Persons in Context: Developmental Processes (Human Development in Cultural and Historical Contexts)

HUMAN DEVELOPMENT IN CULTURAL AND HISTORICAL CONTEXTS

Persons in context

HUMAN DEVELOPMENT IN CULTURAL AND HISTORICAL CONTEXTS General Editor: Urie Bronfenbrenner Associate Editor: Glen H. Elder, Jr.

Adolescent mothers in later life Frank F. Furstenberg, Jr., J. Brooks-Gunn, and S. Philip Morgan

Persons in context Developmental processes Edited by NIALL BOLGER, AVSHALOM CASPI, GERALDINE DOWNEY, MARTHA MOOREHOUSE

The right of the University of Cambridge to print and sell all manner of books was granted by Henry VIII in 1534. The University has printed and published continuously since 1584.

CAMBRIDGE UNIVERSITY PRESS Cambridge New York New Rochelle

Melbourne

Sydney

CAMBRIDGE UNIVERSITY PRESS Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, Sao Paulo Cambridge University Press The Edinburgh Building, Cambridge CB2 2RU, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521355773 © Cambridge University Press 1988 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 1988 This digitally printed first paperback version 2007 A catalogue record for this publication is available from the British Library Library of Congress Cataloguing in Publication data Persons in context : developmental processes / edited by Niall Bolger . . . [et al.]. p. cm. - (Human development in cultural and historical contexts) Includes index. ISBN 0 521 35577 X 1. Child development. 2. Socialization. 3. Child psychology. 4. Child psychobiology. I. Bolger, Niall. II. Series. RJ131.P39 1988 303.3'2-dcl9 87-36765 ISBN-13 978-0-521-35577-3 hardback ISBN-10 0-521-35577-X hardback ISBN-13 978-0-521-03584-2 paperback ISBN-10 0-521-03584-8 paperback

Contents

Preface Contributors

page vii xii

1.

Development in context: research perspectives Niall Bolger, Avshalom Caspi, Geraldine Downey, and Martha Moorehouse

2.

Interacting systems in human development. Research paradigms: present and future Urie Bronfenbrenner

25

Children, families, and communities: ways of viewing their relationships to each other Jacqueline J. Goodnow

50

Human development and social change: an emerging perspective on the life course Glen H. Elder, Jr., and Avshalom Caspi

77

3.

4.

1

5.

Family process: loops, levels, and linkages Gerald R. Patterson

114

6.

On the constructive role of problem behavior in adolescence Rainer K. Silbereisen and Peter Noack

152

7.

The sociogenesis of self concepts Robert B. Cairns and Beverly D. Cairns

181

8.

Putting persons back into the context Daryl J. Bern

203

9.

How genotypes and environments combine: development and individual differences 217 Sandra Scarr

vi

Contents Author index Subject index

245 251

Preface

This volume is a product of a study group convened by the Society for Research in Child Development at Cornell University to assess the progress that social scientists have made in understanding the processes linking persons and contexts in the course of development. The contributors to this volume represent the disciplines of developmental, personality, and clinical psychology; behavioral genetics; and sociology. They are also identified with a wide range of methodological approaches including longitudinal studies, laboratory experiments, field observations, and the sequential analysis of social interactions. In addition, each contributor has made a distinctive theoretical and empirical contribution to an interactional perspective on human development. In preparation for the conference, the participants were asked to circulate working papers summarizing their approaches. To stimulate discussion during the conference, a less usual procedure was adopted. Each participant presented the viewpoint of another group member rather than a commentary on his or her own work. The successful dialogue that emerged in the course of the conference continues in this volume. The central theme of the volume is how to relate different environmental contexts to one another and to individuals in the course of development. In the first chapter, Niall Bolger, Avshalom Caspi, Geraldine Downey, and Martha Moorehouse provide a map of the territory to be explored as they enumerate the various linkages between different environmental influences and individual development. Environmental influences range from social conditions at the macro level, like a prosperous or depressed national economy, to the immediacy of a child's playful exchange with a parent at the micro level. Chapter 1 reviews different perspectives on the processes that operate within and between these different levels and links them to one another and to other recent advances in the study of development in context. The chapters that follow gradually shift from macro to micro levels of environmental analysis. Earlier chapters concentrate on the stable and changing vii

viii

Preface

aspects of the macrosocial environment and how they influence persons in their proximal environments. Later chapters focus on developmental processes operating at the proximal level, bringing into the foreground the daily lives of individuals, the activities they pursue, their interactions with one another, and their personal attributes. The volume concludes with a discussion of the contributions that individuals make to their own development as they select and shape the very environments they experience. This organizational framework allows the contributors to provide a variety of perspectives and empirical accounts of the essential features of the developing person, the changing environment, and the processes that link them across time. Urie Bronfenbrenner describes existing and emerging research paradigms and discusses their strengths and weaknesses as tools for analyzing development in context. He argues that researchers must make use of more complex conceptual schemes that have the power to explicate relationships among person, process, and context. Especially needed, in his view, are "person-process-context" models that can determine whether a developmental process, like the facilitation of the child's development by parent-child activities, varies as a function of both the context and the characteristics of the individuals involved. Equally important are designs that reveal the effects of stability and change in environmental systems by incorporating a temporal dimension in their analysis of the environment. Bronfenbrenner further reminds us that we need to be equally evenhanded in our treatment of the individual. Calling for new conceptions of developmental outcomes that are integrative and sufficiently complex, he outlines requirements for both a subjective, experiential component and an objective, behavioral component. Jacqueline Goodnow pursues this theme in chapter 3. She considers how children's access to different settings, actual and perceived, changes with development and is facilitated or constrained by the nature of their communities. She critically evaluates some widely held but untested assumptions about desirable relationships among children, their families, and their communities. These assumptions include the ideas that exposure to a wide variety of settings enhances development, continuity between settings is desirable, and the world is open for exploration and enjoyment by every individual. She then examines the developmental processes underlying the relationships among children, their families, and their communities and emphasizes models that are useful in conceptualizing the nature of access between people and settings. Several questions receive special attention here: Is access to information and to settings restricted or freely available within the larger community? How do families conceive of the world outside the home and how do they convey these beliefs to their children? What contributes to individual differences and developmental changes in children's perception of access to settings and social roles? In chapter 4, Glen Elder and Avshalom Caspi bring a life-course perspective

Preface

ix

to their analysis of human development in a changing society as they consider how historical contexts shape individual life patterns and how individual development is related to social transitions in the age-graded life course. Examining social changes in the Great Depression, they document the antecedents and lifecourse implications of problem behavior for children growing up in stressful times. Their model of control cycles in the evolving life course details some of the complex causal linkages between changing social conditions, dynamics of family relationships, personality characteristics of family members, and the later personality development and life patterns of these individuals. In particular, this work identifies factors that mediate and moderate the influence of stressful times on children's lives, features that are also examined and elaborated in chapter 5 by Patterson. Gerald Patterson is not only a scientist who studies self-sustaining destructive interactions but a social engineer whose clients expect him to intervene and deliberately produce discontinuity and change. His chapter provides a methodological framework that illustrates conceptual points about the development and maintenance of family interaction patterns. It also shows how analyzing the microdynamics of social interactions in the family can help describe and explain enduring patterns of antisocial behavior. This perspective on development raises issues regarding the factors that elicit specific instances of action and reaction, what Patterson has termed microsocial processes. His model incorporates variables outside the immediate family environment and connects these to family processes in order to account for behavioral change along a discernible trajectory. Patterson's data also support the notion, advanced by several other contributors to this volume, that children's difficulties in one context are likely to produce difficulties in other contexts. In this chapter he elucidates the mechanisms underlying these developmental progressions. Rainer Silbereisen and Peter Noack's conceptual framework, "development as action in context," denotes a model of development that entails two critical issues: (a) Development is seen as the outcome of a person's own intentional, goal-oriented action aimed at adjusting individual goals and potentials to contextual demands and opportunities; (b) such action produces not only change in the individual, but change in the context of development as well. The contextual changes thus induced continually provide opportunities for new action aimed at further development. In chapter 6, these authors apply this distinctive European "look" to a developmental phenomenon of worldwide concern: the interplay between person and context in the psychogenesis of substance use in adolescence. Using a multilevel assessment strategy that integrates field and observational data with a cohort-sequential design, Silbereisen and Noack examine the relationship between self-esteem, cathartic leisure-time settings, and substance use in early and middle adolescence. Robert Cairns and Beverly Cairns consider the question, "Why would a

x

Preface

woman who is destitute, elderly, ill, widowed, hungry, black, and living in abject circumstances in an urban ghetto, report high life satisfaction?" "Why not?" sums up their response. That is, why do we assume that convergence between self-evaluations and the evaluations of others is normative and adaptive? The authors present a strong case for an alternative view based on a developmental analysis of the functions of social cognitions concerning the self. Their core premise is that adaptation is not synonymous with accuracy. Instead, modestly inflated views of one's competence may serve to enhance both well-being and interpersonal adaptations as they provide room for plans, dreams, as well as recourse from discouragement. When, then, would correspondence occur? Again, when it is developmentally functional for the individual. In their chapter, the Cairnses summarize propositions concerning when and why reports from various sources converge or diverge and draw on empirical evidence from their unique multimethod, multimeasure, longitudinal study of childhood and adolescence. Daryl Bern argues that development cannot be understood fully unless one adopts a perspective that attends to the role of the person's differentiating psychological characteristics. A personologist, Bern argues that our fundamental scientific task is to convert observations of particular persons behaving in particular ways in particular situations into assertions that certain kinds of people will behave in certain kinds of ways in certain kinds of situations. In chapter 8, he presents a commentary on research paradigms in personality psychology that complements Bronfenbrenner's commentary on research paradigms in developmental psychology, and concludes by providing guidelines for renewed attention to personological concerns in developmental research. Putting the person back into the context, Bern argues for a morphogenic theory of response-style variables that could explicate how the person's intrapersonal patterning of styles changes and develops in interaction with particular kinds of environments over time. In the final chapter, Sandra Scarr also addresses the problem of individuality and development by posing the question, "How do organisms and environments combine to produce human development?" Scarr suggests that to understand development and individuality we need a theory about the ways in which people determine their own experiences. She proposes that the individual's genotype both determines the impact that environments will have on him or her and, increasingly during the course of development, drives the individual's own selection of environments. From infancy to adolescence, children become increasingly effective in creating environments that are compatible with their motivations, needs, and abilities. Moreover, Scarr takes issue with the proposition that genotype-environment interactions have major importance in explaining variability and proposes instead that genotype-environment correlations have

Preface

xi

greater effects on development and individuality. According to this view, the responses that individuals elicit and evoke from others and the actions they take to select and shape their environments are correlated with their genotypes. Indeed, although the search for aptitude-treatment interactions as well as genotype-environment interactions has produced uneven and unreliable results, evidence accumulating in the field of behavior genetics provides a good deal of support for Scarr's proposed model. In combination, the contributions to this volume provide a useful beginning for new considerations of how persons and contexts are linked in the course of development. They show that we need not decide between the study of development out of context and the study of context without development. They confirm the promise and importance of studying development in context. A successful group endeavor requires a gathering of open-minded, curious, and energetic conferees. This they were. It also required the support and assistance of Gerri Jones, who managed the logistics of the conference. Finally, we wish to acknowledge the support of the Society for Research in Child Development in bringing the conference and this volume to fruition. Niall Bolger Avshalom Caspi Geraldine Downey Martha Moorehouse

Contributors

Daryl J. Bern Department of Psychology Uris Hall Cornell University Ithaca, NY 14853

Beverly D. Cairns Department of Psychology Davie Hall University of North Carolina Chapel Hill, NC 27514

Avshalom Caspi (coeditor) Department of Psychology William James Hall Harvard University Cambridge, MA 02138 Geraldine Downey (coeditor) Institute for Social Research University of Michigan Ann Arbor, MI 48106 Glen H. Elder, Jr. Carolina Population Center University of North Carolina Chapel Hill, NC 27514 Jacqueline J. Goodnow School of Behavioural Sciences Macquarie University Sydney N.S.W. 2109 Australia Martha Moorehouse (coeditor) Department of Psychology University of California Santa Cruz, CA 95064

Robert B. Cairns Department of Psychology Davie Hall University of North Carolina Chapel Hill, NC 27514

Peter Noack Institute for Psychology Hardenbergstrasse 28 Technical University of Berlin D-1000 Berlin 12 Federal Republic of Germany

Niall Bolger (coeditor) Institute for Social Research University of Michigan Ann Arbor, MI 48106 Urie Bronfenbrenner Department of Human Development and Family Studies MVR Hall Cornell University Ithaca, NY 14853

xn

Contributors Gerald R. Patterson Oregon Social Learning Center 207 East 5th Avenue, Suite 202 Eugene, OR 97401 Sandra Scarr Department of Psychology Gilmer Hall University of Virginia Charlottesville, VA 22903

xiii Rainer K. Silbereisen Department of Psychology Otto-Behaghel-Strasse 10 University of Giessen D-6300 Giessen Federal Republic of Germany

Development in context: research perspectives

Niall Bolger, Avshalom Caspi, Geraldine Downey, and Martha Moorehouse

Development in context Individuals develop in an ever-changing environment, and students of human development must do more than simply acknowledge the interaction between the two. The contributors to this volume assume the task of translating this homily into an integrative perspective on how individuals and environments are linked in the course of development. In this introductory chapter, we outline the orienting principles that inform the study of development in context and that also unify the contributions to this volume. We begin with a representation of the environment in which development takes place and then examine the role of individuals as producers of their development. Representations of the environment The environment in which individuals develop consists of a complex system of physical, social, cultural, and historical factors that interact with each other and with the developing individual. Most psychologists agree that these factors are influential in regulating development, yet they persist in viewing the environment as an undifferentiated set of factors. The difficulty with this position can be appreciated best by considering where developmental theory would be if we failed to distinguish between cognitive, affective, and behavioral sources of individual change. Would a theory of development that details the structure and process of environmental sources of change add to our understanding of development? Several theorists have suggested as much. The most consistent advocate of a theory of development in context is Bronfenbrenner (1979, this volume). Following Lewin's (1943) field theory and a general systems approach (von Bertalanffy, 1969), Bronfenbrenner has devised an ecological model for the systematic analysis of environmental influences on development. 1

2

N. Bolger, A. Caspi, G. Downey, and M. Moorehouse

Bronfenbrenner conceptualizes developmental contexts as nested, hierarchical structures with influences ranging from those located in the distal, macrolevel setting (e.g., class, culture) to those located in the proximal setting (e.g., family, school, work). The macrosystem comprises the social, cultural, and political structures of society - including its laws, norms, and customs - that define the character, structure, and functioning of proximal settings. According to Bronfenbrenner, three types of proximal settings can be distinguished effectively: the exosystem, microsystem, and mesosystem. The exosystem consists of face-to-face settings in which the individual does not actively participate but that can affect or be affected by the developing person (e.g., the parent's workplace). The microsystem is that part of the environment that an individual is in contact with and can interact with directly in daily life during a certain period of time (e.g., the family, school, leisure setting). It contains those settings in which people can readily engage in face-to-face interaction, and these patterns of interaction, as they persist and evolve through time, constitute the vehicles of behavioral change and individual development. The mesosystem consists of relationships between major microsystem settings at a particular point in the individual's development (e.g., the home-school relationship as it affects children's behavior). This basic conception introduces more conceptual rigor to what is often viewed by psychologists as an amorphous and unclearly differentiated set of influences. It implies that individuals are not independent and that, over and above any commonalities at the biological level, they share experiences and influences owing to their membership in higher-level systems, such as families, organizations, communities, societies, and historical periods. Indeed, the surprises that emerge when we examine the course of development in different cultural settings and in different social strata continue to reinforce the point that developmental processes cannot be fully understood without information on the many levels of social organization in which human action and development are embedded (LeVine, 1981). Bronfenbrenner's ecological model provides a useful guide, offering several analytic challenges that are addressed in this volume. Our first task is to examine the linkages between different levels of the environment. Assuming a hierarchical organization of the environment, we can expect influences on development to originate at different levels of analysis. For example, all individuals are embedded in a proximal system that is subject to change over time, the family. And the family, in its functioning as a socializing entity, also depends on certain characteristics of the larger social system in which it is embedded. Psychologists, however, often avoid probing how these higherlevel structures influence lives. Of course, before we can understand developmental processes, we must understand each level in terms of its own principles of organization. Proximal

Development in context: research perspectives

3

settings, such as schools and families, have unique influences on development that cannot be explained solely in terms of the higher-level social structures in which they are embedded or in terms of the characteristics of the individuals in the immediate setting. Thus, the second challenge for developmentalists is to account for and describe the unique effects of these proximal settings. A third challenge is to determine how development is affected by the features of the immediate environment, which covary because they are part of the same macrosystem. For example, neighborhood and school quality, family structure, and the quality of parent-child interactions are likely to covary because of the common influence of a higher-level system in which they are embedded: social class. Considered individually, each of these factors may predict certain developmental outcomes. But these factors come as a package, and therefore we must examine the joint effects of multiple, simultaneous environments on development. It might be said at this point that we are staging a play that has all sets but no character parts. Of course, a realistic theory of development in context should lead us to expect active individuals who possess the capacity to influence their own development. A fourth challenge, then, is to determine how individuals shape the very settings that influence their development and psychological growth. A fifth challenge is to deal with the fact that both persons and environments change over time. Developmentalists have traditionally focused on changes over time in individuals and have neglected the changing configurations of the environments in which individuals are embedded. To study behavior as it typically develops in natural circumstances, however, we must examine the mutual interplay of individuals and their environments over time, a feature that Bronfenbrenner (this volume) has now explicitly recognized by adding a chronosystem to his ecological model. A sixth challenge is to assess the implications of the above issues for interventions aimed at enhancing human development.

Macrosocial influences on development We cannot begin to understand human development without considering opportunities and constraints in the social structure and the general culture pattern or without taking into account possible and probable sequences arising from social and historical change (Elder, 1975; Riley, 1985). As Mills (1959) noted: The biographies of men and women, the kinds of individuals they have become, cannot be understood without reference to [social and] historical structures in which the milieux of their everyday life are organized. . . . [They] carry meaning not only for individual ways

4

N. Bolger, A. Caspi, G. Downey, and M. Moorehouse

of being, but for the very character - the limits and possibilities of the human being, (p. 157) How are social structural aspects of the environment related to the psychological characteristics of individuals? Our understanding of the relationships between macrosocial phenomena and individual development has been impeded by a general failure to explicate and test theories of how these linkages occur. An adequate analysis involves attending to three analytic principles. First, we must adequately understand the nature of the social structure or system in question. As we shall see later in the chapter, complex social phenomena almost always have multiple dimensions or components, and we must identify those that are most relevant for understanding observed differences in individual functioning. Second, we must recognize that the effects of social structures or systems are transmitted through stimuli that impinge directly on the individual. Thus, a major task in the study of social structure and social change is to trace how macrosocial processes affect increasingly smaller social structures and ultimately those proximal settings that directly impinge on the individual. Third, we must understand how these proximal settings affect individual development. Unfortunately, few studies to date have incorporated these analytic principles (Elder, Caspi, & Burton, 1988). To do so, researchers will have to seek more interdisciplinary cooperation, especially because they are apt to find themselves in need of tools that are not adequately developed within a single discipline. The most impressive effort along these lines is that of Kohn and his associates, who have studied the influence of social class on individual behavior and personality for over two decades (Kohn, 1977; Kohn & Schooler, 1983). What has been particularly exciting about Kohn's work is the attempt to show that the influence of social class emanates from class-associated conditions of life, notably occupational conditions. By outlining a multidimensional model of stratification and specifying the structural imperatives of jobs that define the social and psychological realities of work, Kohn and his colleagues have shown that certain features of work (e.g., the amount of autonomy, the degree of routinization, and the amount of substantive complexity experienced on the job) give rise to values of either autonomy or conformity in workers. On the one hand, the greater the freedom experienced on the job and the more complex and challenging the work, the more likely is the worker to place a high value on self-direction. On the other hand, the more constraining and routine the work, the more likely is the worker to value conformity. Kohn has also expanded his examination of the consequences of the structural imperatives of work to a wider range of psychological variables, such as intellectual flexibility. In a series of studies, he has shown that substantively

