Web-based Learning: Technology And Pedagogy - Proceedings of the 4th International Conference

Web-b sedLearning: Technology and Pedagogy This page intentionally left blank editors Reggie Kwan Open University...

Author: International Conference on Web-based Le | Reggie Kwan | Joseph Fong

28 downloads 693 Views 11MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

Web-b sedLearning: Technology and Pedagogy

This page intentionally left blank

editors

Reggie Kwan

Open University of HongKing

Joseph Fong City University of Hong Kong

Proceedings of the 4th International Conference

Web-b sed Learning: Technology and Pedagogy 1 - 3 August 2005

Hong Kong

vp World Scientific N E W JERSEY * L O N D O N * SINGAPORE

BElJlNG * SHANGHAI

HONG KONG

-

TAIPEI

CHENNAI

Published by

World Scientific Publishing Co. Re. Ltd. 5 Toh Tuck Link, Singapore 596224

USA office: 27 Warren Street, Suite 401-402, Hackensack, NJ 07601 UKoffice: 57 Shelton Street, Covent Garden, London WC2H 9HE

British Library Cataloguing-in-PublicationData A catalogue record for this book is available from the British Library.

WEB-BASED LEARNING Technology and Pedagogy Proceedings of the 4th International Conference Copyright 0 2005 by World Scientific Publishing Co. Re. Ltd All rights reserved. This book, or parts there05 may not be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording or any information storage and retrieval system now known or to be invented, without written permission from the Publisher.

For photocopying of material in this volume, please pay acopying fee through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA. In this case permission to photocopy is not required from the publisher.

ISBN 98 1-256-430-6 (pbk)

Printed in Singaporeby Mainland Press

Conference Organizer W

Hong Kong Web Society

In Cooperation with W

City University of Hong Kong

W The Hong Kong Polytechnic University

Premier Sponsors W

ACM Hong Kong Chapter H BEENET / WebCT

W W W W

IEEE Hong Kong Chapter

K.C. Wong Education Foundation

Hong Kong Pei Hua Education Foundation Limited

Education and Manpower Bureau of Hong Kong Government

Exhibitors W H

BEENET

Thomson Publishing W

V

WebCT

I


Foreword Hong Kong Web Society has been organizing an annual international conference on Web-based Learning since 2001. The conference has attracted many academic scholars and industrial practitioners to submit papers and share discussions during their papers presentation in the conference. Overall, more than 100 papers are reviewed by the conference program committee in each conference. These papers cover the advancement technology of infrastructure as well as the learning behavior and the teaching effectiveness of elearning. Basically, we can summarize them into four areas as follows: eLearning for adaptive learning In order to speed up the learning curve of learners, we aim to supplementcLearning (class room learning) with eLearning (online learning through Internet). The students can do self learning from web-based learning exercises on the Internet, which will assess the students academic level and provide them with suitable online exercises to work on. As a result, the students can learn everywhere and any time through Internet. eLearning for teacher helpers In order to reduce teachers’ work load on authoring exercises for their students, on getting students feedbacks and on communicate with students more effectively, many eLearning systems facilitate these functionalities to the teachers as their helpers. For example, eLearning authoring tool can help teachers to prepare exercises, and blog journal file can help teachers getting students feedback on a particular learning problem or subject. eLearning for distance learning Besides traditional class room learning, students can also access learning facilities remotely. For example, a virtual cyber laboratory can help students to perform their laboratory exercises at home through Internet. An online tutorial session can help students access to their tutors through their notebook computers, As a result, students can learn in a very comfortably environment without traveling time, cost and hassle. eLearning technology infrastructure In order to make eLearning a success, technology infrastructure is a must. In fact, eLearning is more cost effective whenever it can be accessed by mass learners vii

viii

because individual cost will be less with increased number of learners for eLearning while the eLearning development time and effort is fixed. In other words, once an eLearning package has been on production, it can be reused for many times with minimum operation cost. However, such mass learners operations need the support of technology. As a result, research on eLearning technology infrastructure is another important area for researchers to explore. Finally, we appreciate the authors who have submitted so many papers to this conference and make it a success in the world of elearning.

Joseph Fong Co-chair of the ICWL 2005 Applied Track

ICWL 2005 Conference Committee Honorary Co-Chairs

Keith Chan, The Hong Kong Polytechnic University Frances Yao, City University of Hong Kong

Conference Co-Chairs

Kamal Karlapalem, IIIT (India) Qing Li, Hong Kong Web Society

Program Co-Chairs

Ronnie Cheung, The Hong Kong Polytechnic University Rynson W. H. Lau, City University of Hong Kong

Organization Co-Chairs

Hong-Va Leong, The Hong Kong Polytechnic University Marian Choy, City University of Hong Kong

Communications Coordinator Taku Komura, City University of Hong Kong Treasurer

C. W. Ngo, City University of Hong Kong

Publication Co-Chairs

Wenyin Liu, City University of Hong Kong Ming Cheung, City University of Hong Kong

Media Chair

Frederick Li, The Hong Kong Polytechnic University

Application Co-Chairs

Joseph Fong, City University of Hong Kong Reggie Kwan, The Open University of Hong Kong Kenneth Lau. ICON Limited

'htorial Chair

Howard Leung, City University of Hong Kong

Publicity Co-Chairs

L. F. Kwok, City University of Hong Kong Qun Jin, Waseda University, Japan

Local Arrangements

Edward Ho, The Hong Kong Polytechnic University ix


Welcome

Welcome to Hong Kong and ICWL 2005. We are thrilled to have ICWL back in Hong Kong after ICWL 2003 in Melbourne and ICWL 2004 in Beijing. For the out-of-towners, we wish you have fun in Hong Kong. Now, ICWL is in its fourth year. We are ecstatic to witness the advances of the field. In the past four years, different tools, models, and applications have been developed, deployed, and evaluated. We look forward to meeting with all of you to exchange ideas and to take web-based learning to new heights. We would like to take this opportunity to thank every single author for contributing to the field of Web-based learning. We are also grateful for all reviewers for their effort in reviewing the many papers. We also owe it to all the colleagues who took care of the logistics and made this book and ICWL 2005 a success. Finally, we would like to express our most heart-felt thanks to all the organizations that support this conference. We are grateful to the Hong Kong Web Society for hosting this conference; to BEENET and WebCT, Pui-Hua Education Foundation, K.C. Wong Foundation, City University of Hong Kong, the Polytechnic University of Hong Kong, and the Education and Manpower Bureau of Hong Kong Government for their sponsorshp.

Reggie Kwan Editor and ICWL2005 Applied Track Co-Chair

xi


International Program Committee Brian dAuriol

University of Texas, USA

Howard Beck

University of Florida, USA

Stephane Bressan

National University of Singapore, Singapore

Wentong Cai

Nanyang Technological University, Singapore

Jiannong Cao

The Hong Kong Polytechnic University, HK

Keith Chan


Shi-Kuo Chang

Chang University of Pittsburgh, USA

Arbee Chen

National Tsing Hua University, Taiwan

Marian Choy

City University of Hong Kong, HK

Jo Coldwell

Deakin University, Australia

Tharam Dillon

Sydney University of Technology, Australia

Guozhu Dong

Wright State University, USA

Ling Feng

University of Twente, Netherlands

Joseph Fong


Andrzej Goscinski

Deakin University, Australia

Edward Ho


Runhe Huang

Hosei University, Japan

Weijia Jia


Qun Jin

Waseda University, Japan

Ryoichi Komiya

Multimedia University, Malaysia

Taku Komura


Reggie Kwan

The Open University of Hong Kong, HK

L. F. Kwok


Dik Lee

HKUST, HK

Sue Legg

University of Florida, USA

H.V. Leong


Clement Leung

Victoria University of Technology, Australia

Howard Leung


Frederick Li

The Hong Kong Polytechnic University, HK xiii

xiv

Keqin Li

State University of New York at New Paltz, USA

Minglu Li

Shanghai Jiao Tong University, China

Wenyin Liu


Xiaofeng Meng

Renming University, China

John Murnane

Melbourne University, Australia

C. W. Ngo


Zhiyong Peng

Wuhan University, China

Geoff Romeo

Monash University, Australia

Timothy Shih

Tamkang University, Taiwan

Chengzheng Sun

Griffith University, Australia

Lily Sun

The University of Reading, UK

Boleslaw Szymanski

Rensselaer Polytechnic Institute, USA

Changjie Tang

Sichuan University, China

Guoren Wang

Northeastern University, China

Limin Xiang

Kyusan University, Japan

Simon Yip

The Chinese University of Hong Kong, HK

Kang Zhang

University of Texas at Dallas, USA

S. Q. Zheng

University of Texas at Dallas, USA

Contents Foreword

vii

ICWL 2005 Conference Committee

ix

Welcome

xi ...

International Program Committee

Xlll

PART ONE Tools 1.

2.

3.

4.

5.

6.

The InterTEST Web-Based Multiple-ChoiceTesting Software: Developing and Evaluating On-Line English Language Tests Philippos Pouyioutas, Victoria Kalogerou, and Maria Poveda

3

WebLec: A Full-scale Lesson Management Support System Developed by University Students Satoshi Uchida

13

On or Off the Slate: University Library E-Reserve Supports Web-based Learning Ophelia Cheung

23

Design and Implement of The Web-based Virtual Laboratory for On-Campuse Circuit Courses Gu Rong, Zhu Miaoliang, and Dong Yabo

29

A Mechanism for Knowledge Map Construction on Personalized E-Learning Platform: A Semantic Approach S.M. Huang, H. E Hsueh, and H. E Jiang

37

Yet Another Platform for Web-based Learning 7: S.Li, S. M. Wong, and Reggie Kwan

xv

47

xvi

PART TWO Models 7.

Blended Learning Approach: A Strategy to Address the Issue of Declining Enrollment in Mechanical Programs and A Promising Model in Teaching AutoCAD in Arabic R. Bedri and M. 0.Al-Nais

57

8.

Concept Maps and Learning Objects Leonel Iriarte Navarq Manuel Marco Such, Pedro Pernias Peco, and Daniel Moron Martin

67

9.

The Construction of Web-based Mastery Learning System Hsien Tang Lin, Zhi Feng Liu, and Shyan-Ming Yuan

75

10.

Enhancing Problem-based Learning by E-learning: A Study with the Teaching of Data Structures and Algorithms Kent K. T Cheung, Alan Z K. Chan, and Paul K. 0. Chow

83

How Should Online Tutors be Trained? A Four-level of Evaluation for E-moderating Programme Eva Bang

93

11.

12.

Methodology for Developing Dynamic Web Authoring System for E-Learning fin Fei Yeung and Joseph Fong

103

PART THREE Applications 13.

14.

15.

Creation of a Library of Learning Objects (LO) from Pre Existing Contents Leonel Iriarte Navarq Manuel Marco Such, Pedro Pernias Peco, and Daniel Moron Martin

117

A Mobile Agent Assisted Learning Resource Service Framework based on SOAP Wu Di, Yang Zongkai, and Cheng Wenqing

125

Teaching Stroke Order of Chinese Characters by Using Minimal Feedback Kerry Bang and Howard Leung

135

xvii

16.

An On-Line Programming Environment with Automated Assessment 145 Ronnie Cheung

17.

Webits and Quality Control in Marking of Examination Scripts Philip Tsang, Reggie Kwan, Andrew K. Lui, and Henry Lo

155

PART FOUR Human Factors

18. Social Capital Creation and Reciprocity in Online Learning

19.

Platforms Andrew K. Lui, Yannie Cheung, and Reggie Kwan

165

Are Teachers in Hong Kong Ready for e-Learning? Teddy Koon Keung So

175

20. Implementation Issues on the Specification for Service Quality Management of e-learning Yi Zhang, Zhiting Zhu, Chengling Zhao, Zongkai Yang, and Sanlan Lu

183

21. A Comparative Evaluation and Correlation Between Learning Styles and Academic Achievement on E-Learning Daniel Su

193

22. A web-based environment for better administration of distance learning courses S. C. Ng, S. 0. Choy, R. Kwan, and X C. Tsang

Author Index

203

213


PART ONE Tools


THE INTER TEST WEB-BASED MULTIPLE-CHOICE TESTING SOFTWARE: DEVLOPING AND EVALUATING ON-LINE ENGLISH LANGUAGE TESTS PHILIPPOS POUYIOUTAS’, VICTORIA KALOGEROU’ and MARIA POVEDA2

‘

Department of Computer Science, Intercollege 46 Makedonitissas Avenue, Nicosia 1700, CYPRUS pouyioutns.p~intercolle~e.nc.cv, kaloperou.v~inte~collepe.nc.cy Department of Computer Science, University of Cyprus 75 Kallipoleos Street, Nicosia 1678, CYPRUS [email protected] Abstract. In this paper, we explain the InterTest system, a WEB based educational software, developed at Intercollege, to meet the need of an on-line multiple-choice assessment tool. We also report our findings from its use in developing and using on-line the Intercollege English Placement Test. We compare the newly automated process of on-line testing with the manual process used so far from the point of view of the faculty/staff administering the process and from the point of view of the students. Our main conclusion emanating from this comparison is that the InterTest software is a very useful teaching/leaming/assessmenttool that can be utilized by faculty members to enhance the teaching and learning environment of the College.

1. Introduction InterTest [ 1,2] is a WEB based educational software developed at Intercollege, a College of Higher Education in Cyprus, to meet the need of an on-line multiplechoice assessment tool. The software has been integrated into the InterLearning [3,4,5,6,7] web-based educational software, which allows faculty members to develop and post on the WEB, educational material. InterTest allows the creation, delivery and assessment of multiple-choice tests and supports amongst others, on-line multiple-choice tests management, adaptive multiple-choice tests, extended reporting capabilities (summary and detailed statistics), possibility to use multimedia content (images, audio-files, animations, etc.) and listen-andanswer questions and finally, “session state” technique in order to be able to recover after a broken connection or a user workstation failure, or even after a server failure. InterTest has been pilot used during the beginning of the 2004-2005 academic year, in automating the process of placing our new students into their

4

level of English Language proficiency and hence into appropriate English language courses. During the previous academic years, new students were taking a paper-based multiple-choice English Placement Test (EPT) that was subsequently marked by the English Language lecturers. With the newly automated process, students are now taking on-line the EPT in the computer labs and their tests are marked automatically by the InterTest software. Statistics regarding the student results are instantly created by the system. In this paper, we explain the InterTest system and report our findings from its use in developing and using on-line the EPT. We also compare the newly automated process of on-line testing with the manual process used so far and address their advantages and disadvantages both from the point of view of the facultyhtaff administering the process as well as from the point of view of the students. In Section 2, we introduce the InterTest software. In Section 3, we explain the Intercollege English Placement Test (EPT) and show how InterTest was used to implement the new on-line version of the test, namely the NEPTON test. In Section 4 we present the results of our survey amongst students and faculty/staff with regards to NEPTON. Finally, we present our conclusions and address our current and future work. 2.

The InterTest Software

InterTest is a web-based multiple-choice assessment tool that allows the creation, delivery and assessment of multiple-choice tests. These tests can either be generated as paper-based tests or tests that are available on the web and can be taken on-line. InterTest supports amongst others the following: on-line multiple-choice tests management 0 adaptive multiple-choice tests pen-and-paper tests (the system generates paper-based tests and answersheets which can be scanned and be processed by he system) 0 sophisticated test-generation - assessment algorithm (questions can be divided by levels and assessment can be programmed considering the levels and types) 0 extended reporting capabilities: summary - details - question 0 possibility to use multimedia content (images, audio-files, animations, etc.), for example: listen-and-answer questions 0 “session state” technique in order to be able to recover after a broken connection or user workstation failure, or even after server failure. The InterTest interface showing the TestIQuestions Editor is given in Figure 1. The main functions of the tool as shown in the top horizontal menu are: 0 Item Editor for generating the questions and the tests

5

Slide Manager for programming the behavior of the system for adaptive tests, allowing three levels of difficulty 0 User Manager for managing administration privileges of authorized users 0 Student Manager used if InterTest is to work as a stand alone application 0 System Setup to set up of levels, question types, language, etc. The Questions Editor (QE) shown in Figure 1 allows the creation of on-line and paper-based tests. For the on-line version formatting is done according to HTML formatting rules. The QE includes a built-in web based HTML editor, which provides basic fimctionality, including scripting in HTML codes. Figure 1 is part of a screen shot of the QE showing how an on-line question is created using InterTest. We show how a test and the questions are presented to the user through the Questionviewer (Figure 2). The Questionviewer interprets the script of each question and dynamically generates interactive components: radio buttons, check boxes, drop-down lists, etc. retrieving data from the question bank. Each question can contain more than one item as well. Responses can be programmed by providing answers and weights. To validate the answer weights are used. If one choice only is allowed then the correct answer must have a weight of 1 and all other answers a weight of 0. If more than one choice are allowed SUM (e.g. for questions like “Choose all correct answers”), the system will consider the Sum of all the weights. 0

3.

TheNEPTON

English is the language of instruction at Intercollege. Due to the wide range of our students in relation to their English skills, we offer many classes to assist them to acquire the required level of understanding in the target language. For this reason we offer 4 language classes starting from beginners going up to higher intermediate level, namely BENG-50, BENG-80, BENG-90, BENG-100 and 2 classes of academic English, namely ENGL-100 and ENGL-101. When students register at Intercollege, they are required to take a placement test in English in order to be classified in the relevant English Course. For this reason, a paper-based English Placement Test (EPT) was created to facilitate the process of the placement of the students. This test consists of two components, one including a set of 64 multiple choice questions in order to check the knowledge of the grammar of English, and the second asking students to produce a 200 words essay in order to check the competence in completing a written task and evaluate their expression. The distribution of the students in the appropriate course is done after the evaluation of the two components by the lecturers of the English Department. This method, despite its success for many years, has demonstrated certain flaws. First of all it was an additional responsibility on the

6

Responses

Fig. 1. The InterTest Interface - The Questions Editor - Answers and Weights

lecturers. The weekly schedule designed to assign every day’s EPTs to a group of 3 people worked but during particular points of the academic year (beginning of the Fall or the Spring Semester, or during the summer) when lecturers had already their weekly teaching loads, it was particularly difficult to finish marking the placement tests. During busy times, lecturers had to spend a day in order to complete the marking so that the results would be ready on the following day for the students to be distributed normally in classes and to be allowed to take other courses according to their academic paths. The increasing number of the students, the heavy workloads of the faculty along with the need for improvement demanded a modification of the existing system. So an on-going research started in order to create an archive of the multiple-choice questions relating to the different levels, to create the database and the software in order to make this task realistic and ready. This has resulted in the development of the NEPTON test, using the InterTest software. NEPTON allows the creation of various types of multiple-choice questions and tests, online adaptive testing, management of administrative users, management of students and detailed and summary reporting of results. NEPTON was put to use for the first time during Summer 2004 in order to place students for the Fall 2004 semester into their English classes. Some adjustments relating to

7

organization, timetable arrangements, staff distribution were needed for the new test to run smoothly. These adjustments were concerned with the availability of labs and lab assistants, training of staff and the variant computer skills of students. Students are presented the test using the interface shown in Figure 2. They can move from question to question either using the navigation bar on the top of the screen (showing the numbers of the questions available on the test) or the (>, Next) button at the bottom of the screen, where they can also see the remaining time for the test. Finally, they can review questions and change answers with the system logging all actions. Before starting the test, an on-line how-to-use tutorial is given to the students.

,

Choose

an answer

I f you can use the computer, you can do

............. things

0 lot5 C many r much C plenty C too

Fig. 2 The NEPTON Test - Sample Question -Using Multiple-choice Answers

NEPTON maintains very useful statistics in detailed and summary form. The details include for each student personal information, test date, results level (essay, multiple-choice test, combined result and Final decision), time started, time finished and duration of test. The summary show for each level (BENG-50, etc) the number and percentage of students assigned to it. The statistics can be sorted out by datehime of test.

8

4. The NEPTON Test - Faculty/Staff and Student Evaluation The NEPTON test was in general enthusiastically accepted on behalf of the students. After obtaining a date for the particular test they could take it and their results were ready on the same or the following day. The test being adaptive, required only the necessary time on behalf of the students, which meant that students having little knowledge of the questions didn’t have to spend the time to attempt and answer all of them. Their scores were printed and then the invigilators attached their written document to be marked by the language instructors. The English Language faculty members were also released from the tedious job of marking hundreds of paper based multiple-choice tests. Changing the existing system inevitably caused a number of limitations/problems due to the different nature of the system. To begin with, the test has to take place in the computer labs; this restricts the number of students to the number of the PCs available (our computer labs have a maximum sitting capacity of 20) and the test is subject to possible hardware failure/problems. We actually faced such a problem in one of our tests when a server failure occurred. The students had to wait half an hour to continue their test after system recovery. Furthermore, the test invigilation now requires at least two people in each room; one person familiar with the NEPTON test to help students with specific test questions and a lab assistant to deal with specific hardware/sofhvare problems. This, is combination with the restricted size of the computer labs, has increased the human resources needed for invigilation. Finally, the test invigilators need now better training in order to deal with the on-line test. We conducted a survey amongst the students to get feedback for NEPTON. Figure 3 below (Y -> Yes, U -> Unsure, N -> No) shows the student responses to a questionnaire answered by around 100 new students. The students were generally happy with NEPTON. As the results show, only 5% of the students were not familiar with computers and therefore the majority did not have any problems in using computers and in answering the questions. Furthermore, the answers show that students in general, believe in multiple-choice testing and online testing, although there seems to be some questioning as to the effectiveness of NEPTON in placing them in the correct level of English.

9

Fig. 3. The Results of the Student Questionnaire

We also conducted a survey amongst the English Language faculty to get feedback from their own perspective. The table below shows the faculty members’ responses to the questionnaire distributed to them. The percentages do not add to 100 due to the fact that some questions were not answered. The main conclusions from above are that the English Language faculty believe: 1. that the NEPTON process is an improvement to the manual process used so far, especially when it comes to the efficiency and effort needed for producing student results 2. NEPTON may prevent more students from cheating 3. the essay part of the process cannot be eliminated 4. multiple-choice testing seems appropriate for their classes 5. IntertTest should be considered for developing on-line testing.

10

choice testing for your classes? Fig. 4. The Results of the Faculty Questionnaire

Conclusion In this paper we have presented the InterTest web-based multiple-choice testing software and addressed its use in automating the English Placement Test at our College. This resulted in the development of NEPTON. We have also presented the student and faculty members beliefs with regards to NEPTON specifically, and on-line testing, in general. Our main conclusion in this paper is that the NEPTON process is an improvement to the manual process used so far and that the InterTest software is a very useful teaching/learning/assessment tool that can be utilized by faculty members at Intercollege to enhance the teaching and learning environment of the College.

11

References 1. Pouyioutas, P, Kalogerou, V & Christou, C (2005a) Using the InterTest Web-Based Educational Software in Automating the English Placement Test at Intercollege, International Conference on Methods and Technologiesfor Learning, Italy, 2005,369-375. 2. Pouyioutas, P., Poveda, M., Apraxin, D. & Kalogerou, V. (2005b) The InterTest Multiple-Choice Web-Based Software, IASTED International Conference on Web-Based Education, Switzerland, 2005,436-441. 3. Pouyioutas, P., Apraxin, D., Ktoridou, D. & Poveda, M. (2004a) The InterLearning Web-Based Educational Software, Proc. IASTED International Conference on Web-Based Education, Austria, 595-600. 4. Pouyioutas, P., Poveda, M., Apraxin, D. & Kalogerou, V. (2004b) InterLearning - A Comparison Analysis of Students’ Perspectives, Proc. International Conference on Applied Computing, Portugal, Vol. 2,70-74. 5. Pouyioutas, P., Poveda, M. & Apraxin, D. (2004~)Students’ Evaluation of the InterLearning Software, I dh World Conference on Educational Multimedia, Hypermedia and Telecommunications, Switzerland, 22642269. 6 . Pouyioutas, P., Poveda, M. & Apraxin, D. (2004d) Design, Development and Evaluation of Web-Based Educational Software, International Conference on e-Society, Spain, 267-275. 7. Pouyioutas, P., Poveda, M. & Apraxin, D. (2004e) The Impact of WebBased Educational Software: Off-the-shelf vs. In-House Developed Software, International Journal of the Information Technology Impact, 3(3), 121-130. 8. Questionmark 2004. httu://www.questionmark.com/ 9. WebCT Software (2004) httu://www.webct.com/


WEBLEC: A FULL-SCALE LESSON MANAGEMENT SUPPORT SYSTEM DEVELOPED BY UNIVERSITY STUDENTS SATOSHI UCHIDA Department of fndustrialEngineering and Management, Faculty of Enginem’ng, KANAGA WA University E-mail: [email protected] UFU : http://www.inf.ie.kanagawa-u.ac.jp/ Abstract: This paper introduces a full-scale lesson management system called WebLec that is currently used by several universities. Although there are numerous similar systems, WebLec includes multiple features. WebLec has been continuously updated and implemented over three years by the educational staff of a university. In order to raise productivity, a new database processing technique called an Object-based Relational Database was integrated into the system. WebLec has been used since 2002, and in the first half of 2004 the system boasted 3,400 student registrants at Kanagawa University and Shizuoka University, among others. The development techniques and usage in several universities is discussed in detail.

Key Words: Lesson Management System, Web application, Object based RDB 1. Introduction WebLec is an Internet-based system that provides class attendance management, examination management, report submission, questionnaires, and a web page creation function. WebLec was also developed for the purpose of reducing a teacher’s office workload and preventing human mistakes in a lecture presentation. Although there are a number of similar systems currently on the instruction market, WebLec boasts the following features. (1) Even though university students developed the majority of the system, WebLec is a professional product and can be practically applied. WebLec 3.0 was used by several universities including Kanagawa University. As of January 2004, about 3,400 students and 25 teachers in 50 classes had used the system. (2) The opinions of both teachers and students were considered during the development of WebLec. Because the development staff were also WebLec users, the system is essentially the ultimate in end-user-computing. WebLec is continuously updated, with a new version released every year. WebLec 4.0 is the latest version, which was 13

14

released in January 2005. (3) The development costs of WebLec are minimal because it is developed primarily by university students as a graduate research project. In addition, the annual updates are also created at a minimum cost. This continuous development process in necessary to ensure a top-of-the-line product. (4) The project is implemented using Java Servlet and JSP (Java Server Pages) technology developed by Sun Microsystems, Inc. The database was created using PostgreSQL. In order to enhance the programming power and flexibility of the system, the author designed a mechanism for the object database system in postgreSQL and implemented an automated code generator into the system. This technology allows university students to implement WebLec at a practical usage level. 2. Outline of WebLec WebLec contains three subsystems: a university information management subsystem, a teacher support subsystem, and a student support subsystem. 2.1 University information management subsystem A WebLec manager registers university information such as the university name, faculty names, department names, etc. in this subsystem. Each university has its own regulations, including its university calendar and holiday schedule, university time-table, and seat information for each classroom. This type of information is stored and managed within WebLec.

2.2 Teacher support subsystem In this subsystem, a teacher registers his or her specific information such as the teacher name, e-mail address, etc. All information stored in the database is converted to an encrypted message for safety reasons. A teacher registered in WebLec can log into the system and register class information such as the course schedule, course registration information, room information, and set various permissions for a course. The teacher can create a web page for a course, generate examination questions, administer tests, and assign reports using WebLec. The teacher also has the ability to administer a questionnaire.

2.3 Student support subsystem In this subsystem, students register their own information, including student IDS, names, e-mail addresses. All information stored in the database is then converted to an encrypted message for safety reasons. Class registration is

15

performed over the Internet using WebLec. Teachers are capable of maintaining an attendance register for each student. Therefore, students must confirm their attendance information in the system. Students are also able to take examinations, submit reports, and answer questionnaires over the Internet. JABEE, the Japan Accreditation Board for Engineering Education, is an accreditation system for engineering education in Japan that was established on November 19, 1999. JABEE requires a minimum of 1,800 contact hours between teachers and students. Thus, taking attendance is essential for ensuring the required contact time. 3. History of Weblec WebLec 1.0, which was a basic prototype system, was implemented in May 2002. In September 2002, WebLec 2.0 was implemented and an experimental system was deployed at Kanagawa University, Yokohama campus. In December 2002, WebLec 2.1 was implemented and actually used by the university. WebLec 2.1 was available to the Kanagawa University teachers, and many teachers adopted this system. However, as the number of students registered in WebLec increased, the performance of the system decreased. Therefore, the system and its database processing system were redesigned to solve this problem. WebLec 3.0, which employed these improvements, was released in October 2003. In December 2003, the actual WebLec 3.1 system had been implemented. After WebLec 3.1 had been used for over a year, WebLec 4.0 was implemented in January 2005. 4. WebLec functions WebLec contains the following functions: attendance management, examination management, questionnaire management, web page creation, and customization.

4.1 Attendance management Teachers can define periods for attendance and lateness for their classes via the Internet. Students must log into the WebLec system in order to record their attendance at the beginning of the class. WebLec is capable of outputting the seating list with student names and their attendance status, as shown Fig. 2. A teacher can then use this list to identify the location of a student. In classrooms without computers, attendance can be taken by distributing an attendance card and a lateness card, which are output by WebLec for every student. The cards have unique identification numbers and a student’s

16

attendance record can be retrieved by inputting the identification number into WebLec after a lecture. The List o f Attendance

Fig. 1 The list of attendance for each class Based on this information, a teacher can readily obtain an attendance list for each class in real time, and students with poor attendance records can be identified. 4.2 Report submission management A student can submit a report using WebLec by uploading the submission file via the Internet. Teachers can specify the submission period, and after accepting the reports, can grade and comment on the report via the Internet. Students can then read the teacher’s comments using an Internet interface. The system also generates the list of students who have submitted reports.

4.4 Questionnaire management Teachers can administer questionnaires to a specific class or to all students registered with WebLec. In addition, WebLec 4.0allows for the questioning of students who are not currently registered. The questionnaire management system also includes an analysis function, including factor analysis and CS analysis. The analysis is performed automatically by simply clicking a button. 4.3 Examination management Teachers can prepare multiple-choice test questions and administer examinations in a specified period and classroom using WebLec. The examination is carried out via the Internet, and the teacher can obtain a variety of statistical information on the test results.

17

Unfortunately, there are a lot of potential problems when testing over the Internet, and it is easy for a student to cheat on an examination compared to a paper-based examination. In order to avoid this situation, the order of questions is rearranged for each student in WebLec 4.0.This version also includes an analysis function for an examination. A teacher can use this function to improve the contents of an examination.

Kanasara Unircrsny Vohohama Campus Seal l i d bury 18. 2006 T h u r d v Period 1 Computer Aided DericdTerhor: Salmrhi Uehida) Room: 21-111

4.5 Web page creation A web page can be designed for a specific class, and files can be made available for upload to students via the web page. The web page can also include pictures. Teachers can also create a web page with the system even if they don't know html coding. 4.6 Customization WebLec provides the ability to customize a variety of information specific to the university, including campus information, faculty and department information, and timetable and classroom information. 5. implementation of WebLec All members of the project group are university students, with the exception of the author. Therefore, careful instruction is essential. This issue will be discussed in detail in Chapter 6 . This chapter will discuss the implementation technique and project structure, which is important because numerous members of the development staff are replaced every year. Therefore, the development environment must be consolidated. WebLec cannot be

18

implemented without using techniques that reduce the programming difficulties. Because communication with the students is very important to product development, a communication system was created. 5.1 Object-based Relational Database and its code generator

Consider an examination, for example. An exam contains various questions, and each question could be a single-choice type, a multiple-choice type, or a descriptive type. Using this method, the system's data structure is too complex for a university student to implement. So, instead, an Object-bused Relational Database was designed to solve the problem. This technique can directly store a data object to an SQL database. Fig. 3 shows the flow of the Object-based Relational Database. When the serialize function of the Java programming language is employed, this process is not entirely difficult to implement. However, because the serialization and deserialization program is slightly complex, a code generator for this system was designed and integrated. Using the code generator, the programmer only writes the class design specification. Then, using the specification, the code generator generates Java source code that defines the class, and then stores this information in a relational database and later searches from this relational database. 5.2 Communication system with students As the development of the system progressed, communication between the teacher and students became increasingly important. In order to communicate as quickly as possible, Sh@y Unit, which is a groupware system for a teacher and students, was developed by the team's students. Using Shifty Unit, a message can be sent to the forty-student development team, regardless of their locations. The message is sent to a cellular phone as an e-mail and the person who receives the message can answer the message directly using his or her cellular phone. This is very powerful tool for timely communication.

19

hexadecimal conversion

I UserID

123 User name “uchida” Password “pass12345“ Nam “Satoshi Uchida”

I

I

Class object

1store Relational Database

4 “

Search

AB23F45C24085DAC

&data

I

I

UserID 123 User name “uchida” Password “pass12345”

Class object Fig.. 3 Object Serialization

6. Instructing students using an e-learning system WebLec has been developed and improved by five students every year since 2002. The WebLec program exceeds 100,000 lines in Java and JSP. Therefore, each student has to write more than 20,000 lines of program code. However, most students have not had any experience writing practical software, so instructing the students is very important. A Java e-Learning system was developed to create an environment for studying Java programming technology. The system includes a video lecture on a basic introduction to Java programming that is available to the public on the Internet as a video stream. The lecture is based on a primer for the Java programming language. The video lecture consists of about 38 hours video and a PowerPoint presentation containing about 5,000 slides. An examination can also be taken upon registration. The test contains approximately 850 questions on Java programming at four levels,

20

including super-beginner,beginner, intermediate, and advanced. No restrictions have been applied to viewing or listening to the Java programming video lecture. It is rare in Japan for a video lecture for an entire book to be made available to the public over the Internet. The contents of the video lecture were arranged in detail before the video was captured with students to ensure that people who want to study can learn efficiently. The video lecture contains short five to eight minute video clips to facilitate study. The video lecture includes Chapter 13 of "An exhaustive introduction to Java programming - basic course -" by Damp Publications. 33mm*7T-flbea3

Fig. 2 Screen shot of Java e-Learning system for Java programming

7. Conclusion In recent years, the software development environment has become remarkably powerful. This means that end user computing can become even more realistic. This paper describes an attempt to develop a web-based system by university students as a graduate research project. Due to the low maintenance costs, advice from teachers and students can be continuously integrated into updated WebLec versions. The developed system has also proved a powerful tool for the developing team students. Various video lectures on the Java e-Learning System, CASL e-learning System, C e-Learning System, etc, are available. A combination of video lectures and examinations is an effective means for instructing many students at a time. More than 40 students work on the development team, and it would be nearly impossible to adequately instruct them without employing this e-Learning system. 8. Future work Plans are currently in progress to combine the WebLec and e-Learning systems. This will enable the teacher to understand the student study

21

situation in greater detail. In order for WebLec to be made available to a larger market, the commercialization of WebLec is currently scheduled by Educe Co., Ltd.

