Advances in School Psychology 9781317535805, 9781138848719

Citation preview

PSYCHOLOGY REVIVALS

Psychology Revivals is a new initiative aiming to re-issue a wealth of academic works which have long been unavailable. Encompassing a vast range from across the Behavioural Sciences, Psychology Revivals draws upon a distinguished catalogue of imprints and authors associated with Routledge and Psychology Press, restoring to print books by some of the most influential academic scholars of the last 120 years. For details of new and forthcoming titles in the Psychology Revivals programme please visit: http://www.psypress.com/books/series/psyrevivals/

ISBN 978-1-138-84871-9

,!7IB1D8-ieihbj!

www.routledge.com  an informa business

Advances in School Psychology Edited by Thomas R. Kratochwill, Stephen N. Elliott and Maribeth Gettinger

Psychology Revivals

PSYCHOLOGY REVIVALS

Advances in School Psychology Volume 8

Edited by Thomas R. Kratochwill, Stephen N. Elliott and Maribeth Gettinger

Psychology Revivals

Advances in School Psychology

Originally published in 1992, this title is the last in a series of books on school psychology. It contains diverse contributions relevant to school psychology, research, theory and practice at the time. Including chapters on alternative intervention strategies for the treatment of communication disorders, strategies for developing a preventive intervention for high-risk transfer children, a review of sociometry and temperament research, a review of the recent advances in research in training behavioral consultants at the time, and an overview of school-based consultation to support students with severe behavior problems in integrated education programs.

Page Intentionally Left Blank

Advances in School Psychology Volume VIII

Edited by Thomas R. Kratochwill Stephen N. Elliott Maribeth Gettinger

First published in 1992 by Lawrence Erlbaum Associates, Inc. This edition first published in 2015 by Routledge 27 Church Road, Hove BN3 2FA and by Routledge 711 Third Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business © 1992 by Lawrence Erlbaum Associates, Inc. The right of Thomas R. Kratochwill, Stephen N. Elliott and Maribeth Gettinger to be identified as author of this work has been asserted by them in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Publisher’s Note The publisher has gone to great lengths to ensure the quality of this reprint but points out that some imperfections in the original copies may be apparent. Disclaimer The publisher has made every effort to trace copyright holders and welcomes correspondence from those they have been unable to contact. ISBN: 978-1-138-84871-9 (hbk) ISBN: 978-1-315-72604-5 (ebk)

ADVANCES in SCHOOL PSYCHOLOGY Volume VIII

edited by

THOMAS R. KRATOCHWILL STEPHEN N. ELLIOTT MARIBETH GETTINGER The University of Wisconsin-Madison

m 1992

LAWRENCE ERLBAUM ASSOCIATES, PUBLISHERS Hove and London Hillsdale, New Jersey

Copyright© 1992 by Lawrence Er1baum Associates, Inc. All rights reserved. No part of this book may be reproduced in any form, by photostat, microform, retrieval system, or any other means, without the prior written permission of the publisher. Lawrence Erlbaum Associates, Inc., Publishers 365 Broadway Hillsdale, New Jersey 07642

Library of Congress Cataloging in Publication Data Advances in School Psychology ISSN 0270-3920 ISBN 0-8058-1118-4 Printed in the United States of America 10 9 8 7 6 5 4 3 2 I

This series is dedicated to improved services to children and to schools, and especially to Tyler Thomas. Thomas R. Kratochwill Dedicated to my sons, Dustin Rhodes and Andrew Taylor, and to my loving wife, Anita. Stephen N. Elliott To my boys, Brad, Trevor, and Grady. Maribeth Gettinger

Page Intentionally Left Blank

Contents of Previous Volumes

Volume I Advances in School Psychology: A Preview of the Contents and an Overview of the Chapters Thomas R. Kratochwill School Psychology in the 1980's: Some Critical Issues Related to Practice Beeman N. Phillips School Psychology: An Instrumental Service for the Handicapped David A. Sabatino Issues Surrounding Diagnosis of Learning Disabilities: Child Handicap Versus Failure of Child-School Interaction Gerald N. Senf Mental Health Consultation Joel Meyers Program Evaluation and School Psychology: Perspectives, Principles, Procedures Charles A. Maher Advances in Criterion-Referenced Assessment Anthony A. Cancelli and Thomas R. Kratochwill Path-Referenced Assessment in School Psychology John R. Bergan Naturalistic Observation in Design and Evaluation of Special-Education Programs Gene P. Sackett, Sharon Landesman-Dwyer, and Victor N. Morin Treatment of Hyperactive, Aggressive, and Antisocial Children Mark A. Stewart and Howard B. Ashby Volume II Advances in School Psychology: An Overview of the Chapters Thomas R. Kratochwill

vii

viii

CONTENTS OF PREVIOUS VOLUMES

School Psychology Training and Practice: The APA Perspective Walter B. Pryzwansky Competence Based Education and School Psychology Barbara A. Mowder Larry P. and PASE: Judicial Report Cards on the Validity of Individual Intelligence Tests Donald N. Bersoff The Rights of Children: A Challenge to School Psychologists Maynard C. Reynolds An Integrated Review of Almost a Decade of Research on the McCarthy Scales Alan S. Kaufman Neuropsychological Assessment Raymond S. Dean Vocational Guidance Consultation in School Settings Michael E. Bernard and Frank D. Naylor Facilitation of Prosocial Behavior in Children Jerry D. Harris, Nancy Eisenberg, & James L. Carroll Self-Control in the Classroom: Theoretical Issues and Practical Applications Richard N. Roberts and Margaret L. Dick Current Approaches to Physical Fitness Education Steven N. Blair, Russell R. Pate, and Bruce McClenaghan Volume III Advances in School Psychology: An Overview of the Chapters Thomas R. Kratochwill School Psychology Training and Practice: The NASP Perspective Ann W. Engin and Reid Johnson International School Psychology: Cross-Cultural Perspectives Frances M. Culbertson Vocational School Psychology Thomas H. Hohenshil Organicity as a Construct in Psychological Diagnosis Joan F. Goodman Assessment of the Gifted Child Sylvia Rosenfield Social Skill Assessment and Teaching in the Schools Gwendolyn Cartledge and JoAnne F. Milburn Reading Research: What it Says to the School Psychologist Nancy Roser and Diane L. Schallert Thinking Through Academic Problems: Application of Cognitive-Behavior Therapy to Learning Dana E. C. Fox and Philip C. Kendall Steps to Beginning a Preventive Health Education Program in the Schools RonaL. Levy Type A Behavior, Achievement Striving, and Their Childhood Origins Judith M. Siegel and Karen A. Matthews

CONTENTS OF PREVIOUS VOLUMES

Volume IV Advances in School Psychology: An Overview of the Chapters Thomas R. Kratochwill Multidisciplinary Tearns in the Schools: Current Status and Future Possibilities Charles A. Maher and Roland K. Yoshida Behavioral Assessment in the School Setting Rosemery 0. Nelson Assessment of Children's Fears: A Critical Review Billy A. Banios and Carol C. Shigetomi Depression in Childhood and Adolescence: Diagnosis, Assessment, Intervention Strategies and Research William M. Reynolds Behavioral School Psychology: A Conceptual Model for the Delivery of Psychological Services Francis E. Lentz and Edward S. Shapiro The Pine County Model for Special Education Delivery: A Date-Based System Gerald Tindal, Coren Wesson, Stanley L. Dena, Gary Germann, and Phyllis K. Mirkin Acceptability of Classroom Intervention Strategies Joseph C. Witt and Stephen N. Elliott Volume V Advances in School Psychology: An Overview of the Chapters Thomas R. Kratochwill Improving Report Writing in School Psychology Raymond L. Ownby and Fred Wallbrown Crime and Violence in Schools: A Challenge to Educators Keith Baker Behavioral Assessment of Academic Skills Edward S. Shapiro and Francis E. Lentz, Jr. Grade Retention Frederic J. Medway and Janet S. Rose The School Psychologist in Medical Settings Steven I. Pfeiffer, Raymond S. Dean and Sylvia Shellenberger Teacher Stress: Causes, Effects, Intervention Susan G. Forman and Marc A. Cecil Volume VI Advances in School Psychology: An Overview of the Chapters Thomas R. Kratochwill Current Status of Professional Training and Practice: Implications for Future Directions Joseph E. Zins and Michael J. Curtis Licensing, School Psychology and Independent Private Practice David P. Prasse Family Assessment and Intervention in the School Setting Cindy I. Carlson

ix

X

CONTENTS OF PREVIOUS VOLUMES

Client Resistance to Parent Training: Its Therapeutic Management Patricia Chamberlain and David V. Baldwin Characteristics of Rural America and Their Interface with School Psychology James McLeskey, Jack Cummings, and E. Scott Huebner Issues in the Conceptualization, Classification, and Assessment of Social Skills in the Mildly Handicapped Frank M. Gresham and Daniel J. Reschly Adaptive Behavior and the Mildly Handicapped Daniel J. Reschly and Frank M. Gresham Volume VII Advances in School Psychology: An Overview of the Chapters Thomas R. Kratochwil/ Gender Issues and School Psychology Jane Close Conoley and Mary Henning-Stout Medication and School Children: Effects on Behavior and Learning Marion I. S. Huettner, Teresa A. Hutchens, Becky L. Rosenthal, and George W. Hynd Law, Psychology, and Education Beeman N. Phillips School Psychology, Assessment, and the Computer: An Analysis of Current Relationships and Future Potential Jack J. Kramer and Terry B. Gutkin Research on the History of School Psychology: Recent Developments, Significance, Resources, and Future Directions Thomas K. Fagan

Contributors

(Number in parentheses indicate the pages on which the author's contribution begins.) Cathy L. Alpert (7)

Ann P. Kaiser (7)

James R. Barclay (79)

Thomas R. Kratochwill (115)

David Betts (45)

Cecile Lardon (45)

Pamela Carrington Rotto (115)

Luanna H. Meyer ( 153)

Laura Filippelli (45)

Lisa Neuson (45)

Mary Louise Hemmeter (7)

Diane Salmon (115)

Rachel E. Janney (153)

Susan M. Sheridan (115)

Leonard A. Jason (45)

Andrew M. Weine (45)

Joseph H. Johnson (45)

xi

Advisory Board

James L. Carroll Wichita State University Wichita, KS Carolyn T. Cobb Division of Student Services, SOP! Raleigh, NC Thomas K. Fagan Memphis State University Memphis, TN Daniel J. Reschly Iowa State University Ames, lA Sylvia Rosenfield University of Maryland College Park, MD Edward S. Shapiro Lehigh University Bethlehem, PA James E. Ysseldyke University of Minnesota Minneapolis, MN xii

Contents

Preface

xv

Advances in School Psychology: An Overview of the Chapters Thomas R. Kratochwill, Stephen N. Elliott, and Maribeth Gettinger Introduction References

1.

1 6

Alternative Intervention Strategies for the Treatment of Communication Disorders in Young Children with Developmental Disabilities Ann P. Kaiser, Mary Louise Hemmeter, and Cathy L. Alpert Components of the Language System Communication Disorders 13 Three Current Models of Language Intervention 16 Case Examples 28 Conclusions 35 References 38

2.

1

9

Developing, Implementing, and Evaluating a Preventative Intervention for High Risk Transfer Children Leonard A. Jason, David Betts, Joseph H. Johnson, Andrew M. Weine, Lisa Neuson, Laura Filippelli, and Cecile Lardon Tutoring Studies

7

45

54

xiii

XiV

CONTENTS

Conclusion References 3.

4.

70 72

Sociometry, Temperament, and School Psychology

79

Preparation of School Psychologists in Behavioral Consultation Service Delivery

115

School Consultation to Support Students with Behavior Problems in Integrated Educational Programs

153

James R. Barclay The Nature of Sociometry: A Brief Overview 80 The Nature of Sociometric Methods 84 Research with Early Versions of Sociometry 88 The Theoretical Basis of the BCAS 95 Temperament and Sociometry 100 Persistent Problems with the Use of 105 Sociometries The Future of Sociometric Assessment 108 References 11 0

Thomas R. Kratochwi/1, Susan M. Sheridan, Pamela Carrington Ratto, and Diane Salmon Current Research in Training Consultants 116 Issues and Future Directions in Consultation Training and Research 133 References 14 7 5.

Luanna H. Meyer and Rachel E. Janney Current Services for Students with Challenging Behaviors 154 Alternative Mainstream-Support Models 155 An Educational Consultation Approach 163 Discussion 186 References 190 Author Index Subject Index

195 205

Preface

Each of the previous volumes of the Advances in School Psychology has pointed to its serial focus. Volume VIII is the final book in this series format. As editors we will continue to produce a series of individual, thematic volumes on school psychology. In the tradition of the Advances the current volume includes diverse contributions relevant to school psychology research, theory, and practice. Volume VIII includes chapters devoted to alternative intervention strategies for the treatment of communication disorders, strategies for developing a preventive intervention for high-risk transfer children, a review of sociometry and temperament research, a review of recent advances in research in training behavioral consultants, and an overview of school-based consultation to support students with severe behavior problems in integrated education programs. Special appreciation is extended to our advisory board members and the authors of the current chapters for their valuable contributions. We also want to express appreciation to our families for their support of our professional work. We again thank Karen Kraemer for her assistance in production of the volume. And finally, a special note of thanks goes to Larry Erlbaum for his support over the years. Thomas R. Kratochwill Stephen N. Elliott Maribeth Gettinger

XV

Page Intentionally Left Blank

Advances in School Psychology: An Overview of the Chapters

Thomas R. Kratochwill Stephen N. Elliott Maribeth Gettinger The University of Wisconsin-Madison

INTRODUCTION

As school psychology moves through the 1990s concerns have been expressed about the viability of the profession to serve the mental health needs of children and schools. The problem is not one of practitioner competencies or the quality of services but, presumably, is based on the declining numbers of practitioners available to fill existing job vacancies. The problem recently has received increased attention in national publications such as the National Association of School Psychologists Communique (Connolly & Reschly, 1990) and has been embedded in various publications in professional journals (e.g., Reschly & Connolly, 1990). The problem is not a new one and, as Connolly and Reschly indicate, has been predicted, but perhaps, not to the degree expected (Brown & Minke, 1984, 1986; Fagan, 1988; McMaster, Reschly, & Peters, 1989). The problems pertaining to the shortage of school psychologists have some potential negative consequences for the profession as identified by Connolly and Reschly (1990): (a) untimely reductions in the availability of psychological services to children and youths; (b) reduction in the quality of psychological services; (c) deterioration in the ratio of psychologists to students; (d) decline in standards as schools place increasing pressure on credentialing authorities to permit practice by less qualified, but available persons; (e) loss of positions in schools and universities (positions that stay open for over a year are often re-allocated to areas where well qualified people are available); and (f) undermining efforts to improve graduate education and to expand doctoral level opportunities for students and current practitioners. The conclusions of Connolly and Reschly (1990) and others about a practi-

2

KRATOCHWILL, ELLIOTT, GETTINGER

tioner shortage raise a number of concerns about what to do in the future. Although some suggestions were offered by Connolly and Reschly (1990), the problem actually may be linked to a number of broader issues that need to be addressed by the school psychology leadership. To begin with, it seems clear that the length of time it takes to complete a graduate program in school psychology has increased for both specialists and doctoral level training. Can we expect individuals to devote 5 to 7 years to graduate training, or perhaps more, with additional specialty training or licensure/certification to enter the school psychology market? Alternatives to our conventional training format need to be considered, at least for some students. For example, would it be possible to streamline graduate programs and develop alternative models for training to produce graduate students of the same caliber but much quicker? Unfortunately, in all of the debates regarding doctoral versus specialist training there are few, if any, empirical guidelines for addressing this question but, see Phillips, 1990, for some suggestions). It is also unlikely that framing the question in a specialist versus doctoral context will be productive for the field (Kratochwill, 1982). Nevertheless, major effort should be made to consider alternative training formats. Conceptually related to the training issue is the concern pertaining to salary level. It is not surprising to learn that the number of students applying to school psychology programs might be smaller than in the past. It appears that entry incomes and increases in professional income within the profession are greatly limited compared to other areas of professional psychology (e.g., pediatric psychology, child clinical, industrial organizational, etc.). The leadership in school psychology must continue to lobby for increased financial and other benefits associated with the profession to attract students and maintain individuals in current practice. Another potential issue is that the profession has for the past decade received considerable criticism for the kinds of roles and functions that school psychologists serve. Criticism has come both from within and outside the profession in terms of traditional role functions of psychologists. Recent survey data shared with us by G. Batsche and H. Knoff suggest, however, that the practice of school psychology really has not changed much over the past decade, at least in terms of the importance given to assessment functions (G. M. Batsche, personal communication, February 26, 1991). If we hypothesize that external factors control the profession, would we expect individuals to enter and continue in our profession given the rather static nature of psychological service delivery systems? More effort may need to be directed towards how to change schools as systems and make role functions more diverse. Also related to role activities is the potential for job stress and burnout among school psychologists. Although we do not know of specific research investigating this dimension of school psychology practitioners, it is likely that with the pressure put on schools and the large increase in mental health case loads for

ADVANCES IN SCHOOL PSYCHOLOGY

3

school psychologists, considerable stress is likely. The contribution that job stress and burnout have on people entering and leaving the profession remains unclear. A research agenda pertaining to learning about and treating job stress and burnout seems essential. Finally, it would appear useful for school psychology training programs to consider partnerships with other training program disciplines. For example, specialized tracks within school psychology might be allowed within pediatric psychology programs, child clinical programs, as well as others that focus on training psychologists to work with children and families. We obviously have witnessed a splintering of psychologist groups that work with children and families. A productive relationship with other speicalities could be useful in serving the mental health needs of children. Some consideration must be given to practitioners entering related fields who work with children. A common assumption in the school psychology literature is that the mental health needs of children are best served within the school as a setting. But do we know that this is accurate? In fact, the number of individuals in child clinical, pediatric psychology, and related psychology professions has increased and many of these individuals serve the mental health and academic needs of our children well. An assessment of the services these people provide would be useful. Moreover, it is likely that many of these individuals are actually providing more treatment services for children than has been true among school psychologists, especially given recent survey data suggesting that psychologists are primarily involved in doing assessment (G. M. Batsche, personal communication, February 26, 1991). A broader range of issues than just the shortage of practitioners may actually be of some concern to the profession of school psychology. Some of these issues pertain to reorganizing our thinking while others necessitate gathering some information to make informed decisions about future directions in training and recruitment of students into the profession. This agenda will be very important and is a critical priority among our leaders within our school psychology profession. This volume of the Advances in School Psychology addresses a number of dimensions pertaining to services to children and to schools. In Chapter 1, Kaiser, Hemmeter, and Alpert review alternative intervention strategies for the treatment of communication disorders in young children with developmental disabilities. The authors indicate that four tasks must be accomplished for effective language intervention: (a) provide the conditions for learning the linguistic forms of language; (b) facilitate the development of social communication; (c) link newly acquired forms to the context for social use; (d) facilitate the child's continued acquisition of new aspects of the language system. The authors further review three major domains of intervention strategies that can improve children's language skills including direct instruction of language skills with behavioral teaching technology, adaptations of behavioral technology through the use of milieu or incidental teaching, and interventions that facilitate development of

4

KRATOCHWILL, ELLIOTI, GETIINGER

social communication during interactions with adults. Specific examples of research programs in these areas also are reviewed. The authors note that although no single model of intervention has proven completely effective in remediating language deficits, a strong impact model involves one in which social communication is linked to interactions with the care provider and where incidental teaching is focused in advance of the child's current language level. School psychologists play an important role in this process by becoming knowledgeable about the potential for intervention programs, cooperating with speech and language professionals who are working in the area, and assisting in the identification of children who could profit from language intervention. In Chapter 2, Jason, Betts, Johnson, Weine, Neuson, Filippelli, and Lardon review a program in preventative intervention for high-risk transfer children. School psychologists traditionally have not been as active in prevention programs as individuals in the community psychology area. Nevertheless, many of the programs for children are carried out in public school settings. In the chapter, Jason and his colleagues take the perspective that providing at-risk transfer students support in social relationships and competencies is a prevention program with great potential in militating against transitional difficulties. The authors present an approach for developing, implementing and evaluating a program in this area. The program is focused on academic needs of at-risk children who transfer schools. The authors believe that children who transfer to a new school are at high risk for later academic problems. Their approach, which involves an orientation plus tutoring intervention, is designed to provide children social support and academic competencies to deal with transitions. Although their program provides a very promising approach for helping youngsters, they are careful to indicate that considerably more work needs to be done in this area to make programs more individualized and to deal with the complexities of schoolbased intervention projects. School psychologists will benefit greatly from the applied nature of this program and the potential for implementation in multiple school environments. In Chapter 3, Barclay presents an overview of sociometry, temperament, and its relationship to school psychology. As Barclay notes, sociometry is not a new procedure but the premise of the chapter is that this technique has great potential for discovering and analyzing patterns of friendships within group settings. In the chapter, Barclay concludes that sociometric choices co-vary and are related to a number of personality group measures and individually-administered instruments. Sociometric choices are related to ratings of efficiency and competence, they are moderately related to the structure of intelligence and achievement tests, and they have positive correlation with measures of motivation and self-concept, and to social economic status. Sociometric choices are strongly related to behavioral observations of children in the classroom and in the context of a broader assessment framework, can provide assessment of individual differences with

ADVANCES IN SCHOOL PSYCHOLOGY

5

implications for treatment. When paired with academic measures, sociometry provides an overall estimate of personal and social status of children in the classrooms. The author builds a strong case for the role of school psychologist in implementation of these strategies. In Chapter 4, Kratochwill, Sheridan, Rotto, and Salmon provide an overview of training behavioral consultants. The chapter focuses on practical, theoretical, and research considerations in this area. After providing a rationale for the importance of research on training, the authors review current research on training behavioral consultants, including conventional pre-service program training and workshop in-service training; they then review competency-based pre-service and in-service approaches. These latter approaches include micro-consultation, conventional behavioral consultation training, self-instructional training, and mentorship training formats. Thereafter, the authors focus their attention on future directions in consultation training and research. Among the numerous issues raised are the importance of training consultees in consultation, using alternative technology such as interactive video procedures in training, moving beyond conventional teacher-only focused consultation service, re-evaluating the content of training, the nature of supervisory relationships in consultation training, and discussing both micro-relationship issues as well as macro-system issues in the context of future directions. The chapter provides several research directions for future efforts on preparation of school psychologists in this alternative service delivery model. In the final chapter in the volume, Meyer and Janney describe a school-based consultation model to support students with severe behavior problems in integrated educational programs. The model incorporates a number of program components that have been related to positive outcomes in children. They present an approach that can be replicated by school psychologists and report actual program outcomes associated with school districts that they were involved with in treating serious student needs in integrated sites. The authors put their program in context noting that there is little empirical data on effective outcomes with students with special needs. The major thrust of their chapter is that students with serious emotional disturbance and those who exhibit serious behavior problems should receive services in the public school setting. However, they note that if this model is to work effectively numerous resources must be provided to assist teachers in working with these individuals. Their program includes a number of components which are student-focused, utilize local expertise, display a commitment to integrated service delivery and non-aversive behavioral interventions, voluntary participation by teacher and parents, on-going technical assistance and in-service training, minimally intrusive district and building level administration support, special content expertise by a consultant in both nonaversive interventions and educational best practices, implementation of a user-friendly data system, and regular and on-going meetings with staff to solve problems regarding

6

KRATOCHWILL, ELLIOTT, GETTINGER

psychoeducational services for the child. The program is presented in great detail and should be of great interest to school psychologists who wish to examine alternatives to traditional services for youngsters displaying behavioral problems.

REFERENCES Brown, D. T., & Minke, K. M. (1984). Directory of school psychology training programs in the limited states. Stratford, CT: National Association of School Psychologists. Brown, D. T., & Minke, K. M. (1986). School psychology graduate education: A comprehensive analysis. American Psychologist, 14, 1328-1338. Connolly, L. M., & Reschly, D. J. (1990). The school psychology crisis of the 1990s. Communique, 19, I and 12. Fagan, T. K. (1988). The historical improvement of the school psychology service ratio: Implications for future employment. School Psychology Review, 17, 447-458. Kratochwill, T. R. (1982). School psychology: Dimensions of its dilemmas and future directions. Professional Psychology, 13, 987-989. McMaster, M. D., Reschly, D. J., & Peters, J. M. (1989). Directory of school psychology graduate programs. Washington, D.C.: National Association of School Psychologists. Phillips, B. N. (1990). School psychology at a turning point: Ensuring a bright future for the profession. San Francisco, CA: Jossey-Bass. Reschly, D. J., & Connolly, L. M. (1990). Comparisons of school psychologists in the city and country: Is there a "rural" school psychology? School Psychology Review, 19, 534-549.

1

Alternative Intervention Strategies for the Treatment of Communication Disorders in Young Children with Developmental Disabilities

Ann P. Kaiser Mary Louise Hemmeter Cathy L. Alpert Vanderbilt University

By the time a normally developing child reaches the age of 12, his/her vocabulary may exceed 50,000 words. More importantly, as a relatively competent speaker and listener, the typical 12-year-old will be able to understand almost all of what is said to him/her and to describe most experiences using a complex syntactical system that maps causal and logical relationships, marks the relative importance of various ideas, and combines words to multiply the number of meanings that can be expressed using vocabulary alone. The 12-year-old also will be able to communicate according to an elaborate set of pragmatic rules on the basis of his/her metalinguistic skills. In using productive verbal skills, he/she will access rule-based knowledge related to phonological production, morphology, syntax, semantics, and pragmatics simultaneously and make a myriad of small adjustments based on the conditional rules that modify each of these systems separately and in relationship to one another. To communicate fluently requires auditory and perceptual processing, phonological production, semantic and episodic memory, and the capacity to respond to multiple, transitory, complex social and linguistic stimuli within milliseconds. The language and communication skills the typical 12-year-old exhibits represent a specific human capacity that is unequaled. Human society is defined, both literally and metaphorically, by the social linguistic communication. Whether one ascribes to a linguistic determinism view of culture (Whorf, 1956) or simply regards language in its written and oral forms as the primary means of communication within the culture, the critical role of communication skills in human transactions is readily apparent. Even more impressive than the taxonomy of skills related to the performance of complex social verbal communication is the rapidity and ease with which most

7

8

KAISER, HEMMETER, ALPERT

children acquire these skills. By the end of their third year of life, children have mastered the essentials of complex communication. Although expansions of vocabulary and refinements of syntax and pragmatics will continue throughout the preschool and primary school years, 3-year-olds are surprisingly competent communicators having already mastered a vocabulary of 1,000 words, most of the phonological system, enough of the morphological and syntactic systems to communicate primary social meaning and many causal and attributional meanings. All this is accomplished without formal instruction. Apparently, the social function of language in the context of everyday conversational interaction is sufficiently motivating for the normally developing child to learn the critical skills needed for mastering the basic forms and functions of the communication system. Many children have difficulty acquiring one or more aspects of the communication system. In 1988, approximately 1,114,410 school-aged children were diagnosed as speech and language impaired (Annual Report to Congress, 1988). This figure, although representing the largest single area of services within special education, probably underestimates the occurrence and treatment of speech and language related problems in the school context. A diagnosis of language impairment in the early grades is frequently replaced with a diagnosis of learning disabilities in later grades. While difficulties in acquiring communication may be the singular disability of concern in school based intervention, such difficulties may also be one of a cluster of related skill deficits included in learning disabilities or mental retardation and not specifically labeled or treated in children in later grades. Acquisition of adequate communications skills is essential to the social and academic functioning of school-aged children. In the early childhood years, concern is focused on the acquisition of language and communication skills as a critical developmental milestone. Serious delays in acquiring these skills may subsequently impact older students' ability to learn in the school instructional context. Language learning difficulties frequently carry over to difficulties in learning to read (Maxwell & Wallach, 1984). For students beyond the 3rd grade, both independent acquisition of new knowledge through reading and limitations in their verbal abilities may further reduce their competence and/ or performance in academic content areas. In addition, communication deficits directly impact children's relationships with peers and sometimes their ability to relate to adults within and outside the school setting (Brinton & Fujiki, 1989). In summary, the development of communication skills is a keystone to children's academic and social development and thus, effective, early intervention to remediate language deficits is a critical issue for school psychologists, teachers, speech and language therapists and for parents. This chapter is organized into six major sections. We begin by describing the components of the language and communication system; next, the nature of language disorders is discussed. These discussions are intended to provide the

1.

COMMUNICATION DISORDERS

9

reader with background for the discussion of current views of language learning, models of intervention and examples of their application which are covered in the subsequent three sections. The conclusion offers a brief summary of the current issues in research and practice related to treating communication disorders. COMPONENTS OF THE LANGUAGE SYSTEM Language is a rule-based system for social communication that relies on cognition, social interaction, and the use of a linguistic symbol system to express meanings. Language has both receptive and productive aspects that depend on separate but related capabilities for encoding and decoding linguistically mapped social meanings. Spoken language can be described in terms of six major components: phonology, morphology, syntax, semantics, lexicon, and pragmatics. These components comprise the form of language (phonology, morphology, and syntax), its content or meaning (lexicon, semantics) and its use (pragmatics). In this chapter, we are largely concerned with spoken language rather than signed or written language. Thus, the components of the language system are defined in terms of spoken language and important exceptions for signed and written language are noted as such. Phonology is the rule-based system for the production of the sounds of a particular language. Phones are the actual sounds produced by the speaker; phonemes are classes of sounds that are related, yet perceptually distinct as a result of their production characteristics. The sounds produced by the speaker can be distinguished on the basis of their place and manner of production. In English, there are about 45 phonemes; 21 are classified as vowels and 24 as consonants. Speech production is a complex and dynamic process requiring fine motor control of the lips, mouth, tongue and breath as well as their coordinated use to change the shape of the vocal tract. The acoustic signals that are produced as a result of air passing through the vocal tract are perceived as speech sounds (Owens, 1988). Difficulty in articulating the sounds needed for spoken language can result from delays in motor development, malformations of the oral-peripheral mechanism (for example, a clef palate), or from specific brain damage incurred at any point in the life span. Deficits in articulation skill are not necessarily related to other language learning abilities or to overall cognitive development. Children who experience delays in articulation may develop normally in terms of cognitive skills and use of language in social interaction. Furthermore, children with great difficulty in producing spoken language may have highly developed receptive skills that allow them to comprehend and learn complex language. Use of alternative modes for production, such as signs or computer-based voice synthesizers, can provide a production mode for children whose motor abilities will not support the development of spoken language.