Development in context: research perspectives

5

complex work positively affects the intellectual flexibility of workers and contributes to the development of autonomy values. A core feature of this research program on job conditions is the general premise that work influences personality and behavior through a type of learninggeneralization process by which job influences are translated to other domains of life, such as marriage, parenting, and leisure. Men should thus function differently in the father role according to whether they have experienced job conditions that foster self-direction. This hypothesis has been successfully tested in both the United States and Poland. In both societies, family social position appears to influence the valuation of self-directed values for children through the occupational self-direction of parents. Parental values, then, tend to be extensions of the modes of behavior that are functional for parents in their occupational spheres. This transformation of work-generated values into parental practices constitutes an important link between the occupational and family contexts of socialization. The model and tests are not complete, but they have given us an unparalleled account of how social structural conditions are linked to aspects of psychosocial functioning in different contexts. To extend Kohn's model to research on maternal employment as it affects children's development would seem to be a natural priority. Instead, maternal employment has simply been added to the list of what Bronfenbrenner has termed "social address" variables as researchers continue to compare children whose mothers hold jobs outside the home with those whose mothers do not. The assumption here is that a mother's work-related absence, rather than her work conditions, is the critical variable influencing child development. Moreover, even when features of the mother's work status have been considered, her personal satisfactions have been emphasized and the structural imperatives of her work conditions have been ignored. Moorehouse (1986) has attempted to identify some of the influential aspects of maternal employment in a study of first graders and their families. This study related two aspects of the mother's employment - her work hours and the continuity of her work history - to parent-child activities and to the child's school performance and social adjustment. The results suggest that discontinuity in employment, especially an increase to full-time hours, is associated with more negative social and cognitive school outcomes, but only when joint mother-child activities are infrequent. The findings concerning discontinuity in employment are important for several reasons. Previous studies of maternal employment have sampled primarily women working full-time and on a stable basis. However, since discontinuity rather than continuity is the more typical pattern of employment for women with

6

N. Bolger, A. Caspi, G. Downey, and M. Moorehouse

school-aged and younger children, this approach may have led us to misidentify some of the common effects of maternal employment on children's development (Moen, 1985). More important, this study highlights the shortcomings of developmental research on social structure that focuses exclusively on men's class position and their job conditions. The social class position of women has traditionally been defined in relation to their husbands' occupational status. This definition is valid for women who are homemaking exclusively, but it is inappropriate for the majority of women who are now in the labor force. Indeed, a pressing issue for research on social structure and socialization is to examine how features of work are translated into the experiences of both fathers and mothers and how the combined effects of these experiences are transmitted to the next generation through family socialization practices. These limitations notwithstanding, Kohn's important contribution has been to stress a more general problem with the use of structural concepts, such as social class, that psychologists clumsily borrow from sociology. Much has been written in developmental psychology about the effects of social class as measured by a single composite index combining education, occupation, and income. Aggregating variables as such, however, tends to obscure what may often be differential effects of these component variables. Moreover, Kohn's research program provides an important reminder that the characteristics associated with different social addresses are quite wide ranging. Much has been written about differences in the quality and quantity of parentchild interactions across social addresses. The Gurins (Gurin & Gurin, 1970; Gurin, Gurin, Lav, & Beattie, 1969) have argued that it is more difficult for working-class parents to transmit certain values to their children owing to classbased constraints that may limit their ability to ensure their children are rewarded consistently for certain efforts. In addition, the environmental characteristics associated with different social class backgrounds range from the quality of housing conditions to the nature of toys and the number of books available to the child. Thus it may be that these characteristics exert a direct influence on the perceptual, cognitive, and motivational development of children by providing a substrate of stimulation beyond that produced through interpersonal interactions (e.g., Wachs, 1979; Wohlwill, 1983). Structural conditions, then, may impinge on the proximal settings of development in multiple ways and along a variety of dimensions. Indeed, as Elder and Caspi emphasize in their discussion of social change and human development in chapter 4, the crucial issue remains that of specifying the components and psychological processes linking structural conditions to individual development. To accomplish this task we need a more refined view of the proximal environments that impinge directly on the individual. Some advances along these lines have been made by recent research on schools and families.

Development in context: research perspectives

1

Proximal influences on development The number of proximal settings in a complex society is almost limitless. Here we focus on two settings, school and family, to illustrate problems and promises in research linking the proximal environment to individual development. The school as a developmental context. School socialization is by and large the result of direct instruction, buttressed by a system of social reinforcement, expectancy effects, and social comparison processes. But the quality of education obtained by children depends on more than just teacher-student interactions in the classroom. These interactions are structured by the type of school children attend and by the overarching educational policies of the counties and states in which the schools are located. In contrast to the early reports of Coleman (1966) and later Jencks and his colleagues (1972), recent research suggests that the school does matter, and what matters most is the quality of cultural and social organization, the attitudes, values, and mores. In an impressive study of 12 secondary schools (Rutter, Maughan, Mortimore, & Ouston, 1979), data from interviews and questionnaires completed by students, classroom observations, and official records were used to derive a "school process" score dealing with items such as style of classroom management, patterns of rewards and punishments, and student opportunities for responsibility and participation. The differences between schools were systematically and strongly associated with student outcomes in these schools, including disruptive behavior, truancy, damage to school property, and examination failure. It remains unclear just how the variables making up the "school process" score combine to create a school environment that fosters particular patterns of student behavior and academic progress, although Rutter and his colleagues suggest that these qualities determine the nature of the values and norms of behavior set by the school and thus affect students' motivation to behave in certain ways. As Rutter's work demonstrates, proximal environments have unique influences on development that do not emanate either from macrolevel social structures or from the additive characteristics of individuals in the school setting. School effects cannot be attributed entirely to ecological factors because publicly funded schools in the same catchment area differ with respect to school process variables. Moreover, school effects cannot be accounted for by selection factors whereby some schools admit a higher proportion of disruptive students. Rutter and his colleagues tackled this latter question by asking whether student characteristics at intake correlate as strongly with the school process variables as do student characteristics at the time of leaving school. The finding that student intake measures were only weakly correlated with school variables suggests that

8

N. Bolger, A. Caspi, G. Downey, and M. Moorehouse

school functioning and teacher performance were not shaped by the characteristics of the students admitted to the school. Rather, the direction of causal influence was from school factors to student behavior. A related body of research has examined the developmental implications, particularly in the area of social mobility, of the ways schools structure educational opportunities (Hauser, Sewell, & Alwin, 1976; Rosenbaum, 1976). In a case study of a school in a primarily white working-class town, Rosenbaum (1978) has shown that students' uninformed self-selection into educational tracks during high school constrains their long-term educational achievement and social mobility. Students who selected themselves into noncollege tracks jeopardized their chances of attending college not only because of the different training they received but also because of an implicit assumption that their performance on school tests was less credible than that of college-track students. This assumption was reflected in the lower weighting their grades received in calculating class rank. Rosenbaum's study illustrates that the social structure of proximal settings influences individual lives and also suggests that lack of access to information can affect developmental outcomes, a point elaborated by Goodnow in chapter 3. An issue that has unfortunately been neglected in research on schools as contexts for development concerns the relationship between school characteristics and other community and family factors, and how these jointly influence what happens in a given setting, such as the classroom. As Goodnow makes clear in this volume, we need to broaden our examination of person-environment interactions and assess the joint effects of the multiple settings in which the individual is an active participant. Linkages between developmental contexts. By applying person-environment models to the context of multiple environments, researchers can address fundamental questions about the extent to which settings containing the developing person conflict with or complement each other, and about the role of compatibility between settings in the process of development. Some of these issues are addressed in Hansen's (1986) study of the combined effects of family and classroom interactional styles on children's school performance. Family interaction patterns were classified as cohesive, coercive, or laissez-faire using definitions analogous to Baumrind's (1966) distinction between authoritative, authoritarian, and permissive styles. Teacher reports provided data for a similar classification of classroom styles. To measure academic performance, children's current grade point averages (GPA) were compared with their averages for the previous two years. Not surprisingly, the findings indicate that family style is the best predictor of GPA for any given year. However, it was the fit between classroom and family styles that predicted changes in GPA. A match predicted an increase in GPA whereas a mismatch was associated with a

Development in context: research perspectives

9

decrease in performance. This work suggests the importance of congruence between developmental contexts, but it does not take into account the characteristics of the children who occupy these settings. Promising work along these lines is found in Epstein's (1983) assessment of the match or mismatch between family, school, and personality characteristics and its effects on students' academic development. This longitudinal study followed adolescents from their last year of middle school through their first year of high school in order to examine the joint effects of two setting factors (the extent to which the formal organization of the school emphasizes student participation and the extent of decision-making opportunities for students at home) and a person factor (student self-direction and independence) on students' academic performance. The findings provide evidence for the benefits of the "fit" and for disadvantages of the "lack of fit" of persons and environments (French, Rodgers, & Cobb, 1974; Pervin, 1968; Thomas & Chess, 1980). For example, each type of student is rewarded in high-participatory schools, but those students who come from high-participatory families and who are also initially high in independence have the highest grades. In addition, highly independent students who are mismatched with both socializing environments (i.e., they come from low-participatory families and also attend low-participatory schools) are among those with the lowest grades. The findings are especially interesting because they show how research designs that incorporate multiple developmental settings can reveal the power of compensatory environmental experiences. Thus, we find that school environments are most important for highly independent students who come from families that do not emphasize participation in decision making. Epstein's study also illustrates some directions that research on home-school relations might profitably take and provides provocative findings from a promising research paradigm: the mesosystem component of Bronfenbrenner's (1979) ecological model that emphasizes the relationships between different developmental settings. Unfortunately, studies documenting the joint effects of processes occurring within and between two or more settings in which the developing person is an active participant are still in short supply (see chapter 3). The neglect of these issues is especially disappointing given the prevalence of "reconstituted" families and the attendant number of residential and school moves that today's children can expect to experience during the course of development. The family as a developmental context. The family is, of course, the context most closely associated with developmental processes. It is the first socialization context that most of us experience and it is the place where we develop our initial sense of self. Two advances in recent years have renewed interest in research on the family. First, the operational model of the parent-

10

N. Bolger, A. Caspi, G. Downey, and M. Moorehouse

child relationship has been expanded from dyadic to polyadic forms. Second, findings from human behavior genetics studies have converged on the conclusion that nonshared environmental influences represent a critical source of environmental variance for producing individual differences. The recent growth of family systems theories in developmental psychology, as well as the development of conjoint family therapies for alleviating children's psychiatric problems, has strengthened the interest in interactional processes and relationships within the family (Minuchin, 1985). The basic concept of systems approaches is that the family constitutes a social group in which the functioning of the whole is qualitatively different from the sum of its parts. This is so because properties of the family as a whole derive from the properties of the relationships between individuals in the family and not just from the characteristics of the individuals as separate persons (Hinde, 1979; Hinde & Stevenson-Hinde, 1988). Thus, the nature of the home environment and family dynamics can best be understood in terms of a synthetic framework that links family members in a network of interactions and personal relationships (Belsky, 1984). The first systematic studies to incorporate this framework documented the role of the father in the mother's treatment of the child (Clarke-Stewart, 1978; Parke, 1979). Another such study demonstrated that a child's classroom performance varies systematically as a function of whether or not the teacher had taught the child's older sibling, and if so how well that older sibling had performed (Seaver, 1973). Other recent examples are provided by the sibling rivalry or jealousy phenomenon, whereby children with particularly warm and close relationships with their mothers are more likely to develop negative reactions toward their younger sibling (Dunn & Kendrick, 1982), and by the sibling deidentification phenomenon whereby children in the same family react in ways that emphasize differences from their siblings (Schachter, 1982; Tesser, 1980). Findings from the literature on marital and parent-child relationships in a wide range of domains, including developmental psychology, family sociology, and clinical psychology, have also drawn attention to the interdependence of family relationships. For example, in a study of families with school-aged children, Brody, Pillegrini, and Sigel (1986) found different patterns of parent-child interaction in a problem-solving task among couples varying in marital quality. In the case of low marital quality, mothers tended to impose structure on the interaction, and fathers were more intrusive and provided less feedback about their children's problem-solving efforts. Other studies investigating different age periods and different domains of development support the general conclusion that positive marital adjustment facilitates parental sensitivity to the child (e.g., Feldman, Nash, & Aschenbrenner, 1983; Goldberg & Easterbrooks, 1984). The interdependence of family members is further illustrated in a study of the

Development in context: research perspectives

11

effects of a depressed adult on the psychological well-being of other adult family members. Coyne and his colleagues (1987) found that over 40% of adults living with a depressed person in the midst of an episode were sufficiently distressed to require intervention. In contrast, these elevated rates of distress were not found among adults living with a depressed person who was not currently in the midst of a depressive episode. Moreover, these elevated rates of depression were primarily mediated by the formidable objective and subjective burden placed on the family by the depressed person's behavior. In conjunction with findings regarding the impaired parenting of depressed women (e.g., Zahn-Waxler, 1981), these findings illustrate the influence of individual problems on the functioning of other family members through family relationships. Beyond the family, recent research has also documented the implications of social network membership for individual development. For example, Kessler and McLeod (1984) have proposed that women's increased vulnerability to depression may be due to their greater sensitivity to negative life events in the lives of other network members. There is suggestive evidence that women are more likely than men to hear about the life events of others and then to assume more of the burden associated with the latter's problems and crises. Emerging from these new microsystem models and findings is a more dynamic and differentiated concept of the immediate environment. Instead of studying the child in isolation from his or her family context, researchers are increasingly adopting family-based dyads and triads as their fundamental units of analysis (Hinde & Stevenson-Hinde, 1988). Such studies provide persuasive evidence that many developmental outcomes depend on properties of higher-order relationships rather than on properties of family members considered individually (Maccoby, 1982). This dependence is well illustrated by Patterson's (1982) model of violence among family members, suggesting that the density of coercive exchanges in the family covaries with intensity. He has labeled this process the anarchy progression. Given that one family member increases his or her coerciveness beyond the normal range, other members become increasingly at risk of being hit. In addition, as more dyads within the family become more coercive, the risk for physical assault increases commensurately. A view of the proximal environment as a system of interdependent members and relationships also has critical implications for intervention strategies. Consider the highly aggressive child. Traditional treatments have asked how we can change the characteristics of the child in order to reduce aggression. These individual-centered approaches have met with minimal success, however. Therefore Patterson (1982) asked a different question: What is the aggressive child's microsystem (i.e., family) like, and how can we change it? The success rate of

12

N. Bolger, A. Caspi, G. Downey, and M. Moorehouse

their interventions indicates the relevance of Bronfenbrenner's concern in chapter 2 that child development and personal functioning be viewed in a systemic and ecological perspective. Clearly, a conceptualization of the environment as an organized system of parents and children interacting in multiple and changing environments demands an eclecticism that is open to new conceptual and methodological approaches (Applebaum & McCall, 1983). The extant evidence suggests that an emphasis on family relationships, as opposed to the structural characteristics of the family, may deliver new advances in our understanding of the family as a developmental context. Some of the most important findings about environmental influences have also emerged from behavior genetics research (Plomin, 1986; Plomin & Daniels, 1987). This research converges on the conclusion that, for many variables of interest, the correlations between siblings in the same family are not much higher than those between nonrelated individuals. Scarr and Grajek (1982) make explicit the implications of these findings: Upper middle-class brothers who attend the same school and whose parents take them to the same plays, sporting events, music lessons, and therapists, and use similar child rearing practices on them are little more similar in personality measures than they are to working class or farm boys, whose lives are totally different. Now, perhaps this is an exaggeration of the known facts, but not by much. Given the low correlations of biological siblings and the near zero correlations of adopted siblings, it is evident that most of the variance in personality arises in environmental differences among siblings, not in the differences among families, (p. 361) These findings are rich in implications for research and theory. In terms of offering a practical guide to research, they suggest that family studies should observe more than one child per family in order to identify environmental factors that make children in a family so different from one another. Traditional environmental research has attempted to relate measures of the family environment to measures of the behavior of one child per family, but the yield from these efforts has been disappointing (see Maccoby & Martin, 1983, for an assessment). It may make more sense to focus on environmental sources of differences between children in the same family, for these differences in experience (and perhaps perception) appear to be the factors that drive behavioral development. The importance of nonshared environments also suggests a need for a theoretical reconceptualization of environmental influences on development. As Scarr points out in chapter 9, individual differences in adoption and twin research arise from genetic and environmental differences within families, not from environmental differences between families. Our psychological theories, however, address only between-family differences. To develop more powerful accounts of environmental influences we will have to identify experiences that are not shared

Development in context: research perspectives

13

by family members and then relate these nonshared environmental factors to differences in siblings' behavior. Thus far, only a few relevant studies have been reported (e.g., Daniels, Dunn, Furstenberg, & Plomin, 1985). These studies suggest that siblings in the same family experience different environments and that differential parental treatment is related to differences in siblings' adjustment. But if, as this evidence suggests, most of the environmental variance that affects behavioral development is of the nonshared variety, the foremost problem is to determine whether nonshared environmental variance is systematic or stochastic, and, if it is systematic, to ascertain what gives rise to the different environments children experience. Are the relationships between nonshared environments and behavior mediated by the individual? As we shall see in the next section, a good deal of evidence has accumulated for child-directed socialization effects. The individual's effects on the environment An important process linking individuals to their environments is provided by the individual's effects on the environment. Indeed, the relative weights assigned to Lewin's formulation, B = f(PE), have undergone a great deal of change in the last several decades. We began with an oversocialized conception of human nature that assigned priority to the environment, B = f(P F a m i l y (explosive, Relations etc.)

IB

G2 Generation

Problem ^C^ ^ Unstable Relations

Figure 4.4: Intra- and intergenerational processes linking problem behavior and unstable family relations

we would need to follow members of each lineage unit over their life span as they age and interact with one another. Intergenerational relations at a point in time could thus be viewed in relation to both individual aging and a relational history of self-other interaction. Our work in this area represents a compromise between the ideal model noted above and the realities imposed by the Berkeley data archive. The Berkeley Guidance Study archive includes members of four generations: The cohort of children (G3), born in 1928-9, which has been studied prospectively up to the middle years, along with their grandparents (Gl), parents (G2), and children (G4). Because not every generation was studied prospectively, we utilized ordered dyads as the primary domain for our intergenerational analyses and not the full sequence of generational units. In the first phase of our study, we examined grandparents (Gl) and parents (G2). This pattern was replaced by parents (G2) and children (G3) in the second phase. The last part of the picture centered on the grown-up children (G3) and their offspring (G4). Across the four Berkeley generations, we find substantial documentation of a reinforcing dynamic between problem behavior and unstable weak ties in the family. Aversive family patterns mediate the influence of unstable parents on offspring, and they are reproduced in the next generation through the development of offspring who are least able to sustain and nurture enduring relationships. From Gl to G2 generations, unstable personalities are reproduced in part through marital tension and parent hostility. This socialization environment linked unstable grandparents to a similar style of behavior in their middle-aged sons and daughters. Likewise, unstable parents in the G2 generation markedly increased the likelihood for ill-tempered difficult children by increasing marital tensions and arbitrary tension. In this cohort, the pattern was also fueled by stressful events associated with the Great Depression. The pattern finally repeats itself in the G3 generation during their active years of parenting. Indeed, weak

102

Glen H. Elder, Jr., and Avshalom Caspi

family ties were characteristic of adolescents in the fourth generation (G4) who ranked highest on lack of self-control. The intergenerational cycle of problem behavior and problem relationships resembles a general dynamic across the four Berkeley generations. The evidence reviewed here - on the enduring structure of early personality and on the intergenerational replication of aversive dynamics - provides insights about the origins, transmission, and sequelae of problem behaviors. The course of individual differences, however, is probabilistic; disadvantages - personal, familial, or environmental - are clearly not immutable, and for this reason "biographies cannot be written in advance" (James, 1890). Equally deserving of analysis, then, are the factors that enable individuals to surmount the limitations of a problematic childhood. What are the mechanisms that enable some men and women to turn around their fortunes and lives?

Turning points in the life course Change may be explained by the fact that at times in people's lives a fortuitous event is encountered that intervenes in the developmental process rather than simply confirms an established mode of relating to the world. Bandura (1982) has referred to this process in an essay on chance encounters. Some experiences channel behavior into predictable and unprofitable directions, but other experiences may direct people toward new life paths. Little research has documented such processes, however. A central aim of our research is to identify key turning points in the life course of children from deprived backgrounds. According to project research, three life transitions represent potential change or turning points in which the life course is recast: the quality of the first marriage, entry into college, and entry into military service. The quality of the first marriage bears directly on the development of a primary support system that can provide direction and soften the blows of adversity. The stabilizing influence of an understanding spouse on adults with a deprivational history has been documented by Robins (1966) and Quinton and Rutter (Quinton, Rutter, & Liddle, 1984; Rutter & Quinton, 1984). The role of college and military service has not received as much systematic attention, but both events may open up a new opportunity system of significant figures, incentives, and goals.

Entrance into higher education. As we follow the Berkeley and Oakland children from deprived families to the middle years, one early transition stands out above all others as the critical line of demarcation between achieve-

Human development and social change

103

ment and continuing disadvantage - entrance into higher education. College entrance opened the door in both cohorts to career lines with advancement opportunities, and college men from deprived families generally made the best of them (Elder, 1974; Elder & Rockwell, 1979). Even when college amounted to little more than a year or so, it made a significant difference in the career entry and progress of men who grew up in deprived homes. In the case of women, college entry created social opportunities for advancement through marriage. From early adulthood to middle age, the men who succeeded in entering college also managed to advance well beyond worklife expectations based on their background, rising to the occupational level by midlife of men from more affluent homes. But when coupled with less schooling, Depression hardship forecast an adult life that had some parallels with the 1930s. These men from deprived homes fared poorly in work (more instability, less advancement) than the nondeprived. They also ended up with a lower occupational position by midlife (age 40).