References [l] Satoshi Uchida, “The Development of the web-based system for supporting lecture,” the conference of information technology for university instructions, Japan University Association for Computer Education (September, 2002) (in Japanese) [2] Satoshi Uchida, Naokazu Yamaki, “WebLec: A Full-scale Lesson Management Support System And It’s Usage In Several Universities,” Journal of Japan e-Learning Association, Vol. 7 (2004) (in Japanese) [3] Satoshi Uchida, “An exhaustive introduction to Java programming basic course -” by Dempa Publications. (July, 2003) (in Japanese)


ON OR OFF THE SLATE: UNIVERSITY LIBRARY E-RESERVE SUPPORTS WEB-BASED LEARNING OPHELIA CHEUNG Ryerson University Library Toronto, Ontario, Canada Ryerson University Library in Toronto, Canada, embarked on an E-Reserve pilot project in 2004, providing electronic links to journal articles and scanned documents for course readings. Insufficient time allowed by faculty for copyright clearance, excessive copyright cost and the difficulty of achieving a single sign-on by users to access eReserve in Blackboard, the University’s courseware, were the major issues. Despite setbacks, E-Reserve continued to undertake innovative projects, using the new E-Reserve module in Blackboard to integrate course readings with faculty course pages, utilizing SFX open linking to connect citations to full-text articles, and experimenting with Web streaming of videos in Blackboard.

1. E-Reserve in University Libraries

The Internet has increased university faculty and students’ access to a myriad of resources outside the walls of the library. Traditional Library services, including Print Reserve, are not flexible enough to meet the needs of the current group of faculty and students, particularly with the increase in “distributed learning” in universities. Like many other libraries, Ryerson Library has re-thought and redefined its services in relationship to the educational needs of its users. An electronic Reserve system, providing a virtual repository for copyright-cleared digitized course readings, available remotely 2417, was a recent step to keep up with student and faculty demand. Ryerson University, in Toronto, Ontario, Canada, is a leading centre of career-focused education, offering more than 40 undergraduate and graduate programs, with over 20,000 FTE (hll-time equivalent students), and with an ever-growing Distance Education (DE) department. E-Reserve service is especially advantageous for DE students and Web-delivered courses. Users do not have to come on campus during Library opening hours to borrow short-term loan materials from Print Reserve, and worry about fines if they cannot return by the due date or time allowed. Print Reserve keeps only a limited number of copies for borrowing but E-Reserve offers multiple access and simultaneous use. 23

24

E-Reserve has been in existence since the early 199Os, and an estimated 250 E-Reserve systems were in production in the United States in 1999.' In Canada, and in the province of Ontario, the libraries we surveyed in 200213 were mostly in their third or fifth years of E-Reserve, but some did not go beyond linking to full-text articles from licensed databases, due to the extra cost and work of obtaining copyright permission for scanning documents. Ryerson Library began e-linking in 2003 at faculty requests. New personnel and a new administrative team structure allowed us to embark on a more ambitious EReserve pilot project in 2004: In 2004, we became more proactive, checking all reserve requests against our Library's subscribed databases and automatically linking the citations to full-text articles wherever available. No duplicate backups in paper format were kept in Print Reserve. We also, for the first time, sought copyright permission for materials for which we did not have digital rights (e.g. book chapters), before scanning them on E-Reserve. When I took on the E-Reserve responsibility in late 2003, I conducted an extensive research on E-Reserve operations in North America, including best practices and case ~ t u d i e s .I~ was assisted by the Copyright Lead Hand Technician, a newly created staff position for E-Reserve. We learned from others that copyright and systems support were major concerns. The time for copyright clearance ranged from weeks to months. Rights holders could be difficult to locate. There were long waiting periods for responses, and occasionally no responses. The average cost was Can $30 to $90 per item per course per year. However, some publishers charged a few hundred dollars per item. Who paid the cost - the Department, Library or students, was a major decision. The choice of an E-Reserve system was another issue. Ideally, the system should be user-friendly for students, flexible enough for the Library to restrict access to a course or group, if required by rights holders, and robust enough for tracking statistics. The lessons learned from other E-Reserve operations helped us steer away from anticipated roadblocks but we were not immune to new challenges, some of which forced us to re-assess the direction of the pilot project.

2. Lessons Learned 2.1. Copyright Issues

Although we emphasized on our Web site that we discouraged digital course packs due to the cost and time for copyright clearance, the very first request from DE, our targeted user group for the pilot project, was a course pack of 29 readings, mostly book chapters. The copyright cost would amount to approximately Can $2,500 to digitize the chapters for use in three semesters.

25

Eventually, DE proposed seeking administrative and financial approval in the following year for a special student levy to defray such cost. We had better luck with subsequent requests. We only turned down a few faculty requests because the copyright fees were too expensive for our small E-Reserve budget from Library funds. Almost 50% of the faculty E-Reserve requests were not completed. One major reason was faculty had not allowed sufficient lead-time for seeking copyright permissions. Some were mistaken in thinking that materials used in an education setting were “exempted” from copyright. Some thought that the permissions obtained by the University Bookstore to compile print course packs for sale to students were transferable to digital course delivery. For those who attempted to create their own links to library databases in their Web course pages, they were not aware that the URL (Uniform Resource Locator) of web sites might break because they were not created to be “persistent” (also known as stable, durable, constant or static URL). What we could have done better was to step up our education of faculty users. We should have made sure they understood fully the copyright implications -the time and cost involved, and the alternatives available ( e g Print Reserve or print course packs). Our Web page and FAQ may not be effective enough. We should tap the resources of faculty listserv, faculty conferences, new faculty orientations, workshops by librarians and promotion through subject librarians. Another complication was created by our collaborative programs with other academic institutions. Due to database licensing agreements, access is limited to users in our university community. Some faculty would insist on posting materials only for Ryerson students and asked other students in the collaborative program to access the databases directly, if available at their home institutions. Others abandoned the E-Reserve access altogether, which was unfortunate. While we cannot do much with licensing agreements, we might be able to warn faculty in advance of these complications so that they consider alternatives, such as resorting to course packs. 2.2. System Issues

We had looked at how other libraries chose their E-Reserve systems! In the end, we decided on utilizing the Millenium Media Management module in our integrated library system to manage E-Reserve files. The module allows libraries to share media files with its patrons (e.g. WORD, PDF, TXT, Powerpoint), sounds, movies, HTML files, URLs and images (e.g. jpeg, gif, tiff). Keeping E- Reserve part of the integrated library system has the advantage of easier searching and transferring of data within the system, and staff training. However, we had to compromise our local needs. The publisher codes in Media

26

Management have the function of tracking E-Reserve usage for the purpose of paying royalties, if required. Nevertheless, to turn that code on, students on campus will have to authenticate once before accessing library resources, and twice if they log-on from off-campus through the proxy server. The code was never used for the obvious reason of inconvenience to users, and we never obtained usage statistics for E-Reserve access. The Library authentication presented another obstacle to a single sign-on for Blackboard users. E-Reserve was initially offered only via the Library catalogue. Students could look up course readings by course codes or names of instructors. However, DE courses were mostly delivered on Blackboard, a courseware for faculty and students to deliver lecture notes, complete assignments, conduct virtual discussions and share files and resources. DE had repeatedly enquired why students accessing from home had to go through a library authentication after they had logged-on to Blackboard with their passwords. The same passwords were used for e-mail and lab accounts but not for library authentication. The Systems Librarian in the Borrower Services Team (of which E-Reserve is part) explained that single-sign-on is a complex issue. Innovative Interfaces Inc. (111), our integrated library system vendor, has made some progress, such as introducing a new External Patron Verification package using LDAP (Lightweight Directory Access Protocol) in the new Silver release, which can be used to validate library patrons against an external LDAP server. However, it cannot be used to authenticate in the reverse mode i.e. Blackboard patrons could not authenticate against the Library’s patron database. A workaround created by some libraries might involve the creation of PHP scripts to link the student information system with a portal server, a web server, and the Library server. Custom CPIP connections would be required and the student information system might have to be compatible with either Oracle or LDAP. Without significant support in programming and funding, it seems that we might have to wait until 111, our systems vendor, develops its own solution to link to the campus portal. Meanwhile, faculty told us that students accessing from home had to log-on several times to move between E-Reserve files. These students might have closed their browsers or exited Blackboard unnecessarily, or used some older versions of Netscape, not compatible with Blackboard. What we could have done was to advise faculty in advance to link their Web pages to a Troubleshooting Guide in the Library Web Page for tips on remote access from home.

27

3. New Initiatives to Improve E-Reserve Despite the above setbacks, we have made progress in E-Reserve. 33 documents for 6 different courses were scanned in 2004 after clearing copyright, and 18 documents scanned in the first three months of 2005. Links to licensed databases have gone up by 50%, from 316 links in 2003 to 474 links in 2004. The Nursing Department, in particular, has switched from using print course packs to E-Reserve, as they rely, almost exclusively, upon our licensed databases. In early 2005, less than a year since we started our pilot project, we undertook other pilot projects to anticipate demands and improve E-Reserve service. In January, we partnered with DMP (Digital Media Projects Office, the University department providing assistance in the use and production of multimedia technologies for e-teaching and e-learning), to test E-Reserve in the new Blackboard release. Previously, for Blackboard faculty users, we e-mailed them the durable links for incorporation in their course pages. In the new Blackboard System, the Library will create links and upload reserve files into Blackboard. DMP will still be the system administrator and provide training to faculty wanting to create course pages to access E-Reserve from within Blackboard. Integrating E-Reserve with Blackboard is undoubtedly convenient for student users of courses delivered in Blackboard. The controlled access in Blackboard also provides more efficient copyright control, especially if the rights holders require restricted access to scanned documents. Some workflow issues remained. As long as there are courses that are not Blackboard-based, the Library has to maintain library catalogue access to E-Reserve files for nonBlackboard users. In April 2005, we tested another course using the new Blackboard release and, in addition, employing SFX links, which are vendor-independent, to connect citations to hll-text articles. SFX is the link server from Ex Libris. It uses the OpenURL standard for interoperability between information resources and allows for localization in an open linking environment. Apart from the relative ease of using OpenURL Connector for locating the databases that contain the journal citations and generating &Reserve links, we hope that the central maintenance of SFX links by our Collections Team in the Library will reduce the incidences of broken links, which leave students with only a citation or an abstract when they want to see a full-text article. These were often caused by changes in publishers’ activities (e.g. mergers and acquisitions) or journal content in the aggregator databases. We used to make a direct link to one single database, but SFX offers a menu of multiple database options, so users will have more choice in the event of a broken link. The SFX screen will also provide a Library branding presence in Blackboard. We will not lose the Library identity

28

entirely in the campus portal. We are not the first in integrating E-Reserve with Blackboard but we may be a pioneer in the application of SFX links to EReserve. In April 2005, E-Reserve also worked with DMP to digitize videos for streaming in Blackboard. The Audio Visual (AV) Librarian in our Borrower Services Team is exploring a site license agreement with a major Canadian distributor of educational videos for streamed content and the acquisition of the digital rights for a streamed Nursing video series. If successful, the Webstreaming solution in Blackboard will become a breakthrough in the delivery of AV content, solving the chronic issue of faculty and students competing for access, on-site and off-campus. 4. Conclusion

It has been a year since our E-Reserve project was officially launched. The lessons we learned could be useful for other universities planning to start or improve their E-Reserve operations. In the face of changes, we maintain that the right direction to deal with them is to embrace them. Our team structure has anticipated the need for amalgamation of expertise to meet the challenges ahead - systems, copyright, multimedia, publicity and collaboration with diverse user groups within the University. We are well positioned to cope with these issues, central to the successful delivery of E-Reserve in its existing or a revised mode in the future.

References 1. LU, S., 2001. A model for choosing an electronic reserve system: a pre-

implementation study at the Library of Long Island University’s Brooklyn campus. Journal of Interlibrary Loan, Document Delivery & Information Supply, 12(2): 25-44. 2. CHEUNG, 0. and PATRICK, S., 2004. The gang’s all here: team-work and collaboration facilitate E-Reserve for e-learning through the university library. In: E-Learning in Corporate, Government, Healthcare, & Higher Education Conference Proceedings, pp. 539-542. [E-Learn World Conference held in Washington, D.C., United States, November, 2004.1 3. CODY, S.A., PFOHL, D. and BITTNER, S . , 2001. Establishing and refining electronic course reserves: a case study of a continuous process. Journal of Interlibrary Loan, Document Delivery & Information Supply, 1l(3): 11-37. 4. PEARCE, L., 200 1. Lessons learned: the development of electronic reserves at the University of Calgary. D-Lib Magazine, 7( 11). Online. Available: http://www.dlib.org/dIib/novemberO1/pearce/l 1pearce.html/.

DESIGN AND IMPLEMENT OF THE WEB-BASED VIRTUAL LABORATORY FOR ON-CAMPUSE CIRCUIT COURSES* RONG Gut Zhejiang University of Technologv, No. 6 District Zhaohui Hangzhou, Zhejiang 310014, China YABO DONG, MIAOLIANG ZHU Networking Center, Zhejiang University Hangzhou, Zhejiang 310027, China In this paper, some general considerations of virtual laboratory (VL) are reviewed. The paper is to describe a web-based virtual laboratory for undergraduate circuit course, which has been used in Zhejiang University and EastSouth University. The various planning, design issues in this project will also be discussed. One of the major objectives of this project is the client adopted flash client to enhance the students' learning experience and provides significant impact on users' learning motivation. With the support of this virtual laboratory, the students' uninfluenced learning and experiment process can be carried out at any place of the world where Web access is available.

1. Introduction Nowadays, the development and implement of Web-based education bloomed. As the major application of Web-based learning, Virtual Laboratory (VL) can be treated as an accessorial tool of real laboratory. Real laboratory and experiments can reinforce the students' understanding of subject and developing skills to deal with instrumentation and physical processes, while face-to-face instruction help student to master concept. There is no doubt that nothing will replace synchronous learning through face to face interaction, but it is not always feasible for students to do experiment uninfluenced'. *. Virtual Laboratory (VL) can enable students to improve the skills before going to the actual laboratory, to learn breaking the restriction of ordinary arrangement and enhance the instruction3''. Besides, the

*

'

This work is partially founded by the National "Ten-Five'' Key Technologies R&D Project of China (200 1BA 101A08-03). Ph.D Student, College of Computer Science and Technology, Zhejiang University, China

29

30

VL resources can be shared by many institutions and students worldwide, which can save much money. So VL can effectively help to overcome the barriers imposed by the traditional education by using an innovative combination of a new approach to education and the application of new technologies 1.1. Review of Virtual Laboratory

Virtual Laboratory is a heterogeneous, distributed collaborative and experimental environment, which allows students to work on a common group of projects, to generate and deliver results using distributed information and communication technologies. The prototype of VL is Remote Access Laboratory (REAL) '. Computers are connected to manipulate instruments, which could be remotely accessible. When students log in and control these machines via the network, they are able to control both the computer and the equipment. A video camera is used to live broadcast what is happening in the physical world. It does not matter if the student is in a nearby dorm room or on the other side of the world. The general architecture of REAL is Client/Server(C/S) architecture. With the development of simulation technology, REAL gradually replaced by the Simulation-based Laboratory (SLab), which adopts pure-software simulation components or software-supported common simulation hardware named virtual instrument. The world-wide famous virtual instrument provider is National Instrument (NI).Many VL system adopted products from NI are installed in wide area6. '. With the emergency of Web technology, more and more VL systems are transplanted on the Web platformg.With Browse/Server(B/S) architecture, Web-based SLab systems adopt many new technologies such as CORBA, PORTAL and using Virtual Reality (VR) or Rich Client to enhance the instruction and learner's experiencelO,ll,12 . 77

1.2. GUIs to Enhance Learning Experience of SLab system Graphical user interfaces (GUIs) are being increasingly used in the computer aided education to provide users with a friendly and visual approach to enhance learning experience. In REAL system, using video camera to live broadcast the output of instruments to user. But in SLab system, the output of simulation engine is mainly in the form of binary data or ASCII, which can't bring user a visual feedback. The most important value of the GUIs is that it can post-process the results from the background simulation engine, and convert the abstract data

31

into visual form. A visual illustration of how results change as a function of various variables, or parametric sweeps, reinforces the lessons learned in the experiments. There are various technologies to build the GUIs of SLab system, such as computer graphics, virtual reality and flash. And some successfbl SLab 14* 15. systems have been developed and used in Web-based The system addressed in this paper was about a virtual laboratory, which has been used in Zhejiang University and EastSouth University for on-campus analog/digital circuit instruction. The paper is structured in three parts. The first part introduces the concept of virtual laboratory and review some related technologies. The second part describes some issues about system architecture and GUIs design. Finally, an implement and evaluation of this system is discussed.

2. System Design The VL system was developed by College of Computer Science and Technology, Zhejiang University, who has developed the first VR-based chemical virtual laboratory in 200 114.

2.1. ARCHITECTURE

Although VL system can be used in wide area, the common architecture can be described in four layers: access layer, grid layer, supervision layer and resources 1ayerI6.

0

Access layer consists of tools which enable access to the laboratory resources and the presentation of data. Grid layer provides the services which we can use under distributed environment, such as global scheduling and data transport middleware, which connect user-side front-end to background service. Supervision layer is in charge of local scheduling, resources monitoring and user management. Resources layer which consists of devices to experiment execution and also the necessary software used by system.

32

“/.I

User Rich Clienl

User Rich Clint

User Rich Client

Grid Layer

Web /Application Server

User Managemenl

Lab ApplifaHon Server

AccessLayer

Lab Manegemenl

Supervision Layer

LOcal Scheduler

Resources Layer spce Sirnulalor Database

xspice Simu Iatrx Database

Figurel. The architecture of virtual laboratory

Access layer provides user the function of design circuit, collaborative tools with other students, including instructor. The most important components are: 0 Analoddigital circuit design workspace. Virtual experimental instruments, such as oscillograph, multimeter. These equipments can obtain the output from background server and convert data into visual form. 0 Communication panel for multi-users communication and collaboration. 0 User management panel for user logidlogout and authorization. Grid layer mainly provides the function of user response and user authentication. The most important components are: 0 Web server for responding to users’ request and returning the output data. Global scheduling, which responses for choosing the appropriate laboratory application server and load balancing when possible. 0 Data transfer modules for fetching the user-side circuit information and forwards to Lab application server. Supervision layer specific services are gathered. These services have to be implemented taking into consideration a given device and their specification. The most important components are: 0 Lab application server, which receives the user circuit data, generates circuit netlist file and sent to local scheduler. 0 Learning material system (LMS) for managing laboratory instructional material, including user management and lab management.

33

Resources layer which consists of modules of experiment execution, circuit simulation and also the necessary software and database. These modules are: 0 Local scheduler, with responsibility for arranging the circuit netlist file into a FIFO queue according to each simulator. 0 User database, which is the storage of user information and log file. Spice simulator, answering for simulating analog circuit netlist. XSpice simulator, answering for simulating digital circuit netlist.

3. GUIs Design and Development A friendly, interactive user client enables the instructors to develop a virtual laboratory environment which the intended students can interact with. Modern Electronic Design Automation (EDA) software, such as PSpice, Electronic Workbench, is used to establish virtual laboratory for educational purpose. However, most of the EDA software cannot provide sufficient educational instruction to users, for the lack of friendly user interface, especially web-based, interactive visual interface, which can enhance user experience, and significant impact on users' motivation and learning effect. Concerning more complex and friendly interaction, system involves a user dragging and dropping a wire from a pin of circuit component to assemble a circuit. Afier submit the circuit to background server, students can actually view the waveform together with any relevant diagrams or data instead of the long plain text form paragraph output description, which the students may have difficulty in understanding. Not only will this explain the process more clearly, but also the students would derive more enjoyment from learning in such a rich and creative Internet environment. System adopts flash client to build the system instead of VRML (Virtual Reality Model Language) .Though VRML is a popular web tools for generating the immersing learning material via Internet and there are many helpful cases have been de~eloped'~. The main disadvantage of using VRML is high hardware and network bandwidth requirement and low efficiency of appending new experiment items. Using flash can build user interface with less file size. Furthermore, it is support XML data exchange protocol. System adopts the flash MX2004 to build the user interface. The user interface consists of some circuit instruments and apparatus, such as oscillograph, signal generator, multimeter, and circuit board with a set of diodes, transistors, resistances, potentiometer and other circuit components (shown in Figure 2). Users can generate a wire through drag and drop on any pin of components, build any circuit as they wish, power the circuit and use oscillograph and multimeter to check any pin's waveform and voltage value. When the power is on, users can change adjust the circuit and apparatus, for example, turn the knob of a potentiometer, or change the signal frequency of the signal generator. Once the users do that, the displayed result will be refreshed in real time.

34

Figure 2. The GUIs components of system a: analog circuit design panel b: digital circuit design panel c: multimeter d: voltameter e: oscillograph fisignal generator

4. System Evaluation The system has been tested by more than 200 undergraduate students from Zhejiang University and EastSouth University in circuit course instruction, at 2004 autumn. In order to analysis the instructional effect of this system, we made a survey on students' attitude to system. From the feedback and data from survey, we found more than up to 95% students hold positive attitude.

El Enhance understanding

W Improve operational skill

0 Promote learning interest 0 Inspirit innovation Figure 3. The detail survey result of system evaluation and users' feedback.

35

5. Future Works Because all the users using the system is on-campus students, so the server architecture is centralized and system suppose the users' bandwidth is sufficient. The next step of system is to develop a multi-users collaborative system; especially the serve architecture is distributed. According to the new system requirement, some new feature maybe considered. Firstly, design a distributed multi-server architecture, such as Grid-based. Secondly, because of the communication between servers is via Internet, the network QoS can't be guaranteed, so adoption the mechanism of interest management to reduce and filter the data flow of network is needed. Finally, we need to develop a new scheduling algorithm to realize the load balance between servers real-time according to users' action.

6. Acknowledgement Thanks to Mao Song, Huang Yuewei, Mao Yunjie and Ouyang Yang for their work on system design and development.

References 1. Ingvar Gustavsson, Laboratory Experiments in Distance Learning, International Conference on Engineering Education, 8B 1- 14(2001) 2. Bourne, J. R., Brodersen, A. J., Campbell, J. O., Dawant, M. M.and Shiavi, R. G.,A Model for On-line Learning Networks in Engineering Education, International Journal of Engineering Education, 253-262(1966),. 3. Moure, MJ.,Valdes, MD.,Salaverria, A.,Mandado, E., Virtual laboratory as a tool to improve the effectiveness of actual laboratories, International Journal of Engineering Education,Vol.-20., 188-1921(2004) 4. Budhu, M, Enhancing instructions using interactive multimedia simulations, Simulation, 222-23 l(2001) 5. Guimares,EG,Maffeis,AT,Pinto,RP,Miglinski,CA,Cardozo,E,erge~an,M Magalhaes,MF, REAL-A virtual laboratory builtfiom software components, Proceeding of The IEEE,Vol-9 1.440-448(2003) 6. Ko,CC,Chen,BM,Hu,SY,Ramakrishnan,V,Cheng,CD,Zhuang,Y,Chen A Web-based virtual laboratory on aRequency modulation experiment, IEEE Transactions on Systems Man and Cybernetics Part C-Application and Reviews, Vol-3 1.295-303(2001) 7. Ybarra, Gary A,, Introduction to wireless control and virtual instrumentation using Lab VIEW, ASEE Annual Conference Proceedings, 7PPV998)

36

8. Li, Shuhui; Khan, Abrar A., Developing digital measurement and analysis laboratory in circuits and electronics lab at TAWK,ASEE Annual Conference Proceedings, 3225-3238(2004) 9. Ferrero,A,Piuri, V, A simulation toolfor virtual laboratory experiments in a WWW environment, IEEE Transactions on Instrumentation and Measurement,Vol-48. 74 1-746(1999) 10. Fukui,Y,Stubbings,A,Yarnazaki,T,Himeno,R,Constructing a virtual laboratory on the internet: The ITBL portal, High Performance Computing, Lecture Notes in Computer Science,Vol-2858., 288-297(2003) 1 1. Anido,L,Santos,J,Caeiro,M,Rodriguez,J, An online environment supporting high quality education in computational science, Computational ScienceICCS 2002, Lecture Notes in Computer Science,Vol-233 1.,872-88 l(2002) 12. Wang,JX,Lu,WN,Jia, WJ, A new architecture for web-based virtual laboratory with CORBA, Advances In Web-based Learning - ICWL 2003 Proceedings, Lecture Notes in Computer Science,Vol.-2783., 104113(2003) 13. Kam,Y,Cheong,S,Chien,S,You,A, Design and development of a multimedia interactive labfor distance learning applications in the WWW, Advances in Multimedia Information Processing-PCM 2002, Lecture Notes in Computer Science,Vol-2532., 993- lOOO(2002) 14. Dong, Yabo and Zhu, Miaoliang, The animationforming mechanism in VRbased virtual laboratory, IEEE Region 10 Annual International Conference, 269 - 272(2002) 15. Schmid, C,A remote laboratory using virtual reality on the Web, Simulation,Vol-73., 13-21(1999) 16. Lawenda,M,Meyer,N,Rajtar,T,Okon,M,Stoklosa,D,Stroinski,N,Popenda,L, Gdaniec,Z,Adamiak, RW, General conception of the virtual laboratory, Computational Science-ICCS 2004, Lecture Notes in Computer Science,Vol-3038., 1013-1016(2004)

A MECHANISM FOR KNOWLEDGE MAP CONSTRUCTION ON PERSONALIZED E-LEARNING PLATFORM: A SEMANTIC APPROACH SHI-MING HUANG, HSIANG-YUAN HSUEH, AND HSING-YUN JIANG Department of Information Management, National Chung-Cheng Universiq, Taiwan.

Abstract The concept of e-Learning particularly emphasizes how to establish training scenario and integrate on platforms with heterogeneity manners. It is obvious that e-Learning is considered as an important issue for governments, enterprises, and education organization. However, there is still neglect for such issues. Without a systematical approach for construction of personalized e-Learning platform, the effect of such platform will be decreased since the learning path might not be adequate for various learners and educators. In this study, we proposed a systematic approach for knowledge map construction adequate for SCORM-based e-Learning platforms. With aid of knowledge map construction approach, knowledge map can be systematically constructed in such platform, with more personalized manners. In this article we also applied a simple case for competency management in order for feasibility analysis. With aid of result of this article, it is possible for e-Learning platform construction with more personalized, systematical, and semantic manner for insight knowledge. It can therefore facilitate more on e-Learning process based on an e-Learning system, with more tit on users’ personal requirements for learning. Keywords: e-Learning ,Knowledge Map, Competency, Personalization,SCORM.

1. Introduction As the popularity of Internet, the concept and implementation of e-Learning had become more and more important issues, since Information and Internet Technology dramatically changed the ways for learning, with the role of learning and training in both education organizations and enterprises also dramatically changed. As media of learning is varied on Internet, learning methods are also changed with more and more portable, flexible, and adaptive manners. Internet has therefore been widely adopted as a medium for network-enabled transfer of skills, information, and knowledge. Current evolution of learning platforms, such as Web-based distance learning with aid of Internet Technology, supports significant improvement for online courses and training [ 11. It is indeed that e-Learning is important. The concept of e-Learning particularly emphasizes how to establish corresponding training scenario, and integrate on platforms with heterogeneity manners, such as Knowledge Management or Human Resource Management. With integration mechanism among heterogeneous resources, efficiency and effectiveness of e-Learning will be properly improved. It is obvious that e-Learning is considered as an important issue for governments, enterprises, and education organization [ 2 ] . As for research in e-Learning, it is considered as a very important issue to look for a solution of personalized 37

38

e-Learning platform construction [3], particularly with a systematic manner, since the quality of e-Learning platform will directly impact the effect and efficiency of e-Learning, and the behavior of e-Learning status might be varied from different learners and educators. Currently, the Sharable Content Object Reference Model (SCORM) [4] is considered as hndamental standard of e-Learning platform. However, there is still neglect for construction of personalized platform. It is still developer-based and experience-based approach SCORM-complaint e-Learning to be developed. Without a systematical approach for construction of personalized e-Learning platform, the effect of such platform will be decreased, since the learning path might not be adequate for various learners and educators. In this study, we proposed a systematic approach for knowledge map construction adequate for such SCORM-based e-Learning platforms. With aid of knowledge map construction approach, knowledge map can be systematically constructed in such platform, with more personalized manners. In this study we also applied a simple case for competency management in order for feasibility analysis. With aid of result of this article, it is possible for e-Learning platform construction with more personalized, systematical, and semantic manner for insight knowledge. It can therefore facilitate more on e-Learning process based on an e-Learning system, with more fit on users’ personal requirements for learning. Related Works Definition of e-Learning As for definition of e-Learning, it can be defined as a way for propagation of educational information, with aid of Internet and Information Technology [ 5 ] . The characteristics of e-Learning include shar-ability of learning-related resources, anytime, anywhere, and any-thing learning. Summarily, e-Learning can be defined as learning behavior by aid of Internet and Information Technology. It can also be considered as extension of traditional learning process with aid of management among communities, developers, and expertise. 2.

2.1

2.2 Knowledge Managementfor e-Learning Platforms Knowledge management (KM) is a systematic process that helps users gather, exchange, and distribute various forms of knowledge, for creating maximum performance and productivity [b].With facilitate of Information Technology, for both organization and personal, knowledge can be effectively and rapidly applied to support problem solving. Improved knowledge management process, such as knowledge producing, knowledge storing, and knowledge transferring with knowledge bases, is indeed shortening time of problem solving.

39

On the other hand, the notion of knowledge map had become an important component in knowledge-based system, including e-Learning platforms. Knowledge map is currently considered as core component of KM. It can be defined as a form for knowledge presentation, which can facilitate users to discover knowledge efficiently [7]. Summarily, it is expected to have more widely adoption and implementation on e-Learning environments, transferring current data and information management view to knowledge management view.

Competencyfor Human Resource Management Competency theory has been integrated to e-Learning nowadays. IMS organization [8] defined the concept of competency and Reusable Definition of Competency or Educational Objective (RDECO). Competency denotes relationships among knowledge, attitude and ability, which is related to certain jobs and certain employees. The impact of performance and working behavior of one employee can be reflected from such knowledge. Competency can be strengthening by training. It is expected to be widely applied on human resource management (HRM) [9]. The issue of human resource management and e-Learning integration was discussed due to development of e-Learning. It is also expected to exchang&'metadata more conveniently to improve employee competencies and performance. There are some studies that have been focused on application of competency theory for learning behavior management. For example, [3] proposed a framework for personalized e-Learning platform, based on itineraries and long-term navigational behavior. With an embedded knowledge discovery module from database (KDD), it is possible to look for personal behavior from historical data.

2.3

2.4 PageRank for Resource Weighting The PageRank Technology is proposed by [lo]. In PageRank Technology, the characteristic of web hyperlink can be utilized for reference citation analysis. The rank of one resource can be evaluated as weight values with such analysis. Such technology is particularly accepted in Google [ I11 Search engine. The rank of one web source is depended on reference attribute, that is, page hyperlinks reference to the resources. The rank value of one web resource is higher if there are more pages hyper-links to that resource. With such mechanism, the rank of search result can be properly determined. In this study, we applied such approach to evaluate the rank of learning resources according to personal information, in order for more personalized knowledge map construction.

40

3. Mechanism with a Simple Case Illustration 3.1 Overview The following Figure 1 illustrates the overall mechanism of this study. In the mechanism, learning knowledge documents on Learning Management System (LMS) which are compatible with SCORM 1.3 [4] Content Aggregation Model (CAM) standard are used as system input. Two components are contained in such document: One described learning content, and another described learning components, that is, Sharable content Object (SCO) and assets on SCORM-compatible e-Learning platforms. The competence dictionary is a database with manually-constructed competence information. Knowledge map is constructed systematically as output, from Content Model and metadata in CAM. I-

I

---

1

I

I I

I

I IFigure 1: Overview of our mechanism.

3.2 PageRank Approach for Weighting Evaluation For relationship analysis, PageRank approach is applied in this study. The basic principle of Relationship can be defined as connection status for one component or document. As one learning document i be linked to documentj, the value of PageRank Relationship variable is defined “I”, otherwise, its set as “0”. The status of multiple learning documents from i t o j can be stored in Array A as:

,,

a =

1, If learning courselSCOlassetjbe linked from learning document i.

0, If learning courselSCOlassetjbe not linked from learning document i. ,where i, j E N ...

Additionally, in this study we defined the term bi-directional link relationship between two learning resources, when they both connected to one learning object. Then the PageRank (PR) value can therefore be properly evaluated from PageRank array.