10

KAISER, HEM METER, ALPERT

Morphology and syntax are related rule systems governing the combination of units of meaning within words and within sentences, respectively. Morphemes are the smallest unit of meaning. Free morphemes are words; bound morphemes are markers added to words to modify their meaning. In English, plurality [-s], possession [-s'], past tense [-ed], and manner [-ly;-er] may be marked morphologically. In addition, entire word meanings may be changed by the addition of prefixes [e.g., un-, in-, pre-] or suffixes [e.g., -ness, -ment]. The complexity of meanings and rules related to morphological use varies and thus, some morphological conventions are acquired earlier in language development than others. Syntactical rules provide a hierarchical structure that is the basis for sentence organization. Syntax includes word order rules, sentence organization rules, and rules governing the formation and use of word classes within sentences. Children initially learn a limited set of rules that allow them to produce simple sentences, possibly without full knowledge of the constraints that apply to word order and usage. Early word combination production may be based more on children's expression of relational meaning than in their understanding of the rules governing sentence production (Bowerman, 1976). By the age of 36 months, children show evidence of understanding many of the rules for generating simple sentence forms that resemble adult sentences. From this point through adulthood, language learners continue to acquire increasingly complex rules for generating and understanding meaning in sentences. By age 12, children have acquired an extremely complex set of rules that allows expression of complex causal, relational, and logical meanings in a variety of ways. The development and productive use of morphological and syntactic systems usually are linked to cognitive development. However, the particular difficulties of learning the language system sometimes result in language delays that exceed the child's cognitive delay. When children do not have full access to auditory input, as in the case of hearing-impaired children, or to the full range of social and environmental stimuli associated with linguistic interaction, as may be the case with children living in impoverished environments, rule-based language development may be delayed even when nonlinguistically based cognitive skills are developing normally. While language deficits almost always are associated with mental retardation, the type and degree of delay vary widely even within groups of children having the same etiology of retardation. The acquisition and expression of simple and complex meanings depend on children's semantic knowledge. Semantic knowledge, concept development, and the use of words to map concepts are linked closely in early development. Learning and use of simple noun referents is a progressive process in which children initially have a single example of a concept, which they mark using the referent word. For example, "cat" may refer at first to the child's particular stuffed cat. Next, the child may extend or overgeneralize the meaning of "cat" to include all stuffed animals and all furry four-legged animals found in the household environment. Eventually, the child's concept of "cat" and his use of the

1.

COMMUNICATION DISORDERS

11

word are refined to correspond to the adult meaning and use of this word. The development of lexical knowledge continues with expansion of the number of words in the child's vocabulary and with more subtle refinement of meanings for those words. Early semantic development also includes learning to express relational meanings using two or more words. For example, early two-word utterances typically include agent-action ("Daddy eat"), action-location ("Sit chair") and possession ("Chrissy bottle"). Development of the critical concepts of agent, action, object, location, and attribute depends on the child's ability to formulate categories based on his/her experience with the world. The categorization and conceptual development needed for generative use of word combinations are linked to the child's very early cognitive development. Children with serious cognitive deficits often have difficulty acquiring the conceptual underpinnings of language use. They may acquire modest vocabularies, however, their meanings for words may be more restricted or different from adult meanings for these words. Children who have difficulty acquiring and expressing early semantic relationships (e.g., agent-action, action-location, action-object) usually are limited in their mastery of syntactic and morphological rules required for complex language in conversations. Use of syntax requires the formalization of the early social-relational meanings into a more complex linguistic representational system. While learning the basic representational structure of language as mapped by early semantic and lexical knowledge is necessary for the development of more complex and abstract rules that form the syntactic system, it is not sufficient. Because syntactic rules are numerous and vary in complexity, children with language disorders may master many rules yet still have obvious limits in their language skills. Pragmatics is the set of rules governing the use of language to communicate meaning in the context of social interactions. The rules that underlie the pragmatic use of language reflect both social-cultural rules for interaction and the paralinguistic rules for modifying the nonverbal aspects of communication to enhance meaning. Pragmatic use of language develops from early nonverbal social interactions and builds on the expression of basic social intentions in communicative interaction. Early intentions, including requesting, greeting, commenting, and protesting, are first expressed nonverbally in the child's actions and gestures. Subsequently, single and multiple-word utterances are used to mark and elaborate the expression of social intention. Most languages are accompanied by a complex set of pragmatic rules related to the use of language in conversation. Acquisition of these rules for social communication in specific contexts begins in infancy and continues through adult life. Because pragmatic knowledge is the basis for social communication, children who have limited verbal language skills still may be able to communicate basic social intentions effectively. Communication of social intention without a welldeveloped linguistic system, however, is dependent on the development of a

12

KAISER, HEMMETER, ALPERT

signaling system that is shared by the child and those who interact with him/her. Child-caregiver interactions form the social-interactive basis for acquisition of early social communicative skills and for more complex language learning based in social meanings. Children who regularly interact with a responsive caregiver rapidly learn strategies for communicating with that person. At the same time, they develop the basic communicative skills that are necessary for communicating with unfamiliar persons. Figure 1.1 contains a schematic representation of the interrelationships between the components and subcomponents of the language system and the potential impact of communication disorders on each aspect of children's language use. Three aspects of this schematic are important to understanding the nature of language development, the effects of specific communication disorders, and the importance of effective intervention. First, the three components of the language system (form, content, and use) are based in the child's specific linguistic knowledge, cognitive development and social interaction skills. Thus, disabilities such as mental retardation or behavior disorders may directly result in delays or deficits in the child's communication development. Second, each subcomponent of the language system is a potential area for disruption in the developing child's communication skills. Limited disruption in a single area, such as acquisition of phonology, may have a significant impact on the child's communication performance. Third, while three separate components of the language system can be

BASIS

SPECIFIC LANGUAGE KNOWLEDGE

-

I

' COGNITION AND

SOCIAL

MEMORY

SKILLS

'

I ___

USE

CONTENT

COMPONENTS AND

FORM

SUBCOMPONENTS OF

A) MORPHOLOGY

A) SEMANTICS

THE LANGUAGE

B) PHONOLOGY

B) LEXICON

SYSTEM

C) SYNTAX

IMPACT ON CHILD

INTERACTION

-

A) PRAGMATICS

DECREASED

LIMITED

DIFFICULTY

IN TE LLIGIBI LIT Y

MEANING

IN SOCIAL

COMPLEXITY

VOCABULARY

INTERACTIONS

ACCURACY

EXPRESSION OF RELATIONSHIPS

FIG. 1.1. The interrelationships between the components and subcomponents of the language system and the potential impact of communication disorders on each aspect of children's language use.

1.

COMMUNICATION DISORDERS

13

labeled and illustrated individually, in development and in use, each of the components impacts on each of the others. As indicated by the arrows in Figure 1, at the most basic level, cognitive skill contributes to the child's social interaction skills. For example, at the level of impact on the child's communicative performance, limited expression of meanings (resulting from semantic and lexical content limitations) will contribute to the child's difficulty in social interactions. Both receptive and productive use of communication skills may be affected by skill deficits in any of the subcomponent areas, however, the degree of disruption may differ in the receptive and productive modalities. In the following section, the specific language disorders are described and the impact of these disorders on a child's performance are discussed in greater detail. COMMUNICATION DISORDERS

Because of the complexity of the language system, there are many points where breakdowns in essential processes related to learning can result in communication disorders. In this section, we describe primary language disorders in relationship to the form, content and use of the language system, following Bloom and Lahey's (1978) conventions for the etiology of communication disorders. As noted in the previous discussion, form relates to the phonological, morphological, and syntactic aspects of the language system. Content describes the semantic or underlying representational meaning basis for language. Use refers to the social use of language in conversational contexts. Disorders Related to Form

The form of language has to do with its surface or structural properties. When speech is the mode of expression, its quality is determined by structural attributes including sound production and the use of morphological and syntactical rules. Fluency and characteristics associated with voice, such as pitch, intensity, intonation, and prosody, also contribute to the quality of form. Speech production problems may have an organic or a functional base. Organic problems arise from physical or neurological impairment; functional problems are related to incorrect learning, discrimination difficulties, and short auditory memory span (McCormick, 1984). Children with form disorders may have a specific message (i.e., content) to convey and the intention and social skills (i.e., pragmatic abilities) necessary for communication, but they have difficulty determining or producing the appropriate forms and/or sequences of forms (i.e., sounds, words, grammatical morphemes) to represent their message. Production problems may be related to (a) deficits in content-form linkages (i.e., either the necessary connections between

14

KAISER, HEMMETER, ALPERT

content and form have not been made or problems exist in matching the appropriate learned form(s) with the content), or (b) difficulties in producing the forms that have been selected to represent the message. Disorders of articulation (types of articulation errors include substitutions, omissions, distortions, and additions) are the most common type of speech production problem (Polloway & Smith, 1982). Articulation disorders are common among the general population and they frequently are associated with developmental disabilities such as hearing impairment (Jensema, Karchmer, & Trybus, 1978). A number of studies have investigated articulation deficiencies of the mentally retarded (Bangs, 1961; Schiefelbusch, 1963; Spradlin, 1963). In general, the research in this area suggests that phonological development by mentally retarded individuals is characterized by developmental delay rather than developmental deviance (Bartolucci & Pierce, 1977). Difficulty using standard syntax and morphology also characterizes many children with language delays and disorders. These children may display regular and irregular errors related to the use of grammatical morphemes (to indicate, for example, tense and number), word order, subject-verb agreement, and particular categories of speech (e.g., prepositions, connectives, and articles) appropriately in spontaneous speech. Form disorders have been reported among children described as mentally retarded, aphasic and learning disabled, as well as children with cerebral palsy and hearing impairments. Children who stutter also display a problem with form. Studies on the syntactic and morphologic abilities of mildly retarded individuals indicate that development in these domains does not differ qualitatively from development by nonretarded persons (Johnston & Schery, 1976; Naremore & Dever, 1975; Wheldall, 1976). However, development is slower and, even when retarded and nonretarded individuals are matched for mental age, mentally retarded persons appear to use shorter, less complex sentences than their nonretarded peers (McLeavey, Toomey, & Dempsey, 1982). According to McLeavey et al. (1982), mentally retarded persons may depend on early syntactic forms for a longer time and, while advanced syntactic forms are learned by mildly retarded persons, they are used less frequently.

Disorders Related to Content The content of language has to do with the semantic or meaning dimension of language. Semantic abilities are directly linked to the child's cognitive knowledge about objects, persons, events, and relationships in the world. To communicate this knowledge via oral language, the child must learn to "map" what he/she knows about the world with linguistic forms. The abilities to use linguistic forms and grammatical rules, to link them to content, and to use language for purposes of communication (e.g., pragmatic skills) are tied to the child's level of cognitive functioning. Generally, semantic

1.

COMMUNICATION DISORDERS

15

abilities develop at about the same rate as other aspects of the communication system. Problems associated with the content dimension of language have been linked to limited learning about the perceptual and functional characteristics of objects, actions, and/or events and their relations, due to, for example, visual impairment, restrictions on mobility, and attentional and/ or perceptual deficits and, with more general deficits in learning abilities. Studies examining semantic development by mentally retarded individuals report evidence of developmental lag, but no qualitative differences between mentally retarded children and normally developing children (Leonard, Cole, & Steckol, 1979). Word meanings for persons with mental retardation tend to be more concrete than those of persons who are not retarded. Children with mental retardation, visual impairments, learning disabilities, spina bifida, and children classified as autistic may show delays in semantic development. Disorders Related to Use

In addition to its code or form and semantic (content) base, language has a function. Language, both verbal and nonverbal, is used to affect the environment in particular, intended ways. Pragmatic skill is the ability of speakers to adhere to rules governing the use of language in social contexts (e.g., taking turns, maintaining the topic of conversation, relating new to old information, using listener cues to modify one's behavior) to achieve desired environmental outcomes (e.g., altering the listener's behavior, belief, and/or attitudes) (Bates, 1976). It involves the ability to consider such factors as (a) the desired effect of communication on the environment, and (b) the prevailing social and physical conditions, for purposes of determining what will be expressed, and how, when, where, and to whom it will be expressed. Children with pragmatic disorders may have a message to communicate and a demonstrated understanding of a code and rules for using it, but they have difficulty "saying the right thing the right way at the right time ... " (ErvinTripp, 1971, p. 37). Problems with use may take a variety of forms. Children may show disruptions in basic interactional strategies such as looking at the listener's face/eyes during joint interaction, following the listener's gaze to attend to a focal point, and taking turns during reciprocal interactions. Difficulties in talking about contextually relevant topics and maintaining the topic of conversation are also characteristic of use problems. In addition to talking about things that are out of context, pragmatically impaired children may make abrupt shifts in conversational topics and/or "string together ideas tangentially without regard for the listener's perspective" (McCormick, 1984, p. 106). Normal speakers make adjustments in their own behavior, in part, on the basis of particular listener variables. Children with pragmatic disorders, however, may be less able to use such variables (e.g., age, sex, position, presumed knowledge of the listener) for purposes of making modifications in their communicative attempts.

16

KAISER, HEMMETER, ALPERT

Along these same lines, pragmatically impaired children may have problems receiving and/or interpreting the listeners's verbal and nonverbal conversational cues, and adjusting their behavior accordingly. Failure to modify unsuccessful communicative attempts, and relate new to old information also have been observed in children with use problems. A limited amount of research has dealt with the pragmatic aspects of language functioning in the mentally retarded. Owens and MacDonald ( 1982) reported that the language of retarded and nonretarded children fulfills functions that were expressed earlier through gestures. Further, the distribution of these functions is similar for both groups when they are matched for language development level. Bedrosian and Prutting (1978) reported that the conversational role of retarded individuals seems to be one of nondominance. Bedrosian and Prutting found this to be the case even though subjects with mental retardation displayed the ability to control conversations in ways similar to those of nonretarded adults. Summary

Both children's abilities and the environment contribute to communication development and to communication disorder. Disruptions can and do occur within specific aspects of the communication system, however, the interrelatedness of components often results in problems in one area affecting other aspects of the system. In addition, secondary developmental problems also may arise as a result of communication disorders. Examples of secondary problems include behavior disorders, inattentiveness, academic learning problems, and social isolation. The environment plays a critical role in accentuating or remediating the primary and secondary effects of communication disorders. THREE CURRENT MODELS OF LANGUAGE INTERVENTION

Language intervention is a broad term that describes a teaching process in which some remediation of a communication deficit is the goal. How remediation is accomplished depends on the theoretical and logical basis of the particular language intervention models. Models of language intervention may be viewed along a continuum as shown. The ends of the continuum are represented by therapist-focused models and child-focused models, with a hybrid model representing the middle of the continuum. Several factors determine where, along the continuum, an intervention model is placed. The primary factors include (a) what is trained, (b) who initiates the topic of training, and (c) how training is accomplished. Other factors that may be considered in placing a model are who the trainer is and where the training takes place. At one end of the continuum is a therapist-focused language intervention

1.

COMMUNICATION DISORDERS

17

model typified by didactic behavioral interventions. In a didactic intervention model, the teacher or therapist selects the materials, the reinforcers, and the topic of the interaction. The goal is to train the child to respond to a set of preselected targets using operant learning procedures. On the opposite end of the continuum is the responsive-interaction model where the child determines the focus of the intervention. The goal of the intervention is not to teach the child some set of preselected targets, but to facilitate the child's development by responding to and communicating with the child. The interaction is based on child-selected topics. Child utterances are followed by communicative responses from the adult. The adult continues the child's topic so that the child is encouraged to elaborate on his/her own topic. Overt prompts for elaborated language are not used in this model. A naturalistic or incidental intervention model falls between the didactic model and the responsive-interaction model. It is less therapist-focused than the didactic model. Teaching interactions are based on the child's interest in some aspect of his/her environment rather than a topic predetermined by the therapist or teacher. The naturalistic intervention model incorporates a system of prompts as the adult responds to the child's interest. The behavioral didactic, responsiveinteractive, and naturalistic language intervention approaches are described further in the following section, with emphasis given to their underlying assumptions, strategies for language acquisition, and empirical basis. Each of these models derives from a different theoretical perspective and each emphasizes teaching somewhat different aspects of the communication system. In this section, we review each model, its assumptions and its limitations, and research supporting the use of this model. In the following section, case study examples are used to illustrate how each model of intervention may be applied to meet a child's specific needs for communication training. Behavioral Didactic Language Intervention

Most didactic approaches to language intervention are derived from an operant model of learning and language use (Skinner, 1957). Operant learning theory assumes that language is behavior controlled by particular discriminative stimuli and that it is reinforced by various stimuli that result in the increased use of the consequated response. Any stimulus that functions to increase the likelihood of eliciting the appropriate response may reinforce language use. Although Skinner's analysis of language clearly considers the environment as a critical stimulus for language and functional consequences of language use as the primary means by which language forms are learned, behavioral interventions have relied primarily on didactic instruction outside the natural environment for teaching new linguistic forms (Kaiser & Warren, 1988). The didactic model assumes that new language is best taught when discriminative stimuli, child language targets and child language reinforcers are specifically identified and

18

KAISER, HEMMETER, ALPERT

targeted. Many opportunities for association between the stimuli and the desired response are posited to be necessary for learning to occur. The didactic model also assumes that generalization across stimuli and settings will occur by gradually increasing the number and types of discriminative stimuli and reinforcers to the point where they are indiscriminable from those in naturally occurring conversations. Didactic language intervention is characterized by adult control and a high degree of structure. Teaching usually focuses on the lexical and syntactic components of productive and receptive language and occurs in a clinical setting with the speech therapist and one child present. The teaching format includes massed trials; each trial consists of presentation of a stimulus (a picture, an object and/or a verbal prompt from the therapist), a child response, and a consequence from the child's response. The topic for the trial is predetermined by the teacher. A variety of verbal and nonverbal prompts may be used; stimulus control procedures (shaping and fading) are used to support child learning and generalization. Training typically includes a high ratio of differential reinforcement. This model is based on several assumptions. Communication therapy is assumed to be a highly technical skill that requires an adult with specialized training. Brief but intense instruction is believed to be the most effective means of language training and the location and setting of the training are not significant factors. For the most part, didactic language interventions have used the "train and hope" strategy of generalization discussed by Stokes and Baer ( 1977). That is, training would occur without specifically programming for generalization and, if generalization occurred, it would be an added benefit. Some didactic language training programs (Guess, Sailor, & Baer, 1978; Waryas & StremelCampbell, 1983) do include systematic procedures for promoting generalization. Using multiple exemplars when training targets, training across settings, and systematically teaching rules for new combinations using a matrix training strategy (Wetherby, 1978) may enhance children's generalization of newly training forms.

Research on Didactic Language Teaching. Research on didactic language intervention has demonstrated its effectiveness in terms of teaching new linguistic structures and skills to children with a variety of handicapping conditions (Garcia, Guess, & Byrnes, 1973; Welch & Pear, 1980; Zwitman & Sonderman, 1979). Studies have demonstrated that didactic language training could be used to train simple clausal constructions (Hester & Hendrickson, 1977; Lutzker & Sherman, 1974), elements of noun phrase structure (McReynolds & Engmann, 1974), verb inflections (Guess & Baer, 1973; Schumaker & Sherman, 1970) and complex sentences. Hester and Hendrickson (1977) trained three language-impaired children to produce specific sentence forms following a live enactment as the stimuli for the specified response. The three children successfully learned to use the targeted structure. Schumaker and Sherman (1970) used a combination of

1.

COMMUNICATION DISORDERS

19

operant learning procedures to teach past tense and present progressive inflections to three moderately retarded children. Tokens were used as reinforcers. All three children learned to use the verb inflections. These studies confirm the effectiveness of didactic language interventions in terms of teaching a variety of language targets to children with various degrees of handicapping conditions. Concurrent with these demonstrations of the effectiveness of didactic language interventions was the development of a technology of language intervention which was typical of didactic interventions (e.g., Gray & Ryan, 1973; Guess et al., 1978; Waryas & Stremel-Campbell, 1978). Specifically, these programs incorporated operant learning procedures, mass trials, and adult-controlled interactions. While didactic language interventions are effective in facilitating initial acquisition of language skills, they are less promising in terms of the extent to which they result in generalization of newly trained forms to appropriate communication contexts (Costello, 1983; Spradlin & Siegel, 1982; Warren & RogersWarren, 1980). Some studies on the generalization of trained forms have produced mixed effects (Hester & Hendrickson, 1977; Warren & Kaiser, 1986) while others have reported poor generalization (Mulac & Tomlinson, 1977; Zwitman & Sonderman, 1979). The real test of generalization is whether the trained form is used in natural contexts regardless of the resemblance of those contexts to the training setting. The foregoing studies suggest that generalization may not occur even when only one of two stimuli are varied from the training setting. Issues in Application. The findings of limited generalization are not surprising in that certain characteristics of the didactic approach may actually inhibit generalization. When language is taught using didactic procedures, the functions of language use are those typical of instructional settings rather than those associated with normal social and communication contexts. Thus, the emphasis is on structure rather than functional communication. Didactic interventions are structured such that the interaction is adult-controlled rather than based on the child's interest. This situation encourages rote responding rather than spontaneous language use. These characteristics combine to produce a situation where teaching is the goal rather than communication. Despite the problems associated with this approach, there may be some situations where didactic interventions are the treatment of choice (Carr & Kologinsky, 1983). Didactic interventions may be the most appropriate approach for fine tuning formal language skills with children who already have functional language use and have mastered basic formcontent-use linkages. Didactic interventions may also be used for establishing initial skill acquisition such as vocal and motor imitation, linguistic rules, and labels. Even in these instances, generalization and spontaneous use of communication resulting from training may be limited and specific strategies to enhance generalization may be needed. Cole and Dale (1986) argue that combining naturalistic language procedures with didactic language procedures may prove to be

20

KAISER, HEMMETER, ALPERT

superior to either approach used separately. Use of multiple exemplars, indiscriminable contingencies for language use and sequential modification have been demonstrated to enhance generalization (Anderson & Spradlin, 1980; Campbell & Stremel-Campbell, 1982). In general, didactic instruction appears to be most effective in teaching new forms and least effective in establishing meaning or content and functional use. When didactic instruction is used as primary means of intervention, it should be accompanied by procedures to enhance generalization across settings, trainers and stimuli or other intervention procedures designed to teach functional use.

Responsive-Interaction Interventions Responsive-interaction approaches to intervention derive from empirical descriptions of mother-child interaction suggesting that certain characteristics of interactions facilitate communication development (Kaye & Charney, 1980). Consistent with these observations, an intervention model based on conversational interactions has been developed. Responsive-interaction approaches to intervention are consistent with systems views of interaction and child development. Communication itself is viewed as a system of components that cannot be easily separated for teaching (as required in behavioral didactic instruction). Communication is a function of contingencies and context, each affecting the other. Children learn a language in interaction with their significant others (Bronfenbrenner, 1977; Bruner, 1975; Mahoney, 1975). Thus, communication interventions must involve not only the child, but also his/her significant others. Because the child develops in the context of the environment, joint activities and routines must form the basis for language use. An ecological systems approach develops the intervention around the naturally occurring interactions between children and their significant others (Snow & Ferguson, 1977). In this perspective, language is purposeful behavior; it develops as a means for communicating social messages. Thus, establishing the function of communication is a necessary component of language learning. Early communication skills are assumed to be facilitative of the child's participation in interactions that provide the structure for learning language (Wells, 1986). Responsive-interaction models of intervention assume that language is learned most effectively in interactions with adults who provide models slightly in advance of the child's current abilities. These linguistic models are to be presented at times of maximum child interest and should be related to the child's topic. Ideal interactions for learning language occur when the child intentionally or unintentionally communicates, thus, in responsive-interaction approaches, child communication is followed by adult models of appropriate language (Nelson, Carskudden, & Bonvillian, 1973), or by expansions and repetitions of the child's language (Cross, 1977) and extensions (Newport, Gleitman, & Gleitman, 1977) of the child's language. In addition, the adult complies with requests

1.

COMMUNICATION DISORDERS

21

made by the child thus making the child's language functional in interactions. In this model of intervention, the adult's role is to be a communicative partner rather than a teacher per se. Sincere communication is the goal, therefore, verbal prompts for language practice (e.g., demanding models, such as "Say 'ball'.") are discouraged. The environment may be arranged as a means of eliciting language, but verbalizations directed at the child take the form of a communicative response rather than a prompt. Responsive-interactive teaching differs from behavioral didactic instruction in three ways. First, there is little emphasis on production of new utterances; strategies to elicit child utterances are rarely used and nondirective interaction is stressed (Mahoney & Powell, 1986). Second, the goal of the intervention is to enhance the social communicative relationship of the dyad as a basis for accelerated language learning rather than to teach the child specific language skills directly (Weiss, 1981). Third, the intervention is intended to increase the quality and richness of the language to which the child is exposed, thus improving the likelihood that he/she will learn new language (Girolametto, 1988). Responsiveinteractive teaching is a secondary, rather than primary, model of language intervention (Fey, 1986). Research on Responsive-Interaction Teaching. Research on the responsiveinteraction model is less complete than the research on other language intervention models. Several responsive-interaction programs of language intervention have been developed (MacDonald, 1985; Mahoney & Powell, 1986; Manolson, 1985; Weiss, 1981). Much of the research supporting this approach to interventions has been in conjunction with these programs. The existing models are similar in philosophy, but differ according to who the primary interventionist is (parent or teacher/therapist) and the specific interactional strategies that are used to facilitate social engagement and language use. The models share an emphasis on enhancing the communicative relationship and tumtaking and responsiveness as the basis of building more complex social communication interactions. The INREAL model (Weiss, 1981) was among the first widely disseminated models of responsive-interaction interventions. The INREAL model incorporates the speech therapist into the classroom daily as a resource for the teachers. The therapist or teacher enters into an interaction with the child after closely observing the child and responds to his/her interests and/or communicative attempts. S.O. U .L. (Silence, Observation, Understanding, and Listening) is the strategy used by the adult to earn the right to enter into the child's activity. There are six techniques which the teacher may use to interact with the child: (a) mirroring (imitating the child's nonverbal behavior), (b) self-talk, (c) parallel talk, (d) verbal monitoring and reflecting, (e) expansion, and (f) modeling. These techniques are responses to the child's communicative attempts and/or interests and do not require a response from the child. Use of the INREAL model has been evaluated with young children with a variety of handicapping conditions (Weiss,

22

KAISER, HEMMETER, ALPERT

1981 ). Children in the experimental group showed significantly grater gains on standardized measures of language performance than did the children in the control group. When the INREAL model was used with children having autism or severe emotional handicaps (Rogers, Herbison, Lewis, Pantone, & Reis, 1986), significant increases were reported in the children's attempts to interact, to engage in tumtaking games, and in social initiations. The Environmental Language Intervention Program (ELI) (MacDonald & Gillette, 1984) has five main program components: receptive language training, imitation training, conversation training, structured play, and generalization training. The ELI program was evaluated in a series of studies by MacDonald and his colleagues. Mahoney and Snow (1983) used the ELI model in an investigation of the relationship between sensorimotor functioning and the children's response to early language training. Fourteen children with Down syndrome and their mothers participated in the study. The mothers were trained to implement the intervention program over four sessions. Although the results of the study indicated that the subjects made substantial gains in rate of language acquisition and spontaneous communication during intervention, the lack of an appropriate control group makes it difficult to determine whether the gains could be attributed directly to the intervention. Girolametto ( 1988) used the Hanen Early Language Parent Program (Manolson, 1985) as an intervention for developmentally delayed children. Mothers were trained to use the intervention in eight group sessions. The parents were taught to use a variety of strategies for communicating with their children. They were taught to observe the child and follow his/her lead, establish joint focus on the child's topic, respond contingently to the child, and encourage conversation by promoting tumtaking and using conversational repair signals. In addition, the parents were assisted in planning activities and games as contexts for communication. Although there were no significant differences on posttest scores on a standardized language assessment between the control group and the experimental group, differences were observed in aspects of the parent-child communicative interactions. The parents in the experimental group were more responsive and less controlling of their children. The children in the experimental group were more responsive to their parents, took more verbal turns, and initiated topics more often. The results of research on the responsive-interaction model of language intervention are equivocal. The studies discussed earlier, in addition to studies by Mahoney and Powell (1986) and McConkey and O'Conner (1982), indicate that parents and teachers can be trained to use the components of the responsiveinteraction model which are assumed to facilitate language learning (increased responsiveness to child interest, decreased directiveness, following the child's lead and using the child's lead as a topic for communication, and more balanced tumtaking were observed in the trained dyads), however, evidence of significant increases in children's language is limited.

1.

COMMUNICATION DISORDERS

23

Issues in Application. Responsive-interaction methods are well-suited to teaching the early interactional skills that form the basis for the subsequent development of social communication. Training significant others who interact with the child is the critical aspect of implementing this intervention because change in the child depends on systematic changes in the adult's behavior in social interactions with the child. Responsive-interaction may have limited effects on immediate language development, thus, it is important that the parent or therapist being trained to use these techniques has a clear understanding of the types of changes expected. Because the techniques involved are somewhat different than those typically associated with language intervention, the consultant or trainer must be well-prepared to explain the underlying approach and to model the specific techniques for the parent or teacher who will be using them. These interactional strategies may be a valuable addition to other forms of intervention for children at any language skill level who are having difficulty in social interactions.

Milieu Teaching Milieu or incidental teaching refers to interactions between an adult and child which arise naturally in unstructured situations and are used by the adult to transmit information and give the child practice in developing a skill (Hart & Risley, 1975). In milieu teaching, child initiations, typically requests for assistance from the adult, determine when language teaching will occur. The child's verbal or nonverbal initiations indicate to the adult what is of prepotent interest to the child and, thus, define the topic for conversation and milieu teaching. Experimental applications of milieu teaching typically have included four sequential steps: (a) arranging the environment to increase the likelihood that the child will initiate to the adult; (b) selecting language targets appropriate to the child's skill level and reflecting the child's interests; (c) responding to the child's initiations with prompts for elaborated language consistent with the child's targeted skills; and (d) reinforcing the child's communicative attempts by providing access to requested objects, continued adult interaction, and feedback in the form of expansions and confirmations for the child's utterance. Milieu teaching is designed to use brief, positive episodes of communicatively oriented interactions to provide language models and opportunities for practicing language. Milieu teaching involves the application of behavioral teaching procedures in

24

KAISER, HEMMETER, ALPERT

naturalistic, conversational settings (Hart & Rogers-Warren, 1978). While milieu teaching is firmly rooted in behavioral principles of stimulus control and reinforcement, the interactional aspects of milieu teaching are consistent with descriptions of naturally occurring facilitative language learning contexts (Hart, 1985). Emphasis on the development of responsiveness, tumtaking, mutual engagement, joint attention, and topic elaboration is consistent with developmental analyses of mother-child interactions (Bruner, Roy & Ratner, 1980; Hart & Risley, 1986; Moerk, 1983; Schacter, 1979). In milieu language intervention, the development of social communication and linguistic skills is viewed in terms of teaching functional language (Halle, Alpert, & Anderson, 1984; Warren & Rogers-Warren, 1985). The forms and functions of language are interdependent, but separate, components of the child's communication repertoire. Teaching in the natural environment is presumed to instruct the child about both aspects, to link form and function, and to result in the social use of new language. The milieu language teaching model was designed to compensate for the shortcomings of the didactic model, specifically the failure of trained language to generalize to other settings (Warren & Kaiser, 1986). The milieu model incorporates procedures that are associated with facilitating generalization (Stokes & Baer, 1977): Language is trained during naturally occurring interactions and is reinforced by natural consequences; functional language responses are trained in the natural environment; goals are selected from broad classes of syntactic and semantic forms; and, setting and verbal and nonverbal prompts are used as means of enhancing the potential for stimulus generalization. Components of Milieu Teaching. There are two major components of milieu language intervention: environmental arrangement and the four milieu teaching techniques. Environmental arrangement strategies set the occasion for the use of the milieu techniques. These strategies prompt communicative attempts by the child; the teacher then uses a milieu technique to prompt additional or more complex language. Effective use of milieu techniques is enhanced by arranging the environment so that the adult controls materials that the child finds reinforcing. For example, some of the materials should be placed in the child's view but out of his reach. Access to those materials depends on the child's use of communication to indicate his/her interest in or desire for a specific material. The child's interest in the environment indicates the reinforcer that can be used in teaching language. If the environment is not arranged so that the child is interested, then there are no apparent reinforcers for using elaborated language. Several additional strategies may be used to arrange environments; (a) providing choices; (b) using sabotage; (c) presenting materials with which the child may need assistance; (d) displaying materials with multiple parts; and (e) violating the child's expectations or routines (Ostrosky & Kaiser, 1991 ). Four milieu language teaching techniques form the core of the milieu model. The model procedure is the most fundamental of the four techniques. The model

1.