Military experience in the life course. Life trajectories are formed through the influence of early childhood on subsequent life changes and by events or experiences that enable people to alter the course of their lives. Unpromising origins may set in motion a cumulative sequence of life disadvantages, but they can also be surmounted through opportunities and initiative. An understanding of such change through military service presents a major challenge, especially in relation to the mechanisms that enable men to turn their lives around. As the Oakland and Berkeley men entered the service, they experienced a disjuncture in their lives that resembled the disjunctures of the Depression decade. For the most part, the war years and military service broke with the past, a past defined by hardship, frustration, and limited opportunity (Elder, 1986). Placed within this context, military service (from 1940 through the mid-1950s) acquired special historical significance in relation to life-course development and aging. Indeed, service in the military has emerged as a plausible source of life change. The more we probed the evidence on military service, the more certain we became of the need to consider this event along with that of Depression hardship (Elder, 1986; Elder & Bailey, 1986). In the transition to adulthood, military entry closes options, creates others, and postpones still others. It establishes a moratorium from decision pressures in a structured environment, and rearranges the traditional pathway to adulthood, delaying career progress and commitments. Early entry into the service, in particular, should extend the transition to adulthood, postpone family events, and

104

Glen H. Elder, Jr., and Avshalom Caspi

their responsibilities, and provide a route to future opportunities through separation from home, exposure to new places and people, and access to educational benefits. As a whole, military service may well have placed children from deprived circumstances on a more rewarding trajectory. Consider the notion that there is an optimal time for life events. In the case of military service, the optimal time seems to be early entry, as soon after high school as possible. On the "early" track, the service comes before family and serious work careers. In many cases, men who enter later do so after marriage, children and launching a career. Service time would be disruptive, but it could also build upon accomplishments and enhance continuity. From one perspective, among others, early military service represents a moratorium or holding operation relative to the age-graded career, a legitimated "time out" from commitment pressures. This concept of service time appears in the account of a young Berkeley man at age 30. Uncertainty about future directions made the army especially attractive. "I didn't know what I wanted, though I probably wanted to get away from home and community. I didn't know what to take in college. I recall arguing with myself about maybe I'd grow up and settle down a bit if I got into the service." Coupled with this moratorium experience is the developmental payoff of a physical break from home and greater independence, as well as exposure to male models in a structured program. At first thought, the timing of military service in wartime seems unlikely to be a matter of choice. However, military service does combine elements of choice and constraint. Within the limits of the draft and local draft board policy, young men could decide whether to enter immediately or to stay out as long as possible, perhaps aided by a student or work deferment. Assuming that life choices are made within the context of a life history, we hypothesized that early entry among the Oakland and Berkeley men is most strongly linked to military ideologies and felt deprivations that made joining up an attractive option. In both cohorts, the early entrants were characterized by attitudes favoring the military role and by disadvantages of one kind or another. Early entrants in the Oakland cohort were more likely than the late entrants to come from large families and from the less successful ranks in school. They were also more positive toward the U.S. military. In the Berkeley cohort, early entry was most common among those who held positive attitudes toward the armed forces, and among those who experienced at least one of the following disadvantages - an economically deprived background in the 30s, academic trouble or difficulty, and low self-esteem. These factors were less predictive of military service as a whole. Military service entailed a delay in the timing of family events (marriage, first birth, etc.) among veterans in both cohorts and especially among the early

Human development and social change

105

entrants. The late entrants were more likely to have entered the service with a wife and child. Military service also led to a delay in the completion of education. As a whole, the early entrants tended to follow a less conventional path of adult achievement than the late entrants, and this observation applies to both cohorts. The early joiners had less education and they did not fare as well on occupational achievement by midlife. However, they frequently acquired additional education after military service and managed to narrow the gap in life achievement relative to the late entrants. On the basis of background, the early entrants should have ranked higher on marital discord and divorce than the late entrants, given their more disadvantaged background. But we find no such difference in either cohort. Marital stability consistently favors the early veterans. An important explanation for this change in outcome involves the consequences of a large delay in marriage. The early veterans married at a much later time than did the nonveterans, a timing that increases the prospect of marital permanence. In both cohorts the developmental influence of military service appears among veterans who entered the service at a relatively early age. According to social and psychological evidence, these men show the greatest development and career achievement when measured from baseline. At midlife, a larger percentage of the early veterans linked military service to personal growth and improved life chances, when compared to the late veterans. Military service was clearly a turning point for a good many of these men, though a precise understanding of the mechanisms awaits more detailed studies. Control cycles and the evolving life course: a concluding note Human development in a changing society poses many demanding questions about the life course, such as how it evolves or changes. One approach to the study of change focuses on the societal or institutional level. A changing society alters the course of people's lives by changing social institutions and complex organizations. Major school reform is one example of such change (Featherman & Sorenson, 1984). The extension of mandatory public education to older age groups adds an element of standardization to the pathway from childhood to the adult years. Another approach focuses on individual maturation and aging, emphasizing the behaviors and choices of individuals and how these shape life pathways or trajectories (Scarr, and Silbereisen & Noack, this volume). Particular choices reflect individual socialization and development, as expressed through family, school, and neighborhood. Both of these perspectives are linked in a third approach. The evolving life course of individuals and families is a product of the interplay between individual

106

Glen H. Elder, Jr., and Avshalom Caspi

Resources Figure 4.5: Discontinuity between soaring claims and available resources (historical examples: initial stage of rapid economic growth, World War II)

and environmental changes. Age-graded schools establish a range of options for moving through the educational system. The option followed leads back to the individual. Some students qualify for the honors track, whereas others remain in general education. Studies in the Social Change Project have relied on this interactive approach to study families and individuals across time in rapidly changing environments. The connecting thread across levels of analysis involves the core idea of control, as initially developed by W. I. Thomas in his writings on crisis situations. Thomas (1910) viewed control of desired outcomes as a function of the changing relation between claims relative to resources. Control is achieved when resources make it possible to fulfill claims. In Thomas's formulation, resources include the potential means to claim satisfaction, both psychological and social. A widening gap between claims and resources creates a disturbance of habit; habitual ways of behaving no longer meet expectations. Figure 4.5 shows how this gap may result from soaring claims during periods of rapid economic growth. Even in the late 1930s, aspirations rose quickly and spurred feelings of frustration. Durkheim (1951, p. 248) refers to this dynamic as the "malady of unlimited aspiration." The more one has, the more one wants. Figure 4.6 shows that a gap may also arise when resources decline sharply, for example, through income loss during the Great Depression. Both aspiration gain and resource loss entail a loss of control over desired outcomes, but the solution to each problem differs. Resource loss requires efforts to create more income or resources; it also calls upon reductions in level of expectation and expenditure. Soaring aspirations call for limits on demand, as well as efforts to increase resources or income. The same dynamic applies to individuals. Both of these new conditions, from soaring aspirations to loss of resources, represent a transition to a new situation and a loss of control potential. They also represent well-known disjunctures between family or individual change and the

Human development and social change

Ti

107

T2 Claims-,

§Lack of control! -"Crisis"!

1 Resources^ Figure 4.6: Loss of control: Discontinuity between claims and resource loss (historical examples: economic depression, severe inflation)

surrounding environment. In our studies of Depression families, adaptations to drastic income loss were efforts to regain control potential, to regain a balance between claims and resources in the achievement of desired outcomes. The adaptive options available to families were determined by the times (for example, the public aid option changed markedly over the 1930s). The option selected had much to do with what families and individuals brought to the Depression decade, both psychological and social resources and vulnerabilities. The choice also had much to do with the nature of the economic crisis or loss. Some men lost income through unemployment, while others lost similar amounts through reductions in work week or through loss of customers. The jobless sought jobs as many of their wives and children earned small wages on odds and ends; the shortweek condition led men to carry more than one job, if possible; and customer loss motivated efforts to retain customers, for example, in providing larger charge accounts. What we see in family and individual adaptations is a process by which the social life course is constructed. Loss of control sets in motion a life-course dynamic involving efforts to regain control. Cases of soaring aspirations and loss of resources can be found across the programmed and idiosyncratic transitions of the life course. One of the better examples comes from the literature on marriage and the pitfall of idealization. Rising expectations of what marriage will offer place the new relationship under greater strain and increase prospects for disillusionment. The profound and often devastating economic loss following divorce illustrates the implications of resource loss. Retirement also provides examples of this disparity. But perhaps even more relevant to life transitions is the loss of control potential that, as shown in Figure 4.7, arises from discontinuity between resources before and after a transition, or between claims before and after an event. In Thomas and Znaniecki's landmark study, The Polish Peasant, discontinuity is vividly illustrated by the process of migration to America from the

108

Glen H. Elder, Jr., and Avshalom Caspi

Claims-^

WLack of control! -"Crisis"!

-Resources*^ 'Some inappropriate for adaptive requirements of new situation.

Figure 4.7: Loss of control: discontinuity between resources before and after a transition (historical examples: rural European immigrants in urban America)

feudal countryside of 19th-century Poland. Immigrants brought skills to the new land that were not useful in an urban, industrial environment. The same could be said for claims or aspirations. Unplanned transitions in the life course often place individuals in predicaments where they lack the essential skills to cope effectively. Even with education for prospective parents, the birth of a first-born places them in a situation where they feel ill-equipped to meet expectations, theirs and others. Within the age-graded life course, transitions to new roles entail some loss of control potential and prompt efforts to regain control through adaptations of one kind or another. A predictable transition does not ensure control over desired outcomes. The control cycle dynamic in constructing the life course rests on the link between losing control and efforts to restore control over life outcomes. This connection has been documented by studies of personal control and reactance behavior. Reactance feelings occur whenever one or more freedoms or expectations are threatened or eliminated. Such feelings motivate efforts to regain control. The Brehms (1982, p. 375) refer to the overwhelming evidence for such motivation in their review of the field. "A problem threatens control, whether it is a matter of understanding, attitudinal disagreement or explicit loss of control. It is the threat to control (which one already had) that motivates an attempt to deal with the environment. And the attempts to deal with the environment can be characterized as attempts to regain control." On the larger screen of the life span, defining the immediate situation (see Figure 4.1) has much in common with Thomas's ideas regarding life planning and life organization. Life plans are based on concepts or definitions of the future life course. Revised thinking about the future prompts new life-course planning the situation has changed. As a means of controlling social reality for personal needs, the individual in Thomas's analysis devises general schemes of situations with goals, a life organization that provides a plan for subsequent action. This

Human development and social change

109

organization represents a continuing "project" of construction and reconstruction as a new experience is encountered that cannot be assimilated. The more drastic the change in life experience, the more one's life organization must be recast and the more the transition marks a probable turning point in trajectory. Major transitions, then, are times that increase the likelihood of a reorientation in life structure, direction, and purpose. They are times that make life planning a conscious act. Four general points from our studies summarize the relation between control cycles and the evolving life course: 1. The disparity between claims and resources may occur through rising claims, declining resources, and a discontinuity between acquired and needed resources. 2. The experience of losing control over one's life situation evolves from the preceding disparity. 3. Efforts to restore control or alternative modes of resignation involve adjusting claims, resources, or both in terms of their relation. 4. The life course may be altered through new lines of adaptation and their consequences. The precipitating event for this process is an event that substantially changes the balance between claims and resources, the actor's control potential. Options depend on conditions - the structured situation; responses to life change and the loss of personal control entail choices among given options, and this constraint is one way that a social institution shapes the life course. Action alternatives also depend on the resources people bring to the new situation. Control emerges as a central theme across many examples of linkages in the Social Change Project. We have seen how the connection between "losing control and life disadvantage" derives from a cumulative dynamic in which selfdefeating efforts to regain control produce even greater life disadvantages. Force within the family initiates a dynamic that can eventually make it ineffectual; force generates counterforce. Another example comes from the decision to enter military service, which can be viewed as one element of the dynamic in regaining a sense of control - an effort to gain the mastery and recognition so long denied the children of misfortune. Conclusion This chapter describes a program of research that began 25 years ago in the archives of the Institute of Human Development on the campus of the University of California, Berkeley. Although few analysts at the time even used the term "life course" in their research, the data archives of the Institute called for an approach that is now labeled the life-course perspective, an approach that views people and families over time as a dynamic process in context. The unparalleled

110

Glen H. Elder, Jr., and Avshalom Caspi

archives of the Institute served as a mentor in life-course thinking, as did the intellectual currents of the time. The approach we have followed in this program of work reflects the times as much as it matches the requirements of the problems at hand. Sociological insights are fashioning a distinctive approach to human development by relating it to social structure and history. Twenty years ago we saw this imagination at work in probing differences between middle- and working-class parents on values and behavior (Kohn & Schooler, 1983). More recently, we see its guiding hand in shaping a view of development across the life span and in specific historical settings. Growing awareness of an ever-changing environment is giving overdue visibility to matters of time, context, and social dynamics in the study of human development. Problems of social change and human development have gained significance as research foci, along with the concept that people make social history even as they are influenced by it. Views of this dialectic have emerged in the form of a life-course perspective that is based on the premise that aging entails interacting processes (social, psychological, and biological) from birth to death, and that life-course variations among individuals and cohorts are shaped by historical conditions. The dialectical perspective of life-course analysis, relating social history and life history, brings a number of issues to the fore that call for fresh thinking. Unfortunately, in most studies, structural and psychological variables still have a timeless quality. Social structures are seldom rooted in historical contexts that give them particular meaning, nor are they linked to psychological characterizations of the individual. Yet we know that the psychological effects of a social structure depend on historical context. Three emphases distinguish our theoretical orientation to the study of lives from the usual study of life-span development and aging. The most distinctive feature is our emphasis on social change and its effects on individual behavior and family patterns. Related to this theme is attention to linkages between social change and life patterns, and to the effects of differential life paths on developmental processes. A third feature is the focus on intracohort processes, as opposed to comparison of whole cohorts. At present, life-span research and developmental psychology are largely descriptive, relying on cohort comparisons instead of intracohort comparisons and offering little understanding of social change, the life course, and their relationship. Greater social science imagination (Mills, 1959, p. 12) is needed to "grasp history and biography and the relations between the two within society."

References Bandura, A. (1982). The psychology of chance encounters and life paths. American Psychologist, 37, 747-55.

Human development and social change

111

Barrett, J. E. (Ed.). (1979). Stress and mental disorders. New York: Raven. Belsky, J. (1984). The determinants of parenting: A process model. Child Development, 55, 83-96. Bleuler, M. (1974). The offspring of schizophrenics. Schizophrenia Bulletin, 8, 93-107. Block, J. (1971). Lives through time. Berkeley, CA: Bancroft Books. Block, J. H., Block, J., & Morrison, A. (1981). Parental agreement-disagreement on child-rearing orientations and gender-related personality correlates in children. Child Development, 52, 965-74. Brehm, S. S., & Brehm, J. W. (1982). Psychological reactance: A theory offreedom and control. New York: Academic Press. Bronfenbrenner, U., & Crouter, A. C. (1983). The evolution of environmental models in developmental research. In P. Mussen (Ed.), Handbook of child psychology: History, theory, and methods (Vol. 1). New York: Wiley. Caplan, G. (1976). The family as support system. In G. Caplan & M. Killiken (Eds.), Support systems and mutual help: MultidiscipUnary explorations. New York: Grune & Stratton. Caspi, A., Elder, G. H., Jr., & Bern, D. J. (1987). Moving against the world: Life course patterns of explosive children. Developmental Psychology, 22, 303-8. Caspi, A., Elder, G. H., Jr., & Bern, D. J. (1988). Moving away from the world: Life course patterns of shy children. Developmental Psychology. Caspi, A., & Elder, G. H., Jr. (1988). Childhood precursors of the life course: Early personality and life disorganization. In E. M. Hetherington, R. M. Lerner, & M. Perlmutter (Eds.), Child development in life-span perspective. New York: Cambridge University Press. Caspi, A., & Elder, G. H., Jr. (1988). Emergent family patterns. The intergenerational construction of problem behavior and problem relationships. In R. A. Hinde & J. Stevenson-Hinde (Eds.), Relationships within families. New York: Oxford University Press. Cavan, R., & Ranck, K. (1939). The family and the Depression. Chicago, IL: University of Chicago Press. Durkheim, E. (1951). Suicide. New York: Free Press. Eichorn, D. H. (1981). Samples and procedures. In D. H. Eichorn, J. A. Clausen, N. Haan, M. P. Honzik, and P. A. Mussen (Eds.), Present and past in middle life. New York: Academic Press. Elder, G. H., Jr. (1973). On linking social structure and personality. American Behavioral Scientist, 16, 785-800. Elder, G. H., Jr. (1974). Children of the Great Depression. Chicago, IL: University of Chicago Press. Elder, G. H., Jr. (1975). Age differentiation and the life course. In Annual Review of Sociology (Vol. 1). Palo Alto, CA: Annual Reviews. Elder, G. H., Jr. (1978a). Approaches to social change and the family. In S. Boocock & J. Demos (Eds.), Turning points (special supplement to the American Journal of Sociology, 84). Chicago, IL: University of Chicago Press. Elder, G. H., Jr. (1978b). Family history and the life course. In T. Harevan (Ed.), Transitions. New York: Academic Press. Elder, G. H., Jr. (1979). Historical change in life patterns and personality. In P. B. Bakes & O. G. Brim, Jr. (Eds.), Life-span development and behavior (Vol. 2). New York: Academic Press. Elder, G. H., Jr. (1984). Families, kin, and the life course: A sociological perspective. In R. D. Parke (Ed.), Review of child development research: The family (Vol. 7). Chicago: University of Chicago Press. Elder, G. H., Jr. (Ed.). (1985). Life course dynamics: Trajectories and transitions, 1968-1980. Ithaca, NY: Cornell University Press. Elder, G. H., Jr. (1986). Military timing and turning points in men's lives. Developmental Psychology, 22, 233-45. Elder, G. H., Jr., & Bailey, S. (1986). The timing of military service in men's lives. In J. Aldous & B. Klein (Eds.), Families and lives. New York: Guilford.

112

Glen H. Elder, Jr., and Avshalom Caspi

Elder, G. H., Jr., Caspi, A., & Downey, G. (1986). Problem behavior and family relations: Life course and intergenerational themes. In A. Sorensen, F. Weinert, & L. Sherrod (Eds.), Human development and the life course: Multidisciplinary perspectives. Hillsdale, NJ: Erlbaum. Elder, G. H., Jr., Caspi, A., & Van Nguyen, T. (1986). Resourceful and vulnerable children: Family influences in hard times. In R. K. Silbereisen, K. Eysferth, & G. Rodinger (Eds.), Development as action in context: Problem behavior and normal youth development. New York: Springer. Elder, G. H., Jr., Liker, J. K., & Cross, C. (1984). Parent-child behavior in the Great Depression: Life course and intergenerational influences. In P. B. Baltes & O. G. Brim, Jr. (Eds.), Lifespan development and behavior (Vol. 6). New York: Academic Press. Elder, G. H., Jr., Liker, J. K., & Jaworski, B. (1984). Hardship in lives: Historical influences from the 1930s to old age in postwar America. In K. McCluskey & H. Reese (Eds.), Life-span developmental psychology: History and cohort effects. New York: Academic Press. Elder, G. H., Jr., & Rockwell, R. (1979). Economic depression and postwar opportunity in men's lives: A study of life patterns and health. In R. G. Simmons (Ed.), Research in community and mental health. Greenwich, CT: JAI Press. Elder, G. H., Jr., Van Nguyen, T., & Caspi, A. (1985). Linking family hardship to children's lives. Child Development, 92, 310-30. Emery, R. E. (1982). Interparental conflict and the children of discord and divorce. Psychological Bulletin, 92, 310-30. Featherman, D., & Sorensen, A. (1984). Social transformation in Norway and change in the life course transition into adulthood. Acta Sociologica, 26, 105-26. Feldman, K. A., & Newcomb, T. M. (1969). The impact of college on students. San Francisco, CA: Jossey-Bass. Garmezy, N. (1981). Children under stress: Perspectives on antecedents and correlates of vulnerability and resistance to psychopathology. In A. I. Rabin, J. Aronoff, A. M. Barclay, & R. A. Zucker (Eds.), Further explorations in personality. New York: Wiley. Glueck, S., & Glueck, E. (1950). Unraveling juvenile delinquency. New York: The Commonwealth Fund. Goode, W. J. (1971). Force and violence in the family. Journal of Marriage and the Family, 33, 624-36. Gottman, J. M. (1979). Marital interaction. New York: Academic Press. Hinde, R. A. (1979). Toward understanding relationships. Ontario/London: Academic Press. James, W. (1890). The principles of psychology. New York: Macmillan. Kagan, J. (1979). Family experience and the child's development. American Psychologist, 34, 88691. Kohn, M., & Schooler, C. (1983). Work and personality: An inquiry into the impact of social stratification. Norwood, NJ: Ablex. Liker, J. K., & Elder, G. H., Jr. (1983). Economic hardship and marital relations in the 1930s. American Sociological Review, 48, 343-59. Lippmann, W. (1914). Drift and mastery. New York: Mitchell Kennedy. Lynd, R. S., & Lynd, H. M. (1937). Middletown in transition: A study on cultural conflicts. New York: Harcourt & Brace. Macfarlane, J. W. (1938). Studies in child guidance. I. Methodology of data collection and organization. Monographs of the Society for Research in Child Development, 3, (Serial No. 6). Macfarlane, J. W., Allen, L., & Honzik, M. P. (1954). A developmental study of the behavioral problems of children between twenty-one months and fourteen years. Berkeley, CA: University of California Press. McCord, J. (1979). Some child-rearing antecedents of criminal behavior in adult men. Journal of Personality and Social Psychology, 37, 1477-86.