41

3.3

Learning Document Construction

The first phase is document construction. It is responsible to construct learning documents with competence identification. There are three steps, as follows: Step 3.3.1:semantic extraction Firstly, we extract semantically patterns of Meta-data for identify learning documents, and identify the keyword of domain knowledge in this course through semantic analysis. We can therefore parse the description of Human Resource B. The pseudo code algorithm can be shown in the following Figure 2. B.pur h r p IS any non-atom 69ntence w cwrpnt h e node hsn Choose the j r s t ley? hand sentence sepator as c m n t mot node; The left-hand-ade ojmntencej m m mot is the ky?-hand-m& sub-node ojmot; The nghl-hand-ado ojssntencej m m mot ISthe nght-hand-srdaarbnode o,fmot; h x i - ~ mpamlepmcess ~ e j o r ley?-hand-adosub-nodo o j m t ; hrui-61~ sopuate processfor r z g h l - h d m d e arbnode ojmot;

End

Figure 2: Algorithm for separating sentence into a semantic tree.

Step 3.3.2:semantic comparison After getting the keyword of domain knowledge, the semantic will be then compared with the competence dictionary according to HR-XML framework [ 121. There are two measurements for comparison of semantics: 0 Naming-conflict: Denotes the synonym or homonym condition for compared semantics. 0 Structure-conflict. Denotes the conflict condition for structure or expression conflict for compared semantics. Types of structure conflict include data type conflict, data dependency or foreign key conflict, and keyword conflict. It is occurred when different parameters are set in different learning resources, with the same or similar semantics. Step 3.3.3: content identij2ation Finally, we mark available learning documents with meta-data for competence identification for further relationship analysis in next step.

3.4 Knowledge Map Construction The second phase is knowledge map construction. It is responsible to construct learning Knowledge Map for Competence Evaluation. Firstly, it is necessary to translate the manifest information of one document to corresponding conceptual connection model. Step 3.4.I : connection initialization Firstly, it is needed to perform relationship connection analysis to “type of relationship” information described by Learning Objects Metadata (LOM) in CAM. With description of recourse connection in “Relation.Resource”

42

information, result of connection analysis can be stored in a knowledge connection status table for further relationship analysis. From our case illustration for Human Resource B, assume it is pre-defined for goals of this course to improve the following competences: Coordination, Persuasion, and Negotiation. And the connection status can be summarized as the following Table 1. Notice that in this case the PageRank evaluation is not necessary, since learning resources are connected only with one-way-directional tree. The connection status model among learning resources can still be generated, as shown in the following Figure 3. In this case we can use degree of connection for one learning object as value of weighting. For example, it is obvious that four courses is related and connected to SCO A. It is therefore that the rank for SCO A is 4.

Human resource B Communication skill Social communication C Art of communication

Coordination, Persuasion, and Negotiation. Persuasion and Negotiation. Social perceptiveness, Persuasion and Negotiation. Persuasion and Negotiation.

A, B, C A, D, E A, E A, C

Figure 3: Knowledge Connection Status model from the case illustration.

Step 3.4.2: relationship analysis Finally, in this step PageRank approach is applied to look up personalized list of important learning objects for certain competences. Such information will be output as form of knowledge map. In this step we applied approach from [lo] to calculate final PR value for SCO learning objects: PR(A) = (1-d) + d(PR(tl)/C(tl) +

...+ PR(tn)/C(tn))

43

In above equation, d denotes damping factor, which is practical PR partial value when the course link to another course. It is set that d is 0.85 in this study. PR(A) denotes sum of PR value of all external course from tl to t,. According to PageRank mechanism, therefore, PR(A) gets PR partial value from every external course. PR(t,) denotes PR partial value of each external course. C(t,) denotes total link degrees of each external course. Thus, 85% PR partial value will be retrieved in every external course. That is, the parameter of 0.85 has average distributed to all external courses. The PR value of our case illustration is shown as the following Table 2. As stated above, the PR value of every course could be evaluated by PageRank mechanism. As one learner attempts to look for more Negotiation skill, the learning path adequate for certain learner, that is, knowledge map, can therefore be generated with form of PR value list.

Feasibility Analysis: a Simple Case Scenario In this study, we applied the O*Net competency database [ I3 I to construct our prototype system. An expression language for O*Net is also provided for definition of competency information for each employee or job [ 131. The metadata and data dictionary of O*Net is pre-defined. Currently the definition version is 7.0. In this study, descriptors are applied for measurement of personalized categorization. Furthermore, in order to adequate for characteristics of e-Learning platform, we applied three factors as input for analysis: Skills, Knowledge, and Abilities. There are 60,456 data items in Knowledge factor, 64,120 in Skill factor, and 95,264 in Abilities factor. It is enough for various competence information expressions. 4.

5. System Implementation:An SCORM Approach In this study, we applied Open Knowledge Initiative (Om)[ 141 architecture for our prototype system. In such architecture, we divided the e-Learning platform into four layers: infrastructure, service API, services, and educational applications. The underlying infrastructure contains fundamental services, such as file service and database service, play as role of resource repository. The service API is tools for accessing of SCORM compatible resources. The service in this platform contains knowledge management and competency management modules, for implementation and manipulation of our mechanism. Finally, the application layer contains interfaces for learning functionalities.

44

6. Conclusion In this study, we proposed a systematic approach for knowledge map construction adequate for such SCORM-based e-Learning platforms. With aid of knowledge map construction approach, knowledge map can be systematically constructed in such platform, with more personalized manners. We also applied a simple case for competency management for e-Learning on Human Resource Management issues in order for feasibility analysis. The prototype is recommended to be constructed according to OKI framework, With aid of result of this article, it is possible for e-Learning platform construction with more personalized, systematical, and semantic manner for insight knowledge. It can therefore facilitate more on e-Learning process based on an e-Learning system, with more fit on users’ personal requirements for learning.

7. Acknowledgement The work presented in this study has been supported by National Science Council, Taiwan, R.O.C, under Grant No. NSC 93-25244- 194-004. We deeply appreciate their financial support and encouragement. Reference [ l ] Zhang, D., Nunamaker, J. (2004), A natural language approach to content-based video indexing and retrieval for interactive e-learning, IEEE Transactions on Multimedia, VI (3), 450-458. [2] Brennan, M. (2003), US. Corporate and Government e-Learning Forecast 2002-2007. IDC. [3] Mor, E., et. al, (2004). e-Learning Personalization based on Itineraries and Long-Term Navigational Behavior, Proceedings of the 13th international World Wide Web conference on Alternate track papers and posters, 264-265. [4] Advanced Distributed Learning, web site available at: httr,://www.adlnet.org. [5] Rosenberg, M., (200 l), E-Learning: Strategies for Delivering Knowledge in the Digital Age, Powells (pub.). [6] Weidner, D. (2002). Using connect and collect to achieve the KM endgame, IT Professional, IV (l), 18-24. [7] Beep Knowledge System: Knowledge Map Search, web site available at: htt~://www.beepknowlednesystem.orn~nowled~eMa~.as~ [8] IMS Global Learning Consortium, web site available at: http://www.imsglobal.org. [9] Hall, B. (1997). Web-based training cook book, New Work, John Wiley and Sons.

45

[ 101Lawrence, P., et. al, (1998). The PageRank Citation Ranking: Bringing

Order to the Web, Stanford Database Group Publication Server, web site available at: httr>://db~ubs.stanford.edu:809O/cgi-bin/makehtml.cgi?document= 1999166. [ 111Google Technology, web site available at: httr>://www.google.com/technology.

[ 121HR-XML Consortium, Inc., web site available at: http://www.hr-xmI.org. [ 13]0*NET Online, website available at: http://online.onetcenter.org [ 1410pen Knowledge Initiative, web site available at: httu://www.okipro-iect.org


YET ANOTHER PLATFORM FOR WED-BASED LEARNING T. S. LI, S. M. WONG and REGGIE KWAN School of Science and Technology The Open University of Hong Kong, Hong Kong

Abstract In this paper, we will introduce an online education platform which enables the teacher to organize teaching materials, assignment booklet release, classes scheduling, interactive communications with students and updated stop presses on the Web. This platform is especially useful for a distance learning environment.

Introduction We have seen rapid progress in online learning platforms over the last decade. There have been quite a number of online education platforms like WebCT [ 11, Lotus Domino [ 2 ] , Blackboard [3], Learn Wise [4]and First Class [5] in the market. The main feature of the online platform is to provide learners with an effective and efficient learning environment through the use of web-based discussion forums, online materials, self-help tests and electronic assignment submission. Many teachers have found these platforms very useful in terms of organizing online discussion forums, publishing teaching materials [6]. However, these platforms are not quite as useful when they are used in a distance leaning education. In this paper, we will first discuss what extra features that are needed for the distance education environment. Then we will introduce a platform we developed that addresses these requirements.

Extra Features that are needed for distance education An online teaching platform must provide the following features: A tool for teachers to upload teaching materials; An online discussion forum. Some platforms like Blackboard also allow students to work together and then the teacher is able to observe how they work. However, this facility is not very popular at the time being because: In a conventional face-to-face institution, it is better to have students actually work together face-to-face [7]. In a distance learning institution, students might come from different backgrounds and therefore some may not be able to afford the high 47

48

speed internet connection, some may not have a quiet environment at home that they can concentrate at the work. Although real time interactive learning environment is the way to go in the future, right now it is not as useful as other teaching media like static and dynamic web pages, animation, discussion forum etc. Most of these online educational platforms were built with conventional educational institutes in mind, so they lack some of the features that are important in distance educational institutions: A way for the teacher to organize teaching schedules, assignment submission dates, classes. These features are important to distance educational institutions because we need to have a way to alert students of what are going to happens. If we have the electronic records of all the teaching schedules, assignment submission dates etc, then we can send email or other messages to alert students about anything that is to happen. This feature is less important in conventional educational institutes because students go to school everyday and submit assignment regularly. So they usually are aware of the classes or assignment deadlines. However, students studying in the distance learning mode are usually tided up with a day time job and therefore need special reminders from time to time. 0 Electronic submission of assignment and online marking facility. Again, this feature is not very useful for conventional institutions because students would go to school everyday and therefore they can submit the assignment to their teacher in person. However, using electronic submission of assignment would not only allow distance learning students to submit assignment more easily, it also has some byproducts. For example, when students submit their assignment electronically, it is possible to add an automatic plagiarism detection module [8].

The main features of our online learning platform The platform has four main modules: 0 module for organizing teaching schedules; 0 module for publishing teaching materials; 0 module for electronic assignment submission and marking; 0 module for tracking student's assignment submission progress. The platform has two different views: 0 a teacher's view which allows the teacher to organize the course;

49

0

a teacher's view which allows the teacher to organize the course; a student's view which allows student to view different schedules, materials and to submit assignments.

Module for organizing teaching schedules In this module, we allow the teacher to set teaching schedules like class timetable, assignment deadlines. In the Open University of Hong Kong (OUHK), we have two different types of face-to-face classes, one is tutorials and the other is surgeries. Students are divided into tutorial groups so that about 35 students are in one group. The page to enter the teaching schedule is shown in Figure 1. This page contains a table that allows the teacher to key in different topics, assignment deadline (assignments are known as TMA in OUHK), tutorial and surgery times. After keying in the time for tutorials and surgeries, there are pages to key in the tutorials and surgeries details like tutors in charge, teaching contents and venue. Since all the information is stored in a database, it is possible to send alert messages to all students and tutors to inform them of upcoming events like tutorial classes, assignment deadlines etc. The student's view of the teaching schedules is shown in Figure 2.

50

Electronic Assignment Submission As mentioned above, electronic assignment submission is a very important component on a distance learning platform. If many students are trying to submit the assignment at the same time, then it is possible that the upload rate become unacceptably low. Student will be very frustrated if they start up the submission process minutes before deadline but cannot finish after the deadline.

1 Presentation Schedule of MT311(2001-OCT)

H

Figure 2 Teaching schedule

There are several ways to solve this problem. The first way is to add a registration step before the actual uploading. Students are given a small program which they run at home. The program will read in the assignment file and produce an MD5 hash code. Then, student will use this hash code to register with the online platform. The time when student register the hash code is considered to be the time of submitting the assignment. Then students can have a number of days to upload the assignment. In this way, they do not have to rush to upload the assignment.

51

The second way is to break up the assignment into small chunks and then upload the chunks one by one. In this way, a students need to run a small program on their computer and then the program will read in the assignment, generate a hash code and register this with the server, then finally it will send the assignment in small chunks. Students can stop the program at any time and restart the program to upload any chunks that have not been uploaded. Again, the submission time is taken to be the registration of the hash code. The program will communicate with the server to determine the delay between sending two consecutive chunks according to the current server loading.

Publishing teaching materials It would be simple to publish materials if it is pure text or html that does not include pictures or other documents that are hyperlinked. In this case, we just need to set up a form for the uploading of text or html. However, if the materials contain pictures and other documents that are hyperlinked, then this method would have problem. One method to solve this problem is to have the pictures and other hyperlinked documents uploaded first and then each of them is referred by a URL. Then the html file would refer to the URL and then uploaded to the platform. However, this method requires the user to keep track of the URL of each picture and require the uploading of each picture to be done individually. In our platform, the user uploads a zip file that contains a file named index.htm1 and all other relevant files. It should contain directory information. The server will then expand the zip file in a dedicated directory together with all the files in appropriate subdirectories. Of course, all reference to pictures in the zip file must be using relative links. In this way, all the html files and pictures can be uploaded in one go. Currently, teaching materials are divided into the following categories: 0 0

0 0 0

0

PDF files Online activities Self-tests Online reading Errata What's new.

Obviously, PDF files, online activities, self-tests, online readings are usually prepared before the class starts while what's new, errata will be published from time to time when there is a need. So when we publish an errata or a what's new item, we want to show that time when it is uploaded. Figure 3 shows the page where the user can input the contents for errata. You can see that there is a field

52

to allow the user to upload a zip file. Right below that field is a text area for the user to enable a pure text version of the errata. Figure 4 shows the students' view of the errata.

Uploadhtml canterts

Errata 4.1 posted at 14-Mu-05 23:21

[

Delete Errata1

3

J

Page 4 In the matcnd. you were told how to allow an appler access a local iile when the applet IS executed unog applemcwer Ifyou want to by this ~n a browrcr, then you need the follormng semng

Figure 3 The page for entering errata

Conclusions This paper introduces an improved online learning platform which is especially suitable for distance learning education. Our platform offers a better attractive environment to help teachers organize effective online learning materials and scheduling for distance learners. The system also incorporated with a more flexible system for electronic submission of assignments in which the student could upload assignment independent of time and space.

53

B m t a for unit 4

I

Posted at 14Mar-05Z3:21 Page 4 In the material, yw were told how to allow m amlet m e s a Iml file when the *let is executedusing applebiewer If you wmt to by this in a browser, t k n you need the follwing Setting

In wder to allow appleh rurning in browsers to have mess to y a r local files, you need to have a file called java policy sitbrg In y w home directory If you a e using XP and y w use^ mme on that machine Is abc,then y w home directory sharld be

Figure 4 Student's view of the errata

References 1.

WebCT, http://w.webct.com

2. Lotus Domino, http://www-306.ibm.com/software/lotus/

3.

Blackboard, http://www.blackboard.com

4. Learnwise, http://www.learnwise.com 5.

Firstclass, http://www.firstclass.com

6. Learning Technology Group, Centre for the Enhancement of Learning and Teaching, "Report on E-Learning survey 2004",

http://domino.lancs.ac.uk/CELT~earnTech.nsf/O/C7CA62ECAO89DA BB80256EC800384B7 1/$FILE/survey2004SA.doc?openelement.

54

7.

http://~.iped.vxu.se/forskn/projekt/wm/texts/NLrepor.htm

8. S. C. Ng, T. S. Li and H. S. Ngai, "Plagiarism Detection of Programming Assignments in Distance Learning", Proceedings of the 21st ICDE World Conference on Open Learning and Distance Education

(ICDE 2004), Hong Kong, Feb 18-2 1 2004.

PART TWO Models


BLENDED LEARNING APPROACH: A Strategy to Address the Issue of Declining Enrollment in Mechanical Programs and A Promising Model in Teaching AutoCAD in Arabic R . BEDRI and M.O. AL-NAIS College of Technology at Hail, P.O. Box. 1690 Hail, Saudi Arabia Tel: +966 6531 7705 ext 240 Far: +966 6531 7704 e-mail: r bedri@,vahoo.com ABSTRACT This study explores the various strategies that can be envisaged and deployed in view to address some critical issues including declining enrollment, excessive drop-out, and attrition. Furthermore, this contribution reports on preliminary findings emerging from an ongoing analysis aimed at understanding the impact of Blended Learning Approach on the teaching of AutoCAD and the opportunity to extend the use of this new technology to more courses.

1. Introduction

The forces of Information and Communication Technologies (ICT) on one hand and globalization of education on the other hand are slowly, but dramatically, transforming education from narrow, local, face-to-face-basedtype of education into a global, technology-based, student-centered learning in which online education is a key component [I]. This reshaping of education has inexorably started and concerns every aspect and types of learning including vocational training and technical education. To our standpoint, globalization should mean erasure of both physical and virtual barriers without elimination of intrinsic cultural values and ethos. In this perspective, every nation, be it a third world one, should welcome and grab the unique opportunity bestowed by the ICT technologies to address critical issues pertaining specifically to education and which are exasperated in the poor countries by demography and limited resources. In the kingdom of Saudi Arabia, the General Organization for Technical Education and Vocational Training (GOTEVOT), to name but this vital institution, has allocated significant resources to vocational education and this commitment appears to be set to continue for the foreseeable future. Specifically, this organization has invested and continues to invest heavily in information technology assets, including hardware, software, telecommunicationsand training in the field of emerging technologies. 57

58

In our college of technology at Hail, run by the GOTEVOT, we are experiencing unacceptable dropout rates and a dramatically declining enrollment in the department of mechanical technology, despite the availability of scholarships for all Saudi students. This preoccupying situation should be interpreted by our fellow educators and instructors in the mechanical technology department as a reminder of the need to improve and redefine our educational programs. Learners and instructors alike continue to become comfortable and adept with information technologies and e-learning applications despite inertia and reluctance to the change manifested by the opponents. Negative and positive biases persist among instructors, administrators,and support personnel regarding the efficacy of delivering curricula through the Internet. Today rapid technological development requires innovative educational approaches. If we wish to continue to communicate effectively in the contemporary environment, we are bound to constantly update our visual literacy and our technological knowledge. To address the issues of attrition and declining enrollment, we have come up with few strategies that we are projecting to adopt. These encompass the proposal of new computer-related mechanical specialties and the gradual introduction and adoption of innovative educational approaches. Pedagogical strategies that are based on partial use of state-of-the-art technologies such as Internet and computer-based teaching (CBT) coupled with the traditional classroom should correspond to the highest level of student's convenience. This teaching format referred to, in the literature as Blended Learning [2,3,4,5,6], can constitute a judicious compromise between the two worlds. This setting has been successfilly experienced in teaching AutoCAD in our department. 2. Diagnosis and analysis At the college of technology, students enroll in specialized programs which are normally two academic years in length, during which they take 72 credit hours of course work. These diploma programs are offered in Hail and many other colleges run by GOTEVOT in Saudi Arabia. At the Hail College, there are three different technical specialties and some commercial ones from which students may choose. Technical specialties are offered in the fields of Production, Computer and Electrical Technology. In the college at Hail, we have been experiencing lately a noticeable reduction in students' enrollment in the mechanical technology department whereas we have been noticing an incredible rush to the department of computer technology and to a lesser extent to the department of electrical technology. This inexplicable situation has prompted us to undertake an objective analysis in order to cogitate and elucidate the motives responsible for the students'

59

disinterest and aversion of mechanical programs and identify the repulsive causes that alienate and turn them away from the department. 2.1. Societal perception of mechanical technology careers

In our view one possible reason why our high school students may hesitate to enroll in mechanical technology programs is their negative perception about technology careers coupled with lack of understanding of the skills required to succeed in mechanical technology careers. This perception is the product of social cultural values and attributes, some deeply rooted in Middle Eastern history and some product of the oil-boom experience. The major contemporary cultural and social features of the Middle East have influence on the vocational education system. [7]. Sociologists all agree to testify that attitudes and perception toward specific careers have an enormous impact on the choices that high school students make, and in fact, attitudes sometimes play a greater role in career choice than classroom education, job opportunities, fhture career growth, and long term learning potential [S]. 2.2. Students' biases

Careers in the fields of computer technology are now more attractive to young people because of the abundance of available jobs and competitive salaries in both the private and public sectors where the computer has become ubiquitous. On the other hand, manual labor is also spoken of as honest and decent work in the Islamic teachings. All of these factors have contributed to the shift in outlook on such professions. 3. Strategies to promote mechanical technology programs We believe with confidence that it is possible to shape positive attitudes toward mechanical technology programs and attract qualitatively and in numbers the best students by elaborating and adopting innovative strategies and devising ingenious mechanisms. This would revolve around three principal axes: Creation of a mechanical technology awareness club Proposal of two new training programs Adoption of a blended learning approach 3.1. Mechanical technology awareness club

This mechanism would be committed to the promotion of the mechanical technology careers through the education of people and the erasure of stereotypes and to create awareness toward the importance of the technology professions in Hail. Moreover, strong links must be sought and strengthen with

60

Hail chamber of industry and commerce, and key private sector employers who can play a major role in raising public awareness toward this profession. 3.2. Proposal of new training programs To date, our department is offering just one program of training leading to a diploma in production. We believe that proposing new wave programs would be very efficient and opens new perspectives to the prospective students who would be exposed to alternatives and not be confronted anymore to the single choice available. This undoubtedly would contribute in attracting more students to the department. Acknowledging and taking into account the students' attitudes and biases, Mechatronics and CAD have the potential to win the students' favors. 3.2.1. Diploma in industrial mechatronics Training in industrial mechatronics aims at qualifying the trainees on technical as well as personal skills and knowledge in order to meet the demands for assembly, setup, maintenance and design of automated industrial systems controlled by computer [9]. Besides the general subjects, technical aspect would focus on and span three blending components namely: 0 The mechanic/machining field. The electdelectronic field. 0 Control and PLC technology. 3.2.2. Diploma in CAD Staying in tune with the general policy of the GOTEVOT, this new specialty would contribute to respond favorably to the national demand for skilled labor force in this field. Besides the usual general subjects, an emphasis must be placed on engineering graphics, descriptive geometry, CAD fundamentals, applied mechanics , strength of materials, mechanical drafting, tool design, and mechanical design project [lo]. 3.3. Blended Learning Approach Simply put, Blended Learning can be described as a program where more than one delivery mode is being used with the objective of optimizing the learning outcome and cost of program delivery. However, it is not the mixing and matching of different learning delivery modes by itself that is of significance, but the focus on the learning and business outcome [4]. A study by Peter Dean and his colleagues [ 5 ] found that providing several linked options for learners, in addition to classroom training, increased what they learned. In 2002, Harvard Business School faculty DeLacey and Leonard [6] reported that students not only learned more when online sessions were added to traditional courses, but

61

student interaction and satisfaction improved as well. Our initial thought was that an online approach would be a useful addition to face-to- face teaching. So, with some colleagues in our department, we decided to experiment with some online techniques. We started by designing, developing and deploying on the web electronic courseware. We selected for this purpose four core subjects: materials testing [ll], AutoCAD [12], metrology and quality control [13]. These electronic courses were not meant to replicate nor supplant their on-campus analogs but to complement them. In this contribution, we will confine our analysis on the transformed model of teaching AutoCAD. There are two overall objectives to this transformation: 1. Increase learning impact, and 2. Increase learner satisfaction and ease of use. 3.3.1 First experience in the use of Blended Learning Approach Having been directly involved in the teaching of AutoCAD in the department of mechanical technology, for some time now, we have attempted to analyze informally the reaction of our trainees and students in their learning process of this software. The analysis has revealed the following issues [ 141: 0 Communication handicap: the graphical user interface with its menu bars and commands being in English, a minimum language command is required to be able to perform well and deal with the necessary and inevitable interaction sequences. 0 Each trainee, having his own pace of learning, making the whole process of teaching difficult in a stand-deliver-format. 0 In general, the first lab sessions are well accepted by our trainees as they constitute their first encounter with the software, so curiosity stimulates further their enthusiasm in discovering this software. 0 Annoyance and disinterest rapidly catch up with some trainees in the following sessions especially when detailing the different drafting commands and the more involved user interaction sequences. Our trainees are curiously and unexpectedly more engaged and manifest interest when topics on 3D modeling and rendering are considered. Furthermore, it has been observed with regret, that most of the lab sessions' time is consumed interpreting the different commands and translating from English to Arabic the different aspects of drawing movements and interactive sequences than teaching the use and applications of the software. All these shortcomings and hindering barriers have prompted us to seek innovative teaching approach that may permit to transmit and channel the drafting skills to frustrated and disabused students and adopt new paradigms that

62

assist in transforming the whole process of learning to a more engaging and enthusiastic one. The blending configuration adopted encompasses four blending components: On-campus lab sessions: a four-hour-lab-session weekly. The CAD laboratory is equipped with 25 Pentium 111, PCs, all connected through a local area network (LAN) to a server administered by the instructor. A software (Net-Op) is installed to permit the instructor to monitor on the server screen the students' activities, demonstrate and explain drawing drills, select and communicate directly with one or a bunch of students, let a student's work be projected directly on the screens, give full control to a student to expose his work, etc. Computer-Based Teaching through the use of a CD-ROM developed by the GOTEVOT. This product includes a well conceived first course on AutoCAD, and the students are encouraged to acquire a copy fiom the department as well as a printed version of this course. Assignments are purposely chosen fiom this CD-ROM in order to give the students the opportunity to be exposed to this course and reap an added benefit. Furthermore, in order to instill teamwork spirit into the students, these assignments are shared by small groups of four students. Web-Based Arabic CAD-Tutor: flee access anywhere, anytime at one's convenience, transcending all the barriers of space, time, and what's more the language hampers. This Tutor as will be argued amply in the next chapters, does not replicate the CD-ROM course nor does it duplicate the classroom lectures but is conceived as a supplement to both. In fact, it is meant to broaden and deepen the scope of the students' knowledge in the sense that more advanced topics are tackled besides the basics of AutoCAD that are presented in a simplified step by step fashion. VHS tapes are made available to whoever needs to revisit elements of engineering or technical drawing that are usually taken respectively in the first and second semester of the curriculum, whereas AutoCAD is taught in the fourth and final semester. 3.3.1.1. Design phase of the web-based Arabic CAD-Tutor

Acknowledging the above analysis outcome, we have set out to design an educational tool that takes into account these findings and caters for most of the shortcomings of the traditional teaching format. As is widely accepted today, elearning is the best vehicle for supplementing the knowledge beyond the classroom. Specifically, having noted that English language constituted the main obstacle precluding the comprehension of the software by our students, we

63

decided to conceive the electronic courseware in Arabic [ 121. This pedagogical tool is meant to enhance the drafting skills of the student and even broaden his scope of knowledge by exposing him to more advanced topics not necessarily addressed in the classroom. 3.3.1.2. Content development Developing an online course can require the assembly of many types of data forms: text, graphics, animation, videos, applets, etc. It also requires an environment or architecture for residence. Finally, it requires the glue or properties that will facilitate assembling these together [ 151. The web-based Arabic CAD-Tutor is structured in three hyperlinked groups of lectures: The first group is designed as an introductory course on AutoCAD, providing the basics of computer aided drafting. A thorough presentation of the graphical user interface is given with all the details about the menu bar, the floating and docked icons bars, the drawing and modifying commands together with a comprehensive explanation on the interactive sequences associated with them. The second group of lectures address various important issues such as the different coordinate systems, absolute, cumulative or relative, Cartesian and polar, user coordinate system, object snap tools, dimensioning, layers, orthographic projections, etc. The third group of lectures deal with 3D modeling, rendering, Boolean operations, animation, etc. 3.3.1.3. Web environment and hosting In order to create a web environment for this Tutor, and to make the structuring of content and combining of media elements into a cohesive program, we opted for Microsoft's FrontPage. The Tutor has been temporarily hosted at Tripod Lycos [12] and will be promptly transferred to the college web site as soon as this latter will become fully functional and operational. 3.3.1.4. Quafity controffor the Tutor The main quality control for a web-based product is the so-called combined process of alpha and beta testing. These tests are conducted to ensure the following attributes [ 161: accuracy of content, ease of use, instructional integrity, technological soundness and user satisfaction Without going into detail about the alpha and beta testing performed on the CAD-Tutor, it is worthwhile mentioning however that the alpha test which intervened in the development phase of the project, was undertaken to verify the accuracy of the content. The primary testers were the subject experts involved in

64

this project with the association of some selected professionals well versed in this field. As for the beta testing, surveys were pilot tested on fifty five students. A combination of structured and open-ended questions were cast to give respondents the opportunity to express their feelings, motives, or behavior spontaneously. The choice of open-ended questionnaire was dictated by the fact that such a type of survey generally produces more self revelations by subjects and the ability to express thoughts and opinions, without inhibition. The questionnaire yielded some valuable information: More than a fifth of the respondents feel uneasy when navigating the Tutor. 0 About half of the students admit not to have access to the web at home and connect from cyber cafe. 0 A third of them don't possess a PC. 0 Less than a fifth of them think that the Tutor is a waste of time. 0 More than sixty per cent welcome the Tutor and wish to see it more developed. 3.3.2. Evaluation of the transformed method of teaching The same questionnaire used to evaluate the Tutor included further items pertaining to the testing of the new learning approach as a whole. This has produced the following abridged groups of answers: Half of the respondents are not sure on which of the delivery modes to concentrate most efforts and require more guidance. 0 Most of them are more concerned about the topics on which they would be examined in the final test. 0 More than half of the students appreciate the VSH tapes and find them helpful. Two third of them think that the problem of language is no longer a major handicap. 0 Less than a third prefer to stick with the traditional form of teaching despite the difficulties encountered. Moreover, it has been observed that the rate of failure at the final test had sensibly declined if compared with the results obtained in the past. In the light of these findings and acknowledging students' feedback, it has been decided to persevere in this direction. However, it has been decided to market more the benefits of the transformed method of teaching to learners and their supervisors, and to provide instructions on how to get access to the net and to the CAD-Tutor. Besides, It has been decided to take hrther actions in order to help the most reluctant students to embark willingly and eagerly on this transformed learning format. Sending and receiving e-mail is arguably the most

65

basic and ubiquitous skill in technological communication. It is often the first and easiest skill that computer novices learn. As such, we have convened to capitalize on the judicious use of e-mail as a confidence builder for the most bewildered students. An e-mail exchange is an ideal way to get started in this manner. We solicit an initial e-mail contact from each of our students during the first week of the course. This enables us to have their Internet addresses on record. It also allows us to greet them with individual replies. Through such exchanges, students can essentially "road-test their connectivity" to us and to the CAD-Tutor. This exchange provides them with a prime opportunity to practice sending and receiving e-mail during the startup week of the course.

4. Comments and conclusions It would be pure speculation to foretell anything concerning the impact that would have the introduction of the two new training programs, i.e., CAD and Industrial Mechatronics, on the level of enrollment in the mechanical department. However we can only anticipate a positive effect if we take into account the biases and the reshaped attitudes of the prospective high school students once the strategies have been deployed and by marketing the mechanical technology programs through public awareness and outreach. Preliminary evidence suggests that the blended learning approach as applied to the teaching of AutoCAD demonstrates relative student satisfaction and greater success on assignments and final tests compared to the traditional version of the course. This experience should be extended to other subjects. At least three more subjects are potentially candidates for this transformative teaching setting, namely Materials Testing, Metrology and Quality Control, owing to the fact that electronic versions of these courses have already been deployed on the web, while the other ingredients can readily be made available.

References 1. http://interact.hmnet.ordwebcommissionhdex. htm 2. M. Abramovici, K. Borilski and A. Stekolschik, "Blended Learning in Product Design Education and Training", International Engineering and Product Design Education Conference, 2-3 Sept. 2004, Delft, The Netherlands. 3. G. Prendergast, "Blended Collaborative Learning: Online Teaching of Online Educators" available at: h~://www.~lobaled.com/articles/Ger~d~ender~ast2004.html 4. H. Singh and C. Reed, "Achieving Success with Blended Learning", 2001 ASTD State of the Industry Report, American Society for Training and development, March 2001.

66

5 . P. Dean, M. Stahl, D. Sylwester and J. Pear, (2001). Effectiveness of

Combined Delivery Modalities for Distance Learning and Resident Learning. Quarterly Review Of Distance Education, 2(3), 247-254. 6. B.J. DeLacey and D.A. Leonard, (2002). Case study on technology and distance in education at the Harvard Business School. Educational Technology and Society, 5(2), 13-28. 7. K. MELLAHI, "Human Resource Development through Vocational Education in Gulf Cooperation Countries: the case of Saudi Arabia", Journal of Vocational Education and Training, Vol. 52, No. 2,2000. 8. U.G. Gupta and L.E. Houtz, "High School Students' Perceptions of Information Technology Skills and Careers", Journal of Industrial Technology, V O ~16, . NO.4, Aug- OCt. 2000. 9. httu://www.gtzth.orglibran//files/gtzth 1998 4480 4 1.html 10. Xin-Ran Duan, "A model Curriculum for Computer Aided Design (CAD) Associate Degree Programs", Journal of Industrial Technology, Vol. 20, No.2, Feb-May 2004. 11. R. Bedri and H. Al-Moteiry, 2004, "Web Based Materials Testing Dedicated Tutorials", available at: http://saidhail.tripod.com 12. R. Bedri, 2004, "Web Based Arabic CAD Tutor" available at:

httu://ramdanhail.trieod.com/ 13. M. Aichouni, 2004, "Dimensional Metrology Home Page", available at http://hctmetroloe;v.tripod.com 14. R. Bedri and M.O. Al-Nais, Development of an Arabic on-line CAD Tutor", 3rdSaudi Technical Conference and Exhibition, Conference Proceedings, Vol. 4, pp. 302-307, Riyadh, Saudi Arabia, 11-15 December, 2004. 15. R. Bedri and M.O. Al-Nais, "Web Based Dedicated Tutorials", First Baha Technical Meeting, Conference Proceedings, Computer Systems and Distance Learning, pp. 171-176, Al-Baha, Saudi Arabia, May 34,2004. 16. B. Lesniak, "Putting it to the Test: Quality Control for e-Learning Courses", The eLEARNING DEVELOPERS JOURNAL available at: http://www .elearningGuild.com

CONCEPT MAPS AND LEARNING OBJECTS LEONEL IRIARTE NAVARRO Agrarian University of Havana, Habana , Cuba Email: [email protected] MANUEL MARC0 SUCH Languages and Computer Science Systems Department at the Universityof Alicante, Alicante , Spain, Email: [email protected] PEDRO PERNIAS PECO Languages and Computer Science Systems Department at the University of Alicante, Alicante , Spain, Email: [email protected] DANIEL MORON MARTfN Languages and Computer Science Systems Department at the Universityof Alicante, Alicante , Spain Email: [email protected]

ABSTRACT Concept maps constitute one of the tools frequently used in learning management as they offer the possibility to personalize learning, share knowledge and re-enforce learning to learn skills. At the same time, many initiarives or standards are being developed rapidly to make the contents in dinerent learning management systems and learning environments compatible. This paper states the need to combine the technique of concept maps with initiatives that package contents developed by IMS to produce more portable and powerful content. A model to create tools for learning management is proposed.

