell igence Scale (WAIS) to the Wechsler Memory Scale ( W M S ) . Journal of Behavioral Assessmelll, 6, 37-4 3 . Ryan, J . 1 . , & Schneider, J . A . ( 1 986) . Factor analysis o f the Wechsler Adult Intelligence Scale - Revised (WAIS-R) i n a brain-damaged sample. Journal of Clinical Psychology, 42 ,

962-964.

Saccuzzo. D. P. , & Lewandowski, D. G. ( 1 97 6 ) . The WISC as a diagnostic tool . JOUT/wi o/Clinical Psychology, 32 , 1 1 5 - 1 24 . Saigh , P. A . ( 1 98 1 a ) . The effects o f positive examiner verbal comments on the total WISC-R performance of institu­ tionalized EMR students. Journal of School Psychology, 1 9, 86-9 1 .

Saigh, P. A . ( 1 98 1 b) . The validity of the WISC-R examiner verbal praise procedure as a concurrent predictor of the aca­ demic achievement of intellectually superior students . Journal of Clinical Psvchologv. 3 7. 647-649. Saigh . P. A . . & Payne. D. A. ( 1 976) . The influence of examiner verbal comments on WISC performances of EMR students. }oumal of School Psychology, 14. 342-345 .

REFERE NCES

964 Salagaras, S . , & Nettelbeck, T. ( 1 98 3 ) . Adaptive b ha\ ior of mentally retarded adolescents attending schoo l . American Journal of Mental Deficiency, 88, 57-68. Salend . S . J. ( 1 983). Mainstreaming : Sharpening up follow-up. Academic Therapy, 18, 299-304. Salend , S. J . ( 1 984) . Selecting and evaluating educational assess­ ment instruments. Pointer, 28, 20-2 2 . Sameroff, A . J . (Ed . ) . ( 1 978) . Organization and stabil ity of newborn behavior: A commentary on the Brazelton Neonatal Behavior Assessment Scale . Monographs of the Society of Research in Child Developmenr, 43, (5-6, Serial No. 1 77 ) . Sameroff, A . J . ( 1 979) . The etiology of cognitive competence: A systems perspective . In R . B. Kearsley & I . E. Sigel (Eds . ) , Infants ar risk: Assessmenr of cognirive fi.mcrioning ( pp. 1 1 5- 1 52 ) . Hillsdal e , NJ: Lawrence Erlbaum . Sandova l , J . ( 1 979) . The WISC-R and internal evidence of test bias with minority g roups. Journal of Consulring alld Clinical Psychology, 4 7, 9 1 9-927. Sandoval , J . , & Miille, M . P W. ( 1 980) . Accuracy of judgments of WISC-R item difficulty for m inority groups. Journal of Consulring and Clinical Psychology, 48 , 249-253. Sandova l , J . , Zimmerman, I . L., & Woo-Sam , J . M. ( 1 983). Cultural differences on WISC-R verbal items. Journal of School Psychology, 2 1 , 49-55. Sapp, G . L . , Horton, w. , McEl roy, K . , & Ray, P ( 1 979, Apri! ) . A n analysis o f ABIC score patterns o f selected Alabama school children. In Proceedings ofrhe Narional Associarion ofSchool Psychologisrs/Calijornia Associarion of School Psychologisrs and Psychomerrisrs . San Diego. Satterfield, J . H. ( 1 97 3 ) . EEG issues in children with minimal brain dysfunction. Seminars in Psychiarry, 5 , 35-46. Satterfield, J. H . , Cantwe l l , D . P , & Satterfield. B. T. ( 1 974 ) . Pathophysiology o f the hyperactive child syndrome. Archives of General Psychiarry, 3 1 , 839-844. Satterfield, J . H . , Cantwell, D. P , Saul , R . E . , & Yusin, A . ( 1 974) . Intel ligence . academic achievement. and EEG abnor­ malities in hyperacti ve children. American Journal of Psychia­ rry, 131 , 39 1 -395 . Satterly, D . ( 1 98 1 ) . Assessmenr in schools. Oxford. England: Blackwell . Sattler, J . M . ( 1 966) . Comments on Cieutat's " Examiner difrer­ ences with the Stanford-Binet I Q . " Perceprual and Mocor Skills , 2 2 . 6 1 2-6 1 4 . Sattler, J . M . ( 1 969) . Effects o f cues and exam iner influence on two Wechsler subtests. Journal of Consulring alld Clinical Psychology, 33, 7 1 6-72 1 . Sattler. J . M . ( 1 970) . Racial " experimenter effects" in experimen­ tation, testing , interviewing, and psychotherapy. Psvchologi­ cal Bul/erin , 73, 1 37 - 1 60 . Sattler, J . M . ( 1 973a) . Examiners' scoring sty Ie. aceu racy, abil ity, and personality scores . Joumal of Clinical Psychology. 2 9. 38-39 . Saltier. J . M . ( 1 973b) . Racial experimenter ffeets. In K . S . M iller & R . M . Dreger ( Eds . ) . Comparari\'e studies of blacks

and whires in rhe Un ired Srares ( pp. 8-32 ) . New York: Seminar Press. Sattler, J. M . ( 1 976) . Scoring di fficulty of the WPPSI Geometric Design subtest . Joumal of School Psychology, 1 4, 230-234. Sattler, J. M . ( 1 977) . The e ffects of therapist-client racial sim­ ilarity. In A. S. Gurman & A. M. Razin ( Eds . ) , Effecrive psychorherapy: A handbook of research ( pp. 252- 290) . Elmsford : N Y : Pergamon Press. Sattler, J. M. ( 1 97 8 ) . Review of McCarthy Scales of Children's Abilities. In O . K . Buros ( Ed . ) , The eighrh mental measure­ menrs yearbook ( pp. 3 1 1 -3 1 3 ) . H ighland Park , NJ : Gryphon Press . Saltier, J . M . ( I 982a) . A g e effects on Wechsler Adult Intel­ l igence Scale Revised tests . Jou rnal of Consulring and Clinical Psychology, 50, 785-786. Saltier, J. M. ( 1 982b) . Assessment of children's intelligence and special abiliries (2nd ed . ) . Boston: Allyn and Bacon. Saltier, J . M . ( 1 982c ) . The psychologist in court : Personal reflec­ tions of one expert witness in the case of Larry P. er al. v. Wilson Riles , er al. School Psychology Review, 1 1 , 306-3 1 8 . Sattler, 1 . M . , & Altes, L . M . ( 1 984) . Performance o f bilingual and monolingual H ispanic children on the Peabody Picture Vocabulary Test - Revised and the McCarthy Perceptual Per­ formance Scale . Psychology in rhe Schools, 2 1 , 3 1 3-3 1 6. Sattler, J . M . , Andres, J . R . , Squire , L. S . , W isely, R . , & Maloy, e . F ( 1 978). Examiner scoring of ambiguous WISC-R re­ sponses. Psychology in rhe Schools , 1 5 , 486-489 . Sattler, J . M . , Avila, V , Houston, W. B . , & Toney, D . H . ( 1 980) . Performance of bil ingual Mexican A merican children on Spanish and Engl ish versions of the Peabody Picture Vocabu­ lary Test . Journal of Consuiring and Clinical Psychology, 46, 782-784 . Sattler, J . M . . & Bowman, G . E. ( 1 98 1 ) . A comparison between the Koppitz and SOMPA norms for the Koppitz Developmental Bender-Gestalt Scoring System. School Psychology Review, 10. 396-397 Sattler. J. M . . & Gwynne, J. ( 1 982a) . Ethnicity and Bender Visual Motor Gestalt Test performance . Journal of School Psychology. 20. 69-7 1 . Sattler. J . M . . & Gwynne, 1 . ( l 982b) . White examiners generally do not impede the intell igence test performance of black chil­ dren: To debunk a myth . Journal of Consu/ring and Clinical Psychology. 50. 1 96-208 . Sattler. J . M . . H i l l i x . W. A . . & Neher, L. A . ( 1 970) . Halo effect in examiner scoring of intell igence test responses. Journal of Consulring and Clinical Psychology. 34 , 1 72- 1 76 . Sattler. J . M . & Kuncik, T. M . ( 1 976). Ethnicity, socioeconomic status, and pattern of WISe scores as variables that affect psychologists' estimates of " effective intell igence . " Journal of Clinical Psychology, 32 , 362-366. Sattler, J. M . . & Martin, S. ( 1 97 1 ) . A nxious and nonanxious examiner roles on two wIse subtests. Psychology in rhe Schools . 8. 347-349. Sattler. J. M . . Pol ifka. J. C . . Polifka, S . . & H ilsen, D. E. ( 1 984 ) . _.

.

REFERENCES A longitudinal study of the WISC-R and WAIS-R with specia ! education students. Psychology in the Schools. 2 J , 294- 295 Sattler, J. M . , & Ryan, J. 1 . ( 1 973a) . Scoring agreement on tht� Stanford-Binet. Journal of Clinical Psychology, 29. 3 5 - 3 8 . Sattler, J . M . , & Ryan , J . J . ( I 973b) . Who should determine the scoring of WISC Vocabul ary responses? Journal of Clin ica.l Psychology, 29, 50-54. Sattler, J. M . , & Squire, L . S . ( 1 98 2 ) . Scoring difficulty of th,e McCarthy Scales of Childre n's Abilities. School Psychology Review, I I , 83-88 . Sattler, J . M . , Squire, L . S . , & Andres, J . R. ( 1 977). Scorin ", discrepancies bctween the WISC-R Manual and two scorin", guides. Journal of Clinical Psychology, 33, 1058- 1059. Sattler, J. M . , & Theye, F. ( 1 967) . Procedural , situational . an interpersonal variables i n individual intelligence testing. Ps � ­ chological Bulletin, 68, 347-360. Sattler, J . M . , & Winget, B . M . ( 1 970) . Intelligence testing procedures as affected by expectancy and !Q. JOtlrnal of Clinical Psychology, 26, 446-44 8 . Sattler, J . M . , Winget, B . M . , & Roth, R . J . ( 1 969) . Scoring difficulty of WAIS and W ISC Comprehension, Similaritie s , a n d Vocabulary responses. Journal of Clinical Psychology, 25, 1 75 - 1 77 . Satz, P , & Bullard-Bates, e . ( 1 98 1 ) . Acquired ',phr.sia i n chil ­ d ren . In M . T. Sarno ( Ed . ) , Acquired aphasia (pp. 399-426) . New York: Academic Press. Satz , P. , & Moge l , S. ( 1 962) . An abbreviation of the WAI S for clinical use . Journal of Clinical Psychology, 18. 77-79. Satz , P , Van de Riel. H . , & Moge l , S. ( 1 967) . An abbreviation flf the WISC for c l i nical use . Journal of Consulting Psychology, 3 1 , 108. Saunders , B . T. , & Vitro, E T. ( 1 97 1 ) . Examiner expectancy and b ias as a function of the referral process in cognitive asses�­ men! . Psychology in the Schools, 8, 1 68- 1 7 ! . Sawyer, R . N . ( 1 968). A n investigation of the rel iability of the French Pictorial Test of Intell igence . Journal of Educational Research , 61 , 2 1 1 -2 1 4 . Sawyer, R . N . , Stanley, G . E . , & Watson, T. E . ( 1 979) . A factor analytic study of the construct validity of the Pictorial Test of I ntell igence. Educational and Psychological Measurement . 39, 6 1 3-623 . Scarr, S . ( 1 97 8 ) . From evolution to Larry P , or what shall we do about IQ tests? IllIelligence, 2, 3 25-342 . Schachar, R . , Sandberg, S . , & Rutter, M . ( 1 986) . Agreement between teachers' ratings and observations of hyperactivity. i nattentiveness, and defiance . Journal ofAbnormal Child Psy­ chology, 14, 3 3 1 -345 . Schachter. F. F. , & Apgar, Y. ( 1 95 8 ) . Comparison of preschool Stanford-Binet and school-age WISC IQs. Journal of Educa­ tional Psychology, 49, 320-323 . Sch i l ler, J . J . , DeSimone , J . R . , Gross , R . . Hoey. J . A . . McGuire , 1 . P , Smith, E . A . . & Torres . P A . ( 1 982 ) . A s.creening instrument for the assessment of dysnomia of chil­ dren. Clinical Neuropsychology, 4, 22-25 . Sc midtke, A . , & Schaller, S . ( 1 980) . Comparative study of

965

factor structulre (of I Raven's Coloured Progressive Matrices. Perceptual mild MlorcJr Skills, 5 1 , 1 244- 1 246. S,chmits . D . W . & Bteckenbaugh , L . ( 1 979 , March) . A com­ parison oJ WPPSi! arId McCarthy scores in a preschool popula­ tion . Paper p'rese:nte;d at the meeting of the National Associa­ tion of Scho I Ps;ychlologists/Cal ifornia Association of School Psychologists antd Psychometrists , San Diego . Schooler, D . L . . 8leebe , M . e . , & Koepke, T. ( 1 978) . Factor analysis of W ISC-R scores for children identified as learning disabled, educable mentally impaired , and emotionally im­ paired . Psychology in the Schools, 1 5 , 478-485 . Schopler, E . , Reichller, R . J . , DeVe l l i s , R . E , & Daly, K. ( 1 980). Toward obje tive classification of childhood autism: Child­ hood Autism Rating Scale (CA R S ) . Journal of Autism & De­ velopmental Dis rders, l a, 9 1 - 1 0 3 . Schopler, E. , Reichder, R . 1 . . & Renner. B . R . ( 1 986) . Th e Child­ hood Autislll Raring Scale (CA RS) . New York: Irvington. Schroeder, H . E . , & Kleinsasser, L. D . ( ! 972) . Examiner bias : A determinant of children's verbal behavior on the WISe . Jour­ nal of Consultin and Clinical Psychology, 39, 45 1 -454. Schuler, A. L . . & Goetz, L. ( 1 98 1 ) . The assessment of severe language disabil ities : Communicative and cognitive consid­ erations. AnalYSis and Intervention in Developmental Dis­ abiliti
REFERENCES Sue. D . W ( 1 98 1 ) . Counseling rhe clliluraill' dijfere// r : Thcon' & pracrice. New York: W iley. Suess, J . F , Cotten . P D o o Gu stave, F P . & Sison . P D o o J r ( 1 98 3 ) . The American A ssociation on Mental Deficiency ­ Adaptive l3ehavior Scal e : A l lowing c red it for alternative means of communication. A merican An//als of rhe Deaf. 128. 390-393. Sullivan, P M . ( 1 978). A comparison ofadmin isrrario// modifica · rions 011 rhe WISC-R Performance Scale wirh differelll careg(l ­ ries of deaf children . Unpubl ished doctoral d issertation. Uni­ versity of Iowa. Sullivan, P M. ( 1 982). Administration modifications on the WISC-R Performance scale with d ifferent categories of dea:­ children . American Annals of the Deaf: 1 2 7. 780-788. Sulzer-Azaroff, B . , & Reese , E. P ( 1 98 2 ) . Applying beha rioral analysis: A program for developing professional cOlllperence . New York: Holt, Rinehart and Winston . Sundberg, N . D . , Tapli n , 1 . R o o & lYler. L . E. ( 1 98 3 ) . lnrroduc­ rion lO clinical psychology: Perspectives, issues, and COlllribu­ rions 10 human service . Englewood C l i ffs, J : Prentice-Hal l . Suppes, P ( 1 974) . A survey of cognition in handicapped chil ­ dren. Review of Educarional Research , 44, 1 45- 1 76. Sutter, E. G . , & Bishop, P C. ( l 986a ) . Factor structure of the WISC-R and ITPA for learning-disabled , emotionally dis­ turbed , and control children. lournal of Clinical Psvchology. 42 , 975-978 . Sutter, E . G . , & Bishop, P C . ( 1 986b). Further investigation of the correlations among the WISC-R , PlAT, and DAM . Psy­ chology in rhe Schools, 23. 365-367 . Sutton , G . W , Koller, 1 . R . , & Christian. B. T. ( 1 982 ) . The Stanford-Binet mental age and the WISC-R test age: A com­ parison study. Psychology in rhe Schools, 1 9, 287 -289 . Svanum, S . , & Bringle, R . G . ( 1 98 2 ) . Race, social class. and predictive bias : An evaluation using the WISC, WRAT, and teacher ratings. IllIelligence, 6, 275-286. Swanson, E . N . , & DeBlassie, R . ( 1 97 1 ) . Interpreter efrects on the WISC performance of first grade Mexican-American chil­ d ren. Measurement and Evaluation in Guidance, 4, 1 72- 1 75 . Swanson, F L . ( 1 970) . Psychotherapists and children : A pro­ cedural guide . New York: Pitman. Swanson, H. L. ( 1 985). Assessing learning di abled children's i n tellectual performance : An information processing perspec­ tive . In K . D. Gadow ( Ed . ) , Advances in learning and behav­ ioral disabilities (Vol . 4, pp . 225-272 ) . Greenwich, CT: JAI Press. Sweet, R . c . , & Ringness , T. A. ( 1 97 1 ) . Variations i n the i ntelligence test performance of referred boys of differing racial and socioeconomic backgrounds as a function of feed­ back or monetary reinforcement . Journal of School Psychol­ ogy, 9, 399-409 . Swe rdl ik, M . E. , & Schweitzer, 1 . ( 1 97 8 ) . A comparison of factor structures of the WISC and W ISC- R . Psvchology in rhe Schools , 15. 1 66- 1 72 . Swiuky. H . N o o Haywood . H . c . , & Rotatori , A . F ( 1 98 2 ) . Who

969 are the e\'�re l .. alnd p rofoundly mentally retarded? Education and Trai//i//� o( r he Mentally Retarded, 1 7, 268-272. Taddonio. R. O . ( 1 973 ) . Correlation of Leiter and the visual subtests of the I l l inois Test of Psychol ingu istic Abilities with deaf elementary school children . Journal of School Psychol­ ogy, I I , 30-35. Tal lent, N . ( 1 963 ) . Clinical psychological consultation . En­ glewood Cliff . J : Prentice-Hall . Tal lent. N o o & Reiss . W 1 . ( 1 959a ) . Multidiscipli nary views on the preparat ion of written clinical psychological reports : l . Spontaneous uggestions for content. Journal of Clinical Psy­ cholog\·. 15. 2 1 8-22 1 . Tallent, N o o & Reiss, W J . ( l 959b) . I I . Acceptability of certain common content variables and style of expression. lournal of Clinical Psvchology, J 5 , 2 73-2 74 . Tallent , N . , & Reiss, W J . ( 1 959c ) . I I I . The trouble with psycho­ logical reports. Journal of Clinical Psychology, 15, 444- 446. Taylor, E . M. ( 1 96 1 ) . Psychological appraisal of chi/drell with cerebral defects. Cambridge, M A : Harvard University Press. Taylor, H. D . , & Thweatt, R . C . ( 1 972). C ros -cultural develop­ mental performance of Navajo children on the Bender-Gestalt Test. Perceptual and MOlOr Skills, 35, 307-309. Taylor. H. G. ( 1 98 3 ) . M B D : Meanings and misconceptions . Jou rnal of Clinical Neu rops ych o logy , S , 27 1 -287. Taylor, R . L . . & Ivimey, J. K . ( 1 980) . Predicting academic achievement : Preliminary analysis of the McCarthy Scales . Psych oloRi ca l Reports, 46, 1 23 2 . l'd ylor, R . L . , & Partenio, r . ( 1 984) . Ethnic d i fferences on the Bender-Gestalt :' Relative effects of measured iriteJ l ige·nce. Journal of Consulting and Clinical Psychology, 52 , 784-788 . 1:1)'lor, R . L . Slocumb, P R o o & O'Neill, J . ( 1 979). A short form of the McCarthy Scales of C h i l d ren's A b i l ities : Meth­ odologic;d ,HId c l i n ical applicat ions. Psychology in the Schools, 16, 347-350. Tcbeleff, M o o & Oakland , T. D . ( 1 977, August) . Relationships bPlweell the ABIC, WISC-R and achievement . Paper presented at the annual meeting of the American Psychological Associa­ tion, San Francisco. Teeter, A . , Moore, C . L . , & Petersen, J. D . ( 1 98 2 ) . WISC-R Verbal and Performance abilities of native American students referred for school learning problems. Psychology in the Schools. 1 9, 39-44 . Teglas i , H o o & Freeman, R . W ( 1 983) . Rapport pitfalls of begin­ ning testers. Journal of School Psychology, 2 1 , 229-240. Tel legen , A . , & Briggs , P F ( 1 967 ) . Old wine in new skins : Grouping Wechsler subtests into new scales. Journal of Con­ siliting Psvchology, 3 1 , 499-506. Temple. C. M. ( 1 984a) . New approaches to the developmental dyslexias. Advances in Neurology, 42, 223-232. Temple, C . M . ( 1 984b) . Surface dyslexia i n a child with epilepsy. Neuropsychologia , 22, 569-576. Tennent. G . , & Gath . D. ( 1 975). Bright del i nquents: A three­ year fol low-up study. British Journal of Criminology, I S , 386-390. Terman . L. M. ( 1 9 1 1 ) . The Binet-Simon Scale for measuring

970

intell igence : Impressions gained by its application. Psy cho­ logical Clinic, 5, 1 99-206. Terman, L. M . ( 1 9 1 6) . The measuremenr ofintelligence. Boston: Houghton M i ffl i n . Terman , L . M . ( 1 92 1 ) . A symposium. Intelligence and its mea­ surement. Journal of Educational Psychology, 1 2 , 1 27- 1 3 3 . Terman, L . M . ( 1 925). Genetic studies ofgenius ( Vol. 1 ) . Stan­ ford , CA : Stanford University Press. Terman, L . M . , & Childs, H. G. ( 1 9 1 2) . A tentative revision and extension of the Binet-Simon Measuring Scale of Intelligence. Journal of Educational Psychology, 3 , 6 1 -74 , 1 3 3- 1 43 . 1 98-208, 277-289. Terman, L . M . , & Merri l l , M . A. ( 1 960) . Stanford-Binet Inrel­ ligence Scale. Bosto n : Houghton M i ffl i n . Terman , L . M . , & aden, M . H . ( 1 959) . The gifted group at midlife . Stanford, CA: Stanford University Press. Terrell , F , Taylor, 1 . , & Terre l l , S . L . ( 1 978). Effects of type of social reinforcement on the inte l ligence test performance of lower-class black children. Journal of Consulting and Clinical Psychology, 46, 1 53 8- 1 539. Terrell , F , Terre l l , S . L., & Taylor, 1. ( 1 98 1 ) . Effects of type of reinforcement on the intelligence test performance of retarded black children. Psychology in the Schools, 18, 225-227. Tew, B. 1 . , & Laurence, K. M. ( 1 98 3 ) . The relationship between spina b ifida children's intelligence test scores on school entry and at school leaving: A preliminary report. Child: Care, Health and Development, 9, 1 3 - 1 7 . Tharp, R . G . , & Wetze l , R. 1 . ( 1 969). Behavior modification in the natural environment. New York: Academic Press. Thomas, A . , Hertzig, M. E . , Dryman, I . , & Fernandez, P ( 1 97 1 ) . Examiner effect in IQ testing of Puerto Rican working­ class children . American Journal of Orthopsychiatry, 41 , 809-82 1 . Thomas, P 1 . ( 1 977) . Administration o f a dialectical Spanish version and standard Engl ish version of the Peabody Picture Vocabulary Test. Psychological Reports, 40, 747-750. Thompson, A. P , Howard, D . , & Anderson, 1 . ( 1 986). 1\.vo- and four-subtest short forms of the WAIS-R: Val idity i n a psychi­ atric sample. Canadian Journal of Behavioural Science, 18, 287-293 . Thompson, P L . , & Brassard, M . R. ( 1 984) . Val idity of the Woodcock-10hnson Tests of Cognitive Ability : A comparison with the WISC-R in LD and normal elementary students. Journal of School Psychology, 22, 20 1 -208. Thompson, R. 1 . , 1 r. ( 1 98 1 ) . The diagnostic utility of Ban­ natyne's recategorized WISC-R scores with children referred to a developmental evaluation center. Psychology in the Schools, 18, 43-47. Thomson , M . E . ( 1 982). The assessment of children with spe­ cific reading difficulties (dyslexia) using the British Ability Scales. British Journal of Psychology, 73, 46 1 -478. Thorndike, E . L . ( 1 927). The measurement ofinrelligence. New York: Bureau of Publications, Teachers College, Columbia University. Thorndike, R . L. ( 1 968) . [Review of R . Rosenthal and L . 1acob-

REFERENCES

son, Pygmalion in the classroom] . A merican Educalional Re­ search Journal. 5 , 708-7 1 1 . Thorndike, R . L . . & Hagen, E . P ( 1 97 7 ) . Measuremel1f and evaluation in psvchology and education (4th ed . ) . New York : Wiley. Thorndike, R . L . , Hage n , E . P , & Sattler, 1 . M . ( 1 986a). Guide for administering and scoring. the Stanford-Binet Intelligence Scale: Fourth Edition . Chicago: Riverside Publishing. Thorndike , R . L . , Hage n . E . P . & Sattler, J. M. ( 1 986b ) . Technical manual. S/{/nford-Binet Il1felligence Scale: Fourth Edition . Chicago: Riverside Publishing. Thurstone, L . L . ( 1 9 3 8 ) . Primary mental abil ities . Psychometric Monographs, No . 1 . Tiber, N . , & Kennedy, W. A . ( 1 964) . The effects of incentives on the intell igence test performance of different social groups. Journal of Conslilting Psychology, 28, 1 87 . T i ffi n , 1 . ( 1 948) . Manual for the Purdue Pegboard. Chicago : Science Research Associates. Tillman, M. H . ( 1 967a) . The performance of blind and sighted children on the Wec hsler Inte l l igence Scale for Childre n : Study I . Il1fernational Journal for the Education of the Blind, 1 6, 65-74.

Tillman, M . H . ( 1 967b). The performances of blind and sighted children on the Wechsler Intell igence Scale for Childre n : Study I I . International Journalfor the Education of the Blind, 1 6 , 1 06- 1 1 2 . Til lman, M . H . ( 1 97 3 ) . I ntelligence scales for the blind: A review with implications for researc h . Journal of School Psy­ chology, 1 1 , 80-87 . Tillman, M . H . , & Bashaw, W. L. ( 1 968). M ultivariate analysis of the WISC scales for b l ind and sighted childre n . Psychologi­ cal Reports, 23, 523-526. Tillman, M. H . , & Osborne , R . T. ( 1 969). The performance of blind and sighted children on the Wechsler Intell igence Scale for Childre n : Interaction e ffects . Educalion of the Visually Handicapped, I , 1 -4 . Tomsic , M . , & Rank i n , R . J . ( 1 98 5 ) . Selecting gifted children with the Siosson I ntelligence Test: 1 9 8 1 vs. 1 96 1 norms. Psychology in the Schools, 22 , 1 02 - 1 03 . Torgesen, 1 . K . ( 1 979). What shall w e d o with psychological processes? Journal of Learnillg Disabilities, 1 2 , 5 1 4- 52 1 . Torgesen , J . K . ( 1 980) . Conceptual and educational impl ications of the use of efficient task strategies by learning disabled childre n . Journal of Learning Disabilities, 13, 364-3 7 1 . Torgesen , 1 . K . ( 1 9 8 1 ) . The relationship between memory and attention in learning disabil ities. Exceptional Education Qllar­ terly, 2( 3 ) , 5 1 -5 9 . Torgesen , 1 . K . ( 1 982a ) . The learning disabled child as an inactive learner: Educational implications. Topics in Learning and Learning Disabilities, 2 ( 1 ) , 45-52 . Torgesen, J. K . ( 1 982b). The study of short-term memory i n learning disabled childre n : Goals , methods, a n d conclusions. In K . D. Gadow & I. Bialer (Eds . ) , Advances in learning and behavioral disabilities ( Vol . I , pp. 1 1 7- 1 49). Greenwich, CT: JAI Press.

REFERENCES

Torrance, E . P ( 1 966) . Torrance Tests of Creative Thillking . Princeton. NJ : Personnel Press . Torrance. E . P , & Myers. R. E. ( 1 970) . Creative learning and teaching. New Yor k : Dodd . Mead. Tramil l , J. L . , Tramil l , J. K . . Thornthwaite . R . , & Anderson. F ( 1 98 1 ) . Investigations into the relationships of the WRAT, the PlAT, the SORT, and the WISC-R in low-functioning referrals . Psychology i n the Schools. 18. 1 49- 1 5 3 . Trites, R . L . . Blou i n , A G . A . , & Laprade , K . ( 1 982 ) . Factor analysis of the Conners Teacher Rating Scale based on a large normative sample. Journal of Consulting and Clinical Psy­ chology, 50, 6 1 5-62 3 . Trueman, M . , Lync h , A . , & Branthwaite, A . ( 1 984) . A factor analytic study of the McCarthy Scales of Children's Abilities. British Journal of Educational Psychology, 54. 3 3 1 -335 . Tsushima, W T , & Stoddard, V M . ( 1 986) . Predictive validity of a short-form WPPSI with prekindergarten children: A 3year fol low-up study. Journal of Clinical Psychology, 42 , 526-527. Tuddenham , R . D . ( 1 962) . The nature and measurement of intell igence. In L . J . Postman (Ed . ) , Psychology in the making (pp. 469-525 ) . New York: Knopf. Tufano, L. G. ( 1 976) . The effect of effort and performance reinforcement on WISC-R IQ scores of black and white EMR boys . Dissertation Abstracts International, 36, 596 1 A . (Uni ­ versity M icrofilms No . 76-6455) lurnbul l , W W ( 1 979) . I ntel l igence testing in the year 2000. Intelligence, 3, 275-28 2 . Tyler, L . E . ( 1 965). The psychology of hulnan differences ( 3 rd ed . ) . New York: Appleton-Century-Crofts . Uebersax, 1 . S . ( 1 982 -83 ) . A design-independent method for measuring the rel iabil ity of p sychiatric diagnos i s . JOllrnal of Psychiatric Research , 1 7, 335-342 . Ulibarr i , D . M . , Spencer, M . L . , & Ri'vas, G . A . ( 1 98 1 ) . Language of proficiency and academic achievement : A study of l anguage proficiency tests and their relationship to school ratings as predictors of academic achievement. National Asso­ ciation for Bilingual Education (NABE) Journal, 5, 47-80. Ungerer. J. A . , & S igman, M. ( 1 98 1 ) . Symbolic play and lan­ guage comprehension i n autistic children. Journal ofthe Amer­ ican Academy of Child Psychiatry, 20, 3 1 8-337. University of Chicago Press. ( 1 98 2 ) . The Chicago manllal of sryle ( 1 3th ed . ) . Chicago: Author. Urbach , P ( 1 974 ) . Progress and degeneration in the " IQ debate . " British Journal of the Philosophy of Science, 2 5 , 99- 1 35 , 235-259. Urbina, S. P , Golden, C. 1 . . & A riel , R. N. ( 1 982). WAISI WAIS-R: Initial comparisons. Clinical Neuropsychology. 4, 1 45- 1 46 . Uzgiris, I . c . , & Hunt . J . McV ( 1 97 5 ) . Assessment i n infancy: Ordinal scales ofpsychological development . U rbana: Univer­ sity of Illinois Press. Vacc, N. A . , & Atwel l , B . ( 1 980) . Relationship of the Adaptive Behavior Inventory for Children and intell igence . Psychologi­ cal Reports, 4 7, 402 .