Human development and social change

113

McCubbin, H. I., Joy, C. B., Cauble, A. E., Patterson, J. M., & Needle, R. H. (1980). Family stress and coping: A decade review. Journal of Marriage and the Family, 42, 855-971. Merton, R. K. (1968). Social theory and social structure. New York: Oxford University Press. Mills, C. W. (1959). The sociological imagination. New York: Oxford University Press. Moen, P., Kain, E. L., & Elder, G. H., Jr. (1983). Economic conditions and family life: Contemporary and historical perspectives. In R. Nelson & F. Skidmore (Eds.), American families and the economy. Washington, DC: National Academy Press. Parke, R. D., & Deur, J. L. (1972). Schedules of punishment and inhibition of aggression in children. Developmental Psychology, 7, 266-9. Patterson, G. R. (1982). Coercive family process. Eugene, OR: Castallia. Pearlin, L. I., Lieberman, M. A., Menaghan, E. G., & Mullen, J. T. (1981). The stress process. Journal of Health and Social Behavior, 22, 337-56. Quinton, D., Rutter, M., & Liddle, C. (1984). Institutional rearing, parenting difficulties, and marital support. Psychological Medicine, 14, 107-24. Riley, M. W., Johnson, M., & Foner, A. (1972). Aging and society. New York: Russell Sage Foundation. Robins, L. N. (1966). Deviant children grown up. Baltimore, MD: Williams & Wilkins. Robins, L. N. (1978). Sturdy childhood predictors of adult antisocial behavior: Replications from longitudinal studies. Psychological Medicine, 8, 611-22. Robins, L. N. (1984). Longitudinal methods in the study of development. In S. A. Mednick, M. Harway, & K. M. Finello (Eds.), Handbook of longitudinal research: Birth and childhood cohorts (Vol. 1). New York: Praeger. Rutter, M. (1979). Protective factors in children's responses to stress and disadvantage. In M. W. Kent & J. E. Rolf (Eds.), Primary prevention of psychopathology: Social competence in children (Vol. 3). Hanover, NH: University Press of New England. Rutter, M. (1980). Changing youth in a changing society. Cambridge, MA: Harvard University Press. Rutter, M., & Garmezy, N. (1983). Developmental psychopathology. In P. Mussen (Ed.), Handbook of child psychology: Personality and social development (Vol. 4): New York: Wiley. Rutter, M., & Madge, N. (1976). Cycles of disadvantage: A review of research. London: Heinemann. Rutter, M., & Quinton, D. (1984). Long-term follow-up of women institutionalized in childhood: Factors promoting good functioning in adult life. British Journal of Developmental Psychology, 18, 225-34. Ryder, N. (1965). The cohort in the study of social change. American Sociological Review, 30, 84361. Schaffer, R. (1977). Mothering. Cambridge, MA: Harvard University Press. Simmons, R. G., Blythe, D. A., Van Cleave, E. F., & Bush, D. M. (1979). Entry into early adolescence: The impact of school structure, puberty, and early dating on self-esteem. American Sociological Review, 44, 948-67. Thomas, W. I. (1910). Sourcebook for social origins. Boston: Badger. Thomas, W. I., & Znaniecki, F. (1918-20). The Polish peasant in Europe and America (Vols. 1 & 2). Chicago, IL: University of Chicago Press. Volkart, E. H. (1951). Social behavior and personality: Contributions ofW. I. Thomas to theory and social research. New York: Social Science Research Council. Werner, E. E., & Smith, R. S. (1982). Vulnerable but invincible: A study of resilient children. New York: McGraw Hill. Zigler, E., & Child, I. L. (1969). Socialization. In G. Lindzey & E. Aronson (Eds.), The handbook of social psychology (Vol. 3). Reading, MA: Addison-Wesley.

Family process: loops, levels, and linkages

Gerald R. Patterson

The relation of family to the culture contains many contradictions. On the one hand, the family's function is to make the next generation ready for socialization by agents outside the family. On the other hand, it is the function of a small number of families to prepare individuals who will function as agents of change. The family must somehow retain its sense of identity while one half of its members prepare to enter and the other half prepare to leave society. This tiny, transitory social system floats in a societal sea that is constantly making contradictory demands for changes and, simultaneously, for constancy and equilibrium. Clinicians and developmentalists alike would agree on the necessity for the study of change. The official imprimatur for such studies has been issued and reissued for decades. It is not a lack of interest or commitment that prevents our working in this area; it is simply that we do not know how to proceed. This chapter outlines some metaphors that seem appropriate to the study of process. Data will be presented at several points to illustrate the utility of thinking about it in this way. As employed here, the term process refers to behavior changing over time along a discernible trajectory. A plot of the changes over time in children's height and weight would be one example of a process. Changes over time in musical skill, or in the skills that maintain acquaintances, friends, or building intimate relationships would be other examples. As illustrated in Figure 5.1, each individual is moving along in multiple processes at any given point in time. We assume that children find themselves in many different processes, but make continuous movement in only a restricted number, and by middle age very few individuals make a commitment to initiate a new process. Commitments to initiating activities such as music, sports, getting an education, or starting a profession would all be examples of socially acceptable processes. Many young people are initiated to such processes, but for most, they do not move very far. For example, graduation from high school completes the education process for many, and few graduates read a book after that time. 114

Family process: loops, levels, and linkages

115

Figure 5.1: That's life

Because of a long-term commitment to clinical interests, I have focused my investigations on what might be called pathologic processes in families. This preoccupation is reflected in the present report by its emphasis on the coercion process; the details of these preoccupations are summarized in Patterson (1982) and Patterson, Reid, and Dishion (in press). Levels Each process implies the presence of some underlying mechanism(s) that determine the continuous changes in behaviors over time. In the maturation and aging process, a number of biological and environmental factors come to mind. However, if we consider processes such as learning a musical skill, getting an education, becoming neurotic, or developing conduct disorders, then the mechanisms are not so self-evident.

116

Gerald R. Patterson

Social processes such as those represented in the development of conduct or character disorders are complex affairs, in which changes take place simultaneously at several levels. There are at least two primary levels that must be examined in order to understand how the process is initiated and why the changes continue over time. Each process must differ in terms of what the relevant levels, or mechanisms, might be, but for the coercion process, it is hypothesized that the initial changes in behavior are molecular (or microsocial) in nature. This would mean that the process starts at the level of dyadic exchanges. The second, accompanying hypothesis is that many of these initial changes occur when one or both members are subjected to high levels of stress. According to the model (Patterson, 1982, 1983), increases in the daily level of stress are associated with commensurate increases in the irritability for social interaction exchanges with other family members. Persons with dispositions to antisocial and depressive behaviors are presumably more vulnerable to these increases. In a sense, dyadic exchanges become a kind of transformer whose function is to transform forces impinging from outside the family into disrupted patterns of social exchange. The third hypothesis is that for the process to move to a second stage, it is necessary for the social patterns altered by the dyadic exchange to generalize to other social settings. For example, the coercive patterns acquired by the child in his or her family interactions generalize to include interactions with his or her peer group. Data from the first cohort of a longitudinal study on juvenile delinquency being conducted at the Oregon Social Learning Center in Eugene were used to test the hypothesis that the effects of dyadic training in the home generalize to the peer group. The findings were consistent with the hypothesis. The correlation between the measure of peer (report) aggression and the construct measuring observed child coercion (start up 4- continuance + synchronicity) in the home was .31 (p < .001). The fourth hypothesis identifies the mechanism that keeps this stage of the process moving. It requires that the newly generalized behaviors elicit some predictable pattern of reactions from a significant constituency in the social environment. For example, the reaction of peers to musical or athletic skills taught at home might have a great deal to do with the child's continuing in these processes. The noncompliance and coerciveness of the child with a conduct disorder produce the same short-term reinforcement at school as at home. In effect, these immediate reactions sustain and perhaps extend the home-based training program. However, the members of the peer group also tend to reject the coercive child in the long run. To test this hypothesis, the child coercive behaviors observed in the home were correlated with the data from the peers describing their relation with the child. The findings were consistent with the hypothesis, with a correlation of — .35 (p < .001). It is the long-run reaction that makes the

Family process: loops, levels, and linkages

117

significant contribution to the maintenance of the problem child in the conduct disorder process. It may also contribute to speeding the process up, and the child becomes increasingly deviant relative to his or her peer group. It seems that data focused on two very different levels are required to understand the coercive process. Data at the molecular level would require observation data collected in sequential form in natural settings. The analyses would identify action-reaction patterns that occur repeatedly enough to be identifiable. The data describing stressors and the reactions of the peer group would be based in large part on more macrosocial measures such as self-reported stress, peer nomination data, and teacher reports. Short-term and long-term effects As a general case, we believe that two kinds of mechanisms are required for a process to go into action. At the molecular level, there needs to be immediate short-term effects that significantly alter the behavior of one or both members. As noted above, a second mechanism could be the reinforcing reactions to coercive behavior at school that would keep the child in the process. It does seem that there are some second mechanisms for the coercion process that actually make the individual more vulnerable to reinforcement provided at home and by the peer group. As noted above, not one but two reactions of the social environment relate to the process. In the new setting in which the short-term coercive and avoidant behaviors pay off, this maintains the process. But the fact that the peer group also rejects the child serves a long-term function. It is hypothesized that long-term reactions such as this function as positive feedback loops. The function of these mechanisms is to maintain and to speed up the process. Over a period of months and years, the effect of peer rejection or isolation would be to create a situation of limited opportunity. According to Youniss (1980), the function of the peer group is to teach the child cooperation with equals, reciprocity in social exchange, and empathy. The adolescent who has failed to acquire these skills will be perceived as increasingly childlike by peers and employers alike. In a culture that demands increasing social skills (work and relational) of its young adults, this individual will become increasingly salient as unskilled. He or she will find him- or herself using coercive and avoidant behaviors at increasing rates. Relatively speaking, he or she is identifiable as being increasingly deviant. In this sense, one might say that the process speeded up. This is also the sense in which one might think of the long-term effects of peer rejection as functioning as a positive feedback loop. The remainder of this report is divided into a discussion of the microsocial and the macrosocial levels.

118

Gerald R. Patterson

Microsocial mechanisms for change It is the social interactional perspective (Cairns, 1979; Gottman, 1981; Lamb, Suomi, & Stephenson, 1979; Patterson & Reid, 1984) that implicates dyadic exchanges as an important mechanism for change in families. In the dyadic exchange, there are literally hundreds of "trials" each day. If one observed the dyad interacting in the home on each of 16 days, it would be possible to calculate the mean daily values for the likelihood of person A reacting in a coercive fashion to person B. Figure 5.2 illustrates what such values might look like. Previous analyses have shown that individual differences are quite stable over a 1-year period. For example, in one study the stability correlation was .74 for a 12-month test/retest interval. The first estimate was based on three sessions and the second on two sessions. We believe that some of the unaccounted-for variance in such estimates is really noise or errors of measurement, but also that there is a significant component that reflects genuine shifts in interactional patterns over time. In our experience, daily plots of coercive exchanges show a good deal of fluctuation. In fact, we believe this ebb and flow of likelihoods is the daily bill of fare for most dyads. The assumption, again, would be that not all of the variance in daily fluctuations is due to errors of measurement, but that some of it is in fact determined. Most shifts in likelihood could not be detected by the participants, but only by trained observers and careful statistical analyses. Figure 5.2 shows that for the first 8 days there was no change in the slope. During that time, all the likelihoods fluctuated about a shared grand mean. What is important for our purposes is the slope described after Day 8. Notice that now the values are shown to fluctuate about a nonzero slope. Given more data points, it would be possible to use a time-series analysis (Gottman, 1981) to demonstrate that this is a significant shift in dyadic exchange. What is interesting about this is that, given observational data collected in sequential form, it is possible to specify precisely the initiation of change in dyadic process. The assumption is that for the target child in a longitudinal study, many of the basic changes leading to prosocial and deviant outcomes begin in the dyadic exchanges with the mother and the siblings. What begins in a simple shift in probability values at the level of dyadic exchange may generalize to a surprisingly wide spectrum of effects, what members of the dyad feel about each other, and the kinds of attributions they make about each other. If one has an empirical anchor point on the question of when the dyad process shifts, one can focus on the question of which variables determine that shift. This is a kind of bootstrapping operation, beginning with precise descriptions of action-reaction sequences occurring in the real world, then moving to a precise statement about when minute changes have occurred.

B.

o

Probability A coercive to B

120

Gerald R. Patterson

For convenience, the examination of recurring dyadic exchange patterns and their accompanying probability values are labeled microsocial analysis. This is also essentially what most writers, such as Cairns (1979) and Lamb et al. (1979), mean by a social interactional approach. It describes the likelihood of a certain reaction by person B given a certain action by person A; it may also include more complex three-step patterns (Patterson, 1982). Rather than a finite pattern such as A-B-C, some formats such as the sequential lag analysis described by Sackett (1978) examine the likelihood of event A occurring at each interval or event from tx through tl6. Given the microsocial anchor point illustrated in Figure 5.2, one is in a good position to examine several different questions. For example, how do these changes at the microsocial level relate to changes in more macrosocial (e.g., trait-like) behaviors? It would also be possible to use the reference point to search for the variables that produce the dysjunction in process. Given the contemporary developments in code systems that accurately describe interacting dyads in sequential form (e.g., Dishion et al., 1983), then it seems reasonable to suppose that we can specify when a process is changing. Given information about baseline levels, we could also specify the magnitude of change over time. However, there is another way of thinking about how much things have changed over time. It is possible that most processes move through recognizable stages, or sequences, in a progression. It is our strong impression that children in the coercion process move through definable stages, not only in the development of symptoms (Patterson & Dawes, 1975) but also in more complex ways as well. The assumption is that there are definable sequences, transitive in form, that characterize both the microsocial and macrosocial levels. This possibility will be examined in a following section. Before that can be explored, however, it is necessary to understand how the coercion process produces significant changes in dyadic exchanges. This material is examined in the next section. How do changes in microsocial exchanges come about? As suggested by Skinner (1969) and others, people can and probably do use positive reinforcers to alter each other's behavior all the time. However, if one looks only at immediate outcomes, the contingent use of aversive stimuli may be a more efficient method for changing behavior. But in the long run, these aversive contingencies have some negative outcomes that are extremely disruptive to family process. It is a sad fact that many adults in our society choose to maximize short-term gains by employing aversive contingencies at very high rates, to the detriment of those living with them. In the long run, the contingencies work against the person who employs them, but the fact that the payoffs are

Family process: loops, levels, and linkages

121

immediate, while the negative outcomes are lagged, creates a reinforcement trap for the unskilled and unwary. The details of the coercion mechanism emerged slowly, beginning with our first sequential analysis of family interaction data (Patterson & Cobb, 1971). The data showed that aversive consequences suppressed the child's ongoing prosocial behaviors, but increased the likelihood that the child would repeat prior coercive behavior! This was a surprising finding. We decided to begin a systematic search for the antecedents and the consequences that controlled children's ongoing coercive behaviors (Patterson, 1977; Patterson & Cobb, 1973). After a decade of effort, it finally became clear that in addition to rewards and punishments, there was a very important third contingent arrangement that also altered the performance of social behaviors. This arrangement was also the key to understanding coercion in families. In the operant literature, this third arrangement is called negative reinforcement (Hineline, 1977). Generally, it is a variant of escape-avoidant conditioning defined by the sequence: aversive stimulus - response - termination of aversive stimulus. Because many investigators are confused by the term negative reinforcement (it is not punishment), we use the rather awkward phrase escapeavoidant conditioning as the appropriate label. The experimental findings and field study data related to this mechanism are spelled out in detail in Patterson (1982, ch. 7). Escape-avoidant conditioning and punishment contingencies are at the heart of the coercion mechanism, but their long-range effects tend to slip by unnoticed, lost in the flood of ongoing social interaction. In keeping with the speculations by Langer (1984) and the experimental studies reviewed by Nisbett and Wilson (1977), we believe that most of the minutiae of changes in probability during social exchanges occur outside the awareness of the individual. Each of us has better things to do with our limited channel capacity than to try to focus on whether this reinforcer or punisher occurred and what the effect was on our behavior. As shown in the experimental tests of the reinforcement change without awareness hypothesis, it is a relatively simple affair to arrange for an individual to be attending to one set of stimuli while his or her ongoing behavior is altered by some other set of contingencies s/he is not tracking (Gewirtz & Boyd, 1977; Rosenfeld & Gunnel, 1985). One of the characteristics of dyadic interchanges that makes them so difficult to track and store lies in the multiplicity of effects, introduced by a single aversive event. One of the most fascinating features of social interaction is the possibility of simultaneous short- and long-term outcomes for a single event. After observing it in family after family being treated because of an antisocial child, we came to label the simultaneous outcomes as the reinforcement trap. From the situation illustrated in Figure 5.3, one may obtain five different

122

Gerald R. Patterson Mother

Mother

Mother

less

scolds

talks

t0

Child whines

Da

V

like| sc

If child whines, y

mother more likely

ld

Child stops whining

1

t0

Jalk

Child more likely to whine Day 2

Figure 5.3: The reinforcement trap (from Patterson, 1982)

outcomes from a single aversive exchange. The situation there tells a familiar story. There have been many previous trials in which the mother nagged the child about a messy room. The scene opens with mother scolding. As previously rehearsed, the child begins to whine. For the sake of illustration, assume that this time the child's whining effectively terminates mother's scolding. She now talks to the child in a soothing manner. Thus, the child's whining produced mother's talking. In the example, the mother and child both behave in such a way as to maximize short-term gains. The child behaves in such a way as to maximize the likelihood that the mother will terminate her scolding. The mother behaves in such a way as to maximize the likelihood the child will immediately stop whining. This second outcome is in keeping with findings from the laboratory analogue study by Dengerink, Schnedler, and Covey (1978). In a Buss-type shock situation, when the other member terminated attacks, subjects quickly reduced their level of aggression. In a family setting, does the coercer terminate the attack within seconds of victim compliance? In a study of three mother-child pairs, Woo (1978) found that during baseline the mean duration of the child's whining was 28 seconds. Given an aversive stimulus followed by the child whining and mother complying, the child stopped whining in an average of 4.6 seconds! What makes this interesting is that both participants failed to see that for both of them there were long-term effects that would, in turn, produce increases in future misery. The mother did something that stopped the child's whining, but in so doing, she increased the likelihood of the child whining in the future. This is an operational definition of what is meant by the reinforcement trap. The child has trained the mother to stop trying to teach social skills; in the long run, this will cost him dearly. This lack of agreement between short-term and long-term effects makes it extremely difficult for the participants to understand what has happened. It is, in other words, an ideal situation for behavior change without awareness.

Family process: loops, levels, and linkages

123

In the short term, both members reduced the number of average events impinging on them. It should be noted, however, that the child's room remained a mess. The mother lost because she opted for an immediate reduction in her "pain" schedule, not knowing, of course, that it was at the expense of an increased likelihood that messy rooms will occur in the future. The findings from the laboratory and field studies summarized in Patterson (1982) are all consistent with these ideas. More recent studies by Snyder and Patterson (1986) showed that when mother submitted to coercive child behaviors, it increased the probability that the child would behave coercively in future interactions. The coercion mechanism has recently been used to investigate dyadic process in a wide range of pathologies. For example, Biglan et al. (1984) showed that mothers' depressed behaviors may serve the useful function of suppressing spouse aggressive behaviors. Gottman and Levenson (1986) have begun the analyses of the function of spouse affect during conflict, which suggest that what works in reducing conflict may become strengthened over time. Microsocial structure A network of ideas is involved in this next problem. The first part concerns the possibility that there are changes in the structure of social interaction accompanying reinforcing contingencies. This implies that there are not one, but two, outcomes of programmatic reinforcement: (1) increases in the likelihood of occurrence for the reinforced responses, and (2) regular structural changes that accompany these programs, a kind of bonus as it were. These structures are transitive in nature. In a perfectly transitive progression, all individuals classed in an advanced stage would have passed through each of the prior stages; however, not all individuals in a given stage move on to the next one. The second assumption about these structures is that the longer the child has been involved in the process, the further he or she has moved into the progression. The third assumption is that the further into the progression, or the more advanced the stage, the more likely the training is to generalize to macrosocial levels (including traits and settings). The microsocial process starts after there has been an increase in the rate of coercive behaviors for one or both members of the dyad. When the rate of coercive responses is well above the mean for a period of several weeks, then certain structural changes in the dyadic exchanges begin to emerge. We assume that the startup variable is the first to show an increase. Startup means that even though the other member of the dyad is in a positive or neutral state, the child is likely to initiate an aversive behavior. Figure 5.4 summarizes the data from a sample of the first 44 families in Cohort 1. It can be seen that 72% of the boys did this one or more times with the siblings during the three observation sessions.