COMMUNICATION DISORDERS

25

procedure differs from traditional modeling in that the child's interest determines the topic of the model. There are three primary goals associated with the model procedure: (a) training generalized imitation; (b) building tumtaking skills; and (c) establishing a basic vocabulary. When using this procedure, the teacher first establishes joint attention by focusing attention on the topic of the child's interest. When the teacher has control over the desired material, he/she presents a model related to that material. If the child imitates the model, the teacher immediately expands the child's response and offers the desired material. The mand-model procedure was developed to program for generalization of language skills from one-to-one language training sessions to the classroom (Rogers-Warren & Warren, 1980). Generalization was programmed by providing opportunities for the child to use trained language in the classroom. Other goals associated with the mand-model procedure are establishing joint attention as a cue for verbalization, training turntaking skills, training the child to request information upon request, and training the child to respond to a variety of verbal cues. When the child shows interest in some material, the teacher establishes joint attention by focusing attention on the child's topic of interest. Once joint attention is established and the adult has control of the desired material, the teacher presents a mand. A mand is an instruction to verbalize and may be present as a question, "What do you want?"; an instruction, "Tell me what you want"; or a choice, "Which one do you want, the cookie or the juice?" If the student responds correctly to the mand, the teacher expands the child's utterance and provides the desired material. If the student responds incorrectly or incompletely, additional mands or models may be presented, however, the interaction always ends with the child having access to the desired material. The time-delay procedure is used to teach children to respond to environmental cues other than verbal prompts and to provide opportunities to initiate the conversation. The general steps of the time-delay are: (a) the adult establishes joint attention and delays (focuses attention, pauses briefly and waits for the child to verbalize) following child interest in some specific aspect of the environment; (b) the adult praises and expands correct child verbalizations; (c) the adult uses verbal prompts following incomplete, incorrect or unintelligible child responses. The final procedure, the incidental teaching procedure, is used to elicit more elaborate language and to improve conversational skills. The most important step of this procedure is to arrange the environment so that the child will show interest in the environment and make requests for access to or assistance with materials or activities. Once the child makes a request, indicating what his/her reinforcer for elaborated language will be, the teacher delays, mands, or models for a more elaborated child response. If the child responds correctly, the teacher expands the child's response and provides the desired material. Incorrect responses are followed by corrective mands and models. However, teaching continues only as long as the child remains interested in the topic material. When the child begins

26

KAISER, HEMMETER, ALPERT

to lose interest, the teaching episode should be terminated using corrective feedback. The integration of the four milieu language intervention procedures with the environmental strategies into a comprehensive intervention package is the goal of milieu teaching. None of the techniques alone would provide the training necessary for a language-delayed child to function normally in his/her everyday environments (Halle, 1982). Environmental arrangement is essential to maintaining the naturalness of the intervention. The model and mand-model procedures should be used to teach new forms and functions, and to teach generalized use of these forms and functions by increasing opportunities for their use in a variety of settings. The time-delay procedure may then be used to teach the child to produce those forms and functions in response to environmental cues rather than only producing them in response to verbal prompts. Finally, the incidental teaching procedure is used to encourage the child to use more elaborate language. The language level of the child, as well as the demands of the environment should be assessed as a means of developing a milieu language intervention plan for a child. Research on Milieu Language Teaching. Researchers have demonstrated the effectiveness and efficiency of milieu teaching with persons with varying handicapping conditions under a variety of conditions using a range of intervention agents (teachers, aides, institutional staff, siblings, parents) to train technical and syntactic language targets (Warren & Kaiser, 1986). Hart and Risley ( 1968) trained disadvantaged preschoolers to use color adjectives when requesting materials. The classroom was arranged so that adults mediated the environment by providing materials contingently. Incidental teaching resulted in significant increases in the use of the target adjectives in the training setting and the children continued to use their targets even when use of targets was no longer required to obtain play materials. Hart and Risley (1974) replicated the findings of the first study. Disadvantaged preschoolers were trained to use nouns, adjectives, and compound sentences during a freeplay setting. A multiple baseline design across the three targets was used to demonstrate the effectiveness of milieu teaching. Hart and Risley ( 197 5) replicated the findings of their earlier studies and provided some additional information about applications of incidental teaching. Continued prompts were not necessary to generate a variety of language responses or to maintain the responses that had been trained using incidental teaching. In addition, they provided evidence that incidentally trained language generalized to peers and other adults. Hart and Risley (1980) compared data from their 1975 study with data obtained on two groups of preschool children. One group consisted of middle-class children who were attending a university-based classroom and the second group included children attending an inner-city Headstart class. After training, the disadvantaged children used language that resembled that of the comparison group of middle-

1.

COMMUNICATION DISORDERS

27

class children, as indicated by both frequency of language use and rates of vocabulary growth. Halle and his colleagues (1979; 1981) adapted the incidental teaching procedure by adding a nonverbal prompt called a delay. They demonstrated the effectiveness of the time delay procedures in two studies. In the first study (Halle, Marshall, & Spradlin, 1979), institutional staff were trained to present a delay before giving food trays to six mentally retarded adolescents. When the time delay procedure was implemented during breakfast, an increase was noted in verbal requests both at breakfast and at other meals. In the second study, Halle, Baer, and Spradlin (1981) trained preschool teachers to use the time delay procedure with developmentally delayed preschoolers. Increases in verbal initiations during both training and generalization sessions were noted. Rogers-Warren and Warren (1980) also adapted incidental teaching by adding a verbal mand as an instruction to verbalize. The effectiveness of the mandmodel procedure was demonstrated in two studies. Rogers-Warren and Warren (1980) trained preschool teachers to use the mand-model procedure during free play. Generalization of previously trained words and structures was noted, and increases in vocabulary and complexity of utterances were also observed. In a second study, Warren, McQuarter, and Rogers-Warren (1984) used the mandmodel procedure to train single and multiword utterances to language delayed preschoolers. Increases were observed in rate of initiations, responsiveness to adult verbalizations and overall complexity of language. Increases were observed in a second setting where incidental teaching was not in effect and the increases in the training setting were maintained even when the intervention was faded. Alpert and Rogers-Warren ( 1983) trained mothers of developmentally delayed preschoolers to use the three techniques described above as well as a fourth technique, the model procedure. Six mothers successfully learned to use all four of the procedures and subsequent changes in the children's language skills were noted. Several other studies have been conducted to examine the effectiveness of milieu teaching. McGee and her colleagues have used incidental teaching to teach receptive object labels and prepositions to children with autism (McGee, Krantz, Mason, & McClannahan, 1983; McGee, Krantz, & McClannahan, 1985). Oliver and Halle (1982) used a time delay procedure to teach a mentally retarded child to use signs. Appropriate sign use and generalization to novel trainers and opportunities were noted. Cavallaro and Bambara (1982) compared an incidental teaching procedure with a question plus labeling procedure. The results indicated that incidental teaching was more effective in increasing requests in a preschooler with language delays.

Issues in Application. For milieu procedures to be effective, they must be implemented by people who interact with the child in naturalistic conversational settings-the classroom teacher, parents, and significant others. Since effective milieu teaching requires correct and frequent use of the procedures to teach the

28

KAISER, HEMMETER, ALPERT

child's targets, specific training in how to use the procedures correctly may be required. The existing research on training parents, teachers, aides and others to use the milieu techniques suggests that such is a lengthy process. Studies by Alpert and Kaiser (in press) and Hemmeter and Kaiser (1986) provide evidence that specific training and feedback on the milieu techniques are necessary for correct implementation by parents. The Mudd and Woolery study (1987) and Kaiser, Alpert, Hancock, Hemmeter, and Ostrosky study (in preparation) indicate that feedback and in vivo coaching, in addition to instruction, are critical to training effective use of the milieu techniques. In the existing studies, changes in the child's language have resulted form very frequent and correct use of the milieu techniques. It is not known how intense or exact milieu teaching must be to produce reliable effects. In addition, recent research (Yoder, Kaiser, & Alpert, 1989) indicates that children learning complex syntactic skills may benefit from the more frequent drill and practice provided by behavioral-didactic training. Limitations of Research Reviewed

Much of the research discussed here has involved young children with mental retardation or developmental disabilities. Our own research has focused primarily on this population as well. However, many children receiving speech and language services are not mentally retarded, and speech and language deficits may be their primary disability. While the research reviewed here focuses on children with mental retardation, the procedures and findings reported here are applicable to children with less severe disabilities. In particular, milieu teaching and responsive-interaction have been used extensively with language delayed children (cf. Hart & Risley, 1968; Mudd & Wolery, 1987; Warren et al., 1984; Weiss, 1981 ). Results reported for these two interventions applied to children with language delays are generally very positive. It may be the case that either procedure may be somewhat more effective for children with milder disabilities than for children with mental retardation. CASE EXAMPLES Case 1: Behavioral Didactic

Background. David was 7 years, 7 months old when he was referred for language intervention. David is functioning in the moderate range of mental retardation. In addition, he has mild cerebral palsy (CP). The CP has primarily affected his fine motor skills. David is the youngest of five children and lives at home with his siblings and his mother and father. While his parents are very interested in David's education, they both work full-time and have little time to

1.

COMMUNICATION DISORDERS

29

implement interventions while at home with David and his four siblings. He attends a special education classroom in a public school serving both handicapped and nonhandicapped children. He attends gym, music and art classes, and lunch with nonhandicapped children. David has been enrolled in some type of program since he was diagnosed as mentally retarded at the age of 18 months. David has speech therapy twice a week for 30 minutes at school, and he sees a private speech therapist twice a week for an hour. The speech therapy at school takes place in a small room next to the library. A recent speech and language evaluation indicated that David has deficits in both his receptive and expressive language but that the deficit in expressive language is much more significant than the receptive deficit. David's communication skills are adequate for functional social exchanges, but his formal syntax language skills impair his overall performance in social and academic exchanges. The evaluation indicated that he has present progressive verb forms (e.g., eating, driving, playing), several prepositions (in, on, over, under, beside), regular plural, some irregular past tense verb forms (e.g., came, went, sat), possessives, and the uncontractible copula (e.g., "he is" in response to "Who is sick"). However, he does not use the regular past tense (e.g., walked, played, worked), articles, possessive pronouns, the uncontractible auxiliary (i.e., "she is" in response to "Who's playing"), or the third person singular verb marker. Observations of David indicated that he uses language functionally, and quite effectively, with both peers and adults. He uses a variety of sentence forms appropriately, as well as a variety of social-communicative functions.

Intervention Prescription. David's strengths, in terms of language, are in the areas of meaning and use. His weakness is in the development of specific forms. For this reason, the speech therapists met with the parents and teachers and decided to use a didactic approach to target specific formal language deficits. The parents felt that they were unable to implement programs at home because of the demands of their jobs and the demands of raising their five children. Thus, it was decided that both speech therapists David currently sees would be given a list of David's targets with examples of each to use in the classroom. Both parents and teachers were encouraged to focus on the target skills to promote generalization which is commonly a problem in didactic interventions. In addition, the therapists, parents, and teachers worked together to select targets that would most improve David's language use. The targets chosen were regular past tense, possessives, and articles, all of which are developmentally just beyond David's current language skills. The therapists developed a set of photographs and objects to be used during the therapy sessions. At least 10 trials for each target occurred during each session with more occurring as time permitted. A trial consisted of the presentation of the stimulus (the photo or the object) and the verbal prompt, followed by a child response or a second prompt. For example, the child was shown a pho-

30

KAISER, HEMMETER, ALPERT

tograph of a person looking down at an ice cream cone on the ground and asked "what happened?" The correct response was "the boy dropped the ice cream." Models and questions were used, but the models were faded once the child understood what the appropriate answer should be. Positive reinforcement followed a correct response, and corrective feedback followed an incorrect or incomplete response. One of the primary shortcomings of didactic interventions is that they often fail to result in generalized outcomes. A number of strategies designed to facilitate generalization were included in David's intervention plan. First, David's intervention plan was implemented by a private speech therapist. In addition, the school speech therapist agreed to conduct sessions in the classroom rather than taking David out for his sessions. Thus, training occurred across settings, teachers, and responses. The speech therapists set criteria for successful mastery of the targets and agreed that neither therapist should move on until the child had reached criteria in both settings. Criteria were: (a) the child will produce the target response on four out of five trials on three consecutive days, and (b) the child will produce the target response at least three times spontaneously on three consecutive days. When criteria were met in both settings, the speech therapists met to discuss additional goals. It was important for the therapists to keep the parents and the teachers informed of any changes and to elicit their participation in developing new targets. Outcomes. David's speech therapists talked on the phone at the end of each week to combine the frequency data they had collected. At the end of 8 weeks, David was producing the regular past tense and articles at criteria level when prompted during the training session. However, he was not producing them spontaneously, and he was not producing possessives when prompted. It was decided that both therapists needed to focus on possessives, but also continue to prompt articles and past tense verbs until he used them spontaneously.

Case II: Responsive-Interaction Background. Robert is 2 years old and was diagnosed at birth as having Down syndrome. Cognitively, Robert is functioning at 16 months and motorically at about 23 months. He attends an early intervention program 2 days a week for 3 hours. During the day, he spends the rest of his time at home with his mother and 4-year-old sister, and occasionally with a babysitter. His father works long hours and often travels for 3 or 4 days at a time. Robert's mother reports that the father is very good with Robert, when he has a chance to spend time with him. The early intervention program that Robert attends does not have a full-time speech therapist. There is a speech therapist consultant who tests the children

1.

COMMUNICATION DISORDERS

31

once a year, and sees the children once every 6 weeks to observe progress and make recommendations for program changes. Robert does not receive any speech therapy, although the classroom teachers report that they work incidentally on language during the hours that he is in the classroom. Robert does not seem to be very interested in toys at this point. He appears to have a short attention span, going from toy to toy, using them inappropriately. He often is observed just sitting on the floor, not playing with anything or anybody. Social praise and attention do not appear to reinforce Robert, except when they come from his mother. Robert's language was recently assessed and his receptive and expressive language were found to be delayed. Receptively, he is functioning at 15 months, and expressively, he is functioning at about 12 months. His mother reports that he has a few words, most of which are only intelligible to the mother and the sister. However, the teachers report that he only makes a few sounds. He is nonverbally imitative and is beginning to imitate some speech sounds.

Intervention Prescription. The language consultant decided that a responsive-interaction intervention using the mother as the intervention agent would be the most appropriate intervention for Robert. Robert was not yet interested in materials or activities, and thus, a milieu approach was somewhat difficult to implement because Robert's preferred natural reinforcers were difficult to identify. Second, Robert finds his mother more reinforcing and attends to her more than to his teachers and other adults. Finally, responsive-interaction is designed to develop a contingent tum taking relationship which forms the basis of language use. Robert does not yet take turns in conversations and thus, opportunities for language use during interactions with his mom are minimized. Observations of Robert and his mother indicate that she talks without allowing time for Robert to take his tum, many of her verbalizations are instructions and, in her attempts to interest him in some material or games she fails to follow his lead and interests. Also, it was observed that when the mother placed demands on Robert to talk (i.e., giving prompts), Robert "tuned her out" by looking away from her. Robert is more interested in playing little games with his mother such as peekaboo and paddy cake. He laughs when his mother tickles him and is generally responsive to her. The language consultant met with Robert's mother for three 90-minute sessions to train the mother to use the responsive-interaction strategies. The language consultant helped the mother to select toys and play activities that she knew were reinforcing to Robert, as the context for the intervention. She was trained to use a number of strategies including: (a) establishing joint attention, (b) parallel talk, (c) expansion, and (d) modeling. Mirroring (nonverbal imitation) and parallel talk were selected to provide the mother with ways of responding to Robert since the majority of his communication was nonverbal. Mirroring and parallel talk in response to nonverbal behavior also provided the context for

32

KAISER, HEMMETER, ALPERT

teaching turntaking. Expansions of Robert's utterances and modeling of appropriate language were selected as strategies for providing appropriate language models without placing demands on Robert to communicate. All of these strategies are used to increase the child's attention to and engagement with the environment. The mother also talked primarily in utterances of two words or less. The mother's utterances were then appropriate language models based on Robert's language level. A set of sample utterances were developed from words in Robert's receptive and/or expressive vocabularies. This set of utterances was then given to the mother as examples of utterances to use when interacting with Robert. However, the mother was not restricted to using only the sample utterances. Although specific language targets usually are not selected when using responsive-interaction, goals for the intervention were determined for Robert and his mother. The primary goal was to facilitate communicative interactions that were fun and interesting for both Robert and his mother. Specifically, the goals for the mother were to decrease her use of instructions, follow Robert's lead and respond to his attempts to communicate, and allow Robert to take his tum either verbally or nonverbally. These goals were selected based on the assumption that they would result in increased responsiveness, more balanced conversations, and increased language use by Robert. Increased expressive vocabulary was a shortterm goal for Robert, with early two-word utterances being the long-term goal. Robert's expressive vocabulary was too small to begin trying to teach him twoword combinations. The mother was encouraged to expand his one-word utterances as a means of providing models of language that were slightly more advanced than his current level. The intervention lasted approximately 12 weeks with weekly sessions with the language consultant. These sessions were used to provide feedback to the mother and to assess the effects of the intervention on Robert's communication skills. In addition, the mother was encouraged to videotape at least 20 minutes of play sessions in which she used the responsive-interaction strategies. These tapes were to be used to provide additional feedback. At the end of the 12 weeks the language consultant and the mother met to reevaluate the intervention. Outcomes. After 6 weeks of training, the language consultant indicated that the mother was using the responsive-interaction strategies correctly with Robert. The language consultant noted that the mother had dramatically reduced her use of instructions, and that she had become more responsive to Robert's communicative attempts. Consequently, the tumtaking was more balanced. Robert had become more responsive, but was still taking his turns nonverbally. The language consultant decided that it was time to concentrate on Robert's skills. The consultant reviewed with the mother Robert's goals and ways to incorporate these goals into the interaction. At the end of 12 weeks, the mother and consultant met to discuss Robert's progress. Robert's verbalizations had increased by about 50%,

1.

COMMUNICATION DISORDERS

33

and his initiations had increased as well. The consultant felt that the mother should continue to work with Robert, meeting with the consultant on a monthly basis. These monthly meetings were used to reevaluate Robert's language progress.

Case Ill: Milieu Teaching Background. Barby is 3 years, 2 months old. She is diagnosed as developmentally delayed resulting from premature birth and health impairments during her first 2 years. Barby's developmental delays are manifest in both motor and cognitive areas where she is estimated to be functioning at about 2.5 years and 2.1 years respectively. She attends an early intervention program for 4 hours in the morning, and then a bus takes her to a daycare center where she stays until her mother picks her up at 3:00p.m. Her mother is the primary caregiver until the father arrives home at 6:00 p.m. The early intervention program Barby attends enrolls only handicapped children. Barby has two half-hour speech therapy sessions each week. During these sessions, Barby and the therapist sit in a comer of the room and go through a series of articulation exercises. Barby's daycare center has mostly nonhandicapped children and the teachers have very little training related to children with special needs. Barby is a social child and the other children seem to enjoy playing with her. She has difficulty sharing toys and participating in prolonged independent and cooperative play. She often gets angry, and takes toys away from other children, or hits them if they refuse to give the toy to her. The teachers at Barby's early intervention program chose to implement a milieu language intervention program with Barby. They followed a series of steps to develop an intervention plan for her. Following a formal assessment of Barby's language, they conducted an ecological assessment of the environments in which Barby spends most of her time. The purpose of the ecological assessment was to determine what communication skills would be most functional for Barby in each of the environments. The third step was to collect a 1-hour sample of Barby's spontaneous language during play and snack time. The final step was to interview those people who spend the most time with Barby. In Barby's case, those people were the early intervention teachers, the bus driver, the daycare teachers, and her mother and father. The results of the formal language assessment indicated that Barby exhibited only a slight delay in receptive language, but she exhibited a significant delay in expressive language (more than 16 months). Her mean length of utterance (MLU) was 1.34 indicating that she spoke in mostly one-word utterances with a few two-word utterances. Barby spoke frequently, about twice per minute during the language sample but less than 5% of her utterances were directed to peers. When Barby combined words, her articulation was poor; 70% of her two-word utterances were at least partially unintelligible.

34

KAISER, HEMMETER, ALPERT

The results of the ecological assessment and interviews with her significant others provided the following information. For the most part, Barby does not need to use communication to get things she needs. At home, at daycare and in the early intervention program, Barby obtains desired materials by pointing to them. The one exception is that the early intervention teacher requires her to label objects during group time. Barby's parents indicated that their main concern was that Barby is not talking. They reported that she seems to know the names of many things, and understands what is said to her, but she will not talk. The daycare workers indicated that Barby often throws temper tantrums because another child will not give her the toy she wants. The daycare worker seems to think that the other children do not understand what it is that Barby wants. The teachers' primary concern is that Barby is not progressing in language development as rapidly as they think she should. When she entered the program a year ago, her language was basically the same as it is now.

Intervention Prescription. After consultation with the family, the daycare workers, and the speech therapists, the teacher selected goals for the milieu intervention program. The first goal was to increase two-word combinations. The teacher felt that there were many opportunities to teach two-word combinations at home, at school and at daycare. The second goal was to increase Barby's verbal requests. Barby makes requests nonverbally at a very high rate, but seldom makes verbal requests. The speech therapist and her parents also wanted improved articulation to be a goal. The speech therapist agreed to implement the milieu intervention plan as a part of her ongoing speech therapy intervention. The first step in the intervention plan was to arrange the environment in ways that would encourage Barby to attend to it and make requests. Preferred materials were placed on shelves so that the materials were out of her reach but within her view. Materials with which Barby needed assistance and materials with multiple parts were displayed prominently. The arrangement of the environment was planned not only to encourage requests, but also to increase the opportunities for teaching language. It was decided that the model procedure would be used infrequently. The teacher and therapist agreed that it would be better to allow Barby to attempt a response without being given the exact form. The mandmodel procedure and the time-delay procedure were used primarily to prompt two-word combinations. The mand and the delay were be gradually faded so that spontaneous use of targets might be observed. It was agreed that these procedures should be used only when Barby is interested in and attending to some material or activity. Nonverbal requests in particular were used as opportunities for incidental teaching of two-word combinations and for encouraging verbal requests. Models were included at the end of each procedure as a means of improving articulation. Content for two-word combinations was selected from Barby's receptive and productive vocabulary so that the first requests for twoword utterances required her to use only words she already knew and used in

1.

COMMUNICATION DISORDERS

35

single-word utterances. Early functional two-word combinations based on the basic relationships between object and events in the environment were selected for training. These included request plus object (want+ noun), object plus action (noun + verb), and agent plus action (noun + verb). A list of examples based on Barby's existing vocabulary and the actual objects she prefers was prepared and given to each of her significant others who would implement the training. The speech therapist trained Barby's significant others to implement the four milieu procedures and to arrange the environment during a 3-hour workshop session.

Outcomes. A simple data collection system involving tallying episodes of milieu teaching and recording Barby's responses was provided for each intervenor. The intervention was conducted for 8 weeks and then reevaluated. The speech therapist checked with each intervenor on a weekly basis and collated the data collected by all interveners to monitor Barby's progress. At the end of 8 weeks, Barby was regularly using 10 two-word combinations, mostly request forms involving "want + noun." The second 8 weeks of intervention were then designed to target the use of other two-word combinations and to encourage Barby to use the request form (want + noun) with her peers in the daycare setting.

CONCLUSIONS

The design of strategies for effective intervention to improve the use of language by children with serious communication deficits has proven difficult (Nelson, 1989). The particular challenge has been interventions that impact the child's social and linguistic development to the extent that new uses of functional, appropriate language in conversational settings result. To be effective, language intervention must accomplish four tasks: (a) provide the conditions for learning the linguistic forms of language; (b) facilitate the development of social communication; (c) link newly acquired forms to the contexts for social use; and (d) ultimately, facilitate the child's continued acquisition of new aspects of the language system. Three strategies for improving children's language skills have been researched: (a) the direct instruction of language skills using behavioral teaching technology (cf. Baer & Guess, 1971; Wheeler & Sulzer, 1970); (b) adaptations of behavioral technology to teach language in more naturalistic environments through the use of milieu or incidental teaching (cf. Halle, Marshall, & Spradlin, 1979; Haring et al., 1987; Hart & Risley, 1975; Rogers-Warren & Warren, 1980); and (c) interventions that facilitate the development of social communication during interactions with adults (cf. Girolametto, 1988; Mahoney & Powell, 1986; Weiss, 1981). These three strategies represent a continuum of approaches

36

KAISER, HEMMETER, ALPERT

to language intervention in terms of their theoretical bases, their primary means for increasing language skills, and their relative emphasis on linguistic and social communication skills. These three approaches were reviewed in terms of their theoretical tenets, their prescribed approaches to language intervention, and empirical evidence of their effectiveness. The contributions of each to understanding the linkages between the first three tasks of language intervention were also addressed in intervention-related research. Strategies for teaching new linguistic forms are fairly well established in the behavioral literature (Guess et al., 1978; Stremel & Waryas, 1974). Three different approaches for linking form to social communicative use have been researched. The first, growing from a behavioral perspective, has emphasized the facilitation of generalization from training to conversational settings (Campbell & Stremel-Campbell, 1982). The second, responsive-interaction interventions, has sought to facilitate social communication as a basis for subsequent language development and use (Seitz, 1975; Weiss, 1981). The third intervention, milieu teaching, uses behaviorally based techniques for teaching new linguistic forms in their functional social contexts (Kaiser, Hendrickson, & Alpert, 1991). The fourth task for language intervention, facilitating the child's continued acquisition of language, is essential to the development of communicative competence. This task is the least understood and least addressed in intervention research. Shatz (1987) recently described three strategies for continued language acquisition used by normal children in the natural environment. Elicitation, expansion and entry processes are "bootstrapping operations" by which children use what they already know about the syntactic, semantic and pragmatic aspects of the language system to learn more. Elicitation processes are indicated by the child's active seeking of specific information about form or meaning from his/her listeners. Expansion processes include language practice, language play and spontaneous repairs in which the child makes substitutions of words into frames. Entry operations-the child's use of rote phrases, familiar exemplars or canonical sentence patterns-suggest that the child is learning and storing larger than word unit patterns with associated meanings in memory. The normal child is an active learner who selectively uses the input available, maps it into the existing knowledge base, and seeks additional information about the aspects of the language system which he/she currently is acquiring. A rich social and linguistic environment is necessary, but since the child is selective about the input needed at any given time, the environment does not have to fit perfectly the child's current need for input. Young children with communication disorders may be limited in strategies for soliciting appropriate input as evidenced by their less frequent engagement with people and the environment, by lower levels of responsiveness (Maurer & Sherrod, 1987; Rondal, 1978), and generally slower acquisition of linguistic forms. In particular, when the learner has less well developed cognitive capacities, immediate language learning must be externally supported. Supporting the child's learning through environmental language inter-

1.

COMMUNICATION DISORDERS

37

vention may not be sufficient to allow the child to independently acquire new language skills. Specific intervention to promote the development of language learning strategies may be needed. When the three prominent intervention models are considered in relation to the four tasks of effective language intervention [i.e., intervention results in (a) changes in linguistic skill, (b) changes in communicating behavior, (c) links linguistic form and social use; (d) teaches strategies for the continued acquisition of new skills], a pattern emerges. The limitations of didactic instruction relative to facilitating the development of social communication, linking form to social context and the continued acquisition of new language are readily apparent. Both responsive-interaction and milieu teaching potentially impact on all four criteria. Responsive-interaction teaches social communication but is weakest in regard to teaching new linguistic forms. Milieu teaching clearly can teach specific language skills and facilitate the development of specific social communication skills such as responsiveness and requesting but its impact on general social communication is unknown. Both milieu and responsive-interaction teaching offer some promise to impact the child's continued acquisition of new skills. Milieu teaching promotes use of generalized imitation as a means of acquiring new forms. Responsive-interaction proposes to improve the child's attention to mother input while enhancing the quality of that input. Neither milieu teaching nor responsive-interaction has been assessed in terms of its impact on continued acquisition. Components of milieu teaching and responsive-interaction may be needed to satisfy the four criteria of successful intervention. In that no single model of intervention has proven fully effective in remediating the social, linguistic, and language learning deficits of young children with communication disorders, alternative intervention strategies are needed. The premise that social-pragmatic and linguistic aspects of language use develop separately but should be taught concurrently is supported by evidence from both normally developing children (Bates, Bretherton, & Snyder, 1988) and from children with language disorders (Snyder & Silverstein, 1988). Based on the research to date, an optimal model of intervention appears to be one in which social communication is built through interactions with an invested caregiver and salient, but limited, episodes of incidental teaching are focused slightly in advance of the child's current language level. Since many potential combinations of existing intervention strategies are possible, future research must be directed toward understanding the instructional components of these intervention strategies and their specific effects on each aspect of the child's communication system. The focus on generalized, functional usage and on social communication that has emerged in the last decade prescribe outcome measures that directly represent the child's use of newly learned language in social conversation. The development of measures to indicate changes in children's strategies for learning language poses a particular challenge but one that, if met, will inform both our understanding of children's

38

KAISER, HEMMETER, ALPERT

language learning and the practice of language intervention. For the practitioner, the use of combinations of intervention strategies, adapted to the specific needs of the child and changing across the child's course of therapy is highly recommended. Hybrid techniques, such as a milieu teaching, may represent the most effective alternative available based on current research findings. The task of learning language is complex and life long. The design of effective interventions is apparently equally complex. The progress in research on children's language learning and language intervention during the last 20 years has provided a foundation for practice, but further research and innovative practice are needed to guide our intervention efforts toward comprehensive language interventions for young children.