97 1

Valencia. R. IR . ( ( 1983) . Stabil ity of the M cCarthy Scales of Children's Abillities over a one-year period for Mexican­ American clnilMen . Psychology in the Schools, 20, 29-34 . Valencia. R . R. ( 11 984 ) . Rel iability of the Raven Coloured Pro­ gressive Maltric;es for Anglo and for Mexican-American chil­ dren. PsycholMJY in the Schools, 21 , 49-52. Valencia. R. R . , & Rothwell, 1. G. ( 1 984) . Concurrent validity of the WPPSI with Mexican-American preschool children. Edu­ cational & Psychological Measurement, 44, 955-96 1 . Valus. A . ( 1 9 6). Achievement-potential discrepancy status of students in LD programs . Learning Disability Quarterly, 9, 200-205. Vance , H. B . . Fuller, G. B . , & Lester, M. L . ( 1 986). A com­ parison of the M i nnesota Perceptual Diagnostic Test Revised and the Bender Gestalt. Journal of Learning Disabilities, 1 9, 2 1 1 -2 1 4 . Vance, H . B . , Huelsman, C . B . , Jr. , & Wherry, R . 1 . ( 1 976) . The hierarchical factor structure of the Wechsler Intell igence Scale for Children as it relates to disadvantaged white and black children. Journal of General Psychology, 95, 287-293 . Vance, H . B . , Kitson, D . , & Singer, M . G . ( 1 985). Relationship between the standard scores of Peabody Picture Vocabulary Test - Revised and Wide Range Achievement Test. Journal of Clinical Psychology, 41 , 69 1 -693. Vance , H . B., & Wallbrown, F H. ( 1 97 7 ) . H ierarchical factor structure of the WISC-R for referred children and adolescents. Psychological Reports, 41 , 699-702. Vance, H. B . , & Wal lbrown, F. H. ( 1 978). The structure of intelligence for black childre n : A hierarchical approach . Psy­ chological Record, 28, 3 1 -39. Vance, H. B . , Wal lbrown , F H . , & Fremont, T. S. ( 1 97 8 ) . The abil itie of retarded students: Further evidence concerning the stimulus trace factor. Journal of Psychology, 100, 77-82 . Vaildenberg, S . G . , & Volger, G . P. ( 1 985). Genetic detenninants of intell igence. In B. B. Wol man ( Ed . ) . Handbook of illlel­ ligence: Theories , measuremelllS, and applications (pp. 3-57). New York : Wiley. VanderVeer, B . , & Schweid, E . ( 1 974) . Infant assessment: Sta­ bil ity of mental functioning in young retarded childre n . Ameri­ can Journal of Mental Deficiency, 79, 1 -4 . Van Etten, G . , A rkell , c . , & Van Etten, C . ( 1 980) . The severely and profoundly handicapped: Programs, methods, and mate­ rials . SI . Loui s : C . V Mosby. Van Hagen, J . , & Kaufman, A . S . ( 1 975 ) . Factor analysis of the WISC-R for a group of mentally retarded c h ildren and adoles­ cents. Journal of Consulting and Clinical Psychology, 43, 66 1 -667 . Van Melis-Wright, M . , & Strein, W ( 1 986) . Materials review: A comparison of the K-ABC global scales and the Stanford-Binet with young gifted children. Topics in Early Childhood Special Education , 6. 88-9 1 . Vega, M . , & Powe l l , A . ( 1 970) . The effects of practice on Bender Gestalt performance of culturally disadvantaged childre n . Florida Journal of Educational Research , 1 2 , 45-49. Vellutino, F R . , & Shub, M . 1. ( 1 982) . Assessment of d isorders

972 in formal school language: D isorders in reading. Topi C J ill Language Disorders, 2 (4), 20-33 . Vernon , M . C. ( 1 974) . Deaf and hard of hearing. In M . V Wisland ( Ed . ) , Psychoeducational diagnosis of exceptional children ( pp . 1 90-2 1 2) . Springfield, I L : Charles C Thomas. Vernon, M. c . , & Brown , D. W. ( 1 964) . A guide to psychologi­ cal tests and testing procedures in the evaluation of deaf and hard-of-hearing children. Journal of Speech and Hearing Dis­ orders, 2 9 , 4 1 4-423 . Vernon, P A . ( 1 983). Speed of information processing and general intelligence . Intelligence, 7, 5 3-70 . Vernon, P E. ( 1 950). The structure ofhuman abilities. New York: Wiley. Vernon, P E . ( 1 965). Ability factors and environmental i nflu­ ences. American Psychologist, 20, 723-73 3 . Vernon, P E . ( 1 979) . Intelligence: Heredity and environment. San Francis'co : W. H . Freeman . Wachs, T. D. ( 1 975). Relation of infants' performance on Piaget scales between twelve and twenty-four months and their Stanford-Binet perfonnance at thirty-one months. Child De­ velopment, 46, 929-935 . Wahler, R. G . , House, A. E . , & Stambaug h , E. E . ( 1 976). Ecological assessmem of child problem behavior: A clinical package for home, school, and institutional settings. New York: Pergamon Press. Wal ker, R. E . , Hunt , W. A . , & Schwartz, M . L. ( 1 965) . The difficulty ofWAIS Comprehension scoring. Journal of Clinical Psychology, 2 1 , 427-429. Wallbrown, F. H. ( 1979, March). Afactor analyticframeworkfor the clinical imerpretarion of the W1SC-R. Paper presented at the meeting of the National Association of School Psychol­ ogists/California Association of School Psychologists/ Psy­ chometrists, San Diego. Wal lbrown , F. H . , Blaha , J . , Wallbrown, 1 . D . , & Engin, A. W. ( 1 975) . The hierarchical factor structure of the Wechsler Intel­ l igence Scale for Chi1dren - Revised . Journal of Psychology, 89, 223-2 3 5 . Wal lbrown , F. H . , Blaha, J . , & Wherry, R . J . ( 1 973). The hierarchical factor structure of the Wechsler Preschool and Primary Scale of I ntell igence . Journal of Consulting and Clinical Psychology, 41 , 356-362 . Wallbrown, F. H . , & Fremont, T. S . ( 1 980) . The stability of Koppitz scores on the Bender-Gestalt for reading disabled children. Psychology in the Schools, 1 7, 1 8 1 - 1 84 . Wapner, w. , & Gardner, H . ( 1 979). A study o f spell i ng aphasia. Brain and Language, 7, 363-374 . Warner, M . H . ( 1 983) . Practice effects, test-retest reliability and comparability of WA IS and WAIS-R: Issues in the assessmellt of cognitive recovery in detoxified alcoholics. Unpublished doc­ toral dissertation, University of Georgia, Athens, GA. Wasik, B. H . , & Wasik, J . L. ( 1 976) . Patterns of conservation acquisition and the relationship of conservation to intelligence for children of low income. Perceptual and Motor Skills, 43, 1 1 47- 1 1 54 . Wasik, J . L . , & Wasik, B . H . ( 1 970). A note o n use ofthe WPPSI

REFERENCES in evaluating intervention programs. Measurement alld Eval­ uation in Guidance, 3, 5 4-56. Watkins, M. W. ( 1 980) . Intellectual and special aptitudes of tenth grade EMH students. Education and Training of the Mentally Retarded, 15, 1 39- 1 42 . Watson , B . U . ( 1 983 ) . Test-retest stab i l ity o f the H iskey­ Nebraska Test of Learning Aptitude in a sample of hearing­ impaired children and adolescents . Journal ofSpeech & Hear­ ing Disorders, 48, 1 45 - 1 49 . Watson , B . U . , & Goldgar, D . E . ( 1 985). A note on t h e use o f the Hiskey-Nebraska Test of Learning Aptitude with deaf chil­ dren. Language, Speech , and Hearing Services in the Schools, 1 6, 53-57 . Waugh, R. P ( 1 978) . The I . T. PA . : Ballast or bonanza for the school psychologist. Journal of School Psychology, 1 3 , 20 1 -208 . Weaver, A . S. ( 1 968) . The prediction of first grade reading achievement in culturally disadvantaged children . Dissertation Abstracts, 28, 3789 A . (University M icrofilms No. 68-2 , 887) Webb, L . 1 . , DiClemente, C . C., Johnstone, E. E . , Sanders, J . L . , & Perley, R . A . (Eds . ) . ( 1 98 1 ) . DSM-III training guide. New York : Brunner/Maze! . Wechsler, D. ( 1 939) . The measurement of adult imelligence. Baltimore: Williams & Wilkins. Wechsler, D . ( 1 949) . Manualfor the Wechsler Illtelligence Scale for Children . San Antonio: The Psychological Corporation. Wechsler, D. ( 1 955) . Manual for the Wechsler Adult Imelligence Scale. San Antonio: The Psychological Corporation. Wechsler, D . ( 1 958) . The measurement and appraisal of adult illtel/igence (4th ed . ) . Baltimore: Williams & Wilkins. Wechsler, D . ( 1 967) . Manual for the Wechsler Preschool and Primary Scale of 1ntelligence. San Antonio: The Psychologi­ cal Corporation. Wechsler, D. ( 1 974) . Manualfor lhe Wechsler 1mel/igence Scale for Children - Revised. San Antonio: The Psychological Cor­ poration . Wechsler, D. ( 1 98 1 ) . Manual for rhe Wechsler Adult Intelligence Scale - Revised. San Antonio: The Psychological Corpora­ tion . Wehman, P ( 1 979). Curriculum design for the severely and profoundly handicapped. New York : Human Sciences Press. Weiner, S. G . , & Kaufman , A. S. ( 1 979). WISC-R vs. WISC for black children suspected of learning or behavioral d isorders. Journal of Learning Disabilities, 1 2 , 100- 1 0 5 . Weisz, J . R . , & Yeates, K . O. ( 1 98 1 ) . Cognitive development i n retarded and nonretarded persoris: Piagetian tests o f the sim­ i l ar structu re hypothesis . Psychological Bulletin , 90 , 1 5 3- 1 78 . Weitz, R . ( 1 976, September). Forensic applications of psycho­ logical testing. In L. W. Field (Chair), Comribution ofpsycho­ logical testing to clinical practice. Symposium presented at the meeting of the American Psychological Association, Washington, DC. Wells, C. F., & Rabiner, E . L . ( 1 973). The conjoint family

REFERENCES

diagnostic interview and the family index of tcn. Ion. Fanlll., Process, 1 2 . 1 27- 1 44 . Werner, E. E . , Honzik . M . P , & Smith . R . S . ( 1 968) . Predi tio of intell igence and achievement at ten year ' from twenty months pediatric and psychologic examinations. Child Devel­ opment, 39, 1 063- 1 075. Wesma n , A. G . ( 1 968) . Intelligent testing. American Psychol­ ogist, 23. 267-274. Wesman. A. G. ( 1 972) . Symposium: Tests and counsel ing : I . Testing and counseling: Fact and fancy. Measuremelll and Evalumioll in Guidance, 5. 397-402 . Werrer. 1 . . & French, R. W ( 1 973) . Comparison of the Peabod} Individual Achievement Test and the Wide Range Achievement Test in a learning disability cl inic . Psychology in The Schools . 1 0, 285-286. Whalen, C. K . , & Henker. B . ( 1 976). Psychostimulants and children: A review and analysis. Psychological Bulletin, 83. 1 1 1 3- 1 1 30 . Whelda l l , K . , & Jeffree, D. ( 1 974) . Criticisms regarding the use of the E . P. Y.T in subnormality research. British Journal of Disorders o/ CommunicaTion , 9. 1 40- 1 43 . White , D . R . . & Jacobs, E . ( 1 979) . The prediction o f first-grade reading achievement from WPPSI scores of preschool chil­ dre n . Psychologv in The Schools, 16, 1 89- 1 92 . White , O . R . , & Haring, N . G . ( 1 978 ) . Evaluating educational programs serving the severely and profoundly handicapped . I n N . G . Haring & D . D . Bricker ( Eds . ) , Teaching The severely handicapped ( Vol . 3 , pp. 1 5 3-2(0) . Seattle: American Asso­ c ia t ion for the Education of the Seve re l y / P rofoundly Handicapped . White , T H . ( 1 979) . Correlations among the WISC-R. PlAT, and DA M . Psychology ill The Schools, 16, 497-50 1 . Wh itman , T L. . Sribak , J . W , Rutler, K . M . Richtf'f, R . , & Johnson, M . R. ( 1 982). I mproving classroom behavior in mentally retarded children through correspondence training. Journal of Applied Beh(/\'ior Analysis, 15. 545-564 . Whitmore, J . ( 1 985). New challenges t o common identification practices. In J . Freeman (Ed . ) , 771e psychology ofgifted chil­ dre n : Perspecti l'es on development and educaTion ( pp . 93- 1 1 3 ) . New York: Wiley. W hyte, J . , Curry, C . , & Hale, D . ( 1 985) . Inspection time and intelligence in dyslexic children. Journal of Child Psychology and Psvchia:ry, 26. 423-42 8 . W iebe, M . 1 . . & Watkins, E . O. ( 1 980). Factor analysis o f the McCarthy Scales of Children's Abil ities on pre chool children. Journal of School Psychology. 18, 1 54- 1 62 . W iederholt. 1 . L. ( 1 978). Review of the I l l i nois Test of Psycho­ l i nguistic Abilities, revised edition . In O. K. Buros ( Ed . ) . 77le eighth mental measuremenrs yearbook ( pp . 580-583 ) . High­ l a nd Park, NJ : Gryphon Press. W ied I. K. H . . & Carlson. J. S . ( 1 976). The factorial structure of the Raven Coloured Progressive Matrice Test . EducaTional and Psychological Measuremellt. 36. 409-4 1 3 . W iener. G . . & M ilton. T ( 1 970). Demographic correlates of low

973 b i r t h we:ghl' . r4merican Journal of Epidemiology, 91 . 2ffJ-272 . W/ikl er, L . . Wa. ow M . , & Hatfield, E. ( 1 98 1 ) . Chronic sorrow revisited: Parent vs. professional depiction of the adjustment of parent of mentally retarded childre n . American Journal of OrrhopsychiaTry, 5 1 , 63-70. �v'ikoff, R . L. ( 1 978). Correlational and factor analysis of the Peabody Individual Achievement Test and the WISC-R. Jour­ lIal of ConsulTing and Clinical Psychology, 46, 322-32 5 . Willerman, L. . & Fied l e r, M . F ( 1 97 7 ) . I ntellectual l y pre­ cocious p reschool children: Early development and later intel­ lectual accomplishments . Jcurnal cfGenetic Psychology, 131 , 1 3-20. Williams, R. L . ( 1 970). From dehumanization to black intellec­ tual genocide: A rejoinder. Clinical Child Psychology News­ lerter, 9. 6-7 . Will iams, R . L. ( 1 97 2 , September). 771e BITCH- IOO: A culture­ specific test. Paper presented at the meeting of the American Psychological Association, Honolulu, Hawaii . Will iams, S . E . ( 1 983). Factors influencing naming performance in aphasia: A review of the l iterature. Journal of Communica­ tion Disorders, 1 6, 357-372 . Willis, J . , & Shibata, B . ( 1 97 8 ) . A comparison of tangible reinfOl'celflent and feedback eff
of measuring inte l ligence. Journal of Applied Psychology, I , 1 1 1 - 1 22 . Yeudal l , L . T. ( 1 979, February) . Neuropsychological concomi­ tants of persistent criminal behavior. Paper presented at the annual meet i ng of the Ontario Psycholog ical Association, Toronto, Canada. Yeuda l l , L. T. , Fromm, D . , Reddon , J . R . , & Stefanyk, W O . ( 1 986) . Normative data stratified b y age and sex for 1 2 neuro­ psychological test s . Journal of Clinical Psychology, 42 , 9 1 8-946. Young, R. M . , Bradley-Johnson, S . , & Johnson , C . M. ( 1 982). I mmediate and delayed reinforcement on WISC-R perfor­ mance for mentally retarded students. Applied Research in Mental Retardation , 3, 1 3-20. Young, Y H. ( 1 974) . A black American socialization pattern. American Ethnologist, 1 , 405-4 1 3 . Yoshida, R . K . . & Meyers, C . E . ( 1 975) . Effects of label i ng as educabl e mentally retarded on teachers' expectancies for change in a student's performance. Journal of Educational Psychology, 67, 52 1 -527. Ysseldyke, J. E., & Sabatino, D . A. ( 1 97 2 ) . Identification of statistically significant d i fferences between scaled scores and psychol i nguistic ages on the I T PA . Psychology in the Schools, 9, 309- 3 1 3 . Ysseldyke, J . E . , & Samuel , S . ( 1 97 3 ) . Identification of diag­ nostic strengths and weaknesses on the McCarthy Scales of Children's Abilities. Psychology in the Schools, 10, 304- 3 1 5 . Yudin, L . W ( 1 966) . A n abbreviated form o f the WISC for use with emotionally disturbed childre n . Journal of Consulting Psychology, 30, 272-27 5 . Yule, W , Berger, M . , Butler, S . , Newham, Y , & Tizard , J . ( 1 969) . The WPPSI : An empi rical evaluation with a British sample . British Journal of Educational Psychology, 39, 1 - 1 3 . Yule, W , Gold. R . D . , & Busch. c . ( 1 982 ) . Long-term predic­ tive validity of the WPPS I : An I I -year follow-up study. Per­ sonaliry and Individual Differences. 3 , 65-7 1 . Zachary, R . A . , Crnmpton, E . , & Spiegel , D. E . ( 1 985) . Es­ timating WAIS-R IQ from the Shipley I nstitute of Living Scale. Journal of Clinical Psychology, 41 , 5 32-540. Zahn-Wa x le r, C . . McKnew, D . H . , Cummings, E. M . . Daven­ port , Y B .. & Radke-Yarrow, M . ( 1 984) . Problem behaviors and peer interactions of young c h i ldren with a manic­ depressive parent. American Journal of Psychiatrv. 1 4 1 , 236-240 . Zaide l , E. , ZaideL D. W , & Sperry. R . W ( 1 9 8 1 ) . Left and right intell igence : Case studies of Raven's Progressive Matrices following brain bisection and hemidecortication. Cortex, 1 7. 1 67- 1 86 . Zajonc . R . B . ( 1 976) . Family configuration and intell igence. Science. 1 92 . 227-236. Zampard i . M. G . ( 1 978). School psychologists' opiniollS and practices regarding the Buckley amendmelll. Paper presented at the meeting of the American Psychological Association, Toronto, Canada. Zigler, E. ( 1 982, August ) . On the definition and classification of

975

REFERENCES mental retardaTion . Paper prc e nted at the meeting of the

Amcrican Psychological Associat ion . Washington . DC Zigler. E . & Bal l a . D. ( 1 98 1 ) . Recent issues in developmental approach to mental retardation . In M . P. Friedman. J . P Das. & N . O'Connor ( Eds . ) . llllelligence and learning (pp. 25-38). New York: Plenum. Zigler, E . , Bal la. D . , & Hodapp. R . ( 1 984) . On the definition and classification of mental retardation . American Journal of Men­ .

ral Deficiency. 89. 2 1 5 -230.

Zigler. E . , & Butterfield . E . C . ( 1 968 ) . Motivational aspects of changes in IQ test performances of culturally deprived nursery school children . Child Del·elopmel1l. 39. 1 - 1 4 . Zigler, E . , & Farber. E . A . ( 1 985 ) . Commonalities between the intellectual extremes: Giftedness and mental retardation . In F D . Horowitz & M . O'Brien ( Eds . ) . 77le gifTed and ralellled: Del'elopmelllal perspecTil'es ( pp . 387-40 8 ) . Washington . DC : A merican Psychological Association. Zigler. E . , & Muenchow. S . ( 1 979) . Mainstreaming : The proo f is i n the implementation . A/llerican PsychologisT. 34. 993-996.

Zilmm,erman I . L. , COV in , T. M . , & Woo-Sam, J . M. ( 1 986). A longitudinal cf1.lmpJari .son of the WISC-R and WAIS-R. Psy­ cho logy in the Sc/lIoo is, 23, 1 4 8- 1 5 1 . Zillnm.erman. I . L. , c& Woo-Sam , J . M . ( 1 970) . The utility of the Wec hsl e r Pres,cholol and Primary Scale of Intelligence in the pulY l ic chool . JOlAmai of Clinical Psychology, 26, 472. Zilmm.erman. I. L . . & Woo-Sam, J. M . ( 1 985). Clinical applica­ tio s. In B. 3. Wol man (Ed . ) , Handbook of intelligence: Theories, measurements , and applicaTions (pp. 873-898) . New York: W ley_ Zim . J . E . . & Barnett, D . W. ( 1 984). A val idity study of the K-ABC . the WISC-R, and the Stanford-Binet with nonreferred childre n . Journal of School Psychology, 22, 369-37 1 . Zintz. M . Y. ( 1 962 ) . Problems of classroom adjustment of Indian childre n i n public elementary schools i n the Southwest. Sci­ ence Educarion . 46, 26 1 -269. Zuelzer. I . 8 . , Stedman , J. M . , & Adams, R. ( 1 976). Bender Ge talt score� in fi rst-grade children as related to ethnocultural bac kground. socioeconomic class, and sex factors. Journal of Consulting and Clinical Psychology, 44, 875 . '

_ APPEN DIX G WE CHSL E R PRESC H O O L AN D PRI MARY SCALE O F I NTELLI G E N C E REVISE D (WPPSI-R)

Thollgh! ollce ( / l l '(lk(,lI('d do('s 1I0! sllllllbi'l: - Thomas Carlyle

Standard izat ion Dev iation IQs. Scaled Scores. and Test-Age Equivalents R e l i a b i l ity Val id ity I nt e rcorre l at ions Between S ubtests and Scales Factor A n a lysis I Q Ranges and Subtest Scaled-Score R anges Normat i ve Changes on A n i ma l Pegs Comparison of the W P P S I - R and the W PP S I A d m i nistering t h e W P P S I - R W PPS I - R Subtests I nterpre t i n g the W PP S I - R A ssets o f the W P P S I - R L i m itations of t h e WPPS I - R Psychol ogical Evaluation Test You r S k i l l S u m m a ry

976

WECHSLER PRES C H O O L A N D PRIMAR Y SCALE OF

I�JTELLIC ENCE-·

This appendix describes the latest version of th,� Weer sler Preschool and Primary Scale of In tell ige nce - Rev i sed (WPPSI-R) [ D . Wechsler ( 1989 ) . Wechsler Prese/lOul und Primary Scale ofInrelligence - Rel'ised (San Antonio: The Psychological Corporat ion) J . The test is one of the m ajor instruments for assessing the cognitive abil ity of y ung childre n , The content of this appendix is based primarily on the WPPSI-R manua l . a factor analysis for each age level of the scale. and research based on the prior edition of the test . The append ix should be read in conjunction with Appendixes J and K on the W ! S C - I I I and Chapter 8 on the W I SC-R, Because the WISC - R (WISC-I I I ) and the WPPSI-R are so simi lar. psychometric . cl inical. and psychoeducat ional approaches used on the W I S C - R ( W ISC- I I I ) can also b e appl ied t o t h e WPPSI-R , However. because the WP PSI (and the W P PSI-R) has not been as widely used or researched as the WISC-R (WISC- I 1 I ) . more caution i s needed i n interpreting test findings and in generating hypotheses about the impl ications of children's performance on the WPPSI-R, The WPPSI-R (see Figure G- l ) was published 22 year after the original versio n . The length of t i me between rev isions is somewhat long, but it is comparable to that for the other Wechsler scales . The WPPSI-R is more elaborate

F i :gure G· I .

chan its predece ) o r. It has a more extensive age range ( from 3 year ' , 0 m onths to 7 years , 3 months ) and one Cilddit ional subtest (Object Assem bly ) . The new age range i,s approximately two years wider than that of the original iV PPSI -one year downward and one year upward , Unfor­ tunately, the WPPSI-R is not completely di stinct from the WISC-II I ; some items overlap with those on four WISC-Ill subtests ( Picture Completion , Mazes, Vocabu­ I ary. and Information) , As noted later in the append i x , this overlap is a limitation i n the retesting of c h i ldre n . It would have been preferable to have the scale completely distinct from the other Wechsler scales . The WPPSI-R contains 12 subtests ( see Exhibit G-l ) , 6 In the Performance Scale and 6 in the Verbal Scale . Five of the six subtests in each scale are designated as the standard subtes!s, They are Object Assembly, Geometric Design, lock Design , Maze s , and Picture Completion in the Per­ ormance Scale and I n formation. Comprehension . Arith­ metic, Vocabulary, and S i milarit ies in the Verbal Scale. The optional subrests are Animal Pegs in the Performance cale and Sentences in the Verbal Scale . Nine of the 12 subtest s are similar to those in the WISC- I I I (Object As­ sembly, Block Desi g n , M azes , Picture Completion , I nfor­ mation . Comprehension , Arithmet i c , Vocabulary, and

Wechsler Preschool and Primary Scale of I ntel ligence-Revised. Frul11 thc

Illl e!ligel/ce - Re \ ·ised. Copyright C

1 989

977

RE' VISE D (WPf'SI-R',)

Wl!chs!er Pre.l ch",,! tII/(1 Prill/III'\' 5
17

14

14

12

X

Thlal Verbal Tests Scaled Score

IQ

Performan{'E.> Score Verbal Score Full Scale Score

"See the Manual fur a rli:'l'u�:"lon of lhl' ;o;i,l!Tllficam« ' of difft'r(>!lc(>!"> b(>lw(>(>n ;o;tur€'� fin lhr tc:,l:,
SI·R) l

1 00 7

s(�'holol prograrrns, pe;rhaps tapping important developmen­ wt! o,r matu ratilOnali falctors needed for chool success i n the 1cJwer grades. 4. Good adll1/ill i,5. tralive procedures . The prescribed p'rocedure

-

3 40

-1 . 6�

.I . thing l ike " Now we'll do something di fferent . " At the end of the test, thank the child for comi ng and for being cooperative ( i f appropriate) . 10. On I n formation item 8 . i f the child fi rst says "5. " ask " How many counting the weekend?" If the child then says " 2 ."' ask " How many altogether. a l l week?"" I I . On A rithmetic, you may repeat an item only once. 1 2 . On Comprehension. the Record Form is marked with a n asterisk next to those items that require you 10 ask for a second response. 1 3 . On Digits Forward and Digit> Backward. ;t!\\ays ad· minister both trials of each series. cao the dig ih at the rate of

one per second and at an even pace - that is. no chunking - and drop your voice inflection sl ight l y on the last d igit in the sequence.

1 4 . On P icture Completion, place the St i m u l us Booklet flat on the table close to the c h i l d . Note that the t i me l imit is 20 seconds. If you are not s u re whether the chi ld's verbal response is correct , say "Show me where you mean . " You must know the cautions (top of page 61 in the W I SC - I I I manual) and when to use them. On item 29, g ive credit to "rib" or " spoke " (personal commun ication , Lawrence Weiss, The Psychological Corpo­ ration, September 199 1 ) . A l t hough the letters are missing on the buttons of the telephone ( item 23 ) , The Psychological Corporation advised that you not g ive credit to a response mentioning this omission because it is an unessential missing part ( personal communication, Lawrence Weiss, The Psycho­ logical Corporation, January 1 992 ) . I f t h is is the fi rst unessen­ tial missing part mentioned by the child , say "Ye s . but what is the most i m portant part that is missing?" The c h i ld can turn the pages of the test booklet if you are s u re that he or she w i l l al low you to set the pace. 1 5 . On Picture A rrangement, when you are doing the sam­ ple, move each card down to a new row ral;ler than shifting the cards w ithin their original row. Place the cards at least 3 or 4 inches away from the edge of the table. 16. On Block Design, when you demonstrate the samples. put them together slow ly. Be careful not to cover the blocks with your hand: the ch i ld needs to see what you are doing. Make the designs so that they a re i n the appropriate d i rection for the child. This means that you w i l l be making the designs upside dow n . Don't make a design right side up and then turn the whole t h i ng a round to face the c h i l d . 17. On Symbol Search . o n l y items t h a t have been at­ tempted within 120 seconds a re counted as correct or incorrect ( personal communication , Lawrence We iss, The Psychologi­ cal Corporation, January 1 992 ) . 1 8 . To fac i l itate the scoring of the Coding subtest , write at lhe end of each row of the Cod ing scoring template the cumulative total number of symbols up to and including that row. For the Coding A template. the n u m bers that should be written at the ends of rows I through 8 , respectively, are 3 , I I , 1 9 . 2 7 , 3 5 , 43 . 5 1 . and 59. O n the Cod i ng B template. lhe cumu lative totals are 14 . 35 . 56 ; 77. 98, and 1 19 : these numbers should be w ritten at the ends of the first through sixth rows . After you w rite the numbers, laminate the lem­ plate to prolong its l ife (cf. Danielson , 199 1 ) . (In January 1 992, Aurelio Pri fitera , from The Psychological Corporation. informed me that the second printing of the templates w i l l incl ude these numbers . ) 1 9 . O n subtests that rece ive bonus points for speed. make (Erhibil continues /leXI page)

1 05 7

ADMINISTERING THE Wise-III

Exhibit 1-2 (cont.) s u re that the scorc you c i rcle on the Record Form corrc,ponlJ, to the t i me taken by the child to complete the itcm . III n( ) ( a ,c

D f a l short form. , " riHe S F beside the appropriate I Q . I f IQs

v. e r·e prorated .