124

Gerald R. Patterson Transitional probabilities

Alternate path probability values Figure 5.4: Microsocial progression: target child and sibling exchanges

Given that the child engaged in one or more startup behaviors, the likelihood that he would move on to the next stage in the structure was .59. The comparable conditional for boys' reactions to mothers was .64. This next stage was defined as synchronistic negative exchange. Given the sibling reacted to him in an aversive fashion, then the target subject reacted in kind. Dyads who display frequent negative synchronistic reactions have made considerable progress into the coercion process. For example, Levenson and Gottman (1983) found that couples with high negative synchronicity in a laboratory task were extremely likely to be dissatisfied with their marriage. As might be expected, boys who had high negative synchronistic scores were also described as highly antisocial by parents (Patterson, 1982), and in another study by teachers, peers, and parents (Patterson et al., in press). Given that the child engages in one or more synchronistic reactions, the likelihood that he will move on to the next stage was .79; the comparable value for his reactions to mothers was .64. This next stage, continuance, was defined as the likelihood that the child would follow one coercive response with another (regardless of what the other person did). The significance of this stage lies in the

Family process: loops, levels, and linkages

125

fact that longer-duration coercive exchanges set the stage for increases in the amplitude of responses. It is during the extended exchanges that one or the other member is likely to try a higher-amplitude response, and when s/he does, the odds are very good that the other will withdraw. In keeping with this formulation, the analyses by Reid, Taplin, and Lorber (1981) showed that hitting was more likely to occur during coercive chains longer than 18 s. This held for both normal and clinical samples, but as might be expected, the family members in the clinical samples were more likely to engage in the longer chains. Given that the child engages in one or more continuance reactions with siblings, then the likelihood of engaging in physical aggression with some family member (most likely siblings) was .55. The finding provides support for the earlier analyses of data from clinical samples by Reid et al. (1981). The last bit of information contained in Figure 5.4 describes the likelihood that a child arrived at a given state via some path other than the progression described there. Notice that the odds were generally less than one in five that if a child engaged in that behavior, he did not also engage in the prior behavior in the progression. The data presented in Figure 5.4 are consistent with the notion of a microsocial structure that accompanies family-based training in coercion. What might the implications of this be? It is hypothesized that there is a parallel progression for antisocial traits such that the further one progresses in the microsocial progression, the further one advances in the trait progression. Presumably, it is simply the amount of time in the microsocial process that determines how far into the trait progression one moves. This possibility will be examined in the following section. Traits define progressions Antisocial process is characterized by simultaneous movement along both microsocial structural progressions and a corresponding movement along an antisocial trait progression. Both are transitive in nature. Best of all for our purposes is that both are eminently measurable. We believe traits to be relatively stable across time and across settings. We have expanded the meaning of the term slightly to have it serve as a referrent for the menu of settings likely to be selected by the child. For example, it would be expected that the aggressive child would more likely choose unsupervised activities rather than supervised ones, and excitement-provoking settings rather than intellectual ones. In the present context, a trait should include in its definition the general reactions the (trait) behavior elicits from the social environments in which they are practiced. For example, the careful studies of peer interactions by Coie and

126

Gerald R. Patterson

his colleagues attest to the consistency of peer reactions to coercive behavior (Coie & Kupersmidt, 1983). This is also in keeping with our own studies noted earlier, where child coercion in the home correlated .34 with peer rejection at school. The other characteristic to be included in defining a trait would be a specification of its position in the transitive sequence. The trait progression reflects the fact that there is an underlying shared process determining all the traits in the sequence. One of the problems in using progression is that, as yet, there is no effective means for deciding when a progression exists, or even for deciding where one should look. The technique proposed by Guttman (1944) initiated a great deal of research activity, and that activity attests to the fact that he posed the right question. But as Robinson (1973) and many others have pointed out, the statistical basis he used for deciding on the existence or nonexistence of a progression proved to be an absolute disaster. As we shall see later in the discussion, if one can live with the idea that there is no single index for deciding about an entire progression, then it might be said that a simple application of conditional probability analyses is a serviceable way of describing each juncture in the progression, as well as the likelihood of alternative routes to the same stage. The data used to illustrate the progression in Figure 5.5 were taken from the first 50 families in Cohort 1 from the longitudinal study. Both the mother and the father filled out the Oregon Child Aggression questionnaire when the boys were 10 years old. If one or both parents indicated that noncompliance, temper tantrums, or fighting occurred at least once a week, it was counted as evidence that the child had progressed that far. The child was also included in that stage if one parent claimed once a week but the other once a month. The child was placed in the stealer stage if both parents agreed that he stole at least once per month or more. Five of the children had been involved in one or more court offenses at the time of the study or during the ensuing 12 months following the study. The base rates for the sample of fourth-grade boys are shown in the lower line of the box for each symptom. The parents describe surprisingly frequent occurrences of noncompliance and temper tantrums for boys of this age (a temper tantrum per week!). We were also surprised to see that 33% of the boys stole at least once a month. The transitional probability values describe the likelihood that, given the child was in one stage, he had also been described in the next stage in that progression. For example, two thirds of the boys described as noncompliant had also been described as having temper tantrums. Notice that the majority of children with temper tantrums did not move on to fighting. A substantial number of children who were fighters were also stealers, and about one third of the children who were stealers had court offenses.

Family process: loops, levels, and linkages

127

Transitional probability values

Alternate path probability values

Figure 5.5: Microsocial progression

The third bit of information required to understand progression is the probability value defining the likelihood of an alternative path. For example the p value for fighting describes the fact that 82% of the boys who were fighters were also described as having temper tantrums. If one believes these particular data, then 18% arrived at the fighting stage from some other path. This probability value is of particular interest when it describes the alternative path for stealing. The general model stipulates that the majority of the cases referred for treatment would follow the path described in Figure 5.5 (noncomply - fight - steal) (Patterson et al., in press). In that model, about one quarter of the clinical referrals are thought to have passed along a sneaky aggression path that does not include fighting, and where the children have an acceptable level of social skills. The present findings reaffirm the need to study that alternative path because the majority of the stealers in this sample do not seem to fit the coercion model. The alternative model described in Patterson (1982) was based on the idea of a distancing parent who could cope with discipline confrontations but was

128

Gerald R. Patterson

Table 5.1. Relating microsocial to trait progression Trait sequence Fight {.22)a

Microsocial progression p(X/ Startup) p{XI Startup + Synch) p{XI Startup + Cont) p(X/Startup + Synch + Cont + Phys Agg)

Steal (.33)

Mom

Sibling

Mom

Sibling

.22 .32 .44

.34 .33 .38

.38 .63 .47

.45 .50 .36

.60

.56

.40

.33

Court offense 11) Mom

Sibling

.40

.40

° Figures in parentheses indicate baserate.

not involved enough to effectively monitor the child. According to the earlier study, the trait progression defining this path seemed to be noncomply - lie steal - set fires. Integrating microsocial and trait progressions It was implied earlier that in some sense the microsocial and the trait progressions were tied together. Other things being equal, the longer the time spent in the coercion process, the further info the microstructure the child would progress. Presumably, the further the child progresses at this level, the greater the risk for movement through the trait progression. The data from the first 44 families in Cohort 1 were analyzed to form a preliminary basis for examining the hypothesis. Prior studies suggest that the key microsocial exchanges probably take place between the target child and both siblings and mother. For this reason, the two dyadic exchanges were analyzed separately. Each stage in the microsocial progression was analyzed separately to determine its relation to each stage in the trait progression. The findings are summarized in Table 5.1. The baserate value for each trait is given at the head of each column. For the analysis to be of interest, the conditional probability value should exceed the baserate by a good deal. Because this analysis is just illustrative, no effort was made to make a statistical comparison of the two values. It became clear very quickly in this example that the microsocial progression provided no basis for predicting transition to the noncomply and temper tantrum stages in the trait progression. The baserate values for these two behaviors were

Family process: loops, levels, and linkages

129

simply set too high; they also exceeded the conditional probability values from the microsocial progression. Evidently some artistry is required to systematically test the relation between two progressions. Note, however, the steady increment in prediction to the fight stage for both mother and sibling exchanges. Given that one knows the child used one or more startups does not provide a solid basis for predicting fighters, but the fact that the child has moved from startup to synchronicity to continuance seems predictive. If one knows the child uses physical aggression during the observation sessions, it provides a conditional that more than doubles the baserate values. We simply do not know what to make of the next column describing the situation for stealing. Knowing the child used startup plus synchronistic reactions seems to be a useful basis for predicting stealing; moving further into the coercion process seems to function as a suppressor! It seems to be an iteration of the earlier theme that there is a second path leading to stealing, and perhaps to court offenses as well. Given the small sample size, no effort was made to even calculate the various conditionals for the court offense column. The last one in the column, however, was simply too suggestive to omit. It seems plausible to assume that there might well be an empirical linkage between the two levels. The pilot data are consistent with the idea that the variable that connects the two is really time in process. If the dyad has reached the high-amplitude stage in the microsocial progression, the child would be more likely to have moved far into the trait progression, for example, to fighting. Stages The idea of process can be approached in several ways. The least interesting would be to think of a single behavior increasing in rate over time. Most clinical processes consist of much more than an increase in rate of deviant behavior over time. The hundreds of hours of clinical histories presented by the families treated at Oregon Social Learning Center suggest a sequence of stages. The idea is that arrival at each new stage places the child at even greater risk to continue on toward a career of antisocial behavior. The sequence of stages that characterizes the clinical histories is summarized in Figure 5.6. As shown there, the first stage consists of the microsocial training of a target child in the home by family members. They teach him to use aversive behaviors, to avoid responsibility, and above all else to be noncompliant. These behaviors generalize to the school setting. The reactions of the persons in that setting define Stage 2. The coercive interpersonal style quickly leads to rejection by the normal peer group. In this sense, the trait of coercion might include as part of its definition the social environmental reaction, "peer rejec-

Gerald R. Patterson

130

Stage 5

Stage 4

Stage 3

Stage 2

Stage 1 Figure 5.6: Stages model (from Patterson, Reid, & Dishion, in press)

tion." As noted in Figure 5.6, the reaction of the social environment should also specify that the parents tend to reject the child, and for the same reasons that peers do. The correlation between parental and peer rejection was .43 (p < .001). The obdurate noncompliance includes noncompliance to classroom

Family process: loops, levels, and linkages

131

setting rules, homework assignments, and responses to teacher instructions and requests. This, in turn, places the child at grave risk of academic failure. As shown in Figure 5.6, the effects of rejection by parents is thought to lead to the development of poor self-esteem. Rejection by peers and academic failure are thought to lead to the development of depressed mood. These are outcomes or effects produced by what had previously happened at Stage 2. The assumption is that these outcomes, as well as those at Stage 2, would function as positive feedback loops that serve the function of keeping these children in the process. The boys have been in the process for some time. They are both antisocial and socially incompetent. They have failed at almost everything important that they have attempted. They feel they are not good people, and find themselves increasingly caught up in dysphoric mood swings. Should they continue in the process, they will become at risk for membership in the cadre of likefellows that comprise the deviant peer group. This support group will facilitate their entry into substance abuse and delinquency. The sequence model is designed to describe what the data should look like when generated in a well-designed longitudinal study. The data have been collected for the first two waves for two cohorts of 100 families each. The families are living in the high-crime areas of Eugene, Oregon; they began their third wave of data in the fall of 1985. In the following sections, the relations among variables that should characterize each of the first three stages will be examined by using data from the first wave. Each of the stages in Figure 5.6 defines which constructs should covary at any point in time. If the correlational structure does not fit, then the models are wrong and would not be suitable for the longitudinal analyses that will follow. The findings relevant to each of the first two stages are briefly examined as illustrations. Stage one: micro social and macro social together It would seem useful at this point to attempt to put the microsocial and the macrosocial levels together. The structural model presented here examines the relation between two microsocial variables defining the Parental Inept Discipline and Child Coercion constructs on the one hand and an across-setting measure of Antisocial Behavior on the other. It is assumed that outside stressors or a lack of prior training in family management skills can lead to a situation in which the microsocial exchange between parent and child becomes badly disrupted. The relation between outside stressors (employment, illness, daily hassles) and parenting skills has been explored in a series of modeling studies reported in Patterson (1986a), Forgatch, Patterson, and Skinner (in press), and Viken (1986). We are also interested in

132

Gerald R. Patterson

exploring the Elder hypothesis that the effects of stress on child discipline confrontations may be mediated by the father's general disposition to be irritable (Elder, Van Nguyen, & Caspi, 1985). The effort to replicate the Elder mediational model for stress was pursued in a study of single-parent families (Forgatch et al., in press). The analyses of these data showed clear support for the Elder hypothesis. For single mothers, the effect of stress on discipline practices was mediated by their general disposition to be irritable. The Stage 1 model, basic training for antisocial behavior, requires a demonstration that antisocial behavior occurring in the home and the school is significantly related to coercive exchanges between parent and child. In Figure 5.7, the parent contribution to the microsocial exchange is to be found in two of the four indicators that define the Inept Discipline construct (likelihood of nattering, likelihood of explosive discipline). The child's contribution is defined by the Child Coercion construct (likelihood of startup and duration of coercive chains). Note that the relation between the parent and child constructs is assumed to be bidirectional (e.g., as the child becomes more coercive, he is even more difficult to discipline). There are major problems associated with the estimation of bidirectional effects. In that we are still struggling with this problem, the precise values for these paths remain in doubt. In the model in Figure 5.7, multiple indicators were used to define each of the three latent constructs. The data from Cohort 1 served as the derivation sample, and Cohort 2 as the replication sample. The findings are summarized in Figure 5.7. The numbers on the diagonals connecting the constructs to the indicators refer to the factor loadings. Notice the substantial stability in loadings across cohorts for 10 of the 11 indicators (mother reported discipline will be dropped). According to the clinical model, inept discipline sets into motion the microsocial process whereby siblings and mothers, and to a lesser extent fathers, train the antisocial child in the basic antisocial skills. This, in turn, makes it more difficult to discipline him or her. All the path coefficients for the bidirectional effect were found to be significant for both cohorts. The findings are therefore consistent with the clinical model. In the theory-driven model, the malleable determinant for antisocial behavior is thought to be parental discipline. In the treatment procedures, training parents to use nonphysical forms of punishment effectively reduces the level of antisocial behavior. The model requires a strong path coefficient connecting the Antisocial Behavior and Inept Discipline constructs. As shown in Figure 5.7, the path with the largest coefficient in the entire model is the one connecting these two constructs. Two chi-square tests showed a satisfactory fit of the data to the overall model. For both cohorts, the model accounted for about 40% of the variance in the criterion measure of antisocial behavior. This more than meets our requirements

EXPLOSIVE DISCIPLINE

OBSERVED STARTUP TO PARENTS

OBSERVED PARENT NATTERING

OBSERVED STARTUP TO SIBLINGS OBSERVED DURATION OF CHILD COERCION

OBSERVER IMPRESSIONS MOTHER REPORT DISCIPLINE STRATEGIES

PARENT REPORT

ANTISOCIAL BEHAVIOR *p environment effect is called evocative because it represents the different responses that different genotypes evoke from the social and physical environments. Responses to the person further shape development in ways that correlate with the genotype. The third kind of genotype —> environment effect is the active niche-picking or niche-building sort. People seek out environments they find compatible and stimulating. We all select from the surrounding environment some aspects to which to respond, learn about, or ignore. Our selections are correlated with motivational, personality, and intellectual aspects of our genotypes. The active genotype —» environment effect, we argue, is the most powerful connection between people and their environments and the most direct expression of the genotype in experience. Developmental changes in genotype —> environment effects The second proposition is that the relative importance of the three kinds of genotype —» environment effects changes over development from infancy to adolescence. In infancy much of the environment that reaches the child is provided by adults. When those adults are genetically related to the child, the environment they provide in general is positively related to their own characteristics and their own genotypes. Although infants are active in structuring their experiences by selectively attending to what is offered, they cannot do as much seeking out and niche building as older children; thus, passive genotype —> environment effects are more important for infants and young children than they are for older children, who can extend their experiences beyond the family's

230

Sandra Scarr

influences and create their own environments to a much greater extent. Thus, the effects of passive genotype —» environment effects wane when the child has many extrafamilial opportunities. In addition, parents can provide environments that are negatively related to the child's genotype to compensate for the child's perceived deficiencies or problems. Parents with a child who is aggressive may attempt harsher discipline to curb the child's unacceptable behavior. Parents with a child who does not learn to read easily may provide compensatory practice in reading. Although parents' genotypes usually affect the environment that they provide for their biological offspring, it is sometimes positive and sometimes negative and, therefore, not as direct a product of the young child's genotype as later environments will be. Thus, as stated in proposition 3, genotype —•» environment effects increase with development, as active forms replace passive. Genotype —> environment effects of the evocative sort persist throughout life, as we elicit responses from others on the basis of many personal, genotyperelated characteristics from appearance to personality and intellect. Those responses from others reinforce and extend the directions our development has taken. High intelligence and adaptive skills in children from very disadvantaged backgrounds, for example, evoke approval and support from school personnel who might otherwise despair of the child's chances in life (Garmezy, Masten, & Tellegan, 1984). In adulthood, personality and intellectual differences evoke different responses in others. Similarities in personal characteristics evoke similar responses from others, as shown in the case of identical twins reared apart (Bouchard, 1984). These findings are also consistent with the third proposition. Accounting for observations from family research In an earlier report (Scarr & McCartney, 1983), we reviewed evidence about how monozygotic twins come to be more similar than dizygotic twins, and biological siblings more similar than adopted siblings on nearly all measurable characteristics, at least by the end of adolescence (Scarr & Weinberg, 1978). We also reviewed the evidence on declining similarities between DZ twins and adopted siblings from infancy to adolescence. Most fascinating to laypeople are the unexpectedly great similarities between identical twins reared in different homes. Neither extreme genetic determinism nor naive environmentalism can account for these observations from research on twins and families. Our theory of genotype —> environment effects can account for these data by predicting the degree of environmental similarity that is experienced by the co-twins and sibs, whether they live in the same family or not. The expected degree of environmental similarity for a pair of relatives can be thought of as the product of a person's own genotype —> environment path and

How genotypes and environments combine

231

path b

path x

path x

path a Figure 9.6: A model of environmental similarity based on genetic resemblance

the genetic correlation of the pair. Figure 9.6 presents a model of the relationship between genotypes and environments for pairs of relatives who vary in genetic relatedness. Gj and G 2 symbolize the two genotypes, Ex and E 2 their respective environments. The similarity in the two environments (path a) is the product of the coefficient of each genotype with its own environment (path JC) and the genetic correlation of the pair (path b). On the assumption that individuals' phenotypes are equally influenced by their own genotypes, the similarity in the environments of two individuals becomes a function of their genetic correlation. This model can be used to describe the process by which MZ twins come to be more similar than DZ twins, and biological siblings more similar than adopted siblings. For identical twins, for whom b = 1.00, the relationship of one twin's environment with the other's genotype is the same as the correlation of the twin's environment with her own genotype. Thus, one would certainly predict what is often observed: that the hobbies, food preferences, choices of friends, academic achievements, and so forth of the MZ twins are very similar (Scan* & CarterSaltzman, 1980). Kamin (1974) proposed that all of this environmental similarity is imposed on MZ co-twins, because they look so much alike. We proposed that the home environments provided by the parents, the responses that the co-twins evoke from others, and the active choices they make in their environments lead to striking similarities through genotypically determined correlations in their learning histories. The same explanation applies, of course, to the greater resemblance of biological than adopted siblings. The environment of one biological sib is correlated to the genotype of the other as one half the coefficient of the sibling's environment to her own genotype, because b = 0.50, as described in Figure 9.2. The same is true for DZ twins. There is a very small genetic correlation between adopted siblings in most studies that arise from selective placement of the offspring of similar mothers in the same adoptive home. More important for this theory, however, is the selective placement of adopted children to match the intellectual characteristics of the adoptive parents. This practice allows adoptive parents to create a positive, passive genotype-environment correlation for their adopted