REFERENCES Alpert, C. L., & Kaiser, A. P. (in press). Training parents as milieu language teachers. Journal of Early Intervention. Alpert, C. L., & Rogers-Warren, A. K. (1983, March). Mothers as incidental language trainers of their language disordered children. Paper presented at the Gatlinburg Conference on Mental Retardation, Gatlinburg, TN. Anderson, S. R., & Spradlin, J. E. (1980). The generalized effects of productive labeling training involving common object classes. Journal of the Association of the Severely Handicapped, 5, 143-157. Baer, D. M., & Guess, D. (1971). Receptive training of adjectival inflections in mental retardates. Journal of Applied Behavior Analysis, 4, 129-139. Bangs, T. (1961). Evaluating children with language delay. Journal of Speech and Hearing Disorders, 26, 6-18. Bartolucci, G., & Pierce, S. J. (1977). A preliminary comparison of phonological development in autistic, normal, and mentally retarded subjects. British Journal of Disorders of Communications, 12, 137-147. Bates, D. M., Bretherton, 1., & Snyder, L. (1988). From first words to grammar: Individual differences and dissociable mechanisms. New York: Cambridge University Press. Bates, E. (1976). Pragmatics and sociolinguistics in child language. In D. Morehead & A. Morehead (Eds.), Normal and deficient child language (pp. 411-463). Baltimore: University Park Press. Bedrosian, J. L., & Prutting, C. A. (1978). Communicative performance of mentally retarded adults in four conversational settings. Journal of Speech and Hearing Research, 21, 79-95. Bloom, L., & Lahey, M. (1978). Language development and language disorders. New York: Wiley. Bowerman, M. F. (1976). Semantic factors in the acquisition of rules for word use and sentence construction. In D. Morehead & A. Morehead (Eds.), Normal and deficient language development (pp. 99-179). Austin, TX: Pro-Ed. Brinton, B., & Fujiki, M. (1989). Conversational management with language-impaired children. Rockville, MD: Aspen. Bronfenbrenner, U. (1977). Toward an experimental ecology of human development. American Psychologist, 32, 513-531. Bruner, J. S. (1975). The ontogenesis of speech acts. Journal of Child Language, 2, 1-19. Bruner, J., Roy, C., & Ratner, N. (1980). The beginnings of requests. InK. E. Nelson (Ed.), Children's language (Vol. 3, pp. 101-126). New York: Gardner Press.

1.

COMMUNICATION DISORDERS

39

Campbell, C. R., & Stremel-Campbell, K. (1982). Programming "loose training" as a strategy to facilitate language generalization. Journal of Applied Behavior Analysis, 15, 295-301. Carr, E. G., & Kologinsky, E. (1983). Acquisition of sign language by autistic children: Spontaneity and generalization effects. Journal of Applied Behavior Analysis, 16, 297-314. Cavallaro, C. C., & Bambara, L. (1982). Two strategies for teaching language during free play. Journal of the Association for the Severely Handicapped, 1, 80-93. Cole, K. N., & Dale, P. S. (1986). Direct language instruction and interactive language instruction with language delayed preschool children: A comparison study. Journal of Speech and Hearing Research, 29, 206-217. Costello, J. M. (1983). Generalization across settings: Language intervention with children. In J. Miller, D. E. Yoder, & R. L. Schiefelbusch (Eds.), Contemporary issues in language intervention (pp. 275-297). Rochelle, MD: American Speech-Language-Hearing Association. Cross, T. G. (1977). Mother's speech adjustments: The contributions of selected child listened variables. In C. Ferguson & C. Snow (Eds.), Talking to children: Language input and acquisition (pp. 151-188). Cambridge, England: Cambridge University Press. Ervin-Tripp, S. (1971). Sociolinguistics. In J. Fishman (Ed.), Advances in the sociology of language (pp. 37). Mouton, The Hague. Fey, M. (1986). Language intervention with young children. San Diego: College-Hill Press. Garcia, E., Guess, D., & Byrnes, J. (1973). Development of syntax in a retarded girl using procedures of imitation, reinforcement, and modeling. Journal of Applied Behavior Analysis, 6, 299-310. Girolametto, L. E. (1988). Improving the social-conversational skills in developmentally delayed children: An intervention study. Journal of Speech and Hearing Disorders, 53, 156-167. Gray, B., & Ryan, B. (1973). A language program for the nonlanguage child. Champaign, IL: Research Press. Guess, D., & Baer, D. M. (1973). An analysis of individual differences in generalization between receptive and productive language in retarded children. Journal of Applied Behavior Analysis, 6, 311-331. Guess, D., Sailor, W., & Baer, D. M. (1978). Functional speech and language training for the severely handicapped. Lawrence, KS: H & H Enterprises, Inc. Halle, J. W. (1982). Teaching functional language to the handicapped: An integrative model of natural environment teaching techniques. Journal of the Association for the Severely Handicapped, 7, 29-37. Halle, J., Alpert, C. L., & Anderson, S. R. (1984). Natural environment language assessment and intervention with severely impaired preschoolers. Topics in Early Childhood Special Education, 4, 35-56. Halle, J. W., Baer, D. M., & Spradlin, J. E. (1981). Teachers' generalized use of delay as a stimulus control procedure to increase language use in handicapped children. Journal of Applied Behavior Analysis, 14, 387-400. Halle, J. W., Alpert, C. L., & Anderson, S. R. (1984). Natural environment language assessment and intervention with severely impaired preschoolers. Topics in Early Childhood Special Education, 4, 35-56. Haring, T. G., Neetz, J. A., Lovinger, L., Peck, C., & Semmel, M. I. (1987). Effects of four modified incidental teaching procedures to create opportunities for communication. Journal of the Association for Persons with Severe Handicaps, 12, 218-226. Hart, B. (1985). Naturalistic language training strategies. In S. F. Warren & A. Rogers-Warren (Eds.), Teaching functional language (pp. 63-88). Baltimore: University Park Press. Hart, B. M., & Risley, T. R. (1968). Establishing the use of descriptive adjectives in the spontaneous speech of disadvantaged preschool children. Journal of Applied Behavior Analysis, 1, 109-120. Hart, B. M., & Risley, T. R. (1974). Using preschool materials to modify the language of disadvantaged children. Journal of Applied Behavior Analysis, 7, 243-256.

40

KAISER, HEMMETER, ALPERT

Hart, B. M., & Risley, T. R. (1975). Incidental teaching of language in the preschool. Journal of Applied Behavior Analysis, 8, 411-420. Hart, B. M., & Risley, T. R. (1980). In vivo language intervention: Unanticipated general effects. Journal of Applied Behavior Analysis, I2, 407-432. Hart, B. M., & Risley, T. R. (1986). Incidental strategies. In R. L. Schiefelbusch (Ed.), Language competence: Assessment and intervention (pp. 213-225). San Diego: College-Hill Press. Hart, B. M., & Rogers-Warren, A. K. (1978). Milieu teaching approaches. In R. L. Schiefelbusch (Ed.), Language Intervention Strategies (Vol. II). Baltimore: University Park Press. Hemmeter, M. L., & Kaiser, A. P. (1986, October). The effects of environmental arrangement on parent-child language interactions. Paper presented at the Division of Early Childhood (DEC) Conference, Louisville, KY. Hester, P., & Hendrickson, J. (1977). Training functional expressive language: The acquisition and generalization of five-element syntactic responses. Journal of Applied Behavior Analysis, 10, 316. Jensema, C., Karchmer, M., & Trybus, R. (1978). The rated speech intelligibility of hearing impaired children: Basic relationships and a detailed analysis (Series R., No. 6). Washington, DC: Gallaudet College, Office of Demographic Studies. Johnston, J., & Schery, T. (1976). The use of grammatical morphemes by children with communication disorders. In D. Morehead & A. Morehead (Eds.), Normal and deficient child language (pp. 239-258). Baltimore: University Park Press. Kaiser, A. P., Alpert, C. L., Hancock, T. B., Hemmeter, M. L., & Ostrosky, M. M. (in preparation). The effects of training parents of preschool handicapped children to use milieu teaching: A longitudinal analysis. Vanderbilt University, Nashville, TN. Kaiser, A. P., Hendrickson, J. M., & Alpert, C. L. (1991). Milieu Language teaching: A second look. In R. Gable (Ed.), Advances in mental retardation and developmental disabilities (Vol. 4, pp. 63-92). London: Jessica Kingsley Publishers. Kaiser, A. P., & Warren, S. F. (1988). Pragmatics and generalization. In R. L. Schiefelbusch & L. L. Lloyd (Eds.), Language perspectives: Acquisition, retardation, and intervention (2nd ed., pp. 393-442). Austin, TX: Pro-Ed. Kaye, K., & Charney, R. (1980). How mothers maintain dialogue with two year olds. In D. Olson (Ed.), The social foundations of language and thought: Essays in honor of Jerome S. Bruner (pp. 146-162). New York: Norton. Leonard, L. B., Cole, B., & Steckol, K. F. ( 1979). Lexical usage of retarded children: An examination of informativeness. American Journal of Mental Deficiency, 84, 49-54. Lutzker, J., & Sherman, J. (1974). Producing generative sentence usage by imitative and reinforcement procedures. Journal of Applied Behavior Analysis, 7, 447-460. MacDonald, J. D. (1985). Language through conversation: A model for intervention with language delayed persons. InS. F. Warren & A. K. Rogers-Warren (Eds.), Teaching functional language: Generalization and maintenance of language skills (pp. 89-122). Austin, TX: Pro-Ed. MacDonald, J., & Gillette, Y. (1984). Conversation engineering: A pragmatic approach to early social competence. Seminars in Speech and Language, 5, 171-184. Mahoney, G. (1975). An ethological approach to delayed language acquisition. American Journal of Mental Deficiency, 80, 139-148. Mahoney, G., & Powell, A. (1986). The transactional intervention program teacher's guide. Farmington, CT: Pediatric Research and Training Center. Mahoney, G., & Snow, K. (1983). The relationship of sensory motor functioning to children's response to early language training. Mental Retardation, 2I, 248-254. Manolson, A. (1985). It takes two to talk: A Hanan early language parent guidebook. (Available from the Hanen Resource Centre, Suite 4-126, 252 Bloor St., West, Toronto, Ontario, Canada, M5S 1V6) Maurer, H., & Sherrod, K. B. (1987). Context of directives given to young children with Down

1.

COMMUNICATION DISORDERS

41

syndrome and non-retarded children: Development over two years. American Journal of Mental Deficiency, 91, 579-590. Maxwell, S. E., & Wallach, G. P. (1984). The Language-learning disabilities connection: Symptoms of early language disability change over time. In G. P. Wallach & K. G. Bulter, Language learning disabilities in school-age children (pp. 15). Baltimore, MD: Williams & Wilkins. McConkey, R., & O'Conner, M. (1982). A new approach to parental involvement in language intervention programmes. Child: Care, Health, and Development, 8, 163-176. McCormick, L. (1984). Perspectives on categorization and intervention. In L. McCormick & R. L. Schiefelbusch (Eds.), Early language intervention (pp. 99-116). Columbus, OH: Charles E. Merrill. McGee, G. G., Krantz, P. J., Mason, D., & McClannahan, L. E. (1983). A modified incidentalteaching procedure for autistic youth: Acquisition and generalization of receptive object labels. Journal of Applied Behavior Analysis, 16, 329-338. McGee, G. G., Krantz, P. J., & McClannahan, L. E. (1985). The facilitative effects of incidental teaching on preposition use by autistic children. Journal of Applied Behavior Analysis, 18, 1731.

McLeavey, B. C., Toomey, J. F., & Dempsey, P. (1982). Nonretarded and mentally retarded children's control over syntactic structures. American Journal of Mental Deficiency, 86, 485-494. McReynolds, L., & Engmann, D. (1974). An experimental analysis of the relationship of subject and noun phrases. ASHA Monographs, 18, 30-47. Moerk, E. L. (1983). The mother of Eve as a first language teacher. Norwood, NJ: Ablex. Mudd, J. M., & Woolrey, M. (1987). Training head start teachers to use incidental teaching. Journal of the Division for Early Childhood, 11, 124-134. Mulac, A., & Tomlinson, C. (1977). Generalization of an operant remediation program for syntax with language-delayed children. Journal of Communication Disorders, 10, 231-244. Naremore, R., & Dever, R. (1975). Language performance of educable mentally retarded and normal children at five age levels. Journal of Speech and Hearing Research, 18, 82-95. Nelson, K. E. (1989). Strategies for first language teaching. In M. L. Rice & R. L. Schiefelbusch (Eds.), The leachability of language (pp. 263-310). Baltimore, MD: Brookes Publishing Co. Nelson, K. E., Carskaddon, G., & Bonvillian, J.D. (1973). Syntax acquisition: Impact of experimental variation in adult verbal interaction with the child. Child Development, 44, 497-504. Newport, E., Gleitman, L., Gleitman, H. (1977). Mother I'd rather do it myself: Some effects and non-effects of motherese. In C. Snow & C. Ferguson (Eds.), Talking to children: Language input and acquisition (pp. !09-150). Cambridge, England: Cambridge University Press. Oliver, C. B., & Halle, J. W. (1982). Language training in the everyday environment: Teaching functional sign use to a retarded child. The Journal of the Association for the Severely Handicapped, 8, 3. Ostrosky, M. M., & Kaiser, A. P. (1991). Arranging preschool classroom environments to promote communication. Teaching Exceptional Children, 23(4), 6-10. Owens, R. E., Jr., (1988). Language development: An introduction. Columbus, OH: Charles E. Merrill. Owens, R., & MacDonald, J. (1982). Communicative uses of the early speech of nondelayed and Down syndrome children. American Journal of Mental Deficiency, 86, 503-510. Polloway, E. A., & Smith, J. E. (1982). Teaching language skills to exceptional/earners. Denver, CO: Love Publishing Company. Rogers, S. J., Herbison, J. M., Lewis, H. C., Pantone, J., & Reis, K. (1986). An approach for enhancing the symbolic, communicative, and interpersonal functioning of young children with autism or severe emotional handicaps. Journal of the Division for Early Childhood, 10(2), 135148. Rogers-Warren, A. K., & Warren, S. F. (1980). Mand for verbalization: Facilitating the display of newly-taught language. Behavior Modification, 4, 361-382.

42

KAISER, HEMMETER, ALPERT

Ronda!, J. (1978). Maternal speech to normal and Down syndrome children matched for mean utterance length. In C. E. Myers (Ed.), Quality of life in severely and profoundly mentally retarded people: Research foundations for improvement (pp. 193-265). Washington, D.C.: American Association on Mental Deficiency. Schacter, F. F. (1979). Everyday mother talk to toddlers: Early intervention. New York: Academic Press. Schiefelbusch, R. (Ed.). (1963). Language studies in mentally retarded children. Journal of Speech and Hearing Disorders, Monograph Supplement No. 10. Schumaker, J., & Sherman, J. (1970). Training generative verb usage by imitation and reinforcement procedures. Journal of Applied Behavior Analysis, 3, 273-287. Seitz, S. (1975). Language intervention: Changing the language environment of the retarded child. In R. Koch & F. de Ia Cruz (Eds.), Down syndrome (pp. 157-179). New York: Bruner/Maze!. Shatz, M. (1987). Bootstrapping operations in child language. In K. E. Nelson & A. van Leek (Eds.), Children's language, Vol. 6 (pp. 1-22). Hillsdale, NJ: Lawrence Erlbaum Associates. Skinner, B. F. (1957). Verbal behavior. New York: Appleton-Century-Crofts. Snow, C., & Ferguson, C. (Eds.). (1977). Talking to children: Language input and acquisition. Cambridge, England: Cambridge University Press. Snyder, L. S., & Silverstein, J. (1988). Pragmatics and child language disorders. In R. L. Schiefelbusch & L. L. Lloyd (Eds.), Language perspectives, acquisition, retardation, and intervention (2nd Ed., pp. 189-222). Austin, TX: Pro-Ed. Spradlin, J. (1963). Language and communication of mental defectives. In N. Ellis (Ed.), Handbook of mental deficiency (pp. 8-31). New York: McGraw-HilL Spradlin, J. E., & Siegel, G. M. (1982). Language training in natural and clinical environments. Journal of Speech and Hearing Disorders, 47, 2-6. Stokes, T. F., & Baer, D. M. (1977). An implicit technology of generalization. Journal of Applied Behavior Analysis, 10, 349-367. Stremel, K., & Waryas, C. (1974). A behavior-psycholinguistic approach to language training. ASHA Monographs, 18, 96-130. Tenth Annual report to Congress on the implementation of the Education of the Handicapped Act. (1988). U.S. Department of Education. Warren, S. F., & Kaiser, A. P. (1986). Generalization of trained language by young language delayed children: Longitudinal analysis. Journal of Speech and Hearing Disorders, 51, 239-251. Warren, S. F., McQuarter, R. J., & Rogers-Warren, A. K. (1984). The effects of mands and models on the speech of unresponsive socially isolate children. Journal of the Association for the Severely Handicapped, 8, 18-31. Warren, S. F., & Rogers-Warren, A. K. (1980). Current perspectives in language remediation: A special monograph. Education and Treatment of Children, 5, 133-153. Warren, S. F., & Rogers-Warren, A. K. (1985, July). Language remediation with young developmentally delayed children. Paper presented at the Interamerican Congress of Psychology, Caracas, Venezuela. Waryas, C. L., & Stremel-Campbell, K. (1978). Grammatical training for the language-delayed child: A new perspective. In R. L. Schiefelbusch (Ed.), Language intervention strategies (pp. 145-192). Baltimore: University Park Press. Waryas, C. L., & Stremel-Campbell, K. (1983). Communication training program. New York: Teaching Resources. Weiss, R. S. (1981). INREAL intervention for language handicapped and bilingual children. Journal of the Division for Early Childhood, 4, 40-52. Welch, S. J., & Pear, J. J. (1980). Generalization of naming responses to objects in the natural environment as a function of stimulus modality with retarded children. Journal of Applied Behavior Analysis, 13, 629-643.

1.

COMMUNICATION DISORDERS

43

Wells, G. ( 1986). Language, learning and teaching: Helping children to make knowledge their own. In F. Lowenthal & F. Vandamme (Eds.), Pragmatics and education (pp. 57-80). New York: Plenum. Wetherby, B. (1978). Miniature languages and the functional analysis of verbal behavior. In R. L. Schiefelbusch (Ed.), Bases of language intervention (pp. 397-448). Baltimore: University Park Press. Wheeler, A., & Sulzer, B. (1970). Operant training and generalization of a verbal response form in a speech deficient child. Journal of Applied Behavior Analysis, 3, 139-147. Wheldall, K. (1976). Receptive language development in the mentally handicapped. In P. Berry (Ed.), Language and communication in the menta//y handicapped (pp. 36-55). Baltimore: University Park Press. Whorf, B. L. (1956). Language, thought and reality. Cambridge, MA:MIT Press. Yoder, P. J., Kaiser, A. P., & Alpert, C. L. (1989, March). A comparison of didactic and milieu teaching in sma/1 groups of handicapped preschoolers of varying developmental levels. Paper presented at the American Speech-Language-Hearing Foundation Conference, San Antonio, TX. Zwitman, D., & Sonderman, J. (1979). A syntax program designed to present base linguistic structures to language-disordered children. Journal of Communicative Disorders, 13, 232-237.

Page Intentionally Left Blank

2

Developing, Implementing, and Evaluating a Preventative Intervention for High Risk Transfer Children

Leonard A. Jason David Betts Joseph H. Johnson Andrew M. Weine

Lisa Neuson Laura Filippelli Cecile Lardon

DePaul University

Approximately 6 million children between the ages of 5 and 13 transfer to new schools each year (Cornille, Bayer, & Smyth, 1983). For years, concerned educators and parents have sought to learn the effects of such mobility on children's adjustment and achievement. Although the literature is somewhat mixed on this important issue, it is possible to identify a subset of transfer students who are at-risk for academic failure. For such children, transfer represents a stressor that combines with vulnerabilities in academic areas and other environmental life stressors. A majority of school systems have developed orientation programs to ease the transition process. Among strategies used are helping the newcomers reach higher academic levels through tutoring, providing buddies who inform the transfer students about the schools' customs and routines, and offering group discussions to share experiences. Although the proliferation in recent years of these programs is an encouraging development, few if any of these interventions have been evaluated systematically. A primary thesis of this chapter is that primary preventive programs can be organized around key milestones (e.g., transferring to a new school) with the purpose of helping vulnerable youngsters navigate transitional states and adapt to new settings (Feiner, Farber, & Primavera, 1983). Providing at-risk transfer students supportive relationships and at-risk transfer students supportive relationships and competencies represents a type of preventive intervention that during the past 20 years transition and crisis theorists have recommended as having rich potential in ameliorating transitional difficulties (Caplan, 1965; Hirschowitz, 1976; Parkes, 1971 ).

45

46

JASON ET AL.

Scheduled and Unscheduled Transitions

Two general types of school transfers exist: scheduled and unscheduled. A scheduled transfer involves a student entering school or graduating from elementary school to a middle or high school (Elias, Gara, & Ubriaco, 1985). School entry has been found to be associated with increased levels of stress, tension, and irritability (Klein & Ross, 1965). For the transition into high school, Felner, Primavera, and Cauce (1981) found that frequently transferring White and Black youngsters evidenced grade-point average decreases of one half of a grade point. Stewart et al. ( 1982) found that, upon entering a new school, children often feel sad, abandoned, and helpless. Several promising interventions have been developed for these types of scheduled transitions. Signell (1972), for example, provided parents small discussion groups prior to and during kindergarten to help them assist their children in overcoming fears associated with beginning school. Unfortunately, no formal evaluation of this intervention occurred. Felner, Ginter, and Primavera (1982) attempted to ease homeroom teachers assumed guidance and administrative responsibilities for the program students, and the experimental program children took their primary classes with other program children. This effort to build social support among the youngsters was successful. At the end of the school year, project children evidenced significantly better attendance, grades, and self-concept when compared with controls. Two additional scheduled transition projects are in progress. In the first, Hirsch (1985) is studying students' transition into high school, and examining the relationships between dimensions of social networks, and children's coping and mental health. Elias et al. (1986) have taught elementary school children social problem-solving strategies and are now assessing whether these skills will help the children in their transition into middle schools. Unscheduled school transitions involve individual youngsters switching from one school to another. Researchers have posited the possible negative effects of unscheduled transitions on student performance and behavior. Several investigators report that increased school mobility is associated with multiple psychopathological disorders (Millet, Pon, Guibaud, & Auriol, 1980; Stufflefield, 1955). Even preschoolers who transfer to a new school evidence increases in negative statements and activity level, and changes in eating and sleeping patterns (Tiffany, 1984). Frequent movers describe themselves as more insecure, inconsistent, complaining and critical (Shaw, 1979), and as having fewer welldeveloped peer ties (Douvan & Adelson, 1966) when compared to less-frequent movers. In Kantor's (1965) study, mobile families had significantly more child behavior problems than families who did not change residence. School children's negative attitudes toward newcomers probably further intensify problems in the social and psychological domain (Schaller, 1975). Negative effects of a transfer have also been found by Brockman and Reeves ( 1967), Collins and Coulter

2.

A PREVENTIVE INTERVENTION FOR HIGH RISK CHILDREN

47

(1974), and Pedersen and Sullivan (1964). It is not surprising that, from a list of 35 life events, parents of elementary age youngsters rank school mobility as the fourth most significant childhood problem (Coddington, 1972a, 1972b). Frequently transferring youngsters also experience problems in academic areas (Bensen, Haycraft, Steyaert, & Weigel, 1979; Brockman & Reeves, 1967; Schaller, 1974). In an inner-city school, Levine, Wesolowski, and Corbett (1966) found that no child with an A average had previously enrolled in more than two schools, whereas 38% of children with D and F averages had attended two or more schools. If a child is already academically disadvantaged, the effects of mobility might be particularly severe (Turner & McClatchey, 1978). These concerns are also shared by school counselors who admit to having frequent problems in accommodating the academic deficiencies of transfer students (Comille et al., 1983). Other investigators, however, have failed to find adverse effects of school transitions (Barrett & Noble, 1973; Cramer & Dorsey, 1979; Goebel, 1978; Greene & Daughtry, 1961; Kroger, 1980; Moore, 1966; Perrodin & Snipes, 1966; Schaller, 1976; Ziller & Behringer, 1961 ). A failure to control for Socioeconomic status (SES) (Lacey & Blane, 1979; Price, 1965; Tetreau & Fuller, 1942) helps account for differences in these groups of studies. Children from lower SES backgrounds are more likely to have difficulty adjusting to school transfers whereas children from middle to upper class backgrounds are more likely to profit from the varied experiences involved in changing schools (Glidewell, Kantor, Smith, & Stringer, 1965). In Long's (1975) study, mobility was associated with increasing likelihood of being enrolled below the modal grade level; however, this relation did not hold for children of college graduates. Morris, Pestaner, and Nelson (1967) found that the negative effects of mobility on achievement were most prominent with low SES children. Downwardly mobile families, according to Morris et al., are less equipped to deal with the pressures of new and demanding situations. They also have difficulty transmitting the values associated with achievement (e.g., planning and meeting longrange goals, meeting standards of excellence, postponing gratifications in the interests of long-range self improvements), and their children have more problems adapting to new social and educational milieus. Identifying At-Risk Children The research raises two key questions: how to prevent the negative consequences of school transfers, and how to identify those transfer students with increased risk of becoming disorganized and demoralized following the transfer (Freedman, 1964). In addition to being in a low SES family, youngsters may handle the challenge of coping with a school transfer inadequately if they are also confronted with multiple stressors (Shaw & Pangman, 1975) and lack requisite personal and social resources (Goebel, 1981; Gilliland, 1958; Holland, Kaplan,

48

JASON ET AL.

& David, 1974; Jason, Thompson, & Rose, 1986; Turner & McClatchey, 1978; Whalen & Fried, 1973). Several investigators make clear the detrimental effects of multiple risk factors. For example, Rutter (1979) found that if children had only one risk factor (e.g., severe parental discord, low social status, overcrowding), they were no more likely to have psychiatric disorders than children with no risk factors. However, if there were two risk factors, the probability of disorder increased fourfold, and with even more risk factors, rates of disabilities further escalated. In another study, Sandler and Ramsay (1980) found that maladapting children and normal controls significantly differed on two measures of stressful life events: family troubles and entrance events (i.e., addition of a new person to the family, move to a new environment). During a school transfer, youngsters are faced with the task of gaining peer and teacher acceptance, learning about the school's rules and regulations, and meeting new academic standards; those children who fail in these tasks often have fewer personal resources and are confronted with multiple life stressors that detract from their focusing on the task of mastering the demands of the new setting. The above research strongly indicates that children at high risk for school difficulties and failure are from lower SES backgrounds and are exposed to multiple life stressors. If such children received the necessary resources and support, patterns of long-term poor achievement and school failure might be averted. In support of this proposition, Cowen, Lotyczewski, and Weissberg (1984) found that high-risk children with high resources had better adjustment scores than high-risk children with lower resources. A variety of studies measuring social support and resources (Burke & Weir, 1978; Cauce, Feiner, & Primavera, 1982, Durlak & Jason, 1984; Lorion, Work, & Hightower, 1984; Nair & Jason, 1985) theoretically support the notion that a supportive environment could aid high-risk youngsters appreciably in coping with a milestone transition such as a school transfer.

Selecting a Developmental Period Although the leading developmental theorists of psychopathology in children differ in their theoretical viewpoints, they converge on several issues relevant to high-risk transfer students (Achenbach, 1974). For example, Piaget and other developmental psycholinguists indicate that at different ages, children think in qualitatively different ways, and undergo enormous changes in social roles. Children also pass through a sequence of maturational stages of development. During the first few years of school (the primary phase), children are interested in writing numbers and counting; they engage in imaginative play but without much consideration to plot or sequence (Jarvis & Wooton, 1969). The intermediate phase, which lasts until grade 6, is the age beyond childhood and before adolescence. During this phase, children learn to spell about 1,500 common words, their reading vocabulary expands to 6,000 words, and they

2.

A PREVENTIVE INTERVENTION FOR HIGH RISK CHILDREN

49

develop insight into simple fractions. Sex differences between boys and girls show up more markedly, and strong attachments are formed with same sex friends. The children increasingly become involved with peers and begin to move away from their parents (Ausubel, Sullivan, & lves, 1980). During this intermediate phase, belonging to a group has a strong attraction and serves a needed function for children (Jarvis & Wooton, 1969). This developmental issue is particularly important for transfer children, because they report being most worried about establishing new friendships after a move (Schaller, 1975). Kistner and Torgesen ( 1987) describe a complex set of events as children move from the primary to the secondary phase. In kindergarten and I st grade, children are provided a lot of social reinforcement for good classroom conduct, and they are rarely criticized for the accuracy of their academic work. In addition, at this primary phase, children view their abilities as determined by their efforts so that even if they repeatedly fail, they do not interpret this to mean that they will always fail. Intermediate age children are more likely to receive positive teacher feedback for academic performance than good behavior. Low achievers will receive less reinforcement and consequently lower their perceptions of their abilities. As children reach the intermediate age, academic failures are more likely to lead to debilitating beliefs of their abilities. In addition, as these children begin to compare themselves to normative groups, they are more likely to evaluate themselves negatively. The transfer child in the intermediate phase may feel isolated due to the lack of acceptance by peers. Whether the move is from one state to another, or a public school to a parochial school, the child will feel insecure and lonely for a period (Hurlock, 1964). The new peer group has different values and expectations and may not approve of behavior judged acceptable by the old group. Transfer children lack what children of the intermediate period crave, friends who want to play with them and accept them as a member of a group. Those who transfer with gaps in their school work are even more vulnerable since school performance indicates to children how they compare to classmates. School performance influences children's self-esteem, and leads to whether children feel adequate, or a failure and an outcast. Levine (1983) points out a number of developmental acquisitions during early childhood: honing of selective attention; retaining, retrieving, and displaying increasingly complex sequences of information; developing the ability to preserve and find previously encountered data; mastering concrete operations; and interpreting social feedback. According to Levine (1983), children can acquire these capacities and sustain their self-esteem when they have a record of mastery, obtain reinforcement from parents, feel in control, and receive positive feedback from peers. Children who transfer into a new school need to feel they can acquire knowledge and keep up with peers. Those who cannot master the new skills might give up rather than be humiliated by trying and failing. Even though the transfer

50

JASON ET AL.

children try to find social acceptance beyond the family, the home remains a critically important arena of support. Transfer children need to receive from their parents regular reinforcement and credit for what they have achieved in school. If transfer children's academic lags prevent them from receiving these supports, the children might stop making an effort to succeed and begin associating with similarly alienated peers (Rhodes & Jason, 1988). Intermediate age children may be particularly vulnerable to school transitions. An intervention developed by the authors and described below focuses on intermediate age children. These youngsters are in the process of expanding their cognitive skills, developing new social competencies, and forming close friendships. Of course vulnerable children at any age level could profit from interventions designed to ease the transition into a new school, but several studies suggest that transfer children in the first few grades are less at-risk for developing academic and social problems than intermediate age youngsters (Barrett & Noble, 1973; Ziller & Behringer, 1961). In addition, by the time some children matriculate to higher grade levels, repeated academic failure and lowered selfesteem already have combined to produce ingrained patterns of self-defeating behavior and negative belief systems that are difficult to modify. By intervening in the intermediate phase, parents, teachers, and mental health workers can prevent school failure and later drop out. In an urban sample, 37% of students whose reading and math skills were below the national norms at 6th grade went on to drop out of high school (Kyle, Lane, Sween, & Triana, 1986). In another study, Hess ( 1987) indicated that the roots of school failure and later drop out could be traced to the 3rd grade, a time when children begin to fall behind in reading skills, and never are able to eliminate these deficits. Forty-four percent of high school dropouts report schoolrelated reasons for leaving school (e.g., performed poorly, disliked school, was expelled or suspended) (Rumberger, 1983). Many school dropouts are unemployed, and even those with full or part-time jobs are only able to obtain lowskilled jobs. Most regret their decision to leave school prematurely (Bulletin, 1983). Early intervention programs that develop more positive school experiences and attitudes during critical transition points will prepare children for the more rigorous demands of junior high and high school, thus preventing the disappointments and failure that typify the backgrounds of school drop-outs. Sroufe and Rutter (1984) state that broad-based indicators of failure (e.g., achievement problems, inadequate peer relations) during school years strongly predict later adult disorders. Vaillant and Vaillant (1981) found that the best predictors of positive adjustment in adulthood were the willingness and capacity to work in childhood (e.g., regular participation in school activities). Because achievement is associated with social competence, achievement motivation and delay of reward, underachievement statistically predicts all major forms of adult maladjustment (Kohlberg, Ricks, & Snarey, 1984). Furthermore, Kohlberg et al. stated that because factors other than intelligence are set in the first few years of

2.