20. Keep materia l s . othe r than t hose needed for the te 't . olr

\VlrIlC

PRO beside each appropriate I Q .

2 9 . I f fewer [1'lan five subtests we re ad m i n istered i n the

g i ve bOllus point, for O-poi nt answers .

Vcr'bal section or fev, er than five s u btests in the Performance

the table ( for example. soda cans. pocketbook . and key , j .

secn ion . use the Tel l cge n and Briggs short-form procedure

2 1 . Carefu l l y sco re each protocol and recheck your

des,cribed on pag,e 1 3 8 of this text to compute t he I Q . This

SCl)i"­

i n g . If you fa i led to q uest ion a response when you should ha\ e

pro,cedure i� the roost rel iable one for prorating I Q s . You can

and the response is obviously not a 0 response. g i ve the child

also consu lt Tahle L - 1 2 , L- 1 3 , L - 1 4 . or L- 1 5 . These tables

the most appropriate score based on the chi ld\, actual

pro'vide est imate

response. 2 2 . If a subtest was spoiled. w rite spoiled by the subtest total score and on the front cover of the Record Form next

to

F u l l Scale Deviation Quot ients fo r several

b i nation, r s orr forms based on t he Tel legen and B r iggs

co

pro,cedure.

:10. M ake a profile of the exa m i nee's scaled scores on the

the name of the subtes t . If the sublest was not admini,tercd .

froI:lt of the Record Form by plot t i ng the scores on the g raph

write NA in the margin of the Record Form next to the subtest

prolv ided . 3 1 . Look up tlue confidence i nterva l s for the F u l l Scale I Q ,

name and on the front of the Record Form . 2 3 . Add the raw sco res fo r each subtest careful ly. Make

Ve r'bai Scale IQ. a.nd Performance S c a l e IQ i n Table L - I or L - 2

s u re that you g i ve c redit for items not admini stered below the

i n Append i x L o f this tex t . Use Table L- I , u n less y o u are

start i ng-point items. Be sure to add correctly the poinb a" o­

ma ' i ng a long-term pred ict ion : i n that case use Table L-2 . A l l

c i ated w it h the c i rcled numbers on the Record Form for a l l

comfidencc interv als i n Tables L - I a n d L - 2 are based o n the

slJbtests that have bonus points.

exami nee's exact age g roup as wel l as on the total sample.

24. Transfer subtest scores to the front of the Record Form

NOlrmally. the confidence i ntervals are not used with the Ve r­

ullder Raw Scores. A fter t ransfe r r i ng a l l raw scores to the

bal or Performance I Q s , u nless these are t he o n l y IQs re­

front page of the Record Form. check to see that the raw scores

por ted . W rite the confidence i ntervals on the front cover of the

on the front page match those noted i n s ide the Record Form

Record Form in t he space provided . :'; 2 . Look up the pe rcent ile rank and class i ficat ion fo r each

for each su btest . 2 5 . Transform raw scores into scaled scores by USing Table

of t he IQs in Tables B C - I and BC-2 on the i nside back cover of

A . I on pages 2 1 7 to 249 of the W I SC - I I I manua l . Be sllre to

t h i_' text . You can also use Tables A . 2 . A . 3 , and A . 4 (pages

use the page of Table A . I thar i, appropriate for the chi ld's :1tlc

251 to 254 ) i n t ilt: W I S C - I I I manual for the perc e n t i l e ranks

and the correct row and column fo r each t ransformation .

and Table � 8 ( page 32

2 6 . Add the scaled scores for the five standard Ve rbal subte,t, to compute

the

, U I I ! o r the ,cabJ ,wres.

Du no!

u,c

i n the W I SC - I I I manual ) for

cla�sification , .

JJ. I r y ou

w a l l l t t l obtain test-age equiva l e nt s . u s e Table

Digit Span to compute the Ve rbal score u n k�s you have

A . 9 on page 2 .1 9 uf the W I SC - l I l manu a l . They can be p l aced

substituted it for another Ve rbal suntest . Add t he ,caled score,

( i n parcnthesc�j in the right margin of the box that con t a i n s the

for the five standard Performance subtests. Do not include

>caled sco re, . For tcst-age equ ivalents abo\'e t hose i n the

Symbol Search or Mazes to compute the Performance score

tabl c. lise the highest test -age e q u i valent and a p l u s s i g n . For

u n less you have substituted Symbol Sea rch fo r Cod i ng or

test -age equivalents below those in the table. use the lowest

Mazes for another Performance s u bte s t . You can i ncl ude both

test -age equ i valent and a minus s i g n .

Mazes and Symbol Search to compute the Performance score

3 4 . I f you wan t t o . you c a n e n t e r t he factor scores on the

if you have substitu ted Symbol Search fo r Coding and M azes

front cover of the Record Form. The Dev iation Quotients

fo r some other Performance subtest . S u m thc Ve rbal and the

associated with the factor scores advocated i n this text can be

Performance subtest scaled score� to obtain the sum for the

obta i ned from Tab les A . 5 and A . 7 i n t he W I SC - I I I manual fo r

F u l l Scale. Recheck all of your add it ions. 27. Convert the sums of scaled scores for the Verbal . Performance. and F u l l Scales by use of the appropriate con­

Verbal Comprehe n s ion and Processi ng Speed . respe c t i vely. and

from

Table

L- 1 3

in

Append i x

L

for

Perceptual

Organi zation .

version tables in Append i x A (pages 2 5 1 to 254) in the W i Se­

3 5 . In summa ry. read the d i rections verba t i m . pronou nce

III manual . Use Table A . 2 for the Verbal Scale. Table A . 3 for

words clearly. q u e ry at the appropriate t i me s . start with the

the Pcrformance Scale. and Table A . 4 for the F u l l Scale. Be

app ropriate ite m . d i scont i nue at the proper p l ace. p l ace items

s u rc to u,e the corrcct table fo r the appropriate scale. Record

properly before the c h i l d . use correct t i m i n g . and fol low the

the IQs on thc front of the Record Fo r m .

speci fic guidel i ne ., i n the manual for adm i n istering the test.

2 8 . Recheck a l l o f your work. I f t h e IQ was obtained b y use

APPENDI X I

1 0 58

WECHSLER INTELLIGENCE SCALE FOR CHILDRE N - I I I (WISC-III)

Exhibit 1-3 Administrative Checklist for the WISC-1 I 1 A DM I N ISTRATIVE CHECKLIST FOR T H E W[SC-III Name of examiner:

Dale:

Nallle of examinee:

Name of observer:

( Note. If an ilem is 1101 applicable, mark NA 10 Ihe leji of Ihe IlUmbel: ) Picture Completion I . Reads directions verbatim

2. 3. 4. 5. 6. 7. 8. 9. [0. I I.

12.

[3.

14. 15.

16.

17. 18. 19. 20. 21.

Reads directions clearly Pronounces words i n queries clea rly Starts w ith appropriate item P l aces booklet flat on table. close to child Begins timing after last word of instructions Gives a maximum of 20 seconds o n each item Gives correct answer for sample item and items I and 2 i f child fails these items Does not give correct answer on items 3 -30 G ives the prompt "Yes. but what's missing?" no more than one t i me Gives the prompt "A part is missing in the picture. What is it that is missing?" no more than one time G i ves the prompt "Yes. but what i s the most important part that is missi ng" no more than one time Inqu i res correctly on items 6 . 1 3 , 2 1 . 2 3 , 26. and 28 when certain responses are given G ives credit t o correct responses made after the prompt Administers items i n reverse order, when the fi rst or second item administered is fai led, to children ages 8 to [6 until they pass two consecutive items Gives credit t o items not administered when those items precede two consecutive successes G i ves credit for a correct oral o r pointing response G i ves I point credit for each correct response Gives no credit for correct responses given after time limit Records 0 or I point for each item Discontinues subtest after 5 consecutive fai l u res

Circle aile

Yes Yes Yes Yes Yes

No No No No No

Yes

0

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

0

Yes

No

Yes

No

Yes

No

Picture Completion (conI.)

22 . Adds points correctly

No

Yes

No

Yes

No

Yes Yes

No

Yes

0

Yes

No

Yes Yes Yes Yes

No No No No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Comment s :

Information I.

2. 3. 4. 5. 6. 7. 8.

9.

10. II.

12. 13.

Yes

Circle One

14. 15.

0

16.

Reads directions verbatim Reads d i rections clearly Reads items verbatim and clearly Starts with appropriate item G ives sufficient t i me for child t o respond to each question Gives correct answer for item I i f child fails item Does not give correct answer for items 2-30 Says " Explain what you mean" o r "Tell me more about it" for responses that are not clear Gives prompts when the child's response suggests that the child h a s misheard o r misunderstood the exact meaning of the question Repeats question if child says h e ( she) does not understand it Inquires correctly on items 4 . 8 , 1 6 , 1 8 . 1 9 . 2 1 . 24. 26. 2 8 . 29, 30 when certain responses are given Gives credit to correct responses made after the prompt (or inqui ry ) Admin isters items i n reverse order, when the fi rst or second item administered is failed . to children ages 8 to 16 until they pass two consecutive items Does not ask leading questions or spell words Gives credit t o items not administered when those items precede two consecutive successes Gives I point cred it for each correct response

(Exhibil colllillues neXI page)

1 059

ADMINISTERING THE Wise-III

Exhibit 1·3 (cont_) Circle aile

Information (conL) 1 7 . Records 0 o r I poi n t fo r

1 8 . Discont i nues subtest

fa il u re s

1 9 . Adds p o i n ts

each item afte r 5 consecut ive

correctly

Ye,

No

Yes

No

Yes

No

C o mme n ts :

I . Reads d i rections ve rb at im Reads d i rections clearly 3 . Correctly selects ei th er Cod i n g A o r B 4 . P rovides two no. 2 graphite pe nci l s w i t hou t e ra s e rs S . Removes Coding Re s p on se Sheet from the Record Form 6 . P rov i d es a smooth work surface 7 . Pr(wides p,xtra Coding Response Shep,t f("lf Ie ft-handed ch i Id ren 8 . Points to the key w h i le read i ng the fi r s t part of the instructions 9 . P o i nts to t he proper forms (e.g . . star. bal l . t r i a n gle circle, box with two· l i n e s ) w h i le reading i n st ru ct ion s 1 0 . Follows d i rections in manual for po i n t i ng to sample items while readi n g di re ct ion s I I . Praises c h i l d s successes 01 1 eac h sample item by saying ye s or "right" 1 2 . Corrects chi ld's mistakes on sa m pl e i t e ms 2.

"

"

1 3 . Does not begi n sublest u n l i l c h i ld clearly

understands the task

1 4 . G i ves proper i nstruct ions after

Yes

No

Yes Yes Yes

No No No

Yes Yes Yes

No No No

Yes

No

Yes

No

Yes

No

3 .., 1 9

Yes

No

item 7

Yes

No

Yes Yes

No No

Yes

0

Yes

No

Yes

No

Yes

No

Yes

No No No No

number of incorrect boxes from n u m be r atte mpted 2.5 . G ives t ime-bolnus c redits appropriately on Cod i ng A �6. G ives no t ime -bonus credits on Coding B 27. Rec o rd s c rrect Humber of items

No No

Yes

No

2 . Reads d i re cti o n s clearly

No

4 . Begins w i t h

u nderstands the task Provides proper ca u t ion the fi rst time child omits item or do c s on l y one type: "Do them in order. Don' t s k i p any. " Then p o i nts to t he next item and says "Do this one nex t .

1 6 . G i ves caution about omitting or s k i pp i ng

item only one t i me Re m i nd s c h i ld 10 con t i nue u n l i l told

( w hen needed )

Docs not t i me t he s a mp le it em s

Beg in s t i m i ng

10

20 . A l lows 1 20 seconds

2 2 . Uses scori n g stenc i l t t l scor.:: subte,t

Sim ilarities I . Reads d i rec t i o n s

verbat im

3 . Reads items �'erba t i m a nd clearly

Yes Yes

No

Yes

No

sa mp le item and then item I . reg a rd less of hild's age 5 . Gives sufficient t ime for c h i l d t o respond to each quest ion

Yes

No

6. Gives correct answer for items I and 2 if child fa i l s items 7 . Does not give correct a nswer for items

Yes

No

8.

Yes

No

Yes

Yes Yes Ye s

No

No

No

No

Ye s

No

Ye s

No

stop Yes

No No

Yes

No

Yes

i m me d i a t e l \' a ft e r

c o m p let i n g i n s t r uc t i on s

21 Record, t i m.:: accu rat.:: l y

No

child

co m p l e t es the sample items a nd

19.

No

Yes Yes

'

18.

Yes

2 4 . Adds number of correct boxes o r s u bt rac t s

Ye s

,

1 7.

2: 3 . Pia e s a mark through each i ncorrect b ox

C o m me n ts :

Coding

15.

Circle One

Codling (conI . )

Yes

Ye s Yc,

No

No No

l)

Gi\'es a n c x amp le of a 2 -point re sponse if a I -point re s pon se is given on i tem 6 or

every response fol lowed in the W I SC- I I I manual by a ( Q ) . even if it is a O - po i nl re�ponse 1 0 . Querie� vogue respon s es I I . Says Ex p l a i n what you mean " or "Tell me more about it- and no other statement to query a rc�ponse 1 2 Docs not query a clearcut response, e s p ec i a l l y one that is not fol l owed in the W I SC-I l l manual by a (Q) 1 3 . Asks " Now w h ich one is it')" each time a child's response contains both a correct and an i ncorrect a n swe r 1 4 . Gives I point credit for each correct re s pon se to it e m s 1 -5 1 5 . G i ves l o r 2 p o i n t s cred it for each correct rcspon Example: A three-subtest short form composed of the A rithmetic. Vocabul ary, and Block Design subtests is admin­ istered to a 6-year-old child. The child obta ins scaled scores of 7 , 1 2 . and 1 3 on the t h ree subtests. The four steps are as follows: T he three scaled scores are summed to yield a composite score of 32. 2. The correlations between the three subtests are obtained from Tab!e C . I (page 270) of the WISC-III manual (Arith­ metic and ocabulary, . 50; Arithmetic and Block Desig n , . 5 1 ; Vocabulary a n d Block Design, . 5 3 ) . These are summed to yield 1 . 54 (f.r k ) . j 3 . The appropriate row i n Table C-36 i n Appendix C i s the third one under the heading "3 Subtest s . " The values for the constants a and b are 2 . 0 and 40, respectively. 4 . The formula I .

Deviation Quotient

=

(composite score X a) + b

is used to obtai n a Deviation Quotient of 104

1 < 32 x 2 0) + 40J .

Computing the Reliability Coefficient of the Short Form

The follow i ng formula is used to obtain the reliability of the short form: rss

where

r,s

rii

ri}

k

=

f.rii + 2f.rij

k + f.2rij

= rel iabil ity of the short form

=

= =

rel iabil ity of subtest i correlation between any subtests i and ) number of component subtest s .

Example: The rel iability o f the two-subtest combination of Vocabulary and Block Design is calculated in the fol lowing way. given rii ( Vocabulary) = . 87 , rii (Block Design) .87, and rij (Vocabulary and Block Design) = .46. =

r

Sf

=

1 . 74 + . 92 2 + .92

=

2 . 66

--

2 .92

=

.91

(Exhibit continues next page)

1 0 70

APPENDIX I

Exh ibit 1-4 (cont.) Computing the Validity Coefficient of the Short Form

The fol lowing formula is used to obtain the validity of the short form :

where

r�w =

modified coefficient of correlation between the composite part and the composite whole rjl = correlation between any subtest j included in the part and any subtest I included in the whole, where any included correlation between a subtest and itself is represented by its rel iability coefficient rij = correlation between subtests i and j rim = correlation between subtests I and m k number of component subtests I = number of subtests included in the whole.

WECHSLER INTELLIGENCE SCALE FOR C H ILDREN-III (WiSe-III)

To obtain EEr){, total the fol lowing three sums: (a) the sum of the reliabil ities of the component subtests, ( b) twice the sum of the intercorrelations among the component subtests (2Eri) , and ( c ) the sum of the intercorrelations between any compo­ nent subtest and any noncomponent subtest . Example: The validity of the two-subtest combination of Vocabulary and Block Design is calculated in the fol l owing way, given rii ( Vocabulary) = . 8 7 , rii ( B lock Design) = . 87 , and rij ( Vocabulary and Block Design) = .46. In this example, all 13 WISC-II I subtests are used . I n other cases, only the 10 standard subtests may be used .

r� w = ��=::::;C7::::;7-:-r:;:;;;=��;:::;;:� 1 . 74 + . 92 + 9 . 74 J2 + 2 ( . 4 6) -J1 3 + 2 ( 2 8 . 9 1 ) 1 2 .40

----

( 1 . 7 1 )( 8 .42)

=

Atkinson recommended that reduced-item short forms not be used.

Vocabulary and Block Design Short Form A popular screen i ng short form consists of Vocabulary and Block Design. These two subtests have excellent rel iabil ity, correlate highly w ith the Full Scale over a wide age range, and are good measures of g. You can use Table L- 1 2 i n Appendix L t o convert the sum o f scaled scores o n these two subtests d i rectly to an estimated Deviation Quotient.

Information, Vocabulary, Picture Completion, and Block Design This short-form combination , wh ich contains two Verbal subtests and two Performance subtests, has h igh rel iability (rll = . 9 3 5 ) . It takes longer to adm inister than the Vocabu­ lary and Block Design short form , but provides more clinical and diagnostic information . You can use Table L-14 in Appendix L to convert the sum of scaled scores on these fou r subtests d i rectly to an estimated Deviation Quotient.

Other Useful Short Forms Other short forms discussed in the l i terature for the WISC-R , which you can also use for the WISC-II I , are Similarities and Vocabulary ( Fe l l & Fel l , 1 982 ) ; Sim­ i larities and Object Assembly (Fell & Fel l , 1982 ) ; S im­ ilarities , Vocabulary, and Block Design (Karnes & Brown ,

=

1 2 . 40

--

1 4 .40

= . 86

Source: Adapted from Tellegen & Briggs ( 1967 ) .

1 98 1 ) ; Similarities , Vocabulary, Block Design, a n d Object Assembly (Karnes & Brown, 198 1 ) ; Similarities, Vocabu­ l ary, Block Design , and Picture Completion (Clarizio & Veres , 1 984) ; Sim ilarities, Object Assembly, and Vocabu­ lary ( Dirks, Wessels, Quarfot h , & Quenon , 1980) ; I n for­ mation , Comprehension , Block Design , Picture A rrange­ ment , and Coding (Kennedy & Elder, 1982 ) ; and Arithmetic , Vocabulary, Picture Arrangement, and Block Design (Kaufma n , 1 976b) . If you want to give a short form that you can administer quickly and score relatively easily, consider the fol lowing combinations ( re l iabil ities shown i n parentheses): two subtests - Information and Picture Completion ( ru = . 8 7 ) ; three subtests - Information, Similarities, and Pic­ ture Completion (rtX = . 9 1 ) ; four subtests - Information , Similarities, Picture Completion , and Object Assembly ( rxx . 9 1 ) . These short forms , although they do not fal l into the IO-best combinations, are effective a s screening tools and have i n the i r favor, as noted earl ier, quick admin­ istration with a m i nimum of scoring problems. The four subtests have relatively high correlations with the Ful l Scale ( r's of . 72 , . 72 , . 5 8 , and . 5 6 , respectively) . You can use Tables L- 1 2 , L- 1 3 , and L- 1 4 . respecti vely, to convert the sum of scaled scores on these combinations directly to an estimated Deviation Quotient. The footnotes at the bottom of these tables indicate which columns to use for these short-form combination s . A word o f caution i s in order, however. Even though the Information and Picture Completion short form is more convenient in terms of time and scoring than Vocabulary =

CHOOSING BETWEEN THE WISC-III AND THE WPPSI R AN:J d�,'''''' E E�\J T HE Wise-II ANJO ,rHE WAIS-R

and Block Design, you do not get any informdtion about how the child copes with a less-structured task ( Vocah u­ lary) or use problem-solving strategies ( Block Desig n) when you use this short form. For the extra time involved , you w i l l get more valuable cl inical information by usi ng Vocabulary and Block Design in a two-subtest short form ( or in a longer short-form combination ) . I recommend that you consider using Vocabulary and Block Design in a shon form i f you need to use one.

Comment on Short Forms of the WISC-I1I (and Short Forms of Other Tests as Well) Short forms save time and are useful screening devices . but they have many disadvantages. First , you usually obtain less stable lQs with short forms than you do with the fu l l battery of subtests. Second , you lose information about cognitive patterning (that is, the pattern of strengths and weaknesses and the pattern of variabil ity among subtest scores) . Third , you lose the opportunity to observe the exami nee's problem-solving methods over a range of situa­ lions . Fou rth , you lose information about nonverbal abil ity when you admi nister short forms composed of Verbal subtests only, and you lose information about verbal abil ity when you admi n ister short forms composed of Perfor­ mance subtests 6 nly. Final ly. the internal consistency rel ia­ bility of the IQ is diminished when you el iminate subtests. If you are think ing about using a short form, weigh the time �aved againsl lh val idiry l ost . In add ition . comidEf what k i nd of decision you will make on the basis of the short-form scores. The most efficient testing strategy for a particular situation will depend . i n part . on the goal of the eval uat ion - whether it is for a general assessment of intel­ l igence. classification , selection , or screening. Even when you adm inister all of the subtests, the IQ you obtain on any inte l l igence tests is merely an estimate of the abilities possessed by a ch ild . When you use a small num­ ber of subtests. the esti mate may be far less adequate than that provided by the Full Scale. Additionally, educational and clinical situations call for more. rather thall less,

extensive cognitil"e evalllation. Consequently. the Full Scale should be adminisfered 10 maximi�e the diagnostic information \'011 can obtain and 10 minimize placement errors. I encourage YOU 10 administer the Full Scale. unless there is some compelling reason to administer a short form. Incl uded among these reasons would be situations in which the child was ready to quit test ing or the physical capa­ bil ities of the examinee made some of the subtests inappro­ priate. I do not recommend short fo rms for an." placement . edllcationa l . o r clinical decision -making pllrpuse.

1 07 1

ClHOOSI NG B,ETW E E N T H E W i Se-I I I

AIND T H E W P'PSI-R A N D BETW EEN T E WISC I I I A N D T H E WAIS-R -

Thle WISC-III overlaps w ith the WPPSI - R for ages 6-0-0 to 7-.3- 1 5 and wito the WAI S-R for ages 16-0-0 to 16- 1 1 -30. Thle overlap in ages between the WISC-II1 and the WPP SI-R and between the WISC-III and the WAI S-R is eSlPe ially helpful in retest situations . For example, you ca.n r te t a chi ld first administered the WISe-III at age 6 ye:ars with the WPPSI-R at any time du ring the next 15 mlonth s . Similarly, you can retest a 16-year-old adolescent w 0 i nitiall y was given the WAI S - R with the WISC-IlI up umtil his or her seventeenth b i rthday. However, because the WISC - I l l and t he WPPSI-R share many items in common, thle two tests are not truly independent i n struments . In the overlapping age ranges, Atkinson ( personal com­ muni ation . January 1 992) compared the WISe-III w ith the WPPSI-R and the WISe- I I I with the WAIS-R on sev­ eral criteria, including mean subtest rel iabil ity, F u l l Scale re l i abil ity, mean subtest floor, mean subtest ceil ing , item gradient (refers to number of items needed to reach the mean and the relationship of raw seore points to scaled­ sc ore points) , Full Scale floor, and Ful l Scale cei ling. H i s re-commendations are d iscussed below.

WISC-lII vs. WPPSI-R A gE con sid m ions should be taken into account in evaluat­ ing the choice of either the WISC-I11 or WPPS I - R . a . For afW� 6-0 10 6-Jl, the W ISe-III and W PPSi-R are comparable in all respects, except for measu res of item grad ient s . (tern gradient statistics suggest the fol lowing: • the WPPSI-R is a beller choice for children with be/ow-a I'erage ability • the WISC-1lI is a beller choice jor children with above-a verage ability • either test is adequatefor children with average ability

b. For ages 7-0 to 7-3 , the WISC-III has superior sub­ test rel iabi l ities, h igher subtest ceilings, and better item gradients both above and below the mean. In other respects the two tests are comparable. Therefore, •

the WISC-1lI is a beller choice for all children at 7-0

10 7-3

The fol lowing example i l lustrates how you can obtain a more thorough sampling on the WPPSI-R than on the W ISC -I II for a 6-year-old child with below-average abi l ity. To obta, n a scaled score of 5 on Information. a 6-year-ch i ld

1 07 2

AP'PENDIX I

needs a raw score of 16 on the W PPSI-R , but only a aw score of 2 on the W ISC-III . Atk i nson's recommendations are summarized i n the fol ­ lowing chart :

Recommendations for Selecting W I SC-I I I or WPPSI-R A biliry

Ages 6-0 to 6- 1 1 7-0 to 7-3

level

Below average

Average

Above average

WPPSI-R W iSe-III

either test Wise-III

WISC-III WIse-III

These recommendations d i ffer somewhat from those presented in the WISC-III manual . The WISC-II1 manual (page 8) recommends that in the overlapping ages you use the WISC- I I I for childre n w ith average or above­ average abil ity and the WPPSI-R for children with below­ average ability. I suggest that you fol low Atk i nson's recommendation s .

WISC-III vs. WAIS-R For ages 1 6-0 to 16- 1 1 , the WISC-I I I , in comparison to the WAIS-R, has better subtest rei iabilities, lower subtest floors, better item gradients below the mean, a lower Full Scale IQ floor, and a higher Full Scale IQ ceiling . There­ fore , Atkinson noted that •

the WISC-/11 is a beller choice than the WA IS-R

At kinson's recommendations are summarized in the fol­ lowing chart :

Recommendations for Selecting W I S C - I J I or WAIS-R Abilit\ level Ages

1 6-0 to 1 6- 1 1

Beloll' average

A verage

AbOl'e a verage

Wise-III

Wise- I I I

W ise- I I I

T h e following example i l l ustrates how you c a n obtain a more thorough sampl ing on the W I SC - I 1 I than on the WAIS-R for a 16-year- , 8-month-old adolescent with below-average abil ity. On I n format ion , to obtain a scaled score of 5, the adolescent needs a raw score of 14 on the WISC-III but only a raw score of 4 on the WA IS-R . This recommendation differs from that presented in the WISC - I I I manua l . The WISC - I I I manual recommends that you use the WISC-I II for adolescents w ith be low-average abil ity. but makes no recommendation for adolescents with

WECHSLER INTELLIGENCE SCALE FOR CHILDREN- I I I ( W ISC-III)

average or above-average ability. I suggest that you fol low Atkinson's recommendation s .

Comment on Choosing the WISC-III or WPPSI-R and WISC-III or WAIS-R The previous recommendations were based on inte rnal psychometric data. The issue of validity still needs to be addressed . In the fi nal analysis, the choice of a test in the overlapping ages should depend on the val idity of the inferences that you can make from scores on it. To this e nd , val idity studies that compare the WISC-III with the WPPSI-R and the WISC-II I with the WAIS-R in their overlapping age ranges, using samples of both normal and exceptional childre n , would be helpfu l .

A D M I N I STE R I N G T H E W E C H S L E R TESTS TO H A N DICAPPED C H I LD R E N

For handicapped childre n , you w i l l need to evaluate the child's sensory-motor abilities before you administer one of the Wechsler tests. I f you find that the child has a visual , hearing, or motor problem that may interfere w ith his or her ability to take one or more subtests, do not administer those subtests. Table I -1 5 shows the physical abil ities that an examinee needs to take a Wechsler test . Obviously, if you give the directions orally, the child must be able to hear what you say. On most of the Performance subtests , the child must be able to see the items and use his or her hands to solve the problems. I f you want to admin ister a Wechlser test to a child with a physical d isabil ity, you w i l l need to admin ister the subtests without providing c ues to the child. If your modifications go beyond simply permitting the child to respond in a different manner or using alternative procedures to present the items, the results may be inva l id .