232

Sandra Scarr

children in early childhood, when the theory asserts that this kind of correlation is most important. In fact, the selective placement estimates from studies by Scarr & Weinberg (1977) can account for most of the resemblance between adoptive parents and their children. In addition, adoptive parents, like their biological counterparts, can provide negative genotype-environment correlations that ensure that their several children will not differ too much on important skills, such as reading. The most interesting observation is the unexpected degree of resemblance between identical twins reared in different families, and most often in different communities and even different cultures. With the theory of genotype —» environment effects, their resemblance is not surprising. Given opportunities to attend selectively to and choose from varied opportunities, identical genotypes are expected to make similar choices. They are also expected to evoke similar responses from others and from their physical environments. The fact that they were reared in different homes and different communities is not important; differences in their development could arise only if the experiential opportunities of one or both were very restricted, so that similar choices could not have been made. According to previous studies (Newman, Freeman, & Holzinger, 1937; Juel-Nielsen, 1980; Shields, 1962) and the recent research of Bouchard and colleagues at the University of Minnesota (Bouchard, 1984), the most dissimilar pairs of MZs reared apart are those in which one was severely restricted in environmental opportunity. Extreme deprivation or unusual enrichment can diminish the influence of genotype on environment and therefore lessen the resemblance of identical twins reared apart. Changing similarities among siblings The second discrepant set of observations from previous research concerned the declining similarities of dizygotic twins and adopted siblings from infancy to adolescence. It is clear from Wilson's (Matheny, Wilson, Dolan, & Krantz, 1981; Wilson, 1983) longitudinal study of MZ and DZ twins that the DZ correlations for intelligence of .60-.75 are higher than genetic theory would predict in infancy and early childhood. For school age and older twins, DZ correlations were the usual .55. Similarly, the intelligence correlations of two samples of late adolescent adopted siblings were zero compared to the .25-.39 correlations of the samples of adopted children in early to middle childhood (Scarr & Weinberg, 1978). How can it be that the longer you live with someone, the less like them you become? The theory put forward here predicts that the relative importance of passive versus active genotype-environment correlations changes with age. Recall that passive genotype-environment correlations are created by parents who

How genotypes and environments combine

233

provide children with both genes and environments, which are then correlated. Certainly in the case of DZ twins, whose prenatal environment was shared and whose earliest years are spent being treated in most of the same ways at the same time by the same parents, the passive genotype —> environment effect is greater than that for ordinary sibs. Biological and adopted siblings do not, of course, share the same developmental environments at the same time because they differ in age. The passive genotype-environment correlation still operates for siblings, because they have the same parents, but to a lesser extent than for twins. MZ twin correlations for intellectual competence do not decline when active genotype-environment correlations outweigh the importance of the passive ones, because MZ co-twins typically select highly correlated environments anyway. DZ pairs, on the other hand, are no more genetically related than sibs, so that as the intense similarity of their early home environments gives way to their own choices, they select environments that are less similar than their previous environments and about as similar as those of ordinary sibs. Adopted sibs, on the other hand, move from an early environment, in which parents may have produced similarity, to environments of their own choosing. Because their genotypes are hardly correlated at all, neither are their chosen environmental niches. Thus, by late adolescence, adopted siblings do not resemble each other in intelligence, personality, interests, or other phenotypic characteristics (Grotevant, Scarr, & Weinberg, 1977; Scarr & Weinberg, 1978; Scarr, Weber, Weinberg, & Wittig, 1981). The early environments of biological siblings, like those of adopted children, lead to trait similarity as a result of passive genotype -» environmental effects. As biological siblings move into the larger world and begin to make active choices, their niches remain moderately correlated, because their genotypes remain moderately correlated. There is no marked shift in intellectual resemblance of biological sibs as the process of active genotype —» environment influence replaces the passive one. Summary In summary, the theory of genotype —> environment effects proposed by Scarr and McCartney (1983) describes the usual course of human development in terms of three kinds of genotype-environment correlations that posit cooperative efforts of nature and nurture. Both genes and environments are constituents in the developmental system, but they have different roles. Genes determine much of human experience, but experiential opportunities are also necessary for development to occur. Individual differences can arise from restrictions in environmental opportunities to experience what the genotype would find compatible. With a rich array of opportunities, however, most differences among people arise

234

Sandra Scarr

from genetically determined differences in the experiences to which they are attracted and that they evoke. In addition, the theory accounts for observations from research on twins and families that no other developmental theory can. How do genotypes and environments combine? An argument for correlations over interactions One must distinguish environmental events that on the average enhance or delay development for all children from those that account for variation among children. There can be "main effects" that account for variation among groups that are naturally or experimentally treated in different ways. Within these groups of children there still remain enormous individual differences, some of which arise in response to the treatment. It is rare that the variation between groups approaches the magnitude of differences within groups, as represented in the pervasive overlapping distributions of scores. In developmental psychology, we have usually been satisfied if the treatment observed or implemented produced a statistically reliable difference between groups, but we have rarely examined the sources of individual variability within the groups. Two major hypotheses represented in this volume by Bronfenbrenner and his group, and by the present chapter, disagree about the principal sources of individual variability within groups. Bronfenbrenner proposes that genotype-environment interactions play an important part in explaining variability. I propose that g-e interactions may be present, but of minor importance, and that g-e correlations have greater effects on development and individuality. Linear and nonlinear effects Most often, the same treatments that alter the average performance of a group seem to have similar effects on most members of the group. Otherwise, we would find a great deal of variance in person-treatment interactions; that is, what's sauce for the goose would be poison for the gander. For the kinds of deprivation or interventions studied most often in developmental psychology, the main effects seem not to change the rank orders of children affected. The main effects are real, but they are also small by comparison to the range of individual variation within groups so treated or not. Some children may be more responsive than others to the treatment, but I doubt that there are many situations in which disordinal interactions are the rule. Very few children lose developmental points by participating in preschool programs or gain by being severely abused in adolescence. The search for aptitude-treatment interaction (Cronbach & Snow, 1971) and genotype-environment interactions (ErlenmeyerKimling, 1972) has not produced dramatic or reliable results (Plomin, 1986).

How genotypes and environments combine

235

In studies of adoptive and biologically related families, the correlation of children's IQ scores with the educational level of biological parents is about .35, whether or not the parents rear their children (Scarr & Weinberg, 1983). Adopted children on the average have higher IQ scores than their biological parents as a result of the influence of their above-average adoptive parents. Taken together these findings support the claim that treatments can have main effects without overcoming genetic differences in children's responsiveness to those environments. Adopted children have IQ scores above those of their biological parents, yet the correlations of adopted children are higher with their biological than adoptive parents (Scarr & Weinberg, 1977, 1978, 1983). The average effects of treatments, such as adoption, seem to increase the mean IQ scores, but they do not seem to alter the rank order of individual differences, on which correlations depend. In the theory just presented, emphasis was placed on the correlation of genotypes with their environments, on the nonrandomness of people, places, and experiences. Other theorists, such as Bronfenbrenner (in this volume) prefer to think in terms of person-situation interactions, such that different people respond differently to different events in their lives. Similarly, Elder and Caspi (in this volume) emphasize the different responses of men to stresses, depending on their prior characteristics. The theoretical preference for what I will call genotype-environment (g-e) interactions has a long and honorable history. There just is not much empirical support, however fondly one looks (Plomin, 1986; Erlenmeyer-Kimling, 1972). The quest for aptitude-treatment interactions has a similarly disappointing history (Cronbach & Snow, 1977). Few consistencies can be found unless one is willing to deprive bright children of suitable instruction and bore them to death with task-analyzed learning that is appropriate for the retarded. Under those circumstances, which no real-life teacher would implement, one can find consistent decrements in the performance of bright children and consistent increments in the learning of dull ones (Snow, 1982). I submit that is not the result we were seeking. Under normal instructional conditions, bright children learn more than dull ones; the main effect of intellectual differences overpowers any interactional one. Yet, one is intuitively drawn to the idea that different genotypes have differential responsiveness to different environments. Perhaps the problem is that in ordinary life people selectively expose themselves to correlated environments, thereby largely obviating any necessity to demonstrate g-e interactions. Children who are terrible at sports do not usually go out for the varsity team. Children who will make the varsity do not usually place themselves in remedial physical education. Thus, we fail to observe the interaction that could occur if varsity athletes were demoralized by slow walks around the soccer field and the phys-

236

Sandra Scarr

ically ungainly were required to run 20 laps around the field. There must be something to the appealing idea of interactions; perhaps they rarely occur in the real world of self-selection and natural assortment. In animal studies where genotype-environment interactions are regularly reported, investigators use highly inbred strains (and their crosses) and restricted versus more normal environments in their tests. It may be that there are no human analogues to such animal studies, so that, conversely, the animal results do not apply straightforwardly to human populations. In human behavior genetic studies, estimates are seldom made of genotypeenvironment interactions or correlations. In large part, this is because there are no powerful tests of such terms. Twin designs are not suitable for estimating g-e interaction, because, even if it exists, it could be assigned to either the genetic or environmental term. If both MZ and DZ co-twins react equally similarly (or equally dissimilarly), any g-e interaction will be part of the environmental estimate. If the greater genetic resemblance of MZ twins leads them to have more similar reactions to the same events than DZ twins, any g-e interaction will be included in the genetic term (Plomin, 1986). Because we do not know in advance if g-e interaction exists or how it works on individual differences, we cannot use the twin method to find out. Adoption studies offer a more favorable but still weak test of g-e interaction. Unlike animal studies, where the genotypes can be controlled through generations of selective breeding and environments manipulated at the will of the investigator, human populations do not lend themselves to such tidy analyses. And the adoption study is only a partial solution to the problem of estimating g-e interactions in human populations. Here are the reasons - first, the traditional ones from quantitative genetics, then my developmental ones. Where's the interaction? In biologically related families, the resemblance between parents and children is potentially composed, in traditional variance terms, of g, e, cov g-e, and g-e interaction. In adoptive families, the variance terms are limited to e alone; because there is no g-connection, there also cannot be g-e connections between parents and adopted children. Measures of the natural or birthparents of adopted-away children correlate with their offspring because of shared genotypes only. Because there is no e connection, there can be no g-e connections either. Thus, a cross-tabulation of IQ values for adoptive and natural parents can potentially estimate the main effects of g and e and reveal any effects of g-e interaction by nonlinearities in the combined main effects. The intellectual genotypes of adopted children can be very roughly estimated

How genotypes and environments combine

237

Table 9.3. Adopted children's IQ scores by adoptive parent IQ and natural parent education (Transracial Adoption Study) Adoptive midparent IQ Low Mid High S.D.

12

N

X

S.D.

99.8 106.2 106.5 104.2 (9.51)

16 14 8

107 107.8 106.6 107.1 (8.02)

22 33 31

106.7 105.8 110.2 107.6 (9.06)

18 12 18

104.5 106.6 107.8 106.3 (8.66)

(9.08) (8.23) (8.69)

Note: F Natural parent education = 2.58, p environment effects. Child Development, 54, 424-35. Scarr, S., Weber, P. L., Weinberg, R. A., & Wittig, M. A. (1981). Personality resemblance among adolescents and their parents in biologically related and adoptive families. Journal of Personality and Social Psychology, 40, 885-98. Scarr, S., & Weinberg, R. A. (1977). Intellectual similarities within families of both adopted and biological children. Intelligence, 1, 170-91. Scarr, S., & Weinberg, R. A. (1978). The influence of "family background" on intellectual attainment. American Sociological Review, 43, 674-92. Scarr, S., & Weinberg, R. A. (1980). Calling all camps! The war is over! American Sociological Review, 45, 859-64. Scarr, S., & Weinberg, R. A. (1983). The Minnesota adoption studies: Malleability and genetic differences. Child Development, 54, 260-7. Shields, J. (1962). Monozygotic twins brought up apart and brought up together. London: Oxford University Press. Snow, R. E. (1982). Education and intelligence. In R. E. Sternberg (Ed.), Handbook of human intelligence. Cambridge, MA: Cambridge University Press. Teasdale, T. W., & Owen, D. R. (1985). Heredity and familial environment in intelligence and educational level - a sibling study. Nature, 309, 620-2. Toulmin, S. (1973). Human understanding. Chicago, IL: University of Chicago Press. Wachs, T. D. (1983). The use and abuse of environment in behavior-genetic research. Child Development, 54, 416-23. Watson, E. H., & Lowrey, G. H. (1967). Growth and development of children. Chicago, IL: Year Book Medical Publishers. Wilson, R. S. (1983). The Louisville twin study: Developmental synchronies in behavior. Child Development, 54, 298-316.

Name index

Acland, H., 7 Acock, A. D., 62, 63 Adams, G. R., 174 Ageton, S. S., 143 Allen, A., 181, 204 Allen, L., 90 Allen, V. L., 154 Allport, G. W., 14, 205, 206 Alwin, D. F., 8 Applebaum, M. I., 12 Arnold, J., 137 Arnold, L., 198 Arthur, J., 123 Aschenbrenner, B. G., 10 Bachman, J. G., 172 Bailey, S., 103 Baldwin, A. L., 41, 42 Baldwin, J. M., 184 Baltes, P. B., 40-1 Bandura, A., 69, 102, 183, 190, 213, 228, 240, 241 Bane, M. J., 7 Bank, L., 137, 138, 139, 143 Barker, R. G., 41-2, 51, 52, 66, 67 Barrett, J. E., 93 Bartholomai, F., 28 Baszormenyi-Nagy, I., 53, 65 Bates, J. E., 144 Baumrind, D., 8 Bayley, N., 31^2 Beattie, M., 6 Beck, S. J., 196 Becker, H., 155 Beckmann, M., 44 Belsky, J., 10, 60, 95 Bell, R. Q., 13, 33, 229 Bern, D. J., 14, 15, 73, 98, 181, 204, 205, 210-11, 212 Bern, S. L., 208, 212

245

Bengston, V. L., 62, 63 Bentler, P. M., 137, 152, 164 Berger, P. L., 58 Bergeron, G., 181 Best, J. A., 173 Biglan, A., 123 Binet, A., 181 Bleuler, M., 81 Block, J., 87, 89-90, 95, 192, 214 Block, J. H., 95, 214 Blyth, D. A., 88, \64nl Bott, E., 53 Botvin, G. J., 174 Bouchard, T. J., 223, 230, 232 Bourdieu, P., 58«1 Bowers, K. S., 181 Boyd, E., 121 Bradley, R. H., 219 Brake, M., 155 Brandtstadter, J., 153, 154 Brehm, J. W., 108 Brehm, S. S., 108 Brody, G., 10 Bronfenbrenner, U., 1-2, 3, 5, 9, 12, 28-9, 32, 33, 35, 36, 38, 40, 42, 50, 52, 66, 71, 73, 89, 93, 145, 154, 155, 188, 203, 206, 234, 235 Brooks-Gunn, J., 17, 188 Brown, K. S., 173 Buckley, S., 155 Bugenthal, D. B., 71 Bullivant, B., 56 Burns, A., 55, 56, 59, 65, 66, 67, 72 Burton, L. M., 4 Burton, R. V., 137, 181, 182, 191 Bush, D. M., 88 Buss, D. M., 14, 217, 218/, 242 Cairns, B. D., 15, 16, 55, 181, 184, 191, 194, 195, 196, 198

Name index

246 Cairns, R. B., 15-16, 17, 55, 118, 120, 181, 182, 184, 187, 191, 194, 195, 196, 198, 199 Caldwell, B. M., 219 Calkins, V., 198 Campbell, D. T., 20 Campbell, J. D., 137, 181, 182, 191 Capaldi, D., 134 Caplan, G., 92 Carlton-Ford, S., 164«1 Carter-Saltzman, L., 231 Carver, C. S., 153 Cashmore, J., 54, 57, 63, 71 Caspi, A., 4, 6, 15, 36, 44, 46, 51, 60, 81, 82, 87, 92, 94, 97, 98, 99, 100, 132, 163, 210-11, 212, 241 Cattell, R. B., 209 Cauble, A. E., 91 Cavan, R., 92 Chamberlain, P., 141 Cheek, J. M., 204 Chess, S., 9 Child, I. L., 89 Clarke-Stewart, A., 10 Clausen, J. A., 62 Cobb, J. A., 121, 134 Cobb, S., 9 Cochran, M., 35 Cohen, D., 7 Cohen, J., \12n5 Cohen, P., \12n5 Cohen, S., 164 Coie, J. D., 125-6 Coleman, J. S., 7 Cook, T. D., 20 Cooley, C. H., 183, 185 Cottrell, L. S., 211 Covey, M. K., 122 Coyne, J. C , 11 Crockenberg, S. B., 37-8 Cronbach, L. J., 217, 234, 235 Cross, C , 92, 94 Crouter, A. C , 28-9, 33, 38, 42, 89, 154, 155 Daniels, D., 12, 13 Dantchik, A., 204 Darley, J., 14 Darwin, C , 217 Dasen, P., 69 D'Avernas, J. R., 173 Dawes, R. M., 120 DeFries, J. C , 15, 224t, 227, 228, 237 deLone, R. H., 38 Delugach, J. D., 17

Dengerink, H. A., 122 Deur, J. L.,94 Diener, E., 14 Dishion, T. J., 115, 120, 130, 133/, 134, 138, 139, 144/, 210 Dodge, K. A., 17 Dolan, A. B., 232 Douglas, M., 70 Downey, G., 36, 44, 46, 94, 99, 100 Dreher, E., 156, 157 Dreher, M., 156, 157 Duner, A., 152 Dunn, J., 10, 13 Durkheim, E., 106 Dusek, J. B., 172 Dwyer, J. H., 136, 137, 138-41, 143, 165 Easterbrooks, M. A., 10 Eckenrode, J., 37 Edwards, M., 53 Eichorn, D. H., 82 Eick, K., 155 Eigenbrodt, J., 155 Elder, G. H., Jr., 3, 4, 6, 15, 36, 42, 43, 44, 46, 51, 60, 77, 79, 81, 82, 84, 85-6, 87, 89, 91, 92, 94, 97, 98, 99, 100, 103, 132, 210-11, 212, 235, 241 Elliott, D. S., 143 Emery, R. E., 91, 95 Emmons, R. A., 14 Endler, N. S., 209 Engfer, A., 44 Entwisle, D. R., 55 Epstein, J. L., 9 Epstein, S., 181, 184, 192 Erlenmeyer-Kimling, L., 234, 235 Eron, L. D., 62 Esveldt-Dawson, K., 196, 198 Eyferth, K., 152, 155, 156, 162 Eysenck, H. J., 209 Farrington, D. P., 17, 145 Fazio, R. H., 14 Featherman, D., 105 Feldman, K. A., 91 Feldman, S. S., 10 Fend, H., 169 Ferguson, L. L., 191 Ferguson, T. J., 194 Fischer, A., 172 Fischer, R. C , 172 Fisher, R. A., 30 Flaherty, J. F., 172 Flay, B. R., 173 Fleischman, M. J., 137

247

Name index Foner, A., 80, 84 Forehand, R., 196 Forgatch, M. S., 131, 132 Foucault, M., 58 Frame, C. M., 17 Freeman, F. N., 232 French, J. P., 9 Freud, S., 213 Friedman, L., 123 Fuchs, W., 172 Fuhrer, U., 155 Funder, D. C , 181 Furstenberg, F. F., Jr., 13, 17-19, 61, 62, 63 Galton, F., 32 Gamble, W., 60 Garbarino, J., 66, 71 Gardner, K., 120 Gardner, W., 147 Gariepy, J.-L., 191 Garmezy, N., 81, 88, 95, 230 Gewirtz, J. L., 121, 181 Gibbons, F. X., 204 Gibson, J. J., 155 Gintis, H., 7 Glueck, E., 94 Glueck, S., 94 Goldberg, W., 10 Goode, W. J., 95 Goodenough, D. R., 212 Goodnow, J. J., 8, 54, 55, 56, 57, 59, 63, 65, 66, 67, 71, 72 Gottlieb, G., 227 Gottman, J. M., 94, 118, 123, 124, 141 Grabow, S., 155 Grajek, S., 12 Green, J. A., 182 Green, K. D., 196 Greenberg, J., 183 Greenwald, A. G., 182, 184 Griffin, W. A., 147 Groeneveld, L. P., 20 Grotevant, H. D., 233 Guiller, H., 146 Gulotta, T., 174 Gunnel, P. K., 121 Gurin, G., 6 Gurin, P., 6 Guttman, L. A., 126 Handel, G., 62 Hannan, M. T., 20 Hansen, D. A., 8-9 Hareven, T., 53 Harms, G., 155, 162

Harter, S., 134, 182 Hartmann, H., 53 Hartshorne, H., 181, 182, 205, 212 Hauser, R. M., 8 Havighurst, R. J., 153, 156, 157 Hayduk, L. A., 55 Heath, S. B., 55 Heider, F., 53, 185 Henri, V., 181 Herjanic, B., 198 Hess, R. D., 62 Heyns, B., 7 Hinde, R. A., 10, 11, 53, 93 Hineline, P., 121 Holahan, C. J., 155 Holzinger, K. J., 232 Honzik, M. P., 90 Hood, K. E., 16 Hops, H., 123, 134 Homey, K., 212 Huba, G. J., 152, 164 Huesmann, L. R., 62 Huizinga, D., 143 Hunt, J. McV., 209 Jackson, B., 68 James, W., 102 Jaworski, B., 91 Jencks, C . , 7 Jennings, M. K., 62 Jessor, R., 152, 154, 164, 174 Jessor, S. L., 154, 164 Johnson, L. D., 172 Johnson, M., 80, 84 Jones, D. R. W., 155 Jones, E. E., 181 Joreskog, K. G., 42, 165 Joy, C. B., 91 Juel-Nielsen, N., 232 Kagan, J., 31«2, 88 Kain, E. L., 81 Kamin, L. J., 231 Kaminski, G., 155 Kandel, D. B., 152, 164 Kantor, B., 20 Kaplan, H. B., 163 Kastner, P., 153, 157 Kazdin, A. E., 196, 198 Kelley, H., 181 Kelly, G., 208 Kendrick, C , 10 Kendrick, D. T., 182, 204 Kendzierski, D., 14 Kenny, D. A., 139