1. Introduction

With the outburst of new technologies and the appearance of powerful methods and tools for knowledge management, new virtual communities have been created which go beyond the limits of traditional teaching processes while offering both professors and students a number of teaching services. Despite its success, the learning process has been affected, as many professors and institutions have tried to apply the same learning structures found in traditional education to the new circumstances. In generating new materials while considering these new developments, it is necessary to change our way of thinking and to include the idea of making the student responsible for his intellectual autonomy and to manage the knowledge he needs [l]. It is very important to adapt the new content to each student’s characteristics, and at the same time, the professor should coordinate all the activities of the virtual community to expand the knowledge frontier, promoting the sharing of experiences. 67

68

In the seventies, the theory of Concept Maps appeared in Cornell University in United States as an answer to the learning theory developed by Ausubel that relates the evolution of students‘ previous knowledge to construct new knowledge. Ever since, this technique has been a useful tool for teachers, researchers in topics related to education, psychologists, sociologists, students and other areas that manage large quantities of information [2]. A Concept Map constitutes a schematic summary of what has been learned; it is hierarchically organized and represents all abstraction levels. The design and creation of Concept Maps is done through these automated systems that are available to the users in diverse forms; each has its own format, structure and style. Most of them export to HTML format, making browsing through map nodes possible. The use of these tools generates much knowledge that remains in the “invisible web”, as each professor or institution produces its own maps and they are not published or organized adequately, so that they can be reused some other time. Many times a map’s concept is linked to a website where the student cannot find the information he needs, causing it to lose its essence. This happens because when it is being designed, the professor has no alternatives, which allow him to look for precise information that is oriented toward concept learning The existing software offers no possibility for introducing highly complex knowledge representational structures where information can be generated in a dynamic way and cooperative learning is present; the maps represent sequences or hierarchies due to their own nature. Many compatibility and learning structure problems we have stated have been solved in the creation of standards that permit the documentation, search, and distribution of educational contents that are generated in different environments [3]. IMS developed by the Global Learning Consortium [4] is one of the most important standards from which SCORM was developed by Advanced Distributed Learning Initiative [ 5 ] and the Institute of Electrical and Electronics Engineers [61* Standards are based on the definition of learning objects as any digital object that can be reused [7]. To achieve reusability and interoperability in educational materials, different specifications that provide schemas to classify and package contents have been created. Some specifications that define standard structures to exchange contents and learning structures have arisen, making it possible to convert these structures to learning objects. The best-known ones are those developed by IMS [4]. The first and most widely spread one is IMS Content Packaging (IMS-CP) [l 11, which permits the organization of hierarchical content structures. Next, IMS Simple Sequencing Specification was proposed (IMS-SS) [ 121 which is based on IMS-CP and permits the definition of sequences of educational activities and the conditions in which each activity should be done. Last, IMS Learning Design (IMS- LD) [ 131 has been developed which includes the previous models and incorporates the possibility of modelling learning from the Educational Modelling Language (EML) [ 141 specifications created in the Open University of Holland.

69

IMS LD permits modelling of more complex learning structures in which many users can develop educational activities at the same time, depending on their conditions and with the possibility to exchange information and messages. These are the most complete specifications that exist at the present time to achieve the standardization and reusability of learning structures; but they are highly complex for use by professors and in their implementation through software. There are different research projects and institutions that generate applications to implement these specifications in their different levels [15]; nevertheless, most of them concentrate just in implementing the specifications and not in developing interface in which professors do not need to know them in detail. For this reason, this article proposes a model for both designers and users to take advantage of concept maps for modelling and managing learning as well as the benefits of IMS LD to obtain more complex and reusable learning units. Before defining our model, we will make reference to some general concepts about concept maps and the principal elements of IMS LD specifications. 1.1. Concept Maps (CM) Concepts, linking words and prepositions are the basic elements of a concept map. Concepts are also called nodes, which make reference to anything that can be brought about or that exists. According to Novak [ 161, concepts are mental images, which provoke within us the words or signals we use to express regularities. Linking words join concepts and the type of relationship between them. The preposition is the semantic unit that joins concepts. In the concept map, elements are graphically organized, forming semantic chains and the knowledge is lineally and hierarchically organized which make holistic units in such a way that when one activates itself, the rest do as well. (Fig 1).

Figure 1. Graphic Representation of a Concept Map

Concept maps can be described in abstract, where each of the interconnected nodes can be seen as an organized hypergraph .

70

Due to the nature of concept maps, they are principally represented by hierarchically organized structures. However, spiderwebs, sequential structures or system maps which include inputs and outputs that feed into each of the map concepts can be used. There are other types of concept maps like the hyper-medial ones that have some concepts interrelated through linking words. There is a lot of software that allows professors and students to design concept maps. Many of them are free and offer the user many possibilities in designing and using them, thus making it a widely spread technique for learning management. 1.2. Standardization of learning structures As we explained above, there are different specifications that permit modelling and describing the content structure so that it is both portable and reusable in different educational environments. Below we make reference to the main elements that make up the "IMS Learning Design" (IMLD- LD). IMS-LD permits the description of the tasks and activities structure within a learning unit. As in all specifications, XML is used to describe the units. There is a precise and rigorous documentation [ 131 about each specification component, so it is not necessary to concentrate on that. An IMS-LD based learning unit is composed of obiectives, roles (whether students or professors), learning objects, and learning activities organized through different structures formed by services [ 181. (Fig 2 )

components roles

learner' staff* activities learning-activity*

environment-ref* activity-description

s u pport-activi t y '

activity-structure' CsequenceIselection> envlron m e n t-ref' activity-ref* activity-structu r e - r e f * environ rnents environment' l e a r n I ng -object' learning-service' mall-send* conference'

method

play;ct* role- parts* r ole-r role. ef activlty-ref ~~

Figure 2. Components of IMS-LD specification

In IMS-LD it is necessary to understand the different roles of users in carrying out the activities organized in coherent structures [ 181. (Fig. 3 )

71 method

play

Role-part 1 Role-part 2 Role-part 4

corn ponents

Figure 3. IMS-LD Functioning Schema

These specifications have three levels of implementation. Level A was explained above and is the one that defines the vocabulary and the metadata necessary for the specification. Level B permits the association of properties and conditions for carrying out the activities. Level C allows communication and messages to be sent among the learning unit components. Optimally, a professor would be able to write a learning unit using the schemas defined in the specifications, but this is impossible due to the conceptual complexity and levels of abstraction that are necessary; for this reason, some tools and projects have emerged to aid design of learning units using the specifications as a basis. The majority of existing software products, currently in evaluation, have complex and inadequate interface, which causes the users' rejection. Function and component libraries are also created, they permit software developers to create facilities to use these specifications in the applications they develop [15].

2. Interoperable Concept Maps As we have explained above, research shows the effectiveness of concept maps to improve learning, as well as the experience acquired by professors and students in using them; nevertheless, it is important to point out that in the design phase, the professor needs to associate the map concepts with the information so that the CM does not lose its meaning. It is very difficult to do in both personal libraries and in the Internet especially when the educational information is not is not well classified. When the map is exported to a web site making it visible to the browsers, it loses its map structure and the possibility to reuse it (completely or just a part of it) in generating a new learning unit. That is why we propose a model for professors and designers, which combine the design of a concept map with the opportunities that learning objects offer to guarantee that they can be searched as well as their portability. Our model is based on the possibility of applying a T transformation to a traditional concept map to make it a more powerful and compatible one in different environments enriched with the facilities that specifications offer.

72

It is supposed that the users have one or more concept maps created by any of the available tools and that they can access the repositories of learning objects (LO), so that one can establish a correspondence between each node or concept of the CM with a learning object. Each link is considered a condition to fulfil the IMS-LD(Fig.4).

services, internet Orienal Conceptual Map

Interoperable Conceptual Map

Learning Object

Figure. 4.Relationship between a Concept Map and IMS-LD specifications

As observed in the figure, during the transformation, a concept can be associated to documents in different formats that the designer might have, obtained from the internet as traditionally done, or taken from a repository of learning objects for managing the information for this concept in the most efficient way. The learning object (LO) can be simple or a learning structure compatible with the specifications of the design we just described. There are XML search structures and algorithm procedures that define a learning object; in addition there are alternatives, which can be found to achieve specific levels of intelligence in the managers of LO in a way that they can create new learning objects from the information existing in the repository. We are working on the creation of some tools and services that automate the management of a repository of LO and we will conduct research to apply the concept of Ontologies in learning management basing our work on the theory of learning objects, which could be very useful in this process. Nodes can make reference to services such as CHAT and email or they may ask for a specific complex computer task. They can be given properties to characterize them or the concept, as well as to define indicators that can be used as conditions or restrictions in the transition from one node to another. Applications based on the previous model can be designed, its input could be the concept map and the output could be the concept map enriched with the benefits of IMS-LD specifications. This guarantees its portability and power, as we have just stated.

73

We have created an application called COMALO (from Concept MAps to Learning Objects) that generates learning structures compatible with IMS-CP and IMS-LD design specifications in levels A and B. This application uses a repository of LO and an XML structure that represents the concept map that the user introduces through an interface that is close to his knowledge, without being obliged to know in detail the terminology used in the specifications. We are working on an application that permits the automatic conversion of some maps obtained from different tools in COMALO format, in a way that users can make minimum changes and incorporate the advantages of the specifications. COMALO permits searches in the repository and to associate LO to nodes. Besides, it guarantees the definition of properties and transitions subjected to restrictions. Although the application is still being developed, it is possible to use it in the generation of concept maps that meet IMS LD specifications in levels A and B. 3. Conclusions and Recommendations

In our paper we have proposed an alternative that can be the basis for teachers and s o h a r e developers in the design of applications for knowledge management, as there are the necessary conditions for implementing and putting into practice the recommendations we have made. In this way, concept maps, reusables in different educational environments can be produced. The combination of both theories will allow users to choose more precise, classified and structured information according to the complexity of the learning processes of the present time. As we have explained, the present tools that implement the design and content packaging specifications are still oriented towards the users that know them thoroughly; this hinders their development and causes rejection from other users, even including the most advanced. The use of concept maps as a basis for design will undoubtedly achieve better results. The effectiveness of the COMALO application will be evaluated throughout the design of Spanish courses using the repository of LO of the Spanish Virtual Classroom (AVE). It is necessary to work in the graphic interface of this application to offer better design facilities and to develop interpreters, which permit the use of IMS-LD generated structures specifically in levels €3 and C where few results have been seen so far. References 1.

2. 3.

UAantes G . , E-learning: cambiando paradigmas en capacitacidn. 2003, El principe.com. http://www.elprincipe.com/teleformacionhotas/index 14.shtml Estrada , V., Mapas Conceptuales. 1998, MES. Morales, R., Capacitacidn basada en objetos reusables de aprendizaje. 200 1 . http://www.umb.edu.co/umb/sitiopedagogia/lecturas/tendencias.pdf

74

4.

5.

6. 7.

8. 9.

10.

11.

12. 13. 14. 15. 16.

17. 18.

IMS Global Learning Consortium, IMS Learning Resource Meta-Data Specijkation: Version I . I Final Specification. 2000, IMS Global Learning Consortium. http://www.imsproject.org/metadatdindex.html ADL, Sharable Content Object Reference Model Version 1.2. 200 1 , Advanced Distributed Learning. http://www.adLnet.org LTSC, Draft Standard for Learning Object Metadata Version 6. I . 200 1, Learning Technology Standards Committee. http://ltsc.ieee.org/doc Willey ,D., Connecting learning objects to instructional design theory: A de3nition, a metaphor, and a taxonomy,. 2001. http://reusability.org/read/chapters/wiley.doc Instituto Cervantes , Pagina Web del Instituto Cewantes. 2004. http://www.cervantes.es/ Marco, M.y.Moron, D., Sistema de produccidn multimedia para la WEB, ejemplo aula virtual del espafiol. 2003: MIC, CUBA. http:/www.infollnaticahabana.co.cu/eventos/eventos/Educacion/default. htm Iriarte, L., Marco, M., Moron , D., Creacidn automatizada de una Biblioteca de Objetos de aprendizaje. 2003, Memorias del X CONGRESO INTERNACIONAL DE INFORMATICA EN LA EDUCACION, IMS Global Learning Consortium, IMS Content Packaging. Versiort 1.1.3 Final SpeclJication. 2003. http://www.imsglobal.org/content/packaging/cpv1p 1p3/imscp_infov1p 1p 3.html IMS Global Learning Consortium, IMS Simple Sequencing Information and Behavior Model. Version 1.0 Final Specification. 2003. http://www.imsglobal.org/simplesequencing/ssv lpO/imsss-infovl p0.html IMS Global Learning Consortium, IMS Learning Design Best Practice and Implementation Guide.Version I.0 Final Specification. 2003. http://www.imsglobaI.org/learningdesign/ldvl pO/imsld-bestv 1pO.html OpenUniversiteItNetherlands, Educational Modelling Language. 2004.

http://eml.ou.nVeml-ou-nl.htm Britain, S., A Review of Learning Design: Concept, Specifications and Tools. 2004, A report for the JISC E-learning Pedagogy Programme. http://www.jisc.ac.uk/uploaded_documents/ReviewLeamingDesi~.doc Novak, J.D., Aprendiendo a Aprender. 1988: Ediciones Martinez Roca, S.A. Gaines, B.e.S., Collaboration Through Conceptt Maps. 1995. Olivier, B., IMS Learning Design Public Draft .An Overview. 2003, CETIS. http://www.brookes.ac.uk/research/odl/alt-n103~resen~tions/LD presentation.ppt

THE CONSTRUCTION OF WEB-BASED MASTERY LEARNING SYSTEM* HSIEN TANG LIN Department of Computer and Information Science, National Chiao Tung University 1001 Ta Hsueh Rd., Hsinchu, Taiwan 30050, R. 0. C

ZHI FENG LIU Graduate Institute of Learning and Instruction, National Central University No.300, Jungda Rd, Jhongli City, Taoyuan, Taiwan 320, R.0. C SHYAN-MING YUAN Department of Computer and Information Science, National Chiao Tung University 1001 Ta Hsueh Rd., Hsinchu, Taiwan 30050, R.0.C For the learning of consecutive courses, student’s learning outcome may be affected by learning performance of previous course. In other words, if student does not learn something well, say integer addition in mathematics education, then poor learning performance of following consecutive course, say integer multiplication can be expected. Bloom advocated concept of mastery learning in 1960s [l]. Bloom claimed that giving sufficient learning time and good lecturing quality, almost every student learns everything taught by teacher well. In this manner, if student gets mastery at every learning unit, then it may solve problem stated above. The very important part of mastery learning is remedial learning. In remedial learning, teaching content will be tailored to meet each individual student’s need. It will put a lot of burden on teacher if students who need remedial learning are not a few. In this paper, a web-based mastery learning system is introduced to make mastery learning process more practical and more effective. An experimental study has been done on an Institute of Technology in Taiwan. The result showed students appreciated this system, and almost all students got mastery based on predefined criteria

1. Introduction

The progression in information technology benefits almost every domain. Incorporating information technology into education to strengthen teaching quality and student’s learning performance has been a hot issue. Among possible models of information technology in education, web-based model is most popular. The characteristics of web-based learning are: 1:) studentcentered learning, 2:) promoting active learning, 3:) beyond time and space boundary, 4:)providing multi-dimensional learning environment. The Internet has penetrated every corner to the extent that anything without a network capability, information technology (IT) in particular, will be thought of as outof-date. IT provides easy access to information and resources anytime and anywhere across time and geography boundaries. A popular pedagogical topic

*

This work is partially supported by grant of Science Education Department of the National Science Council of the R.O.C. under NSC 93-2520-S-009-005 project.

75

76

of recent discussion has been how to reinforce the quality and effectiveness of teaching and learning by incorporating IT. In school education, especially in higher education, three different models for such incorporation have been found appropriate: complementary, mixed, and complete. With the first approach, IT has only a subordinate lecturing role that complements that of the traditional classroom format. With the second, IT shares a more or less equal role with the traditional format. With the third, the IT format entirely replaces the traditional format, as is found for example with the Open University. In this paper, a web-based mastery learning system is presented. This system integrates characteristics of network which make mastery learning more practical and effective. Bloom advocated mastery learning in 1960s. It has become a famous and widely using learning approach since that. In traditional manner, teacher should spend a lot of efforts to conduct remedial learning which is core of mastery learning. To let student gains most, remedial learning should be tailor-made to address each individual student’s need. It is difficult to achieve in traditional way, but is easily in web-based approach. In the following sections, the background and literature review of mastery learning will be discussed, and then is description of web-based mastery learning system, finally experimental result of system is presented.

2. Review of Related Literature 2.1. Mastery learning

The general concept about teaching and learning had radical change after Carroll [2] advocated a new teaching concept. His teaching concept suggested that teaching should focus on individual student needs different time to learn same material. This suggestion is in contradictory with traditional model which allocating same period of time to all students to learn same material. In fact, Carroll [3] claimed aptitude is a major measurement of learning time. He used a formula called LR (Learning Rate) to stand for degree of learning: LR = f( time spent learning / time need to learn)

Carroll’s viewpoint based on all learners have potential to learn anything well, but require different period of time to achieve. When taking learner’s aptitude as content of learning rate, student is not longer a good or bad learner, but a faster or slower learner [4].Carroll also pointed out two factors which affect student’s learning rate: student’s perseverance and chance to learn. The former factor is controlled by students themselves, in other words, how much time they spend on learning. The later is how long or how much learning material teacher allocates for student to learn in classroom or after class. The concept of mastery learning was advocated by Benjamin Bloom [l] [ 5 ] . When Bloom participated a research called “effectiveness of individual

77

difference to learning”, he was deeply affected by Carroll’s concept. He gave a further elaboration: 1:) because aptitude might be used to forecast learner’s learning rate, a predefined learning content might be set 2:) therefore, attention should be paid on teaching variables controlled by teacher, such as opportunity of learning and quality of teaching, 3:) finally, teacher should guarantee every student can achieve predefined course objective. In summary, Bloom claimed giving sufficient time and good quality of teaching, almost every student learns well. The features of master learning are: 0 pointing out what to learn and how to assess, 0 allowing student learns in his or her own pace, 0 assessing student’s progression, and providing correct feedback or remediation, evaluating whether student achieve final learning criterion Mastery learning theorem makes teacher’s teaching responsibility a radical change. Student’s fail in learning should blame on teacher’s teaching but not on student’s deficiency of ability. In this kind of learning environment, the challenge of teaching becomes to provide sufficient learning time and suitable teaching strategy. In this way, all students can achieve same degree of leaning [5] [6]. Mastery learning has been widely applying on school teaching and training, the research results showed that this approach improves effectiveness of teaching [7] [8]. In other hand, mastery learning has theoretical and practical deficiency. People indeed have difference on ability, and are prone to different achievement. Besides, the establishment of mastery learning needs a lot of time and effort; it prohibits teachers and school administrative to establish such learning system. Assessment in mastery learning is also a criterion-reference assessment. In criterion-reference assessment, student’s score is not for comparing with other students but with a presetting criterion. The objective of mastery learning requests all students achieve this criterion. Speaking in terms of mastery learning is mastering this subject. In literature, the criterion is 95% in the highest and 80% in the lowest.

2.2. Persuasive Technology Fogg 191 defined persuasive technology as any interactive computing system designed to change people’s attitudes or behaviors. The emergence of the Internet has led to a proliferation of web sites designed to persuade or motivate people to change their attitude and behavior. Web sites are the most common form of persuasive technology today. With regarding to education issue, computing system, especially web-based system, provides tremendous advantages on persuading student to learn. By incorporating simulation or multimedia content into learning material student gets easily understanding. The earliest signs of persuasive technology appeared in the 1970s, when a few computing systems were designed to promote health and increase workplace productivity. One of the earliest examples is a computer system named Body Awaremess Resource Network (BARN), developed in the late 1970s [9]. This

78

pioneering program was designed to teach adolescents about health issues such as smoking, drugs, exercise, and more, with an ultimate focus on enhancing teens’ behaviors in these areas. Fogg [9] stated six distinct advantages computing system over human persuaders: Be more persistent than human beings Offer greater anonymity Manage huge volumes of data Use many modalities to influence Scale easily Go where humans cannot go or may not be welcome As stating in previous section, Carroll pointed out one of two factors which affect student’s learning rate is student’s perseverance. In web-based mastery learning system, student’s attitude or behavior may be changed or affected toward allocating more time on learning. Then student’s perseverance may be enhanced. 3. System Design

With the above discussion, it is obviously mastery learning is an effective and practical teaching approach. To relieve teacher’s burden and provide sufficient and convenient environment to student, a web-based mastery learning system is presented in this paper. In his system, the process is a cyclical approach which consists of unit lecturing, formative assessment, remedial learning, and advanced learning. In traditional mastery learning process, remedial learning usually takes about 2 times. If any student does not achieve mastery level after second remedial learning, the process is terminated. With the capability of networked system, remedial learning can repeat as many times as it needs. The reason is remedial learning is done by networked system instead of teacher. One of many advantages about Networked system over traditional approach is networked system is more persistent than human beings [ 9 ] . The remedial learning may repeat as long as student does not achieve mastery level. The approach also realizes what Carroll mentioned student’s chance of perseverance and learning. The chance of students’ perseverance and learning can be enhanced through this system; therefore they can achieve course objectives. In this approach, the main purpose of networked technology is to let learning process proceeds smoothly. In traditional classroom teaching activities, there is time limitation. It is hard to allocate suitable time and location for student to interact with teacher or peers after class. With the aids of networked and information technology, student can participate remedial learning beyond time and location boundary. Remedial learning has two dimensions. One is learning activity and another is formative assessment. Formative assessment not only assesses student’s mastery level, but also diagnoses student’s weak point and insufficient part. Therefore, formative and diagnostic aspects are taken into consideration while designing

79

test items. By way of test items on diagnostic assessment, and inspecting student’s answers, then student’s learning problem can be revealed. Once identifying student’s learning problem, suitable or appropriate remedial material can be presented to each individual student. Each test item of diagnostic assessment addresses an atomic concept. In this manner, student’s cognitive process in learning the subject domain can be easily identified. Taking common fraction problem in basic mathematics domain as example, its solving process consists of several atomic concepts. In order to correctly solve this problem, the student should have knowledge of these atomic concepts. If there is misunderstanding on any atomic concept, then it will lead to incorrect answer. While designing test item of diagnostic assessment, teacher identifies all atomic concepts of a specific problem domain, then design a test item for each atomic concept. If student can correctly answer this item, that means he/she has understood it. Furthermore, if student correctly answer all test items relate to a specific problem domain, that means helshe has achieve mastery level for this learning unit or domain. Formative assessment can be in traditional way or webbased. System presented in this paper adopts a mixed model [ 101. In this model, activity such as learning or assessment may be traditional way or web-based relies on practical consideration. If teacher prefers traditional way, then assessment is in paper and pencil form. After collecting and grading student’s answer, if any student does not achieve mastery level, teacher will select suitable remedial material on the system, and asks student to take learning. The preferable way is in web-based form. Teacher puts all test items into system. At the end of each lecturing unit, students are requested to take formative assessment on the web. System will automatically select remedial learning material for students, if they do not achieve mastery level. No matter formative assessment is in traditional form or web-based form, further formative assessment after remedial learning is on the web. Test items for any formative learning after remedial learning are randomly selected by system. Each item has accompanied with a hint or further explanation. If student’s answer is incorrect, system will present the related hint to student. This will make student gain immediately assistant. 4. Experiment Design

In order to evaluate effectiveness and functionality of this system, an experimental has been conducted on an Institute of Technology located in northern Taiwan. The experimental course was principle of electrical circuitry. Students attended this class were junior and fail on previous study. This kind of classes sometimes is called summer school which means open in summer vacation. Totally, 24 students attended this class in which one was female and

80

others were male. Because of these students had studied same subject but fail on final score, it’s more suitable to introduce mastery learning into this class. As per definition of mastery learning, first step is to determine what student should learn. Then, divide learning material into several lecturing units. In this experiment, it is divided into 12 units. For each unit, teacher prepared lecturing material, test items of formative assessment, remedial learning material, and hint for every test item. Normal lecturing was on classroom, assessment was webbased form, and further as well as remedial learning was also through webbased system. Each unit started with unit teaching. At the end of teaching, student was requested to perform web-based formative and diagnostic assessment. It this experiment, mastery level was defined as correctly answering 85% of test items. If student achieved this level, he/she has mastered this unit, and was encouraged to take further advanced study. In contrast, if student did not achieve this level, he/she was requested to attend remedial learning. The remedial learning material was not exactly same as unit lecturing material. The reasons are (1) in different lecturing approach (traditional and web-based), the lecturing material should be in different form to let student learned most, (2) because of student did not accomplish last learning, the learning material should be amended, (3) the remedial learning material should focus on what student did not understand, in other word did not answer correctly. As mentioned above, diagnostic assessment is part of formative assessment, student’s weak point or misunderstanding can be identified through test items. It is teacher’s responsibility to clarify a test item should link with which course concept, and what remedial material should be presented if student gave incorrect answer. System presented in this paper provides a convenient and systematic way to let teacher accomplishes this task. Student who did not get mastery, should arrange and allocate time to attend remedial lecturing after class. Student could attend formative assessment after remedial learning. If student still did not get mastery, the above remedial process should be repeated until student gets mastery. System provides another two features to assist student perform remedial learning. Student can discussed with classmates through discussion board or contacted teacher while encountering difficulty on remedial learning. To encourage students who have gotten mastery share acquaintance with classmate, system will give extra score to these students. For some students who want to apply scholarship or get high rank, this strategy will provide motivation. With regarding to contact teacher, this system provide a more effective way. Student raised problem on the system, and system will deliver this message with predefined channel such as email and SMS (Short Message Service), etc. set by teacher. The purpose is to let teacher gets this message and provides assistance to student as soon as possible. Teacher can set a learning deadline on this

81

system. If there is any student does not get mastery over this deadline, system will automatically inform teacher. Then, teacher may contact student to provide necessary assistance or teaching. 5. Result and Discussion At the end of course, a satisfactory survey was conducted. In summary, about 94% of students claimed this system has positive effect on their learning. 95% of students claimed this system changed their learning attitude toward more positive. These results agreed on Fogg’s viewpoint that computing system may be used as a persuasive technology to change people’s attitude. With regarding to provide instant and specific hint to each item student gave wrong answer, about 91% of students thought it gave great help. This is a strength part of web-based mastery learning. In traditional approach, it is hard to accomplish especially if such students are not a few. All of students spent more time on learning. The average time increased on learning is about 2.5 hours per week. These extra hours do not squeeze learning time of other courses. This is another evidence that student’s attitude has been changed and willing to spend more time on learning. At the end of course, not all 24 students got mastery, two of them fail in final score. The possible reasons are as follows. (1) This was a s u e r school class, students were fail at last semester, that means some of them may have serious learning problem which may not be cured by mastery learning. (2) As stated previously, people indeed have difference on ability, and are prone to different achievement. (3) Although computing system can be a persuasive tool, but not all students can be persuaded. Persuasive effect of this system may not work for some of students. Several issues need to be pointed out. Whether student gets mastery is checked by formative assessment. Besides, if student does not get mastery, then should attend remedial learning. Therefore, formative assessment should have functionality of diagnostic assessment. How to design a suitable and adequate test items to diagnose student’s real learning difficulty, so that suitable remedial learning material can be presented to student is an important issue. For each subject, domain experts or teachers may form a task force to design lecturing units, test items and remedial learning material. Secondly, the modality of learning material on the web may affect student’s learning performance. Fogg’s study has confirmed this point [9]. In Fogg’s research, attractive material is more persuasive than unattractive one. This is a critical issue on network- or web-based learning system. Same as first point, if teacher or domain expert can spend more effort to design learning material, it may be more effectiveness for student’s learning. Finally, teacher may aware student’s learning situation in face to face lecturing circumstance, and necessary intervention or assistance may be applied instantly. In this manner, student may follow teacher’s lecturing step as close as possible. In web-based learning circumstance, it lacks in such facility or channel. How to

82 provide a mechanism to aware students learning status on the web is real time manner, it is an issue need to be addressed.

References 1. Bloom, B. S. Learning for mastery. Evaluation Comment, 1 (2) 1-5 1968 2. Carroll, J. B. A model of school learning. Teacher College /Record, 64, 723-

733. 1963 3. Carroll, J. B. The Carroll model: A 25 year retrospective and prospective view, Educational Researcher, 18(1) 26-31 1989

4. Guskey, T. R. Implementing Mastery Learning (2nd ed.). Wadesworth

Publishing 1997

5. Bloom, B. S. All Our Children Learning New York McGraw-Hill 1981 6. Levine, D. Improving Student Achievement Through Mastery Learning Programs San Francisco: Jossey-Bass. 1985

7. Block, J. H. Efthim, H. E., & Burns R. B. Building Effective Mastery Learning Schools. New York: Longman. 1989 8. Slavin, R. E. Mastery learning reconsidered. Review of Educational Research,

57(2), 175-214. 1987 9. Fogg, B. J. Persuasive Techology: Using Computers to Change What WE Think and Do. CA: Morgan Kaufmann 2003 10. Lin, H. T., Kuo, T. H. & Yuan, H. M. A. Web-based Customized Instuction Strategy for Engineering Courses. In R. Cheung R. Lau, Q. Li (Ed.) New Horizon in Web-based Learning (pp 61-70), World Scientific Publishing. 2004

ENHANCING PROBLEM-BASED LEARNING BY ELEARNING: A STUDY WITH THE TEACHING OF DATA STRUCTURES AND ALGORITHMS KENT K. T. CHEUNG Department of Computing, Hong Kong Polytechnic University Hung Hom, HKSAR ALAN Y. K. CHAN AND K. 0. CHOW Department of Computer Science, City University of Hong Kong Kowloon Tong, HKSAR

Many science subjects involve concepts that are abstract and difficult. In addition to explaining and visualizing the concepts, there is a need to teach the students when and how to apply the concepts to real applications. This paper discusses a study of augmenting the learning process with a web-based quiz system based on the ProblemBased Learning strategy (PBL). With a PBL approach, it is possible to teach difficult concepts and theory in lectures and lead the students to discover how to apply the theory in practice by solving the given problems. A deployment of the system has been applied to a course on data structures and algorithms, which is a difficult subject of computer science. Feedback from the students after the course indicates that the proposed system and teaching method is able to enhance the learning process by clarifying difficult concepts and arouse interests of the students on the subject.

1.

Introduction

Many science subjects involve concepts that are abstract and difficult. Instructors thus tend to pay more attention to clearly explain and illustrate the concepts. However, it is also important to emphasize on students’ applying the concepts to solve problems in real situations. Difficult concepts that are hard to understand and apply are pervasive in many subjects such as computer science and medical science. The data structure and algorithm design is an example of this class of subjects in computer science. This is a difficult subject partly due to the difficult mathematics involved and partly due to the difficulty of applying the theory. Moreover, application of the concepts and theories is also difficult but is often overlooked. The paper investigates a deployment of a web-based quiz system that is designed to augment the teaching process with the Problem-Based Learning (PBL) [ 11 strategy. 83

84

Problem-based learning has been widely applied to many scientific subjects such as medicine[2] and physiology [3]. As the name implies, PBL is a learning strategy that leads students to learn through a set of problems related to a particular topic that is being taught. As a tool that facilitates PBL, the web-based quiz system is designed to enhance the effectiveness of communications between instructor and students. As will be discussed in the next section, the proposed system not only serves to distribute the problems, but is also responsible for helping instructor to give feedback to students. By posting the feedback and all feasible solutions on the web, students are able to learn from the feedback and the cited suggestions from other students. The system has also incorporated games by design so that students’ interests on the subject may be aroused. The paper is organized as follows. Section 2 describes the web-based quiz system, called Master Solvers’ Club (MSC) in details. A discussion of the design objectives and the way that these objectives are satisfied is also given. Section 3 presents an analysis of MSC in terms of a survey conducted after the course. Finally, the paper conclusion is given in section 4.

2.

Master Solvers’ Club (MSC)

In this section, an overview of MSC is discussed. Instead of providing the implementation details of MSC, which are straightforward, we discuss the design objectives in this section. This is followed by a discussion of how these objectives are achieved in MSC.