Verbal Scale Subtests You can administer all of the Verbal Scale subtests orally to a child who can hear. If the child cannot hear but can read , you can type the I n formation , Comprehension , Simi lar­ ities , and Vocabulary questions on cards and show the cards to the child one at a time. However, visually present­ ing the Arithmetic and Digit Span items poses more diffi­ culties because of the time l imits involved in the Arithmetic subtest and because v isual presentation of the items seems drastical ly different from oral presentation , especially with Digit Span items. There fore, you may have to omit D igit

ADMINISTERING T H E WECHSLER TESTS TO HANDI,CAF'PE[) C -1ILDREtN

1 0 73

Table 1-15 Physical Abilities Necessary and Adaptable or Subte'sts on the"! Wechsler S,cales Physical abiliry

Subresr Information Comprehension Arithmetic Similarities Vocabu l a ry

Digit Span Picture Completion Coding . Digit Symbol Picture Arrangement Block Design Object Assembly Symbol Search Mazes Sentences Animal Pegs Geometric Design

Vi sion

S S S S S

N

/-leG/rill g

Oral speech

A rm-hand use

A A

A A A A A A

W W W W W W

0

P or W

IA..

A

A

R

A

R

.'\

R

A

0

A

R

A

N

R

R

A

N

R

R

.'\

0

P or W

R

A.

N

R

A 0

W A

N

R

S R R

A.

A. ....

R

NOle. The code i s as follows: A - This a bi l i t y is requ i red for standard administration. but the sobtest is adaptable. N - This ability is not required. 0- E xa m i ne e s who are able to speak can say their answers. p- Examinees who are able to point can po i n t to their answer,. R - This a bili t y is required. Adaptation is not feasible if- thi function is ·absellt or more · t han mildly i mpa i red . $- Examinees who are able to read can be shown the question�. If the cxarnin('e cannot read , hearing is necessary. If neither the a b i l ity to read nor the ability to hea r is present. the subtest should not be admi n istered . W - Examinees who are able to write can write theIr answer, .

Span when you test deaf childre n . I f the child can not respond orally. you can accept written replies (or those typed on a typewriter or computer) to any of the Verbal Scale subtests.

by writ ing or typing. You can adapt Symbol Search by pointing to each item and hav ing the ch ild say (or type) whether the symbol is or is not in the array. Appendi xes D and M provide detailed i nstructions for administering the Performance Scale subtests to a deaf chi ld .

Performance Scale Subtests Adaptations of the Performance Scale subtests center on the child's method of responding . You can give the Picture Completion subtest only to a child w ho has adequate vision and who can describe the missing part e ither orally or in writing (or typed on a typewriter or computer) or by pointing to it. You cannot easily adapt Block Design. Object Assembly. Coding. Digit Symbol ( WA IS-R). Mazes. An imal Pegs (WPPSI-R) , and Geometric Design (W PPSI-R) for a ch ild whose arm-hand use is severely impaired . However. you can adapt the Picture A rrange­ ment subtest for a child who has an arm-hand impairment . I n this case. ask the child to tell you in what order he or she wants you to arrange the cards . This can be done orally or

Advantages o f Tho Separate Scales The div ision of the Wechsler subtests into Verbal and Per­ formance Scales is helpful in testing handicapped c h i ldre n . You c a n admi nister t h e Verbal Scale to a blind child or t o a child with severe motor handicaps . And you can administer the Performance Scale to a hearing-impaired child or to a child who has little or no speech . I f you also administer the Verbal Scale to a hearing-impai red child, you can compare the child's pe rformance on the Verbal and Performance Scales t evaluate the child's verbal deficit . However, in such

cases do not include the Verbal Scale in th e computation of the IQ; only use the Performance Scale as the best estimate of the hearing-impaired child's intellectual ability.

1 0 74

A PPENDIX I

Unknown Effects of Modifications Without empirical findings , there is no way of knowing how the suggested modifications affect the reliabi lity and validity of the scores. Yet , when you cannot use tandard procedures because handicaps prevent the child from c o m ­ prehending the instructions or manipulating the material s, you may need to use such modification s . When you use

modifications, consider the resulting score only as all ap­ proximate estimate of the score that the child migh t obtain under standardized procedures. Timed Subtests A study of the WISC-R indicated that speed of correct response on the WISC-R Picture A rrangement, Block De­ sign, and Object Assembly subtests significantly relates to childre n's chronological age and to their problem-solving ability (Kaufman, 1 979b) . Older childre n solve the tasks more quickly than younger childre n , and those who solve the problems quickly also tend to solve more problems than those who solve them slowly. Because speed plays only a limited role in enabling children below 1 0 years of age to earn bonus points, it is reasonable to administer these subtests to 6- to lO-year-old orthopedically handicapped children who are able to manipulate the materials ; you will not unduly penalize these children for failure to earn bonus points (Kaufman , 1979b ) . These findings likely hold for the WISC-III, although we need research before we can accept them without question.

ASSETS O F T H E WiSe-I I I

The WISC-III is a well-standardized test, with excellent reliabil ity and adequate concurrent and construct validity. The 1 3 subtests are divided into a Verbal and Performance section , and the test provides a Verbal , Performance, and Full Scale IQ. The division into a Verbal and Performance section is especially helpful in clinical and psychoeduca­ tional work and aids in the assessment of brain-behavior relationships. A valuable feature of the test is that all children take a l:omparable battery of subtests . For chil­ dren with sensory impairments, the Verbal Scale can be administered to bl ind children and the Performance Scale to deaf childre n . The following are assets of the WISC - I 1 I : I . Excellent standardization . T h e standardization pro­ cedures were excel lent , sampling fou r geographical re­ gions, both sexes, White, Black, H ispanic, and other chil­ dren, and the entire socioeconomic statu range. The standardization group wel l represents the nati on as a whole for the age groups covered by the test .

WECHSLER INTELLIGENCE SCALE FOR CHILDREN-III (WISC-III)

2 . Ercellent overall psychometric properties . The WISC-III has excellent reliabil ity for the IQs generated by the Verba l , Performance, and Full Scales . The few studies available suggest that the WISC-III has adequate concur­ rent and construct val idity, although we need more research to evaluate the validity of the latest edition of the test . 3 . Useful diagnostic information . The WISC-III pro­ vides diagnostic information use ful for the assessment of cognitive abil ities of elementary- and high-school age chil­ dren who are functioning within four standard dev iat ions from the mean ( ± 4 SD) . It also furnishes data l ikel y to be helpful in planning special school programs. perhaps tap­ ping important developmental or matu rational factors needed for school success, especially in the lower grades . 4 . Good administration procedures. The procedures described in the WISC-III manual for administering the test are excellent . The examiner act ively probes the chi ld's responses to evaluate the breadth of the child's knowledge and determine whether the child really knows the answer. On items that require two reasons for maximum credit, examiners ask the child for another reason when he or she gives only one reason . These procedures ensure that the test does not penalize the child for fail ing to understand the demands of the questions. The emphasis on probing ques­ tions and queries is extremely desirable. 5. Good manual and interesting test materials. The WISC-III manual is easy to use ; it provides clear directions and tables . Examiners are aided by instructions pri nted in a color that differs from that of other text material . The manual provides he l pful abbreviation for record i ng [he child's responses, such as "P" for Pass. " F" for Fai l . " Q" for Query. "OK" for Don't Know, "NR" for No Response, "Inc . " for I ncomplete, and "R" for Rotation . The test mate­ rials are also interesting to childre n . 6. Helpful scoring criteria. Wechsler and The Psycho­ logical Corporation carefully prepared the criteria for scoring responses. The Similarit ies and Vocabulary scor­ ing guidel i ne s , for example, detai I the rationale for 2-, 1 - , and O-point scores. Several examples demonstrate the ap­ pl ication of the scoring principles . The scoring guidelines present several examples , and those thought to need further inquiry are indicated by a "Q . " 7 . Extensive research and clinical literalUre . There is a vast amount of research and case material on the WISC-R that you can use to interpret the WISC-II I . However, the composition of the WISC-III differs from that of [he WISC-R, particularly with regard to factor structure and subtest specificity. These differences mean that the inter­ pretations used for the WISC-R may not always apply to the WISC-III . We also need research to evaluate the valid­ ity of hypotheses derived from the WI SC-I I I .

CONCLUDING COMMENT ON THE WISC-III

L I M I TATI O N S OF THE WISC· I I I

Although the WISC-I1I is ove rall a n excel lent instrument , some problems d o exist. I . Limited floor and ceiling. The test is not applicat Ie for severely retarded or extremely g i fted children . 2 . Low reliability oJ individual subtests . Re liability co­ efficients for the individual subtests are lower than . 80 at most ages . In these cases, the scores may not be depend­ able . In addition , during the standard ization of the sca:e. The Psychological Corporation d id not obtai n test-retest scores for the Coding and Symbol Search subtests for each age level of the test; therefore, we do not know the accuracy of the rel iabil ity estimates (based on adjacent ages) for t he ages at which the two subtests were not readministered . 3 . NonuniJormity oj subtest scaled scores , Because the range of scaled scores on all subtests is less than 19 at some of the older ages, there may be some minor problems in profile analysis at the upper extremes of scores. All sub­ tests , however, do have a range of I to 1 7 for all ages.

4 . Difficulty in interpreting norms when you substitute a supplementary sllbtest Jor a regular subtest. With the norms based on only the 10 standard subtests, you have no way of knowing precisely what the scores mean when you substitute one of the supplementary subtests (Digit Span. Mazes, or Symbol Search) for a regular subtest. Make a substitut ion of this kind, therefore, only in unusual circum­ stances and label the results "tentative" when you report the scores . 5 . Possible difficulties in scoring re�'fJoll5 e.\·. Work w ith the WISC-R suggests that Simi larities. Vocabulary, and Comprehension may be difficult to score. The WISC-III manual c ites a study in wh ich there was high agreement among examiners in the scores they gave 10 these subtests (and to the Mazes subtest as wel l ) . These results are encouraging, but researchers need to replicate them . I recommend that you consult colleagues when you are hav­ ing trouble scoring responses . 6. Large practice effects on the PerJormance Scale. The large practice effects on the Performance Scale, close to I standard deviation, suggest that the WISC-III may give misleadi ng scores in retest situations. This is especially so when the retest interval is less than 9 weeks . This means that the WISC-III may not be useful to gauge change and progress in children retested within a short time. (See pages 537 and 540 for fu rther discussion of practice ef­ fects . ) Care fu ll y consider whether you want to use the WISC-III for a retest when you have previously given the test and what the retest results may mean if you use the test . 7 . Lack oj independen ce . At least 23 items overlap on the W ISC-III and WPPSI-R, primarily on the Information. .

1 07 5

Vo)cabulary, Pic:ture Completion , and Mazes subtests. This ovcerlap is ullfortl_nate for at least two reasons. F i rst, it means that the W ISC-III and W P PSI-R are not i ndepen­ de:nt parallel fo ms at the overlappi n g age levels (6-0 to 7-3 ye:ars ) . Second , it means that c h i ldren tested with the W ' PPSI-R and then with the WISe-III (or vice versa) have an advantage on the second test because of practice effects . Om the next revision of either test, The Psychological Corporation should ensure that there are no overlapping items on the two tests. 8. Problems Jor children who do not place a premium 01'1 speed. Because many of the subtests place a premium on spleed (all Performance Scale subtests and A rithmetic ) , the te:st may penalize children who are from a minority group th:at does not place a premium on speed (see Chapter 1 9) or any children who work in a s low, deliberate. and thoughtful m.anner. 9. Poor quality oJsome test materials. Danielson ( 1 99 1 ) pointed out that the Coding and Symbol Search templates are poorly constructed and may rip and disintegrate quickly. He also noted that the puzzle pieces i ncluded in the WISe-III Object Assembly subtest are of much l ighter construction than are those i n the WISC-R and have a tendency to be jarred apart as additional pieces are added to each puzzle. Care should therefore be taken when dealing with h ighly compUlsive children who may spend an inordinate amount of time trying to make sure that all of the edges touch one another. It is also very unclear why the horse in this subtesl is gray on both sides. Certainly, this has to i ncrease degree Or confmion and the assem bl y time for children who inadver­ tently turn one of the pieces over. One of the features on the WISC-R Object Assembiy subttst was that the front and back of each piece was clearly a different color. (p. 23)

CONCLU D I N G COM M E NT O N THE WISC· I I I

The WISC-I1I w i l l l i keJy b e well received by those w h o use tests to evaluate c hi ldre n's i ntellectual ability. It has excel­ lent standardization, rel iability, and concurrent and con­ struct validity, and much care has been taken to provide useful administrative and scoring guidelines . The manual is excellent, and much thought has gone into the revision . A valuable addition to the manual would have been data about the standard errors of measurement of IQ scores on the Verbal , Performance, and Full Scales at IQ levels of 70, 100. and 130 (if not others) . The W ISC-III will l i kely serve as a val uable instrument in the assessment of childre n's intelligence for many years to come.

1 076

APPENDIX I

T H I N KING T H RO U G H T H E I S S U ES

Although the Wechsler tests were by no means unique when they were developed , they did m a ke a suhstantial impact on the assessment field . Why do you t ink the Wechsler tests were so successful as c l inical assessment instruments? The Wechsler tests do not provide intercha ngeable scores. What problems do you foresee i n using d ifferent Wechsler tests w ith children in the overlapping ages where either of the two tests is appropriate? Here are some other issues that you m ight want to con­ sider. In what cases would you want to use the W ISe-III factor scores? What explanation do you h ave as to why the WISe-III Verbal Scale s ubtests are better measures of g than the Performance Scale subtests? W hat can you do to develop proper W ISC-III administrative techniques? When do you think W ISC-III short forms would be appro­ priate to use? How would the l i mitations of the W ISe-II I affect its clinical usefulness?

Question : What letter comes after T? Answe r : Well , that's easy ; V

S U M M ARY

I . The Psychological Corporation published the WISC- I l l in 1 99 1 , 1 7 years after the former edition, the WISC-R. The WISC-II I is similar to its predecessor, w ith 73 percent of the items retained, plus the original Coding subtest , which was slightly modified . Symbol Search, a supplementary subtest, is new. The WISC- I I I is appl icable to children from 6-0-0 to 1 6- 1 1 -30 years of age. Standardization of the scale was excellent and included White, Black , Hispanic , and other children. 2. The WISC - I I I provides Deviation IQs for the Verbal , Performance, and Ful l Scales (M = 1 00 , SD I S ) and standard scores for the 1 3 subtests (M 1 0 , SD 3). 3 . A lthough Wechsler objected t o the use of mental ages in the calculation of IQs, the WISC- I I I manual includes a table of test-age equivalents for the scaled scores; these are essentially mental-age scores . 4 . The internal consistency rel iabil ities o f the Verbal , Perfor­ mance, and Full Scales are excellent (average ro- of .95, . 9 1 , and . 96 , respectively) . Subtest internal consistency reliabilities range from . 69 to . 87 , and test-retest reliabilities range from . 57 to . 89 . Rel iabilities for the three scales are h igher than those for the individual subtest s . =

=

=

5 . Standard errors o f measurement are 3 . 53 for the Verbal Scale, 4 . 54 for the Performance Scale. and 3 . 20 for the Full

WEC HSLER INTELLIGENCE SCALE FOR C H I LDREN-III (WiSe-III)

Scale. Most confidence can be placed in the Full Scale. fol lowed by the Verbal Scale and then the Performance Scale. 6 . Shifts in IQ due to practice effects (after approximately a three-week interval) were about 8 IQ points higher on the Full Scale, 2 IQ points higher on the Verbal Scale, and 1 2 IQ points higher on the Performance Scale. 7. Confidence intervals based on the obtained score and its standard error of measurement should be used when you want to describe the measurement error associated with a child's current functioning. Confidence intervals based on the estimated true score and its standard error of estimation should be used when you want to describe the measurement error associated with the best long-term measure of the child's functioning relative to other children in a particular group. The former method produces symmetrical confidence intervals around the obtained score ; the latter method produces asymmetrical confidence intervals around the obtained score. The asymmetry is most pronounced for IQs furthest from the mean. 8. Studies in the W I SC-III manual suggest that the WISC-III has acceptable concurrent, criterion, and construct validity. Me­ dian correlations with measures of achievement and school grades range from the upper . 30s to the low . 70s. Correlations with the WISC-R , WPPSI-R, and WAI S-R are in the . 70s to . 90s for the Verbal , Performance, and Full Scale IQs. 9 . Based on the limited studies reported in the WISC-II I manual , the W ISC-II I tends to provide lower IQs than does the WISC-R by about 5 to 9 points, lower IQs than the WAIS-R by about 4 points , and higher IQs than the WPPSI-R by about 4 points. The various Wechsler scales do not appear to provide interchangeable IQs. 10. The Verbal Scale subtests correlate more highly with each other ( Mdn r = . 55 ) than do the Performance Scale subtests ( Mdn r = . 3 3 ) . Correlations between the Verbal subtests and the Ver­ bal Scale ( Mdn ,. . 72 ) are higher than those between the Perfor­ mance subtests and the Performance Scale ( Mdn r = .45). The Verbal subtests also have higher correlations with the Full Scale (Mdn r = . 68) than do the Performance subtests ( Mdn r = .56). I I . A factor analysis of the WISC-II I standardization data indicated that three factors account for the test's structure: Verbal Comprehension , Perceptual Organization , and Process­ ing Speed. The best measures of g are Vocabulary, Information, Similarities, Arithmetic, Comprehension, and Block Design . 1 2 . Because several WISC-I1I subtests have a n adequate de­ gree of subtest specificity, interpretation of profiles of subtest scores generally is on fi rm ground. 1 3 . You can use the Deviation IQs associated with the Verbal and Performance Scale IQs as factor scores. You can obtain somewhat purer factor scores by using (a) I n formation, Sim­ ilarities, Vocabulary, a"d Comprehension for Verbal Comprehen­ sion; (b) Picture Coml ,letion, Block Design, and Object Assem­ bly for Perceptual Organization; and (c) Coding and Symbol Search for Processing Speed. 14. The subtest scaled-score range is from I to 1 9 , but not at all ages. After II years, the ceiling level ranges from 17 to 1 9 scaled­ score points, depending on the subtest and age. =

1 077

KEY TERMS. CONCEPTS. AND NAMES

1 5 . The WISC-I 1 I does not adequately asses� the cognill \ e abil ity o f children who are either severely retarded r cxceptiun­ ally gifted . Although the WISC-III manual shows a range of F .1 1 1 Scale I Q s from 40 t o 160, the range o f the test i m re limited at some ages . 16. The WISC-Ill differs from the WISC-R in both major and minor ways. The most noticeable changes are the addition ot a new subtest (Symbol Search). a new factor ( Processing Speed) in place of the Freedom from Distractibil ity factor. and full co or illustrations for all pictures. New items have been added or changed on every subtest. and numerous administrative changes have been made. 1 7 . Developing proper administrative procedures early in your testing career is an important step in becom ing a competent clin ician . 1 8 . Beginning examiners tend to make administrative errors. These i nclude fail ing to fol low the scoring rules , to complete the Record Form properly, to adhere to directions, to probe ambig­ uous responses, and to fol low starting point and discontinuance procedu res. 1 9 . Follow the standard order of administering the subtests in all but the most exceptional circumstances. 20. To ensure that scoring is standardized at the starting and discontinuance points, credit any items failed below the starting­ point items when the child passes the starting-point item and do not credit any items passed above the discontinuance-point items when the child fails the discontinuance-point items . 2 1 . The WISC- l I l requires the use of many probing questions and queries. Give spoiled responses a score of O. 2 2 . Certain modifications in test procedure ' have been found to increase children's scores on the W ISC- R . S im i lar modifica­ tions probably will affect WISC-II I scores. Use such modifica­ tions only afler the standard administration . 2 3 . Scoring WiSe-i l l Similarities, Vocabulary, and Compre­ hension subtests requires considerable ski l l . A careful study of the scoring criteria can help to reduce scoring errors. 24. Short forms of the WISC-Il l , although practical, have serious d isadvantages. Short-form lQs may be less stable, impede profile analysis, and result in misclassifications . Do not use short forms for any placement, education, or clinical decision-making purpose. If you need to use a short form for screening purposes, follow the procedures advocated by Tel legen and Briggs to deter­ mine Deviation IQs . Table L- I I in Appendix L shows the best short-form combinations of two, three, four, and five WISC-Ill subtests. Appendixes L- 1 2 , L- 1 3 , L- 1 4 , and L- 1 5 show the esti­ mated IQs associated with these short-form combination . The best two-subtest short-form combination is Vocabulary and Block Design. 25. Although you can v iew the WISC-Ill and the WPPSI-R as alternative forms for children aged 6-0 to 7-3 years , the two tests are not independent because many items overlap. You can give both the W I SC-Ill and the WAIS-R to adolescents aged 1 6-0 to 1 6- 1 1 years. I recommend that for children aged 6-0 to 6- 1 1 you give the WPPSI-R to those with below-average ability, the WISC­ I I I to those with above-average abil ity, and either test to children

\\ illh average ah.l lty. For all children aged 7-0 to 7-3, give the W· ISC-I 1 1 . I rcwmmend that you give the W I SC-HI to all adoles­ ceilltS aged 16-0 to 1 6- 1 1 . 26. Children must be able to hear to take most WISC-Ill (and W' PPSI-R and WA IS-R) Verbal subtests, although they may use vi!sion as a substitute modal ity for some subtests. Arm-hand use is a prerequisite for most of the Performance subtests, although . Qlme adaptations are possible. Testing handicapped children is falcil itated by the arrangement of subtests into a Verbal Scale and a Pe:rformance Scale. You usually w i l l need to use special proce­ dUires, described in Appendixes D and M , to administer the Performance Scale to deaf childre n . 2 7 . The assets of the WISC-I I I include its excellent standard­ iz.ation , excellent rel iabil ity, and good val id ity ; usefulness as a di.agnostic instrument; good administrative procedures; good m.anual and interesting test materials; helpful scoring criteria; arud extensive research and clin ical l iterature with the prior e ition . 2 8 . The limitations of the WISC-II1 include limited range of IQs (40 to 1 60 ) , low reliabil ity of individual subtests, nonunifor­ m ity of scaled cores, difficulty in interpreting norms when you st.t bstitute a supplementary subtest for a standard subtest, diffi­ c ity in scoring some subtests, l arge practice effects for the Performance Scale subtests, overlap of items with the WPPS I - R , problem for children who do not place a premium on speed , and quality of some test materia l s . 2 9 . Overa l l , the WISC-IIl represents a major contribution to the field of intel l i gence testing of children. It serves as an impor­ tant instrument for this purpose.

t< EY TERMS, CO N C E PTS, AN D N A M E S

WISC-I11 Verbal Scale (p. 103 1 ) W ISC- I 1 1 Performance Scale ( p . 103 1 ) WISC-III standardization sample (p. 103 1 ) WISC-I11 Deviation I Q ( p . 1 034) W ISC-I 1 I scaled scores (p. 1 034) Prorating procedure (p. 1034) W ISC-II 1 test-age equivalents (p. 1034) Rel iabil ity of the WISC- I I I (p. 1035) Standard errors of measurement of the WISC-I 1 I (p. 1 03 5 ) Test-retest rel iabi l ity o f t h e W ISC-II I ( p . 1 036) Confidence intervals for the W ISC-III (p. 1037) Estimated true scores (p. 1037) Standard error of estimation ( p . 1038) Asymmetrical confidence intervals (p. 1039) Concurrent validity of the W I SC - I I I (p. 1040) Predictive validity of the WISC-III (p. 1042) Construct validity of the W I SC-I1I (p. 1043) W I SC-II I subtest intercorrelations (p. 1043) WISC- I I I factor analysis (p. 1 044) WISC-I I I Verbal Comprehension factor (p. 1 044) WISC-I l I Perceptual Organization factor ( p . 1 044) . W ISC-II I Processing Speed factor ( p . 1044)

1 0 78

p, �P"NDIX I

WISC - I 1 1 Freedom froI I I Di�,tract ibil ity factor (p. 1046,) WISC-III subtests as measure of g ( p . 1048) Subtest speci ficity on the WISC-III ( p . 1 048) Factor scores on the WISC-III ( p . 1049) WISC-I I I subtest scaled-score ranges (p. 1050) Starti ng-point scoring rule on the WISC-III (p. 1065 ) Discontinuance-point scoring rule on the WISC-III ( p . 1060) Repetit ion of items on the WISC-III ( p . 1066) Use of probing questions on the WISC-III ( p . 1066) Spoiled responses on the WISC- I 1 1 ( p . 1066 ) Modify i ng standard procedures on the WISC-III ( p . 1067) Short forms of the WISC - I I I (p. 1067 ) Yudin's abbrev iated procedure for the WISC-III ( p . 1068 ) Choosing between Wechsler tests ( p . 1 07 1 ) Admin istering the WISC-III t o handicapped children ( p . 1072) A ssets of the WISC-III ( p . 1074) L i mitations of the WISC-III (p. 1075 )

WECHSLER INTELLIGENCE SCALE FOR CHI LDREN-III (WISC-III)

9. Discuss the assets and limitations o f the W I SC -I I I .

REFERENCES

Danielson , G. I. (1 99 1 ) . An initial reaction to the W I SC-III . COl/llnullique, 20(4) . 2 3 . Flynn. 1 . R . ( 1984 ) . The mean IQ o f Americans : Massive gains 1932 to 1978 . Psvchological Bul/erin , 95 , 29-5 1 . Flynn, 1 . R . ( 1 987 ) . Massive I Q gains in 1 4 nations: What IQ tests really measure. Psychological Bul/etin , 101, 1 92-2 1 2 . Glutting. 1 . 1 . , McDermott, P. A . , & Stanley, 1 . C . ( 1 987 ) . Resol ving differences among methods o f establ ishing confi­ dence l i mits for test scores. Educational and Psychological Measurement , 45, 607-6 1 4 . G ranier, M . 1 . & O'Donne l l . L . ( 1 99 1 ) . Children'S W I S C - I I I scores: Impact o f parent education and home environment. Paper presented at a poster session of the 99th Annual Conven­ tion of the American Psychological Association, San Fran­ cisco, CA . Kaufman, A. S. ( 1 990) . Assessing adolescent and adult iluelli­ gence. Needham , M A : Allyn and Bacon. S i l verstein, A. B. ( 1990a) . Notes on the rel iabil ity of Wechsler short forms. Journal of Clinical Psychology, 46, 1 94-196. Silverstein, A . B . ( l 990b ) . Short forms of individual intell igence tests. Psychological Assessment: A Journal of Consulting and Clinical Psychology, 2, 3- I I . S late, 1 . R . , & Chick, D . ( 1 989) . WISC-R examiner errors : Cause for concern. Psychology ill rhe Schools, 26, 78- 84 . S l ate, 1. R . , & Hunnicutt, L . C . , l r. ( 1 988) . Examine r errors on the Wechsler Scales. Journal of Psychoeducational Assess­ ment, 6, 280-288. Wechsler, D. ( 199 1 ) . Manual Jor the Wechsler Intelligence Scale Jor Children - Third Edition . San Antonio: The Psychological Corporation. .

STUDY Q U ESTI O N S

I . Discuss the WISC- I I I . I nclude i n your discussion the fol­ lowi ng issues: standardization, Deviation IQs. test-age equiva­ lents, reliabil i ty, and val idity. 2 . D iscuss two procedures for developing confidence intervals. 3. Describe and interpret the intercorrelations between W I S C - I I I subtests and scales. 4. Describe and interpret WISC- I I I factor analytic findings. 5 . D iscuss W I S C - I I I adm inistrative considerations. 6. Discuss WISC-III short forms, including their value� and l imitations. 7. For overlapping ages, how would you go about choosing between the WISC-III and the W PPSI-R, and between the WISC-III and the WAI S- R? 8 . Identify the most important factors to consider in adminis­ tering the WISC-III ( and other Wechsler tests) to handicapped childre n .

_ APPENDIX J

-

Wise-I I I SU BTESTS

Infio rmation

Order and simpiijicalion are fhe jirsf sfeps IOward fhe masfery of a subjecf - fhe aCluai enemy is fhe unknown . - Thomas

Similarities

Mann

Arithmetic Vocabulary Comprehension Digit Span Picture Completion Codin g Picture A rrangement Block Design Object Assembly Sy mbol Search Mazes Th inking Through the I ssues Su mmary

I 79

APPENDIX J

1 080

This appendix provides information that will help you administer, score, and interpret the 13 W ISC-II I subtests. It gives the rationale, factor analytic fi ndings, reliability and correlational h ighlights, administrative suggestions, and interpretive suggestions for each subtest . The factor analytic fi ndings and reliability and correlational data dis­ cussed in this appendix are based on the WISC-III stan­ dardization sample (as presented in the WISC-III manual ) . (See Appendix I for information about the factor analysis . ) Reliabilities for the Coding and Symbol Search subtests are test-retest correlations , whereas those for the remain­ ing I I subtests are spl it-half correlations corrected by the Spearman-Brown formula . For the 1 2 subtests that are common t o both the WISC-II I and WISC-R , Table C - 1 3 in Appendix C summarizes those abil ities measured by each subtest, background factors influencing performance, impl ications of high and low scores, and instructional suggestions to improve a child's abilitie s . Table L-16 in Appendix L summarizes similar information for the WISC-III Symbol Search subtest. Therefore, both Table C- 1 3 and Table L-16 are useful references for report writing and deserve careful study. I f you are interested i n the W ISC-III Structure-of-Intellect classifications for the Arithmetic , Vocabulary, Compre­ hension , Digit Span , Picture Completion, Picture Ar­ rangement , Block Design, Object Assembly, Coding , and Mazes subtests, see Table C - 1 2 in Appendix C (page 822 ) ; these classifications are based on Guilford's model of intel­ ligence (see Chapter 3 ) . When you consider the abil ities measu red by the subtests , recognize that all of the subtcs[s require the child to pay attention , to hear, to l iste n , to understand directions, and to retain these directions in mind while solving problems. Many of the WISC-III subtests have enough subtest specificity at most ages (see Table 1 - 1 2 in Appendix l) to prov ide rel i able estimates of specific abilities, or at least to permit development of hypotheses about the underlying cognitive functions that may be measured b)4-a subtest. The subtests that have sufficient specificity at most ages are Information , Similarities, Vocabulary, Digit Span , Picture Completion, Picture Arrangement, Block Design, Symbol Searc h , and Mazes. The other subtests have ample or adequate specificity only at a few age s . Combinations o f subtests provide the best estimates of specific abilities. For example, the Verbal Scale IQ, de­ rived from a combination of five subtests, yields more accurate data about a child's verbal skills than does a single subtest , such as Vocabulary. Similarly, Coding and Sy mbol Search together provide more information about speed of processing than does either subtest alone. For each subtest, this chapter poses questions to help you

Wise-III SUBTESTS

observe and interpret the child's pe rformance. The answers to these questions will serve as a database for testing any clinical hypotheses you may form . After the test is admin­ istered , you will often have questions about the child's performance. For example, you may have questions about the quality of the child's response s , the pattern of fail ures, or how close to a solution the child was . By recording the child's performance carefully, you will be in a better posi­ tion to answer any such questions that arise later. This appendix provides suggestions for testing-of-limit s for several o f the WISC-III subtests . Testing-of-l imits should be done after you administer the entire test fol low­ ing standard procedure s . Testing-of-limits is useful for several purpose s , including fol lowing up leads about the child's functioning obtained during the standard adminis­ tration , testing clinical hypothese s , and evaluating how the child performs when he or she receives additional cues. This appendi x , unlike the WISC-III manua l , u ses the term "mental retardation" instead of the term "intellectual deficiency" to describe children who may be signi ficantly below average in their intellectual ability. "Mental retarda­ tion" is the term used in DSM Ill-R and by the American Association on Mental Retardation. Consequently, I be­ l ieve that "mental retardation" is the preferred term for describing children who are function i ng two or more stan­ dard deviations below the mean. Let us now turn to a discussion of each of the 1 3 WISC-III subtests.