Name index

248 Kersell, M. W., 173 Kessler, R. C , 11, 164 Kidd, K. K., 227 Knight, R., 57, 71 Knorr, G. L., 134 Kohlberg, L., 183, 208 Kohn, M., 4, 5, 6, 89, 110 Kraines, R. J., 62 Krantz, J. Z., 232 Kumka, D., 46 Kuo, Z. Y., 184 Kupersmidt, J. B., 126 Lamb, M. E., 118, 120 Langer, E. G., 121 Larson, R. J., 14 Lautrey, J., 57, 69 Lav, R. C , 6 Ledingham, J. E., 181 Lee, C. L., 144 Lefkowitz, F., 62 Lehmann, H. C , 204 Lerner, J. V., 172 Lerner, M., 53 Lemer, R. M., 172 Levenson, R. W., 123, 124 Levin, H., 137 Levine, A., 137 Le Vine, R. A., 2 Levinger, G., 53 Lewin, K., 1, 13, 25, 26/zl, 27, 28, 31, 40 Lewis, C , 56 Lewis, M., 73, 188 Liddle, C , 17, 102 Liebermann, M. A., 93 Liker, J. K., 91, 92, 94 Linville, P. W., 182 Lippmann, Walter, 78 Lipsett, L. P., 41 Loar, L. L., 196, 198 Loeber, R., 138, 139 Loehlin, J. C , 15, 227, 228, 237 Logan, J. A., 152 Lorber, R., 125 Lorence, J., 46 Lowrey, G. H., 227 Lynd, H. M., 78 Lynd, R. S., 78 Lytton, H., 15-16, 182 McAvay, G., 31 McCall, R. B., 12 McCall, R. R., 16 McCartney, K., 15, 210, 226, 230, 233, 239, 242 Maccoby, E. E., 11, 12, 66, 137

McCord, J., 94 McCubbin, H. I., 91 Macfarlane, J. W., 84, 90 McLeod, J. D., 11 Madge, N., 88 Magnusson, D., 152, 154, 183, 199, 209 Margulies, R., 164 Markus, H., 182 Marlowe, J. H., 137 Marsden, D., 68 Martin, J. A., 12 Martin, S. S., 163 Masten, A. S., 230 Matheny, A. P., Jr., 232 Maughan, B., 7 May, A., 205, 212 May, M., 155 May, M. A., 181, 182 Mayr, E., 217 Mead, G. H., 183 Mead, M., 55 Medrich, E. A., 155 Megargee, E. I., 213 Mehan, H., 73 Menaghan, E. G., 93 Merton, Robert, 79, 80 Michelson, S., 7 Mills, C. W., 3-4, 110 Mills-Byrd, L., 199 Minuchin, P., 10 Mischel, W., 181, 204, 206-7, 213 Moen, P., 6, 66, 71, 81 Moorehouse, M., 5 Moos, B. S., 219 Moos, R. H., 219 Morgan, S. P., 17 Morrison, A., 95 Mortimer, J. T., 46 Mortimore, P., 7 Moss, H. A., 31^2 Muchow, H., 155 Muchow, M., 155 Muchowski, L., 157 Mullen, J. T., 93 Murray, H. A., 209 Myers, I. B., 212 Nash, S. C , 10 Neckerman, H. J., 191 Needle, R. H., 91 Nerlove, S. B., 227 Neugarten, B. L., 62 Newcomb, T. M., 91 Newman, B. ML, 156, 157 Newman, H. G., 232 Newman, P. R., 156, 157

249

Name index Niemi, R. G., 62 Nisbett, R. E., 121, 181 Noack, P., 14, 70, 105, 154, 156, 157, 160, 162 Nurius, H., 182 Olweus, D., 145, 209 O'Malley, P. M., 172 Osteen, V., 123 Ouston, J., 7 Owen, D. R., 224t Parke, R. D., 10, 73, 94 Passeron, J., 59 Patterson, G. R., 11, 15, 16-17, 94, 96, 99, 115, 116, 118, 120, 121, 122/, 123, 124, 127, 130/, 131, 132, 133/ 134, 135/ 136, 137, 138, 139, 141, 142, 143, 144/ 145, 146, 210, 211, 213, 241, 242 Patterson, J. M., 91 Payne, R. T., 155 Pearlin, L. I., 93 Perrin, J. E., 198 Perry, C. L., 174 Pervin, L. A., 9 Peterson, A. C , 153 Peterson, D. R., 204 Piaget, J., 184, 227 Pillegrini, A. D., 10 Plomin, R., 12, 13, 15, 219, 220, 224r, 227, 228, 234, 235, 236, 237, 240 Poole, M., 56, 68 Preissing, C , 155, 162 Pysczczynski, T., 183 Quinton, D., 17, 102 Ranck, K., 92 Reese, H. W., 41 Reich, W., 198 Reid, J. B., 115, 118, 120, 125, 130/ 133/ 137, 141, 144/ 210 Reitzle, M., 165, 169 Richtermeier, A., 155, 162 Riley, D., 37 Riley, M. W., 3, 80, 84 Robbins, C , 163 Robins, E., 60 Robins, L. N., 90, 97, 102 Robinson, J. P., 126 Rockwell, R., 103 Rodgers, W., 9 Rogosa, D., 139, 142 Roizen, J., 155 Rollins, B. C , 62 Rosenbaum, J. E., 8

Rosenfeld, H. M., 121 Rotter, J. B., 213 Rubin, V., 155 Rubin, Z., 66 Rule, B. G., 192 Russell, Graeme, 54 Rutter, M., 7, 17, 87, 88, 95, 102, 146 Ryan, K. B., 173 Ryder, N., 79-80 Sackett, G. P., 120 Salkind, N. J., 155 Sameroff, A. J., 145 Sarason, S. B., 19 Scarr, S., 12, 15, 31, 105, 182, 209-10, 219, 220, 223, 224r, 225, 226, 227, 228, 230, 231, 232, 233, 235, 237-8, 242 Scarr-Salapatek, S., 43 Schacter, F. F., 10 Schaefer, W. S., 31«2 Schaffer, R., 92 Schaie, K. W., 40-1 Scheier, M. F., 153 Schlundt, D. G., 17 Schnedler, R. S., 122 Schneewind, K. A., 44 Schneirla, T. C , 184 Schocken, I., 17 Schoggen, P., 52 Schooler, C , 4, 89, 110 Schroer, S., 169 Schuck, J. R., 137 Schuhler, P., 157 Schulz, R., 19 Schwabe, H., 28 Schwartzman, A., 181 Schwarzer, R., 169, 172 Sears, R. R., 137, 182 Sears, S., 181 Seaver, W. B., 10 Seifer, R., 145 Sewell, W. ML, 8 Sharrock, W. W., 58 Shennum, W. A., 71 Sheridan, A., 58 Sherman, L., 123 Shields, J., 232 Shweder, R. A., 57, 58 Sidman, M., 141 Sigel, I. E., 10 Silbereisen, R. K., 14, 70, 105, 152, 153, 154, 155, 156, 157, 162, 163, 165, 169 Simmons, R. G., 88, \(An\ Skinner, B. F., 120, 207 Skinner, M., 131, 132 Sloman, J., 66

Name index

250 Smith, M., 7 Smith, R., 145 Smith, R. S., 88 Smith, T. E., 63 Snipper, A. S., 227 Snow, R. E., 234, 235 Snyder, J. J., 123 Snyder, M., 14, 204 Sorbom, D., 42, 165 Sorenson, A., 105 Spark, G., 53, 65 Spencer, M., 56 Spiga, R., 153 Spyrou, S., 120 Stattin, H., 154 Stephenson, G. R., 118, 120 Stevenson-Hinde, J., 10, 11 Steward, M., 61, 64-5, 70 Stodolsky, S., 58, 67 Stone, C. P., 41-2 Stouthamer-Loeber, M , 134, 136, 138 Stringfield, D. O., 182, 204 Strober, M. H., 53 Sullivan, H. S., 211 Suomi, S. J., 118, 120 Swann, W. B., Jr., 14 Syme, S. L. , 164 Tal, M., 11 Taplin, P. S ., 125 Teasdale, T. W., 22 Tedin, K. L . , 6 3 Tellegan, A,., 230 Tesser, A., 10 Thibodeaux, S., 10 Thomas, A.., 9 Thomas, D. L., 62 Thomas, D. S., 209 Thomas, W. I., 60, 78-9, 106, 107-8, 209 Thorsen, C , 123 Thrasher, F. M., 155 Toulmin, S., 219 Troll, L. E., 62 Tulkin, S. R., 34-5, 45, 46 Tuma, N. B., 20 Turnbull, J., 11 Uchtenhagen, A., 173 Valsiner, J., 67 Van Cleave, E. F., 88

Van Nguyen, T., 36, 46, 51, 60, 81, 82, 87, 92, 132 Vaughn, R., 164 Vernon, P. E., 204 Viken, R., 131 Volkart, E. H., 78 von Bertalanffy, L., 1 Vosk, B., 196 Wachs, T. D., 6, 238 Wachtel, E. F., 19 Wachtel, P. L., 14, 19, 210, 211 Wallwork, E., 184 Watson, E. H., 227 Weber, P. L., 233 Weinberg, R. A., 224f, 230, 232, 233, 235, 237-8 Weinrott, M., 137 Werner, E. E., 88, 145 West, D. J., 17, 145 Whiting, B. B., 66 Willis, P., 155 Willoughby, T. L., 198 Wills, T. A., 164 Wilson, R. S., 232 Wilson, T. D., 121 Witkin, H. A., 212 Wittig, M. A., 233 Witty, P. A., 204 Wohlwill, J. F., 6, 155 Woo, D., 122 Woolever, F., 35 Wortman, C. B., 11 Wright, H. F., 51, 52, 66, 67 Wright, S., 42 Wrong, D., 13 Yarrow, M. J., 137 Yarrow, M. R., 181, 182, 191 Younger, A., 181 Youniss, J., 117 Zahn-Waxier, C , 11 Zank, S., 165, 169 Zanna, M. P., 173 Zetterboom, G., 152 Zigler, E., 89 Zimmer-Hofler, D., 173 Zinnecker, J., 172 Znaniecki, F., 78, 107-8

Subject index

abstract thinking, 187, 189, 192 academic failure, 131, 134-6, 142, 143 accentuation hypothesis, 60, 91 access, in description/perception of settings, 67-70; meanings of, 66-7; observations prompting questions about, 64-6; perceptions of, 68, 70, 71; and qualities of people, 70-3 access to information, 8, 58, 67; blocked by adults, 58; effect on development, 58 achievement: prospects for, 98, 103; training for, 89 action-reaction sequences, 117, 118, 120 action-theoretic approach, 153 adaptation(s): in attempts to regain control, 108, 109; to economic hardship, 82, 96, 107; new, 225; options and resources in, 85-6; problem behavior as, 90 adaptive functions: of self cognitions, 184, 190, 199-200; of self evaluations, 191 adaptive resources: marital integration as, 92 adjustment, 212, 219-20 adolescence: change in, 187-9; constructive role of problem behavior in, 152-80; insider status in, 58; self concepts in, 194-5; selfother ratings in, 192-3; leisure time as context of development in, 154-6; effects of Depression in, 84, 87-8; as mother, 17-18; and peer group, 117; role in shaping own development, 153 adopted siblings, 223, 230-3, 235 adoption studies, 12, 236-8, 242 adoptive families, 224; parent-child IQ correlations in, 235; parent-child resemblance in, 236-8 adulthood, 42; transition to, 97, 103, 104, 152, 154 adults: vested interests in advice of, 56, 57 age, 100; and effects of economic hardship on development, 86-8; and perceptions of set-

251

tings, 71-2; as personological construct, 40-1; in society and history, 79, 80; social meaning of, 77 age-graded roles, 97-9 age strata, 80 aggression, 95; etiology/treatment of, 241-2; irritability and, 96; microsystem and intervention in, 11-12; self-other ratings of, 194, 197; social networks and, 16-17 alcohol use, adolescent, 160, 161; and selfesteem, 167, 169-73 alienation, 173-4 anarchy progression, 11 antisocial behavior, 94; determinants of, 142; discipline and, 146-7; feedback loops in, 142-3; and homework completion, 134; in school, 142-3; stages in, 129-36, 210; temperament and, 145; training for, 132-4, 138; see also aggression; coercion process; problem behavior antisocial traits: progression in, 125-8, 129 aptitude-treatment interaction, 234, 235 arbitrariness: in parenting, 94, 96 aspirations, 106-7, 108 attitudes: and behavior, 14; intergenerational transmission of, 91, 99-102 authoritarian personality, 208 authority, 211 autonomy values, 4-5 aversive behavior: learned, 129, 211; see also antisocial behavior; problem behavior aversive stimuli, 120-1, 122 behavior(s): effect of economic hardship on, 91; as function of person and environment, 203, 208-9; generalized from family to peer group, 116-17; intergenerational transmission of, 91, 99-102; across life span, 9 9 100, 115-16; nomothetic classification of, 207-8, 212, 213; school, 7-8

252 behavior genetic studies, 10, 12, 15; family resemblances/parent-child effects, 223-5; genotype-environment interactions, 236 behaviorism, 206-8, 213, 242 beliefs: and family management skills, 145-6; about self and reality, 189-91 Berkeley Guidance Study, 15, 78, 84-5, 8 5 9, 89-105 Berlin Longitudinal Study on Adolescent Development, 155-6 bidirectional influences, 42-3; in coercive exchanges, 132, 147; person/context, 163-4 biological famlies: correlations within, 224; genotype —> environment effects in, 229; parent-child IQ correlations in, 235; parentchild resemblance in, 236 biophysical change, 187-8 catastrophes, 240-1 cathartic activities, 163, 174, 175; as mediator between self-esteem and substance use, 164, 167, 169-72 causal attribution, 181 causal models, 42-3, 79; effects of economic hardship on development, 87, 92; and theory construction from observations, 219 cause/causality: in genotype/environment combination, 220, 221, 223 Centre for Contemporary Cultural Studies, Birmingham, 155 change, 3, 91, 93; in development, 40, 44, 229-30; family and, 114, 138; in life experience, 109; microsocial mechanisms for, 118-36; in personality, 97; process of, 114, 115-16; social /individual, 79; see also social change Chicago School of Sociology, 78 children: access to wider society, 65, 66; as agents of society, 65; in Depression, 86-9; difficult, 144-5, 241-2 (see also aggression); effect of stress on, 81-9; and genotype-environment combination, 219-20; influences on family relations, 93-4; relationships with families and communities, 50-76 cigarette use, 160, 161, 162, 164; and selfesteem, 167, 169-73, 174, 175 class; see social class class-theoretical models, 27-32 coercion process, 8, 115, 116-17; dyadic exchanges in, 118, 147, 241-2; mechanisms in, 117; microsocial structure and, 123-5, 128-9; production of change in dyadic exchanges, 120-3; stages in, 120, 123-5, 129-36, 142; testing stage model of, 136— 43

Subject index cognition: in self-organization, 183; see also self cognitions cognitive development, 69, 184, 187, 189; cultural differences in, 227; Piagetian theory of, 57; sex typing in, 208 cognitive style, 212 cohort(s), 110; as concept for studying life course, 80; confounding effects of age and, 41; and historical events in life course, 84 Colorado Adoption Project, 237 communication: changes in abilities in, 187, 188; self cognitions in, 184, 185 community(ies), 2, 66-7; description of, 52; and parent-child interactions, 64-5; relationships with children and families, 50-76 competence, 85, 91; in stressful situations, 81 competition, 51, 53, 65-6 conduct disorders, 116-17; see also antisocial behavior; problem behavior context(s), 2; family in, 143-6; hierarchy of systems of, 38-9; in human development, 77; and interventions, 20; in life-course perspective, 77, 79, 85, 100; process and, 207-9; in process-person-context model, 36, 38 continuity: in development, 40, 44; interactional/cumulative, 98-9; in personality, 97 control, 59, 106-8; over access, 51; force in, 94; locus of, 37; perceived, 19 control cycles (model): and evolving life course, 105-9 control potential, 109; loss of, 106-7 coping mechanisms: social support and, 164 coping strategy, 96; problem behavior as, 152, 154, 174-5 cross-sectional research design, 41-2 cultural differences: in cognitive development, 227 cultural values: transmission of, through parents, 61-4 culture, 50, 240; continuity in, 51; description of, 52; influence on development, 3-6; macrosystem as blueprint of, 39; and personality variables, 212-13; relation of family to, 114; of schools, 7 cumulative continuity, 98-9, 210 danger: perceptions of, 69-70 defense mechanisms, 212, 213 delinquency, 78, 116, 131, 134, 138; risk of, 145-6 depression, 143; effects on family relations, 11; in mothers, 123; rejection and, 131, 136; self appraisal of, 183 deprivation, early, 85; effects of, 17, 77; fami-

Subject index ly, 82-4, 87-8, 90-105; see also economic hardship determinism (genetic), 224-5, 230 development: agentic conceptualization of, 14; and average tendencies/individual variation, 218, 225; defined, 58; delayed effects in, 142-3; effect of maternal employment on, 5-6; family income loss and, 82-4; family relations and, 10, 11, 12-13; genetically programmed, 242; genotype —* environment effects in, 226-8, 229, 233-4, 240; holistic models of, 184-5; and individual differences, 217-44; and individual differences in information processing, 225-6; macrosocial influences on, 3-6; opportunities for, 55; parental knowledge about, 220-3; and problem behavior, 153-6, 163; proximal influences in, 6, 7-13; realignment of, in self concepts, 186-9; self-esteem in, 163; separation of genetic and environmental effects on, 227-8; as series of behavioral reorganizations around developmental tasks, 100; shifts in perceptions of settings as form of, 70; stability of, and enhancement of self, 193-4; substance use in, 173-4 development as action in context (concept), 152, 153, 154 developmental change, 40, 44; in genotype —> environment effects, 229-30 developmental integration, 191, 199; self cognitions in, 184-5, 189-90 developmental outcomes, 31; connections between settings in, 55, 59; defining, 44-6; effect of environment on, 34, 50; participation and, 52; research paradigm and, 26 developmental status: correlation of environments with, and individual differences, 2402; measures of, 47 developmental tasks, 97, 100, 153; adolescent, 156, 157; substance use in mastering, 174-5 developmental theory, 1-3, 182-3, 224-6, 228-9, 242; genotype—environment correlations in, 228-9 deviance, 131; in defense of self, 163 discipline, 127, 134, 141; and child behavior, 136-7, 138, 146-7; consistent, 94-5; effects of changes in, 136; effects of stress on, 132; inept, 131-2, 242; marital discord and, 96 disposition(s), 116; and behavioral style, 99; and economic hardship, 91, 93; generality of, 181; and selection of environment, 15210 division of labor: in children-families-communities relationship, 65-6; in family, 53-4 divorce, 80, 98, 105, 107, 146

253 dizygotic (DZ) twins, 15, 223, 230-3, 236 dreams: self cognitions in, 184, 186 drugs: adolescent use of, 157, 160, 163, 173 dyadic exchanges: as mechanism for change in families, 118-20, 147; microsocial structure and, 123-5; production of changes in, by coercion process, 116, 120-3, 128-9, 131

ecological perspective, 1-2, 3, 9, 12, 47, 93 economic conditions, 30; contextual analysis of, 85; effect on family functioning and child development, 51, 53, 82-4, 96; stress imposed by, 59 economic hardship: and development, 44, 89105; linkages to children's behavior, 90, 94-6; pre-crisis histories and, 90, 91-4 education, 7, 8; in life course, 97, 98, 102, 103; and military service, 105 ego: compromising, 189-91, 199 ego defense, 213 emotional support, 60, 61 enhancement model, 50-1, 58-61, 64, 65, 66, 70-1 environment(s), 10, 12, 31, 50; in behavior, 203; in control of self concepts, 183; correlation with individual differences and developmental status, 240-2; and development, 2, 3, 11-13, 32-3, 34, 226, 227-8; and family relationships, 10, 12-13; genotypes and, 15-16, 209-10, 217-44 {see also genotype —» environment effects); individual's effects on, 13-16; and intelligence, 224; linkages between levels of, 2; persons in reciprocal transaction with, 21011, 214; in process-person-context model, 38-40; representations of, 1-13; in research paradigms, 26; shaped and selected by individuals, 99, 210, 214, 2^8, 229, 231, 2356; in social address model, 29; uncorrelated, 240 environmental change, 88, 89, 110; children's response to, 90; differential effects of, 8 5 6; individual in, 105-6; vulnerability to, 86-8 environmentalism, naive, 224-5, 230 environmental similarity: model of, 230-2 equivalence classes, 204, 205, 206-7, 212, 213, 214 escape-avoidant conditioning, 121 events: random vs. correlated, 240-2 evolutionary theory, 217, 225, 242 exosystem, 2, 38-9, 47 experience(s): change in, 109; correlated, 240; directed by genotypes, 15-16, 225, 226, 227, 229, 233-4; events shaping, 240-2;