2.1. Overview In the course that MSC was applied, six set of questions with each composing of 5 questions were given. Each question described a practical scenario and asked the students to make a software design decision. Usually, only a few feasible options might be identified using the theory taught within the course. Students were asked to submit their decisions, and the comments or explanations about their decisions. The comments ftom the students helped to reveal how students understood the problem and the related theory behind the problem. Instead of asking students to read answers and comments from all participants, the instructor was asked to give feedback to students by citing representative comments or answers from students. In this way, the instructor was allowed to summarize common thoughts from students and explained their underlying mistakes or misconceptions.

In addition, the instructor was also responsible to classify students’ answer. In many cases, only a few feasible options are available per question. Although students did not produce identical answers, it was a simple matter to classify the answers into one of the few feasible options. The instructor then gave each option a mark, from the lowest possible of 0% to the maximum of loo%, and depending on how well the decision can solve the given problem. This mark assessed how well a student had solved a question. After all sets of questions had been attempted, the total mark scored by a student might be used as an assessment of the performance of the student on the subject. In addition to be a facility of distributing questions and feedback fiom the instructor, MSC was also designed to be a game in an attempt to arouse students’ interests on the subject. MSC was actually stimulated by a competition published every month in a magazine [4]about the game of bridge. Thus, the student answers were rated by the instructor and a hall of fame was published to praise the best performing students. All students were allowed to read the feasible options and the ratings. Students might verify the classification of his/ her answer since all classifications were shown after the list of all feasible options. In addition, a hall of fame page was also created to show the scores of leading scorers to encourage the students to participate. MSC was designed as a web-based system so that students may access the system any time, anywhere. Being a web-based system, the user interface is nothing more than entry forms for students to submit their answers or web pages for instructor to present the questions and results to students. XML was chosen to store the data of MSC in part due to its flexibility and in part for its less demanding expertise. A XML file is editable by any text editor and is portable across platforms. It is therefore easy for an instructor to manipulate the data stored in the file.

2.2. Design objectives

As discussed earlier, MSC was designed as a tool for enhancing the communication between students and instructor in a PBL learning process. Specifically, MSC was designed to let students: 0 practice their judgments on real scenarios

86

learn how to apportion the importance of conflicting factors learn to identifjl alternatives when facing a problem reveal their misunderstandingsof the theory 0 stimulate the interests in the subject 0 assess how well they perform in the study The above learning objectives are incorporated in the design of MSC. In what follows, we discuss each objective in details and the ways we achieve these objectives. 0

2.2.1. Practice students 'judgments on real scenarios As discussed in the last section, many CS courses were traditionally taught by considering toy problems only. In many cases, this approach is not sufficient because students lack the experience for tackling actual situations, where the best decision is a tradeoff between many conflicting factors. Case study is widely used in many subjects that require practical experience. Unfortunately, the highly abstract and theoretical nature of many CS subjects makes the approach less attractive. Naturally most of the attention must be paid to facilitate the learning of fundamental theory during lectures and tutorials. MSC was a web-based PBL tool that was designed for practical case studies. It is designed to augment the lack of practical experience gained in lectures and tutorials. While the lectures and tutorials continued to be dedicated to explaining the difficult theory, the practical aspects of the subject may be augmented by MSC.

2.2.2. Apportion the importance of conflictingfactors Many design decisions involves a tradeoff between conflicting factors such as speed and memory requirements. In some cases, the speed is more important, but sometimes a program becomes unusable if it requires too much memory. Some other factors should also be considered in other cases. In all, students need to (1) identify the deciding factors and (2) apportion the importance of these factors. Toy problems taught in lectures are usually designed to illustrate a theoretical problem and therefore they are too simple to make students aware of these two needs. Each question posted in MSC outlined an actual scenario and asked students to make a design decision according to the scenario. The question asked students to consider various alternatives and conflicting factors before making the final decision. While these questions might be very difficult for weak students, they

87

helped hard-working students to realize the aforementioned needs for the practical use of the theory taught in lectures.

2.2.3. Identi& alternatives whenfacing a problem When facing a problem, it is natural for some people to choose the first solution that they can think of. Unfortunately, this approach may lead to sub-optimal solution since a better solution may be found after carehl consideration. Sometimes, students are simply not aware of the possibility of alternative solutions, or in other cases students do not have enough knowledge or experience to find alternatives. By presenting to students with problems that involve many potential solutions, students have a chance to face real problems that may require a second thought in order to search for the best solution. MSC improved the learning process by revealing all feasible options to students after instructor had considered all answers given by students. In addition, by discussing the answers and comments made by students, instructor was able to explain why an option was feasible as well as the pros and cons of the various options in response to the actual problems of the students. In this way, students are presented with alternative solutions from instructor and other students. Students may slowly gain experience of identifying alternative solutions after reading the feedback fiom instructor.

2.2.4. Reveal mis-understandings It is important for instructor to identify the areas of weaknesses and misunderstandings of students. Since MSC asked students to make comments and explanations about their own answers, it is possible for instructor to identify the areas of weaknesses and misunderstandings.

2.2.5. Stimulate interests The subject for which MSC was designed was a difficult subject to many students. Difficult subjects tend to be boring from the student viewpoint. MSC was designed in the form of a game so as to stimulate students’ interest. Also, a

88

hall of fame was posted on the system to praise the students who did well in the game. This also helped to encourage students to work hard in the subject.

2.2.6. Assess the performance of students MSC was designed to be a PBL tool with many tough questions posted on the system. As a result, it can also be treated as an assessment tool. In general, it is possible for instructor to assess how well students understand and apply the theory from the comments given.

3. Students’ Feedback MSC was applied to a class of students who enrolled in the subject of data structures and algorithms. As a preliminary study of the proposed teaching tool, a small-scale questionnaire survey was carried out after the course had ended. In the survey, the students were asked to give their feedback about the system. Table 1 summarizes the survey questions in the questionnaire.

Table 1 - Survey questions

1 2 3 4 5

6 7

Relevance of questions (5=most relevant, l=least relevant) difficulty of questions (5=most difficult, l=easiest) Clarity of feedback (5=too long, l=too brief) feedback was responsive to students misunderstanding (5=strongly agree, 1=strongly disagree) activity stimulates interest (5=strongly agree, l=strongly disagree) enhance understanding (5=strongly agree, l=strongly disagree) web-based Q/A system was user-friendly (5=strongly agree, l=strongly disagree)

There were 68 students in the class. 55 filled questionnaires were received to give a response rate of 80.9%. A separate bar chart was created for the student feedback by question as shown in Figure 1. For all questions except question 2, a response of 5 indicates a very positive response. A quick glance reveals that options 3 and 4 are the most popular responses. The responses fiom the students were generally positive. A more careful analysis of the feedback for each question follows. It can be seen that most students found the questions to be relevant from the left-most bar chart. The feedback of question 2 indicates that the questions were quite difficult for the students. Nevertheless, the third bar chart shows that the students were generally satisfied about the clarity of feedback from the instructor. The feedback from the instructor helped to clarify the misunderstanding of the students as shown by the bar chart of question 4.Also,

89

students thought that the activity helped to enhance their understanding of the subject from the bar chart of question 6. As mentioned above, one of the objectives of MSC was to arouse interests of the students. The feedback of question 5 shows that this objective has been achieved. Finally, the userinterface of the web-based system was satisfactory to the students as indicated by the last bar chart.

MSC Questionnaire 40 35 30

225

g B

Ls

20 15

10 5

0

1

2

3

4

5

6

7

Questions Figure 1 - Feedback from students

At the end of the questionnaire, students were asked to give a rating of the general quality of MSC. As shown in Figure 2, most students found that the activity had a good quality. There were fewer than 5 students who disliked the activity.

90

Oveall rating

Excellent

VeryGood

Good

Acc-le

Poor

VeiyPoor

VeiyVeiy Poor

Rating

Figure 2 -Overall rating from student

As a summary, the feedback from students shows that MSC was successful in helping them to understand more about the subject. Misconceptions were also clarified, thanks to the feedback 6-om the instructor. The activity was also able to make the subject more interesting according to the feedback from the student. Most students thought that the activity was very good overall. We are also pleased to note that there were only a few students who did not think the activity was good.

4. Conclusion

We have discussed a study of an online teaching aid, Master Solvers’ Club. This study has indicated that it is possible to supplement theoretical teaching in lecture with practical exercises via online and open-ended quizzes. Feedback from students indicates that MSC is able to stimulate interests from students and enhance the understanding of the subject. Overall, students thought that MSC was a very good teaching aid. Therefore, MSC is a teaching aid worth investigating for the subject in the future. Due to its close relationship with the PBL approach, MSC may also be adapted to any subject that is suitable to be taught by PBL. We regard the reported study as being at a preliminary stage, and the current implementation of MSC is still primitive. In the near future, we aim to provide more functionality for instructors to prepare feedback. For example, it is difficult for an instructor or a student to type a formula or include a diagram. A specially

91

designed editor might be a valuable tool for MSC to be successful. Also, a software framework or design pattern may also be considered when trying MSC in other subjects.

References M. A. Albanese and S. Mitchell, “Problem-based Learning: a Review of Literature on its Outcomes and Implementation Issues”, Acad. Med., 68( l), 1993, 52-81. H. S. Barrows and R. M. Tamblyn, “Problem-based Learning: an 2. Approach to Medical Education”, New York: Springer, 1980. 3. S. Mierson, “A Problem-based Learning Course in Physiology for Undergraduate and Graduate Basic Students Students”, Am. J. Physiol., 275 (Adv. Physiol. Educ. 20): S16-S27, 1998. 4. The Bridge World magazine, online information available: www.bridgeworld.com.4. V. Baran, M. Colonna, M. Di Tor0 and V. Greco, Phys. Rev. Lett. 86,4492 (2001). 1.


HOW SHOULD ONLINE TUTORS BE TRAINED? A FOUR-LEVEL EVALUATION FOR E-MODERATING PROGRAMME EVA TSANG The Open Universityof Hong Kong

Abstract This paper investigates the evaluation of Online Training for Tutors (OTT) Programme by using the Four-level evaluation model. The O’IT online training programme is to train online tutors to improve their knowledge of online teaching and learning, to use computer mediated communication (CMC) tools such as email and discussion boards. The most important is to develop their skills for moderating online conferences. The paper covers four evaluation levels including reaction, learning, behavior and results. In level one, it measures participants’ reaction to the training program. In level two, it measures their e-moderating skills and knowledge. In level three, it measures changes in behavior on the job as a result of training. In level four, it measures the final results that occurs in this training programme.

Introduction The Open University of Hong Kong (OUHK) formally launched its first online courses in the medium of English in 1999, and Chinese online courses in 2000 via the online learning platform. Including courses in both languages of instruction, the OUHK has delivered more than 250 courses online since 2000. However, the communication tools are not very much used on most of them. According to the users’ statistics, it reports that the use of discussion boards is limited. A study of OUHK tutors’ participation in online discussion (Tsang et al, 2002) reports that more than half of tutors logged in fewer than 64 times, which is even fewer than the average of the students. There is a high correlation between tutors’ and students’ participation in online discussion. The more tutors are actively involved in the online discussion, the more their students will participate. It is believed that there are several reasons behind. Mainly the role of tutors is not clearly identified when they provide online tutoring services. However, the part-time tutors at the OUHK most likely have little knowledge of online learning or e-moderation. This parallels the above findings conducted by Shin (2002). In the study, the OUHK tutors were interviewed and perceived the online discussion was challenges to them. The 93

94

challenges mainly involved time management, struggle for students’ expectations of receiving speedy responses, and how to handle the discussion in different level of participation. Obviously, the role of tutors now shifts from the course materials to the online moderated discussion. The role of the tutor will change especially when handling the discussion with e-moderating skills. Therefore, the University provides appropriate support for this change process. How should such online tutors be trained? What should be the objectives of a training course for online tutors and how should we assess and certify them? In order to improve the tutorial training, tutors are required to have a whole set of skills to perform better in the discussion boards. Goodyear et a1 (2001) listed the complete set of competencies that are required for the online tutors to enhance the teaching and learning. While many of the skills that tutors use in traditional tutorials can be transferred to online tutoring, successhl online interaction with students also requires a unique range of social, technical, managerial and pedagogical competencies (Berge, 1995). Theoretical Basis The transformation of tutoring services from face-to-face instructor to online tutor/facilitator is now being to be considered. Gustafson and Gibbs (2000) state explicitly that teaching in an online environment involves far more than simply transferring teaching skills from the classroom. They also illustrate that the successful facilitator will need to learn strategies for developing online ‘antennae”, for humanizing the electronic environment, and new ways to guide students to discuss, critique and reflect together as they engage in the construction of meaning. There are skills and techniques that can help e-moderators to carry their job more effectively (Salmon, 2000). Clearly, these types of strategies are not generally to be found in the repertoire of the average face-to-face tutors, nor would they need to be. However, most of tutors are not really aware of the importance of the online tutoring and do not care about the skills on leading discussions and answering questions on the discussion board. They even may not know the roles of the facilitator for online learning. Tutors’ characteristics are also considered to be related to students’ participation in online discussion. Vrasidas and McIsaac (1999) argues that tutors’ philosophy is an influential factor towards students’ participation. Furthermore, the content of instruction is also believed to be an important factor that determines students’ participation (Tolmie and

95

Boyle, 2000). Discussion and sharing of experience are considered to be more important in courses of some subjects or disciplines. Compared with the roles of conventional face-to-face tutors, the roles involved in online teaching cover widely aspects. Carlson (1989) defines a moderator as one who helps people get started, give them feedback, summarize, weave the contributions of different folks together, get it unstuck when necessary, deal with individuals who are disruptive, or get off the track, bring in a new material to freshen it up periodically, and get feedback from the group on how things are going and what happen from the group on how things are going and what might happen next. Paulsen (1995) notes that moderators, most likely played by tutors, in an online course are expected to play an “organizational role”, a “social role” and an “intellectual role”. Collins and Berge (1996) identifies that online tutors are firefighter, administrator, participant, facilitator, promoter and helper. Harasim et a1 (1997) states that online tutors play the roles of planner, group structurer, facilitator and guide. Rheingold (1998) describes an online moderator as a ‘host’, an ‘exemplar’ and a ‘cybrarian’. Schweizer (1999) discusses “facilitator”, “discussion leader-discussion creator”, and “manager” as essential roles to be played by those designing and teaching an online course. Salmon (2000) mentions that they perform information giving and receiving, development, knowledge construction, access and motivation and socialization. Goodyear (200 1) classifies the roles of online tutors including process facilitator, adviser-counselor, assessor, researcher, content facilitator, technologist, designer and manager-administrator. Given the centrality of tutor roles in guiding distance learners, the above literature on online teaching and learning also puts a great emphasis on tutor training and skill development. In order to support online learners effectively, online tutors need to experience online learning as a student before. Therefore, offering the online tutoring is necessary. The roles of online tutors should be identified. Clearly the most important role of the online tutor is to model effective teaching and accept “the responsibility of keeping discussion track, contributing special knowledge and insights, weaving together various discussion threads and course components, and maintaining group harmony” (Rohfeld & Wiemstra, 1995). In fact, skilled e-moderating is crucial to successful online learning (Salmon, 2000). Salmon (2000) identifies five-step model for this kind of training. They include: access and motivation, online socialization, information exchange, knowledge construction and development. In order to provide the training course to enhance the e-moderating

96

skills of the online tutors, the training programme was designed and put emphasis on leading discussions and answering questions on the discussion board. The training course has been offered since 2003. The tutors working for each presentation were invited to take this course.The focus of the training course " Online Tutor Training (OTT)" is on developing tutor's e-moderating skills, and feedback to skill-based e-moderating activities will be delivered through a moderated module discussion board. Furthermore, the course also provides for online tutors to develop skills and understanding in the design and use of online learning technologies. Learning takes place entirely online. Central to the course is the conference environment, in which participants discussed and practised e-moderating skills, and there is also a web-based course. This online course, which is part of the OUHK tutor orientation training programme, deals with online teaching and learning at the OUHK. This course not only provides the sound self-instructional strategies for designing distance learning materials but also offers the learning environment in which interactive learning can take place. The learning objectives of completing this online training for tutors (OTT) course, tutors will:

+

improve their knowledge of online teaching and learning;

+

log on to the OUHK's electronic learning platform, the Online Learning Environment (OLE);

+

find out more about online learning at the OUHK

+

use computer mediated communication (CMC) tools such as email and discussion boards;

+

develop skills for moderating online conferences with their online tutor groups;

+

understand the roles of online tutors.

The objective of the training course is to help the participants (tutors) to enhance their skills during the process of redefining and developing the crucial role of the online tutors in web-based learning. Not only tutors provide coaching and scaffolding support in structuring and managing a conference

97

activity, but also master e-moderating skills in weaving and summarizing the discussion. There are two versions, a Chinese and an English version, and these are offered to tutors e-moderating Chinese and English online courses respectively. Those that e-moderate both English and Chinese online courses have the option of choosing either the English or Chinese version. Each conference will have a moderator. Instructional designers will also play the role of e-moderators. The moderator’s job includes: (a) create a welcoming and encouraging conference environment; (b) ensure that everyone has access to the conference; (c) stimulate and manage discussion; (d) facilitate opportunities for exploring ideas; (d) provide feedback and resources.

Method The present study evaluates the satisfaction and learning effectiveness of learners’ participation in this course. While many of the skills that tutors use in traditional tutorials can be transferred to online tutoring, successful online interaction with students also requires a unique range of social, technical, managerial and pedagogical competencies. By enhancing tutor training in these aspects, tutors and students are encouraged to make more use of the online interactive tools through the OLE and improve the effectiveness of online learning. A crucial aspect of effective online learning may hold the key to the changing role of the tutors from face-to-face tutorial to online e-moderating skills. The evaluation is to verify if the participants (tutors) can enhance their skills during the process of redefining and developing the crucial role of the online tutors in web-based learning. The Kirkpatrick (1994) four-level evaluation provides a comprehensive system for evaluating effectiveness of the training program. It shaped the research questions and instrumentation developed and used for this multifaceted evaluation. The Kirkpatrick model consists of these four components: (1) Reaction-Are learners satisfied with the online tutor training? 2) Learning-What do learners learn in the training course? (3) Behavior-Do learners apply e-moderating skills and knowledge in their online courses? (4) Results-What are the final results that occurs? Level 1 Reaction

The first level of the evaluation is the reaction which measures

98

participants’ reaction or satisfaction with the training program. It also gauges the degree to which learners were satisfied with the way the training was conducted (Horton, 2001). It is a typical end-of course evaluation which can be sought through web-based surveys. The question “Were you satisfied with this OTT course in general?” seek feedback from participants about both training itself and their perception of the extent to which it will help them in their job. Level 2 Learning

This level of evaluations measure how much learners learned or achieved, According to Barksdale and Lund (2001), the purpose of Level 2 is to determine if the participants can obtain acquired knowledge, developed or enhanced skills, the change of mindset or now knows something he or she did not before the intervention. Since most of the participants are part-time tutors, they may not have a concept of e-moderating skills for online discussion. Therefore, it is quite difficult to identify what individual learning was in place before and what has taken place as a result of training. Questions are grouped into five main themes including the skills level, online learning system, course content, learning activities, study schedule and so forth. Learners have to rate the sections if they are usehl for them. There are three sections including online teaching and learning, emoderating and learning activities. In the section of online teaching and learning, it includes: the online learning environment, the meaning of electronic conference, and the role of an online tutor. In the second section, they include: starting up a conference, encouraging and managing discussion, formulating good questions, weaving and summarizing. In the third section, learning activities include: introducing ourselves, pros and cons of conferencing, using email for teaching and learning, welcome messages, encouraging and managing discussion. Level 3 Behavior

This level determines the impact of training on behavior, on-the-job performance, and the application of learned skill, knowledge, or attitudes. Horton (2001) called this level performance or application and it measures to what degree learners can and do apply learning in their context of their jobs. Since the participants are OUHK part-time online tutors, they get the just-in-time training and concurrently apply what they have trained and learned about the e-moderating skills into their online courses. A set of questionnaires to capture the views of participants (online tutors) of the behavioral changes that have occurred after training can be used. Furthermore, the direct on-the-job

99

observation of online tutors will be employed to see if they are applying the e-moderating skills. As it looks at the degree of change in behavior on the job that is due to the training, it takes several months to complete the evaluations after the training has taken place. It has to give the participants time and opportunity to develop their behavior. In fact, the feedback from stakeholders including students may be important. However, at this level, the participants (online tutors) should be allowed to have time for them to reflect and improve. The question “Will you adopt what you have learnt in OTT to the course you serve as a tutor?” is asked.

Level 4 Results The fourth level is to find out if the results of the training program achieved. It may measure increases in skills, quality improvement, productivity, return on investment and so forth. Kirkpatrick (1 994) identifies that this level is important for these are the reason for having the training program. Furthermore, it measures the business results of learning (Horton, 2001). The common practice for measuring results is using ROI which measures the financial rewards of an initial outlay. However, it is not really appropriate for the OTT training programme. Since the whole training programme is free of charge for participants, the trainers are also the in-house staff (most of them are instructional designers), working as part of their duty lists. Furthermore, the soft data like attitudes, new skills initiative and so forth is very difficult to measure and to convert monetary values. Most likely there are hard data including time, overhead costs, output and so forth. Here, increases in e-moderating skills is emphasied. The results are measured by examining tracking reports and written activity logs and comments.

Results A total of 252 online tutors were enrolled in the OUHK online tutor training course offered in March 2003, November 2003 and May 2004. Same course materials and requirements were presented in these three presentations. Each course lasted four weeks. During the course, tutors were required to participate into five online activities and submit an assignment. At the end of the course, an online evaluation, in which the questions were designed according to Kirkpatrick model, was posted onto the learning platform for tutors to complete. A hundred and eight evaluation forms were collected for a 43% response rate. The descriptive statistics and correlation coefficients among different levels are

100

shown in the following. Table 1: Descriptive statistics of variables under investigation ~

Items

Means

Standard Deviation

3.68’

0.75

OUHK learning environment (OLE) (L2a)

3.66’

0.74

Electronic conference (L2b)

3.57

0.70

Your role as an online tutor (L2c)

3.81

0.72

Starting up a conference (L2d)

3.63

0.89

Encouraging and managing discussion (L2e)

3.70

0.83

Formulating good questions (L2f)

3.65

0.75

Weaving and summarizing (L2g)

3.68

0.75

Planning a conference activity (L2h)

3.58

0.74

Structuring and managing a conference activity (L2i)

3.60

0.85

3.913

0.90

3.27

0.82

Level I

Satisfaction with OTT course in general (Ll)

Level 2

Level 3

Adopting learned skills to the respective courses (L3)

Level 4

Online tutoring skills increased (L4)

’ 1 represents “not at all satisfied” and 5, “very satisfied”;* 1 represents “not at all usehl” and 5, “very useful”; 1 represents “definitely not” and 5 , “definitely”.

101

As shown in Table 1, the respondents were satisfied with the course taken (mean=3.68). In level 2, all items were rated high in general. The skills learnt by respondent in the training course were quite useful, especially the role as an online tutor, encouraging and managing discussion, as well as weaving and summarizing. For level 3, the respondents reported a high intention (mean=3.91) to practice what they learnt in conducting online tutoring. For level 4, the respondents did recongize their emoderating skills have been increased. Conclusion Results generated by the four-level process offered a thorough look at measures of effectiveness for this training programme. The training programme enhances the emoderating skills of online tutors. It seemed to meet its objectives and provided useful just-in-time on job training for tutors. From the evaluation results, among the four levels, the overall satisfaction and feedback from tutors was generally very positive. To further investigate and make improvements in e-moderating programme, some qualitative research into this area should be conducted. References Barksdale, S. and Lund, T. (2001). RapidEvaluation. N Y ASTD. Berge, Z.L. (1995). Facilitating Computer Conferencing: Recommendations from the field. Educational Technology, 35( 1) pp22-30. Carlson, L. (1 989). Effective moderation of computer conferences: Hints for moderators. In Brochet, M.G. (ed), Moderating Conferences. Ontario: University of Guelph. Collins, M. and Berge, Z.L. (1 996). Facilitating Interaction in Computer Mediated Online Courses. Background paper for our presentation at the FSU/AECT Distance Education Conference, Tallahasee F.L., June, 1996. Retrieved 2 May 2005 from httr,://www.nib.unicamr,.br/recursos/distance education/flcc.html. Goodyear, P., Salmon, G, Spector, J.M., Steeples, C., & Tickner, S . (2001). Competencies for online teaching: A Special Report. Educational Technology Research and Development, vol. 49, no. 1, pp65-72. Gustafson, P., & Gibbs, D (2000). Guiding or hiding? The role of the facilitator on online teaching and learning. Teaching Education, vol. 11, no. 2, pp95-2 10. Harasim, L., hiktz, S.R, Teles, L. and Turoff, M. (1997). Learning Networks: A

102

field guide to teaching and learning online. Cambridge: MIT Press.

Horton, W. (200 1). Evaluating E-learning. N Y ASTD. Kirkpatrick, D (1994). Evaluating trainingprograms: Thefour levels. San Francisco: Berrtt-Koehler Publishers. Paulsen, M.F. (1995). Moderating discussions in the electronic classroom. In Zan, L. B and Mauri, P. C. (eds), Computer Mediated Communication and the Online Classroom. NJ: Hampton Press. Rheingold, H. (1998). The art of hosting good conversations online. Retrieved 2 May 2005 from http://www.emoderators.com/moderators/artonlinehost.html Rohfld, R. W & Hiemstra, R. (1 995). Moderating discussions in the electronic classroom. In Berge, Z. and Collons, M. Computer Mediated Communication ad the Online Classroom Vol. 3: Distance Learning. NJ: Hampton Press. Salmon, G (2000). E-moderating: The Key to Teaching and Learning Online. London: Kogan Page. Schweizer, H. (1999). Designing and Teaching an Online Course. Massachusetts: Person Education Company. Shin, N. (2002). Tutors’ Perceptions of Tutoring Online. Staff and Educational Development International, 6(2),ppl89-197. Tolmie, A. and Boyle, J. (2000). Factors influencing the success of CMC environments in university teaching: A review and case study. Computer & Education, vol. 34, ppll9-140.

Tsang, E., et al. (2002). Students’ participation in computer-mediated communication. In Murphy, D et a1 (eds) Advanced Online Learning in Asia. Hong Kong: Open University of Hong Kong Press. Vrasidas, C. and McIsaac, M. (1999). Factors influencing interaction in online course, American Journal of Distance Education, 13(3), pp22-36.

METHODOLGY FOR DEVELOPING DYNAMIC WEB AUTHORING SYSTEM FOR E-LEARNING YIN FEI YEUNG and JOSEPH FONG Computer Science Department, City University of Hong Kong, Hong Kong, csifonp(ii>,citvu.edu.hk Abstract This paper aims for proposing a methodology for the development and implementation of “Dynamic Web Authoring System for elearning”, which is a new type of web authoring system. This system assists creation of a web material that enables close user interaction and supports frequent change of its content. Web materials are dynamically created by following simple instructions provided by this system. The simple usage of the system benefits users who are not familiar with creation of web materials but wish to create a web material with complicated functions.

Keyword:

Authoring system, elearning, Dynamic web page

1. Objectives The objectives of this paper are mainly to develop a web tool namely “Dynamic Web Authoring System for elearning” for educational usage. It has six goals: 1. Accessibility - both educators and learners must be able to access this tool at anytime in anywhere 2. Portability - there are two levels of portability: the system itself and the web contents 3. Reusability - the educators can reuse the existing web contents to regenerate new web contents 4. Dynamic Web Page Generation ability - the server will dynamically generate the web pages for the educators based on their input 5 . Flexibility - educators can attach any related materials as part of the web content 6 . Statistics Monitoring - both educators and learners can review the learners’ performance 2. Introduction This project is to develop a new type of web authoring system. Unlike some of the online web tools, such as Blogger which only allow create web contents with no user interaction. Furthermore, this system consists sets of built-in function to help users to create user interacting web contents. Not like some of the well-known web application tools such as Microsoft Frontpage, Microsoft Visual Studio, and Dreamwaver that only create static web contents. 103

104

In order to describe how this tool can be such a complete web application tool, we need to look at each element and how they are related. Therefore, in the following sections, we will provide details of each elements and how those elements correlated and functioned as the entire system. After viewing the details, you can understand that this tool is the outcome of the best usage of existing systems and network, and not sequence of new technology.

3. System Design In this tool, there are two categories of users: educator^" and “Learners”. Although the system design for these two categories is totally different, the idea of accessibility is the same which is “at anytime in anywhere”. Another feature of this tool is the “dynamic web content generating ability”. Traditionally, web pages are static. Users can only view the web contents i.e. no interaction with the users. However, dynamic web pages that enable user interactions are more suitable for elearning, and therefore, required for creation of effective web learning material. The following sections will clearly demonstrate the elements of system design to achieve our objectives. 3.1. Educators The idea of accessibility can be archived by implementing this system as a web application. Thus, one of the system requirements is a web server. Another requirement is the interacting ability with the educators. This interacting ability purely relies on both server-side and client-side web languages. In addition, reusability and statistics monitoring are two other main aspects of this tool and can be archived by incorporating with a database system. 3.1.1. Web server In this project, the web server must be platform independent so that we can deploy this Dynamic Web Authoring System in any types of platform. We choose Apache as the web server, since it is a stable, portable, and functional supporting various functions and web languages that facilitate creation of web contents. 3.1.2. Web language User-interaction is a key factor to archive the dynamic web content generation. For this particular tool, we will use the PHP language as the server-side language and the Javascript as the user-side language. We use the PHP built-in functions and procedures. It provides five main functions:

105

i. ii. iii. iv. v.

Authentication - allow only authorized educators to login the system File upload ability - allow educators to upload related material to the system Database connectivity - store data into and retrieve data from the database system Web content generation - dynamically generate web pages depending on what educators input Excise Management and monitoring - allow educators to manage and monitor their existing excises by modification, deletion, and viewing.

Javascript is a well-developed and well-known user-side language. Javascript can act as an immediate reactor to check the correctness of the educators’ input in order to reduce the workload of the web server. 3.1.3. Database system Database system in this project is used to facilitate “reusability” of excises as well as “statistics monitoring”. The idea of reusing the questions from the existing excises is to reduce educators’ workload from typing the same information repeatedly. The purpose of statistics monitoring is for the educators to assess how well the students are doing on each excise.. In order to reuse the content of the excises as well as statistics monitoring, we need to store all related information in a database. In this tool, we will use MySQL as the database system. There are four reasons to choose MySQL. 0 Free license - it is free to use and to download (http://www.mysql.com 0 Independence of platforms - it has different versions for different platforms 0 Accessibility - PHP has built-in hnctions to work with MySQL 0 Database management and monitoring - we can use a software called “MySQL Administrator” to visually managing, monitoring, and configuring the MySQL database 3.1.3.1. Data design and structure There are three pieces of information we store in the database which are “educator information”, “excise information”, and “learner performance”. For the educator information, we mainly store the educator’s login id, password and email address. The table for educator information is:

106

Table 3.1.3.1 (1) - educator

and the entity relationship diagram is: id password

email

Figure 3.1.3.1 ( 1) - the ER Diagram of educator The excise information is the most important data in this tool. It consists of three sets of information. The first set is the brief information about the excise itself. The table for the first set is: Name Attribute Description excise -~ id eid an unique number to represent each excise ename the name of the excise explanation cover page specifying the information and/or the purpose of this excise number of questions in the excise nq the physical name of the web page being opened fname bv students ~

Table 3.1.3.1 (2) -excise

is a foreign key in the excise table so that we can As we can see that the key easily find out the owner of each excise. Hence, the relationship between educator and excise is 1 to many as presented below:

107

excise

educator

fname

Figure 3.1.3.1 (2) - the ER Diagram of excise

The second set is the detailed information of the questions on each excise. Its table looks like:

I content type

E

action-c action-w

I the content of the auestion

I

types of the answer such as single answer, multiple answer, true or false, matching and etc action to be taken if the answer is correct action to be taken if the answer is wrong

Table 3.1.3.1 (3) - question

In order to recognize which question refers to which excise, we borrow the key ‘‘eid” from the excise table as a foreign key. The relationship between the excise table and the question table is 1 to many:

108

1

1-1 I

I

excise

I

I 1

question

ename explanation

description

nq

content

fname

type action-c action-w

Figure 3.1.3.1 (3) - ER Diagram of excise and question

The third set is the possible answers for each question. The table should be:

Table 3.1.3.1 (4)-choice

In this set, we need to borrow the key “qid” from the question table as a foreign key, so that we can identify which choice refers to which question. The relationship between the question table and the choice table is 1 to many:

109

question

1

choice

m

Figure 3.1.3.1 (4) - ER Diagram of question and choice The complete entity relationship diagram for the excise information is:

excise

1

n . educator

1

m

choice

L

eid

eid

ename

@

aid

nq

description

content

content

content

correct

fname

type

type

action-c

action

action-w

Figure 3.1.3.1 ( 5 ) - ER Diagram of excise information The last piece of information is the learner performance. The table is:

Table 3.1.3.1 ( 5 ) -performance

110

Two foreign keys are required in the performance table. The entity relationship diagram should be:

question

3.1.4.

1

m

performance . n

1

excise

System architecture

Educators

Figure 5.1.4 - System Architecture (Educator)

111

3.2. Learners Since the learners do the excise via internet, we need to have a web server to support this action. When the learners work on the excise, immediate response which corresponds with their action is required. Moreover, the learner may want to know how well they are doing on each excise. Thus, we use a database to store the performance of the learners on each excise.

3.2.1. Web server For the purpose of portability, this tool generates a web contents in pure html format; hence, there is no particular web server required. 3.2.2. Web language As stated above, all the web contents are in pure html format, and therefore, no server-side language is required. The main concern in this section is the ability of user interaction. While the learner works on the excise, immediate response that corresponds to the learner’s action maybe required. This interaction ability can be archived by the client-side language. We use JavaScript to perform the following tasks: 0 0 0 0

Verify the learner the input Perform an immediate response regarding to the learner’s input Send back the result to the system for performance update Act as a communication channel between the server and the learner

3.2.3. Database design The performance of the learner will be recorded in the database so that the learner can review how well they are doing on each excise. 3.2.3.1. Data design and structure Only one table is required to store the performance information. Please refer to the section 3.1.3.1 for the data design and data structure.