I N FORMAT I O N

The Information subtest requires the child to answer a broad range of questions deal ing with factual information . The subtest contains 30 questions . Included are questions about names of objects, dates, a l iterary figure, and histori­ cal and geographical fact s . The child's age determines which item is used to start testing . Children 6 to 7 years old (and older children suspected of mental retardation) start with item I , 8 to 10 years old with item 5 , I I to 13 years old with item 8, and 14 to 16 years old with item I I . A l l items are scored I or 0 (pass-fail ) . The subtest is not timed and is discontinued after five consecutive failures . Chi ldren can usually answer the questions correctly with a brief, simply stated fact . They need only demonstrate that they know specific facts ; they need not fi nd relationships between these facts .

Rationale The amount of knowledge a child possesses may depend on h is or her natural endowment, the extent of his ? r her

1 08 1

I NFORMATION

education (both formal and informa l ) . and his or her c\. l­ tural opportun itie and pred ilections. In genera l , the I n� r­ mation subtest amples the knowledge that average ch. l­ dren with average opportunities s hould be able to acqui re through normal home and school experiences. The child's responses and comments provide c lues about the child's general range of i n format ion , alertness to the env ironment , social or cultural background, and attitudes toward scho·ol nd school - l i ke tasks ( for example, the child may say Those questlons are hard . just l i ke my teacher asks" ) . You should not necessarily i nterpret h igh scores as i n i­ cations of mental e ffic iency and competence. Children may have acqu i red isolated facts but not know how to u e them appropriately or effectively. However, intellectual dri ve may contribute to higher scores. Successful performance on the Information subtest requires memory for habitual , overlearned material (that i s , information that the child has l i kely been exposed to over and over), especially in older childre n . Thus Information provides clues about the child's abil ity to store and retrieve old data .

:

I.ntlP.rpretive SU!ggestions �

Olte the qua l it) of a child's answers.

Is the child thinking the questions through or simply gue'ssmg? Does the c h i ld give answers confidently or hesitantly? Does the ch ild give pecu l iar responses? If so, what doe:s your inquiry reveal? Are the child's answers imprecise and roundabout? i fffi c ulty in giving precise answers, such as "When it is hot"' for summer or "When it is cold" for winter, may sug.gest word-retrieval difficul t ies . Are the child's answers wordy? Overly long responses r responses fi l led with extraneous information may sug­ gest an ob essive-compulsive orientation - a child w ith this orientation sometimes feel s compelled to prove how much he r she knows . Alternative ly, excessive responses may slmply reflect the ch ild's desire to impress you . The child's �ntire test protocol , plus other relevant information , should be 'onsidered in interpreting such behavior. Does the child seem to be inhibited in making re­ sponses? Inability to recall an answer may suggest that the question is assoc iated with conti ict-laden material . For example, a child may not be able to recall the number of legs on a dog because of a traumatic experience w ith dogs. •

•

•

•

•

•

Factor A nalytic Findings The Information subtest is the second-best measure of g in the test (61 percent of its variance may be attributed to g ­ range of 50 to 69 percent in the I I age groups) . The subttst contributes substantially to the Verbal Compre hension fac­ tor ( Average loading = . 7 5 ) . Specificity is either ample or adequate at 9 of the I I ages: at ages 6 and 7 it is inadequate.

Reliabil ity and Correlational H ighlights Information is a re l iable subtest ( rxx = . 84 ) , with rel iabi l­ lty coefticie nts above . 70 at each of the I I age groups ( range of . 7 3 to . 88 ) . The subtest correlates more highly with Vocabulary ( r = . 70) and Simi larities (r = . 66) than with other subtests . It correlates moderately with both the Full Scale ( r = . 72 ) and the Verbal Scale (r = .75) and to a lesser degree with the Performance Scale (r = . 55 ) .

Administrative Suggestions The Information subtest is easy to administer. The ques­ tions are si mple and direct . Scori ng is usual ly straightfor­ ward : a correct response receives I point and an incorrect response, O . If the child gives two or more answers to a question , ask the child to choose the best answer. Answers shou ld be recorded verbat i m . Encourage a child who is hesitant to respond or guess or take a chance. Give the child c redit if he or she answers Information questions correctly at any time during the entire test .

Examine the pattern of successes and fai lures. Fail u res on easy items coupled with successes on more difficult ones mlly �lIg �i poor motivation , anx iety, temporary ine ffi ­ cie ncy, boredom , or a n env i ronment that has not been consistent . Alternatively, this pattern may indicate a prob­ lem with retrieval of information from long-term memory. When you suspect such a problem , analyze the content of the fai led items - do they deal with numerical information history, science, or geography? C ontent analysis may pro vide clues about the child's i nterests or areas that you m i ght want to inqu i re about after you complete the test . I f you think the child may have word-retrieval problems , you can use a multiple-choice testi ng-of-limits procedure ( Holmes, 1988[R] ' ) . This procedu re may help you d i ffer­ entiate deficits associated w ith word retrieval from those associated with l ack of know ledge . After you complete the test , go back to the item ( o r items) for which the child seemed to have d i fficulty i n retrieving the correct answer. Then give the child three choices . For example , for item 1 3 you might say "Was h e a king o r a n explorer o r a writer?"

�

I References fol lowed by an [ R j are cited in this appendi x : all other references are cited in the Reference section .

--

-

-----

APPENDIX J

1 08 2

B e sure t o vary randomly the position of t h e correct am;wer in the series ( that is, somet imes put the correct answer as the first choice, sometimes second , and sometimes th i rd ) . I f t h e child answers the multiple-choice questions correctly but not the open-ended questions , you may infer that he or she has a word-find i ng di ffi c ulty and not a lack of knowledge. The range of scaled scores from I to 19 at ages 6-0 to 1 3- 1 1 years aids in profile analysis for c h i ldren in this age range. However, profile analysis is somewhat hindered at ages 14-0 to 1 5- 1 1 years and 16-0 to 16- 1 1 years , where the scaled scores range from I to 1 8 and I to I 7 , respectively.

Question : What are the four seasons? Answe r : Footbal l , basketbal l , basebal l , and hockey.

Wise - I I I S U BTESTS

Factor Analytic Findings The S i m i larities subtest is the third-best mcasure of g in the test (60 percent of its variance may be attributed to g ­ range of 5 2 to 65 percent i n the I I age groups ) . The subtest contributes substantially to the Ve rbal Comprehension fac ­ tor ( Average load ing = . 7 5 ) and t o a l imited extent t o the Perceptual Organization factor ( Average loading = . 30 ) . Spec ificity is e ither ample o r adequate a t 6 ages ( 6 , I I , 1 2 , 14, 1 5 . 16) ; a t 5 ages ( 7 , 8 , 9 , 1 0 , 1 3 ) i t is inadequate .

Reliability and Correlational Highlights Similarities is a rel iable subtest ( rxx = . 8 1 ) , with rel iabi l ity coeffic ients above . 70 at each of the I I age groups ( range of . 7 7 to . 84 ) . The subtest correlates most h i ghly with Vocab­ ulary (r = . 69 ) and I n formation ( r = . 6 6 ) . It correlates moderatel y w ith the Full Scale (r = . 72 ) and the Verbal Scale (r . 75 ) and to a lesser degree with the Pe rfor­ mance Scale ( r = . 5 5 ) . =

Administrative Suggestions SI M I LA R I T I ES

The S i m ilarities subtest requires the child to answer ques­ tions about how objects or concepts are al i ke. The subtest contains 19 pairs of word s ; the child must state the simil ar­ ity between the two items in each pair. A l l children start w ith the sample item and then are administered the first ite m . The fi rst five items are scored I or 0 (pass-fail ) , and items 6 through 19 are scored 2 , I , or 0, depending on the conceptual quality of the response. The subtest is not timed and is d i scontinued after four consecutive failures.

Rationale On the S i m ilarities subtest , in addition to perceiving the common elements of the paired terms, the child must bring these common elements together in a concept to answer the question s . Thus the Similarities subtest may measure ver­ bal concept formation - the abil ity to place objects and events together i nto a meaningful group or groups . To do this, the child may need to organize, abstract , and find relationships that are not at fi rst obv ious . Although concept formation can be a voluntary, effortful process, it can also reflect wel l-automatized verbal conventions ( Rapaport , G i l l , & Schafer, 1968 ) . Performance on the Similarities subtest may be related to cultural opportun ities and interest patterns . Memory may also be involved. Success initially depends on the child's abil ity to comprehend the meaning of the task.

Note whether the child understands the tas k . On items I and 2 , g i ve the child the correct response if he or she fai l s to prov ide an acceptable response to these items . I f a child states that he or she does not know the answer. encourage the child to try to answer the quest ion , but do not press him or her unreasonably. When a child gives multiple accept­ able responses to an item , score his or he r best response. When a child gives both a correct and an incorrect response to an item, say "Now wh ich one is it?" Base your score on the answer to this question . If the child gives a I -point answer to item 6 or 7, tel l the child the 2-point answe r ; this is done to help the child give 2-point responses on l ater items . Responses to the fi rst five questions are generally easy to score, but you w i l l fi nd the scoring of items 6 to 1 9 more difficu l t . On these items (6- 1 9 ) . a conceptual response, such as a general classification , receives a score of 2 ; a more concrete response. such as a specific property of the item, receives a score of I ; and an incorrect response receives a score of 0 . A careful study of t h e scoring guide in the WISC - I I I manual w i l l help you become more profic ient in scoring Simi larit ies responses . Study carefu l l y the sample re­ sponses that fol low each item. These sample responses also l ist responses that you need to probe [as shown by a "(Q)" J . Also master the general scoring principles. which eluci­ date the rationale for 2, I. and 0 scores ( see page 84 i n the WISC- I I I manual ) .

1 08 3

ARITHMETIC

If research with the WISC-R hold for the WI Se-HI . many Similarities responses are l ike ly to be difficul t to score. For example, when a group of 1 10 psycholog ists and graduate students scored WISC-R S i milarities, Vocabu ­ lary, and Comprehension subtest responses. the rater's achieved a level of 80 percent agreement i n scoring for only 5 1 percent of the ambiguous Similarities responses (95 omt of 1 8 7 responses) (Sattler, Andres, Squire, W isely, &. Maloy, 1978). In practice, however, it is unl ikely that an one protocol would include many ambiguous responses . Scoring difficulties arise in part from the relatively few examples in the manual and the d i fficulty of establ ishin g precise criteria that apply to all responses . i ncluding idio­ syncratic ones.

Interpretive Suggestions You may gain insight into the logical character of the child-s thinking processes by study ing his or her responses to th'e items on the Similarities subtest . Observe the child's typi­ cal level of conceptualization throughout the subtest . Are the child's allswers on a concrete, functional . or abstract level? Concrete answers typically refer to qual ities of the objects (or stimul i) that can be seen or touched (apple­ banana: "Both have a skin") . Functional answers typicall y concern a function or use of the objects (apple�banana: "You eat them") . Final ly, abstract answers typically refer to a more universal property or to a common classification of the objects (apple-banana: "Both are fruits"). You can tel l , in part , whether the chi ld's response style I S concrete, functional, or abst ract by the numb"rs of ()- , ! - . and 2 - point responses . Responses that are scored 0 o r I point suggest a more concrete and functional conceptual­ ization style, whereas 2 -point responses suggest an ab­ stract conceptualization style. However, a 2 - point respon e does not necessarily reflect abstract thinking ability. It may simply be an overlearned response. For example, there may be a difference between the 2-point response "Both fruits" for apple-banana and the 2 -point response "Artistic expressions" for painting-statue. The former, even though it receives 2 poi nts, may be an overlearned response, whereas the latter may reflect a more abstract level of conceptual ability. Fu rthermore, if the child earns I point on a number of items. the child may have a good breadth of knowledge but not depth. If the child general ly earns 2 points for each correct response but responds correctly to only a few items, the child may have a good depth of knowledge but less breadth . Also note whether the child gives overinclusive re­ sponses. Overinclusi ve responses are those that are so general that many objects are i ncluded in the concept. For

e�xaJmple, the reply " Both contain molecules" t o a question alslu ing for the simil arity between an apple and a banana is ovelrinclusive because it does not delimit the part icular chalracteristics of these two objects . Overinclusive re­ �,pomses may be a subtle indication of disturbed thinking. Observe how the child handles any frustration i nduced by the subtest questions. When the child has d ifficulty ans'we ring the question s , does he or she become negativis­ ttic ,and uncoope rative, or does the child still try to answer rthe questions? A child who responds w ith "They are not alike" may be displaying negativism, avoidance of the task ,dennands, suspiciousness, or a cop i ng mechanism or may justt not know the answer. To determine wh ich of these may acc:ount for the child's response, compare the child's style of reslPonding to the Simi larities questions with his or her sty�e on other subtests . The range of scaled scores from 1 to 1 9 at all ages aids i n pmfi Ie analysis .

Q estion : In what way are an orange and a pear a l i ke? Answer: Both give me hives .

ARITH M ETIC

The A rithmetic subtest requires the child to answer s i mple to complex prob lems invol vi ng arithmetical concepts and nurnerical reasoning . The subtest contains 24 proble m s , with 5 presented o n picture card s , 1 3 presented ora l l y, and 6 presented in written form . The fi rst five and the last s i x Arithmetic items are in the Stimulus Booklet , w h ich also contains the Picture Completion and Block Design stimul i . Many of the arith metic problems are similar to those com­ monly encountered by children in school , although the child cannot use paper and penc i l to solve the proble m s . Chi ldren 6 years old (and older children suspected of mental retardation) start with item I , 7 to 8 years old w ith item 6, 9 to 1 2 years old with item 1 2 , and 1 3 to 1 6 years old with item 14. All items are timed , with items I to 17 having a 30-second time l i mit ; item 1 8 , a 45-second time l imit; and items 19 to 24, a 75 -second time l imit . I tems I to 18 are scored l or 0 and do not have time-bonus points. items 1 9 to 24 are scored 2. I , or 0 (a score of 2 includes a I -poi nt bonus for answering the item correctly with in the fi rst 10 seconds ) . The subtest is discont inued after three consecu­ tive failure s . Arithmetic subtest problems test various ski l l s . Prob­ Icms I and 2 requ ire direct counting of discrete objects .

1 084

APPE N D I X J

Wise-III SUBTESTS

WELCOME TO LAUIJDR{ 101 • THE FIRST TH IIJ� vJE'RE .. lIH­ �t>..NDW P\KI-\U< '

2·7 D R A B B L E reprinted b y permission o f UFS. I nc .

problem5 w ith pencil and paper, t h e fai lure i s not alsscKiated with Lack of arithmetical knowledge ; the errors mayl be associated with attention or concentration diffi­ c.:ult.ies that inhibit mental computation . I f the child fails the i'terms in both situation s , the fai l u res more l i kely reflect di ffilculties with a rithmetical know ledge, although attention and concentration difficulties may be i nt erfering w ith the chihd's abil ity to solve written arithmetic p roblems. I nspect the written work to see whether the child misal igns num­ bers, sequences computational steps incorrectly, or has poo'r mastery of basic arithmetical operat ion s . I f the child mis.aligns numbers while working, spatial difficulties may be i.ndicated. The information you obtai n from testing-of-limits may help you to differentiate between failures due to temporary inefficiency and those due to l i mited knowledge. Success­ ful delayed performance, for example, may i nd icate tem­ porary inefficiency or a slow, painstaking approach to problem solvi ng. During testing-of-l imits, note whether the child passes or fai l s the items . Of course, do not give the child credit on the test for answering any items cor­ rectly during testing-of-limits. The range of scaled scores from 1 to 19 at ages 6-0 to I S- I I years aids in profile analysis . However, from 1 6-0 to 16- 1 1 years, profile analysis is somewhat h indered because the scaled scores range from only 1 to 1 8 .

t:he

Question : I f I cut a pear i n thirds , how many pieces will I

have? Answer: One. Question : (Testing-of-l imits) A re you sure I will have only one piece? Answer: Yes , and I w i l l have the other two piece s .

APPEN DIX J

1 08 6

VOCABU LARY

The Vocabulary subtest requ i res the child to l isten as the examiner reads words al oud and then to define the words. The subtest contains 30 words arranged in order of increas­ ing d i fficulty. The child i s asked to explain the meaning of each word ( for example, "What is a ?" or "What does mean?" ) . Children 6 to 8 years old (and older chil­ dren suspected of mental retardation) start with item 1 , 9 to 10 years old with item 3 , 1 1 to 13 years old with item 5, and 14 to 1 6 years old with item 7 . All items are scored 2, 1 , or O. The subtest is not timed, and it is d i scontinued after four consecutive failures. __

__

Rationale The Vocabulary subtest , a test of word knowledge, may tap cognition-related factors - including the child's learning ability, fund of information , richness of ideas, memory, concept formation , and language development - that may be closely related to his or her experiences and educational e nv i ronments . Because the number of words known by a child correlates with his or her abil ity (0 learn and to accumulate information , the subtest provides an excellent estimate of intel lectual abil ity. Performance on the subtest is stable over time and relatively resistant to neurological deficit and psychological disturbance (Blatt & Allison, 1968 ) . Scores on Vocabulary therefore provide a llseful index of the child's general mental abil ity.

Factor Analytic Findings The Vocabulary subtest is the best measure of g in the test ( 62 percent of its variance may be attributed to g - range of 5 1 to 7 1 percent in the 1 1 age groups) . The subtest contrib­ utes substantially to the Verbal Comprehension factor (Av­ erage loading . 82 ) . Specificity is either ample or ade­ quate at 9 of the I I age groups ; at ages 9 and 10 it is inadequate . ==

Reliability and Correlational Highlights Vocabulary is the most rel iable subtest (ru . 87) in the Verbal Scale. Rei iabil ity coefficients are above . 70 in each of the 1 1 age grou ps (range of . 79 to . 9 1 ) . I t correlates more h ighly w ith Information (r . 70 ) , Similarities (r == . 69 ) , and Comprehension ( r . 64) than w ith the other subtests. . 74) and It correlates moderate l y w ith the Full Scale (r ==

==

==

==

the Verbal Scale (r Performance Scale

==

(r

WiSe-III SUBTESTS

. 78) and to a lesser degree with the . 56).

==

Administrative Suggestions Be sure to pronounce each word clearly and correctly. Be especially careful about how you pronounce the words, because you are not allowed to show the words to the child or to spe l l the m . When you suspect that the child has not heard a word correct!y, have the child repeat it to you . If the child heard the word incorrectly, say the word again . Care­ ful l y record the child's definitions. The scoring system ( 2 , I , or 0 for all items) considers the quality of the response. Award 2 points for good synonyms, major uses, or general classification s , and award one point for vague responses , less pertinent synonyms, or m inor uses. Do not consider the child's elegance of expression i n scoring the response. Vocabulary is one of the more d i fficult subtests to score, and it may not always be easy to i mplement the scoring criteria in the WISC-III manual . I n the study by Sattler et al . ( 1978), 80 percent of the raters gave the same score to only 38 percent of 352 ambiguous WISC-R Vocabulary responses. Probi ng borderline responses and studying carefu l ly the scoring guideli nes i n the W ISC-III manual will help you resolve some of the scoring problems that arise as you give the subtest . Do the best job possible w ith the guidelines give n i n the WISC-III manual . Consulting with a colleague may be helpful i n scoring ambiguous responses. When young chi ldren (or older children who may be mentally retarded) give a 0- or I -point response to the first word of the Vocabulary subtest, tel l them the 2-point an­ swer. Use this procedure, which is designed to encourage 2-point responses , only on the fi rst ite m . Probe responses that suggest regional isms o r slang ( for example, "Give me another meaning for "). The scoring guidel i ne s following each item i n the W ISC-III manual l ist many responses that you should query [as shown by a "(Q)"] ; study these guidelines carefully so that you can recognize responses that need to be probed. The nature of the response should determine whether the i nquiry occurs during or after the standard administration . For example, if the answer clearly define s a homonym of the test item, repeat the question by saying "What else does mean?" However, if the response i s possibly indica­ tive of a thought d isorder, delay probing until you have completed the test. During the testing-of- l imits phase, you might say "To the word you said . Tel l me more about your answer." __

__

__

__

1 087

VOCABU LARY I nt e rpre t i ve Suggestions

The fol lowi ng guideline� are useful for observ i ng antd eval uat ing Vocabul a ry responses (Tay lor, 1 96 1 ) . Write down all of the chi ld's responses , whether they are correct or not . Note whether the child is defin itely fam i l iar w ith tbe word or only vaguely famil iar w ith i t . If the child ex plains a word , is the explanat ion precise and brief, roundabout . or vague and lengthy') Are the child's responses objective. or do they relate to personal e xperiences ') Note whethe r the ch ild confuses the word w ith another one that sounds l i ke it . I f the child does nO! know the meaning of a word , does he or she guess? Does t he child readily say " I don't know" and shake off further demand - , or does the child pause, ponder, or think aloud about the item? If you show the child the words duri ng testing-of­ limits, note whether seei ng the printed word helps the child . Watch for possible hearing difficulties by l istening carefu l l y to how the child repeats words. Has the child heard the words correctly or with some distortion? Note how the child expresses himse l f or hersel f. Does the child fi nd it easy or d i fficult to say what he or she means') Does the child have mechanical d i fficult ies pro­ nounc ing words properly') Docs the child seem u ncerta in about how best to express what he or she thi nks') Does the child lise gestures to illust rate his or her stat ' m nt.s or even depend on them exclusively') •

·

•

•

•

•

I Tf-II N K Tf-IE T E A U l E R 15 MAD AT YOU F O R N OT DO I N G YOU R HOME W O R K . .

PEAN UTS reprinted by permission of U F S . I nc .

.' Note abo rhe content of definition s . Are the words chclsen synonyms for the sti m u l us word (thief: ''A bur­ glalr" ) , or do they describe a n action (thief: "Takes stuff")? DOles the ch i ld describe some particular feature of the obj,ect (donkey : "It has fou r legs" ) , or does the child try to fit ilt into some category ( donkey : "A l iving c reature that is kept in a barn")') . Note any emotional overtones or reference to personal experiences (alphabet : "I hate to write" ) . �he child's responses to the Vocabulary subtest may rev eal something about his or he r language skill s , back­ ground , cultural milieu , social development , l i fe experi­ ences, responses to frustration , and thought processe s . See if you can determine the basis for incorrect responses . It is important to distinguish a mong guesse s , c lang associations (th t i s , responses that appear to be based on the sound of the stimulus word rather than on its mean i n g ) , idiosyn­ cratic associations, and b izarre associations. Whenever a child gives pecu l iar response s , mispronounces words , or ha peculiar inflections, i nquire further. You can occasion­ ally see language disturbances in the word definit ions of children with schizophre n ia or with other severe forms of mental disorder. The range of scaled scores from I to 19 at all ages aids i n p r fi le analysis . Quest ion : What is a chisel? Answe r : When you are cold you get the chise l s . ( Fl umen & Flumen, 1 979)

S�E SAYS S � E M Ar f-1 AV E

T O R E SORT TO CASTI GATION

'

Tf-IEY CAN T DO SOMETH I N G TO YO U I F YOU D O N ' T KNOW W f-1 AT I T M E AN S .

APPENDIX J

1 088 C O M P R E H ENSION

The Comprehension subtest requi res the child to ex plain situations , actions, or activities that relate to events famil­ iar to most childre n . The questions cover several content areas, including knowledge of one's body, interpersonal relations, and social more s . The subtest contains 1 8 ques­ tions . A l l children start w ith the fi rst item. A l l items are scored 2 , I , or O . The subtest is not timed, and it is d iscontinued after three consecutive fai l u res.

Rationale On the Comprehension subtest , the child must understand given situations and prov ide answers to specific problems. Success depends on the chi ld's possession of practical in­ formation , plus an abil ity to draw on previous experiences. Responses may reflect the child's knowledge of conven­ tional standards of behavior, extensiveness of cultural op­ portunit ies , and level of development of conscience or moral sense. Success suggests that the child has social judgment , common sense, and a grasp of social conven­ tionality. These c haracteristics imply an ability to use facts in a pertinent, meaningfu l , and e motionally appropriate manner. Success is based on the ab i l ity to verbal ize accept­ able actions, as wel l .

Factor Analytic Findings The Comprehension subtest is a good measure of g (50 percent of its variance may be attributed to g range of 3S to 63 percent in the I I age groups). The subtest contributes substantially to the Verbal Comprehension factor (Average . 6 8 ) . Spec ific ity is adequate at 6 of the I I age loading groups (6, 8 , 9 , 1 0 , I I , 1 2 ) ; at 5 ages (7, 1 3 , 14, I S , 16) it is inadequate. -

=

WISC-III SUBT ESTS

vided in the manual . In the Sattler et al . ( 1978) study, 80 percent of the raters gave the same score to only 49 of the 187 ambiguous W I SC-R Comprehension responses . On the first ite m , tell the child the correct 2-point re­ sponse if the child gives a less adequate response (that i s , a response scored I or 0) . This procedure is meant to encour­ age the child to give 2-point responses and is al lowed only on the first item . On the eight items (2 , 6 , 7, I I , 1 2 , 1 5 , 1 7 , and 1 8 ) that require two ideas for ful l c redit ( 2 points ) , ask the child for a second idea when he or she gives only one correct idea , so that you do not penalize the child automat­ ically for not giving two reason s . However, on the other items, for which an adequate one-idea answer receives 2 points, do not probe obvious I -point responses i n an at­ tempt to improve the child's score. The most complete or best response receives a score of 2; a less adequate response, I ; and an incorrect response, O. Carefully study the examples followi ng each item in the manual so that you w i l l know wh ich response types need further inq u i ry [these are labeled "(Q)"J . The examples indicate that you should query many 0- and I -point re­ sponses. I f, in response to your query, the child alters his or her response, score the response given to your query rather than the initial response. Additional queries offer you an opportunity to evaluate more thoroughly the extensiveness of the child's knowledge. When a child gives unusual responses, ask him or her to explain the responses . Although your inquiry may give you insight into what the child is thinking, do not routinely inquire after every response. You can conduct an extensive inquiry as part of testing-of- l i mits after you complete the test . Record the child's responses verbatim during the initial presentation of the items and d u ring the inquiry phase so that you have a complete record with which to evaluate the responses .

Reliability and Correlational Highlights Comprehension is a reasonably reliable subtest . 7 7 ) , w ith rel iabil ity coeffi c ients above . 70 at all ( ru ages (range of .72 to . 85 in the I I age groups) . The subtest correlates more highly with Vocabulary (r = . 64) , Sim­ . 5 6) than with the ilarities (r = . 5 9 ) , and Information (r other subtests. It correlates moderately with the Full Scale (r = . 64) and the Verbal Scale (r = .67) and to a lesser degree with the Performance Scale (r = .49) . =

=

Administrative Suggestions The Comprehension subtest is di fficult to score because children may give responses that differ from those pro-

Interpretive Suggestions Responses to the Comprehension questions may provide valuable information about the child's personality style, ethical val ue s , and soc ial and cultural background . Unl ike the Information questions , which usually e l ic it precise answers, the Comprehension questions may elicit more complex and idiosyncratic replies. Because the questions involve j udgment of soc ial situations, answers may reflect the child's attitude s . Some responses reveal understanding and acceptance of social mores, whereas others reveal understandi n g but not acceptance of social mores. A child may know the right answers but not practice them . Some children may mai ntain that they do not have to abide by

1 089

DIGIT SPAN

soc ial conventions , bel ieving that such matter� d o not per­ tain to them personally. A child's repl ies can reveal initiative. sel f-reliance, i de ­ pendence, sel f-confidence. helplessness . indec isive ess . inflexibil ity, and other traits . For example. a child with a dependent personality style may say that he or she would seek help from his or her mother or other when faced itb problem situations. Repl ies to question 8 , which asks th,e child what should be done if a much smaller child starts a fight w ith him or her, may reveal independence. manipula­ tive tendencies, na'ive perceptions of problems, coopera­ tive solutions, host i l ity, or aggression ( Robb. Bernardom . & Johnson , 1972 ) . Note h o w the child responds t o the questions (Taylor. 1 96 1 ) : D o the chi ld's fail ures indicate misunderstanding of the meaning of a word or the impl ication of a particular phrase? Does the child give complete answers or just part of a phrase? Does the child respond to the entire question or only to a part of it? Does the ch ild seem to be objective. seeing various possibil ities and choosing the best way? I s the child indec isive - unable to come to fi n n answers? A re the chi ld's responses too quic k , indicating fa ilure to consider the questions in their entirety? Does the c h i l d recog n ize when his or her ilnswcr� are s u ffic ient? •

•

•

•

DltGlT S PAN :-he' Digit Span �.ubtest , a supplementary subtest , requires t ne child to repeat a series of digits given orally by t he exam iner. The subtest has two parts: Digits Forward , whilch contains series ranging in length from two to nine d igilts , and Digits Backward , which contains series ranging i n length from two to eight digits . There are two series of digilts for each sequence length . D igits Forward is adminis­ tere:d first . fol lowed by Digits Backward . The subtest is untilmed . )igit Span is not used i n the computation of the IQ when the five standard Verbal Scale subtests are administered. All children start w ith the fi rst trial of Digits Forward , and with the sample item of Digits Backward after Digits Forward is completed . A l l items are scored 2 , I , or O . On both Digits Forward and D igits Backward , the subtest is discontinued when the child fail s both trials on any one ite m . Although D i g i t Span is a supplementary subtest, admin­ iste ring it may give you useful diagnostic information . Considering the small investment of time and energy re­ quited to give it , I recommend that you administer Digit Spa n routinely.