A PREVENTIVE INTERVENTION FOR HIGH RISK CHILDREN

! FAMILY~

iSCHOOL \

I ·~COPING I ··.

CHILD

I

FIG. 2.1.

51

/

\

10UTCOMES/

COMMUNI1Y ;

Relationship of risk factors to coping and outcomes.

schooling (e.g., learning, attention, sense of competence), the years 9-12 are a critical period for mental health intervention. Figure 2.1 illustrates factors that influence coping with the new school, including short- and long-term outcomes. Ultimately, lisrel analyses will enable researchers to discover which combination of factors is most important for youngsters' adjustment following the transfer. We believe that SES and parenting factors (which are family variables) and internal competencies of the child (which are child factors), are some of the most powerful markers of how well children cope with the school transition. In addition, school variables and community influences also probably play a role in mediating the children's adjustment into the new school. Intervention Studies While many school systems have been concerned about the needs of transfer students and developed generic orientation programs, few have evaluated such programs (Cornille, Bayer, & Smyth, 1983). Several programs for transfer students have been published. Hirschowitz (1976) described a project where volunteer mothers served as discussion leaders for small groups of school newcomers. No data were collected to evaluate this program. In another project, Levine (1966) trained upper grade children to serve as guides for transfer children, giving them tours of the building, instructions about general rules, and information about building, instructions about general rules, and information about facilities in the school and neighborhood. This innovative program was not evaluated. Finally, in a comprehensive program by Donohue and Gullotta (1981 ), families moving into a community were visited by a volunteer from the parent-teacher association who provided parents with booklets about the school and local resources. In addition, transfer students were provided an in-school buddy during the first week of school. Once again, a formal evaluation has not been conducted for this promising preventive program. Panagos, Holmes, Thurman, Yard, and Spaner (1981) conducted a more empirically based project aimed at helping transfer students, who had previously lived in a large urban area, adjust to a new school in the suburbs. Because most

52

JASON ET AL.

of the transfer children had educational deficits, they were provided special classes to catch up academically with their peers. In addition, parents received a comprehensive orientation program, and teachers participated in special workshops. Although children at year end showed significant gains in academic areas, the lack of a control group makes it difficult to assess whether the influence of the program or the high quality suburban school accounted for the improvements. Study 1: Douglas and Jason (1979). The first author has been interested in area of transitions as a site for preventive interventions for a number of years (Cradock, Cotler, & Jason, 1978; Jason & Burrows, 1983; Zoline & Jason, 1985). The impetus for the first in a series of school transitions studies conducted at DePaul University occurred when a principal approached the first author to ask for help with one of his new teachers who was experiencing classroom management problems with an incoming class of 1st graders (Douglas & Jason, 1979). These new students in the school were engaging in inappropriate behaviors more than 40% of the time. (Independent observers collected behavioral classroom data throughout the study.) This was the students' first exposure to the school setting, and they clearly had not learned appropriate classroom behaviors. Using a multiple baseline design, the researchers provided one-half of the class a home note system; and later, the other children were also provided with intervention. Students could earn a home note (called a "Happygram") if their classroom behavior was at least 75% appropriate. This note could be taken home, where the parent would then provide social rewards. For each group of children (unit of analysis was one half of the classroom), appropriate classroom behavior rose to over 80% with implementation of the home note system. After the home notes were gradually faded out of the classroom, follow-up data indicated that children maintained the gains achieved during the program. The new 1st graders indirectly received positive survival skills (listening to the teacher, sitting in one's seat, etc.) that are prerequisites for learning to occur. Study 2: Bogat, Jones, and Jason (1980) The next study occurred when another principal told the first author that his parochial school would be closing, and he requested help in preparing his children for entering a public school (Bogat, Jones, & Jason, 1980). The 17 2ndthrough 8th-grade transfer students were divided into two groups: (a) an experimental group (E) who were exposed to a peer-led, preventive orientation program prior to the start of the new school year and (b) a control group who received no such program. A second control group consisted of matched children already in attendance at the school into which the parochial students would transfer. The orientation program for Es occurred prior to the beginning of the school year. The Es received booklets containing school information, were given a tour of the

2.

A PREVENTIVE INTERVENTION FOR HIGH RISK CHILDREN

53

school, and participated in a peer-led discussion session that focused on feelings about the private school and its unfortunate closing. During the second day, the children met the principal and vice principal. They then divided into two teams and competed in a treasure hunt game (this provided the youngsters an opportunity to explore all the rooms in the building). Peer guides then led a discussion that focused on two themes: (a) differences and similarities between the public and private school, and (b) feelings about being a transfer student. Prepoint measures were collected prior to the transfer (i.e., at the end of the school year) and postpoint measures were collected 2 weeks after the opening of the next school year. The E group showed significantly greater increases in peerrelated self-esteem than either of the control groups. The E group at postpoint knew significantly more rules than both of the control groups. Measures taken after the first academic quarter following the transfer indicated that the Es also exhibited significantly better school conduct ratings than both control groups. In a student satisfaction questionnaire administered to the E group of students, they highly recommended the program to other transfer students, and they particularly liked the peer-led component. Although these findings are encouraging, small sample size and lack of longer-term follow-up data limit any conclusions. Study 3: Sloan, Jason, and Bogat (1984)

The next study aimed to isolate the beneficial components of the orientation program (Sloan, Jason, & Bogat, 1984). Eighty-seven lst- through 8th-grade transfer students were assigned randomly to one of four groups: discussion only, slide presentation only, discussion and slide presentation, or a no-treatment control. Four peer guides were trained to read the dialogue developed to accompany the slides and were given opportunities to practice various techniques for promoting group discussion. Transfer students in the slide presentation group observed 35 slides that depicted various school locations and important school personnel. In the discussion condition, students introduced themselves and generated two or three words that described their feelings about transferring into new schools. Peer guides expressed support, empathy, and concern for the feeling that the new students experienced. A general discussion about transferring followed. Standardized assessment measures were administered l week before and after the orientation programs. At postintervention, the youngsters in the discussion and slide-discussion group knew significantly more information about the school than the controls, and only the slide-discussion group members evidenced significantly lower anxiety than the controls. The discussion and discussion-slide group members also evidenced significant positive changes in their attitudes toward school. At the end of the first academic quarter, the control children had been tardy 15 days, the slide group children had been tardy 10 days, the discussion group children 7 days, and the slide-discussion group youngsters 5 days. These results indicated that the students' opportunity to share feelings related to a

54

JASON ET AL.

school transfer is a critical step in a child's adjustment to a new school. Peer guides provided the transfer students an effective model for coping in the new setting. By giving the transfer students an opportunity to hear other children express similar fears and concerns about the transition, the program decreased the program children's preoccupation with these issues, bolstered their attendance, and enhanced their ability to retain information about the school. Alternatively, the discussion group's effectiveness might have been due to helping the children to interact and make friends. Study 4: Jason and Bogat (1983) The next study incorporated many of the ideas learned in previous orientation programs (Jason & Bogat, 1983). Fifty-two 1st- through 7th-grade transfer children were assigned randomly to a comprehensive intervention (E) or a control condition (C). The intervention included information, a peer-led discussion group, and a tour. A major addition to this program was assigning a "buddy" to each program child, and coupling these buddy/transfer student pairs in as many activities as possible to establish for the child an immediate social network. As with previous studies, after the program, the Es knew significantly more general information (rules, regulations, etc.) about the school than the Cs. A series of questions probed the nature of the 2-month-old relationship between transfer students and their guides. Two months after the orientation, 65% of the transfer Es said their buddy introduced them to other students in the school, 70% said their buddy sat with them at lunch, 65% said that they asked their buddy questions about school rules, 45% said that they told their buddies some private feelings, 30% said they played with their buddy in the neighborhood, 45% said that their buddy helped them with their homework, 90% mentioned that they had played together on the playground during recess, and 90% described their buddy as a good or very good friend. Yet, on standardized achievement tests, there were no significant differences at the end of the first academic quarter for Es and Cs. From conversation with the teachers and observations of the transfer children, it was clear that some of the transfer children did not have the academic competencies to master required classroom work or the personal resources to feel confident and able to resolve daily problems in their new setting. While these at-risk children reacted positively to an orientation program that provided information, social support, and opportunities to discuss feelings, certain youngsters also appeared to need assistance in mastering the new academic expectations in their new school setting.

TUTORING STUDIES Over a period of years, the first author assessed a variety of tutoring programs for elementary school children (Jason, Christensen, & Carl, 1982). Findings indicate

2.

A PREVENTIVE INTERVENTION FOR HIGH RISK CHILDREN

55

that children who underwent the competency-enhancing programs, when compared to nonprogram children, significantly improved both in achievement areas and classroom behavior (Jason, Ferone, & Soucy, 1979; Jason, Frasure, & Ferone, 1981). Pillen, Jason, and Olson (1988) assigned all 26 1st- through 4th-grade transfer students to a computer-assisted tutoring program or to a control condition. The male transfer students improved their reading scores over time and the girl transfer students improved in their self-esteem. It is clear that boys and girls reacted differentially to the tutoring project, and future projects will need to closely monitor gender effects. Rationale for a Comprehensive Preventive Intervention Published, as well as unpublished, findings indicate that some children tend to have difficulties in the process of coping with the demands of a new setting. Transfer children who are most anxious have significantly less positive attitudes toward school in general, their peers, and the social structure of the school (Sloan, Jason, & Bogat, 1984). Those with high anxiety also have significantly lower self-appraisal (Jason & Bogat, 1983). Transfer youngsters who perceived themselves in negative terms in regard to their new school setting obtain significantly lower scores on math, spelling, and reading achievement tests during the school year (Jason & Bogat, 1983). The anxiety and negative self-perceptions are due to children having inadequate personal resources and support to meet the demands of the new setting adequately. High-risk groups of children can be identified who continue to be at risk even after an orientation program. In the Jason and Bogat ( 1983) study, a high-risk group of children was identified whose scores in reading, spelling, and math achievement were one standard deviation or more below the mean after 2 months of school. These youngsters were not provided remedial help, and their difficulties in school were evident. Tutoring programs have been developed that have been shown to boost academic achievement, and orientation programs have been evaluated as easing the transition into new schools. In addition, high-risk groups of transfer students having multiple risk factors (e.g., life stressors, academic problems, low SES) have been observed to have considerable difficulty in mastering the demands of the new school setting. A comprehensive orientation program having academic tutoring as one component was evaluated to determine its effect in enhancing the personal competencies of at-risk transfer children, and allowing them to master the social and academic demands of their new setting successfully. Setting and Participants The study took place in 20 inner-city coed, parochial elementary schools in Chicago. Pairs of schools were matched for size and racial composition and then

56

JASON ET AL.

one member of each pair randomly assigned to experimental or control conditions. Thus, 10 schools received the experimental intervention and the remaining I 0 schools served as controls. Our project focused on transfer children entering the 3rd, 4th, and 5th grades. By focusing on the intermediate phase, there is less developmental variation among the sample. During a transfer, children in this phase are vulnerable, and if early social and academic problems are not resolved, these children are at risk of developing later impairments in school performance, social relations, and adjustment. Three criteria were used to identify the risk factors. The high-risk children were from a lower SES background (Hollingshead, 1975), evidenced a lag in at least one area of a standardized achievement test (Jastak & Wilkinson, 1984), and experienced two or more life stress events based on parent reports (Sandler, 1980; Sandler & Block, 1979). These indices were selected based on past research that indicated lower SES children have more difficulty with school transitions. Figure 2.2 presents the effect of SES (higher numbers indicate lower socioeconomic status) on the Wide Range Achievement (Jastak & Wilkinson, 1984) test scores for all incoming transfer students during the first year of the project. In addition, youngsters with fewer academic competencies (defined here as lags on achievement tests) are more vulnerable following a school transition. A growing body of research points to cumulative life-stress events as impeding children's ability to focus their attention on the academic demands of school. Low SES children, entering school with mild achievement lags and distracted by several life-events stressors, are high-risk candidates for further falling behind in their studies, and consequently beginning a pattern of school frustration and ultimate failure. When program teachers, during the middle of the year, were asked to list the criteria they would use to select at-risk transfer children, for their first choice, 95% of the teachers picked one of our three indices; for second choice, 95% again picked one of our three measures; and for third choice, 72% picked one of our three measures. The 3rd-, 4th- and 5th-grade transfer students took group achievement and 110 p

106

w

100

R E R A

T

96

R 8

90 86

FIG. 2.2. Prepoint wide range achievement reading scores versus SES.

3

SES

2. .

A PREVENTIVE INTERVENTION FOR HIGH RISK CHILDREN

57

self-esteem tests. For children who did not speak English, a tester who spoke the child's native language administered individualized testing. Immediately after this 1-hour testing session, the orientation program was delivered to children in the experimental group. Any child with low SES, several stressful life events, and a score in the bottom 33rd percentile on at least one area of the achievement test received criterion-referenced tests in areas to be described below. High-risk E children then began twice weekly tutoring sessions. Orientation Program

The !-hour orientation program occurred at the elementary school following pretesting. This session occurred during one of the first few days of the new school year. The act of entering a new school can be conceptualized as the unconditioned stimulus that can elicit the unconditioned response of anxiety. By pairing the entering of school with a positive experience (i.e., the orientation program), the intervention induced a more relaxed emotional state. The theoretical basis for this part of the program comes from Poser's (1970) work on antecedent systematic desensitization, which involves pre-exposing individuals to conditioned and unconditioned stimuli to inhibit the formation of conditioned avoidant responses. The children were placed in groups of 15 to 20 children. Within these groups, the youngsters were provided a review of school rules, personnel, information, and clubs. The students were also encouraged to ask questions about the school and activities. The theoretical basis for this part of the program lies within the literature on psychological inoculation or anticipatory guidance (Caplan, 1965). The children cognitive1y were prepared for stressful situations by anticipating and thinking about the potentially difficult experience of transferring into a new school. By emphasizing the rules and information, the research team familiarized the children with the school: Students then had a better appreciation of the school's expectations, and were better prepared to function without elevated levels of arousal. In a study of various orientation methods, Sloan, Jason, and Bogat (1984) found that a program that included transmission of basic school information as well as a discussion of feelings associated with the transfer experienced significantly reduced anxiety. The school information was conveyed by discussions led by a 6th-grade peer leader, who was a former transfer student. In part, the rationale behind including the 6th grader with these particular topics stems from the modeling literature (Bandura, 1978). A friendly and confident peer guide who modeled adaptive coping responses and portrayed an enthusiastic attitude toward school provided the transfer students a potential model for effective coping in their new setting. In addition, the peer guides' revelations of their own positive experiences following their transfer into the school enhanced the transfer children's sense of efficacy in mastering their own transitions.

58

JASON ET AL.

When the children discussed feelings, the project personnel, rather than the peer leaders, guided the discussion because they possessed more expertise in encouraging self-disclosure and in validating the emotional responses of the children. The transfer students were asked to think about and discuss how their new school compared to their old one. To help the children feel comfortable talking about feelings, project staff conducted a matching game that required the students to match different cartoon characters (portraying different emotions) with differently named feelings. The transfer students were asked next to write down and discuss two feelings they had about entering their new school. The orientation program also facilitated peer relationships. Each transfer child was paired up with a buddy who actively participated in the orientation session. The pairs were involved in various activities that encouraged cooperation and communication. The children interviewed each other about their favorite foods and television shows, and they engaged in the discussions about school personnel, rules and activities. In regard to school rules, five general good conduct rules (e.g., being on time) were discussed with the assistance of the peer leader, and these rules were written down in an orientation book. The pairs of children also considered how they could be good friends to one another. The 6th grader and a member of the project staff role played a situation involving how children could support each other during lunch time. The pairs of students were then asked to role play one of a variety of suggested scenes involving a cooperation task leading to the development of trust. At the end of the session, the children were asked to pretend they would get together the next week after school. They were instructed to interview one another and decide: (a) where they would meet, (b) what games they would play, (c) what they would have for snacks, (d) what they would do if it was raining, (e) who else might join them, and (f) what they would talk about. The children were given 5 minutes to discuss these issues and then reported back to the group about what occurred. This game was modeled after the "interview and report technique," which helps children learn about how one develops a friend (Haas, 1983). Play was chosen as a topic because: (a) children enjoy playing, (b) by interviewing each other about after school play, the children can see how compatible they are, (c) this activity might prompt them to actually get together after school, and (d) by reporting to the group, the children will communicate their interests not only to one another but also to others who may have similar interests. This creates the possibility of including more children in after-school friendship activities. In addition, the children were provided pictures to color which carried the messages, "Help Out New Students" and "Be a Friend." They were also given a page of riddles that they could complete together. Finally, to promote sharing, each transfer student was given a package of four cartoon pencils, and they were asked to give two of the pencils to their buddies.

2.

A PREVENTIVE INTERVENTION FOR HIGH RISK CHILDREN

59

Peer Guide Training In each school, at least one 6th-grade child served as a peer guide during the orientation session. With larger groups, there were additional peer guides, approximately one for every 10 children, five transfer children and five buddies. During meetings principals and teachers were informed of the 6th-grade guides' duties. Although the principal in each school selected the guides, the 5th- and 6th-grade teachers provided input into this discussion. Peer guides were selected who were extroverted, popular with peers, doing well academically, and who were former transfer students, knowledgeable about the school. Prior to the orientation session, the guides were taken from class and given a 30-minute training session. Discussion and role playing were used to help the guides become comfortable with their roles. During the training session, these students were told of their duties as peer leaders. The following responsibilities were given to them. • Help lead the discussion group during the orientation session, which will include some role playing. (They were given a copy of the script to be performed with one of the project personnel.) • Be a "stand in" buddy if the assigned buddy and the transfer child are not getting along during the orientation session. • Be a support for the transfer children. Try to be available if transfer children need help during the year. Then the students received orientation booklets and were informed of their specific duties during the session that included the following: • Introduce yourself to the children. Tell them how long you've been at this school. Also, mention something special about yourself, like your membership in a club. • Help introduce the buddies and transfer children. • Pass out the orientation booklets. The page guides covered in the booklet were: "Welcome," "Teacher's name and room number," "Important people at school," "Rules," "Activities," and "Being a good buddy." The 6th graders were then given general hints about how to lead the group. These hints were: • Try not to talk too much; the idea is to get the students to share. • If students are sad, mad or whatever, try to tell them their feelings are O.K.

60

JASON ET AL.

• If someone is very quiet you could say, "Jane, we didn't hear your ideas, would you tell us what your are thinking?" • If someone wants to talk ali the time you could say, "John, there is so much you want to tell us but why don't we give everyone a tum. If we still have time we'll come back to you." • If someone really does not want to talk, you might say, "Well Joey, it seems you just don't have anything to say right now and that's O.K., but when you do, we would like to hear from you."

The project staff then asked the children if they had any questions, and thanked the students for participating in the training session. Of course, leading this discussion group was a challenging job for the peer guides. A project staff person was at these discussion group sessions and helped the peer guides facilitate the discussion. To train the guides, staff used role-play, instructions, and discussions. Each school's guides were trained at their own school. All 3rd-, 4th- and 5th-grade transfer children atE schools were provided this orientation program. (High-risk children had not been identified when this orientation program occurred.) Buddy System

Each transfer child in E schools was paired up with a buddy in his/her classroom. All transfer students are in need of new friends in the new school, and so both high- and low-risk transfer students were given a buddy. The teachers had the responsibility of selecting and matching the buddies. During teacher meetings, they received a written guide explaining the buddy system along with a form that could be used to write down the names of the transfer children and their respective buddies. The guidelines for selection of buddies included choosing children who were outgoing, enthusiastic, responsive to others' needs, and doing well academically. These characteristics have been found among children who are adept at gaining friendship (Jalongo, 1983). The teachers selected more buddies than necessary in case a buddy was absent on orientation day, or a buddy and a transfer child did not get along. The teachers first asked the buddies if they wanted to be a buddy. Almost ali children readily agreed to serve in this role. The teachers matched each transfer child with a same sex buddy, because children at this age tend to play with peers of their own sex (Gardner, 1982). The teachers provided buddies information about their new roles. The buddies were told, by teachers, that entering a new school is often difficult for transfer students because they have left their friends at the previous school. Teachers emphasized that developing and maintaining friendships at the new school is very important for transfer students. They asked the buddies to introduce transfer children to peers, help the transfer children become involved in outside play

2.

A PREVENTIVE INTERVENTION FOR HIGH RISK CHILDREN

61

activities, and ask the transfer children if any help was needed in academic or other areas. If the buddy and transfer child were not compatible, for whatever reason, the teacher had the option of assigning the transfer child another buddy. As mentioned earlier, the buddy was involved in the orientation session, which was a nonthreatening environment where the children could get to know one another. At the orientation session, the 6th-grade peer leader and project personnel introduced the pairs. If either the transfer child or the buddy were not happy with their match, the project personnel matched one of the extra buddies with the transfer child. This happened infrequently, but nevertheless provisions were made to prevent inappropriate pairing. During the orientation session, the pairs discussed the reasons for rules, matched faces with feelings, and devised ways in which they could be good buddies. Furman and Robbins ( 1985) stated that one of the best ways to promote friendship between two children is to have them participate together in a series of cooperative tasks. This is the rationale for introducing peers in the previously described orientation program. Bierman and Furman (1984) found that to maximize the probability that peers will recognize and accept the new competencies, environmental manipulations (e.g., peer involvement) should supplement skill training (Bierman, Miller, & Stabb, 1987). Involving peers and transfer students in the orientation program facilitated the transfer students' opportunities to gain easier entry into the peer culture. Pairing transfer students and peers was not meant solely to increase the frequency of peer interactions, without regard for the quality or competence of social approaches (Coie, 1985; Dodge, 1983). Provision of a buddy provided the transfers with a friend to help them develop a new social network. Tutoring

Recent research indicates that tutoring might be one of the more powerful strategies for aiding academically at-risk children. Coie and Krehbiel (1984) selected socially rejected children having serious academic problems, and a tutoring intervention not only brought about improvements in academic achievement, but their social preference scores moved form those extremely rejected to those of average status children. As mentioned previously, staff administered a standardized achievement test (Wide Range Achievement Test, WRAT) to help identify at-risk children. Identified high-risk students then took a criterion-referenced measure to assess the youngsters' skills in reading comprehension, decoding, spelling, and math. There is a hierarchy of learning in reading and math. If a youngster misses earlier steps in reading or math, given these subjects' hierarchy of learning, a child may develop a weak foundation for learning more advanced skills, and thus deficits can continue to be debilitating. Ladd ( 1985) suggested that rather than teaching children all possible skills,

62

JASON ET AL.

we should concentrate on the children's individual deficits, and assess those deficits throughout the intervention. Dodge (1985) also recommends the need for individualized interventions that attend to the specific problems of the target children. The criterion-referenced tests enabled investigators to identify precise skill areas needing remediation. These test results, which pinpoint students' prioritized list of instructional needs, were used in designing specific tutoring intervention. These tests covered phonics, math, spelling, and reading. Although the WRAT test identified children with academic weaknesses, the criterion-referenced test helped determine which specific academic skills needed strengthening. As an example, the child might not be able to divide two digit numbers into two digit numbers, and the test would indicate this deficit. The actual intervention involved individual, 40-minute tutoring sessions twice a week. Since tutoring occurred during the school day, the teachers were consulted in choosing a time that was least disruptive to the children's daily activities. The sessions took place in an unoccupied classroom, school library, or teacher's lounge. Tutoring consisted of teaching new skills, practicing recently acquired skills until mastered, and reviewing previously learned skills. Several minutes a day were also spent discussing any subjects or problems brought up by a student. Materials being used in the classroom, focused on this type of division problem, were employed during the tutoring session. Teachers were also instrumental in helping the tutors select appropriate materials to be used for tutoring. The tutors were DePaul undergraduates who had past experience working with elementary school age children and enjoyed working with children, and who had at least a B average at DePaul University, and a generally extroverted and warm personality. There was one tutor for each of the 10 experimental schools. Each tutor worked 20 hours a week and taught an average of six transfer students. Tutors were instructed on how to use reinforcement, introduce material in an interesting way, and pace the sequence of material. Project staff reviewed strategies for helping a youngster relax and feel comfortable in this new situation. Report-writing and making progress charts wee covered during the training. techniques for working with bored youngsters, high activity level children, as well as noncompliant pupils, were also reviewed. If problems such as these did arise, ongoing consultation with the Project Director was provided to the tutor. At least once a week, the supervisor would go to a school and directly observe a tutor's session. The tutor would receive feedback from the supervisor on the quantity and quality of the reinforcement, the effectiveness of the presentation, the tutor's ability to detect and correct errors. The tutors met individually with a supervisor once a week to discuss their target youngsters' progress. In addition, all of the tutors and supervisors met together once every 2 weeks to exchange ideas about working with teachers and students. The tutoring interventions were designed to help remediate the specific academic skill deficits of the high-risk transfer students. The criterion-referenced

2.

A PREVENTIVE INTERVENTION FOR HIGH RISK CHILDREN

63

tests had identified these weaknesses. During tutoring, a student had the opportunity to repeat trials to strengthen areas of weakness before moving on to progressively more difficult skills. Staff had a progress chart for each child so they could monitor and reinforce specific gains in subject areas. Data were collected during every tutoring session. Whenever possible, tutors used materials from the classroom to strengthen the links between what occurred in the tutoring sessions and instruction in the classroom. One of the main components of the tutoring was the Direct Instruction technique. Some of the main tenets of Direct Instruction include: carefully organized materials, high levels of student involvement, rapid rate of teacher presentation, extensive practice of material, immediate corrective feedback, teaching to criterion level, and positive reinforcement (Rosenshine & Berliner, 1978). One technique of Direct Instruction tutoring that tutors used frequently was the ModelLead-Test method. When the material is presented, the tutor demonstrates the correct response first (model), then the tutor and the student say the response together (lead), and then the student says the correct response alone (test). For example, suppose the tutor was correcting a word the student mispronounced. The sequence might look like this: Tutor: The word is "extreme." Our turn. Tutor and student: Extreme. Tutor: Your turn. Student: Extreme. Tutor: All right! Good reading!

The Model-Lead-Test Method can be used with any type of material. Four academic areas were covered during the tutoring sessions. The math curriculum has four learning hierarchies. A student can work on just one or several areas at a time. Three factors are considered when deciding the number of areas on which to work: (a) math deficits of the student, (b) classroom teacher's wishes, and (c) student's deficits in other areas. Students continued to work on a particular step on the hierarchy until they mastered it. The criteria for mastery was the student being able to solve five out of five problems correctly at a reasonable rate. A hierarchy was also used for the phonics curriculum. Decoding rules were taught in conjunction with a list of words that fit the rule. At the beginning of the year, students received a placement test hierarchy. Students progressed to higher levels by successfully reading the 10 words that corresponded to the rule being taught. Words and rules were periodically reviewed to insure that the skills being learned were retained. The spelling curriculum consisted of words taken directly from the weekly spelling units that were covered in the child's classroom. During the first session of the week, the students were tested on the classroom's spelling words. Missed

64

JASON ET AL.

words were practiced orally and in written form until they were mastered. During the second session of the week, the missed words were retested. Missed words were again practiced until they were perfected. Repeated practice improved performance on the classroom's weekly spelling test. Materials used in teaching reading were taken directly from the classroom. The child read orally, while the tutor corrected mispronounced words. The tutor also provided feedback on reading rate and asked comprehension questions. To measure student progress, tutors gave a 100-word assessment each week. This assessment was also used to measure the suitability of the classroom reader for the student. The assessment was given in the following manner. The tutor marked off 100 words of the classroom reader and instructed the child to read the words, while the tutor timed the student and recorded reading errors. After reading the passage, the student was asked five comprehension questions. The type of questions included factual, inferential, and sequential. By using this assessment, the child's rate, accuracy, and comprehension in reading could be determined (ldol-Maestas, Ritter, & Lloyd, 1983). Implementation Issues A variety of implementation and policy issues arose in the development of this prevention intervention, including: developing the curriculum, working with Hispanic families, negotiating with the principals, establishing good relationships with the teachers, working with the children, and supervising the tutors. Each of these issues is reviewed next. Curriculum. The first issue of implementation concerned which subject areas to focus on. After much consideration, we decided to concentrate on phonics, reading, spelling and math. There were many other areas that we could have covered, such as legibility of writing. Although legibility was not a focus, it still could have been covered in one of the four core areas. With time limited for tutoring, we needed to focus on academic areas critical for success in the classroom. As we developed the curriculum, we needed to decide whether to use materials from the classroom or our own materials from outside of the classroom. To the extent that we used materials from the classroom, teachers would react more positively, and there might be even a better chance of new skills generalizing to the classroom. On the other hand, if we used our own materials, we could be more effective in individualizing instruction and in monitoring our progress. However, by solely relying on our materials, we felt that the teachers would be less receptive to what we were trying to accomplish. Our compromise solution involved using some materials from the classroom and some that we constructed. For example, we used the spelling words from the classroom for drill; and for reading, we also used the reader the child was using

2.

A PREVENTIVE INTERVENTION FOR HIGH RISK CHILDREN

65

in the classroom. By using the child's own reader, we could support the work of the classroom teacher. In two other areas, phonics and math, we used materials from our own structured program. A 5th-grade child might be at the 3rd-grade level in math and phonics. Without special tutoring involving curricular material that was not presently being used in the classroom, the child would not have the opportunity to make up these deficits.

Hispanic Families. One implementation issue involved working with Hispanic children and parents who knew little or no English. When interviewing the parents, project staff communicated in Spanish, using a bilingual staff member or family member to interpret. In working with the children, however, we rarely used Spanish for two reasons. First, any tutoring done in Spanish would not support work done in the classroom because the participating school teachers did not speak Spanish. Second, some of the Hispanic parents transferred their children out of bilingual public school programs so their children would learn more English. For those children who knew little English, we adjusted our program to spend a greater proportion of the tutoring time on reading comprehension and decoding skills. Principals. The parochial school system with which we work is characterized by a decentralized authority structure. Because the principals in each school are autonomous and influential, we thought it important to recruit the principals' support for our project in each program school. Their participation was crucial for two reasons because the principals had the authority to grant or rescind permission to conduct interventions in their schools, and their support showed an important approval of the intervention, thus easing our entry into the school system. To help build a solid relationship between project staff and principals, we encouraged a spirit of openness. Specifically, we invited principals to observe actual tutoring sessions. We felt this invitation sent a constructive message to the principals; namely, that we were proud and confident that the intervention was a viable and effective adjunct to the instruction the children were receiving in the classroom. It also communicated that we were eager to receive their feedback about what we were doing. We believe this partnership was essential, as we needed to communicate a genuine respect for their authority and expertise. Several times during the first year, our relationship with a principal helped protect the actual project. For example, on a number of occasions, teachers would request that the project accept transfer students that had entered a school well into the academic year. Due to the lack of personnel resources, we could not accept these children into the program. The principals overwhelmingly supported this decision and helped the teachers understand that our project had limited resources. Without such agreement our relationship with the teachers might have been strained.