112

3.2.4.

Architecture

which are oducaior, exciCc.q&On and choice

Figure 5.2.4 - System Architecture (Learner)

4. System Features The system features are the core part of this tool. There are two distinct sets of system features. One set is written in PHP language running in the server whereas the other is written in JavaScript running in the client machine. 4.1. Server-side Features These features are especially designed for the educators to managing their excises. 0 The first feature is the educator authentication 0 The second feature is to display all the information of the current login educator 0 The third feature is to accept file upload. Note that files will be physically stored in the system. The reason is to minimize the size of the database for faster data retrieval. The uploaded files will be appeared in the database only as reference: The forth feature is to create excise. We divide this feature into 4 sub-functions. The first function is to start creation of new excise. The second function is to store the question details in the database one

113

0 0

by one. The third function is to store the possible answers into the database. The last function is to convert all the information gathered from the first three functions into a set of web pages and save them in the system. The fifth feature is to delete existing excises for house keeping The sixth feature is to modify existing excises for the idea of reusability The last feature is to perform the statistics monitoring for analyzing purpose

4.2. Client-side Features The client-side features are mainly verification of the user’s input and performing corresponding actions based on the input andlor predefined actions. 0 the first feature is to make sure all fields being filled-in for new registration 0 the second feature is to confirm the password of the educator. 0 the third feature is to perform an immediate action when choice is selected. 0 the forth feature is to respond the answer submission. 0 the last feature is to display the answer to the learner when the learner gives up to answer. 5. Conclusion and Future Work This project is to develop a new type of system namely Dynamic Web Authoring System. It is designed for educational usage - one of the teaching tools. It helps the educators to create web contents so that their students can do excise via internet regardless of time and location. Furthermore, this system also provides statistics monitoring. For the educators, the statistics report helps them to make decision on which subjects they should spend more time on with the students. On the other hand, students can recognize their weakness so that they can spend more effort on particular subjects. In section 3, we provide detailed explanations on the individual system requirements and the system design for the development of Dynamic Web Authoring System. We have discussed how such Dynamic Web Authoring System can be developed in detail. In section 4, we introduce each function of this system. In conclusion, this system will be: 0 high accessibility since it is a web application high portability since it is developed with the most common server-side programming language - PHP - and client-side language - JavaScript 0 high reusability since all relevant information is stored in the database

114

highly useful since educators do not need to know everything about writing web pages; only need to follow simple instructions to finish the creation process 0 highly flexible since educators can attach their work as a part of the web contents 0 highly helpful since both educators and learners can view the statistics report to find out the strength and weakness We have studied this system thoroughly in the theoretical level. So in the next stage, we need to work on the actual implementation of this system. Furthermore, we would like to add two more important features: forum and knowledge sharing among educators. 0

Reference [ l ] Sampson, D., Karagiannidis, C., Schenone, A., & Cardinali, F., “An Architecture for Web-based e-Learning Promoting Re-usable Adaptive Educational e-Content”, Educational Technology & Society, ISSN 1436-4522,volume 5 number4,2002. [2] Joseph Fong, Margaret Ng, Irene Kwan and Marco Tam, “Effective e-Learning by use of HCI and web-based workflow approach”, Proceedings of the 2”dInternational Web-based Learning, LNCS 2783, pp271-286, August, 2003. [3] Bacsich, P., Heath, A., Lefrere, P., Miller, P., & Riley, K., “The Standards Fora for Online Education”, D-Lib Magazine, ISSN 1082-9873, volume 5 number 12, December 1999. [4] PHP Manual available from http://www.php.net/manual/en/index.php

[5] Blogger - an easy-to-use web site, where you can quickly post thoughts and interact with people - available from http://www.blomer.com

PART THREE Applications


CREATION OF A LIBRARY OF LEARNING OBJECTS (LO) FROM PRE EXISTING CONTENTS LEONEL IRIARTE NAVARRO Agrarian University of Havana, Habana , Cuba Email: [email protected] MANUEL MARC0 SUCH Languages and Computer Science Systems Department at the University of Alicante, Alicante , Spain, Email: [email protected] PEDRO PERNIAS PECO Languages and Computer Science Systems Department at the University of Alicante, Alicante , Spain, Email: [email protected] DANIEL MORON MART& Languages and Computer Science Systems Department at the University of Alicante, Alicante , Spain Email: [email protected]

ABSTRACT Interoperability is greatly considered in e-learning systems. XML based standards have arisen facilitating the common methods of description, identification and search of LO which are developed under different formats and platforms thus guaranteeingtheir reusability. S C O W standard is imposing in this field. It was developed by Advanced Distributed Learning [6] and based on proposals of organizations like IEEE Learning Technology Standards Committee [lo] Learning Object Metadata Working Group and IMS Global Learning Consortium. The most modern e-learning platforms use the standards previously mentioned to achieve interoperability among the different systems. Nevertheless, there is much usebl content available in different environments that is not structured according to the standards. This provokes serious difticulties for them to be reused. That is why it is essential to look for alternatives that permit to pack such contents in a way that they may be imported and used in the existing platforms. An automated procedure that allows the making of Digital Libraries composed of learning objects starting from previously developed content structures is proposed. Tools for facilitating this process have been developed. It has been applied to contents developed on a platform for teaching Spanish, as a result we have obtained a Multimedia Digital Library of Learning Objects.

Key Words: learning objects, e-learning, interoperability, standards, SCORM , IMS 117

118

1. INTRODUCTION

Education using the New Technologies has had a quick development lately. Many have been the techniques implemented to achieve virtual educative communities, where professors and researchers create thousands of teaching materials and content structures using web pages, e-learning platforms and others. The need to reuse the teaching materials in different software for e-Learning and for different students has provoked the creation of standards that allow the documentation, search and distribution of the produced educative contents [3]. The IMS standard created by the Global Learning Consortium is among the most important ones [7]; and then the SCORM, implemented by Advanced Distributed Learning Initiative [6] and the Institute of Electrical and Electronic Engineers (IEEE). Much iniciatives have developed to complement and facilitate the use of the most used standards, such is the case of CARE0 project that is being carried out in Canada and permits the exertion of educative objects based on the IMS standard. This project has joined efforts with an initiative called ALOHA (Advanced Learning Object Hub Application); it is a metadata server that offers this project special usage possibilities [4]. There are hundreds of initiatives for trying to solve the problem of interchanging objects among different applications; they will continue developing until they conform Ontologies as more complete structures for describing objects [S] which allow a formal representation of a concept, in addition to its semantic and syntactic representation. Apart from what has been previously stated, it is necessary to look for alternatives that allow to redesign the existing content structures and adapt them to the standards, so that they can be included in repositories or learning objects libraries that may be used in the different platforms. Some important steps have been taken on this matter, an example of this is the SCOMAKER tool created by Boxer Technologies AS [9] that allows to produce the necessary structures for making an Office document compatible with the SCORM standard. The procedure for making up learning objects is not very clearly defined, let alone the process for producing them from the contents previously developed under different environments; that is why we propose a procedure that allows tutors, professors and programmers to create learning objects libraries from the previously produced contents.

119

2. HOW TO AUTOMATE THE MAKING UP OF A LEARNING

OBJECTS LIBRARY The proposal is based on the creation of processes that enable the conversion of contents and teaching materials of the origin Learning Management System in learning objects. For creating such processes, it backs up in the standards. The specifications of the standard are defined in XML templates that have the information of the pattern and structure as well as the contents of the objects. Thus, if there are any changes in the standard, or if it is desired to export anything towards a new pattern you just have to adapt the corresponding templates. It is necessary that pedagogues and computer specialists get together in a multidisciplinary work to study the standards, adapt them to the particular conditions and then program the processes that enable the automatic making of the objects. The definition of a learning object is very wide, it is said that any digital object that can be reused is a LO [2]. Then the number of objects to be documented in the previously produced contents would be of many different kinds. The formats have the content of the objects but not the necessary digital information to automate them. This proposal is based on browsing such information in the platforms, data bases where there are learning objects, considering as LO not only the materials but the content structures as well. What follows is an outline of the elements that make up our proposal. Each of them is explained taking as an example its application in the project Virtual Classroom of Spanish (AVE) [8], where a learning objects library has been created.

LO template according to standard: This template is an XML structure that contains the necessary elements to document a learning object according to the standard to be exported. The metadata can have a literal value or make reference to the name of the function in the library of functions (4) that exerts this metadata. Physical Files: They are composed of the set of files that are necessary to obtain the outlet LO. There are two kinds of files; the ones associated to a template and the data files. Templates: They are XML structures that contain the necessary elements for a process ( 5 ) to produce an outlet physical file (7) that accompanies the LO documentation in the defined pattern.

120

Fig 1. Component for Making LO.

In AVE most of the objects are HTM files that have links to videos and Authorware objects; that is why in the previous template only the name of the file referred to in the clause OBJECT can be varied. Files: They are physical files as images, videos, office documents, flash presentations that are referred to in the LO. We have taken into account the possibility to convert many of these files in XML structures, documenting them considering the defined pattern, thus permitting more portability. To achieve this, existing tools can be used for their documentation using different patterns. MS Office documents, for instance, can be converted to SCORM by using applications as SCOMaker or LRN toolkit [ 131 so that this kind of document be completely documented according to the defined pattern. In case that most of the physical materials are of any specific file that is defined in the standard, a template can be made and then, apply the principle explained above. Contents Structure: It is a template where the structure of a specific content is defined and various LO are combined to achieve a teaching objective. It depends on the standard and there may be some, depending on the kind of structure

121

defined and the possibilities of the chosen pattern It is used by the process ( 5 ) to obtain the LO that corresponds with the structure of contents(7.2). It has been determined in AVE, that the basic structure of the contents is an entity called TEMA which contains the different activities to achieve a basic unit of learning. A template to produce this structure in an automated way has been designed. It is based on the IMS standard from which all the structures of themes contents are produced and so courses that contain these themes. Library of functions : All the functions referred to in the templates are in this library. The implementation of such library depends on the structure and programming strategy. There can be one or some libraries. Its use guarantees that the system reacts upon the changes produced in the platforms structures in a simple way. The processes: The processes use the templates described above to make the LO (7), not only the ones that correspond with the description of a material or set of materials by means of a determined pattern but also the ones associated to a learning structure as well. These processes use the conveniences of the XSLT to obtain different outlet formats from the XML original templates and will use the library of functions just described to substitute the references of the templates by the corresponding values. Own Databases: Own Databases are the ones that platforms use for their fulfillment. Libraries of functions use these structures for their implementation. Learning Objects (LO): After applying the processes just described, XML structures that document each LO (7.1) will be obtained according to the defined pattern, the physical files that resulted from applying a template (7.2), as well as the LO that describe the content structures (7.3). The characteristics of the most used e-learning platforms have been analyzed [ 101 and in most of them we observe common features which allow to export the necessary information to apply this procedure. They have: 0

0

0

A databases structure where there is some information about the

teaching administration process, class sessions, teaching materials and learning models. Repositories or teaching material libraries stored in physical files. Structures for implementing different models and learning strategies.

3. CONCLUSIONS AND RECOMMENDATIONS What we propose may be used in different courses previously produced with the aim of obtaining an automated creation of educative objects reusable in more

122

modern platforms that fit with the standards; thus permitting the distribution of courses to environments where the connection levels are still insufficient. The use of these procedures in AVE project guarantees the interoperability of such courses in other educative environments with few economical resources and little time. This is due to the capacity of the elaborated applications to make use of the information existing in AVE database. Such procedures can be used as a basis for the making of LO libraries in other platforms. Using the templates and libraries of functions here proposed guarantees that the procedures and designed applications for making the LO be more resistant to the changes produced in the standards. They also offer the possibility to make new objects based on other specifications. The creation of reusable educative objects is not possible in isolation, the creation of multidisciplinary groups is required for designing the most appropriate strategies for professors to make and use educative objects in a simple way as is the case of the example explained above. A LO repository does not solve the problem at all. It is very important to study the mechanisms for storing and retrieving information from these objects so that they can be found and used. The new LMS assimilates the new standards but the professor needs to find objects and create new ones. It is necessary to use procedures and techniques that permit to produce LO from the existing ones in a dynamic way and with certain instruction from the professor.

REFERENCE Kay, M., XSLT Programer’s Reference, Wrox Press, 2001 Wiley D., Connecting learning objects to instructional design theory: A definition, a metaphor, and a taxonomy,http://reusability.org/read/chapterslwiley.doc, 2000. Morales,R., Capacitacibn basada en Objetos Reusables de Aprendizaje, http:l/~.umb.edu.co/umb/sitiopedagogia/lecturas/tendencias.pdf, 2002 Norman, D. ,Building Digital Books with Dublin Core and IMS Content Packaging, http://www.bncf.net/dc2002/program/~paper10.pdf, 2002 OWL,Web Ontology Language, http://www.w3.org/TR/2002/WDowl-guide-20021104,2003 ADL, Sharable Content Object Reference Model Version 1.2,Advanced Distributed Learning, 2001. http://www.adlnet.org, 200 1. Learning Resource Meta-Data Specification: Version 1.1 Final

123

Specification.

IMS

Global

Learning

Consortium,

http://www.imsproject.org/metadata/index.html, 2000 [8] [9] [ 101

[ll] [ 121 [ 131

[14] [ 151

IMS Learning Resource Meta-Data Specification: Version 1.1 Final Specification. IMS Global Learning Consortium, http://www.imsproject.org/metadata/index. htm1,2000. 2001. AVE, Proyecto AVE,, http://www.fundacionduquesdesoria.es, LTSC, Learning Technology Standards Committee (LTSC) Drap Standard for Learning Object Metadata Version 6.1. http://ltsc.ieee.org/doc/,200 1. SCOMAKER ,The MS Office SCORM-plugin. http://www.scomaker.com/,2002. XHTML:eXtensible HyperText Markup Language.. http://www.xhtml.org, 2001. Microsoft ,Interchangeable Learning Resources. Http:l/www.microsoft.com/eleam/,Microsoft ,200 1 . Soluciones de Eleaming / Formaci6n a Distancia. http://elearning.bankhacker.com/, 2002. M. Marco, Sistema de producci6n multimedia para la WEB, ejemplo aula virtual deI espaiiol. En actas del IX Congreso Internacional de Informhtica en la Educacibn, http://www.informaticahabana.co.cu/eventos/eventos/Educacion/defaul t.htm, 2003.


A MOBILE AGENT ASSISTED LEARNING RESOURCE SERVICE FRAMEWORK BASED ON SOAP WU DI, YANG ZONGKAI, CHENG WENQING Department of Electronics and Information Engineering, Huazhong University of Science and Technolom, Wuhan, Hubei, China 430074 1

Abstract Learning resource service is one of the most important services in an E-Learning environment. To provide high-quality learning resource service for users, the authors of this paper put forward a mobile agent assisted learning resource service framework. The framework employs Simple Object Access Protocol (SOAP) as the message platform for agent communication. In addition, some widespread accepted learning resource metadata and packaging standards, such as Learning Object Metadata (LOM) and Content Packaging (CP) are applied in the framework to standardize resource. Knowledge Query and Manipulation Language (KQML) is applied in the framework to support agents communication. Under this framework, the mobile agent on behalf of a user can discover learning resource autonomously in the environment, so that the user can gain needed resource from anywhere and at anytime. The process of resource discovering is transparent to the user. A prototype application system based on the framework is also introduced in this paper, which gives a sample application case of the framework.

2

Introduction

One of the most important materials of E-Learning is the digital learning resource. It’s impossible to effectively apply an E-Learning system without learning resource. To provide learning resource service for learners and other consumers more quickly and effectively, we need to gain an effective way to build a novel resource service system in E-Learning environment. In traditional resource service system, learning resource is stored in a centralized resource database. Users query the database and discover their needed resource. But it is impossible to collect all resource in one independent database and it is difficult to provide individual services for special users. So we need a new service framework to provide high-quality learning resource service. Mobile agent gives us a new way to satisfy the requirement. In recent years, mobile agent has gained increasing attention. The ambulant and communicative characteristics of mobile agent make it easy to gain user’s profiles and provide smart services for users. So the mobile agent technology can be applied in learning resource service process to build a new resource service framework. In this paper, the authors put forward an agent assisted learning resource service framework based on Simple Object Access Protocol (SOAP) [l]. 125

126

Extensible Markup Language (XML) [2] is applied in the framework to provide protocol binding specification and Knowledge Query and Manipulation Language (KQML) [3] is applied in the framework as an agent communication language (ACL). Two mature technology standards about learning resource - Learning Object Metadata (LOM) and Content Packaging (CP) are applied in the framework to standardize learning resource.

3

Related Work

3.1 XML

XML has come a long way ever since the World Wide Web Consortium (W3C) issued it as a recommendation in February 1998. The standard has already gained widespread acceptance within the developer community, and almost all major application vendors offer extensive support for it [4]. XML provides a new standard way for data exchange on Internet. 3.2 Mobile agent

The application of mobile agent is very public now. Mobile agent is an emerging technology. It makes the design, implement, and maintain distributed system more easily, attracting a great deal of interest from both the industry and the academia. In particular, the mobile agent paradigm has been used to design applications ranging from distributed information retrieval to network management [5]. Mobile agents can present the following attributes [6]: a. Reactive - the capability to respond to changes within agent environment; b. Autonomous - the mobile agent is able to exercise control over its own actions (decisions); c. Goal-oriented - the agents have an intended itinerary, they do not simply act in response to the environment; d. Communicative - the capacity to communicate with other agents or processes, by exchanging information (knowledge); e. Mobile - the mobile agents can transfer themselves from one machine ( Internet host) to another.

127

3.3 SOAP

SOAP Envelope

\,

Figure 1 SOAP messsage structure [7] SOAP is a method of remote method invocation, and is based on XML-FWC further tightening up the applicability of XML to data structure and storage [8]. The structure of SOAP message is shown in Figure 1. The SOAP protocol consists of three main parts [9]: a. An envelope that depicts the contents of the message and how to use it; b. A set of rules for serializing data exchanged between web applications; c. A platform-neutral manner to represent remote procedure calls, that is the way in which queries and the resulting responses to the procedure are represented. 3.4 KQML

KQML is a language and protocol for exchanging information and knowledge. It is both a message format and a message-handling protocol that supports run-time knowledge sharing among agents. KQML offers a variety of message types (performatives) that express an attitude regarding the content of the exchange. Performatives can also assist agents in finding other agents that can process their requests [lo]. Three layers are included in KQML (see Figure 2).

128

Figure 2 Three layers of KQML [lo]

3.5 LOMand CP

Figure 3 Categories of the LOM schema [l 11

LOM is a metadata standard about learning object published by IEEE LTSC (http://ltsc.ieee.org). A metadata instance for a learning object describes relevant characteristics of the learning object to which it applies. Such characteristics may be grouped in general, life cycle, meta-metadata, educational, technical, educational, rights, relation, annotation, and classification categories [12]. The categories of the LOM schema are shown in Figure 3.

129

Figure 4 The structure of CP information model [13] CP is a standard about content organization published by IMS (http://www.imsglobal.org).To satisfy the needs of online learning resource interchange, the content of the resource should be packaged in a known manner and file format, and with sufficient supporting information. CP defines a standardized set of structures that can be applied to exchange content. The structure of CP information model is shown in Figure 4. 4

The Structure of the Framework

4.1 The layered structure of the pamework

Figure 5 The structure of the framework

130

The framework has a layered structure, which includes five layers (see Figure 5 ) : a. Content layer - includes the learning resource physical files (such as media, picture, text, etc.) and metadata information of the resource. LOM is applied in this layer to present the metadata information because it is widespread accepted and has a XML binding specification. b. Package layer - includes the packaging information of the resource. In this layer, the metadata information and the resource physical files are packaged into one content unit (PACKAGE), so that it can be transported and exchanged in an e-Learning environment. CP is applied in this layer because it is mature and widely accepted by manufacturers. C. ACL layer - provides a communication bridge among agents. KQML is applied in this layer to support agent communication, because it is relatively mature and can be bind with XML. d. Message layer - exchanges messages among SOAP nodes. SOAP is applied in this layer because it is bind with XML and based on HTTP, which makes it have the ability to pass firewall. e. Transport layer - provides a basic network infrastructure to transport messages in network environment. The public accepted network transport protocol - TCP/IP is applied in this layer.

4.2 The basic information unit in thepamework

Figure 6 The structure of the basic information unit in the framework

The basic information unit in the framework is a special SOAP message which embodies a learning resource content package and some KQML messages (see Figure 6). The KQML message which contains agent communication information is XML

131

bind and embedded in the SOAP envelope. All the learning resource request and response information are included in it. The learning resource content package is attached as an attachment part of the SOAP message. Learning resource content package is a package interchange file (PIF) which is a single file (e.g., zip, jar, .cab) compressed from a content package directory. The framework has several advantages: a. Smart - the whole process of resource discovering is transparent for users. The agent on behalf of the user can gain target resource by itself or collaborate with other agents. The user do not need to know how to find and where to find, he may not know the process entirely. b. Platform-independent - the framework is bind with XML and utilizes SOAP as message carrier, so systems developed with any programming language and run on any operating system can apply the framework. c. Standard - LOM and CP is applied in the kamework to standardize resource. So all the resource under the framework is standard and can be exchange conveniently among learning management systems (LMS). One problem of the framework is the performance. Because all of the protocols in the framework are bind with XML, the information unit is very large. Additionally, SOAP is not a high-speed protocol. So the performance of the framework is not very graceful. 5

The Application of the Framework

Figure 7 The application system of the framework In an E-Learning environment, the framework can be used to build a distributed learning resource service system (see Figure 7). The application system of the framework is a multi-agent system (MAS). Two types of agents

132

are applied in the system. Learner agents (LA) are software agents whose functions are: (1) to represent and model a human learner; and (2) perform tasks on behalf of human learners [14]. Resource Agents (RA) are the core agents in the system. RA receives the LA'S request and process the request by it self or collaborate with other agents. The request and response message in the system are defined as KQML messages (see Figure 8).

Figure 8 The request and response KQML perforrnatives

When an LA get user's request, it queries the profile database and creates a request KQML message. The message is packed as a SOAP message and transport to a RA in the environment. The RA moves to one learning resource database and queries it, then moves to the other, etc. The process will not be terminated until the target resource is f h d out or no such resource in any resource database. Then the RA will create a response message to make LA get the result. The software language to construct mobile agent system should be object oriented, platform independent, with communication capability, and implemented with code security [ 151. Aglet provided by IBM is a nice solution for agent development because it is developed with Java which makes it can run on any platform. It can be chose to implement a prototype system like the application system introduced above. 6

Conclusion and Future Work

The authors of this paper put forward an agent assisted learning resource service framework based on SOAP. KQML is applied in the framework as an agent communication language, which makes it easy to interoperate between two resource agents. SOAP is applied in the framework as a communication protocol to load message and transport resource content, which provides a communication approach over HTTPEMTP. XML is applied in the framework as a binding specification, which makes the protocols be bind in consistent format and parsed in a uniform avenue. In the future work, we will focus on improving the performance of the framework. Some methods, such as XML data compressing and package

133

predigesting can be considered to get the goal. In addition, we will try to optimize the structure of the framework so that it can be implemented easy, and we will also implement a practical application system based on the framework. Acknowledgement This work is supported by Ministry of Education (MOE), P.R.C. We acknowledge the staff of E-Learning Working Group (ELWG) in our research center (http://itec.hust.edu.cn),and we also thank for the support of MOE. We thank Mr. Peng Qingtao for his provocative discussions about the application of the framework and his perfect development work in the system implementation. References [ 11 World Wide Web Consortium, Simple Object Access Protocol (SOAP) Version 1.2 Part 0: Primer, W3C Recommendation June 24,2003.

[2] Sperberg-McQueen, Eve Maler, Extensible Markup Language (XML) 1.O (Second Edition), W3C Recommendation 6,2000. [3] Yannis Labrou, Tim Finin, A Proposal for a new KQML Specification, TR CS-97-03, February 3, 1997. [4] Jaideep Roy and Anupama Ramanujan, XML Schema Language: Taking XML to the Next Level, IT Professional, March-April 2001. [5]Jon T.S. Quah, Winnie C.H. Leow, Y.M. Chen, Mobile Agent Assisted E-Learning, the First International Conference on information Technology & Applications (ICITA 2002), 2002. [6] Sabin-CorneliuBuraga, Developing Agent-Oriented E-Learning Systems, proceedings of the 14th International Conference on Control Systems and Computer Science, 2003. [7] Sun Microsystem, http://www.huihoo.com/one-and-net /app7/12.1.htm#08, 2004. [8] Paul OConnell, Rachel McCrindle, Using SOAP to Clean up Configuration Management, the 25th Annual International Computer Software and Applications Conference (COMPSAC’Ol), 200 1. [9] Sabin-CorneliuBuraga, Developing Agent-Oriented E-Learning Systems, Proceedings of The 14th Internation- a1 Conference on Control Systems And Computer Science, 2003. [lo] Yannis Labrou, Tim Finin, A Proposal for a new KQML Specication (TR CS-97-03), February 3, 1997.

134

[ 111 Zhongnan Shen, Yuanchun Shi, Guangyou Xu, A Learning Resource Management System Based on LOM Specification, Proceedings of the 7th International Conference on Computer Supported Cooperative Work in Design (CSCWD 2002), Rio de Janeiro, Brazil, September 25-27,2002. [ 121 Learning Technology Standards Committee of the IEEE Computer Society, IEEE Standard for Learning Object Metadata. 13 June 2002. [ 131 IMS Global Learning Consortium, Inc. IMS Content Packaging Information Model Version 1.1.2 Final Specification, 8 August 200 1. [ 141 Teresita Limoanco, Raymund Sison, Use of Learner Agents as Student Modeling System and Learning Companion, Proceedings of the International Conference on Computers in Education (ICCE’02). [ 151 James E. (May 1996) White Mobile Agents; URL: http://www.webtechni ques.com/archives/l996/ OS/white/.

TEACHING STROKE ORDER OF CHINESE CHARACTERS BY USING MINIMAL FEEDBACK KERRY TSANG AND HOWARD LEUNG City University of Hong Kong, Tat Chee Avenue Kowloon Tong, Kowloon, Hong Kong In this paper, we propose a new automatic method to learn or correct the stroke order of Chinese characters. The methodology is based on the ideas of grouping strokes and applying the longest increasing subsequence method to provide a minimal feedback to the user automatically. With minimal feedback, the least amount of information is needed to be provided to the user to correct the stroke order mistakes. Experiments from user studies show that the proposed minimal feedback approach requires less time and achieves similar learning effectiveness compared with normal feedback.

1. Introduction Traditionally, people learnt a Chinese character by reciting it stroke-by-stroke. As time goes by, people write and use the characters many times, so they have a strong memory - a reflex on the characters. Their memory may be refreshed by reading printed text to recollect information about a Chinese character such as the stroke shape, orientation, length, relative position, structure, number of strokes, etc. However, the stroke order that requires the temporal information cannot be recalled in this manner. A person may write a character in the wrong stroke order either because the person forgets the correct way or because that person makes some mistakes the first time he/she learns that character. Studies have shown that children made stroke sequencing errors because sequencing rules for Chinese characters can be ambiguous’. Additional exercises are required to verify whether the user writes the Chinese character correctly and the user is provided with some feedback if the stroke order is wrong. Although a stroke-by-stroke animation of the character can be provided as the feedback each time when a user writes a character in the wrong stroke order, this feedback is the same in all situations so it does not specify where the user makes the mistake. This kind of feedback is quite time-consuming especially if the user does not make too many stroke order mistakes. Often parts of the user’s memorized stroke order are correct, thus it is not efficient to re-learn the entire order from the first stroke to the last stroke. As a result, we propose to provide automatic minimal feedback to the user for correcting the stroke order mistakes. With minimal feedback, the least amount of information is provided to the user to correct all stroke order problems. This minimal feedback is achieved by grouping consecutive strokes in the correct relative order and applying the longest increasing subsequence method to identify the groups with the correct and incorrect stroke order to be feedback to the user. 135

136

Some researches have been focused on determining the stroke order and structure for offline Chinese character recognition and signature verification application^^.^. Regarding handwriting education, we have done some prior work on teaching people to write Chinese characters with automatic analysis based on the shape information4. In this paper, we propose a novel minimal feedback approach in teaching the stroke order. This paper is organized as follows. In Section 2 we provide a description of our methodology and illustrate the details of the algorithms. We describe our experiments and present the results in Section 3. The conclusions and future work are provided in Section 4. 2. Minimal Feedback Methodology

After showing a template character to the user on a pen-based device, the user is required to write the same character with the stylus. As illustrated in Figure 1, the user input character is processed in four stages: matching, grouping, longest increasing subsequence (LIS) and feedback. AAer these stages, the minimal feedback on the stroke order correctness is generated automatically and presented to the user.

Matching -+ Grouping -+

LIS

-

Feedback

2.1. Matching

The objective of this module is to determine whether the user writes a character in the correct stroke order. There are several methods to determine the correct stroke order from the user input such as applying offline techniques to extract the radicals and using heuristic rules to determine the order of the radicals or ordering the strokes with other structural rules’. The stroke sequence can also be estimated by minimizing the total stroke distance using the direction as the cost3. Since in our system, we collect the handwriting with a pen-based device so the temporal information of the strokes is available. Moreover, the template character is present in front of the user for learning that character. As a result, in the matching stage, we can compare the strokes between the user input character and the template character in order to determine whether the user input character has the correct stroke order defined by the template character. The stroke correspondence between the user input character and the template character needs to be determined and it can be obtained by minimizing the total cost between strokes with the Hungarian method5.

137

An example of the stroke matching result between the user input character (U) and the template character (T) is shown in Table 1. The entries in the 2nd row represent the temporal stroke IDS of the template character while the entries in the 1" row represent the corresponding matched temporal stroke IDSof the user input character after finding the stroke correspondence. For example, stroke 0 of user input is matched with the stroke 0 of the template; stroke 5 of user input is matched with the stroke 1 of the template, etc. This results in a numeric sequence [0, 5, 6, 1'2, 3'41 representing the temporal stroke IDS of user input (stroke 0 + stroke 5 + stroke 6 + stroke 1 + stroke 2 + stroke 3 + stroke 4) which is the correct stroke order specified by template character. This result is passed to later stages for grouping and extracting the longest increasing subsequence before generating minimal feedback. Table 1. Stroke matching result.between user and template characters UIO

5

6

1

2

3

4

T I 0

1

2

3

4

5

6

2.2. Grouping

Grouping is performed for the numeric sequence resulting from the matching stage. A group is defined as a set of adjacent strokes in the numeric sequence with stroke IDS forming increasing consecutive numbers. For example, there are three stroke groups (0), (5, 6), (1, 2, 3'4) for the user character shown in Table 1. The time complexity of grouping is O(n) where n is the number of strokes. There are two advantages in grouping the strokes: preservation of relative order within group and reduction in computation complexity. The strokes within a group have the correct relative stroke order because the stroke IDS within a group form increasing consecutive numbers. People tend to memorize relative stroke order rather than the absolute stroke order. This can be easily verified by a simple experiment by first showing a person a known character and asking that person which stroke corresponds to the 31d stroke. The person often needs to trace the character in hisher mind starting from the 1" stroke. On the other hand, if we show the person the first two strokes of a character and ask that person to write the next (31d) stroke, then he/she can write it immediately without much thinking. This is because the stroke order information is stored sequentially in the relative stroke order in people's mind rather than in the absolute stroke order with random access. This observation justifies the grouping of strokes such that people can associate strokes in the relative stroke order within each group as one unit.

138

Another advantage for grouping is the reduction in computation complexity. Considering each group as one unit speeds up the processing of the groups for extracting the longest increasing subsequence in the later stage. Qualitative and quantitative analyses will be provided in subsequent sections regarding this aspect. In addition, the groups are used for providing feedback to the user to identify correct and incorrect stroke order. This reduces the amount of information provided to the user hence minimal feedback can be achieved. 2.3. Longest Increasing Subsequence

After grouping, we determine which groups of the user character have the correct stroke order by using the Longest Increasing Subsequence (LIS) method. As suggested by the name, the LIS is the longest subsequence in which all numbers are in increasing order. Several algorithms637have been proposed to extract the longest increasing subsequence given an input sequence. When the sequence is nearly sorted, i.e., when the elements of the numeric sequence are almost strictly increasing, the performance of the algorithm6will be the worst with O(n2)where n is the number of elements in the input sequence. Another algorithm7is proposed in with time complexity of O(n log n). In our approach, after grouping the strokes, we form a new sequence by taking the first number of each group and use this sequence as the input for the LIS algorithm7. The strokes in the associated groups from the resulting LIS are labeled as the strokes with the correct stroke order and all other strokes in the remaining groups are labeled with the wrong stroke order. For example, the groups formed from the example shown in Table 1 are (0), (5,6), (1,2,3,4). The new sequence [0,5, 13 is formed by taking the first element from each group. The LIS is determined to be [0, 11 corresponding to the strokes (0), (1,2,3,4) which are the strokes determined to have the correct stroke order. The remaining strokes (5,6) are considered as the strokes with wrong stroke order. This information is provided to the user during the feedback stage. 2.4. Feedback

In this stage, according to the LIS and the groups, we show an animation of a correct-ordered character group-by-group for achieving minimal feedback. By using the animation group-by-group in this minimal feedback approach, the user can just pay attention to the parts they made mistakes, thus there is no need to learn the whole character stroke-by-stroke again as in the normal feedback.