•

•

•

Because Comprehension requires considerable verbal expre ssion , the subtest may be sensitive to mild language impai rments and to disordered thought processes. Be alert to language deficits (such as word-finding difficulties) , circumstantial or tangential speech , or other expressive di fficulties . The range of scaled scores from I to 19 at ages 6-0 to I S- I I years aids in profi le analysis. However, at 16-0 to 1 6- 1 1 years , profile analysis is somewhat hindered because the scaled scores range only from I to 1 8 .

Quest ion : Why should chi ldren who are sick stay home? Answer: To take their antibionics . (Fl umen & Flume n , 1979)

Digit Span is a measure of the child's short-term auditory memory and attention . Performance may be affected by the chii d's abi l ity to rel a x , as a child who is calm and relaxed may ach ieve a higher score on the subtest than one who is excessively anxiou s . The task assesses the child's abil ity to reta in several elements t hat have no logical relationship to one another. Because the child must recall auditory infor­ mation and repeat the information orally in proper se­ quence, the task also involves sequenc ing . D igits Forward primarily involves rote learning and memory, whereas D ig its Backward requires transforma­ tion of the stimu l u s input prior to responding. Not only must the child hold the mental image of the numerical sequence longer ( u sually) than in the Digits Forward se­ quence, but he or she must also manipulate the sequence before restati ng it. High scores on Digits Backward may indicate flexibil ity, good tolerance for stress, and excellent concentration . Digits Backward involves more complex cognitive processing than does D igits Forward and has h igher loadings on g than does D igits Forward (Jensen & Osborne, 1979 ) . Because of differences between the two tasks, it is useful to consider D igits Forward and Digits Backward sepa­ rately. Digits Forward appears to involve primarily se-

1 090

APPEN D I X J

quential processing, whereas Digits Backward appears to involve both planning ability and sequential processing . Additional ly, Digits Backward may involve the abil ity to form mental images and the abil ity to scan an internal visual d isplay formed from an auditory stimul u s . However, more research is needed to support the hypothesis about the role of visual ization in D igits Backward performance.

WiSe-III S UBTESTS

good record can help you evaluate the child's performance. A child who consistently misses the last digit in the fi rst series and then successfully completes the second series differs from one who fails to rec a l l any of the digits in the first series but successfully completes the second . Similarly, a child who responds to the sequence 3-4- 1-7 with "3-1 -4-7" is quite d i fferent from the ch ild who says

"9-8-5-6 . " Factor Analytic Findings The Digit Span subtest is a fair measure of g (26 percent of its variance may be attributed to g - range of 18 to 35 percent in the I I age groups ) . The subtest contributes minimally to the Verbal Comprehension factor (Average loadi ng . 34 ) . Specificity is ample at a l l ages . =

Reliability and Correlational Highlights

Digit Span is a reliable subtest (rxx = . 85 ) , with rel iabi lity coefficients above . 70 at each age (range of . 79 to . 9 1 in the I I age groups). The subtest correlates more highly with A rithmetic (r = .43) than with any other subtest . It has a low correlation with the Full Scale (r = . 43 ) , Verbal Scale (r = . 42 ) , and Performance Scale (r . 35 ) . =

Administrative Suggestions Be sure that the child cannot see the digits in the manual or on the Record Form. Read the digits clearly at the rate of one per second , and drop your inflection on the last digit in the series . Practice reading speed with a stopwatch . Never repeat any of the digits on either trial of a series during the subtest proper. Always administer both trials of each series . Give the child c redit for each trial that he or she passes. On Digits Backward, if the child passes the sample two-digit series (on either the first or the second trial ) , proceed to the two­ digit series i n the subtest p roper. I f the child fails the sample series, read the specific d irections in the manual that explain how the series should be repeated . Whenever there is any doubt about the child's auditory acuity, an audiological examination should be requested . Because this subtest contains no cues (that is , you only present several random series of digits) , hard-of-hearing exam­ inees may be espec ially prone to fai lure. You can record the number of digits correctly recalled i n each series b y placing in the Record Form either a mark designating a correct answer above each digit correctly recalled or a mark designating an incorrect answer on each digit missed . An even better procedure is to record the exact sequence given by the child in the available space. A

The scoring system does not distinguish among fai lure patterns. For example, a child who misses one digit i n the eight-digit sequence obtains the same score as a child who misses all e ight digits , even though the second chi ld's performance is more inefficient than the fi rst chii d's performance - perhaps because the second ch i ld had lapses in attention associated with anxiety or other factors.

Interpretive Suggestions Observe whether the child's fai lures involve leaving out one or more digits, transposing digits , interjecting i ncorrect digits , produci ng more digits than were give n , or giving a series of digits in numerical order ( for example, 6-7-8-9 ) . The child who recalls the correct d igits but i n an i ncorrect sequence is more l i kely to have a deficit in auditory sequen­ tial memory than in auditory memory. The child who fails the fi rst trial but passes the second trial may be displaying a learning-to-Iearn pattern or a need for a warm-up to achieve success . Consider the fol lowing questions: Is the chi ld's performance effortless, or does the child seem to use much concentration? Does the child view the task as interesting, boring , or di fficult? Does the child notice his or her errors , or does the child think that his or her answers are al ways correct? Does the child understand the d i fference between Digits Backward and Digits Forward? Are the errors the child makes on Digits Backward similar to or d i fferent from those he or she made on Digits Forward? As the Digits Backward series proceeds , does the child become stimulated and encouraged or tense, anxiou s , and frustrated? Does the child do much better on Digits Forward than on Digits Backward? ( I f so, the child may be overwhelmed by the more complex operations required on Digits Backward . ) Does the child make more errors o n Digits Forward than on Digits Backward? (This may mean that the child sees D igits Backward as more of a challenge and there fore •

•

•

•

•

•

•

•

1 09 1

D I G IT SPAN

Table J-I Median N u m ber of Digits Recalled on Digits Forward and Digits Backward by Age Median For\\'ard

Backward

6

5

3

7

5

3

8

5

9

6

3 4

Age

10 I I 12

() 6 6

4 4 4

6

4

14

6

15

7

5 5

16

7

5

6

4

13

A l l ages

Table B . 7 (page 268 of the W I S e - I I I manual) shows the eX.ten t to which child ren recalled more d igits forward than batckwards and v ice versa . In all age groups and in the total sample, chi ldren recalled more digits forward than back­ ward (Mdn difference = 2 at 10 of the I I age groups and in the total sample. except at age 1 5 , where Mdll differ­ I ) . Thus , raw score differences of 3 points (or e Ice mlore ) between Digits Forward and Digits Backward may be considered noteworthy. The percentage of children in the standardization group who recalled more digits back­ ward than forward was iess than 4 percent in the total grou p . about I percent at 6 to 8 years, less than 3 percent at 9 to I I years , between 4 and 6 percent at 1 2 to 14 years, 1 0 . 5 percent a t 1 5 years, and 9 percent at 1 6 years (see Table J-2 ) . The range of scaled scores from I to 1 9 at all ages aids in profi le analysis . =

Nore. There were 200 children at each age level . Source: Adapted from Wechsler ( 1 99 I [ R I l . Table B . 6 (page 267 ) .

mob i l izes more of his or her resource s - for example . giv­ ing added concentration and attention - to cope with D igits Backward . ) The child may use various methods to recall the digits . For example, the child may simply repeat what he or she has heard ; visual ize the d igits ; say the digits to himself or herse lf; use a hnger to write the digits; or group the digits. Some grouping techniques introduce meaning into the task so separate digits become numbers grouped into hundreds, tens , or other units ( for example, 3 - 1 -7 becomes three hundred seventeen ) . If the child uses grouping , the func­ tion u nderlying the task may be changed from one of attention to one of concentration . A fter you complete the subtest, you might ask the child how he or she went about remembering the numbers. If you do, record the chi ld's response. The WISe-I I I manual does not provide separate scaled scores for Digits Forward and Digits Backward . However. there are two useful tables in the W I S e - I I I manual that show how the standardization group performed on Digits Forward and on Digits Backward . Table B . 6 ( page 267 of the W ISe-III manual) shows the longest Digits Forward span and the longest Digits Backward span recalled by childre n . Across al l age groups . children had a median Digits Forward span of 6 ( range of 5 to 7) and a median Digit s Backward span of 4 (range of 3 to 5) ( see Table J - I ) .

Quest ion : Now I am going to say some more numbers . but this time when I stop I want you to say them backward s . For example, if I said " 8 -4-6-5 .. 9- 1 - 7 , " what would you say? Answer: I'd say " You've got to be kidding'" (Adapted from Flumen & Flurnen , 1979)

Table J-2 Percentage of Chi ldren i n Standardization Group Who Recalled More Digits Backward than Digits Forward by Age Age

Percell l

6

1 .0

7

1 .0

8

.5

9

I .S

10 I I

2.5 2.5

12

4.0

13

4.0

15

10.5

14

16 A l l ages

6.0 9.0

3.8

Nore. There were 200 children a t each age leve l . Source.' Adapted from Wechsler ( 1 99 1 I R I l . Table B . 7 (page 268 ) .

APPE NDIX J

1 092 PICTU RE COM PLETI O N

The Picture Completion subtest requires the child to iden­ tify the single most important missing detail in 30 drawings of common objects, animals , or people, such as a box , cat , and face. The child's task is to name or point to the essential missing portion of the i ncomplete picture within the 20second time l imit. The pictures are shown one at a time. A l l children start with the sample item. Children 6 to 7 years old (and older children suspected of mental retarda­ tion) are then given item I ; 8 to 9 years old, item 5 ; 10 to 13 years old, item 7; and 14 to 1 6 years old , item I I . All items are scored I or 0 (pass-fail) . The subtest is discontinued after five consecutive failure s .

Rationale On the Picture Completion subtest, the child must recog­ nize the object depicted , appreciate its incompleteness, and determine the missing part . It is a test of visual discrimination - the abi lity to d i fferentiate essential from nonessential details. Picture Completion requires concen­ tration, reasoning (or v isual alertness ) , visual organiza­ tion, and long-term visual memory (as the items requi re the child to have stored information about the complete figure) . Picture Completion may also measure perceptual and conceptual abi l ities involved i n v isual recognition and identification of familiar objects. Perception, cognItIon, j udgment, and delay of i mpulse all may influence perfor­ mance. The t i me l imit on the subtest places additional demands on the child . The richness of the child's life experiences also may affect his or her performance on the subtest .

WiSe-III S U BTESTS

o roups (range of . 72 to . 84 in the I I age groups). The b . subtest correlates more highly w ith Rlock DeSign (r = . 52 ) , Object Assembly (r = . 49) , and I n formation (r = .47) than w ith the other sub tests . It has low correla­ tions w ith the Full Scale (r = . 5 8 ) , Performance Scale (r = . 54), and Verbal Scale (r = . 52 ) .

Administrative Suggestions Picture Completion is easy to administer. Simply leave the booklet flat on the table and turn the cards over to show each consecutive picture. If the child has speech diffi­ culties, such as those that occur in aphasia, you can admin­ ister the subtest by having the child point to the place where the part is m issing . The child should be aware that he or she is being timed , because it is important for the child to realize that speed is expected. Usually, allowing the child to see the stopwatch is all that is necessary. However, you should not tell the child that he or she is be ing timed . The WISC-III manual indicates that, if necessary, you may give each of three guiding statements once to help the child understand the requirements of this subtest : (a) If the child names the object pictured , ask the child what is missing in the picture. (b) If the child names a part that is not on the card , ask the child what part in the picture is missing . (c) If the child mentions a nonessential missing part , ask the child for the most important part that is missing . On fi ve items (6, 1 3 , 2 1 , 23, 28) , ask the child to point to the missing part on the card i f he or she gives an ambiguous response. In other cases as wel l , whenever there is any doubt about the child's verbal or pointing response, ask the child for clarification.

Factor Analytic Findings The Picture Completion subtest is a fai r measure of g (44 percent of its variance may be attributed to g range of 32 to 47 percent in the I I age groups) . The subtest contributes moderately to the Perceptual Organization factor (Average loading = . 5 3 ) and to a lesser extent to the Verbal Compre­ hension factor (Average loadi ng = . 39 ) . These results sug­ gest that verbal reasoning may help children to detect the missing part of the pictures. Subtest specificity is either ample or adequate at 10 of the I I age groups; at age 12 it is i nadequate. -

Interpretive Suggestions As you administer the subtest, consider the following : Does the child understand the task? Does the child say anything that comes to mind, or does the child search for the right answer? When the child fai l s , does he or she fi nd fault w ith himself or hersel f or with the picture? What is the child's rate of response - for example, quick and impulsive, or slow and deliberate? . Is the child fearful of making an error, heSitant , or suspicious? . . Is the child aware of being timed? I f so, does the tlmmg make the child anxious or prompt the ch ild to change the pace of h is or her responding? •

•

•

•

•

Reliability and Correlational Highlights Picture Completion is a relatively rel i able subtest (ru = . 77) , with rel iabil ity coefficients above . 70 at a l l age

•

1 09 3

CODING • Does the ch i ld give roundabout definitions of parts (sometimes referred to as circumlocutiotls),1 [Sue re­ sponses may suggest word-retrieval d i fficulties . For e xam­ ple, to item I (correct response is "Ear") , a roundabout response would be "The thing you hear with . " ] • Does the c h i l d point excessively') A child who points excessively or respond with c i rcumlocut ions may have word-finding problems (dysnomia ) .

If the child's performance leaves any doubt about h i s o r her v isual skills, request a v isual examination. Observe whether perseveration occurs. A child displays perseveration , for example, when he or she says " Ear" for each picture portrayi ng an animal (pictures I , 3 , 5 ) . " Ear" is the correct answer for picture I , but not for the s u bse­ quent pictures depicting animal s . Comparing Picture Completion scores with those on Block Design and Object Assembly may help you d istin­ guish between v isuospatial difficulties and visual -motor difficultie s . Picture Completion is the only task on the W ISC-III Performance Scale that does not have a motor componen t . Record each incorrect response verbatim a s wel l as the t i me the child takes to make the response. The child who usually responds in less than fi ve seconds may be more impulsive, more confide nt , and , if correct , brighter than the ch ild who takes rhore time. A child who responds correctly after the ti me I imit ( for which he or she does not receive c redit) may be brighter than the child who fa ils the item even with add i t ional time . Because the pass" fail cor ing makes no provision for such qualitative factors , care­ fully eval uate individual variat ions in each case and discuss these qualitative factors in the report . Delayed correct responses may suggest temporary inefficiency, depression , or simply a slow and diligent approach , whereas extremely quick but incorrect responses may reflect i mpulsivity. A fter you administer the subtest, you can inqu i re about the c h ild's perceptions of the task : "How did you go about com i ng up with the answer?" or "How did you decide when to g i ve an answer?" Query any peculiar answers. The child's behavior during this subtest may provide insight into how the child reacts to t i me pressure. As a testing-of-l imits procedure, you can ask the child to look again at those pictures that he or she m issed . You m ight say "Look at this picture again . Before, you said that was missing . That's not the part that's m issing. Look for someth i ng else . " In some cases, you m a y ask the child t o name the picture, espe cially when he or she missed many items . The range of scaled scores from I to 1 9 at ages 6-0 to 13- 1 1 years aids in profile analysis . However, profile anal­ ysis is somewhat hindered at ages 14-0 to 15-1 1 years and __

16--0 t o 16- 1 1 years , where the scaled scores range from I to 18 and I to 1 7 , respec tively.

Question : What are 2 , 4 , and 6? Answer: That's easy ; CBS, NBC, and ABC .

CO D I N G

True Coding subtest requ i res that the child copy symbol s paiired with other symbol s . The subtest consists of two separate and distinct parts . Each part uses a sample, or key. In Coding A, the sample (or key) consists of five shapes - star, c i rcle, triangle, cross, and square. W ithin each sample shape, there is a special mark (a vertical l ine, tw horizontal l i nes, a horizontal l ine, a c i rcle, and two vertical l ines , respecti vely ) . The child must place within each test shape (which is empty) the mark that is w ithin the sample shape. There are 5 practice shapes, followed by 59 shapes in the subtest proper. In Coding B, the sample (or key) consists of boxes contai ning one of the nulnbers 1 throLigh 9 in the upper pim anu a symbol in the lower part . Each number is pai red wit h a qitferent symbol . T h e test stimul i are boxes contai ning a number i n the upper part and an empty box in the lower part . The child must w rite in the empty box the symbol that is pai red with the number in the sample. There are 7 practice boxes, followed by 1 1 9 boxes in the subtest proper. The t i me l imit for each Coding task is 120 seconds. Cod ing A is given to children u nder 8 years of age, and Coding B is given to c hildren 8 years of age and older. On Coding A, I point is g i ve n for each correct item, and up to 6 additional time-bonus points are g iven for a perfect score. On Coding B, 1 point is given for each correct item, but there are no time-bonus points .

Rationale Coding taps the child's abil ity to learn an unfami l iar tas k . The subtest involves speed and accuracy of visual-motor coordination , attentional skills, v isual scanning and track ­ i n g ( repeated visual scanning between t h e code key and answer spaces) , short-term memory or new learni ng (paired-associate learni n g of an unfamil iar code) , cogni­ tive flexibility ( i n shi ft i ng rapidl y from one pair to an­ other) , handwriting speed , and , possibly, motivation . The

APPEN DIX J

1 094

subtest also involves speed of mental operation (psych mo­ tor speed) and , to some extent, visual acuity. Suc cess depends not only on comprehend ing the task , but also on using pencil and paper skillfully. The subtest is sensitive to visuoperceptual d ifficulties. Coding B may also involve a verbal-encoding process if the child attaches verbal descriptions to the symbols. For example, a child may label a " + " symbol as a "plus sign" or "cross" and the "v" symbol as the letter " V " A child may improve his or her performance when he o r she uses verbal labels to recode the symbols . Consequently, Coding B can also be described as measuring the ability to learn combi­ nations of symbols and shapes and the abil ity to make associations quickly and accurately. Coding A can also involve a verbal-encoding process, but to a lesser degree. Coding A and Coding B thus may involve separate information-processing mode s . The speed a n d accuracy with which the child performs the task are a measure of the child's intellectual ability. At each step in the task the [child] must inspect the next digit, go to the proper location in the table, code the information distin­ guishing the symbol found, and carry this information in short­ term memory long enough to reproduce the symbol in the proper answer box . ( Estes, 1 974 , p. 745) Cod i ng thus can be conceptualized as an information­ processing task i nvolving the d iscrimination and memory of visual pattern symbols .

Wise-III SUBTESTS

correlation with the Full Scale (r = . 33 ) , Performance Scale (r . 32 ) , and Verbal Scale (r = . 29) . =

Administrative Suggestions The Coding subtest items are located in the Record Form . Administer the subtest on a smooth draw ing surface, and tear the page out of the Record Form . Be sure to put the child's name, the date, and your name at the top of the Coding Response Sheet to identify it if it is misplace d . You and the child should each use a no. 2 graphite pencil without an eraser. A child with visual defects or specific motor disabil ities may be penalized on this subtest. Generally, do not give the subtest to a child with either of these d isabilitie s . If you give it, do not count it in the fi nal score. A left-handed child also may be penal ized on the Coding subtest. If the way the child writes causes the child to cover the sample immediately above the l ine of writing, the child will have to l i ft his or her hand repeatedly during the task to view the key. I f the child is left handed, the WISC-III manual suggests that you place an extra Cod i ng Response Sheet to the right of the child's sheet and have the child work with the separate key both during the sample items and during the subtest proper.

Interpretive Suggestions Useful observational guidel ines are as fol lows:

Factor Analytic Findings The Coding subtest is a poor measure of g (20 percent of its variance may be attributed to g range of 10 to 30 percent in the 1 1 age groups ) . The subtest contributes substantial!y to the Processing Speed factor (Average loading = . 74) . Specificity is ample at 10 of the I I age groups ; at age 8 it is inadequate. -

Is the child i mpulsive or meticulous? Does the child d isplay tremor? Does the child's speed increase or decrease as he or she proceeds? A re the child's symbol marks well executed , barely recognizable, or wrong? • Do the child's symbol marks show any distortions? If the child's symbols marks do show d istortions, do the d istortions appear only once, occasionally, or each time the child draws the symbol mark? How many d i ffe re nt symbols are d istorted? Does the child draw the same symbol over and over again even though the numbers change (perseveration)? Is the child being penal ized for lack of spee d , for inaccuracy, or for both? • Does the child understand the task? Does the child understand and proceed correctly after you give an explanation? •

•

•

•

•

Reliability and Correlational Highlights Coding is a relatively reliable subtest (r(X . 79) . Rel iabil­ ity coefficients are above . 70 at 6 ages (6, 7, 10, I I , 14, 1 5 ) reported i n the W ISC-II1 manual ( range of . 70 to . 90 in the 6 age groups ) . Unfortunately, there are no rel iability coeffi­ cients shown in the W ISC-II1 manual for 5 age groups (8, 9, 12, 13, 1 6 ) because no children were retested at these age s . The subtest correlates more highly with Symbol Search (r = . 5 3 ) than with any other subtest . It has a low =

•

•

•

1 09 5

�CTURE ARRANGEMENT

Are t he ch ild's fa i lu re s due to Inadequate fc 'rm p ;ceo­ tion or to poor attent ion'! Does t he ch ild check each symbol w i th the amp�e. or does he or 'he seem to remember the symbols '> Does the child rechec k every sy mbol before m ving Of! to the next one') Doe the child pick out one number only and sk i p the others') Docs the child work smooth ly. or does he r he see m confused at ti me s ' ) I s the child aware of any erro rs'> Do the chi ld's errors occur i n some regu l ar manne r" H ow does the chi ld react to making errors'> Is the child persistent') Does the child need repeated urg i ng ') [s the child bored w ith the task') •

•

•

•

•

A nswers to the above quest ions w il l provide information about the chi ld's attention spa n . method of w rking , a d other behav iors. An i ncrease in speed, cou ple d with cor­ rect c opy ing of symbols . suggests that the ch ild is adj usting to the task wel l . A decrease in speed . coupled with i ncor­ rect copy ing of symbol s . suggests that the child may oe showi ng fatigue. Coding is pa rtic u lar l y useful for evaluating the chilJ's attent ion when you suspect attent ional difficult ies , such as in cases of learning disabil ity or afte r a head inj ury. [f other tests i nd icate that the child has adeq uate response speed a nd visual acu ity. then poor scores on Codi ng are l ike l y to be assoc iated with attel1l ional defi cits and n\)t v isuopercep­

tual d i fficulties per sc . A s low a nd del iberat e appro�ch m.ay suggest dc prc � s i ,'C fcaiU rc� . Dist ort ion o f fmllls Jllay mcan that t h e chi ld has diffi­ c u l t ies with perceptual function i n g . Ask t he child about any symbol that is pecul i a rly written to find out whether it has some symbol ic mea n i ng to him or he r. Persevc rat ion Jllay suggest neurolog ica l d i fficulties that shou ld be i nves t i gated fu rthe r. Boredom might be present w ith a bright child who does not appear (0 be challenged by t he task. The range of scaled scores from I to 19 a t all ages aids in profi le analysis.

Quest ion : What is celebrated on Thanksgiving Day" Answe r : \1y cousi n's birthday.

PIICTU RE A , R A NG E M E NT

Thle Picture A rrangement subtest requi res the child to pla1ce a series of pictures in logical order. The subtest cOJIltain 14 ser ies, or items, similar to short comic strips . Individual cards, each containing a picture, are placed in a sp(ec i fied di ar anged order, and the child is asked to re­ ar ange the pictures in the "right" order to tell a story that makes sense. The number of pictures per set ranges from th ee to i x . One set of cards is given at a time to the child. Eatch i te m i ti ed, w ith 45 seconds for items I to I I and 60 se.conds for items 12 to 1 4 . The only motor action requi red is for the child to change the position of the pictures. All children are started with the sample ite m , after which ch i ldren 6 to 8 years old (and older children suspected of mental retardation) are given item I and children 9 to 16 ye.ars are gi ven item 3. There are two trials for items I and 2. it ems I and 2 are scored 2, I, or O . I tems 3 to 14 are sc red , 4, 3 , 2, or 0, w ith 2 points for the correct arrangement and up to 3 additional time-bonus points for quick executio n . An alternative arrangement on item 14 re 'eives I poi nt only, w ith no time-bonus points . The subtest is discontinued after three consecutive fai l u re s .

Rationale The Picture A rrangement subtest measures the child's abil ity to c o mp rehend and evaluate a situation. To accom­ plish the task . the child must grasp the general idea of a story. AI though a child may sometimes use trial-and-error experimentation on the subtest , the child usually needs to appraise the total situation depicted in the cards in order to succeed . The subtest may be viewed as a measure of nonverbal reasoning that involves planning abil ity, anticipation , vi­ sual organization . and temporal sequencing. The subtest measures the abi l ity to anticipate the consequences of ini­ tial acts or situations, as wel l as the abi l ity to interpret social sit uation s . Some children may generate covert , ana­ lytica l . verbal descriptions of alte rnative story sequences to gu ide them in arranging the stimulus cards. [n such cases, the subtest may measure verbal sequenc ing proc­ esses as wel l . The capacity to anticipate, j udge, and under­ stand the possible antecedents and consequences of events is impor ant in le nding meaningful conti nu ity to everyday exper ie nl es ( B l att & All ison, 1968) .

1 096

APPENDIX J

Factor Analytic Findings The Picture Arrangement subtest is a fair measure of g (36 percent of its variance may be attributed to g - range of 23 to 46 percent i n the I I age groups). The subtest contrib­ utes minimally both to the Perceptual Organization factor (Average loading . 36) and to the Verbal Comprehension factor (Average loading = 34 ) Speci fic ity is ample at all ages. =

.

.

Reliability and Correlational Highlights Picture Arrangement is a relatively

reliable subtest

(r.o: = . 76 ) , w ith rel iabil ity coefficients of . 70 or above at all the ages (range of . 70 to . 84 i n the I I age groups) . The subtest correlates more h ighly with Block Design (r = A I ) , Information (r = .40) , Vocabulary (r = .40) , and Similarities (r = . 39) than with other subtests. It has a relatively low correlation w ith the Ful l Scale (r = 52), Performance Scale (r = .49), and Verbal Scale (r = .45 ) .

Administrative Suggestions Arrange the Picture Arrangement items from the chi ld's left to right i n the order given i n the manual. As y: a gurl ( seven pieces ) , a car (seven pieces) . a horse ( s i x p ieces). a ball (six pieces) , and a face (nine pieces) . T here i - Ol:le sample item: an apple (four pieces) . Items are given one at a time, and the pieces are present ed in the spe c i fi ed disar· ranged pattern indicated in the W ISe-III manual . Eve ry item is administered to ail child ren, beginning with the sample item and continuing with items I th rough 5 . All items are timed . The tirst ite m has a maximum of 110 seconds ; the next two items , 1 50 econds each : and the last two items, i 80 seconds each . For perfect performance, the scores are 6 points for the girl, S points each for the car and the horse, and 7 points each for the bal l and the face. Bonuses of up to 3 points are given for quick performance. The girl , car, and horse items each have a maximum score of 8 , and the ball and face items. 10. Points are also giv en for partially correct assembl ies l n all items.

Rationale The Object Assembly subtest is mainly a test of the child's skill at synthesis - putting things together to form famil iar objects . It requires visual -motor coordinat ion, w ith motor activity guided by visual perception and sensori motor feedback . Object Assembly is also a test of visual organi· zational abil ity, for visual organ izat ion is needed to pro­ duce an object out of parts that may not be immediately recogn izable. TI) solve the jigs"w puzzles, the child must be able to grasp an ent i re pattern by antic i pating the rela­ tionsh ips among its ind iv idual parts. The task requ i re some constructive abil ity as well as perceptual skil l - the child must recognize individual parts and place them cor­ rectly in the incomplete figure. Performance may also be related to rate and precision of motor activity ; persistence, especially when much t rial and error is required ; and long­ term visual memory (having stored information about the object to be formed) .

Factor Analytic Findings The Object Assembly subtest is a fai r measure of g (44 percent of its variance may be att ributed to g - range of 46 to 64 percent in the I I age groups ) . The subtest contributes substantially to the Perceptual Organization factor ( Aver­ age loading = .67) . Spec ificity is ample or adequate at 4 of the I I age groups (6. 9 . 1 3 . 1 5 ) ; at 7 age groups ( 7 . 8 . 10. I I . 1 2 , 14, 16) it is inadequate.