66

JASON ET AL.

We spent much time during our first year communicating with principals under the mistaken assumption that the topics of our communique would reach teachers via the principals. We discovered that principals were often too busy to transmit the messages and information consistently. Thus, we realized that we needed both to maintain our strong relationship with the principals and concurrently develop separate relations with the teachers. One example of positive interaction with a principal is the case of a transfer child who received excellent grades in his old school but was 1 to 2 years behind the skill level of his present class. The parents were angry because the transfer child was now failing basic courses in the new school. The principal asked us to intercede and help the parents understand in which areas the child was behind. We were able to meet with the parents and help them accept the child's academic problems and his need for extra attention. Another principal mentioned to us that children felt special in our program. She said, "Sometimes transfers get lost between the cracks." Positive reactions from all the principals validated the goals and accomplishments of our program.

Teachers. One of the major concerns that surfaced at the end of the first year was that teachers did not feel well informed about what actually went on during the tutoring sessions. To rectify this problem, we designed systematic procedures to ensure ongoing and consistent communication between the project and teachers. First, we prompted tutors to be aware of the importance of speaking with teachers. We encouraged tutors to talk regularly with teachers, at least once every 2 weeks. Because meeting and talking was not always feasible for teachers, we developed a notebook system. Notebooks, usually placed in the teachers' mailboxes, allowed both the tutor and the teachers to exchange written comments about a student's strengths and weaknesses as evidenced in both the classroom and the tutoring sessions. Tutors were required to write comments in the notebook at least once a week but preferably after each session. Lastly, we developed a Daily Summary Sheet. This sheet, which was completed after every tutoring session, included information on reading accuracy, rate, and comprehension, as well as which story was read. It also reported the percent correct in spelling, which words were missed, and the phonics and math levels achieved. At times, several teachers made derogatory comments about our children in front of the children and tutors. For example, one teacher, in the presence of the tutored child, said to the tutor that the child was not too smart. We did not want to criticize the teacher for saying this, but we also did not approve of such remarks. We rarely confronted teachers on these issues, but attempted to highlight the children's positive accomplishments, and we often said to teachers that focused, behavior-based statements were appropriate in giving children feedback. Several teachers objected to the feedback we give to parents because we focused just on accomplishments their children were experiencing in our tutoring sessions, and the teachers wanted us to mention the problems the children were

2.

A PREVENTIVE INTERVENTION FOR HIGH RISK CHILDREN

67

experiencing in their classrooms. In future communications with the parents, we first sought teacher approval, and by providing more balanced summaries that included classroom reports, the teachers' objections faded. Occasionally, we found the reading books used in the classrooms were too difficult, and we needed to find a nonthreatening way to inform teachers that their materials were not a good fit with the children's ability levels. Of course, we were advocates of more individualized instruction, and by providing teachers with information about what we were doing, we began the process of helping teachers select appropriate materials for our target youngsters. As teachers began to see us as experts, and as effective with their children, they were more willing to accept our recommendations. Tutors. In hiring tutors, we thought it important to ask somewhat personal questions while at the same time not discriminating against the prospective tutors if problems they experienced in the past were under control. For example, one of our hired tutors, unknown to the senior project staff, was on medication for bipolar disorder, and this tutor was eventually hospitalized. This episode forced our project staff to reconsider what we needed to ask tutors before hiring them. Having backup tutors was essential, if a tutor was sick. In gaining the trust of a school system, we foremost needed to be reliable. It was just as important to not have too many different tutors coming into the system. Most schools do not want strangers in their settings. If a substitute was to be used, it was best to always use the same person. One case that was difficult for us to handle was the instance of a child who asked a tutor, "Is it o.k. if I call you after school?" The child was reaching out for support, but we could not allow these types of calls without the teacher's permission or without first determining what was occurring at home. In this situation, we resolved to arrange more time for the child to talk with the tutor during school hours. One potential problem in our work was the frequency of reinforcement during tutoring. In classrooms, amount of reinforcement was not high; however, in tutoring the amount of reinforcement provided by tutors was very high. During tutoring sessions, we tried, over time, to reduce the reinforcement density so that the children would be willing to work for less reinforcement than occurs in more naturalistic settings like the classroom. During the tutoring sessions, children were seldom disruptive. Children's behavior in the classroom, however, was sometimes more disruptive and uncontrollable. Even though children seldom "acted-out" in tutoring sessions, we frequently strengthened self-control skills (e.g., attending to the materials, reinforcing oneself for on-task behavior) because these types of behaviors could generalize into the classroom. Children. During each of the tutoring sessions, tutors spent about 5 minutes informally talking with the children. These conversations helped develop the

68

JASON ET AL.

tutor-child relationships. In addition, this special time talking could be used as a reward for hard work during the session. Many of the project children did have special needs and used this informal time to talk about personal issues. Several children whose parents just separated were slightly depressed, and they needed time to ventilate their feelings during the session. Because one child's parents had a drinking problem, it was understandable that the child was having difficulty focusing on school work. Another child could not concentrate because she was thinking about her foster father, who had recently hit her mother. One child spent half of her time with her grandmother, who only spoke Spanish, and the other half of the time with her mother. Another child wrote a note to the tutor saying that he was fed up with school, because he had been bullied by other children. As we become the children's advocates, we will need to devote some time to dealing with these school and home factors that influence the children's abilities to learn. One of our children was not allowed access to the library, and we intervened so that this restriction was eliminated. Once the child had access to the library, the child was more satisfied with school. Here again we envision our role broadening to include not just tutoring but also child advocacy. Working with Parents: Developing More Comprehensive Interventions

During our first year, every 6 weeks we sent parents a progress note that was followed by a telephone call to give them an idea of what we were doing. The children's spelling words were sent home to some parents, who sometimes became actively involved in helping their children. End-of-year results from the first year of the project suggested that parents were important in facilitating their children's school-related adjustment. Data to support this notion were gathered from the Parent-Tutor Assessment (PTA), a measure of parents' tutoring behavior. Based on the PTA and student outcome measures, we found that children who were less supported at home had significantly poorer end-of-year grades in reading, math, and spelling. Moreover, results from a comparison of students with high and low parental support revealed significant differences on sociometric measures of social withdrawal and likability. In comparing the two groups using the Achenbach Child Behavior Checklist: Teacher Report Form, the less parentally supported group had more difficulties in adjustment. Notably, the less supported group of the transfer students demonstrated significantly greater anxiety and nervousness, inattentiveness, social withdrawal, aggression, and unpopularity as compared to the more highly supported high-risk transfers. Again, these findings suggest that parents probably are important contributors to the social and academic adjustment of their children in a new school. In response to the PTA, it is striking that 63% of the high-risk parents an-

2.

A PREVENTIVE INTERVENTION FOR HIGH RISK CHILDREN

69

swered that they were not effective tutors for their children. We were encouraged that 80% of the parents said that they would participate in parent-tutor training if the opportunity was offered to them. These results suggest that most children whom we serve have parents who are willing to assume an educative role but perceive themselves as lacking the necessary skills. During the second year of the project, we included some parents in pilot work through some training in tutoring skills, ongoing phone contact, and monitoring of home progress by the staff tutors. Individual parent-tutor training began during the sixth week of school, coinciding with the beginning of in-school tutoring for experimental students. Project staff trained parents in direct instruction teaching techniques similar to the ones used by the staff. A brief manual explained the corrective feedback and reinforcement procedures. Parents practiced the procedures, and staff provided coaching and support. A brief manual was used to explain the corrective feedback and reinforcement procedures. Project staff also suggested to parents what they could reasonably expect of their children's progress. Also, staff discussed guidelines for immediate reinforcement and contingency management. Parents and children together were encouraged to negotiate nightly and weekly goals to complete homework that were both realistic and flexible. Homework materials primarily were assigned by the classroom teacher in the basic subjects of math, spelling, and language arts. In addition to classroom homework, tutors assigned short stories from the SRA reading laboratory (Parker & Scannel, 1982). The staff tutor matched the difficulty of the stories with the student's instructional level of reading. Students were encouraged to read the stories orally to their parents, and, with supervision, to complete the brief comprehension and phonic activities. Staff familiarized parents and children with a plan for ongoing communication. During the year, parents reported the amount of time spent in tutoring and provided information about completing the SRA card. The students were responsible for brining a homework calendar, signed by the parent, along with the SRA cards, to and from school in a special folder on the days that the staff tutor was present. All classroom assignments were brought to the regular teacher. Staff checked the SRA answer sheets to assess the accuracy of the assignment. Tutors often sent weekly in-school tutoring progress reports to the parents via the special folders. Additionally, tutors phoned parents on a biweekly basis to consult about parent-tutoring strategies and children's progress. During the final part of the training session, the staff presented a behavioral contract which outlined the expectations of the parent-tutoring session. All parents received copies. One of our remaining goals concerns how to engage uncooperative or uninterested parents and children. Overall, results of this pilot work are promising. We tried to foresee major obstacles to gain as much parental cooperation as possible. For example, a number of Hispanic parents who are not fluent in English were encouraged to tutor children in areas such as math, which are not so dependent on the use of

70

JASON ET AL.

English. We devised other home supports by, for example, including as tutors the child's English-speaking older siblings, relatives, and sometimes neighbors. Also, many of the parents are single or have full-time employment that limits their time and resources. Emphasizing the importance of parental support, we tried to help these parents find time during their week to tutor their children. For the intervention's third year, we are involving parents so as to promote further the long-term adjustment of high-risk transfer students. During the third and fourth weeks of the school year, program parents were invited to the schools for meetings that include the principal, teacher, and staff tutor. The purpose of these meetings was to outline the objectives and expectations of the DePaul program. Importantly, these meetings not only inform parents of educational goals but secure their cooperation in sharing responsibilities with the school to enhance their children's adjustment. From our experience, one training session was not enough to ensure mastery of parents' tutoring skills. Thus, during the third year, we are including a home visit in November as part of the parent training program. The visits are conducted by pairs of the project staff, preferably on weekends or weekday afternoons. Such visits promote a more personal and trusting communication between parents, tutors, and children. During the visits, parents receive follow-up training and coaching in tutoring skills. Direct observation of parents who are tutoring their children provide feedback for improving our approach in helping parents become more effective helpers for their children. Additional training either in the home or at school is offered to those parents who may require more guided tutoring practice and feedback. In addition to communicating through phone contact and notes, parents, tutors, and teachers meet at the end of each grading period. The transfer children's progress in academic, social, and behavioral areas is discussed, along with mutually agreed upon plans for improvement. These regular in-school meetings keep everyone fully informed of the student's progress. With teachers, tutors, and parents committed and active in the preventive effort, we hope to have a more comprehensive, potent, and effective intervention for our at-risk children. CONCLUSION

A preventive intervention using tutoring for high-risk children raises many complicated issues. Our project focused on the academic needs of at-risk children who transfer schools. There is no question that the project will need to take into account other factors such as their school and home environments, and their peers. As our intervention broadens and becomes more comprehensive, we hope that our tutors will gradually assume the role of an advocate, coordinating the many available resources to help the children academically, socially, and emotionally.

2.

A PREVENTIVE INTERVENTION FOR HIGH RISK CHILDREN

71

A recent study by Jason, Betts, Johnson, Smith, Krueckeberg, and Cradock (1989) evaluated the first year of this program. Consumers of the program, including the target children, parents, teachers, and principals were overwhelmingly positive in their evaluations, thus providing excellent social validity data on the program's need and usefulness. On the more standardized measures, only in the area of achievement were significant program effects found. Of most interest, it appears that by the posttest, the intervention had helped some children move out of the deviant scoring category. Figure 2.3 shows that at the end of the program, only 58% of program children (E) were at high-risk on the post Wide Range Achievement Test, whereas, for nonprogram transfers (C) 72% were at risk. For a group of nontransfer children (CNT) identified as having academic and social problems at the beginning of the year, 72% were at risk at posttest. Some children who transfer into a new school are high-risk candidates for later academic problems, and our experience with this project somewhat confirms this proposition. The orientation plus tutoring intervention was intended to provide transfer children social support and academic competencies to deal with their transition into a new school better. Many of the tutored children were several grade levels below their peers, and almost all were confronted with a curriculum that was different from their previous school. A transactional model (Sameroff, 1987) demonstrates how such early difficulties can lead to even more problems, particularly with children who have fewer resources (e.g., a high number of life stressors, low SES). Children academically behind may opt to catch up on their own, to fall further behind, or to maintain their relative standing. Those children within the last two categories are at-risk for not meeting a teacher's expectations, and lowered expectations can negatively influence a child's adjustment. If peers also reject the transfer youngster because of lowered academic abilities, the child is even more at-risk for subsequent school and social difficulties (Parker & Asher, 1987). In addition, some transfer low achievers who are frustrated due to repeatedly failing in school might be more likely to aggress against peers (Bursuck & Asher, 1986). School transitions might represent an

78

FIG. 2.3. Effects of low-achievement on behavior, social status, and later adjustment.

£

C

Remained High-Risk on Post WRA T Scores

CNT

72

JASON ET AL.

ideal site for mounting tutoring programs designed to interrupt unfortunate reverberating transactions. In summary, the orientation plus tutoring intervention represents a promising approach for helping transfer youngsters adjust to the demands of a new setting. Interventions need to be individualized to meet the unique needs of each child. Preventive interventions with transfer children represent a rich area for investigators interested in prevention.

ACKNOWLEDGMENTS The authors would like to thank the many school principals and teachers who helped make this project possible. In addition, we are particularly indebted to Louise Perone and Kathryn Karsh for helping develop this intervention and securing cooperation from the school personnel. In addition, we appreciate the continued support of Brother Don Houde, the Director of Administrative Affairs of the Office of Catholic Education. We also thank Scott Oliver and Brenda Greiner for their help in data analysis and in computerizing our system. The dedicated and committed tutors also deserve a special recognition for helping make this project possible. Funding for this project was provided by the National Institute of Mental Health (Grant number MH 40851 ). Requests for further information should be sent to Leonard A. Jason, Department of Psychology, DePaul University, 2219 N. Kenmore Ave., Chicago, IL 60614.

REFERENCES Achenbach, T. M. (1974). Developmental psychopathology. New York: The Ronald Press. Ausubel, D.P., Sullivan, E. V., & Ives, S. W. (1980). Theory and problems of child development (3rd Ed.). New York: Greene & Stratton. Bandura, A. (1978). On paradigms and recycled ideologies. Cognitive Therapy and Research, 2, 79-103. Barrett, C. L., & Noble, H. (1973). Mother's anxieties versus the effects of long distance moves on children. Journal of Marriage and the Family, 35, 181-188. Bensen, G. P., Haycraft, J. R., Steyaert, J.P., & Weigel, D. J. (1979). Mobility in sixth graders as related to achievement, adjustment and socioeconomic status. Psychology in the Schools, 16, 444-447. Bierman, K. L., & Furman, W. (1984). The effects of social skills training and peer involvement on the social adjustment of preadolescents. Child Development, 55, 151-162. Bierman, K. L., Miller, C. M., & Stabb, S. (1987). Improving the social behaviors and peer acceptance of rejected boys: Effects of social skill training with instructions and prohibitions. Journal of Consulting and Clinical Psychology, 55, 194-200. Bogal, G. A., Jones. J. W., & Jason, L. A. (1980). School transitions: Preventive intervention following an elementary school closing. Journal of Community Psychology, 8, 343-352. Brockman, M.A., & Reeves, A. W. (1967). Relationship between transiency and test achievement. Alberta Journal of Educational Research, 13, 319-330.

2.

A PREVENTIVE INTERVENTION FOR HIGH RISK CHILDREN

73

Bulletin. (1983, November). High school dropouts: Descriptive information from high school and beyond. Office of the Assistant Secretary for Educational Research and Improvement. National Center for Education Statistics. Dept. of Education. Burke, R. J., & Weir, T. (1978). Sex differences in adolescent life stress, social support and well being. Journal of Psychology, 98, 277-288. Bursuck, W. D., & Asher, S. R. (1986). The relationship between social competence and achievement in elementary school children. Journal of Clinical Child Psychology, 15, 41-49. Caplan, G. (I 965). Opportunities for school psychologists in the primary prevention of mental disorders in children. In W. Lambert (Ed.), The protection and promotion of mental health in the schools (pp. 9-22). Mental Health Monograph No. 5, U.S. Printing Office. Cauce, A. M., Feiner, R. D., & Primavera, J. (1982). Social support in high-risk adolescents: Structural components and adaptive impact. American Journal of Community Psychology, 10, 417-428. Coddington, R. D. (1972a). The significance of life events as etiologic factors in the disease of children I. A survey of professional workers. Journal of Psychosomatic Research, 16, 17-18. Coddington, R. D. (1972b). The significance of life events as etiologic factors in the disease of children II. A study of a normal population. Journal of Psychosomatic Research, 16, 205-213. Coie, J. D. (1985). Fitting social skills intervention to the target group. In B. Schneider, K. H. Rubin & J. E. Ledingham (Eds.), Children's peer relations. issues in assessment and interventions (pp. 141-156). New York: Springer-Verlag. Coie, J. D., & Krehbiel, G. (1984). Effects of academic tutoring on the social status of lowachieving socially rejected children. Child Development, 55, 1465-1478. Collins, J. M., & Coulter, F. (1974). Effects of geographic movement on the social and academic development of children of army personnel. Australian and New Zealand Journal of Sociology, 10, 222-223. Comille, T. A., Bayer, A. E., & Smyth, C. K. (1983). Schools and newcomers: A national survey of innovative programs. Personnel and Guidance Journal, 62, 229-236. Cowen, E. L., Lotyczewski, B.S., & Weissberg, R. P. (1984). Risk and resources indicators and their relationship to young children's school adjustment. American Journal of Community Psychology, 12, 353-367. Cradock, C., Cotler, S., & Jason, L. A. (1978). Primary prevention: Immunization of children for speech anxiety. Cognitive Therapy and Research, 2, 389-396. Cramer, W., & Dorsey, S. (1979). Are movers losers? The Elementary School Journal, 70, 387390. Dodge, K. A. (1983). Behavior antecedents of peer social status. Child Development, 54, 13861399. Dodge, K. A. (1985). Facets of social interaction and the assessment of social competence in children. In B. H. Schneider, K. H. Rubin, & L. E. Ledingham (Eds.), Children's peer relations: Issues in assessment and intervention (pp. 3-22). New York: Springer-Verlag. Donohue, K. C., & Gullotta, T. P. (1981, Nov.-Dec.). FIT: What corporations can do to ease relocation stress. Mobility, 2, 53-57. Douglas, J. A., & Jason, L. A. (1979). Transitions: Utilizing behavioral technology to facilitate entry into a school and an occupation. Crisis Intervention, 10, 68-79. Douvan, E., & Adelson, J. (1966). The adolescent experience. New York: Wiley. Durlak, J. A., & Jason, L.A. (1984). Preventive programs for school-aged children and adolescents. In M. C. Roberts, & L. Peterson (Eds.), Prevention of problems in childhood: Psychologi· cal research and applications (pp. 103-132). New York: Wiley. Elias, M. J., Gara, M., & Ubriaco, M. (1985). Sources of stress and support in children's transition to middle school: An empirical analysis. Journal of Clinical Child Psychology, 14, ll2-ll8. Elias, M. J., Gara, M., Ubriaco, M., Rothbaum, P. A., Clabby, J. F., & Schuyler, T. (1986). Impact of a preventive social problem solving intervention on children's coping with middle-school stressors. American Journal of Community Psychology, 14, 259-275.

74

JASON ET AL.

Feiner, R. D., Farber, S. S., & Primavera, J. (1983). Transitions and stressful life events: A model for primary prevention. In R. D. Feiner, L. A. Jason, J. N. Moritsugu, & Farber, S.S. (Eds.), Preventive psychology: Theory, research and practice (pp. 199-215). New York: Pergamon. Feiner, R. D., Ginter, M., & Primavera, J. (1982). Primary prevention during school transitions: Social support and environmental structure. American Journal of Community Psychology, 10, 227-290. Feiner, R. D., Primavera, J., & Cauce, A. M. (1981). The impact of school transitions: A focus for preventive efforts. American Journal of Community Psychology, 9, 449-459. Freedman, R. (1964). Cityward migration, urban ecology and social theory. In E. W. Burgess, & D. J. Bogue (Eds.), Contributions to urban sociology (pp. 178-200). Chicago: University of Chicago Press. Furman, W., & Robbins, P. (1985). What's the point? Issues in the selection of treatment objectives. In B. H. Schneider, K. H. Rubin, & J. E. Ledingham (Eds.), Children's peer relations: Issues in assessment and intervention (pp. 41-54). New York: Springer-Verlag. Gardner, H. (1982). Developmental psychology (2nd ed.). Boston: Little Brown & Company. Gilliland, C. H. (1958). The relation of pupil mobility to achievement in elementary school. Unpublished doctoral dissertation, Colorado State University. Glidewell, J. C., Kantor, M. B., Smith, L. M., & Stringer, L. A. (1965). Social structure and socialization in the elementary school classroom. A report of a workgroup to the committee on social structure and socialization of the Social Science Research Council. Clayton, MS: The St. Louis County Health Department. Goebel, B. L. (1978). Mobility and education. American Secondary Education, 8, 11-16. Goebel, B. L. (1981 ). Mobile children: An American Tragedy? Psychological Reports, 48, 15-18. Greene, J. E., & Daughtry, S. L. (1961). Factors associated with school mobility. Journal of Educational Sociology, 35, 36-40. Haas, M. E. (1983). What are the characteristics of a friends? A resource unit for elementary social studies instruction. Arkansas Technical University (ED 242 619). Hess, F. (1987). Bending the twig: The elementary years and dropout rate in Chicago Public Schools. Chicago Panel on Public School Policy and Finance, Chicago. Hirsch, B. J. (1985). Social networks and the ecology of human development: Theory, research and practice. In I. G. Sarason & B. R. Sarason (Eds.), Social support: Theory, research and application (pp. 117-136). The Hague: Martinus Nijhof. Hirschowitz, R. B. (1976). Groups to help people cope with the tasks of transition. In R. G. Hirschowitz & B. Levy (Eds.), The changing mental health scene (pp. 171-188). New York: Spectrum press. Holland, J. V., Kaplan, D. M., & Davis, S. D. (1974). Interschool transfers: A mental health challenge. Journal of School Health, 44, 74-79. Hollingshead, A. B. (1975). Four Factor Index of Social Status. Unpublished manuscript, Yale University, Department of Sociology, New Haven. Hurlock, E. B. (1964). Child Development (4th ed.). New York: McGraw-Hill. Idol-Maestas, L., Ritter, S., & Lloyd, S. (1983). A model for direct, data-based reading instruction. Journal of Special Education Technology, 6, 61-77. Jalongo, M. R. ( 1983). Promoting peer acceptance in the newly immigrated child. Childhood Education, 60, II7-124. Jarvis, 0. T., & Wooton, L. R. (1969). The transitional elementary school and its curriculum. Dubuque, lA: William C. Brown. Jason, L. A., Betts, D., Johnson, J., Smith, S., Krueckeberg, S., & Cradock, M. (1989). An evaluation of an orientation plus tutoring school based prevention program. Professional School Psychology, 4, 273-284. Jason, L. A., & Bogal, G. A. (1983). Evaluating a preventive orientation program. Journal of Social Service Research, 7, 39-49.

2.

A PREVENTIVE INTERVENTION FOR HIGH RISK CHILDREN

75

Jason, L. A., & Burrows, B. (1983). Transition training for high school seniors. Cognitive Therapy and Research, 7, 79-91. Jason, L.A., Christensen, H., & Carl, K. (1982). Programmed versus naturalistic approaches in enhancing study-related behavior. Journal of Clinical Child Psychology, 11, 249-254. Jason, L. A., Ferone, L., & Soucy, G. (1979). Teaching peer-tutoring behaviors in first and third grade classrooms. Psychology in the Schools, 16, 261-269. Jason, L. A., Frasure, S., & Ferone, L. (1981). Establishing supervising behaviors in eighth graders and peer-tutoring behaviors in first graders. Child Study Journal, 4, 201-219. Jason, L. A., Thompson, D., & Rose, T. (1986). Methodological issues in prevention. In L. Michelson & B. Edelstein (Eds.), Handbook of prevention (pp. l-19) New York: Plenum. Jastak, S., & Wilkinson, G. S. (1984). Wide Range Achievement Test-Revised. Wilmington, DE: Jastak Associates. Kantor, M. B. (1965). Social consequences of residential and social mobility for the adjustment of children. In M. B. Kantor (Ed.), Mobility and mental health (pp. 86-122). Springfield, IL: Charles C. Thomas. Kistner, J. A., & Torgessen, J. K. (1987). Motivational and cognitive aspects of learning abilities. In B. B. Lahey & A. E. Kazdin (Eds. ), Advances in clinical child psychology (VoL I 0, pp. 289333). New York: Plenum Press. Klein, D. C., & Ross, A. (1965). Kindergarten entry: A study of role transition. In H. J. Parad (Eds.). Crisis intervention: Selected readings (pp. 140-148). New York: Family Service Association. Kohl berg, L., Ricks, D. , & Snarey, J. ( 1984). Childhood development as a predictor of adaptation in adulthood. Genetic Psychology Monographs, 110, 91-172. Kroger, J. E. (1980). Residential mobility and relationships during adolescence. New Zealand Journal of Educational Studies, 15, 69-80. Kyle, C. L., Lane, J., Sween, J., & Triana, A. (1986). We have a choice: Students at risk for leaving public school. Chicago Area Studies Center and the Center for Research on Hispanics, DePaul University, Chicago. Lacey, C., & Blane, D. (1979). Geographic mobility and school attainment-the confounding variables. Education Research, 21, 200-206. Ladd, G. W. (1985). Documenting the effects of social skill training with children: Process and outcome assessment. In B. H. Schneider, K. H. Rubin, & J. E. Ledingham (Eds.), Children's peer relations. Issues in assessment and intervention (pp. 243-269) New York: Springer-Verlag. Levine, M. (1966). Residential change and school adjustment Community Mental Health Journal, 2, 61-69. Levine, M.D. (1983). Middle childhood. In M.D. Levine, W. B. Carey, A. C. Crocker, & R. T. Gross (Eds.), Developmental-behavioral pediatrics (pp. 108-132). Philadelphia: W. B. Saunders. Levine, M., Wesolowski, J. C., & Corbett, F. J. (1966). Pupil turnover and academic performance in an inner-city elementary schooL Psychology in the Schools, 3, 153-158. Long, L. (1975). Does migration interfere with children's progress in school? Sociology of Education, 48, 369-381. Lorion, R. P., Work, W. C., & Hightower, A. D. (1984). A school-based multilevel preventive intervention: Issues in program development and evaluation. The Personnel and Guidance Journal, 62, 479-484. Millet, L., Pon, J., Guibaud, J. M., & Auriol, G. (1980). "Demenager": Annales Medico-Psychologiques, 138, 212-221. Moore, H. R. (1966). Geographic mobility and performance in high schooL Part II. Journal of Secondary Education, 41, 350-352. Morris, J., Pestaner, M., & Nelson, A. (1967). Mobility and achievement. The Journal of Experimental Education, 35, 74-79.

76

JASON ET AL.

Nair, D., & Jason, L. A. (1985). An investigation and analysis of social networks among children. Special Services in the Schools, I, 43-52. Panagos, J. L., Holmes, R. L., Thurman, R. L., Yard, G. J., & Spaner, S.D. (1981). Operation Sail. One effective model for the assimilation of new students into a school district, Urban Education, I5, 451-568. Parker, D. H., & Scannell, G. (1987). Developmental I Reading Laboratory Kit. Science Research Associates, Inc., p. 9. Parker, J. G., & Asher, S. R. (1987). Peer relations and later personal adjustment: Are low-accepted children a! risk? Psychological Bulletin, 102, 357-389. Parkes, C. M. (1971). Psycho-social transitions: A field of study. Social Science and Medicine, 5, 101-115. Pedersen, F. A., & Sullivan, E. J. (1964). Relationships among geographical mobility, parental altitudes and emotional disturbances in children. American Journal of Orthopsychiatry, 34, 575580. Perrodin, A. F., & Snipes, W. T. (1966). The relationship of mobility to achievement in reading, arithmetic, and language in selected Georgia Elementary Schools. Journal of Educational Research, 59, 315-319. Pillen, B. L., Jason, L.A., & Olson, T. (1988). The effects of gender on the transition of transfer students into a new school. Psychology in the Schools, 25, 187-194. Poser, E. G. (1970). Toward a theory of behavior prophylaxis. Journal of Behavior Therapy and Experimental Psychiatry, I, 39-45. Price, D. 0. (1965). Next steps in studying mobility and mental health. In M. R. Kantor (Ed.), Mobility and mental health (pp. 238-247). Springfield, 1L: Charles C. Thomas. Rhodes, J., & Jason, L. A. (1988). Preventing substance abuse among adolescents. New York: Pergamon. Rosenshine, B. V., & Berliner, D. C. (1978). Academic engaged time. British Journal of Teacher Education, 4, 3-160. Rumberger, R. (1983). Dropping out of high school: The influence of race, sex and family background. American Education Research Journal, 20, 199-220. Ruiter, M. (1979). Protective factors in children's responses to stress and disadvantage. In M. W. Kent, & J. E. Rolf (Eds.), Primary prevention of psychopathology: Social Competence in children (Vol. III, pp. 49-74), Hanover, NH: University Press of New England. Sameroff, A. J. (1987). Transactional risk factors and prevention. In J. A. Steinberg & M. M. Silverman (Eds.), Preventing mental disorders: A research perspective (pp. 74-89), DHHS Pub no. (ADM) 87-1492. U.S. Government Printing Office, Washington, D.C. Sandler, I. N. (1980). Social support resources, stress and maladjustment of poor children. American Journal of Community Psychology, 8, 41-52. Sandler, I. N., & Block, M. (1979). Life stress and maladaption of children. American Journal of Community Psychology, 7, 425-440. Sandler, I. N., & Ramsay, T. B. ( 1980). Dimensional analysis of children's stressful life events. American Journal of Community Psychology, 8, 285-302. Schaller, J. (1974). Residential change and academic performance. Goteborg Psychological Reports, 4(6), 1-20. Schaller, J. (1975). The relationship between geographic mobility and school behavior. Man-Environment Systems, 5, 185-187. Schaller, J. (1976). Geographic mobility as a variable in ex-post facto research. British Journal of Education Psychology, 46, 341-343. Shaw, J. A. (1979). Adolescents in the mobile military community. In S. C. Feinstein & P. L. Giovacchini (Eds.), Adolescent psychiatry. Development and clinical studies (Vol VII, pp. 191198). Chicago: University of Chicago Press. Shaw, J. A., & Pangman, J. (1975). Geographic mobility and the military child. Military Medicine, I40, 413-416.

2.