139

3. Experiments and Results

We carry out some experiments to study the effectiveness of our proposed minimal feedback approach for teaching people stroke order of Chinese characters. There are two parts in the experimental setup. The first experiment is designed to test how much time people need to learn the stroke order of a character by using minimal feedback (group-by-group) and by using normal feedback (stroke-by-stroke). The second experiment is to test whether people will still remember how to write the characters with the correct stroke order after one week. It is assumed that the subjects know how to write the characters without the stroke-production errors such as stroke reversal, concatenation of separate strokes and broken strokes. Three template characters: 1. @/ , 2. @ and 3. i 3 are used in these experiments. These characters are selected because it is common that people write them with the wrong stroke order. For example, for the character @/, people usually write the part “&” first, and then two f .Actually, the strokes for the part ‘‘W’ should be written at different instants. The correct stroke order is shown in Figure 2 .

Figure 2. Stroke order for the character @.

3.1. Reduction in time complexity for grouping

As indicated in Section 2.2, grouping reduces the time complexity for obtaining the LIS. As stated in section 2.3, the time complexity €or LIS is O(n log n). After collecting the user data for the four template characters, the number of groups m iis recorded and the average number of groups is calculated in each case. In addition, the average time complexity for grouping can be calculated by taking the average over the quantity mi log mi.The results from 150 user input characters are summarized in Table 2 in which the reduction in time complexity for obtaining LIS with grouping is shown for each case. It can be seen in the last column on Table 2 that the LIS with grouping can be computed 6 to 12 times faster than the case without grouping.

140 Table 2. Reduction in time complexity for grouping. Char.

No. of strokes

Time Complexity without grouping

1.m 2.g 3.E

9 11

8.59 11.46 4.67

6

Average no. of groups 2 2.5 2

Average Time Complexity with grouping 0.716 1.262 0.722

Speed up factor 12.0 9.1 6.5

3.2. Experiment 1: Time taken _ _ learn the s.. oke order with minimal feedback and normalfeedback

In order to compare the efficiency between using minimal and normal feedback, a PDA program is developed to record the data as shown in Figure 3. There are 18 people in the test group (with minimal feedback) and 18 people in the control group (with normal feedback). The age of the subjects is between 20 and 40. With the program in the PDA, each subject will try to write characters given the template characters. When a template is chosen, the program starts the timer for this template character. Then the user writes the character in the designated square region and presses a button to start the analysis. If there is something wrong about the stroke order, the user will get the feedback. There are two kinds of feedback: normal feedback with stroke-by-stroke animation and the minimal feedback with group-by-group animation. The choice of feedback depends on whether the subject belongs to the test group or the control group. After having learnt the feedback, the user tries that character again. The user repeats this process until the user can write in a correct stroke order and the program will log the time the user spends for each template character.

Figure 3. PDA program for the experiment.

141

Figure 4 shows the correct stroke order of a template character and Figure 5 shows the user input sequence of strokes. After matching the strokes, the array which is the correct stroke storing the matching result is [0,3,4,5,6,7,8,9,10,1,2], order the user input should have. There are 3 groups (0),(3,4,5,6,7,8,9,10),(1,2) and the LIS is [0,3,4,5,6,7,8,9,10]. Strokes 1 and 2 are not in the LIS so they are the strokes with wrong order. With minimal feedback shown in Figure 6,the program shows an animation for each group, with the lst group being (0), 2"d group being (3,4,5,6,7,8,9,10) and the last group being (1,2). The character can be written correctly if stroke 1 and 2 are written at the end. On the other hand, under normal feedback, all the strokes of the user input character will be shown one-by-one according to the template order, as indicated by Figure 4.

Figure 4. Correct stroke order for the character

"w.

Figure 5. Stroke order for the user input character.

Figure 6. Minimal feedback for the user input character

The average time users spent for each character is recordet- and the result is summarized in Table 3. It can be observed that people spend less time to learn about the stroke order of a character by using minimal feedback compared with normal feedback. Table 3. Comparison between normal feedback and minimal feedback in terms of average learning time required. Characters

charOl@

charO2R

charO3g

Average Time under Normal Feedback (sec) Average Time under Minimal Feedback (sec) Percent Improvement

48

44

43

41

10.4%

6.8%

25 20 20%

142

3.3. Experiment 2: Number ofpeople who remember how to write the characters in the correct stroke order

After one week, users in both the test group and the control group are required to write the characters again to test whether they still remember how to write them in the correct stroke order. In Table 4, the average number of trials required to write each character with the correct stroke order is shown. It can be observed that the average number is very close to 1 in all cases meaning that most people can write the characters correctly in a single trial after one week. Using minimal feedback is just as effective as using the normal feedback. Table 4. Comparison between normal feedback and minimal feedback in terms of average number of trials required. Characters Minimal feedback Normal feedback

char01 @I char0295 I .3 1.2 1 1.1

char03 I 1

4. Conclusions and Future Work We proposed a new automatic approach for learning the stroke order for Chinese characters with the notion of minimal feedback. This approach benefits from the processes of grouping the strokes and extracting the longest increasing subsequence to identify the strokes with correct and incorrect stroke order. Experiments show that users require less time to learn the stroke order with minimal feedback than with normal feedback. Both kinds of feedback have similar learning effectiveness in terms of remembering the stroke order after a period of time. It can be concluded that minimal feedback performs better than normal feedback since it requires less time with similar learning effectiveness. Currently our approach only focuses on the stroke order assuming no stroke-production problems during the experiment. As future work, we will add other functions for determining the stroke-production errors such as stroke reversal, concatenation of separate strokes and broken strokes. Feedback related to these cases will be created to enhance the learning capabilities of our system.

Acknowledgments The work described in this paper was partially supported by a grant from City University of Hong Kong (Project No. 700171 1).

143

References 1. N. Law, W.W. Ki, A.L.S. Chung, P.Y. KO and H.C. Lam. “Children’s stroke sequence errors in writing Chinese characters”, Reading and Writing: An InterdisciplinaryJournal 10, 267-292 (1998). 2. Z. Chen, C.-W. Lee and R.-H. Cheng. “Handwritten Chinese character analysis and preclassification using stroke structural sequence”, IEEE Proc. of the 13th Int. Con$ on Pattern Recognition 3, 89-93, (1996). 3. K.K. Lau, P.C. Yuen and Y.Y. Tang. “Stroke extraction and stroke sequence estimation on signatures”, IEEE Proc. of the 16th International Conference on Pattern Recognition 3, 1 19-122 (2002). 4. H. Leung and T. Komura. “Web-based Handwriting Education with Virtual Animated Teacher”, Int. Con$ on Web-basedLearning, 293-300 (2004). 5. R.E. Burkard and E. Cela. “Linear Assignment Problems and Extensions”. Handbook of Combinatorial Optimization, 75-149 (1999). 6. C. Cerin, C. Dufourd and J.-F. Myoupo. “An efficient parallel solution for the longest increasing subsequence problem”, Proc. of the Fifth International Conference on Computing and Information, 220-224 (1993). 7. K. M. Chao. “Dynamic-programming Strategies for Analyzing Biomolecular Sequences”, IMS Lecture Notes Series, Selected Topics in Post-Genome Knowledge Discovery 3, 1-24 (2004).


AN ON-LINE PROGRAMMING ENVIRONMENT WITH AUTOMATED ASSESSMENT RONNIE C. T. CHEUNG Department of Computing The Hong Kong Polytechnic University Hung Hom, Kowloon, Hong Kong. Email: [email protected]. edu.hk Practical sessions are important elements in programming courses. Students spend lots of time in practical programming sessions. Object-oriented programming courses require additional practices in working with design models. Our objective is to encourage active participation for object-oriented programming beyond classroom environments. By using a Web-based learning environment for object-oriented development, it is possible to track all compilation activities associated with every student. We have shown that an easy-to-use development environment improves teacher-student interaction, provides analysis on students’ participation, and helps students with their compilation activities in real-time. We have developed a Web-based learning environment that provides various features to help students to learn Java Servlets. It also helps tutors to provide feedback for programming assignments.

1. Introduction

In recent years, object-oriented programming has become one of the most influential programming paradigms that are being widely used in education and industry [3]. Many universities are introducing object-oriented programming (such as Java servlet programming) early in their courses. In our courses, firstyear students also make use of an object-oriented development environment for learning programming. These courses aim at developing students’ skills in building reliable systems that can easily be extended and maintained. From both pedagogical and computing perspectives, there are many advantages in teaching object-oriented approaches to beginners. Firstly, it encourages well-structured programming practices. Secondly, it allows at an early stage, the introduction of important software development concepts, such as information hiding and reusable component-based development. It also allows us to reduce the problems associated with the paradigm shift for learning from a procedural language to an object-oriented environment [2]. It has been found that many students whose first programming language is a procedural language experience problems in adjusting to the object-oriented paradigm. Finally, it should be noted that object-orientation is currently the most popular programming paradigm adopted by industry. Students should be well aware of the current practices, and this provides a strong motivation for learning. At the same time, Web-based learning is becoming more and more popular in university environments. A Web-based environment allows students to access 145

146

course materials online. To ensure that students understand the concepts in programming courses, practical exercises and assignments are important elements that must be incorporated in the course contents. Although there are many web sites providing practical exercises, most of them do not provide immediate feedback to students. Without a web-based environment for interaction and providing feedbacks, marking of assignments are always delayed. To solve these problems, we have developed a Web-based Learning Environment for OOP Assessment (WEBLOOP) [l]. It provides a Web-based tool for drawing UML class diagrams, a code editor for modifying the Java programs, an online compilation module and a marking module for checking with student submissions. With WEBLOOP, all environment settings (e.g. the class path) are performed on the server side, students do not need to install the software (e.g. Java Development Kit, Tomcat) and set up the environment for running the software on their own computers. They only need to concentrate on learning concepts and practical skills. WEBLOOP also provides a marking module for automating the assessment process. Figure 1 shows a typical learning cycle for WEBLOOP. Students submit Java Servlet programs online and the marker module provides instant feedback by comparing the results with a database of test cases. Students’ performance are recorded in the system automatically through practical programming exercises that can be assessed on-line.

’

Attempt Problem+ Asses smentt‘

Learning+

A Feedback+ ~

Figure 1. The Learning Cycle with Automated Feedback

147

2. Systems architecture and functions The objective of the project is to develop a web-based environment for learning Java Servlet programming. It provides the followings features : 0 0

0 0

0

Support for drawing UML diagrams A visual environment for learning Java Servlet programming Support for online submission of programs On-line program compilation and assessment Automated feedback and monitoring of student activities

2.1. Systems architecture

Figure 2 shows the basic architecture of our remote working environment. To access our software, such as remote compilation/execution server, students log into our remote server to work on their Java source code. The compiled class files and feedbacks are sent to the students PC/workstations at home. The system architecture is a client-server architecture consisting of three components: a class diagram drawing tool, a compilation server, and a web server for maintaining the individual activities of each student.

Figure 2. Systems Architecture

148

2.2. The UML Class Diagram Editor The UML class diagram editor encourages students to design programs using object-oriented concepts. It encourages students to solve programming problems by drawing class diagrams first. In the WEBLOOP environment, no Java programs can be created before the corresponding class box has been created. Figure 3 shows a screen dump of the UML class diagram editor. The system provides options for students to add a class, an abstract class or an interface to the class diagram by clicking on the “create new class” button. The links between classes in the class diagram include specialization and association. Specialization provides an inheritance hierarchy for the classes whereas association shows how two classes are related to each other. The program text associated with each class box can be edited by clicking the correspond icon. The editor is language sensitive and provides specific hnctionality for inserting program codes for different versions of Java.

Figure 3. UML class diagram editor

149

2.3. Java source code generator and compilation server The Java source code generator automatically generates the basic structures for the Java source code for the classes on the class diagram. This generator makes use of information obtained from the methods, from the attributes of the classes, and from the generalization and association between classes. The generator gathers information from the class diagrams each time a user opens the source code editor. Figure 4 shows the Java skeleton program generated by the source code generator. The system also performs a reverse engineering process to students programs to help students to derive class diagrams that are consistent with the program modules.

P %is I multiPk

h? C(rmm%t

public class Student extends Person

-

Computer mycomputer; String t e x t "The s t r i n g i s i n red";

Figure 4. The Java code editor

The compilation server consists of the remote compilation module and the Java Applet and Servlet execution module. It is implemented using Java Sockets. It responds to the incoming requests from students who want to compile the program text associated with a class. The system also checks the dependencies in the class diagram to compile associated classes that have been updated. It identifies the user and saves the user's source code in the user workspace. Compilation and execution can be initiated by clicking the corresponding class icons. The simple edit-compile-cycle implemented through point-and-click interface is important for encouraging active participation beyond classroom environments. When these features are initiated, the system either displays the results or returns appropriate error messages in the information area of the panel. Figure 5 shows a sample dialog for online compilation and execution. The WEBLOOP environment ensures that any classes can be compiled and tested as

150

soon as the skeleton code has been generated. The graphical interface also reduces the need to remember long program names and Web links for execution. This point-and-click execution cycle is more useful in a learning environment than the following conventional cycle for servlet development : 0 0

0 0 0 0

Set up Java, classpath, environment variables and development libraries in local machine Create each Java program using a conventional editor Compile and debug Java programs using the Java Servlet Development Kit (with complicated command line details and parameters) Upload the compiled classes to the appropriate directories Open a Web Brower Type in the URL to execute the Java Servlet classes in the Web server. Document all classes using a systems engineering tool such as the Rational Rose Package [4]

&age uni8tledModel.

ublic int attiibute3; ublio void newOperation30

To compile the source code, users can click If there are errors, once the students have Compile and then Save & Compile on Server modified the codes, messages are displayed on the students’ machines at home. in the menu bar.

Figure 5 . Sample screens of WEBLOOP for on-line editing and compilation

2.4. The Marker Module

After the students finished editing and compiling their programs, they can submit their programming assignment project by providing the subject code and

151

assignment-ID (as shown in figure 3). The marker module helps lectures in testing the programming assignment submissions from students. It generates feedback for students automatically by using a number of test cases for each program. The detailed operation of the marking modules for providing automatic feedback are as follows (as shown in figure 6 ) : a a a a a a a

a

The student submits programming assignments by providing the subject code, assessment-id and user name to the system (Assessment Drop Box). The system compiles the program and stores the assessment file in programming assessment space of the web server. The system returns the submission acknowledgement to the student. The system save student submission record to database. A Testing Module retrieves student program from the assignment web space. The student program outputs are compared with a set of predefined test case from the database. The Maker Module sends test case report to the tutor and the student. The Tutor provide the comment and final grade.

@ Compilation Result

I@

Making Reaulr

Figure 6. The Operations of the Automatic Marking and Feedback Module

2.5. Automated testing and feedback

Before assignment submissions, the lecturer creates test cases for each assessment. These may include the input and output variables. When the students submit their programming assignments, the system retrieves the test cases from the database. For example, if the test case includes three parameters :

152

inputl, input2 and input3, the Web URL for the corresponding submission (using a doGet() method for the parameters) is :

http://158.132.10.186/cyberlab/z~{username}/servlet/QuadraticEq uationServlet?assessment-id=123&test-case-num=2&servlet-na me=QuadraticEquationServlet¶ 1=input 1¶2=input2&par a3=input3 By using the parameters from the Web URL for inputl, input2 and input3, the system compares the results with the corresponding test cases. A test case report is generated as a feedback to each student (as shown in figure 7).

TEST CASE REPORT

Servlet Name : QuadraticEquationServlet Test Case Number: 5

I n p u t Result: 3 / 3 Output Result: 3 / 3 Figure 7. A Test Case Report

3. Conclusion We have developed a Web-based Learning Environment for Object-oriented Programming with automated feedback and assessment. The system provides various features for students to learn Java Servlet programming. Students do not need to install and set up the Java development kits on their local computers. They can work with the system anywhere by launching the application remotely with a browser. Students can also submit programming assignments through the

153

system. It provides automatic feedback by comparing the program output with a set of pre-defined test cases. As part of our hture work, we shall extend the system to provide conferencing facilities to improve communication among students, tutors and lecturers.

Acknowledgments

This project is supported by Grant-no 4-ZO7P of the Hong Kong Polytechnic University.

Reference

[1]

Current version of WEBLOOP is available for public access at http:Ncyber.comp.polyu.edu.hWcyberlablwebljsp

[1]

Decker, R. and Hirshfield, S., Top-Down Teaching: Object-oriented Programming in CS 1 (Brooks Cole, 1993).

[1]

Kiilling, M. and Rosenberg, J., An Object-Oriented Program Development Environment for the First Programming Course, Proc. 27" SIGCSE Technical Symposium on Computer Science Education, ACM, Philadelphia, USA, March 1996,83-87.

[1]

Rational Rose: The Assurance of Quality, http://www.stylusinc.net/technology/rational~rose.shtml


WEBITS AND QUALITY CONTROL IN MARKING OF EXAMINATION SCRIPTS PHILIP TSANG, REGGIE KWAN, ANDREW K LUI, HENRY LO The Open University of Hong Kong

Abstract An important component in web-based learning is the quality control in examination marking. This is of paramount in the case of large web-based courses such as our CT212 Network Programming and Design course, which had over 600 students and involved more than half of dozen script markers. This paper details the principle of examination paper marking, the potential problems of using a marking scheme, how our home-grown web-based conferencing tool, WEBITS, can help in the synchronization of marking among markers, and some recommendations to improve the overall quality of marking from the markers’ perspective.

Introduction The tasks of a marker involve making an assessment of answers for both closed and open-ended questions in network programming and design taking into account a number of objective and subjective aspects of the marking scheme. This paper introduces the performance of markers in examinations from the perspective of markers. This paper also analyses how our home-grown web conference tool, WEBITS, can help in the synchronization of marking among markers.

Examinations in the OUHK The examination process of the Open University of Hong Kong is very rigorous, which has pleasantly surprised many of our external examiners who are associated with traditional universities. The seven key stages of preparation for course examination are encapsulated in the following[I]: Setting the examination paper Nominating and appointing examination script markers Involving external examiners Conducting examination script markers coordination meetings Examination script monitoring 9 The standardization and Award Meeting Submission of course report

. ..

.

The focus of this paper is on the quality of examination marking, and we will therefore concentrate on stage four of the examination process. 155

156

Potential Problems of Marking Scheme

Taking the marking scheme of the examination for CT212 in April 2004 as an example, some common pitfalls of the marking scheme were revealed during the discussion meeting among the markers, as follows: 1. The answers in the marking scheme are not exhaustive. Very often, in tackling the questions, there are a variety of ways to solve a particular question. As a marker, it is important to list out all possible combinations so that the marks are aligned with what answers are acceptable. But still, the list may not be complete. Granting or not after the meeting is always difficult for markers. 2. The answers in the marking scheme are not necessarily the most appropriate. Markers also need to verify all the answers presented in the student’s answer book that may have a creative or even brilliant approach. The marker should not take it for granted that answers not like those in the marking guides are always incorrect. Sometimes, it is necessary to amend and supplement the answers. 3. The answers in the marking scheme may not be concise enough. The suggested answers should not be too vague to allow personal assumptions. As markers, they need to make sure that the marking scheme is precise and concise enough for their use so that it will not cover too many possible answers by any kind of deduction. 4. The answers in the marking scheme are correct but may not be reasonably expected from normal students. This problem is somewhat tricky as there is a need to classify whether some difficult questions are within the scope of the course. While there is a need to introduce some difficulties in the questions for differentiating students, such a level of difficulty, however, may not be expected by students, or even tutors. Hence, markers need to provide feedback to the coordinating staff on whether a particular answer may be too harsh for students. 5 . The questions sometimes may not be clear enough. Students may not know the exact requirements of the questions. Hence, wrong answers may be seen that resulted from wrong interpretation. Markers need to suggest some potential interpretation of questions whenever doubts are found on the questions. The marking scheme should be adjusted accordingly. 6 . The allocation of marks may not be optimal. The marker, based on his or her experience and knowledge, should be able to comment on the allocation of the marks. In this regard, we need to identify the

157

objectives of the questions and assess whether the answer can meet the core objectives. The higher the importance, the higher the marks that should be given. The breakdown should be clear. 7. Answers to the questions may be evolving. Although it is unlikely, there is a need to address the timeliness of the answers to particular questions as technologies are evolving very fast. 8. Although the same marking scheme is delivered to the markers, the scheme may be subject to personal interpretation. In many cases, markers may come from different backgrounds and have different interpretations on both questions and answers. Some points may be considered valid and relevant very differently. Hence, there is a need to align the understanding for both questions and answers. The common ground is critical to ensure fairness. 9. Although a marking scheme is used, it is not uncommon that students will have different wordings to present the same or similar ideas. It is up to the markers to judge whether the answers are equivalent to the suggested answers, which is unavoidable in many cases. Should there be any major deviation, the case should be reported to the course coordinator and all practitioners should share their views so that the marking scheme can evolve, yet remain unified; and 10. It is rather difficult for programming-type questions to have a clear marking scheme as the programming logic can be very different. Although some framework has been set, students may have different levels of accuracy. We cannot expect that a completely executable program can be written in the examination, but what kind of tolerance is acceptable is extremely difficult to define. Running the exam markers coordination meeting The course coordinatorAecturer chairs the exam markers coordination meeting. The aim of the meeting is to standardize the marking by training the examination paper markers to become familiar with the marking scheme, which is usually delivered to the markers two to three days before the coordination meeting. The scenario goes like this. Assume the examination of a course is finished and everything went smoothly. Our Examination Office (EO) collects all the examination scripts from various distributed examination centers. The EO then records the number of students attending, and sorts the examination scripts into equal batches to be collected by script markers. The EO also reserves two

158

batches for the course coordinator to review. These batches are labeled Batch X and Batch T. For Batch X, the course coordinator selects three scripts from this batch as sample scripts. These are used as sample scripts for pilot marking before the coordination meeting. In addition to the three sample scripts, the course coordinator also marks about 10 other sample scripts so that he or she can get a feel of the students’ performance - and see if the marking scheme works. In the meeting, problems the course coordinator and the script markers have with the marking scheme are discussed and collective decisions on making changes and fine-tuning are performed. This helps to eliminate differences in marking standards among different markers. While in the past these meeting were ALL conducted face-to-face, with the Web and with appropriate tools (such as WEBITS), the meeting is no longer bound by bricks and mortar.

WEBITS and Marking Coordination Meeting To facilitate the markers’ meeting, our home-grown interactive conferencing tool is used. The tool is called WEBITS[%]. For those readers who are familiar with commercial products such as Intenvise, our WEBITS contains a subset (but the most relevant) of the hnctionalities of Intenvise, but with a much lower total cost of ownership. Specifically, it contains: Application sharing, audio conferencing, whiteboard, chat-room and video conferencing. The WEBITS design philosophy, as captured in the logo (See Figure l), is that it is the interactions and motivation of all stakeholders that enhance the course.

~-

Interactive Tutoring System

Figure 1: WEBITS logo In preparation for the meeting, the three sample scripts are scanned as pdf for use in the WEBITS meeting. The pdf format can easily reduce a word file by 90 per cent, and thus reduce downloading time for the client. Figure 2 shows a typical WEBITS login session.

159

Through the application sharing facility, all markers can synchronize their view of the Batch X scripts for discussion. For essay typelshort answer questions, markers and the coordinator can see and discuss any problem questions via the voice-chat tool of WEBITS. Any programming code in question can also be handled via real-time telnet application sharing. Figure 3 shows our current WEBITS System Architecture.

Figure 2 WEBITS Login Screen Using the application sharing facilities, we can see how one marks a particular question. This is then open for discussion: agree, disagree, enhancement, revision, trash and remarks. It also has the ability to record the meeting discussion for future auditing purposes. What is Quality of Examination Marking? In subjects like network programming and design in computer science and network programmes, things are often either right or wrong. It is relatively easy to devise assessment criteria for examinations. Even though design issues and recommendations are subjective in nature, there is an identifiable framework of a good answer or an unsatisfactory answer to a particular question. Such a framework can be turned into a marking scheme of a flexible kind, which enables the characteristics of good and less-good answers to be compared and

160

Server

Central Server

Figure 3

WEBITS System Architecture [3]

contrasted. The quality of marking relies on whether such a framework is common and consistent among all entities including the course coordinator, tutors, students as well as external reviewers. Very often, markers face the problem of being too rigid or too flexible for open-ended questions. To avoid this situation, good marking of examination scripts should satisfy the following basic principles of transparency, openness and fairness. For transparency, the whole mentoring process should be transparent to all parties. In this regard, the objectives of the courses and how the students are being evaluated must be delivered and communicated well, not only to the students but also to tutors and external moderators alike. The appeal mechanism is also important for the students to build confidence in the entire quality system. Such guidance on evaluation criteria must be delivered at the very beginning of the course. For openness, it is important to ensure that every stakeholder is provided with the same set of information, and such information should be readily available to all parties. As such, in our CT212 Network Programming and Design course, there are a number of dissemination channels for tutors, students and staff [4]. For example, online discussion board and news announcement pages, online and face-to-face tutorials and surgeries that are open to all students and relevant staff. In particular for marking an examination, it is important to make sure that all markers are provided with the same marking scheme. However, quality

161

assessment is always subjective and is subject to personal interpretation. In order to ensure that the same scheme is delivered to the markers, a discussion meeting is critical for all markers to clear some personal assumptions for alignment. For fairness, all students must be on an equal footing so that the marking can be accurate and consistent. The markers should not be biased by a particular interest, and the following should be observed: 1. All examination scripts should be anonymous for the purpose of marking (e.g. use of student number only or examination number only); 2. Candidates should be assigned to a marker who has no direct relationship (e.g. the tutor should not mark the answer scripts of hisher tutor group); 3. The assessment should adhere strictly to the agreed marking scheme; 4. The marking scheme should be commonly understood and agreed among the markers. There should be no personal assumptions and flexibility; 5. Randomly sampled checking on marked scripts is necessary and should be examined by an n-tier strategy. Quality Assurance and Control Process Having considered the potential problems of marking, an appropriate quality assurance process needs to be in place before conducting the marking process. Proactive and preventive measures are critical here. For example, the discussion meeting and marking of sample scripts are an extremely useful means for this purpose. Besides, quality checking is also vital to ensure accuracy, compliance and consistency. Sample checking and second moderation may be introduced to ensure the performance of the marking. To ensure the quality of a marking scheme, the following draft “To do” list is drawn up for possible improvement of marking scheme: 1. To explore an exhaustive list of answers in the marking scheme;

2. To verify and confirm the answers in the marking scheme; 3. To make the answers in the marking scheme concise and descriptive; 4. To ensure the answers in the marking scheme are reasonable; 5. To clarify unclear questions, looking for different potential interpretation of questions; 6. To adjust the allocation of marks according to the importance of the issues being examined; 7. To allow new answers to the questions as time changes;

162

8. To align the common understanding for both questions and answers; 9. To avoid personal judgment, seeking advice from course coordinating staff should there be any major deviation; and 10. To list the levels of acceptable tolerance. Conclusion

To conclude, from the human perspective, all markers should observe the principles of fair play and transparency. A suite of quality assurance and control processes should be in place. The marking scheme should be normative, descriptive, precise and concise so that the markers can compare themselves against the standard answers as they might perceive then. By weighting the key principles of assessment to reflect the goals of the course, staff involved should align their common understanding and allow the evolution of a marking scheme to ensure its fitness for purpose. From the technological perspective, the Web clearly facilitates the quality assurance process by allowing well planned meetings to be conducted, recorded and possible audited. Acknowledgements

The authors wish to thank the tutors of the courses that we coordinated at OUHK. The tutors have provided constant feedback and inspiration in our teaching. We also want to record our appreciation for the critical comments of Dr Rex Sharman, Education Technology and Publishing Unit (ETPU), OUHK. Thanks also go to Prof. David Murphy, Director of our Distance Learning Research Centre, for constant support not only in the WEBITS project but in many other joint activities. References

1. Course Coordinator Training Manual, OUHK. 2. Murphy, D., & Tsang, P. (2004) WEBITS Research Project Report, OUHK. 3. Chan, J., Murphy, D., & Tsang, P. (2005) WEBIT Users Manual, OUHK. 4. Tsang, P., Fong, J., & Tse, S. (2004) Using E-learning Platform in Open and Flexible Learning, New Horizon in Web-Based Learning, World Scientific, pp.214-225.

PART FOUR Human Factors


SOCIAL CAPITAL CREATION AND RECIPROCITY IN ONLINE LEARNING PLATFORMS ANDREW K. LUI School of Science and Technology Open University of Hong Kong, Hong Kong YANNIE H. Y. CHEUNG Department of Sociology Chinese University of Hong Kong REGGIE KWAN School of Science and Technology Open University of Hong Kong, Hong Kong This paper investigates the creation of social capital in online courses. The advantages of using social capital to abstract the complex set of beneficial factors for online learning and collaboration are exploited. The role of online learning platforms in facilitating social capital creation is highlighted in a strategy based on trust development and reciprocity. The strategy is realized in social interaction components that can be integrated into any online learning platform. The paper reviewed several social interaction components that have been successfully deployed in an online course.

1. Introduction

This paper investigates the creation of social capital in online courses. It specifically explores the system aspect of social capital creation and the role of online learning platforms in facilitating beneficial social interaction processes. This approach aims to share the responsibility expected of online instructors in community cultivation [2]. The remainder of the paper covers the background of this work, which begins with an introduction to social capital. Social capital includes other concepts such as trust, reciprocity, and community, and these concepts will also be briefly discussed. The paper then describes a strategy for social capital creation in online courses, discusses the role of online learning platforms for this purpose, and reviews several prototype components based on the strategy. 165

166

These components will serve as case studies of the strategy. The paper finishes with a general discussion and suggestions for future work. 2. Background on Social Capital

The notion of social capital has been developed to abstract the productive conditions and values embedded in social interaction and communities. High social capital describes a state of social organization and social cohesion [3], and it facilitates collective actions for mutual benefits [4]. These virtues of communities, made possible by social capital, have been shown to correlate positively with effective learning in colleges [ 5 ] , trust in institutions [6] , and access to information and knowledge [7][8]. Social capital offers a convenient handle for dealing with the complexity of social interaction, human characteristics, and conditions of community development. The notion adopts a rational and economic approach in considering the participation and engagement in a community. It has been shown that social capital has good descriptive power in studying a range of social issues, including the development of online learning communities [ 1][9]. 2.1. Social Capital Creation

Many researchers suggest that an increase in social capital is equivalent to an improvement of socially productive conditions [ 11][7][9][lo][ 121. The important conditions include trust, reciprocity, membership scheme, and shared norms, values, and rules. Trust gives the assurance that the relationship is reciprocal - a member can confidently expect obligations to be fulfilled. This encourages altruistic acts for the benefit of the wider community. Membership of a community allows individuals to be identified for sanction, however, if they are found to exploit trust and violate reciprocity. The suspension of membership is usually a sufficiently threatening sanction if access to community resources is tied with membership. The use of sanction can be complemented by institutionalized norms and values. Social norms and values can promote positive attitudes and actions among the members, and they can also provide a yardstick as to whether a member is still committed to the community and considered trustworthy. Trust is generally regarded as the most important indicator of social capital. Trust is the sense of comfort, belief, and confidence that one can rely on a person or a social structure [ 131. A community with an abundance of trust encourages reciprocity, participation, and volunteerism, which allow more donations and other resources to be placed at the disposal of the community-atlarge. Trust itself exhibits reciprocity in that if a person trusts another then it is

167

likely that the trust is mutual. Reciprocity also occurs in cooperation and relation building. Cooperation happens in a group if each trusts others to reciprocate contributions. Reciprocal fulfillment of obligations is a positive experience that breeds trust, and conversely, trust diminishes after a failure of expectations. The cultivation of a cooperative community can take the form of a circular process of reciprocity and trust reinforcement. In the circle, trust encourages cooperation, volunteerism and other forms of reciprocity, and then these activities strengthen trust [ 6 ] . This is a process of positive reinforcement activities, in which social capital increases. The circular process can operate in a destructive mode, however, in that trust can diminish when reciprocity does not happen as expected. This perspective of social capital creation can be succinctly described as follows: encouraging the positive reinforcement process while preventing the negative reinforcement process. This takes deliberate planning to encourage participation in the positive reinforcement process. Reciprocity can be considered in both specific and generalized contexts. Specific reciprocity describes the situation where the fulfillment of obligation is directed to the original provider or contributor of resources. In generalized reciprocity, the original contributors are not concerned with the return of favors from the recipients of resources. Instead they expect to receive something desirable from the community in the future, perhaps intangible goods such as recognition. The generalized form of reciprocity connects individuals with the wider community and it is of more significance to social capital creation than specific reciprocity. 2.2. Social Interaction in Online Environments

A discussion of social capital creation for online learning inevitably involves the issue of social interactions in virtual environments. There have been considerable discussions on the characteristics of online social interactions. Critics often suggest that social interaction through computer mediated communication is impersonal and even hostile [ 141. Conversely, many others report the development of genuine online relationships that are fruitful and intimate [ 181. The reduced cue phenomenon is considered to be an undesirable characteristic of computer mediated communication that social interaction has to overcome. Face-to-face communication includes verbal communication and other visual cues such as facial expression and gesture, and other cues. In online social interaction, the communication facilities determine the permissible modes and forms of communications, which are often restricted to just written

168

text. The reduction in cues decreases the sense or feeling of the other, which is known as the concept of social presence. Social presence is regarded as the basis of the feeling of relation and friendliness [18][16]. There are opposite views claiming that text-only communication leads to high quality academic discussion from a cognitive aspect [34], and it also supports the development of close relations through the hyperpersonal effect [ 151. These cases merely demonstrate the adaptive ability of human beings, and they do not devalue the importance of cues in social interactions. Social affordance [21], sociability [21], social translucence [191, and social awareness [9] are concepts introduced to address the issue of social presence in online environments. The minimalist visualization approach [20] uses a computational method to present visual cues between online participations. The Babble Cookies, for example, uses graphics to indicate the activities of other participants - and who are still participating actively or slipping away. The Group Awareness Widget [2 I] follows a similar approach in which the activities of each online participant from the present to the past are shown graphically. 3. Role of Online Learning Platforms in Social Capital Creation

In this work we investigate the role of an online learning platform in social capital creation. Online learning platforms provide the only channel for social interaction to happen and they can play a progressive role. Based on the trust development model discussed in the previous section, we propose the following three-item strategy of designing social interaction components for online learning platforms: 1. Show Reciprocity. Trust building requires not just the occurrences of reciprocity but also an awareness of reciprocity. Online activities of a reciprocal nature are not visible because of the reduced cue problem. Online learning platforms can include components to display information about reciprocity. 2. Provides Opportunities of Reciprocity. Participation is the precondition of reciprocity and offering a variety of activities encourages individuals with varying needs and concerns to participate. Providing communication facilities and hoping reciprocity will occur is not sufficient. Online learning platforms can include a range of components for personal contribution and sharing of resources. Some components should be concerned with developing initial trust while others with developing deeper trust. The components should also vary in the effort required, the ability required, and

169

the nature of interaction. The more components provided, the more online learners can potentially participate. 3. Reward Reciprocity. Reward can reinforce a desirable activity such as trust building and reciprocity. Online learning platforms can include components that give status or more access rights to those practicing reciprocity. There are two more issues concerning the design of the components. If an online learning platform provides information about online participation, there will be a concern about online privacy. Online privacy is a complex issue in that the perception of privacy can affect the social psychology of computer mediated communication [22]. A simple way to approach this issue is to provide an opt-out option for online learners. The opt-out option would stop the using of an individual’s participatory infomation and at the same time stop their access right to the social interaction components. The other issue is about using a minimalist approach [20] to design the social interaction components. A minimalist approach is economical in the development. In the following minor sections we will discuss several components for online learning environments in the context of social capital creation. Most of the components are not new but they are reviewed from the new perspective of social capital. Specifically, the elements to be reviewed include the nature of the component, the social capital creation capability in the light of the three-item strategy above, the effort or ability required of the user, and the involvement required for implementation. 3.1. Prototype I : Presence Monitor

The Presence Monitor indicates the virtual location of participants currently online and their identity (Figure 1). This component supports the visualization of reciprocity in participation. Current online participants are aware of each other and the nature of their activities through this component. One can determine the identity of the active participants who share the same commitment to participate regularly and to make progress in the course. Together with other participatory information, one can also determine who the lurkers are. The identification of responsible participants allows preferential offering of more help. The use of this component requires virtually no effort from online participants if it is placed on the top level of an online learning environment. The implementation requires the server-side to keep track of the virtual whereabouts of online participants and the client-side to refresh the information through standard HTTP refresh or a JavaScript script.