Re!liability

and

Correlational H ighlights

Ob1j ect Assembly is a marginall y reliable subtest (r n = . 69 ) , with rel iability coefficients above . 70 in only 5 of the I I age groups (6, 9 , 1 3 , 1 5 , 1 6 ) and between . 60 and .69 in the other age groups (range of . 60 to . 76 over the 1 1 age groups) . The subtest correlates more highly with Block Design (r = . 6 1 ) than with any other subtest. It has a somewhat low correlation with the Full Scale (r = . 5 8 ) , a moderate correlation with the Performance Scale (r = . 60) , and a low correlation with the Verbal Scale (r = .48) .

Administrative Suggestions M ake sure that the child does not see the pages of the W ISe-Il l manual that contain pictures of the correctly assembled objects. You can use the Object Assembly Lay­ out Shield, not only to set up the individual puzzle parts, but to shield the manual as wel l . Place the pieces close to the child so that the child does not have to waste time reaching for them . As in other subtests, you may have to as k the child to tell you when he or she is fi nished - "Tell me w hen you have finishe d . " You may need to give this instruc­ tion on items 2 through 5 because the directions i n the WISe-III manual do not include it . Do not give any cues to the child that i ndicate approval or disapproval of his or her performance. Because all items are given to all childre n , young school11 ed children may experience some frustration on Object Assembly. [f so , it would be useful to investigate what e ffects a discontinuance criterion would have on Object Assembly. Would it reduce young children's anxiety level without affecting the rel iabil ity and validity of the subtest? This, of course, is a proposal for research. The subtest should still be administered following the exact procedures stated in the W I Se-III manual.

Interpretive Suggestions Object Assembly is an especially good subtest for observ­ ing the child's thinki ng and work habits . Some children envision the complete object almost from the start and either recognize the relations of the individual parts to the whole or have an imperfect understanding of the relations between the parts and the whole. Others merely try to fit the pieces together by trial-and-error methods. Still others may have initial failure, followed by trial and error and then sudden i nsight and recognition of the object. Observe how the child responds to errors and how the child handles frustration .

APP E N D I X J

1 1 00 • Does the child demand to know w hat the object is before he or she constructs it , or insist that pieces are missing , or say that the object doesn't make sense ( Zim­ merman & Woo-Sam , 1985)? • I f the child has low scores, are they due to temporary inefficiency, such as reversal of two parts, which results in loss of time-bonus credits? • Does the child spend a long time with one piece, trying to position it in an incorrect location? ( I f so, this behavior may indicate anxiety or rigidity. )

A fter you admin ister the subtest, ask the child about any constructions that may be pecu liar or unusual ( such as pieces placed on top of each other ) . You can use testing-of­ l i mits procedures similar to those described for the Picture Arrangement and Block Design subtests after you have administered the entire test . For example, you can intro­ duce a series of graduated cues, such as placing one or more pieces in the correct location. Note the amount of help the child needs to complete the task successfully. The child who needs only a few cues to complete the object may have underly ing perceptual organization skills not evident during the standa rd administration of the subtest and may have better perceptual organization skills than the child who needs many cues. Another testing-of-l imits approach is to ask the child to visualize the object in his or her mind before you lay out the puzzle pieces. For example, say 'Th ink of how a horse looks" and give the child the horse ite m . See if this instruc­ tion helps the child to assemble the puzzle. The range of scaled scores from I to 1 9 at ages 6-0 to I S - I I years aids in profile analysis. However, at ages 1 6-0 to 1 6- 1 1 years , profile analysis is somewhat hindered because scaled scores range only from I to 1 8 .

Question : W hat i s gasoline? Answer: To put on the th ing what takes your temperature so it don't hurt you . ( Fl umen & Flumen, 1979)

SYMBOL S EARCH

The Symbol Search subtest , a supplementary subtest , re­ quires the child to look at a symbol (or symbol s ) and then decide whether the symbol(s) is(are) present in an array of

WISC-1I1 SUBTESTS

symbol s . Symbol Search is not used in the computation of the IQ when the five standard Performance Scale subtests are administered . The subtest consists of two separate parts . Part A is adm inistered to children 6 to 7- 1 1 years old , and Part B to children between 8 and 16 years old . I n Part A there i s one target symbol and three symbols i n the array. The child i s told to draw a slash (I) through the box labeled YES if the target symbol is also in the array. I f the target symbol i s not i n the array, the ch ild should draw a slash (I) through the box labeled NO. The target symbols usually are nonsense shapes and designs, as are the sym­ bols in the array. There are 2 demonstration ( sample) items and 2 practice items. Part A contains 45 items in addition to the 2 sample and 2 practice items. In Part B the re are two target symbols and five symbols in the array. As in Part A, the child is told to draw a slash (I) through the box labeled YES if either of the target symbols is also in the array. If neither one of the target symbols is in the array, the child should draw a slash (I) through the box labeled NO. The target symbols , l i ke those in Part A, are usually nonsense shapes and desig n s , as are the symbols in the array. There are 2 demonstration (sample) items and 2 practice items. Part B conta ins 45 items in addition to the 2 sample and 2 practice items. Some of the symbols in Part A and Part B are the same. Each part has a time l imit of 120 second s . The score o n each part i s the number o f correct items minus the number of incorrect items. There are no time­ bonus credits on either Part A or Part B.

Rationale On the Symbol Search subtest, the ch ild looks at a stimulus figure (target stimulus) , scans an array, and decides whether the stimulus figure appears in the array. The task involves perceptual d iscrimination , speed and accuracy, attention and concentration , short-term memory, and cog­ nitive flexibil ity (in shifting rapidly from one array to the next) . V isual-motor coordination plays a role, albeit minor, because the only motor movement is that of drawing a slash . Part B is more complex than Part A because there are two target stimulus figures instead of one (as in Part A) and five symbols in the array instead of three (as in Part A ) . Most o f the symbOls used i n the Symbol Search subtest will be difficult to encode verbally. However, some symbols may be verbally encoded if children attach verbal descrip­ tions to them. These include, for example, ± (plus or minus ) , L (L shape ) , > (greater than sign) , n (inverted U ) , and I- (a T on its side ) . Research is needed to learn whether children verbally encode these or other symbols and whether the encoding affects the i r performance.

1 101

SYMBOL SEARCH

As in the Coding subtesl, the . 'peed and ae,curacy w ith which the child performs the ta k are a measu e of the child's intellectual abil ity. For each item , the eh ild must inspect the target stimul us, go to t he array, look at the arr ay items and determine whether the t arget stimul s i� preserlt. and then mark the appropriate box (Y ES or NO) once he o r she makes the decision . You can thus concept ual ize Sy 1 bol Search as a task invol ving isual discrimination a.nd visuoperceptual scanning .

ful lly ; those who are compulsive and need to constantly ch(eck the stinmlus figure(s) against those in the array ; and those who are i mpulsive and fail to check the array figures ag:ainst the stimu l us figure(s ) . (Observe the chi ld's work methods. Tel l the child who stops working after the fi rst line to "Continue on the next lime . " Tel l the child w ho fails to turn the page to "Go to the nnt page . " Count these instructions as part of the 2 -minute tinne l imit . I f the child skips lines , tel l him or her to do the limes in order.

Factor Analytic Findings The Symbol Search subtest is a fair measure of g ( 3 8 p,er­ cent of its variance may be attributed to g range of 27 to 48 percent in the i I age groups) . The subtest contribu tes substantially to the Processing Speed factor (Avenge load­ . 62) and to a lesser degree to the Perceptual Organi­ ing zation factor (Average loading . 3 5 ) . Subtest spec i fic ity is ample at 8 of the 1 1 age groups ( 7 , 10, I I , 1 2 , 1 3 . 1 4 , 1 5 , 1 6 ) ; at 3 ages (6, 8, 9) i t i s inadequate. -

=

=

Reliability and Correlational Highlights

Symbol Search is a relatively reliable subtest (rxx . 7 6) , with rel iabil ity coefficients above . 70 at 5 of the 6 age groups reported (7, 10, 1 1 , 14, 1 5 ) ; the one exception is at age 6 , where the rel iability coeRicient is . 69 range of . 69 to . 8 2 in the 6 age groups) . As with Coding, rel iabil ity coefficients are not given for 5 age groups ( 8 , 9, 1 2 , 1 3 , 16) because no chi ldren were retested at theae ages . The ub­ test correlates more highly with Coding (r = . 5 3 ) than with any of the other sllbtests . It has low correlations with the Full Scale (r = . 56), Performance Scale (r = . 5 8 ) , and Verbal Scale (r = . 44) . =

Administrative Suggestions The Symbol Search subtest is in a separate booklet (WISC-III Symbol Search Response Booklet) . You and the child should each use a no. 2 graphite pencil without an eraser. Write the child's name, the date, and your name in the space provided on the Symbol Search Response Book­ let . Administer the subtest on a smooth drawing surface. A child with visual defects or specific motor disabilities may be penalized on this subtest . Generall y, do not give the subtest to a child with either of these handicaps. I f you do give it to a child with these handicaps , do not count it in the final score, even when it replaces a standard Performance Scale subtest. Other types of children also may be pe­ nalized on this subtest, including those who are unable to make quick decisions; those who respond slowly and care-

Il1lterpretive Suggestions Useful observational guidelines are as fol lows : Does the child carefully check the target symbol with those in the array? • Does the child draw the slash ( I ) mark slowly or quickly? • Does the child respond impulsively? Is the child penalized for working slowly? • Does the child make many errors? • Does the child understand the task? • Does the child work smoothly, or does he or she seem confused at times? • Does the child seem to be aware of any e rrors? •

•

Answers to the above questions may provide valuable information about attention, persistence, impulsive ten­ denc ies, compulsive tendencies, and depressive features . I t w i ll b of interest t o compare children who obtain the same score in different ways. For example, two children may get a score of 10, but one child l Ias 10 correct responses and zero errors and the other child has 20 correct responses and 10 errors . These two children l i ke l y have different styles of working. The fi rst child may be a careful and diligent worker, but unwill ing (or unable) to work quickly, whereas the second child may be a quick worker, but rather careless and impulsive. A fter the test is over, for children who make many errors, you may want to go over each item on which an error occurred. You can point to an item on which an error occurred and say "Tell me about your answer" or "Te l l me about why you marked a 'No' (or 'Yes') . " Compare t h e child's response style o n Symbol Search with that on other subtests. If there are differences, try to determine what might account for the m . Consider, for example, the nature of the tasks, the child's motivation , the child's scores on all task s , and when the tasks were administered - that is, at the beginning, middle, or end of the examination . The range of scaled scores from I to 1 9 at ages 6-0 to

1 1 02

1 5- 1 1 years aids in profile analysi s . However, at ages 16-0 to 1 6- 1 1 years, profile analysis is somewhat hindered be­ cause scaled scores range only from I to 1 8 .

APPE N D I X J

Wise- I I I SUBTESTS

tern . To succeed , a child must have visual -motor control combined with speed and accuracy.

Factor Analytic Findings Question : I n what way are a computer a nd T V alike? Answer: They both go on the blink when you need them .

The Mazes subtest is the poorest measure of g in the test ( 1 3 percent of its variance may be attributed t o g - range of 5 to 30 percent in the 1 1 age groups) . The subtest contributes minimally to the Perceptual Organization factor (Average loading . 36 ) . Subtest specificity is ample at 10 of the 1 1 age groups ; at age 1 5 it is inadequate. =

MAZES

The Mazes subtcst, a supplementary subtest , requires the child to solve paper-and-pencil mazes that differ in level of complexity. Mazes is not used in the computation of the IQ when the five standard Performance Scale subtests are administered . Mazes consists of 1 sample problem and 10 problems in the subtest proper. The child is asked to draw a l ine from the center of each maze to the outside without crossing any of the l i nes that indicate wall s . Each maze is presented separately. Children 6 to 7 years (and older children suspected of mental retardation) start w ith the sample maze fol lowed by item I , whereas children 8 to 16 years start with item 4 . A l l items are timed . The first four mazes are given a maximum of 30 seconds each ; the fi fth maze, 45 seconds ; the sixth maze, 6 0 seconds; the seventh a n d eighth mazes, 1 20 seconds; and the ninth and tenth mazes, 150 seconds. The number of errors made determines the child's core. Scores range from 0 to 5 points, with maze s 1 to 6 having a maximum score of 2 ; maze 7 , a maximum score of 3 ; mazes 8 and 9 , a maximum score o f 4; and maze 1 0 , a maximum score of 5 . The subtest is discontinued after two consecutive failures. Although you do not have to adminis­ ter Mazes routinely, administering it to children who are either language-impaired or from a different culture, in particular, may give you useful information.

Rationale To complete the Mazes subtest successfully, the child must (a) attend to the d i rections, which include locating a route from the entrance to the exit, avoiding b l i nd alleys, cross­ ing no l i nes , and holding the pencil on the paper, and (b) execute the task, which involves remembering and follow­ ing the d i rections , displaying v isual-motor coordination, and resisting the disruptive effect of an implied need for speed (Madden , 1974) . The Mazes subtest appears to measure the child's planning ability and perceptual organi­ zational ability - that is, the ability to follow a visual pat-

Reliability and Correlational Highlights Mazes is a relatively reliable subtest ( rxx = . 70) , with reliability coeffic ients above . 70 in 6 of the 1 1 age g roups (6, 7 , 8 , 10, 1 3 , 14) and between . 6 1 and . 68 in the other 5 age groups (9, 1 1 , 1 2 , 1 5 , 1 6 ; range of . 6 1 to . 80 in the 1 1 age groups) . The subtest correlates more h ighly with Block Design (r . 3 1 ) than with any other subtest . It has a low correlation with the Full Scale (r = . 3 1 ) , Performance Scale (r = . 3 5 ) , and Verbal Scale (r = . 23 ) . =

Administrative Suggestions The Mazes subtest is in a separate booklet (WISC-III Mazes Response Booklet) . Administer the subtest on a smooth drawing surface. Use a no . 2 graphite pencil with­ out an eraser to demonstrate the sample item , and give a no. 2 penc i l without an eraser to the chi l d . Write the child's name, the date, and your name in the space provided on the Mazes Response Booklet . When a child makes certain errors, give the child the cues described in the WISC-III manual . The fi rst time these errors occur, tell the child that he or she has made an error and give the child the appropriate cue. The cues are designed to help the child , especially if he or she does not fully understand what to do. A table in the WISC-III manual (page 159) shows how to score the chi ld's perfor­ mance. The sample responses that illustrate the scoring criteria should be carefully studied ( see pages 1 60- 1 64 in the WISC-III manual ) . I n the first printing of the WISC-I1I manua l , there i s an incorrect direction on page 157 for maze 3. The sentence says, "If the child completes the maze within the time l imit with no more than one error, proceed to Maze 4." Disre­

gard this direction because it conflicts with the discontinu­ ance rule. The discontinuance rule states that Mazes should be discontinued after two consecutive failures . Therefore, you should proceed to maze 4 if the child passes maze 1 and maze 2, regardless of the number of errors he or

S U M M A RY

1 1 03

she m a kes on maze 3 . (On

Oc tobe r 2 8 , 1991 I Tlot i Re d T he Psy c ho l o g i ca l C orporat i on of t h i s e rror. Lav.. renlcc We iss a c k n ow l edge d that the sentence is i nco rrect in thle ITlanllal and that it w i l l be changed in the ne x t printin g . ) I nterpret ive Suggestions Consider the fol low ing q u e s t i on s pe rfo r m a nc e . •

as y ou obser\"e the h iJd's

Docs t h e child understand t he task?

•

Does the child sllIdy t he mazes e x t en s i ve i y a nd plan a

•

Do�s the c h i l d show signs of t remor. di fll c ulty i n

route be fore p roc ee d i n g ' )

c on t ro l l i ng t h e penc i l . or d i ffic ul t y i n d ra\\ i n g u n i fNm l ines') •

D oe s the child solve the mazes c o rre c t l y after the t i me

l i mit has e x p i red? •

to

o r t o i m p u l s i v i ty ? D oe s the child say anything that s u g ge s ts a n x i e t y � fo r e x all lple, " The l ittle buy is t rappe d in the enter of t he po o r v is u a l - Ill oto r c oordi n a t i o n •

maze")') These ubservations (and the overall s uccess rate) w ill

g i ve you i n formation abuut t he child's motor p l a n n i n g . s pe e d . e xe c u t io n , i m pUl s i v i t y, and sustained a tte n t i o n .

The range of s c a l e d scores from I to 1 9

at ages 6-0 to However. profile anal­ y s i s is somewhat h i ndered at 1 1 - 0 [ ) 1 2- 1 1 yea r ' and 1 11 0- 1 i y e a r s a ids in p rofi l e a n a l ys i s .

to

[ 6- 1 1 y e ars . where I he 17, respect i ve l y.

and I t o

scaled scores range from 1 to 1 8

whal I say : "Eating too m uch cake and ice c ream ca n g i ve you a stomach a c he . " Answe r : So you have to take an A l ka Seltzer. right? ( Ada pte d from F l u men & Flu m e n . 1 979)

Quest ion :

Liste n . say j u s t

T H I N K I N G THROU GH TH E I SS U E S

In eva l u a t i n g t h e 1 3 W I S C - I 1 1 su bt e s t s , consider the fol l ow­ i n g : W h ich W I SC - I 1 1 s u b tests are the most reliable? I n what ways do t h e W I SC - I I I subtests share common proper­ t ie s . and in what ways do they d i ffe r') What other ki nds 0 s u b t es t s wou ld you l i ke to s e e incorporated in the W I SC ­ l i P Why')

S U M MARY

I . Information measures the child's available information ac quired as a result of native abil ity and early cultural experience. Memory is an important aspect of performance on the subtest. T e subtest is the second-best measure of g and contributes to the Verbal Comprehension factor. Subtest specificity is ample or adequate at most ages . Information is a reliable subtest (r u = . 84 ) . It is easy to administer and score. 2 . Similarities measures verbal concept formation. The sub­ test is the third-best measure of g . Subtest specificity is ample or adequate at 6 of the I I ages. Similarities is a reliable subtest (r u 8 1 ) It is easy to administer but difficult to score. 3 . Arithmetic measures numerical reasoning abil ity. The ubtest is a good measure of g and contributes to the Verbal Comprehension factor and the Perceptual Organization factor. Subtest specificity is ample or adequate at 6 of the I I ages . Arithmetic is a reasonably rel iable subtest (rxx . 7 8 ) . It is easy to administer and score. 4. Vocabulary measures language development , learning ability. and fund of information. The subtest is an excellent measure of g and contributes to the Verbal Comprehension factor. Subtest specifiCity is adequate at 9 of the 1 1 ages. Vocabulary is a reliable subtest (rxx = . 87 ) . It is relatively easy to administer but difficult to score. S . Comprehension measures social judgment: the abil ity to use facts in a pertinent. meaningfu l , and emotionally appropriate mailnCi. The subtcst is a good measure of g and contributes to the Verbal Comprehension factor. Subtest specificity is adequate at 6 of the I I ages. Comprehension is a relatively rel iable subtest . 77 ) . It is easy to administer but difficult to score. (r 6. Digit Span is a supplementary subtest that measures short­ term memory and attention. The subtest is a fair measure of g and contributes to the Verbal Comprehension factor. Subtest specific­ ity is ample at all ages. Digit Span is a reliable subtest (ru . 85 ) . It is easy to administer and score. Administer i t routinely, even though it is not used to compute the IQ when the five standard Verbal subtests are administered. 7. Picture Completion measures the ability to differentiate essential from nonessential details. It requires concentration, visual organization, and v isual memory. The subtest is a fair measure of g . It contributes to the Perceptual Organization factor and to the Verbal Comprehension factor. Subtest specificity is ample or adequate at 10 of the I I ages. Picture Completion is a reasonably reliable subtest (ru . 77 ) . It is easy to administer and relatively easy to score. . Coding measures visual-motor coordination, speed of mental operation . and short-term memory. The subtest is a poor measure of g and contributes to the Processing Speed factor. =

D oe s the child c ross l i nes')

• I f t he child crosses l i nes. is this tenden y relate

I f y o u we re e' v ahu a t i ng a child who had both language and ml tor i m pa i rrne n t s , w h ich W ISC-I I I subtests would be Inl st useful? W h i h ones might be least useful? Why?

.

.

=

xx

=

=

=

1 1 04

APPENDIX J

Subtest specificity is adequate at 10 of the I I age s . Cod Ing is a relatively reliable subtest (ru = . 79 ) . I t is easy to administer and score. 9. Picture Arrangement measures nonverbal reasoning abil ity. It may be viewed as a measure of planning abil ity - that is, the abil ity to comprehend and evaluate a total situation. The subtest is a fair measure of g and contributes to the Perceptual Organization factor and to the Verbal Comprehension factor. Subtest specificity is ample at all ages. Picture Arrangement is a relatively reliable subtest (rn . 76 ) . It is easy to administer and score. 1 0 . Block Design measures spatial visual ization ability and nonverbal concept formation. The subtest is the best measure of g among the Performance Scale subtests . It contributes to the Perceptual Organization factor and to the Verbal Comprehension factor. Subtest speCificity is ample at all ages. Block Design is a reliable subtest (ru . 87 ) . It is somewhat difficult to administer but easy to score. I I . Object Assembly measures v isual organizational ability. The subtest is a fai r measure of g and contributes to the Perceptual Organization factor. Specificity is ample or adequate at 4 of the I I ages . Object Assembly is a marginally reliable subtest ( ru . 69). It is somewhat difficult to administer but relatively easy to score. 1 2 . Symbol Search is a supplementary subtest that measures visual discrimination and v isuoperceptual scanning. The subtest is a fai r measure of g and contributes to the Processing Speed factor and to the Perceptual Organization factor. Subtest specific­ ity is ample at 8 of the I I ages. Symbol Search is a relatively rel i able subtest ( r.xx = .76). It is easy to adm inister and score. 1 3 . Mazes is a supplementary subtest that measures planning ability and perceptual organization . The subtest is the poorest measure of g in the test and contributes to the Perceptual Organi­ zation factor. Subtest specificity is ample at 10 cf the II ages. M azes is a relatively reliable subtest (ru . 70) . It is easy to administer but difficult to score. =

WiSe-III S U BTESTS

KEY TERMS, CONCEPTS, A N D N A M ES

WISC- I I I WISC-II I WISC- I I I WISC-I I I WISC-I I I WISC-III WISC-II I WISC-II I WISC-II I WISC- I I I WISC- I I I WISC- I I I WISC-I I I

Information ( p . 1080) Similarities (p. 1082) Arithmetic (p. 1083) Vocabulary (p. 1086) Comprehension (p. 1088) Digit Span (p. 1089) Picture Completion (p. 1092 ) Coding ( p . 1093) Picture A rrangement (p. 1095) Block Design (p. 1097) Object Assembly (p. 1099) Symbol Search (p. 1 1 00) Mazes ( p . 1 102)

=

=

=

STU DY QU ESTION

Discuss the rationale, factor analytic findings, rel iabil ity and correlational highlights, and administrative and interpretive con­ siderations for each of the fol lowing WISC-II I subtests: Informa­ tion, Similarities , A rithmetic, Vocabulary, Comprehension, Digit Span, Picture Completion, Coding, Picture A rrangement, Block Design, Object Assembly, Symbol Search, and Mazes .

R E F E RENCES

Holmes, 1 . M. ( 1 98 8 ) . Testing. In R . G . Rude l , Assessment of developmental learning disorders ( pp . 1 66-20 1 ) . New York: Basic Books. Wechsler, D. ( 1 99 1 ) . Manual for the Wechsler Intelligence Scale for Children - Third Edition . San A ntonio: The Psychological Corporation.

_ APPENDIX K I NTE RPRETI NG TH1 E Wise-I I I

The gifts ofna/Llre are infinite in their \'ariet.\', (lnd lIIind differs from mind allllost as IIlIleil os bach froll/ body.

- Quinti ian

Profile Analysis Gu idel ines for Interpreting and Communicating W I SC-III F i ndings in the Psychological Report Su ggestions for Describing WISC- I I I Scales and Subtests to Parents and the Referral Source and in the Report Ps ychological Eval uation Test Your Skill Thinking Through the Issues Su mmary

I I I GI S

1 1 06

Appendi x K presents gu idel i nes for interpret ing th WISe-III and for writing reports . The appendix if1l!/Sr bc used in conjuncrion with Chapters 8 and 23 , because these

chaprers present more detailed guidelines for intelprerillg all of rhe Wechsler tests and for writing reports. Because the methods of interpreting the W ISC- I I I - such as the successive level approac h , profile analysis , Performance/ Verba Scale comparisons , and subtest comparisons - are essentially the same for both the W ISC-R and the W I SC-I I I , study carefu l l y the contents of Chapter 8 that pertain to these issues. Appendi x K complements these chapters by focusing on the W I SC- I I I . Appendix K also i ncludes a tra i n i ng exercise designed to sharpen your re­ port writing s ki l l s (Exhibit K-3) and two i l lustrative re­ ports ( Exhibits K- l and K -4 ) . Appendix L contains tables specificall y designed to help you imerpret the W I SC - II I . Other tables that can assist you i n interpreti n g the W I S C - I I I and in writi ng reports are as fol low,: Table C - 1 3 (page 824 ) , which summarizes the abilities thought to be measured by 1 2 of the WISC-I1I subtes:s (with some m i nor exceptions , as noted in the footnote to the table ; Table L - 1 6 in Appendix L presents similar information for Symbol Search); Table C-42 (pages 856- 8 5 7 ) , which summarizes the interpretive ratio­ nales for the Full Scale, Verbal Scale, and Performance Scale cfthe Wechsler batterie s ; and Table C-43 (page 858), which presents activities to improve children'S skills based on Wechsler subtests . On the inside back cover of this text, Table BC-2 shows the classifications associated with WISC-I I1 IQs, and Table BC-I llOWS the percentile ranks for the WISC-I1 I Full Scale, Performance Scale, and Verbal Scale IQs . Percent­ i le ran<s for the three I Q s also can be obtained directly from the W I S C - I I I manual in Tables A . 2 . A . 3 , and A.4 ( pages 25 1 -254) . Table CAl i n Append i x C (page 855) shows :he percentile ranks associated with subtest scaled scores. Table C - 1 2 in Appendix C (page 822 ) , which gives the Structure of I n tellect classifications for the subtests on the W I 5C-R , also can be used for the fol lowing W ISC-IlI subteSb : Arithmetic , Vocabulary, Comprehension , Digit Span , ?icture Completion , Picture Arrangement, Block Design. Object Assembly, Cod i n g , and Mazes .

PRO F I LE ANALYSIS

The seven primary approaches to profile analysis of the W ISC-JI desc ribed below are essentially the same as those describ�d for the WISC-R (see pages 166- 1 7 1 ) However. there ale some differences . First. for p rofile analysis you should u se the tables in

APPEN D I X K

INTE RPRETING THE Wise-III

Appendix L, which cover the W I SC-I I I , i nstead of those in Appendix C, which cover the W I SC-R. Second , the critical values in Appendix L for the W I SC­ I I I are based on the chi ld's speci fic age group rather than an average value (as in the WISC-R ) . Third . factor scores on the W I SC - I I I ( see Appe ndix I ) differ from those on the WISC-R (see Chapter 6 ) i n the follow ing way s : Althoug h the Ve rbal Com prehension fac­ tor is the same on both tests ( I n format ion, Similarities. Vocabulary, and Comprehension ) . the Pe rceptual O rgani­ zation factor is not ( Picture Completion . Block Design , and Object Assembly on the W ISC- I I I and Picture Com­ pletion , Block Design . Object Assembly. and Pictu re Ar­ rangement on the WISC- R ) . In addition . the W I SC - I I I has a Processing Speed factor (composed of Coding and Sym­ bol Search) that is not present on the W I SC-R, and the WISC-R has a Freedom from Distract ibility factor ( A rith­ metic, Digit Spa n , and Coding) that is not present on the WISC-IlI . On the W I SC-I I 1 , Arit hmetic , Digit Spa n . Pic­ ture Arrangement , and Mazes are not incl uded in a factor score. Let us now examine the seven primary approaches to profi le analysis. I. Comparing Verbal and Peljormallce Scale IQs. Table L-3 in Append i x L prov ides the c ritical val ues for compar­ ing the Verbal and Performance IQs for the I I age groups of the W I SC- I I I . These values range from 10 to 13 at the . 05 level and from 1 3 to 1 7 at the . 0 1 level . Thus. an average critical value based on the entire standardizat ion group would be misleading. Therefore. use the values for the chi ld's spec ific age group to evaluate d i ffe rences between the child's Verbal and Performance I Q s . ( Table L-7 in Appe ndix L shows the probabi lities associated with var­ ious differences between the W I S C - I I I Ve rbal and Perfor­ mance Scale IQs . ) 2 . Comparing each Verbal subresr scaled score ({) rhe mean Verbal scaled score. Table L-4 in Append i x L pro­

vides the critical values for each of the I I age groups of the WISC-I I I . Typical values for 6-year-old chi Iclren on the fl ve standard Ve rbal subtests . for example. range frolll 3 . 0 1 to 3 . 5 1 at the .05 level and from 3 . 60 to 4 . 2 0 at the . 0 1 level .