A PREVENTIVE INTERVENTION FOR HIGH RISK CHILDREN

77

Signell, K. A. (1972). Kindergarten entry: A preventive approach to community mental health. Community Mental Health Journal, 8, 60-70. Sloan, V. J., Jason, L. A., & Bogal, G. A. (1984). A comparison of orientation methods for elementary school transfer students. Child Study Journal, 14, 47-60. Sroufe, A. L., & Rutter, M. ( 1984 ). The domain of developmental psychology. Child Development, 55, 17-29. Stewart, A. J., Sokol, M., Healy, Jr., J. M., Chester, N. I., & Weinstock-Savary, D. (1982). Adaptation to life changes in children and adults: Cross-sectional studies. Journal of Personality and Social Psychology, 43, 1270-1281. Stufflefield, R. L. (1955). Children's emotional problems aggravated by family moves. American Journal of Orthopsychiatry, 25, 120-124. Tetreau, E. D., & Fuller, V. ( 1942). Some factors associated with school achievement of children's migrant families. Elementary School Journal, 42, 423-431. Tiffany, F. ( 1984 ). Separation stress of young children transferring to new schools. Developmental Psychology, 20, 786-792. Turner, I., & McClatchey, L. (1978). Mobility, school attainment and adjustment. Association of Educational Psychologists Journal, 4(9), 45-50. Vaillant, G., & Vaillant, C. (1981). Natural history of male psychological health, x: Work as a predictor of positive mental health. American Journal of Psychiatry, 138, 1433-1440. Whalen, R. E., & Fried, M. A. (1973). Geographic mobility and its effect of student achievement. Journal of Education Research, 67, 163-165. Ziller, R. C., & Behringer, R. D. (1961). A longitudinal study of the assimilation of the new child in the group. Human Relations, 14, 121-133. Zoline, S. S., & Jason, L. A. (1985). A preventive parent education for high school students. Journal of Clinical Child Psychology, 14, 119-123.

Page Intentionally Left Blank

3

Sociometry, Temperament, and School Psychology

James R. Barclay University of Kentucky

This chapter reviews the technique known as sociometry and suggests that school psychologists take another look at it. For several decades this technique has not been used much in the schools. Thus an entire generation of school psychologists have come into the schools without a knowledge of this technique and its relevance for evaluating individuals in a social environment. In point of fact, it has been disused for so long that a number of young researchers are "discovering" it again, without any recognition of the research that has been reported.' With increasing demands from parents and schools to provide more services to children in the broad spectrum between handicapped conditions and giftedness, and with the mounting problems relating to academic anomie, drug usage, and undue peer pressure, sociometry ought to be considered an important element in the II am grateful to the editors of this series for allowing me to provide a perspective on 35 years of research relating to the use and application of sociometric techniques in assessment and intervention. This chapter provides a kind of apologia pro vita sua for my own research with the Barclay Classroom Assessment System. I am not doing this for self-serving reasons, but rather to share with new researchers in the social competency field a personal preoccupation with the need to utilize assessment as a means of devising interventions. From a single technique, sociometry has developed into a part of a much more sophisticated assessment procedure, in fact, a system. With a renewed interest in the use of sociometry and in the context of increasing concern about the mental health of our youth, it seems pertinent to provide an overview of the historical development of the technique. In addition, with the current interest and research with temperament constructs, the connection between social perception and temperament needs to be explored. I have attempted to write this chapter in such a manner as to provide the reader with the story of how interest in the problems of dropouts and social maladjustment in a suburban school district 35 years ago led step by step to a fascinating detective-like process that forced me to examine corridors and halls within the research edifice that I had never thought as a student I would be looking into.

79

80

BARCLAY

screening of large numbers of children to determine needs that school psychologists might design using all participants in the school situation. THE NATURE OF SOCIOMETRY: A BRIEF OVERVIEW

Sociometry is a method of discovering and analyzing patterns of friendship within a group setting. Its primary asset to assessment is that it provides both an independent source of individual assessment based on peer perception, and an indirect, but powerful estimate of psychological support, i.e., peer reinforcement. The earliest literature on sociometry was provided by Moreno in a book entitled Who Shall Survive (Moreno, 1934). Moreno, in developing the specific technique of sociometry, saw it as a branch of a larger science which he called "socionomy." He believed that sociometric patterns of choice reflected a profound set of process values that continued to influence individuals throughout life. He believed that such values could be altered and that by altering them, or enhancing them, individuals assumed power positions not only for themselves, but as related to others. For example, Moreno and Moreno (1976) suggested that Hitler exemplified the process of negative leadership evolution. Hitler started as a disillusioned isolate who seized upon negative issues stemming from the first world war and galvanized the hate and avenging tendencies of a nation. On the other hand, Roosevelt, proceeding from a patrician background, focused on national survival issues and created his power base in this manner. In effect, Moreno believed that sociometry was an indispensable method for evaluating the effect of individuals in a group, and what he called the social atom. He believed that the technique had profound possibilities for education and mental health. Another very important component in the development of sociometric techniques was the confluence of both Lewinian field theory with the mental health movement. Sociometry was seen as a viable method for assessing social and psychological support systems of individuals. This was important to develop a course of social action and involvement in developing new self-perceptions in individuals. Lewinian ideas were forged in the consequences of the Nazi holocaust and asserted that both individuals and institutions needed to be re-trained, to learn responsible and internal ways of reacting with others in social justice. The mental health movement assumed a personality theory based on both human intentionality and motivation, and the use of social learning mechanisms that would result in personal responsibility for making judgments, and committment to social action. Thus, sociometry played a key measurement role both in assessing the status quo, the designing of appropriate strategies of intervention, and the evaluation of consequences of those interactions. This was the theoretical background of many psychologists at the beginning of the 1950s. It was overwhelmingly the backbone of school psychology as it began to expand. There were only 519 school psychologists operating in the

3.

SOCIOMETRY, TEMPERAMENT, AND SCHOOL PSYCHOLOGY

81

United States at the beginning of the 1950s (Meacham & Trione, 1967). Even by 1960 the total number had only increased to 2836. In those days, psychologists who operated within the schools were much more psychological consultants in a variety of educational and mental health roles and had much more freedom to contract with administrators regarding their tasks. Most did some individual testing, but in addition, there was a considerable emphasis on personal and social adjustment. Indeed for many young psychologists entering this field after World War II, there was a kind of missionary conviction that re-education was a key factor in the assurance that there would never be another time of horror such as the world had experienced. Many scholars were associated with this movement. Some of them were colleagues of Lewin such as Fritz Heider, but many others were either students of Lewin or contemporary with Lewin's students. Some of these were Eli Bower, Boyd McCandless, William Morse, Merle Bonney, Leon Festinger, Jacob Kounin, Ronald Lippitt, and Norman Gronlund. Lewin's influence was great on these individuals, and three of them, McCandless, Kounin, and Lippitt, did their own doctoral work with him. These individuals pioneered various applications of mental health approaches that reflected the general social learning thrust and were highly applicable to the schools. For example, Heider (1958) urged that phenomenological thinking be used to clarify the nature of human interaction through the acceptance of common experience. Kounin (1970), Kounin and Gump (1961), and Kounin, Gump, and Ryan, 1961) examined the influence of teachers on the misconduct of children, Morse (1965) focused on intervention techniques that could be used by counselors in fresh-air camps and by teachers. Redl and Wineman (1952) explored the establishment of controls from within for the aggressive child. It was reasoned that if much variability in human behavior was due to unfavorable group interaction, then it was important to evaluate the quality of group interaction and specifically those forces of individual differences that interacted both as negative factors and positive ones. Although there was much variability in ideas and programs, there was a common core of concepts that emphasized the postulates of Lewin: (a) learning is an interaction between the person and the environment; (b) perception is the key variable in determining behavior; (c) affective learning does not necessarily correlate with cognitive learning; (d) correct knowledge does not necessarily result in the alteration of false perceptions; (e) and change is a function of responsible involvement in appropriate social contingencies (Lewin, 1935; Lewin & Grabbe, 1945). Sociometric techniques evolved as one means to measure the complex relationships identified by Lewin and Grabbe. Most of them were related to the development of a variety of methods that could be used to estimate the individual's status in a group situation. Several researchers who pioneered work in this technique in the field of education were Merle Bonney (1943a, 1943b;

82

BARCLAY

Bonney & Powell, 1953), M. L. Northway (1952), Norman Gronlund (1951, 1955, 1959), Ned Flanders (Flanders & Havumaki, 1963) and E. Amidon (1963). Although many of these applications were in the field of education, the technique was also used in the military, industry and business to predict a wide variety of personnel problems (Hart & Nath, 1979; Mouton, Blake, & Fruchter, 1955b). A further elaboration of sociometry as a technique extended research into the area of mathematical models. For example, Coleman (1961) analyzed in diagrams cliques in the Elm town High School. Much later, Hunter ( 1978) and Langeheine ( 1978) both provided extensive inquiries into the algebra of predicting consistency in individual and group choices and the development of theorems for illustrating these models. In the late 1960s Barclay recognizing the need for tapping into alternative sociometric judgments related not only to friendship, but skill production and competencies in individuals, developed a computerized base for analyzing sociometric phenomena. This technique was first designed in collaboration with computer technicians at Stanford University in 1968 and subsequently perfected at the University of Kentucky (Barclay, 1972). It still remains the basis of the Barclay Classroom Assessment System (Barclay 1983a). Although a good deal of research was done during the decades of the 1940s, 1950s, and 1960s, this research was heavily directed at social process and mental hygiene goals of education. It therefore had an appeal to those who would improve the overall quality of education for all children rather than focus on a few children with many difficulties. In the 1970s and into the 1980s, sociometric studies fell into decline. A number of reasons are related to this fact. First of all, the influence of Rogerian approaches to counseling tended to denigrate the use of all methods of diagnosis imposed by a psychologist on others. This included all kinds of testing. Second, towards the end of the 1960s the growing popularity of behavioral techniques tended to deprecate all other methods of assessment even though sociometries were consistent with behavioral approaches and could be used as a criterion for selected reinforcement procedures (Barclay, 1967). Third, the emphasis on mental health focusing on internal changes of responsibility, that had been the keystone of Lewinian ideas, gradually declined and disappeared in favor of direct methods of external behavior change. In part this occurred not only because of the immediate outcomes of behavioral interventions, but also because there was an impatience to show change, and the long-term assumptions of the mental health approach seemed unverifiable in the perspective of a renewed thrust for scientific rigor in evaluation. Fourth, with the rise of school psychology as a professional specialty and the passage of PL 94-142, the emphasis in school psychology was directed specifically at special education problems rather than generic mental health ones. Moreover, the more detailed specification of the role

3.

SOCIOMETRY, TEMPERAMENT, AND SCHOOL PSYCHOLOGY

83

of the school psychologist as an adjunct to the special education program, and the increased flow of judicial and legal decisions reflecting negatively against experimental studies within the educational setting. Legal restrictions and the mounting requirement for continuous retesting of children already assessed year after year led to a clear injunction to focus on "safe techniques and methods." Fifth, those individuals who had some experience with sociometries realized that there was a great deal of work that had to be done to develop sociograms and score these data manually. Many of those who used sociometries found them difficult to score, time consuming to analyze, and could not see any relationship to what strategies of intervention might be associated with their use. Although sociometric data could and were scored by computer analysis, those who used the technique either found out about such developments too late or were not aware of their potential for use. Perhaps the greatest single problem with sociometric data has been the lack of a comprehensive theory about the meaning of the phenomena. To be sure, the technique was carefully defined (after a survey of 200 users) as "the field that should be concerned with the quantitative treatment of every kind of interhuman relation and particularly with those involving the expressions of preference or rejection of other members of a group with respect to a choice situation" (Bjerstadt, 1956). This definition, however, rested on the process foundations laid by Moreno and amplified by Lewinian thought. On a more practical level, Barclay (1966a) defined sociometry "as a method for discovering and analyzing patterns of friendship within a group setting" (p. 1070). Given the foregoing changes in emphasis and political-legal redefinition of school psychology, this lack of theoretical development and the emphasis on sociometries as a technique led to its virtual dismissal by school psychologist trainers as a tool with possible consequences for assessing learning, behavior and human development. The fact that sociometry is not considered as a tool for designing interventions is amply documented by a review of interventions by Shapiro (1987). It would appear from his review that only behavioral interventions have any relevance to educational problems. In the 13 journals that publish school psychology studies he found only three nonbehavioral interventions cited out of 597 mentioned in the period from 1981-1986. Sociometry has no role at all in this review. In summary, sociometry had its origins in the work of Moreno. The primary concern of researchers in the early decades was with alternative techniques to measure sociometric data. With the decline of the mental health movement (broadly associated with Lewinian field theory and psychodynamic theory), and the advent of behavioral technology as a basis for assessment, sociometries have declined. In addition, as school psychologists became more involved in the assessment of special education students as contrasted with the earlier emphasis on the overall mental health needs of children, individual analysis became the

84

BARCLAY

overriding specific focus of most school psychologists. A good indication of the status of sociometry as a tool is the fact that the last major book dedicated solely to sociometry was published in 1959 by Gronlund.

THE NATURE OF SOCIOMETRIC METHODS

During the first decades in which sociometric methods were developed, a natural focus was on both the characteristics of the method itself and the reliability and validity of the results. Within the field of education there were many techniques that were explored. Most of them were concerned with tapping into global friendship patterns. Sometimes these patterns were obtained by seeking both positive and negative choices. For example, researchers sought to identify those children who were considered friends either in work or play, and conversely those who were not. From the earliest days on, the use of negative choices was considered as a method to be carefully used. Thus, it did not appear to be wrong to ask children whom they would rather not work and play with, but care needed to be taken not to make the negative sociometric question too specific, e.g., "who would you like to push down the stairs?" Gronlund (1959) suggested that a way around this problem was to ask who children would choose first and last for a given activity. Although there were many varieties of sociometric techniques developed, some calling for children to classify peers in a number of categories such as (a) "my very, very best friends," (b) "my other friends," (c) "not friends but okay," (d) "don't know them," (e) "don't care for them," and (f) "dislike them," (Ohio Social Acceptance Scale, undated), the most common approach was to ask children to identify youngsters with whom they would like to play and work. For example, a Manual of Sociometry for Teachers (Smith, 1951) at the University of Michigan recommended the following format for use with elementary children. Boys and girls, this is a choosing exercise that will help me to find out which boys and girls in our room work and play best together. Answer the questions just as honestly as you can; no one else will look at your answers so be sure to write the person's name that you really want. Please do not mention your choices to any one else.

1. Write the names of the children with whom you would like best to play any of the games that we know. There might be several so write the names of any you would really like. 2. Now, write the names of the children with whom you would like best to work or make something-any work that we do. Again there might be several so write the names of any you would really like.

3.

SOCIOMETRY, TEMPERAMENT, AND SCHOOL PSYCHOLOGY

85

In an example of this method, the teacher gives this to his/her class, and obtains 22 samples of data to be examined with 6 choices from each child, or a total of 132 choices. It is then necessary to tally all of these choices, and from that tally derive a sociogram that consists of concentric circles showing the relative social power of each child. Teachers not only had to spend considerable time in doing this, but they needed to understand certain terms associated with sociometries such as: "star isolate rejectee clique and cleavage." For example, the term star usually referred to an individual who was widely accepted by the group. This individual had a strong social stimulus value within a given group. An isolate was a child who was not actually rejected as such, but because of personality factors was largely ignored. An isolate might also be a child recently included in the classroom and not known well by the other children. A rejectee was a child who was actively disliked and received negative choices. In this particular sociogram no negative choices were used so one cannot determine whether there were any rejected children. The term clique often referred to a special in-group constellation of children who associated with each other, and cleavage referred to divisions within the classroom often by sex, or sometimes by race or other socio-economic considerations. From the description of the process it is apparent that in the early decades of work with sociometries many individuals were discouraged from using the technique simply because of the time consumed in collecting the data, tabulating it, and then creating a sociogram. Once it was completed, moreover, few teachers knew what to do about the results. Barclay, in the mid-1950s working with a suburban Detroit school and wishing to utilize both sociometries and teacher ratings as a rough screening procedure to identify children with learning problems, developed a simplified method for obtaining some of the sociometric data. He asked students a general positive and a general negative sociometric question. Rather than use names, he provided a form with numbers on it and had the teacher write the number of each child on the blackboard or give the child a listing of names and numbers. His instructions were: All of us like some people better than others. On the board is a list of the names of boys and girls in your class, with a number before each name. Who are your best friends in this class? Who are the boys and girls with whom you most like to work and play? Circle the number that goes with your friends' names. You may choose as many or few as you want.

The negative question was couched in this form. As we said before, all of us like some people better than others. Using the numbers next to each child's name, circle those numbers that go with the names of

86

BARCLAY

boys and girls with whom you would rather not play or work. Again choose as many or as few as you want. (Barclay, 1964)·

Scoring of this sociometric instrument was comparatively simple since the total number of positive and negative choices could be easily obtained. To obviate negative values an arbitrary mean of 50. was assumed. The sociometric index for a given child was then the total number of positive choices minus the total number of negative choices with an assumed mean of 50. Thus, if Joe received 10 positive choices and 4 negative ones his score was 56. If Susan received 0 positive choices and 12 negative choices, her score was 38. Interim norms were created on these data from 70 classrooms ranging from the 3rd through the 9th grade. Because this approach was developed in a school system and used as a screening device for more accurate consultation, a teacher rating of each child was also asked for (Barclay, 1964). It was reasoned that the two major sources of within-class reward and modeling came from peers and teachers. However, teacher ratings were known not to correlate highly with the social skills of students (Bonney, 1943a, 1943b; Bonney & Powell, 1953; Gronlund, 1951, 1959). The Barclay teacher ratings were made on a 5-point scale extending from excellent to poor on the following components: (a) overall emotional adjustment, (b) social maturity, (c) depression, (d) aggression, (e) security, (f) stability and (g) conduct. The scoring of the teacher ratings then yielded scores ranging from the highest score of 35 to the lowest score of 7. The results of these two teacher ratings and sociometric scores were then combined on a 9-cell grid where sociometric ratings were plotted on the vertical axis and teacher ratings on the horizontal axis. Figure 3.1 illustrates this arrangement. These cells were helpful in obtaining an overall global view of the child in the Grid Arrangement Sociometric Axis

Cell1

Cell2

Cell3

66th Percentile

Cell4

CeliS

CeliS

-33rd Percentile

CeliS

Cell?

0

33 FIG. 3.1.

Grid arrangement.

Cell9 66 Teacher-Rat1ng Ax1s

3.

SOCIOMETRY, TEMPERAMENT, AND SCHOOL PSYCHOLOGY

87

classroom. Cells 3, 5, and 7 reflected major agreement between peer and teacher ratings as to high, average, and low social status. The other cells showed discrepancies between peer and teacher ratings. Cell 3 has a sample of children highly accepted both by teachers and students. Cell 7 is the most negative of the cells and represents the area where both peers and teachers have rejected or at least ignored the individual. Cell 1 contains those individuals who are leaders among their peers, but who may express values or skills not appreciated by the teacher. Cell 9 is the converse of cell 1. Individuals here have higher teacher ratings but poor sociometric status. These children typically might be viewed as teacher pets. A number of studies were completed with the basic method described above. The basic intention of these studies was to explore ways in which the sociometric phenomenon was related to other social and personality variables. For example, Barclay and Barclay ( 1965) found a significant relationship between lack of social acceptance by peers and teachers in 3rd-grade students who were impulsive and acting-out. Barclay ( 1966b ), in a follow up of 949 children tested originally in 1959 and again in 1963, found that 54% of the female dropouts from school and 65% of the male dropouts had been located initially in a cell of maximum rejection (cell 7). This study confirmed what Kuhlen and Collister (1952) had found about predicting school dropouts from the 6th and 9th grades by use of sociometric indices. Again, these cells were compared for interest patterns in class activities, television viewing habits, music listening preferences, study preferences and habits and hobbies for 1777 elementary and junior high school students (Barclay, 1966c ). It was found that significant differences in preferences accompany high and low social acceptance in the classroom. Several other studies addressed the quality of the environmental conditions expressed by overall averages of teacher ratings and peer ratings for 70 classrooms. Male elementary school teachers tended to have overall lower mean peer ratings and showed lower ratings for children than did female teachers. Young female teachers tended to rate boys the highest and foster the most positive peer interaction, and older female teachers were most positive towards girls (Barclay 1966d). Davis ( 1967) verified the predictive validity of the Barclay method. She found a significant correlation between 445 students tested in 7th and 8th grade by Barclay and retested in lOth grade. Drawing random names from the various cells she presented teachers with the names of 100 students. Teachers confirmed independent of either the pretesting or posttesting results the social acceptance status of students. Kennedy ( 1969) found the system helpful in relating referrals to a mental health facility. Finally, the system was used as a criterion for initiating planned interventions, selective reinforcement and modeling procedures, and change of teachers in three classes. The effect of these interventions could be assessed in terms of the social acceptance of students (Barclay, 1967; Forsythe & Jackson, 1966). Perhaps the most important feature of this latter study was that

88

BARCLAY

the cell arrangement of the teacher-peer grid provided a direct criterion measure for selective behavioral and intervention techniques. What this means is that individuals who were low in social acceptance or teacher ratings on the pretest could be identified by inspecting the sample in cell 7. These individuals could be targeted for selecting reinforcement procedures. If they subsequently moved out of that cell to another more favorable one, this could then be considered a criterion of change. Sociometric choices have also been used with older students. An example of this is the work of Smith (1967) in analyzing the peer support system of 583 college students, 521 nursing students, and 324 high school students. Smith used 42 adjectives, each of which had polar connotations and asked each student to nominate five of their classmates who met the one alternative and five who met the polar opposite. This format was used along with achievement tests, the Edwards Personality Preference Inventory, and high school grade point average (GPA) to predict first year achievement (for the general college students only). He found in a regression using the above predictors and first year GPA as the criterion that peer ratings accounted for 68% of the total variance; academic aptitude tests accounted for 19%; high school grades for 13% and the Edwards Personality Preference Inventory for 0%. One adjective alone accounted for 55% of the variance. That item was quitting. Similar factor structures were found for nursing, college, and high school students. These factors were Agreeableness, Extroversion, Strength of Character, Emotionality and Refinement, with Strength of Character being the strongest factor. RESEARCH WITH EARLY VERSIONS OF SOCIOMETRY

A considerable amount of research was done with early sociometric tests. Because there were so many versions of simple sociometric tests using different numbers of peer choices, different criteria for making choices, positive choices, negative choices, and the use of ratio indices, studies were often related to a specific technique. Nonetheless, a considerable body of research was developed in reliability and validity studies. We look at them briefly. One of the first concerns with sociometry was to determine its reliability. Although many types of reliability were undertaken including: (a) internal consistency or split-half, (b) equivalent or alternate forms, and (c) test-retest, some methods seem to have been used more than others. Lindzey and Borgatta (1954) favored using graphic results to confirm the relative position of individuals within the sociogram. The split-half method was used but criticized by Gronlund as having no special significance (Gronlund, 1959). The alternate forms method involved administering several sociometric questions to the same group at the same time and then correlating the results. Although Kerr (1945) found low positive correlations for associates between eight sociometric questions that re-

3.

SOCIOMETRY, TEMPERAMENT, AND SCHOOL PSYCHOLOGY

89

ferred to different activities, and Gronlund (1955) found a high degree of relationship between the resulting sociometric status scores obtained from three general sociometric questions of work, play, and seating arrangements with five choices allowed on each, the evidence suggested that correlations between sociometric questions was specific to the activity involved. Thus, children may identify certain peers as most popular for attending a party (possibly because the family is more wealthy and can afford good parties), but they may choose someone else as a friend with whom to study. This conclusion suggests that children discriminate between different kinds of friends depending on the stimulus of the question posed to them. The most popular form of reliability has been test-retest. Mouton, Blake, and Fruchter ( 1955a) summarized in detail 27 studies on the consistency of choice status found by test-retest methods. They concluded that choice status of the individual remains consistent over the time he or she is evaluated by others. Gronlund (1959) also reviewed many studies and found substantial agreement of the stability of sociometric status among adolescents. He suggested that over time intervals, those of less than a week, coefficients of approximately .90 might be expected; over several weeks, correlation coefficients of .80 are indicated and for time intervals between 8 and 20 months, coefficients of approximately .60. Barclay (1956) in a 1-year follow up of over 2500 elementary children considering each individual pre and post, confirmed this judgment by obtaining a correlation of .70 in pre and post sociometric status. This was true even though most of the sample had moved from elementary schools to a junior high school and were in totally different surroundings and often with different peers. The method used here was that described earlier with both positive and negative choices obtained. Where sociometric questions are related directly to specific skill areas, Barclay ( 1974a) found that there were pre-post correlations over the course of one academic year of .58 to .77 for artistic-intellectual peer choices (Who can write poetry? Who reads the most books?), .57 to .64 for outdoor-mechanical choices (Who knows how to make things with his or her hands? Who knows how to make model airplanes?), .66 for social skill choices (Who is a good leader? Who knows how to listen to other?), .65 to .71 for enterprising choices (Who always answers question fast? Who talks the most in class discussions?), .50 to .67 for group reticent and shy nominations (Who is shy and afraid to answer the teacher? Who is most quiet in class?), and .58 to .70 on group disruptive and acting-out nominations (Who makes noise in class? Who can't sit still?). In summary, the research suggests that there is a tendency for sociometric results to decline as the time span between tests is increased. Results based on general criteria are more stable than specific criteria. Composite sociometric scores based on several criteria tend to be more stable than scores based on a single sociometric criterion. The use of a limited number of sociometric choices such as 5 positive choices or 3 positive and 3 negative tends to yield higher reliability results than unlimited choices. Finally, extreme sociometric positions

90

BARCLAY

such as strong leadership, rejection, and isolation tend to be more stable than peer choices somewhere in between (Gronlund, 1959, pp. 152-153). In the area of validity there have been many studies completed. Some have held that the construct validity of sociometric choice is valid by definition. For example, Jennings ( 1950) contended that the sociometric test is not an indirect measure of other behavior, but rather a sample of actual behavior phenomena and as such is directly meaningful and need not be validated by relating it to an external and independent criterion. He felt that if each individual discloses his or her preferences on the test in an honest way, the test is valid. On the other hand, even if biases exist-such as in racial discrimination, the reality of these impressions are valid indicators of what is actually happening in a given environment. Although some reading this statement might question how one would know whether biases exist, what Jennings was suggesting is that bias or not, the actual force of what exists is being measured by definition. Concurrent and predictive validity is the most popular approach to the validation of sociometry. Mouton, Blake, and Fructer (1955b) reviewed 44 studies on the validity of sociometric test scores in educational, industrial, and military settings and found that sociometric scores agreed with many outside criteria of performance. Table 3.1 groups a number of representative studies relating to the validity of sociometric choices and provides a brief summary of findings. These studies indicate that sociometric phenomena have been correlated with social worker and psychologist ratings, mental health ratings and teacher ratings. They have shown a relationship to personality tests such as the California Test of Personality, the High School Personality Questionnaire, the Rorschach, the K varaceus Delinquency Proneness Test and other measures of perception such as the Id-Ego-Superego Test, and the Ellis Visual Designs, motivation and selfconcept tests, achievement and intelligence measures. References for these studies are cited in Table 3.1 Sociometric indices have also been related to distinctive interest patterns, the socioeconomic status of parents, and behavioral observations in the classroom. They have been related to military status, have been predictive of dropouts in the school, have been used in experimental studies as criterion measures, and appear to have enduring longitudinal and predictive validity in terms of a variety of criterion measures of effective human behavior. Further Developments

By the mid-1960s it was apparent to many researchers that sociometric methods related to a host of personality, achievement and behavioral variables. The writer's research suggested strongly that sociometric phenomena were related to a variety of personal and social skills and that peers recognized such skills and competencies in their choices of leaders. There was evidence that longitudinal stability existed for such peer choices, and that socioeconomic status and access

TABLE 3.1 Summary of Older Validity Studies Area or Criterion

Mental Health California Test of Personaltiy

Author(s)

Findings

Bedoian (1953)

Low status children lower on mental health ratings.

Bedoian (1953); Barclay (1964)

Significant differences between high- and low-status children.

High School Personality Guinouard and Questionnaire Rychlak (1964)

High sociometric status significantly related to intelligence, self-confidence, enthusiasm, and acceptance of rules.

Rorschach

Tindall (1955)

Significanct relationship between social status and protocols.

Kvaraceus Delinquency Proneness Test

Barclay (1964)

Correlation between sociometric status and delinquent status - .53.

Id-E go-Superego Test

Barclay (1964)

High sociometric status significantly higher on positive ego scores; significantly lower on impulsivity scores.

Ellis Visual Designs

Barclay and Barclay (1965)

High sociometric status significantly higher on perceptual organization.

Teacher Ratings

Gronlund (1955, 1959); Barclay (1965

Low positive association between sociometric status and favorable ratings.

Achievement and Intelligence

Barclay (1964); Gronlund (1959)

Low positive relationships between achievement and sociometric status

Miller (1956)

Superior students most often chosen, followed by average and then retardates.

Gallagher and Crowder( 1957)

.45 correlation bwtween IQ and sociometric status.

Barbe (1954)

Bright children choose friends with slightly lower lOs.

Barclay (1974b)

Group sociometric choices significantly related to selfcompetency scores.

Barclay (1966c)

Sociometric status related to different interests in many areas with popular children more interested in TV and Sports.

Industrial, Military Business

Mouton, Blake, and Fruchter(1955b)

Review of 44 studies show advantages for high status individuals in all areas. Sociometric status clearly predictive of performance.

Prediction of Dropouts from school

Kuhlen and Collister (1952); Barclay (1966b)

Both studies were long-range (involving 1400 ss) and indicate low sociomatric status children significantly more prone to drop out of school.

Sociometry as a Criterion for Change

Amidon (1963); Dineen and Garry (1963); Kranzler, Mayer, Dyer, and Munger (1966); Barclay (1967)

Efforts to change isolate behavior (Amidon); classroom cleavage through seating (Dineen & Garry); use of behavioral reinforcement and modeling to change social status (Mayer et al.; Barclay).

Self-concept and Interest Patterns

91

92

BARCLAY

to skill development activities were factors in such skill development and social acceptance. Moreover, there seemed to be an implicit connection between paternal occupation and child skills (Holland, 1964). Despite this knowledge, the tide was turning against the use of sociometry. It was complex. It took time. Teachers did not know how to use it. Psychologists did not know how to use it either. In addition, the rejection of personality tests and the alarm at their use was generalized to sociometric methods. With increased interest in behavioral methods, only a few researchers saw that sociometry could be a valuable criterion for judging the effects of social interventions. In the mid-1960s, largely due to the earlier research cited, Barclay designed a new system to include peer nominations in specific skill areas, self-reported skill competencies and reinforcers, vocational interests, and teacher ratings. The skill competencies, both as self-rated and peer-rated encompassed outdoor-mechanical skills, artistic ones, intellectual skills, enterprising, social, and conventional competencies. There were 28 sets of nominations altogether. In this system the general friendship criterion was abandoned in favor of a set of specific skill criteria. In addition, the negative question was dropped. Along with these peer nominations, there was a self-report section in which children rated their own skills on the same criteria as peers used (thus providing a comparison of selfreported skills with peer-evaluated ones for each individual), and a teacher evaluation component based on adjective ratings. In addition, interests in various vocations and types of reinforcers were added to the self-report component. Initial studies were done utilizing a project in Alameda County California that focused on reticent and disruptive children. Studies with the data (Ferris, 1968; Stickel, 1968) indicated that this approach discriminated very well between reticent and disruptive children. However, the task in assessment had now not only doubled and tripled, but had become impossible without computer technology. Through the assistance of the Stanford Research Institute Center and specifically Richard E. Snow of Stanford University, a computer analysis system was devised for analyzing the data. Though the process in the late 1960s was highly cumbersome by comparison with present methods, it did provide a model for future development in that it became possible to collect a variety of sociometric choices, summarize them by computer, and interpret them. Not only was this done for the sociometries, but for all of the rest of the data. With the computerization of the data input, the use of sociometries became one component of a matrix of self, teacher, and peer assessment. Now, however, the system that issued from this technology resulted in an entirely new ball game. It was no longer a question of the researcher or psychologist sitting down with a mass of charts and names and trying to understand what was inferred; rather it was a question of programming the computer to become a diagnostician and then reporting such information in a way that it could be used by consumers. It soon became evident that IF such a system was to be used, it had to be very specific in

3.