170

Figure 1. A Presence Monitor with boxes showing various virtual locations such as the offce for download and submit assignments, or the study center where online learning materials are kept. The identity of the online participants is shown when the mouse pointer moves over a location box.

3.2. Prototype 2: Opinion Polls The Opinion Polls component allows online students to express their opinions on an issue (Figure 2). The poll results are displayed only after a vote is cast. This component provides an opportunity to share personal opinions reciprocally. Opinion polls can provide rich information about the opinion, the status, and the need of an online community. The use of this component requires a little effort from online participants. It demands minimal effort if an opinion poll is placed on the front page in an online learning environment. Participation is a simple matter of reading and selecting a choice. The implementation requires a web application development with functions consisting of poll authoring, poll display, data store, and result display. I can confidently write programs with Isnd while structures.

I can confdeentlywrile proarams WiIh ifand while strudures.

Strongly Agree

Figure 2. The Opinion Polls Component offers choices of opinion to a question. An online student must cast a vote first before the poll results are displayed.

3.3. Prototype 3: Interaction Table The Interaction Table shows the interaction frequencies with each other online student (Figure 3). This component can highlight information about reciprocity present in asynchronous discussion. It shows the number of messages written as a reply to each peer, and also the number of messages written as a reply from each peer. It makes clear who have chosen to reciprocate and who have been getting help without making a contribution. This component provides the

171

information flow that allows the acknowledgement of individual contributions and contributors. The use of this component requires virtually no effort from online participants. The implementation involves the extraction of information from discussion forum systems and formatting the information. Asynchronous Discussion Interaction

Figure 3. The Interaction Table displays the amount of messages exchanged with every other online participant. The Reply column shows the number of reply messages to the peers and the Received column shows the number of messages received from the peers.

3.4. Prototype 4: Buddies List

The Buddies List shows the participatory and interaction information of selected peers who are considered close friends (buddies) (Figure 4). This component provides an opportunity for an initial trust to develop into deeper trust. One can monitor the fulfillment of the obligation of being a close friend - regular participation and exchange of information. The buddies list is automatically carried from one course to another to maintain continuity of trust development. The use of this component requires a little effort from online participants in editing the buddies list. When a peer is added to the buddies list, a message is automatically sent to the peer containing an invitation for reciprocal addition of the originator as a buddy. The implementation requires a web application development with functions consisting of buddies-list editing, data store, and data display.

172 Buddies List

Figure 4. The Buddies List component indicates the interaction status between an online student and selected peers. The first column shows the ID or the alias of buddies. The second column of the table shows the participatory status that shows the whereabouts of a peer if online or the previous login time. The third column shows if a mail or a real time alert is received from a peer. Only buddies can exchange mail and alerts. The last two columns show the exchange of discussion messages in asynchronous discussion forums.

4. Conclusion

An investigation of the creation of social capital in online courses has been carried out. The paper reviewed the importance of sociality to learning and the need to cultivate social interactions in online communities. The paper then referred to previous research that the notion of social capital is a convenient handle in developing online communities, and proposed a strategy based on developing trust and encouraging reciprocity. The role of online learning platforms in social interaction was highlighted, and the paper suggested the realization of the strategy by adding social interaction components for online learning platforms. The social interaction components reviewed have been experimentally deployed in the course web site for a couple of courses (Figure 5). An important further task is to investigate empirically how the social interaction components affect trust development and reciprocity in online courses and, in addition, the social health of the courses.

173

Figure 5. Several social interaction components are shown on the front page of a course web site. Some components such as the Asynchronous Discussion Interaction Table are deployed in other pages.

Acknowledgment The authors are gratehl to Dr. Rex Sharman of OUHK for his useful comments about this paper. References 1. Schwier, R.A., (2002). Shaping the Metaphor of Community in Online Learning Environments, the International Symposium on Educational Conferencing. The Banff Centre, Banff, Alberta. 2. Palloff, R., & Pratt, K. (1999). Building Learning Communities in Cyberspace, Josey-Bass Publishers, San Francisco. 3. Timms, D., Ferlander, S., and Timms, L., (2001). Building Communities: Online Education and Social Capital in Szucs A, Wagner E and Holmberg C. eds., Learning Without Limits: Developing the Next Generation of Education. Proceedings of the EDEN 10th Anniversary Conference. 4. Stone, W., (200 1). Measuring social capital: Towards a theoretically informed measurement framework for researching social capital in family and community life, Research paper no. 24, Australian Institute of Family Studies. 5. Smith, M. H., L. J. Beaulieu, and Seraphine, A. (1995). Social capital, place of residence and college attendance. Rural Sociology, 60(3), pp. 363-381. 6. Putnam, R. D. (1993). Making democracy work: Civic traditions in modern Italy. Princeton University Press, New York. 7. Coleman, J. S. (1988). Social capital in the creation of human capital. American Journal of Sociology, 94, pp. 95- 120.

174

8. Prusak, L. and Cohen, D. (2001). In good company: How social capital makes organizations work. Boston, Harvard Business School Press, MA.

9. Daniel, B., Schwier, R.A., and McCalla, G. (2003). Social Capital in Virtual Learning Communities and Distributed Communities of Practice, Canadian Journal of Learning and Technology, 29(3). 10. Putnam R D. (2000). Bowling alone: America's declining social capital. Journal of Democracy, pp. 65-78. 11. Hobbs, G. (2000). What is Social Capital: A Brief Literature Overview, Economic and Social Research. Foundation, Dar es Salaam, Tanzania. 12. Falk and Kilpatrick (1999). What is Social Capital? A Study of Interaction in a Rural Community, University of Tasmania, Australia. 13. Annison, M., and Winford, D., (1998). Trust Matters: New Directions in Leadership. 14. Parks, M.R., Floyd, K., (1996). Making Friends in Cyberspace. Journal of Computer-Mediated Communication, l(4). 15. Walther, J. B. (1996). Computer-mediated communication: Impersonal, Interpersonal, and Hyperpersonal interaction. Communication Research, 23, pp. 1-43. 16. Short, J., Williams, E., and Christie, B. (1976). The Social Psychology of Telecommunications. John Wiley, New York. 17. Hulsmann, T., (2003). Texts That Talk Back - Asynchronous Conferencing: A Possible Form of Academic Discourse? In U. Bernath & E. Rubin (eds.), Reflections on Teaching and Learning in an Online Master Program: A Case Study, pp. 75-120. 18. Walther, JB (1 999). Visual cues and computer-mediated communication: don't look before you leap. Paper presented at the Annual Meeting of the International Communication Association, San Francisco. 19. Erickson, T., Kellogg, W.A. (2002). Social translucence: Using Minimalist Visualizations of Social Activity to Support Collective Interaction, In K. Hook, D. Benyon, A.Munroe (eds) Readings in the Social Navigation of Information Space, Springer. 20. Erickson, T., Laff, M. R., (2001). The design of the 'Babble' timeline: A Social Proxy for Visualizing Group Activity over time. In Proceedings Human Factors in Computer System (CHI 2001),pp. 329-330. 21. Kreijins, K.,Kirschner, P.A. and Jochems, W., (2002). The sociability of computer-supported collaborative learning environment, Educational Technology and Society, 5( 1). 22. Tu, C., (2002). The measurement of social presence in an online learning environment. International Journal on E-Learning 1 (2),pp. 34-45.

ARE TEACHERS IN HONG KONG READY FOR E-LEARNING? SO, KOON KEUNG TEDDY The University of Hong Kong [email protected]

Abstract There is a common understanding that the twenty-first century is a more globalized and knowledge-based era. In response to this rapid change, Hong Kong is trying to transform hersey into an information society. A lot of education reforms were implemented to our schools in the past f a 0 years. Based on the well established hardware infrastructure and the lesson learntfrom the first IT plan (1998 - 2003), the Education and Manpower Bureau released its second Information Technology strategic plan, focusing on the real change of pedagogy, the promotion of life-long learning and e-learning, the use of wireless technology as the extension of the existing wired network, and the new roles of parents as well as students in the life-long learning environment. This research adopted a survey method to conduct the study of e-learning readiness of teachers in schools of Hong Kong. Results indicate that teachers in Hong Kong are not very prepared for using the e-learning technologies in teaching and learning. There are differences in readiness perceived between males and females, secondary school teachers and primary school teachers, and teachers of secondary schools in different bandings. Based on the data found. recommendation is made to improve the situation.

Introduction With the support of new communication technologies, there is a common understanding that the twenty-first century is a more globalized and knowledge-based era, treating knowledge as a commodity. In response to this rapid change, Hong Kong is trying to transform herself into an information society. A lot of education reforms were implemented to our schools so as to equip young people to take the challenge in the past few years. We had our very first five-year Information Technology strategic plan implemented in all primary and secondary schools from 1998 to 2003. In this five-year period all schools set up huge computer networks, having 9 1 and 247 networked computers in primary schools and secondary schools respectively. Moreover, all schools have broadband Internet connections, ranging from 1.5 to 10 Mbps. Although huge resources had been poured into the project of Information Technology in Education, the change of pedagogy I paradigm shift was not obvious. The classroom practice changed only little. Students still learn in a passive way, without much participation and interaction with other students and teachers.

Based on the well established hardware infrastructure and the lesson learnt from the first IT strategic plan, the Education and Manpower Bureau (EMB) released 175

176

its second Information Technology strategic plan, bearing the title “Empowering Learning and Teaching with Information Technology” in July 2004. The focus of the second strategic plan is a bit different from the first one. The first plan emphasized the initial set-up of the hardware infrastructure and teacher training. The second strategic plan focuses on the real change of pedagogy, the promotion of life-long learning and e-learning, the use of wireless technology as the extension of the existing wired network, and the new roles of parents as well as students in the life-long learning environment. According to the report prepared by both IBM and the Economist in 2003, the overall ranking of e-leaming readiness of Hong Kong is 19* out of 60 countries around the world (The Economist & IBM, 2003). One of the important factors influencing the success of e-learning is teacher training. As the way to deliver the online curriculum is new and different from the traditional one, the instructors must be trained to make the most of updated teaching methods. “An ineffective teacher can waste the time of 30 or 40 students. But bad teaching online can touch thousands. ‘We can create mass damage quickly.’” (The Economist & IBM, 2003; p. 12) As the readiness of teachers to use the new technology is critical to the success of implementing e-learning in schools, it is worthy to investigate if and how they are prepared to embrace the new technologies in their teaching and learning activities. The purpose of this research is thus to find out how ready the teachers of secondary and primary schools are to use the new technologies, and what factors are influencing their readiness. It is hoped that the experience gained from this research is beneficial to other countries exploring the use of e-learning technology in new teaching and learning activities. Literature Review e-Learning readiness assessment helps an organization to design e-learning strategies comprehensively and to implement its ICT goals effectively (Kaur, 2004). Learners must be “e-ready” so that a coherent achievable strategy that is tailored to meet their needs may be implemented (infodev, 2001). In sum, this readiness assessment provides key information to organizations to supply solutions that can cater to the specific needs of each learning group (McConnell International, 2000).

Before implementing e-learning programs, organizations need to expand the usual needs assessment process by creating a high-level requirements document that includes: 1. Objectives (macro organizational objectives and micro target learner population objectives); 2. An e-learning readiness score; 3. A list of

177

advantages and potential obstacles to e-learning adoption; and 4. A list of possible e-learning configurations (Chapnick, 2000). Chapnick designed a model for measuring the e-learning readiness of an organization aiming at answering the questions: a. Can we do this?; b. If we can do this, how are we going to do it?; and c. What are the outcomes and how do we measure them? His proposed model groups different factors into eight categories: Psychological readiness. This factor considers the individual's state of mind as it impacts the outcome of the e-learning initiative. This is considered one of the most important factors and has the highest possibility of sabotaging the implementation process. Sociological readiness. This factor considers the interpersonal aspects of the environment in which the program will be implemented. Environmental readiness. This factor considers the large-scale forces operating on the stakeholders both inside and outside the organization. Human resource readiness. This factor considers the availability and design of the human-support system. Financial readiness, This factor considers the budget size and allocation process. Technological skill (aptitude) readiness. This factor considers observable and measurable technical competencies. Equipment readiness. This factor considers the question of the proper equipment possession. Content readiness. This factor considers the subject matter and goals of the instruction. The Ministry of Education of Singapore found that this model is especially useful for principals and HODS who intend to start e-learning in the school (MOE, 2004). However, one of the major drawbacks of this model is that it is designed to measure the readiness of using e-learning in business organizations. It does not fully fit in the school environment. With reference to Chapnick's model, Kaur and Abas (2004) designed a model for measuring the e-learning readiness of the Open University Malaysia. The model consists of eight constructs: learner, management, personnel, content, technical, environmental, cultural and financial readiness. There are findings that show gender differences exist in computer acceptance (Yuen & Ma, 2002; Russell & Bradley, 1997). In his research on 462 middle and high school students, Young (2000) found that there was a significant gender differences in computer attitudes. The male domain scale showed that boys were more likely to have claimed computers as a male area. Russell and Bradley (1997) found that male teachers reported significantly greater

178

confidence with computers than females did. Therefore, they recommended that the design of teacher professional development should take into account the gender difference, taking care of the particular needs of female teachers.

Methodology This research adopted a survey method to conduct the study of e-learning readiness of teachers in schools of Hong Kong. A questionnaire was sent to 200 teachers of secondary and primary schools in the period from December 2004 to January 2005. All items were measured on a five-point Likert scale, with 5 as strongly agree and 1 as strongly disagree. 148 were completed and returned. 13 1 of them were valid. Descriptive statistics, one-way analysis of variance (ANOVA), and Post Hoc tests were applied to analyze the data. Data Findings and Discussion Results show that there are differences in readiness perceived between males and females, secondary school teachers and primary school teachers, and teachers of secondary schools in different bandings. The results of this study are summarized in tables 1 , 2 and 3.

Table 1 : e-learning Readiness of Teachers of Primary vs. Secondary Schools (One-way ANOVA)

It is worth noticing that in many aspects the primary school teachers have significant different perceptions from those of secondary school teachers. Although officially the amount of IT training in terms of time and opportunity offered to both primary and secondary school teachers from the Education and Manpower Bureau (EMB) is the same, primary teachers still consider themselves know less about what e-learning is. Furthermore, they not only do not have enough confidence on themselves in perceiving the abilities of their students, but they also feel that primary students do not have enough IT competencies to use e-learning technologies.

179

Moreover, their confidence in their principals’ understanding and support of using e-learning in teaching and learning is not as high as their counterpart, the secondary school teachers’. This can be originated from the comparatively shorter history of having computers in primary schools. In Hong Kong, secondary schools had to teach computer studies as a subject since 1982. Since then, the teachers and students have chances to use computers. Culture and confidence of using I operating computers were gradually built up in secondary schools. Most of the secondary school teachers are university graduates who most probably have experiences of using computers during their university studies. Not until the year 1998, the year when the first IT strategic plan in education was launched, primary schools in Hong Kong had any computer. However, it took another one to two years to convert normal classrooms into computer laboratories and to complete the wiring, and set up all computer hardware. As a result, in general, primary teachers as well as their principals had only five to six years’ time to learn how to operate computer, to try integrating IT into their teaching, and even less time to explore the use of e-learning. Moreover, most of the primary school teachers are diploma graduates and they did not have much chances of using computers in their pre-service training. This might explain the phenomenon why primary teachers do not have much confidence in their principals and themselves in using e-learning technologies as secondary school teachers do. Regarding the issue of accessibility, it is always a problem in different countries. According to the latest statistics from EMB (2004), the average numbers of computers installed in primary and secondary schools in Hong Kong are 97 and 247 respectively. The computer to student ratios in primary and secondary schools are about 1:9 and 1:s respectively. Hence accessing computers is a bigger problem among primary pupils. Although nowadays there are many public terminals for all users, young primary pupils are considered too young to go around by themselves. From Traditional Asian point of view, for their safety the young kids should either stay in school or at home. As a result, their accessibilities to public computer terminals are further restricted. Gender difference is always a controversial topic. As discussed in the section of literature review, some researches found that there is a gender difference while some did not. This research found that there is a gender difference between male and female teachers. In all three questions ‘‘I know what e-learning is”, “I am ready for integrating e-learning in my teaching”, and “I have enough IT competency to prepare the e-learning materials”, male teachers showed higher

180

confidence than female teachers did, although they received equal amount of IT training from the government. The result of this study agrees with the findings of Yuen and Ma (2002) and Russell and Bradley (1997) that there is a gender difference between male and female teachers in perceiving their IT competencies.

to prepare the e-learning materi aI s

Female

60

3.25

181

There are about 400 plus secondary schools in Hong Kong. They can be categorized into three bands, according to the achievements and the performance in the public examination of the pupils taken from primary schools. The highest achievers will go to band 1 secondary schools, and the lowest go to band 3 schools. While sometimes it is hard to draw a line to distinguish the differences between teachers from band 1 and band 2 secondary schools, together their responses are significantly different from that the teachers of band 3 schools. Significant differences can be found between the teachers from band 3 schools and teachers from band 1, or band 2 schools, or both. Teachers from band 3 schools perceived their schools, themselves, and their students inferior to those from band 1 and band 2 schools. At the school level, they perceived their schools did not have a culture of sharing, which is important to the development of a learning organization. At the classroom level they perceived both themselves and the students were not ready for using e-learning. Another vital factor of using e-learning is the support from the parents. Unfortunately, their perception on the support from the parents is the lowest among the three bandings. Recommendation

In the light of these different perceptions from the primary and secondary school teachers, EMB should consider providing more help / consultancy to primary schools, principals, and teachers. Due to their short history of acquaintance with computer, they need more technical support and in-services training so as to build up their confidence in integrating IT in their daily teaching. It is also recommended that the concept of using IT in education should be emphasized in the principal training program so that principals may really take up the role of leadership in developing the new curriculum. Moreover, it is recommended that the design of teacher professional development should take into account the gender difference, taking care of the particular needs of female teachers. In order to gain support from parents in using e-learning at home, more resources should be invested in schools for conducting parents training programs. Conclusion Although huge resources have been poured into IT in education planning, schools in Hong Kong are still at the initial stage of employing e-learning in their daily teaching and learning activities. It is necessary to conduct a needs assessment in full scale before any e-learning program is actually being launched as it is important to know the factors affecting teachers’ computer use and its implications to teachers’ professional development strategies (Yuen & Ma, 2002).

182

References Chapnick, S. (2000). Are You Ready for E-Learning? Retrieved on Dec. 19, 2004 from http://www.astd.org/ASTD/Resources/dvor/article archives.htm. Education and Manpower Bureau (2004). Empowering Learning and Teaching with Information Technology. Galliers, R. (1992). Information Systems Research: Issues, Methods and Practical Guidelines. Blackwell Scientific Publications, Oxford. infoDev (2001). E-Readiness as a Tool for ICT Development. Retrieved on Feb. 24,2005 from http://www.infodev.org/library. Kaur, K. and Abas, Z. (2004). An Assessment of e-Learning Readiness at the Open University Malaysia. International Conference on Computers in Education (ICCE2004), Melbourne, Australia. McConnell International LLC. (2000). Risk E-Business: Seizing the Opportunity of Global E-Readiness Report. Retrieved on Feb. 24, 2005 from http://www.nicconnellinternational.com/ereadiness/EReadinessReport.htm. Ministry of Education. Are you ready for e-learning? Retrieved on Dec. 19, 2004 from http://www.moe..gov.s.g/edumalI/rd/research summaries.htm. Russell, G. and Bradley, G. (1997). Teachers’ computer anxiety: Implications for professional development. Education and Information Technologies, 2, 17-30.

The Economist Intelligence Unit Limited and IBM (2003). The 2003 e-Learning Readiness Rankings: A White Paper from the Economist Intelligence Unit 2003. Retrieved on Dec. 19,2004 from http://www.eiu.com. Young, B. (2000). Gender Differences in Student Attitudes Toward Computers. Journal of Research on Computing in Education, 33(2), 204-217.

Yuen, A. and Ma, W. (2002). Gender Differences in Teacher Computer Acceptance. Journal of Technology and Teacher Education (2002) 10(3), 365-382.

IMPLEMENTATION ISSUES ON THE SPECIFICATION FOR SERVICE QUALITY MANAGEMENT OF E-LEARNING* YI ZHANG~+ 'Graduate School of Education, Huazhong University of Science and Technology [email protected], [email protected] ZHITING ZHU', ZONGKAI YANG3, CHENGLING ZHA03, SANLAN LU4 'Department of Educational Information Technology,East China Normal University, 3EngineeringCenter of Educational Information Technology, Central China Normal University, 'Department of Electronics & Information Engineering Huazhong University of Science and Technology The emergence of e-learning as a kind of education has led a lot of organizations including schools, colleges and corporations to provide web-based learning and distance learning for lifelong learners. The Chinese e-Learning Technology Standardization Committee is developing a specification for evaluating service quality and management system of e-learning .This paper introduces the background on the specification, CELTS24, and its application information. Further, some critical implementation issues are suggested by a case study of questionnaire and evaluation scale that are used by online and post survey.

1. Introduction

CELTSC(Chinese e-Learning Technology Standardization Committee) is chartered by the Chinese Information Technology Standards Committee and sponsored by the Chinese government. To improve the service quality of elearning in China, the aim of CELTSC is mainly to make and develop standards for enabling interoperability of e-learning system and reusability of e-learning resources. It's also responsible for developing related consistency test software and expanding the influence and application of e-learning standards. As one of the important standard in development within the framework of CELTS, the CELTS-24 specification defines a reference model and scale for service quality assessment in the context of e-learning in college education, compulsory education and profession training to measure and evaluate their service elements as service attitude and related condition in implementing instruction, training and management. This specification develops process elements for e-learning service quality management system according to the scale and a process model. The e-learning institution can improve its service This work is supported by CELTSC project grant. supported by grant 20040075 of Hubei Province Instructional Research and A2004 146 of Hubei Province Educational Science Research.

' Work partially

183

184

quality by the results of assessment and the specification to ensure learner’s proper rights to enjoy high service quality finally [l]. 2. CELTS-24 Specification

2.1. ConceptualModel for Service Quality of e-Learning

The definition for the service quality of e-learning is the overall collection of implicit and explicit characteristics that the service can satisfy the customer. At present, conceptual model for service quality from Parasuraman [2] has had a good effect on commerce, which is practicable in business application. Therefore, we make reference to it and construct a conceptual model for service quality of e-learning, which is as follows (see Fig. 1.).

Assurance e-Learner’s Perceived Service e-Learners

+Validity

4

---, Empathy

t

Specification

,

Management Perceptions of Learners expectations 1

Fig.l.Conceptual Model for Service Quality of e-Learning

2.2. Conceptual Frameworkfor Service Quality of e-Learning According to suggestion from experts and e-learners on service quality of elearning, we design questionnaire for service quality of e-learning and develop an online survey program which is hanged on following websites: CELTSC and Shanghai Distance Education [3]. After analyzing the results of investigation on 41 items by principal component analysis of factor analysis in SPSSl1.0, we find if there are five factors, and cumulative of variance has reached 79.248%,

namely, approximate 80% information can e explained and satisfy the condition for factor analysis Fig.2 30

10

10

D

I

m

o

I

Based on the results from variance analysis of items quantitative analysis of rotation factor loaded matrix and qualitative analysis we extract 5 factors and

25 items from 41 items with incorporation and complification and reaction the questionnaire for service quality of e-leaning which is as follows (see Table 1)

~

~~

Reliability:e-learning providers has the ability to perform promised service dependably and accurately 1.I Reliability of education institution 1.2 Reliability of network system 1.3 Reliability of question answer I .4 Reliability of evaluation Responsiveness: e-learning providers would like to help learners and provide prompt service 2.1 Responsiveness of service request 2.2 Responsiveness of teacher 2.3 Publishing information in time Assurance: Faculty and sfaff engaging in elearning are professional and knowledgeable to let the learner trust them andfeel them reliant 3.1 Integrity o f instruction plan 3.2 Providing related information on courses 3.3 Security of private information 3.4 Technology guidance 3.5 Professional knowledge of the teacher 3.6 Complaint mechanism

Validation of learning resources: e-learning providers can offer credible, effective and rich learning resources. 4.1 Scientificity 4.2 Accessibility 4.3 Integrity 4.4 Real-time 4.5 Selection of media Empathy: e-learning providers understand needs of users and can offer individualized service 5.1 Convenient learning schedule and facility 5.2 Assistant service 5.3 Easy to use 5.4 Customized service 5.5 Care 5.6 Comfort environment 5.7 Interactivity

186

The reliability coefficient of whole scale and each factor in e-learning service quality are calculated by Cronbach alpha and split-half (see Table 2)

Table 2 Reliability Analysis Name of the Factors Reliability Responsiveness Assurance Validation of Learning Resources Empathy Scale

Coefficient of Reliability Cronbach Alpha Split-Half 0.8735 0.8432 0.8022 0.8281 0.8509 0.8408 0.8912 0.8589 0.8872 0.8706 0.9264 0.9064

I

The results show that their reliability are very significant, all reliability coefficient are over 0.8 and total reliability is about 0.9, which indicates the new questionnaire is reliable and repeatable. Therefore, the questionnaire is named the scale for service quality of e-learning. To evaluate validation of questionnaire is involved in three respects[4]. Firstly, content validation, which has been assessed by experts and enterprise representative related to e-learning. Secondly, before performing the investigation, the questionnaire has been passed primary test. Thirdly, factor analysis proves that most of the items attribute to 5 factors. Thus, the questionnaire possess reliable content validation, practicability and efficient constructive validation. 2.3. Process Elements in Service Quality Management System

e-Learning is a process-based activity, in order to ensure its service quality, subsystem in e-learning service quality management system has to be identified, confirmed and analyzed. The process elements in e-learning service quality management system keep to the process approach and process model in I S 0 900 1:2000. Four process elements including management responsibility, resource management, service realization and evaluation are determined, and then corresponding sub-process elements are also made (See Fig.3).

187

r

Continual improvement of e-learning service quality

1

Fig.3. The process model for e-learning

3. Casestudy

This case study examines the implementation of CELTS-24. The study takes post and online survey to investigate. The questionnaire is on the following websites: CELTSC, Shanghai Distance Education, Online Education College Renmin University of China, Distance Education College of East Central Normal University. During half a year more than 900 e-learners have submitted their answers. As a result, 589 available answer sheets are selected.

3.1. Evaluation the Service Quality of e-Learning

According to the questionnaire, the results of expected service (ES) and perceived service (PS) can be gained. If ESPS and there are significant gaps between expected service and perceived service quality of e-learning. Verified by T test, it indicates that these gaps exist indeed, so most e-learners don't satisfy the present service quality of e-learning.

3.2. Effect of Some Variableson Service Quality In order to reveal the reason why e-learning doesn't reach anticipative effect, we analyze 5 demographic indexes including gender, age, region, profession and education, and 3 variables related to learning manner including major, learning place and learning frequency to aim at investigating whether each of them has an effect on 5 factors of service quality on e-learning or not. When age is as independent variable, we divide age into six groups, such as age

Web-based Learning: Technology And Pedagogy - Proceedings of the 4th International Conference

Robotic Sailing: Proceedings of the 4th International Robotic Sailing Conference

Proceedings of the 4th International Conference on Southeast Asia

International Mathematical Conference 1982: Proceedings

Systematic Biology: Proceedings of an International Conference

Proceedings of International Science Education Conference 2009

Medical Textiles: Proceedings of the Second International Conference and Exhibition

Symplectic Geometry and Mirror Symmetry,: Proceedings of the 4th KIAS Annual International Conference

Proceedings of the International Conference on Colloid and Surface Science

Semigroups and Formal Languages: Proceedings of the International Conference

Proc. 4th international conference on data mining

Characterization and Metrology for ULSI Technology 2000: International Conference (AIP Conference Proceedings)

PARLE '92. Parallel Architectures and Languages Europe: 4th International Parle Conference, Paris, France, June 15-18, 1992. Proceedings: Proceedings 4th

Euro-Par '98, Parallel Processing: Proceedings of the 4th International Euro-Par Conference

Euro-Par '98, Parallel Processing: Proceedings of the 4th International Euro-Par Conference

Understanding Pedagogy: And Its Impact on Learning

Proceedings of the Twentieth International Cryogenic Engineering Conference (ICEC20)

Proceedings of the 20th International Conference on Fluidized Bed Combustion

International Mathematical Conference 1982: Proceedings (Mathematics Studies)

p-adic Functional Analysis.. Proceedings of the Sixth International Conference

Proceedings of the 6th SIAM International Conference on Data Mining

Complex geometry: proceedings of the Osaka international conference

Quantum Information V: Proceedings of the Fifth International conference

Proceedings of the International Conference on Experimental Fluid Mechanics (2nd)

Modern Bamboo Structures: Proceedings of the First International Conference

Proceedings of the International Conference on Chinese Enterprise Research 2007

Commutative ring theory: proceedings of the Fès international conference

P-adic functional analysis: proceedings of the fourth international conference

Proceedings of the Sixth SIAM International Conference on Data Mining

Sustainable Automotive Technologies 2011: Proceedings of the 3rd International Conference

7th International Conference on Automated Deduction: Proceedings

Web-based Learning: Technology And Pedagogy - Proceedings of the 4th International Conference

Robotic Sailing: Proceedings of the 4th International Robotic Sailing Conference

Proceedings of the 4th International Conference on Southeast Asia

International Mathematical Conference 1982: Proceedings

Systematic Biology: Proceedings of an International Conference

Proceedings of International Science Education Conference 2009

Medical Textiles: Proceedings of the Second International Conference and Exhibition

Symplectic Geometry and Mirror Symmetry,: Proceedings of the 4th KIAS Annual International Conference

Proceedings of the International Conference on Colloid and Surface Science

Semigroups and Formal Languages: Proceedings of the International Conference

Proc. 4th international conference on data mining

Characterization and Metrology for ULSI Technology 2000: International Conference (AIP Conference Proceedings)

PARLE '92. Parallel Architectures and Languages Europe: 4th International Parle Conference, Paris, France, June 15-18, 1992. Proceedings: Proceedings 4th

Euro-Par '98, Parallel Processing: Proceedings of the 4th International Euro-Par Conference

Euro-Par '98, Parallel Processing: Proceedings of the 4th International Euro-Par Conference

Understanding Pedagogy: And Its Impact on Learning

Proceedings of the Twentieth International Cryogenic Engineering Conference (ICEC20)

Proceedings of the 20th International Conference on Fluidized Bed Combustion

International Mathematical Conference 1982: Proceedings (Mathematics Studies)

p-adic Functional Analysis.. Proceedings of the Sixth International Conference

Proceedings of the 6th SIAM International Conference on Data Mining

Complex geometry: proceedings of the Osaka international conference

Quantum Information V: Proceedings of the Fifth International conference

Proceedings of the International Conference on Experimental Fluid Mechanics (2nd)

Modern Bamboo Structures: Proceedings of the First International Conference

Proceedings of the International Conference on Chinese Enterprise Research 2007

Commutative ring theory: proceedings of the Fès international conference

P-adic functional analysis: proceedings of the fourth international conference

Proceedings of the Sixth SIAM International Conference on Data Mining

Sustainable Automotive Technologies 2011: Proceedings of the 3rd International Conference

7th International Conference on Automated Deduction: Proceedings

Recommend Documents