3 . Comparing each Pel!onl/allCe subresr scaled score 10 rhe meall Performallce scaled score . Table LA in A ppen­

dix L prov ides the crit ical values for each of the I I age groups of the W I SC-II J . Ty pical values fo r 6-year-old ch i 1dren for the five standard Performance su btests . for exam­ ple. range from 3 . 03 to 3 . 62 at the . 05 level and from 3 . 62 to 4 . 34 at the . 0 1 level . 4. Comparing each sllbresr scaled score to rile IIleall subresr scaled score. Table LA in Append i x L provides t he

1 1 07

PROFILE ANALYSIS

crit ical values for each of the I I age groups in ! he W I �SC -I11 for 10, I I , 1 2 , and 13 subtests . Typ ical values fur 6-Yt'�ar-·old chi ldren for the 10 standard subtests, for example , ra:nge from 3 . 43 to 4 . 2S at the . OS level and from 4 . 02 to vevlcr, bec,aus,e it provides base rate i nformation about what Oc(2Ur ed i n t the standardization sample.

Base (The

Rate Verrba l-P(erf(orrnance Differences Prob;abillity-of-·Oc:cu rrence Approach)

Determ.ni Ig hlow fre(que: ntl y a Verbal-Performance IQ dif­ ference of a gi've m,agn)itude occurred in the standardiza­ tion sa m pl e is referred t(o a.s the p robabi l ity-of-occurrence approac h . Thle freqlue nacie s w ith which several Verbal­ Perforrr ance dl i s cre planc' ies a re est imated to have occurred in the n·ormatilve stamdmdization sample are given in the expectaJ1cy talole in 'Tabble JL-8 of Appendix L. The table shows. for ex:ample, (haat ter. R . . 3 7 8 . 915

Ghoze i l . 5 .. 585

108 7 . 1089. 109 1 . 109 7 . 1 100 . 1 103

J..

253

G reen. F . 1 14

E . . 26

1. A. .

2 9 8 . 299

G reenberg. M . T. . 432 G reenberg. R. D . . 567

Foster. S. L . . 492 . 493 . 519. 523

G ibertini. M . . 224

G reenfie l d . J . . 785

Fothe ringhan . J. B . . 656

G i landas. A.

Greenlief. C. L .. 234

Fow l e r. P C . . 1 2 7

G i lbert. 1 . G . . 39. 59. 367. 636. 639

G reenspan.

Fow l e r. R . L . . 1 10

Gill . M

G reenspan. S . 1 . , 4 10

Fow l e s . G . P . 2 2 4

G i l l ingham . W H . . 1 14 G i l more. S. K . . 407

G reenste i n , J . . 3 4 1

Fra i be rg . S . . 636 Fra n k . L. K . . 4 1 9

G i l son. 1 . . 786

Gregg. N . . 2 2 4

Fra n k . R . . 2 2 5

Ginn. F. W . 3 14

Gresham.

Frankenburg. W K . . 3 5 1 . 352 . 9 1 9 . 926

Ginsburg. R. E . . 572

Grice. 1 . E . . Jr. . 195

1..

7 19

M .. 148 , 829. 839. 1082

N.

T. . 410

Greenwood. C . R . . 5 0 3 . 7 7 5

F..

2 2 7 . 402

Fran k l i n . M. R . . J r. . 1 14 . 2 3 2

Giolas.

Frede . M

G illelman . R . . 501 . 906

Griese l . R. D . 76

Gilileman. M . . Glass. A . . 2 2 7

Groff. M .. 127. 627

Franzo n i .

1. B . .

772

C . . 195

Frede rikse n . N . . 5 7 . 58. 60. 79

T. G . 640. Giordan i . B .. 700

641

Gridley. G . . 383 Griese. A. A . . 787

63 1

Freed man . A. M . . 884 Free man . B . J . 63 2 . 910

Glasser. A . J . . 8 2 9 . 839

G rossman . C . 5 . , 787

Free m a n . F 5 .. 4 5 . 8 2 9 . 839

G l u l l i ng . J. J . . 103 8 . 1078

G rossman . F D . . 1 1 3

Fre e m a n . R . D . . 6 1 8 . 786

Goddard. H. H . . 42. 5 8 . 60

G rossman . F M . , 1 2 5 . 1 7 5

Free i l l a n . R. W . 109. 1 3 1 . 1064

Goethe. 3 75

G rossman. H . J . , 3 7 6 . 647

Fre m o n t . T. S . . 364. 655

Goctz. E. T. . 303

G runau. R . V. E . , 76

Gross. R . . 7 1 3

F re nc h . 1 . L . . 3 1 4 . 3 1 5 . 925

Goetz. L . . 660

Guare . 1 . C . . 928

Fre n c h . L. A .. 5 8 . 60

Goffeney. B . . 567

G u i l d . J. 1 . . 503

Fre n c h . R . W . 3 3 3

Goh.

Fried man . L .

Gold. R. D . . 195. 196

c..

224

D. 5 . .

1 10. 1 3 7 . 2 9 8 . 1067. 1068

G u i l fo rd . 1 . 681

p. .

46-48. 50. 56. 59. 60. 3 5 6 .

Frie d r i c h . W N . . -132

Goldberg. 1. S . . 55

Gullo. D. F , 350

Fristoe. M . . 3 7 1 . 92 1

Golden. C. J . 2 2 3 . 700. 703 . 704. 924

Gusdorf. G . . 3 7

Goldenson. R. M . . 4 19

Gussow. J . D . . 5 7 4

Goldgar. D. E . . 3 17

Gustafsson . J . E . . 49 . 50. 5 9 . 60

D. .

714

Frost i g .

M. .

920

F ry.

165

Frnm m .

Fromm-Auch . D . . 700

c..

F u c h s . D .. 88 Fuc h s . L.

5. .

88

F u l ker. D . W . 63

F u l l e r. G. B . . 365

Fu n k . S. G . .

298. 299

Fu re } . W . 500

Goldman. 1 . 1 . . 126

Gustave. F P . 3 7 9 . 3 80

Goldman. R .. 371. 92 1

Guthke . J . . 549

Goldschmid. M

Gutk i n . T. B . . 1 2 7 . 2 1 2 . 2 2 4 . 2 2 7 . 567.

Goltbte i n .

D. .

Goldste i n . G. I

L . . 55

568 . 5 86

5 8 6 . 8 1 4 . 816

. 636

Guy. D. P. 567 G w y n n e . 1 . . 1 14 . 354 . 367 . 570

G o ldwa>ser. E . . 776 Gol land . 1 . H . . 1 3 5 . 1067 Good .

T.

L. . 551

Goodenough .

F.

L. .

Haddad. F A . . 1 2 5 . 3 50

.I I I

Gabrid l i . W F . 626

G,lodman. 1 . D . . -1 1 9 . -1 2 1

Galfney. P D . . 586

Goodstei n .

G ag n e . F . 665

Gmdcn. R. L .

H . A . . .\-1 I . -102 . -10S .

Hadley.

T

J . . 380

Hagek u l l . B . . 3 2 1 H a ge n . E . P . 8 . 2 1 . -1 3 . 60. 1 2 5 . 1 2 6 . 2 -1 6 . -106. 4 1 1

2 6 2 . 9 2 7 . BC-2

1 1 85

NAME INDEX w. .

M . . 785 I . B . . 76. 567 HC lllier,,,, ' n . R . W . 19-1 . 198

599 . 601

Hender",lIl.

Hubble.

H ag i n . R. A . . 2 1 2 . 2 1 3

Hc nJcr,,'li1.

H u berty.

Hagen. 1 . H ah n .

W. K . .

691

Hcnkc-r, IB . .

H a i mes . P E . . 920

Hcnnewl. J. J . . 5 78

H ale . D . . 7 7 Hale. R .

L..

628

V. c . .

L. .

Hami l ton . 1.

D . . 343 C. w. . 134

H u izinga. R . 1 . . 346

Hernand ez . C. E . . 9 1 7

549

>"1 . .

Hwe l l . J

Hamm. H. A . . 628

U S . 1067

H ug h e s .

H ul m e .

c..

7 6 . 80. 569 . 5 7 2 . 592

H u n n icull . L.

Hamsher. K. deS . . 703 . 705 Hancoc k . K. A . . 298

Herzbagcr. S . . 928

Huppertz. J .

Hand l er. B . . 766

He, . . A. K . . 125

H u rley. O.

Hamm i l l . D . D . . 303 . 305 . 346. 598 . 606. 9 1 9 . 920

H andal . P 1.. 1 2 5 . 3 5 0

H andler. L. .

_

Hell e , . P .

766

4ill

H a n so n . D . P . 9 6

B. I( .

H i lscn . D. E . . 126 H i ng te n . J .

H a r r i e s . J . , 787

H i skey.

H a rper. R . G . . 2 2 4

Harris. D. B . . 3 10-3 1 2 . 9 2 1

c..

A. .

N. .

Hart. D . H . . 628

L. c . . 567 . H a r t l ey. R. E . . 419 Hartman. D . E . , 790 H a r t m a n . D. P . 492

752

Hartman, R. K . . 667 H a r t w i g , S. S . . 253 Harver, 1 . R .. 5 7 1

631

H i rsc h . 1 . . S 7 7

H i rshoren. A . . 640 Hin. M. . 571

M. S . .

3 1 6 . 922

1·lobby. K. L . . m . 1068

D. . H od app . R . .

682 73 , 647

Havassy. B . . 569

m P T. . 224

Hoga rty, P S . . (,7 . 70. 83

Hogc, R. D . . 5 1 8 . (,7 1 . 907 Holland . W R . . 58(, Holler-bec i- , C. P . :20 : . 567 Holl iman. W. B . . 1 2 8 . 610 Hol l i nger. C. L . 1 2 5 . 2 5 3 . 350

HOll1at idi s . S . . 625 Honz i k . M

L. .

787

P . 71

Hoover. H . D . . 21

Hopk i n s . K. D . . 636 . 639

Haynes, J . P . 1 2 3 . 2 0 1 . 2 0 4 . 1036

Hopk i n s , T. . 2 5 8 H o p s . H . . 503

Haynes. S .

Horn. 1.

Hayman . P M . . 2 2 4

N..

519

H ay s . J . R . . 6 2 7 Hearn.

1. . .

-17--19. 56. 58-60. 79. 1 1 4 .

244 . 681

C. .

Hay wood . H .

M . . 408 0. . 578

656

Horn.

W F. . 519

Horto n . W. -'83

Hebb. D .

Horwitz. E . . 125

Hecaen . H . . 6 9 1

Hostet l e r. S . " . 96

H e lTe rna n .

Hostetler. S . .

L. .

322

Hega r ty. S . . 549 . 5 5 0 Hei l . 1 . . 201 H e i l bronner. R . L . . 2 2 3 Heine. R .

w. .

r2

H e i neman n . A . W.. 2 2 -1 H e i s ler. V. A . . 7

5

H e l ler. K . A . 5-13

Heller. M . S

. 3-1.'

Hynd.

I acovie l l o , 1. I Ig .

F. L . .

321

M. . 710

I l l ingwort h . R . S . . 7 1

I ngl i s .

J . . 173

I relOn . H . . 4 3 3 . 442 I s aa c so n . R.

V. . 9 1 5

Iscoe. I . 567

Isell. R . D . . 380

I vey. A . E . . 408 I l' i mey. J . K . . 299 Ino. L . D .. 9 1 5

C . N . . 70

Jackson. A . M . . 1 1 0

H a w k i n s , R. Haye s . M .

B . . 125 G . W, 1 2 5 . 2 9 8 . 705

HUllon . J .

Jack l i n .

Holtzma n . W H . . 543

Hawthorne . L. w. . 2 14

w. . 567 1.. . 640 Hutcherson, R . . 343

Hocy. 1 . A . .

H a w k i n s . K. K . . 660 660

T.. 640

Hunt. 1.

H u n t . W A . . 1 1 -1 Hunter. M . . 3 1 4

Hon'ma n .

Holroyd. 1 . . 4 3 2

Halle n , J . T. . 86

:is-1

Houc h i n s . S . . �85

Hou s to n . W. B . 58,6

13-1 ., 1 5 Howard . R . C . 1 1 � . 1036 How ard . D . .

Ho\\ a rd . 1 . L .

HOll c . J. G . 703 Ho\\ c l l . R . . 127

Ho ) .

C . n4

105-1. 1078

H u n t . 1 . McV . 5 5 . 6 1 . 3 2 1 . 922 . 9 3 0

Hoepfner. R . . r6

Holmes . 1 . M . . 108 1 . 1 10 4 . 1 108. 1 1 2 7

Hatfield . E . . 782

P. .

2 2 4 . 475 . 529

Hodges. K . . 127 . 459

298

Harrison . P L . . 2 9 8 . 299 H a r t l age .

W.

H i ltonsmi t h . R.

Hocevar.

519

H a r r i s , S . R . . 90 Harriso n . K .

,85 9?1i

H i l l i x . W A . . 1 1 3 . 5 3 1 . 763

A . . 65 . 6 7 . 8 3 H a rd w ic k . R . S . . 43 Haring, N. G . . 657. 659 Harmer. W. R . . 1 2 5 Harper. D . C . . 3 5 1 . 3 5 2 Hanso n , R .

c . . J r. .

H u n t . E . . 52. 58. 59. 60

Hutchinso n . J . M . . 504

H ighlOll cr. A. D . . 928

S. . 114

H::1sc�e. W. .' . . 5 5 7

H a r r i s . F.

53

H ickoK. I.. A .. 787

Hanesia n . H . . 73 Hanna. G.

HCrrlng.

77

H u mphrey,. L . G . . -15 . -1 7 . 5 5 . 5 6 . 69 . 7 3 .

S . 700 Hcrrn>le i n . R . J . 5 7 7 Her'en. \1 . . -163 Hcr,h. J . B . 1 1 3 H e rl Llg . �1 . E . . 1 1 4

HammCII. E. B .. 2 2 -1

79

H u e l s m a n . C B . . Jr . . 567

HCfj ;nil' . B . -159

'' K r;:sto.n . \1\ . J . 'R6 . Klhan . :\ . ..\ . Gtl7

D A . 1 10 K i.ern.a n . C . . h�7 KilJllkrl in . C . 63 1 Kill1g. J D . . 1�5. 204 Kimg. \ 1 . L . ::'20 K i m g . \\' L . .,20 K i n,h,)urn e . :-'1 . . 446 Ki n,!()n. \ . . 447 . 449 K i r by. J . . 5 2 K i r k . S A . . 345 . 346. 922

Jorm. A . F . 600. 606

K u nclk .

K upersmit h . A . 2 24

Ki. e ffe r.

K irk . W D . . 1 0 4 . 3 4 5 . 3 4 6 .

K i rschenba u m . D .

K ut ner. N. G . . 5 7 5 Lace\. H . M . 752 LaCrosse.

L

J. E . .

114

La Greca. A. M . . 425. 4 3 1 Laham .

S. L ..

657

Lahey. B . B .. 5 1 9 . 9 1 8 . 9 3 0 Lair. C .

v. .

125

A .. 332 S . H . . 640

Lamanna. J . Lamb.

922

S . . 500

Kilche l l . M . M . . 606 K ilson. D.

T.. 1 2 7 T. M . . 166

K u nc e . J .

12 5 . 3 50

Lambert. N . M . . 3 8 1 . 9 1 5 Lamson . F . 6 1 0 Lancy. D . F . 636 Landesman-Dwyer. S . . 654

Kadus h i n . A . . 401

K l anderman. J . . 125

Kagan. S . . 585

K l apper. Z. S . . 102

Lane.

Kahn. H . . 341

K l aric. S. H . . 459

Laosa. L. M . . 586

K l e i n . M . 223

Laprade. K . . 393

Kahn. J.

v. .

322

P.

224

Landino. S. A . . 628

V.

W . 408. 456

Larrabee. G . J . . 705

Kahn. R . L . . 401

K l e inman.

Kalas. R . . 459. 464. 465

K l e inmunlz. D .. 461

Larrabee. L. L . . 1 1 3

K a l i sk i . M. S . . 346

K l e i nsasser. L. D . 1 1 3 . 1 14

Larsen. J . . 2 2 1 . 234

Kanfer. F H . . 450 Kanfer. R . 420. 424. 425

W G. . K l i ng . J. 0 .

224

Larsen. S . c . . 598

2 24

Larson. G . L . . 365

Klett.

K l i nge. V. . 127 . 628

Kaplan. E . . 703

Laurence. K. M . . 196

Kapl a n . H. B . . 198

Kljaj i c . I . . 705

Lawlis. G . F . 567

Kapl a n . H. I . . 884

K l ug rnl n . S. F . 1I0

Lawson. J. 5 . . 173

R..

56

Layzer. D . . 5 7 8

Karnes. F A . 125. 1 3 8

Knapp. J.

Karoly, P . 4 2 2 . 423

K n i g h t . R. G . . 8 1 3

Karon. B.

Knobloch. H . 7 1

Lebov itz. L . . 1 14

Knofl", H . M . . 3 7 3

Lee . R . , 3 1 5

Kobayas h i . K . . 694

Lee. S . W , 1 1 4

P . 580 Kaspar. J . c . . 114

Katz. E. R . . 497 Katz-Ga r r i s . L . . 380

Koc h .

K a u fman . A. S . . 55. 125. 1 2 7 - 1 30 . 1 3 8 , 140.

Kochman.

R.,

71

T..

LeBaron. S . . 4 9 7

Lefever. D . 583

w. .

920

Lehma n . J . , 365 . 3 7 3

147 . 1 7 0 . 1 7 8 . 1 7 9 . 195 . 1 9 7 . 1 9 9 , 2 0 1 .

Koegel. R . L 630

Leibman. R . . 448

204 . 2 5 7 . 298-300. 3 3 3 . 567 . 6 5 5 .

Koenigsknccht . R. A . . 3 1 5

Leigh. J . E . . 598

829 . 8 3 9 . 8 7 9 . 9 2 2 . 986. 1036. 103 7 .

Koenke, K . . 373

Leiter. R . G . . 3 1 3 . 314. 924

Koepke.

Leland, H . . 378. 575. 915

1049.

1070. 1074. 1078

K a u fma n . N. 922

L. .

1 2 5 . 2 9 8 . 300. 3 3 3 . 879 .

T . 128 K o g a n . K . L 102 Kogan. L . . 3 1

T . 568 Koh n . M . . 907. Koh.

Kaye. H . . 76

Kol i n . J. L . . 744

Kazd i n . A . E . . 4 8 8 . 493 . 520

Kol i n . p c . . 744

Kazi mour. K. K . . 383

K o l ler.

.

Lerner. E . . 419 Les iak . J . . 365

Kavajec z . L. G . . 214 Kava l e . K . A . . 609. 640

J. R

Lerner. B . . 563

923

. 1 2 2 , 1 2 5 . 1035

Leskosky. R . . 198 Les ser. G. S . . 5 7 8 Lester. E .

P.

55

Lester. M . L . . 365

Kazu k . E . . 3 5 1 . 919

Komm. R . A . . 567. 579

Levandowsk i . B . . 586

K easey. C. T . 5 5

Konstantareas. M . M . . 625

Levee. R. F . 367

Keat ing. D . P . 55

Kopp.

K e i r. G . . 3 10

Koppitz. E . M . . 3 6 1 - 3 6 7 . 885

Leventhal . M . 7 5 2

K e i t h . T Z . . 2 9 9 . 302 . 303 Keleman . K . 5 . . 503

Korch i n . S. J . . 400. 4 6 7 . 565 Kovacs. M . . 459

Levet t . C . A . . 343 Lev i . A .. 567

K e l ler. H . R . . 4 7 5 . 5 2 9

Kozup J. M . . 7 1 2

Lev i n . H . S . . 693 . 694

Kel ler. J . F . 194

Kraepe l i n . E . . 39. 5 8 . 5 9

Lev i n e . M. N . . 3 14

K e l lerma n . J . . 497

Krahn. G. L . . 420. 425

Lev i ne . S . . 3 8 8 . 390. 9 2 8

K e l ley. C. M . . 46

Kramer. J . 928

Levy. B . B . . 5 7 1

K e l logg. M . 787

Kramer. J. 1 . . 1 2 5 , 350

Lewandows k i . D . G . . 627

K e l l y. M .

P.

223

C. B . .

71

Levenso n . R . L . .

Jr.. .

298

Kramer. R . . 125

Lew i s . J . F . 3 5 2 . 363 . 3 82 . 9 1 5 . 9 2 7

K e l l y. R . R . . 6 3 9

Kratochw i l l . T R . . 110. 341

Lew i s . L . L . . 6 3 9

Kenda l l . p c . . 1 3 7 . 1068 Kendric k . S A . . 572

Krauft . C . c . . 365

Lew i s . M . . 7 1 Lewis. M . L . . 1 3 2 . 202 . 2 2 3 . 2 3 2 . 991

Krau ft . V. R . . 365

Kennedy. L. P . 1 3 9 . 1070

K re b s . E . G . . 1 1 4 . 197 . 198

Lewis-O·Don nel i . M . . 253

Kennedy. T . J r. . 787

K re n t s . H . . 787

Lewkowicz. D . J . . 692

Ken nedy. W. A . . 1 1 0 . 567

K roh n . E. 1 . . 298

Lezak. M . D . . 705

Keogh. B. K . . 7 1 . 367. 542

K ro n i c k . D . . 785

Lichtenstei n . R . . 433 . 442

Kephart. N. c . . 925

Krug . D . A . . 632

Lichtman.

Kershner. J . . 392

Krupansky. R . . 781

Lindamood. c . . 370. 924

M. V.

198

1 1 87

NAME INDEX Lindamood. P. 370. 92.1 Lindemann. S . J . . 56 7 Linde n . J . D . . 40 Linden, K . W . 40 Linder. B. A . . 605 Li ndzey. G . . 578 Li nney. 1. A . . 928 L i ppe rt . J . . 391. 930 Lippold. S . . 2 2 3 L.ittle. J . . 367 Livesay. K. K . . 253 Lloyds. L. L . . 7 1 L oader. P . 447. 449 Lobascher. M . E . . 654 Locke. J . L. . . 37J Lockyer. L. . 631 L.oeh l i n . 1. C .. 578 Loew. D . E .. 125 Lohman . D . F.. 3 10 L o t11ban a . J . H . . 787 Lombard. 1'. J . . 610 Loney. J . . 500 Lng. P :\ . . 57': Longstreth. L. E . . 69 . 77 Lopez. M . . 585 Loprete . L. J . . 114 Lord. C . . 93 L.ord . R. G . . 520 L.orge, 1 . . 315. 919 Lorion. R. P . 9 1 5 . 918 Lose. B . D . . 125 Lotter. v. . 631 Lotyczewsk l . B . S .. 92 L ove l l . K . . 680 Low. M . D . 76 Lowe l l . 1. R . . 328 Lowenfeld. B .. 785 Lubinsk i . R .. 94 Lucas . B . A .. 114 Lucas. D .. 54lJ. 5. 0 Lucas. E . Y. , 549 Lu msden . j . . 347 Lund. J . M . . 1 10 Lupo. 1 . Y. . 2 2 3 Luria. A . R . . 52 Lyle. J. G .. 1 10 Ly man . H . B . . 3 3 2 . 333 Lynch. A .. 299 Lynn. R . . 569 Lytton. H . . 7 10. 7 1 1 .

.

MacArthur. R . S . . 3 10 M accoby. E. E . . 70 Madden . T M . . 163. 1 102 Maffeo. P. . 207 Maha n . T W . Jr. . 1 14 M aheady. L . . 659 Major-Kingsley. S . . 542 Maloy. C. F. . 114. 149 . 1083 Mange l . C . . 785 Mann. M . . 1 1 3 Mann . T.. 1079 Manni . J . L . . 1 2 5 . 354 Mantle. M . . 785 Many. W A . . 34 6 Margol i s . R . B . . 234 Mar·Hay i m . M . . 1 1 3 Marjoribanks. K . . 65. 66. 3

F. c . . Jr. . 332 . 924 Marlamd. S P . J r. . 661 Md ric). M. L . . 365 Marmo,rale A. M . . 367 Ma r shaili . W. . 1 2 5 Mar
Sapp. G. L. . ::' 5 3 . 383

A . . 577 Roach. E . G . . 925 Robb. G . P . 153. 1089

Sarv i s . P

H . . 1 2 5 . 350 G . . 450 Sanertield . B. T. . 625 Sane rficld . J. H . . 6 1 8 . 625 Sanerly. D . . 75 Sa n l c r. 1 . M . . 43. 60. I I I . 1 1 3 . 1 1 4 . 1 2 5 . 126. Saslow.

Robe n s . G . 640

Robens. M . A . . 500 Robe n s . M. C . . 784 Robenson . M . . 1 1 3 R obi n so n . E . J . . 9 1 8 Robinson. H . 13 . . 667 . 675. 678-680

\ ]4- 136. 143. 147 . 149 . 1 5 1 . 153. 158. 159. 166. 2 10 . 2 2 -1 . 2 2 5 . 246. 250. 262. 295 . 35-1 . 3 63 . 367. 46 1 . 5 1 1 . 570. 586 . 610 . 6 8 . 789 . 927. 1067.

Robinson. N . . 90

Robi nso n . V . 787

Rock. D . A . . 5 7 8 Roe. A . . 681 Roedel l . W. C . . 6 7 4 . 675 . 678. 679 Rogers. D. L . . 2 2 3

Roge r s . S . J . . 55

Rohrbec k . C . A . . 928 D . F. 705 Rosen. J . C . . 1 2 5 Rosenberg. L . A . . 114 Rosenberg. S J . . 2 2 3- 2 2 5 . 2 2 7 Rose n feld . J . ( j . . 1 2 6 Roscntha l . R . . 1 \ 3 . 572 Ross. A. 0 . . 7 52 Ross. 1. M . . 1 1 4 Ross. R . T.. 657 Ro,sman. B. B . . 681 Rosso. M . . 12 5

Ro,e.

A. F . 656

Rot h . R. J . 1 1 4 Rot h l isberg. B . A . 2 5 3

C . . 1 1 -1 Rot h ma n . S . -1 5 Rot h\\ e l l . J . G . . 580 Ro\\ land. T . �9R Rot hman.

R(lw k\.

A . . 780

R u b i n . D . C ..

61X

I . . 694

Sanders. J . L . . 558

R i"a,. G.

Rotator i .

D. M . . 659

Sa ndova l . 1 . . 568

Ritvo. A . . 6 32 . 910

Rot h . D . L . 2 3-1

V. . 55 P . 627

Saccuz LO . D.

Sandberg. S . . 393

Riner. D . . 3 15

Ro,zko\\ s k i . M . 1 . . 380.

D . A . . 346 D. L. 1 1 4 . 567

Sabalino.

Sa:�s.

R isucci .

R itvo. E .

Ruhln. E . J . . 6 36. 639 H. IH .. 124. 1 2 6 Ruchalla. E: . . 7 7 Rucker. C . N . . 7 3 1 . 7 44 . 7 5 2 Ruple). W . H . . 477 . 6 10 Ruschlval. M . L . . 194. 195. 199 Ru"c l l . W . 571 RubI O .

07

I08J . 1086. 1088 10')7 . Ion. B l ' - 2 P. 1 1 7 . 234 . 244 . 692 . 1068 Sa u l . R . I:- . 6 1 Saunders . B. T . 1 1-1 Sawyer. R N . 19 7 . 3 15 Scarr. S . . 579 Schachar. R . . 393 Schachter. F F . 1 14 Schafer. R . . 148. X29. 839. 1082 Schal le r. S . . 3 10 Scha l l . E . . 77 Sche rr. S . S . 197 Sch i l le r. 1 . J . . 713 Schimme l . D . 786 Schmidlke. A . 310 Sa! z .

W. 196 A . '217 Schooler. D . L . . 1'28 Sc h oo nove r. R. J . . 78.6 Sc hopler. E . . 6.12 634 Sc hreibma n . L . . 6.10 Schreiner. D . . 627

S c h m i !!> . D .

h

Sc ncider. 1.

Schroeder. H . E .. 113

Schu be n . E. D . . 0-10 Schu l e r. A . L . . 660 Schu l ma n . J L . . 1 1-1 Sch u b l l1gcr.

Sch " arl/ .

F . 6::'6

J . ,\7. 1068 �1 . L . . 11-1. 689

Sch" rodluct·.mo l ment correlation coefficient. see PealfSO Ins r F�rooluct" dJl lllen"on (Guilford) . 47 "'rohile amal y si; SIB FE. 2 76 . 2 7 7 WI.l i male .

Standard t; 1 rllr

ScclJlld i n t c n icv., � . -l51

Sd f-d i , c l l "l l re , . -1 1 2

set!

111.10 Rclialb i l i l y

S d l - fu l l i l l i n g prophec\. 5 5 1

SlanJ�rd prucc:dure,. de parl u re s . 109 - 1 1 2

S e l f-g raphi n g dill a recorll i n g ' ) ,Iem . -187

S l andard Proglre , s i ve 'v1atrice,. 309 . 3 10

Sell la n l l C Evalualion. 306

Siandarci ,core .

Se111 a m i c Producl i o n . 306

d i ,c rep:JnL)

S(: l I l 1 -!-ot rllctureti inten 1('\\

2'2 . r

i

in l e a rn i ng d i sab l i t y. 608

SlandardILa\1()111. see specific lesls

LIIlIi ly. 4-11)

Cll i Te.\ lill·�.

7

Sla n ford - B l ncl I mc l l i g ence Sca l c . e a r l y ed i ­

par�Il" of dc\ c i opmelli a l l y d i ,abled chi 1-

S l a n ford·BlIlel I nl e l l i e nce Sca l e : Form

l i o n , . -1 2 . -\ .1 . 246-248

parellis o f heari n g - i mpaired c h i l d re n . 640. 6-\1

o

�arel1lS Df p re scho l c h i l d r� n . 4-12

g

g

I nt e l l i e nce Sca l e : Fourlh

ad m i n i,!rall'� )f l . 2 5 8 - 26 ! American 1 11d ia n c h i ld re n . 5 8 8

i

I n fo n 1 1 ,, 1 c V:J l ua l i o n .

9 1 4 . 928 9 1 4 . 928 Teachers

35 1 . 3 5 2

l�achL'''. -1-15