254 experience (cont.) individual differences in, 225-6; role in development, 225, 233 family(ies), 2, 3, 10, 39; in context, 143-6; as developmental context, 3, 9-13; and educational/occupational achievement, 9, 31; and genotype-environment combination, 219-20; single-parent, 132, 146; social position of, 5; and state, 65-6; study of change in, 138; as unit(s), 50, 52-4; variable responses to economic hardship, 82-4, 90; see also adoptive families; biological families; parent-child interactions family dependency, 83, 86, 87, 96 family deprivation: effects on development, 82-4, 87-8; in Great Depression, 90105 family discord, 84, 87 family disorganization, 78, 91-2; economic hardship and, 93-4, 96 family interactions: aversive, 94, 100, 101, 102; destructive behavior patterns in, 99, 211; developmental research on, 32; as predictor of grade point average, 8-9; see also parent-child interactions family management, 94, 96, 131; beliefs, values, neighborhood and, 145-6; causal status of, 138-9, 141; and child behavior, 136-7, 146-7; temperament and, 145 family networks, 53 family relationships, 53-4; change in, in response to economic hardship, 83, 84; child influences on, 96; with children and communities, 50-76; effect of economic hardship on, 91-2; interdependence of, 10-12, 16; problem behavior in, 100-2; see also family interactions family research, 12, 229; accounting for observations from, in genotype —» environment effects theory, 227, 230-2, 234, 242 family resemblances, 223-5 family roles, 90 family stress, 93 family systems, 65; analysis of, 53 fathers, 54; and effects of economic hardship, 92-3; influence of work experience of, on intellectual functioning of sons, 89; role in mother's treatment of child, 10; role of, 5, 90 feedback loops, 117, 142-3; positive, 131, 138 fictions: adaptive, 186, 199; self-enhancing, 189 Field Dependence/Independence, 212 field-theoretical formulations, 27, 28 fighting, 127-8, 129, 138

Subject index "fit": person/context, 163, 172; person-environment, 9 functional equivalents: for problem behaviors, 174-5 gender constancy, 188-9 generalization/generalizability: boundaries of, 84; of processes, 38; of scores in stage models, 137-8; of social patterns, 116 genetic differences: and individual differences in experience, 226 genetic effects in development: separation of environmental effects from, 227-8 genotype-environment (g-e) correlations, 234, 239-40 genotype —> environment effects: and changing similarities among siblings, 232-3; developmental change in, 15-16, 229-30; and family research, 230-2; model of, 226-7; propositions in, 228-9; theory of, 225-34 genotype-environment (g-e) interactions, 23440; objections to theory of, 236-40 goals, adolescent, 153, 157-8; leisure-time activities in, 174-5; related to developmental issues, 175-6; pursued in leisure settings, 162; strategies in attainment of, 158— 60; substance use and, 163, 174 goals: self cognitions in, 184, 186 group constraints: and interventions, 19 health: effect of economic stress on, 88; impact of social networks on, 164; views about, 70 health services, 37-8 historical change: influence on development, 3-6 historical contexts, events, 2; cohorts in, 84; effect on life course, 79, 110; life patterns in, 79-80; timing of, 88 identical twins; see monozygotic (M2) twins idiographic approach: content in, 207, 208, 212, 213 idiographic balance: self concepts, 194-5 idiographic/morphogenic approaches to personality, 204-6 individual(s): as agent of change, 114; contribution to own development, 3, 20, 38; effects on environment, 13-19; selection of environment(s), 99, 210, 214, 228, 229, 231, 235-6 individual differences, 183, 206; in coping with stress, 91; correlation of environments with, and developmental status, 10, 227-8, 240-2; development and, 217-44; in evolutionary theory, 225; genetically determined,

Subject index 227-8, 233-4; in intelligence, 223-5; in overcoming limitations of problematic childhood, 102; in perceptions of world, 71-3; predictability of, 16; in response to drastic change and stress, 93; in self-other ratings, 192-3; stability of, 118; theory of, 242 individual-differences approaches to personality, 205-6 individual-family functioning: interactive view of, 100-2 infants, infancy, 227; genotype —> environment effects in, 229-30 information domains: open/restricted, 58 information: managed (prepackaged), 57-8, 60, 63-4, 67; routes to gaining, 60; see also access to information information processing: individual differences in, 209-10, 212, 225-6 instrumental support, 60, 61 intellectual flexibility: work and, 4-5 intelligence, 230; environment and, 219-20; individual differences in, 223-5; parental knowledge and, 220-3 interactional approach: life/social history, 7 9 80; to personality, 203, 208-11; to study of families and individuals, 106 interactional continuity, 98-9, 211 interaction: meanings of, 182, 209, 210-11; phenomenological approach to, 209 interactions: between features of individuals/families and settings, 70; community/individual, 51; face-to-face, 2, 50; interpersonal, 32; intrapsychic, 213-14; see also family interactions intergenerational transmission of behavior and attitudes, 91, 99-102 internal synchrony, 184, 186-7 intervention(s), 3; improving intelligence, 222; individual-environment effects and, 11-12, 19-20 irritability, 116; in family exchanges, 143-6; parental, 96, 132, 147; and response to economic hardship, 91, 93-6 knowledge: openness of, 58; of parent, and intelligence of child, 220-3; and power, 58; relevance of, 58; tagged, owned, 58 legal policies: effect on family, 51 leisure activities: substitutive, 174-5 leisure settings: developmental issues in, 15662; role in mastery of developmental tasks in adolescence, 152, 163-73 life course: age-graded, 108; alteration in, 109; childhood behavior and adult, 90, 97-9; control cycles and, 105-9; families in, 88;

255 military service in, 102, 103-5; occupational attainment in, 46; study of development in, 42-3, 44; turning points in, 91, 102-5, 214 life-stage hypothesis, 80, 84, 85-6; in comparison of Oakland and Berkeley cohorts, 85-9 life transitions, 91, 102-5, 214; in chronosystem paradigm, 42; see also transitions linkages: between antecedents and outcomes, 89-90; between developmental contexts, 8 9; between economic hardship and children's behavior, 90, 93-5; between social change and human development, 77, 78; between social change and life patterns, 110; family patterns as, 83/ longitudinal studies, 31, 77, 78-9, 88; active role of, in life-course perspective, 80, 85; individual's effect on environment, 15; juvenile delinquency, 116, 134; occupational attainment, 46; person-context link in, 16, 17-19; problems of testing stage models in, 136-43; self concepts, 191-8; randomized, 141-2 macrosocial level, 2, 117, 146; behavior changes at, 120; influences on development, 3-6; integration with microsocial, 131, 147 macrosystem, 3, 39 marital discord, 98, 105; in linking economic hardship to children's lives, 94-5; problem behavior and, 100, 101, 102 marital relationships, 10, 143; child influences on, 96; economic hardship and, 84, 91-2 marriage: idealization in, 107; in life course, 97, 98, 102 maturation, 105, 115, 225, 227 measurement, 118; in stage models, 137-8; in study of self, 199, 200 measurement nonconvergence, 181-2 measurement theory, 182-3 menarche, 41-2, 154 mesosystem, 2, 9, 38, 39, 47 Michigan Panel Study of Income Dynamics, 79 micro-environments, 241 micromodels, 82-4 microsocial analysis, 120 microsocial environment: integration with macrosocoial, 131-4, 147 microsocial exchanges: changes in, 120-3; patterns of, 147 microsocial mechanisms for change, 118-36 microsocial processes: effect of discipline on, 136-7 microsocial progression: integration with trait

256 microsocial progression (cont.) progression, 128-9; target child and sibling exchange, 123-5; traits define, 125-8 microsocial structure, 123-5 microsystem, 2, 38, 39 microsystem models, 11, 32-3 military service, 97, 98, 102, 103-5, 109 Minnesota Adolescent Adoption Study, 237-8 models: class-theoretical, 27-32; joint functions in, 26-7; parent effect on child intelligence, 222-3; see also under individual models, e.g., process-person-context model moderating variable strategy, 204-5, 206 monozygotic (MZ) twins, 15, 219, 223, 2303, 236; reared apart, 230, 232 morphogenic approach to personality, 204-6 mother(s), 54, 118; employment of, 5-6; middle class, 34; parenting skills, 145; single, 132, 147; teenage, 17-18, 37 mother-child relations: and effect of economic hardship, 93-4 mother—infant interactions, 34, 35; support networks in, 37 motivation, 187-8 multimethod research, 182; in developmental psychology, 198, 199; in personality, 182-3 multiple settings: joint effects of, 8-9 naturalistic field study(ies): developmental issues in leisure places, 156-62 negative reinforcement, 121 neighborhood(s), 3; and family management skills, 145-6 nomothetic approach to personality, 204, 2057; process in, 207-8, 212, 213 noncompliance, 126, 127, 128-9; and academic achievement, 134; generalization to school, 129-31 nonnormative events, 41 normative experiences, 41, 42 Oakland Growth Study, 78, 81, 82-4; Berkeley cohort compared with, 85-9 observation (method), 32; in field studies of adolescent leisure settings, 156, 160-2; and self-ratings, 181, 182, 199; theory construction from, 218-33 observational learning: in sex typing, 207 occupational attainment, 6, 46; education in, 103 occupational conditions, 4-5 options: adaptive, 85-6, 107; in control cycle, 109; prior to economic loss, 81-2, 91, 93; in reaction to environmental change, 85-6; of school underachievers, 143

Subject index parental competence: contextual theory of, 60-1 parent-child effects, 219-25, 229 parent-child interactions, 3; across social addresses, 6; community in management of, 64-5; and impending arrival of sibling, 41, 42; in process models, 33 parent-child relationship: and agreement on values, 62, 63; and effect of economic hardship, 93-4; effect of marital discord on, 95; mediating stress through, 85; models of, 9— 10, 13, 16; see also family relationships parent hostility, 100, 101 parent training, 137, 141 parenting: and child behavior, 136-7, 139; inept, inadequate, 98, 146, 242; punitive, 94, 95, 100 parenting skills: with difficult children, 145, 242; institutional rearing and, 17; outside stressors and, 131-2 parents, 5, 53-4, 65; adoptive, 231-2; as agents of society, 13, 51, 61-4; and aggressive children, 242 (see also coercion process); distancing, 127-8; and genotype/environment combination, 219-20; perceptions of outside world, 71; rejection of coercive child, 130, 131, 134-6; transmission of knowledge and intelligence, 220-3 pathologic processes: in families, 115, 123 peer contact: and self-esteem, 163; target and frequency of, 156, 157, 158, 160-2 peer group(s), 55, 116-17; deviant, 143, 155; function of, 117; rejection of coercive behavior by, 117, 126, 129-30, 131, 134-6, 142, 143 peer group integration, 156, 157, 162, 174, 175; strategies for, 176-7; substance use in, 172-3 perception(s): of access, 68, 70, 71; of agreement in values (parent-child), 62-3; of power of parents, 65; of settings, 67-70, 71-3 personal attributes model, 29-30, 31, 46 personality, 226, 230; and academic development, 9; continuities in, 97, 102, 210, 21112; idiographic/morphogenic approaches to, 204-6; interactional approach to, 208-11; measurement theory and, 182-3; nomothetic principles in, 204, 205, 208, 212, 213; research paradigms in, 203-16; and selection of environments, 14 person(s): as agent and consequence of changing life paths, 77; and behavior, 209; in reciprocal transaction with situations, 210-

Subject index 11; situation as function of, 209-10; see also individual(s) person/behavior equivalence classes, 204 person-context link: across time, 16-19; and interventions, 19-20 person-context model, 31-2, 46 person-environment models: applied to multiple environments, 8-9 person-situation interaction, 235 persons-in-situations: as unit of analysis, 209 phenotype(s): and environment, 226-7; and genotype(s), 231, 238-9; variation in, 225 phenotype-environment correlations, 239-40 phylogenesis, 182 physical world: self cognitions and, 184, 186 Piagetian theory, 56-8, 69 plans, 108-9; self cognitions in, 184, 186 Pollyanna bias, 188 population approach, 217-18 positive feedback loops, 131, 138 positive reinforcement, 136 poverty, 36-7, 79; see also economic hardship predictability: in developmental paradigms, 29-30 prejudice, 56 prevention: of problem behaviors, 174 problem behavior, 100; and adult life course, 90, 97-9; childhood stress and risk of, 8 1 9; constructive role of, 152-80; as coping strategy, 152, 154, 174-5; development and, 153-6; as effort to master age-typical difficulties, 154, 173-5; environmental selection and, 15; and response to economic hardship, 85, 89, 90, 94, 97; stressful antecedents of, 78; and unstable family relationships, 100-2 problem behavior proneness, 154 process: analyzing: nature of access, 664-73; analyzing: questions about enhancement, 58-61; analyzing: nature of transmission, 61-4; concept of, 129; and content, 207-9, 212-13; defined, 114; in developmental paradigm, 27; family, 114-51; generalizability of, 38; in human development, 77; multiple, 114-15; in past events in evolution of patterns of psychological functioning, 45-6; in process-person-context model, 36; in relationships among children, families, communities, 50-1; short-/long-term effects, 117, 136; in social address model, 29; study of, 146; varies as joint function of context, 36 process/content distinction, 207-8 process-context models, 33-5, 39 process paradigms: human development research, 32-40

257 process-person-context models, 35-40, 42, 43, 47; defining properties of, 36 process-person paradigms, 39 prosocial skills, 137 protective factors, 65, 85, 88, 93; mother as, 92 proximal influences, 6, 7-13, 32-3 psychoanalytic theory: sex typing in, 208 psychobiology, 184-5 public spaces, 155 punishment(s), 121, 132 rational expectations model, 20 rationalization, 187, 189 reactance behavior, 108 reality: self cognitions and, 183-4, 185, 18990, 191, 195-6, 198-9 reason: self cognitions and, 186 reinforcement, 60, 120, 123, 147; of destructive behavior, 211; negative, 121; positive, 136; in sex typing, 207 reinforcement trap, 121-3 relationships, 53; consensus/disagreement about settings in, 70; see also family relationships resilience, 51 resources: claims and, 106-7; prior, 81-2, 91, 93; in reaction to environmental change, 85-6 response-style variables, 212-13; person-centered theory of, 214 responsibility: in children-families-communities relationship, 65-6 risk factors, 85, 88 risk: perceptions of, 70 roles: age-graded, 97-9, 214; allocation, 80; family, 90; of fathers, 5, 90 schema theory, 208 school(s), 3, 10, 55-6, 59, 66, 72; and academic development, 9; age-graded, 106; as developmental context, 3, 7-8; and family/community factors, 8; introduction to, 52; interactional style in, 8-9; and student participation, 9 schooling: state and, 65 school performance: maternal employment and, 5 school reform, 105 selection of environments, 99, 210, 214, 228, 229, 231, 235-6; and individual development, 13-16 selection factors: in school effects, 7-8 selective attention, 228, 229, 232 self-concepts, 69, 70; functional analysis of

258 self-concepts (cont.) properties of, 181-2; functions of, 199-200; inflated, 190; sociogenesis of, 181-202; stability of, 14; therapeutic strategies in, 198— 9; unrealistic, 19, 186 self cognitions: adaptive, 199; developmental and evolutionary functions of, 181-2; enhancement of, 193-4; functions of, 183-6; multiple masters in, 189-91 self-direction, 13-14 self-esteem, 55, 213; leisure activities and, 175; low, 142, 143; rejection and, 131, 134; substance use and, 163-73 self-organization, 183 self-other ratings, 192-3, 197 self: public/private, 196-7; stability of, 200; variables involving, 213 (see also self cognitions) self-ratings: compensatory and self-balancing, 194-5; divergence in, 192-3; and ratings from peers and teachers, 181, 183, 191, 192-8, 199-200; and social consensus, 182-3, 189, 199; stability in, 193-4 self-regulation: of development, 153, 173-4 self-system, 213; construction of, 183 settings, 52; access notions in description/perception of, 58, 67-70; compatibility among, 8-9; community, 66-7; "continuity is good" value judgment, 55; developmental, 154-5; joint effects of, 9; moments of entry/exit, 66; nonopen, 57; qualities of individuals and, 71-3; as units, 50, 52-4; variance in, 58; "wider is better" value judgment, 54-5; "world is open" value judgment, 55-8; see also leisure settings sex typing, 207-8, 212 sexual dimorphism, 187-8 siblings, 10, 15; adopted, 223, 20-3, 235; biological, 230-3, 235; changing similarities among, 232-3; correlations among, 12-13, 230-3 single-parent families, 132, 146 situationism, 206, 207, 208 situations: and behavior, 209; as function of person, 209-10; idiographic classification of, 207; persons in reciprocal transaction with, 210-11 social address model, 28-9, 30, 31, 45-6, 155 social adjustment, 219-20; maternal employment and, 5 social change, 55; human develompent in, 3 6, 77-113; and individual development, 8 1 2; study of lives in, 78-9, 80 social class, 34, 39, 43, 84, 110; and cognitive functioning, 89-90; and effects of

Subject index Depression, 87; effects on development, 4, 6; and life experience, 90; of women, 6 social cognitions, 57, 188, 199; function of, 186; and social maps, 197-8 (see also self cognitions) social consensus, 185; and self ratings, 191, 195-6, 199 social development: measurement theory and, 182-3 social ecology, 187, 188-9 social institutions: effect on life course, 79, 109; as macrosystem, 39 social interactional approach, 118, 120 socialization, 5, 105; age and, 80; effect of socioeconomic status on, 34; in family, 9, 13, 114; family responses to economic hardship and, 94; parent-child, 229; perception of access and, 71; school, 7-8; setting selection in, 66; unstable family relations and, 101-2; work and, 6 social maps, 197-8 social network: and behavior, 16-17; impact on health, 164; and individual development, 11; and stress, 37 social policy: family in, 53; relationships in, 65-6; world views and, 55-6 social skills, 17, 117, 122, 127 social structure(s), 2; age-related change in, 188; of children and adolescents, 197-8; historical context and, 110; influence on development, 3-6; research on, 6 social support, 61 stage models: causal status in, 138-42; coercive process, 129-36; multiple indicators in, 137-8; problems in testing, 136-43 stealing, 126, 127-8, 129, 205 stress, 61; and behavior change over time, 116; economic conditions and, 60; effect on childhood, 81-9; effects on discipline, 132; and family dynamics, 93-6; parenting behavior and, social networks and, 37; vulnerability to, 51 substance use in adolescence, 176; constructive role of, 152, 173-4, 175; context of, 1546, 158, 160, 162; in mastering developmental tasks, 174-5; as problem behavior, 154; roles of, 153; and self-esteem, 163-73; stages in, 173 supervision of children, 93, 128 synchronistic model, 139, 141, 143 temperament, 212; difficult, 144-5, 242 temper tantrums, 98, 210, 211, 212; in trait progression, 126, 127, 128-9 theory construction: from observations, 21823

Subject index therapy: family, 10; regarding self concepts, 198-9; see also intervention(s) time/temporality: in delayed-incremental processes, 142; in effects of economic deprivation in childhood, 84, 88; historical/life, 80; in human development, 40-4, 77; in lifecourse perspective, 77, 79, 85, 100; of life events, 104-5 trait approach to personality, 204, 205, 206 trait progression, 125-8; integration of microsocial progressions with, 128-9 trait psychology, 208 traits, 212; progression in antisocial, 125-8, 129 trait similarity: in siblings, 233 transitions, to adulthood, 97, 103, 104, 152, 154; home-to-school, 35-6; life course, 91, 102-5, 107-8, 214; major, 109; into settings, 52 transmission: of behavior and attitudes, 91, 99-102; nature of, 61-4 transmission model, 5 0 - 1 , 61-4, 65, 66, 71 treatment(s): average effects of, 234, 235; vs. etiology, 241-2; phenotypes and, 238, 239; and within-group variability, 234, 235 turning points, 91, 102-5, 214; see also transitions

259 twins, 224; reared apart, 242; see also dizygotic (DZ) twins; monozygotic (MZ) twins twin studies, 12-13, 15-16, 234, 236, 242 typological approach, 206, 217-18 urban life-space, 154-5 value judgments: in children-family-community relations, 50, 54-8 values, 5, 110; and family management skills, 145-6; intergenerational agreement on, 623; in occupational attainment, 46; and self cognitions, 186; substance use and, 174 violence: family, 11 vulnerable child (the), 81 well-being, sense of, 143 withdrawn behavior, 210, 211, 212; and adult life course, 97-9; and selection of environment, 15 women, 83; see also mother(s) work: and development of conformity/autonomy, 4-5; in life course, 97, 98 world view(s), 55-60 youth gangs, 155