SOCIOMETRY, TEMPERAMENT, AND SCHOOL PSYCHOLOGY

93

recommendations. Neither teachers, guidance counselors, nor school psychologists had the time to use the older techniques. Further it was extremely important to build into the computer screening system a means for identifying quickly those children who needed some kind of an intervention. As a result, a system was developed that attempted to measure three major inputs simultaneously: (a) the self-report of children including their own evaluation of their skills and interests in occupational areas and reinforcers; (b) peer nominations in 22 areas identical to the self-report skills; and (c) teacher ratings of these children using adjective descriptions. In the 20 years since this system was initiated, a great deal of research has been done with it. Most of the research has focused on developing simple strategies of intervention related to social or personal skill deficits. Later on, the research has integrated findings from temperament-treatment interactions for differential suggestions for children who manifest alternate temperament styles. Because this research has been detailed in the Manual of the Barclay Classroom Assessment System (Barclay 1983a), it is only summarized in here to provide the reader with the extent and specificity of this system. Table 3.2 provides this summary. One of the most important considerations for use by school psychologists was to determine a system for analyzing the output that would be directly relevant to interventions by school psychologists, special education personnel, and teachers. Through many years of experimentation and incorporation of ongoing research studies, the program that drives the Barclay Classroom Assessment System (BCAS Barclay, 1972, 1983a) became a computerized diagnostician that scored the data, read it, integrated it, and, using a number of programmed algorithms made and reported these decisions (the details of these procedures are presented elsewhere, Barclay, 1988b). Thus for example, a significantly low score on positive peer ratings and an absence or low number of positive teacher descriptors leads to a computer judgment that the psychological support system of the child is poor. Many peer ratings of shy, reticent, or socially withdrawn behavior joined to similar judgments by the teacher in selecting key reticence adjectives leads to the computer judgment that the child is probably reticent and shy. The output therefore provides readers with a comprehensive diagnostic view of a specific child in a specific environment. The research that has been done with the system not only provides descriptive characteristics of the individual, but pin-points areas of concern for teachers, principals, elementary counselors, or school psychologists. It suggests strategies of intervention for reticence, for increasing peer or teacher support, for verifying impressions that a given child is disruptive or gifted, and for alternative learning strategies. Based on research with alternative treatments in an aptitude-treatment paradigm method and using effect size scores as pre-post measures of change, it also provides interim hypotheses (for further evaluation) about what kinds of curricula or counselor interventions may work best with different children (Bar-

TABLE 3.2 Research with the Barclay Classroom Assessment System Area or Criterion

Discriminability

Author(s)

Findings

Tapp and Barclay (1974) Multimethod, multitrait validation. Barclay (1983a)

Systems comparison congruent with Bennett study in England (1976).

Barclay and Wu (1980)

Comparison of Chinese and American samples show cross-cultural validity of factors and constructs.

Barclay (1974c)

Consumer reactions to use of BCAS.

Barclay, Barclay and Stilwell (1972)

Sociometrtic status of children related to social status of fathers.

Barclay (1974b)

Relationship of seven problem areas compared by age and sex across three school districts.

Ferris (1968); Stickel (1968)

Significant differences in sociometries found for reticent and disruptive children.

Analysis

Barclay, Stilwell, Santoro, and Clark (1972)

Results of independent behavior observation of 700 children over ten days show significant relationships to BCAS categories.

Education

Kehle and Guidubaldi (1978); Barclay and Kahle (1979)

Discusses EMR and LD placements of 115 Ohio classrooms using BCAS and negative effects of mainstreaming.

Barclay, Phillips and Jones (1983)

Development of an index for identifying gifted using the BCAS with findings of 75% against individual tests.

Barclay (1974a, 1974b)

Discusses district-wide analysis and identification of target areas for intervention.

Stilwell and Barclay (1976, 19n. 1979)

Results of experimental intervention in affective domain using the BCAS as a criterion.

Barclay, Covert, Scott, and Stilwell (1975)

Three-year follow-up of behavioral, open, and traditional teaching methods and their consequences for learning in Terre Haute Indiana.

Barclay (1983b)

Meta-analysis of six different approaches to intervention using the temperament X treatment interaction.

Barclay (1983c)

Rationale for a prevention approach for school psychologists using district-wide screening.

Barclay (1987)

Discussion of temperament and its relationship to the CPI, 16PF, and Strelau.

Barclay (1988a, 1988b)

Analysis of the basis of temperament and derivation of temperament from computer analysis.

Barclay and Barclay (1986)

Early childhood skill analysis and competency.

Cohen (1983)

Relationship of BCAS to attribution theory.

Relations

Systems and Change

Temperament and Prevention

94

3.

SOCIOMETRY, TEMPERAMENT, AND SCHOOL PSYCHOLOGY

95

clay 1983b). In summary, the system was designed and researched to provide overall screening of elementary school children based on self, peer, and teacher judgments, AND to provide some specific suggestions for intervention. Where the BCAS differs from other student assessment techniques is that it attempts to provide a suggested intervention. The nature of the intervention is consistent with research studies, and would not be beyond the range of good psychometric judgment on the part of well-trained counselors and psychologists if they have the time to do this for large groups of children. However, to obtain this information, all children must be tested, even though not all reports require interventions. A set of decision rules specify the grouping of data for individuals and groups, the determination of whether a child's problems are primarily related to achievement or personal-social developmental areas or both. More examples of some of these decision algorithms are illustrated in a previous article (Barclay 1983c). THE THEORETICAL BASIS OF THE BCAS

Given the composite nature of the BCAS, a series of factor analyses over a period of years were used to develop and refine a method for summarizing the major dimensions of individual differences measured by the BCAS. Barclay (1972) factor analyzed 23 of the BCAS scales using samples of 1938 males and 1768 females. Analyses were done separately by sex and contrasted two methods of factor analysis: (a) principal axis with varimax rotation, and (b) multimethod factor analysis devised by Jackson (1969) to control for method variance. As was expected in the principal axis method, self-report scales, peer-nominations scales, and teacher-rating scales formed separate factors. The multimethod factor analysis of Jackson appeared to be the most valuable one for determining how the three sets of inputs related to common factors. In this procedure, within-method variance is minimized in favor of between-method variance. From this procedure six multimethod, multitrait factors emerged: taskorder achievement, impulsivity, reserved-internality, physical energy, sociability, and enterprising-dominance behavior. A second-order factor analysis of these multimethod and multitrait factors led to two factors: energy and sociability. When these components were reduced to vertical and horizontal axes, a grid emerged wherein individuals could be placed in terms of the values of their first-order factors. Some 64 combinations of the first-order factor scores have been plotted and these are depicted in the manual of the BCAS. To illustrate the hierarchical development of the overall system, a number of figures have been prepared. Figure 3.2 provides the final output of the system (or typology) indicating the loading of the second-order factors of energy and sociability on the type. The four quadrants represent four types of children who through many analyses have been characterized as thinkers, leaders, followers,

96

BARCLAY

IV TYPE

Thinkers

Ill 2nd ORDER FACTORS:

Energy

Leaders

Sociability

111 st ORDER FACTORS:

+ Task Order Ach.

FIG. 3.2.

II lmpu!sJvity

Ill Reserved Internal

IV Phy. Energy

v

Sociability

VI

Enterprising

First- and second-order components of temperament.

and agitators. These terms are not used in any pejorative labeling sense, but rather as summary descriptors of types of behavior perceived by peers and teachers and by unique personal attributes in self-report. In addition, it should be pointed out that there are two other groups of children who represent solid citizens of the classroom and are positioned in the middle of the grid. For further clarification, Figs. 3.3-through 3.8 diagram each of the multimethod multitrait primary factors of the BCAS. Direct inputs into the factor are indicated by the term "primary" and correlational characteristics derived from other research and particularly comparisons with Bennett's study (that are discussed following clarification of the figures) are identified as "associative." What is important to recall here is that no child is ever positioned in one of these quadrants simply by one method alone, but is so classified by factor loadings that represent subtle but important contributors to this typology. This fact is an important consideration in terms of both interventions and outcome evaluation studies. The results of a number of studies indicate that interventions must be type specific and evaluations must focus on changes in the type, not changes in the classroom. Where changes do occur, particularly on factor scores, there is a reflection not only of self-report changes, but changes in peer acceptance and affiliation, and changes in teacher attitude. As is true in other factor analytic studies, specific scales load on these major factor scores, and thus the entire assessment leads to composite judgments about

I

l! Factor Score

i· . . .

E.BIMAB:i

I Scale

I

TASK-ORDER ACHIEVEMENT

Interest in Artistic Occupations (ARTS) Interest in Intellectual Occupations (!NT) Competencies in Artistic and lnlellectual Skills (SAl)

. . . . . . 0

• Peer Nominations in Artistic and lnte!lectual Skills (GAl)

.. Teacher

Ratings of Stab!e-E xtroverted Behavior (SAN)

.

dSSQC!di!::iE Extroversion Neuroticism Contentiousness Self· Evaluation Anxiety Motivation Asocial Behavior Conformity Above Average Ach . on Standardized Tests

• Predictive of Nelson Reading Test and lTBS Scores • • Predictive of Stanford Reading Scores NOTE: 0 "" No loading or correla!Jon: + means IS associated with or has a significant Ioad1ng on this factor: · s1gn the oppos1te.

FIG. 3.3 Primary and associative components of factor I task order achievement as contributors to both second-order factors of energy and sociability.

I

I! Factor Score

IMPULSIVITY

I 8SSOCIAT!VE

1 Scales

Self Social Competencies (SSq Interest in High Status Occupations (ST) Peer Nominations of Reticent Behavior (GR) Peer Nominations of Disruptive Behavior (GD) Teacher Ratings of Unstable Extroverted Behavior (CHL)

Extroversion Neuroticism Contentiousness Sell Evaluation Anxiety Motivation Asocial Behavior

4

Conformity Behavioral Ratings of Disruptive Behav1or

NOTE:

0 "" No loading or correlation; + means is associated with or has a significant loading on this factor; - s1gn the opposite.

FIG. 3.4. Primary and associative components of factor II impulsivity as contributors to both second-order factors of energy and sociability.

97

98

BARCLAY

i

!I Factor Score

I RESERVED- INTERNAL

ASSOCIATIVE

1 Scales

Competencies in out-door Mechanical Skills (SAM)

Extroversion

Interests in Out-Door Mechanical Occupations (REAL)

• Interests in Conservative vs. Bold Occupations (CVR) Peer Nominations in Outdoor Mechanical Skills {GAM) Peer Nominations in Reticent-Shy + Behavior (GR) Teacher Ratings of Unstable-Withdrawn Beh.

Neuroticism Contentiousness Self-Evaluation Anxiety Motivation Asocial Behavior Con!ormtty

(MEL)

Negatively Predictive of Stanford Arithmetic and Language Scores NOTE:

a = No

loadings or correlations; + means is associated with or has a signlicant loading on This Factor; - sign the opposite.

FIG. 3.5. Primary and associative components of factor Ill reservedinternal as contributors to second-order factors of sociability and ener-

gy.

the overall as well as the specific characteristics of the individuaL Details of this process are provided in the manual of the BCAS. One of the major validations that can occur with a system such as this is comparison with another similar but totally independent system. In 1976, Bennett, in England, collected data on a number of children. 2 He used achievement tests, the Junior Eysenck Personality Inventory (Eysenck, 1965), self-report

2 A further commentary of a personal nature needs to be made on this matter. I was in England at the time of the publishing of the Bennett report. Because of the highly similar findings to my own research, I contacted Bennett who immediately invited me to visit him at the University of Lancaster. When I arrived there, he showed me a worktable where he had placed the earlier manuals of the BCAS and some of my research papers. He told me he had studied my system and had considered using it. Because he felt that the American system might not be appropriate for British children, he conceptualized another system based on the same inputs. The fact that he was able to construct an equivalent system and obtain very similar findings, is a strong argument for the construct validity of the BCAS system itself as well as a workable example of the application of the multimethod, multitrait approach in assessment. The fact that Wu (Barclay & Wu, 1980) could obtain similar results in Taiwan, plus the results of Bennett's study should suggest to future researchers that new systems could be developed the same way.

3.

SOCIOMETRY, TEMPERAMENT, AND SCHOOL PSYCHOLOGY

I

1! Factor Score

PHYSICAL ACTIVITY

I Scales

99

I

ASSOCIATIVE

Interests In Intellectual Occupations (!NT)

Interests in Social Occupations (SOC) Interests in Coventional and Clerical Occupations (CO NV)

Peer Nominations in Outdoor Mechanical Skills (GAM) Peer Nominations in Reticent-Shy Behavior

Extroversion Neuroticism

Contentiousness Self Evaluation Anxiety Motivation

Asocial Behavior Conformity

(GR)

Peer Nominations in Disruptive Behavior {GO) Teacher Ratings of Stable-Introverted BehaYior (PHL)

NOTE:

0 = No loading or correlation; + means is associated with or has a significant loading on this factor; - sign the opposite.

FIG. 3.6 Primary and associative components of factor IV physical activity as contributors to both second-order factors of energy and sociability.

data, teacher ratings, and a sociometric device. In the sociometric measure, Bennett asked students to nominate those peers who were most like or least like a series of descriptions of behavioral stereotypes (e.g., "the children that I most often work with"). He also collected achievement data. From all these data, Bennett identified 8 cluster types which show a considerable degree of overlap with the BCAS types. Given the similarity between measures and groups, the results of the two systems were compared using rho correlations of rank. There were many significant correlations (the full tabular accounts of these correlations are reported in the manual (Barclay, 1983a, pp. 94-95). Thus, from a comparison between the BCAS and a similar but less extensively researched battery used by Bennett, there are agreements between types of students identified, relationships between similar personality factors of student subgroups, and mean gains in achievement. This finding provides unique evidence for construct validity and the generalizability of BCAS descriptive categories across culture and measures (Bennett, 1976).

100

BARCLAY

- Energy

I

II Factor S core

I Scales

i i.

Sociability

SOCIABILITY

ASSOCIATIVE

~ Competencies in Socia! Primary Skills (SSC)

+

Interests in Status Occupations (ST)

+

Peer Nominations of Artistic-

+

+

Peer Nominations of Social Skills (GSC)

+

Contentiousness +

Extrovert Behavtor (SAN)

*

• *

Self-Evaluation Anxiety

+

Motivation

+

Conformity

+

Above ave. Ach. on Standardized Tests

Asoctal Behavior

Teacher Ratings of UnstableWithdrawn Behavior (MEL)

Teacher Ratings of Stable-

Extroversion

Neuroticism

Intellectual Skills (GAl)

Teacher Ratings of StableIntrovert Bel'1avtor (PHL)

I

+

+

Predictive of Nelson Reading and ITBS scores.

Predictive of Stanford Reading scores.

(1) Loadings on second-order factor of Energy-Activity are reversed NOTE: 0 = No loading or correlation; + means is associated with or has a Significant loading on this factor; - sign the opposite.

FIG 3.7. Primary and associative components of factor V sociability as contributors to second-order sociability and energy factors (1 ).

TEMPERAMENT AND SOCIOMETRY In recent decades the recognition that temperament is a formidable biologically based variable in the formation of individual differences is reasonably established. Temperament is biologically based (Allport, 1937; Strelau, 1983), heritable, but not exclusively genetic (Buss & Plomin, 1975). Current descriptions accurately depict the old Galen dimensions divided along stable-unstable, and extroverted-introverted characteristics (Eysenck, 1965). Temperament is present in infant character and remains relatively stable across the lifespan (Kagan, 1980; Thomas & Chess, 1977). It is explanatory of differences in the approach-withdrawal continuum, reactivity, excitation and inhibition and differential styles for coping with stress, isolation, and excitement (Barclay, 1983b, 1987; Strelau, 1983, 1988). What seems to be emerging from all of this research is that temperament is a construct of great importance that lies behind the facade of observable traits such as sociability, aggression, and flexibility. Temperament is viewed as a biologically endowed set of templates or filters that predispose, but do not

3.

SOCIOMETRY, TEMPERAMENT, AND SCHOOL PSYCHOLOGY

l

ll Factor Sea re

I Scales

r ENTERPRISING-DOMINANCE

EBlMABY Competencies in Enterprising Skills {SE)

Peer Nominations in Enterprising SkH!s (GE)

+

+

Extroversion Neuroticism Contentiousness

+ +

Self Evaluation Anxiety

Motivation

+

Teacher Ratings of U nstable Withdrawn Behavior (MEL)

0

Asocial Behavior

+

Piers-Harris Self-Concept

Conformity

4

Teacher Ratings of Stable- Extroverted Behavior (SAN)

I

ASS:QC!eiiYE

Interest in outdoor and Mechanical Skills (REAL)

Peer Nominations in Social Skills (GSC)

101

+

NOTE: 0 = No loading or correlation; + means 15 associated with or has a significant load1ng on this factor; s1gn the oppos1te.

FIG. 3.8. Primary and associative components of factor VI enterprising-dominance as contributors to both second-order factors of energy and sociability.

determine the behavior of individuals as they respond to both internal and external stimuli. It is seen as more concerned with the how of behavior than the what of behavior. It is viewed as a crucial factor in all emotional life, even though, as Goldsmith remarked in a recent plenary conference on temperament, we cannot define it more absolutely than we can the construct of intelligence (Goldsmith, Buss, Plomin, Rothbart, Thomas, Chess, Hinde, & McCall, 1987). Within the range of current temperament research, three trends are of particular importance to school psychology: (a) the explanation of physiological factors as related to behavior, (b) the consistency of factor analytic research in identifying the major dimensions of temperament, and (c) the impact of limited temperament-treatment-interaction studies on intervention outcomes. With regard to the explanation of physiological factors, Jan Strelau has enunciated an integrative theory of temperament which he terms the "regulative" theory of temperament. Based both on Pavlovian concepts and Western research drawn from Eysenck, Gray and others, he distinguishes two major types of individuals derived from temperament differences. One is termed "high reac-

102

BARCLAY

tive" and is characterized by high sensitivity to stimulation, and low endurance. The other is called "low reactive" and is characterized by low sensitivity to stimulation, and high endurance. These two types differ from each other in considerable ways with high reactives becoming confused and overstimulated by strong bursts of stimulation, and low reactives becoming bored and even anxious by deprivation of a continued flow of stimulation. In enunciating this theory, Strelau draws upon Hebb's notion of optimal level of arousal (Hebb, 1955), and Gray's ideas about arousability (Gray, 1964, 1972). Strelau believes that the two major physiological forces central to individual differences are cortically based in activity and reactivity (inhibition). Strelau sees the core elements of temperament thus as activity and reactivity which are "relatively stable features of the organism, primarily biologically determined as revealed in the formal traits of reactions that form the energy level and temporal characteristics of behavior" (Strelau, 1983, p. 171 ). Basically, high reactives and low reactives tend to differ in levels of activity and reactivity or inhibition. For example, if energy-activity can be represented by a river, high reactives are threatened by too fast a flow, and thus set up dams (via reactivity or inhibition). They are eager to maintain a constant flow of stimulation (and much of it may originate in internal mechanisms), but they cannot cope with high levels of stimulation, particularly if they cannot control the level. Low reactives on the other hand not only require a higher level of external stimulation, but enjoy it. Kagan recently reviewed literature and research with children relating to excitation and inhibition (Kagan, 1989). He declared that the forces of inhibition and excitation can be viewed operationally in retiring children and extroverted ones. He found that there is a genetic base related to blood factors and saliva in these two types of children. He has indicated that the continuing research that has been done both in the specific area of temperament and with monozygotic twin studies confirms that individual differences are not only a result of cognitive abilities, but are influenced by central emotional filters that specify both the conditions and the amplitude of arousal, sensation-seeking, and inhibition. Though admittedly these components exist and do influence human thinking and behaving, they are not easily measurable as Kagan, Strelau, and Eysenck have indicated by their use of physiological, chemical, blood, and drug measures. In point of fact, the situation is analogous to the intelligence studies where the "g" factor is considered a primary component, but difficult to measure. With school children, and particularly masses of children, the major methods of assessing temperament will have to be related to parental and teacher or psychologist observation, and self-report from inventories. Temperament has been measured by a variety of inventories where temperament variables are considered as test traits. Some current examples of such tests and research reported with them are the Dimensions of Temperament Survey (DOTS) (Lerner, Palermo, Spiron, & Nesselroade, 1982; Windle, Hooker, Lenerz, East, Lerner, & Lerner, 1986), The Temperament Assessment Battery (Martin, 1984; Martin,

3.

SOCIOMETRY, TEMPERAMENT, AND SCHOOL PSYCHOLOGY

103

Paget & Nagle, 1983;, Martin & Holbrook, 1985; Martin, Drew, Gaddis & Moseley, 1988), and the EASI (emotionality, activity, sociability, impulsivity) scale of Buss and Plomin ( 1975). In addition, the scales of Thomas and Chess (1975) have been used to assess early child development characteristics. These tests are similar, but perhaps more specifically oriented to temperament than many other general personality test batteries. Another approach is to view temperament variables as factors. This has the advantage of reducing a number of trait variables that may or may not truly represent temperament to a few factors. Thus, the second aspect of research to be cited here relates to the use of factor-analysis as a method to abstract from traits the overall primary factors. Royce (1973) pioneered this approach doing multiple factor analyses of existing factor structures from a number of research studies. Basically, he was looking for the invariant secondary or tertiary factors that related to primary ones and hence, to traits. The writer, following the research of Stroup and Mandersheid ( 1977) who had found that both the 16 Personality Factor Test and the California Psychological inventory loaded on common second factors, administered the above two tests and the Strelau Temperament Inventory to a group of 86 freshmen students. The Strelau also loaded on the same factors, suggesting that all three tests, the first based on factor analysis, the second on empirical classification, and the third on physiological studies all were related to the same factors (Barclay, 1987). The commonality of these factors drawn from three different tests, together with the previous factor studies of the writer with the BCAS, and work done by Steger with three self-concept measures, suggested strongly that factor analytic methods over different tests could yield dimensions similar to high reactive and low reactive individuals as outlined by Strelau. In addition, some preliminary work by Sears and Barclay ( 1989) with different types of habitual drug users, suggested similar results. No matter how significant or impressive the foregoing chains of research may be, the real need in temperament research is to establish the fact that different temperament groups respond differently to treatments. McCord and Wakefield (1981), using Eysenck's theoretical framework relative to introverts and extroverts, hypothesized that elementary school children classified as extroverts would learn mathematics principles easier through a good deal of stimulation and praise. Conversely, they believe that a mild form of threat would increase mathematics learning in introverts. Introverts in Eysenck's approach correspond roughly to the higher reactives of Strelau, and extroverts correspond to low reactives. They reported significant results in this study with introverted or high reactive children gaining more in arithmetic skills through mild teacher threat, and extroverts or low-reactive children gaining more in arithmetic skills through praise. Barclay (1983b) reported 6 studies that were done on a pre-post basis with the BCAS. These included use of traditional learning, behavioral approaches, the "open classroom," and mastery learning. In addition, two different methods of social intervention including an existential and discipline approach were evalu-

104

BARCLAY

ated. The core argument of this study was that when pre-post results for each of these studies using the BCAS as the criterion measure were evaluated, there were nonsignificant results when total classroom groups were compared pre-post. However, when the temperament groups were analyzed pre-post, there were many significant results suggesting that different treatments affected different temperament groups in a significant way. To detail this study would take the argument of this section too far afield, but in general the conclusions suggested that high reactive children (high sensitivity, low endurance) as determined by BCAS group 1 were more field-dependent, object-oriented and profited from small group problem-solving, individual research and summative organizer techniques of learning. Low reactive children (low sensitivity, high endurance) as determined by BCAS group 2 are often field-independent, people-oriented, and profit from a method that takes learning in smaller steps, but with considerable interaction from others. Low reactives tend to do better in learning situations that have social relevance, whereas high reactives tend to do better in learning situations that have intrinsic personal relevance. High reactives tended to do better in small structured seminars, independent study, computer learning and typical lecture-recitation sessions. Low reactives, particularly those identified as impulsive (BCAS group 4) tended to have problems in behavior control, cognitive motivation, and poor attitude. For these children, social reinforcement, and mastery learning appeared to be very relevant approaches. Further insight into high and low reactive behavior can be found in the work both of Hart and Wainwright, and Ban dura. Hart ( 1980) in developing postulates of sociometry focuses his theory on the need for affiliation that all individuals show in one way or another. Wainwright ( 1980) maintains that observation often focuses on specific bodily movements, i.e., spontaneity, verbal styles, nonverbal communication and kinesthetics, and the chemistry of speech production and facial expressions. These products of perception are reflective of emotional styles of thinking that reflect central nervous system components. Thus, the field space, and the particular valence of an individual on another individual is a direct consequence of temperament characteristics interpreted in perception. The role of social learning in this complex equation has been clarified by Bandura (1978) wherein he commented on the correction of the Lewinian paradigm of personality by suggesting that the effect of past learning influences both environmental and personality factors. However, the consistency of sociometric choices indicates that personality factors do not change very much unless substantial interventions are implemented. This means that within the predominant temperament pattern, stable learned methods need to be patterned and reinforced. When behavior patterns do change, they are more like stable and socially relevant variations on a theme than a new motif. The core elements relate to both shy-retiring and socially withdrawn or immature behavior, and excitable-aggressive behavior, more commonly generalized as inhibition and excitation. Cor-

3.

SOCIOMETRY, TEMPERAMENT, AND SCHOOL PSYCHOLOGY

105

relational studies between sociometric indices of social power and Strelau's excitation show a moderate positive correlation, but correlations between Strelau's index of excitation over inhibition (showing individuals whose excitation score is higher or lower than inhibition) with sociometric indices is highly positive and significant (Barclay, 1988a). Given the foregoing discussion of the biological components of temperament, it appears that sociometric choices really tend to identify the components of excitation, inhibition and mobility found by Strelau (1983). Children at a very early age are able to detect through their own perception of reality individuals who are active and energetic versus passive and phlegmatic. They are also able to ascertain individuals who are restricted and inhibited in outlook, fearful of engaging in any new activity. Thus the suggestion is made here that sociometric phenomena are naive, but highly valid indicators of temperament as they assess the quality and intensity of excitation, inhibition, and mobility. In summary, aside from the more recent research on the biological underpinnings of sociometry and relationship to social learning, it would appear that sociometry can be linked to temperament: (a) on the basis of the almost identical factor structure obtained within the BCAS system to those structures obtained for adults (Barclay, 1987); (b) from the connection obtained between direct measures of high and low reactivity as indicated in correlations between excitation, inhibition and mobility on the one hand and sociometric data on the other; and (c) finally, from preliminary evidence obtained for differential temperament-treatment outcomes (Barclay, 1983b). PERSISTENT PROBLEMS WITH THE USE OF SOCIOMETRICS

Although it is apparent from this review that sociometries provide a powerful independent and supplementary source of assessment of children, and one that adds substantially to the variance in measurement received from teacher judgments and self-report, it is equally true that it is seldom used in school psychology. Nonetheless, there is a resurgence of interest in sociometries in early childhood: specifically in preschool and kindergarten age children. There was a special issue of the Merrill-Palmer Quarterly (1983) devoted to the topic (see for example, Coie & Dodge, 1983; Dodge, Schlundt, Schoken, & Delugach, 1983; Hymel, 1983; Ladd, 1983; Rubin & Daniels-Beirness, 1983; and Cairns, 1983). Other research relating to this topic is reported by McConnell and Odom ( 1986), and Bullock, Ironsmith, and Poteat (1988). In the literature of school psychology, however, the reasons for its lack of use have been outlined earlier, and may also include failure to read child development literature by school psychology researchers. There are, however, three other issues that mediate the use of sociometries by school psychologists. They

106

BARCLAY

are: (a) ethical considerations, (b) the generalized distrust of group data in school psychology practice and failure to recognize computer developments in this area; and (c) the lack of recognition of the importance of temperament in school psychology interventions designed for prevention. Ethical issues in the use of sociometry were never believed to be an issue in earlier decades simply because there was a more generalized trust of school psychology in the schools and research was more acceptable without current legal prescriptions. In the 1950s and 1960s most users believed that negative questions should either be very carefully worded or not used. However, in the last few decades there has been an increase in caution for the use of personality type tests with school students. Most schools are reluctant or have specific prohibitions against the use of personality tests such as the MMPI with students. Junior versions of these tests are likewise suspect. Parents, society, and the law have all looked at some personality tests as the invasion of the rights of students and parents. Where do sociometric phenomena fit into the overall question of assessment of students? Specifically is the sociometric device a personality probe? Unquestionably, the response to this question is what the specific sociometric item seeks. For the most part sociometric indices do not tap personality traits in the sense of a test inventory, but they can be used to generate relevant temperament data that appear to be anchored in competence. Relative to general friendship or dislike questions, Cairns (1983) views the issue in terms of what is gained and what is lost by certain techniques. He does not believe that children should be encouraged to make up semipublic hate lists or slam books for whatever reason. But he points out that failure to use the sociometric question may deprive us of some very important information about social structure. Certainly, this view is consistent with much earlier thinking. He suggests that there is "frightfully little empirical information" on what are the effects of sociometric studies on both students and teachers. There is some information about the effects on students relative to the sociometric technique. Ratner, Weissberg, and Caplan (1986) conducted a study of 32 6th graders 2 months after an assessment battery that included a class sociometric rating scale. Each individual was interviewed about his/her reactions to these measures. Specifically they examined the students' reactions to the sociometric measure, the degree to which ratings were discussed among classmates and what might have been the differential impact of sociometric testing on popular and unpopular children. Contrary to their expectations, they found few negative reactions to the sociometric ratings. Many of the students remembered the sociometric ratings and reported that the measure was discussed afterwards. However, only 3 of the 32 subjects liked the sociometric the least of all the measures. None of these students mentioned ethical or social considerations such as the potential negative effects of giving or receiving low rating. Additionally, there were no significant differences between popular and

3.

SOCIOMETRY, TEMPERAMENT, AND SCHOOL PSYCHOLOGY

107

unpopular classmates in the tendency to dislike the sociometric measure. Overall, these findings alleviate some concerns about the potentially negative effects of sociometric testing, but several clinical and research issues remain to be investigated. (Ratner et a!., 1986).

Barclay (1974c) collected information about the feedback of information derived from the BCAS to parents, teachers, and children. From studies done in several settings where parents viewed the computer reports of their children, many comments were obtained about the importance of this information. Teachers, in addition, have found the reports very useful in conducting case-conferences with parents particularly since the reports provide a basis for continuing advising whenever problems arise. The judgment suggested here relative to the ethics of sociometry is that it can be helpful in providing relevant social information and temperament data about children. The reports that are generated from the computer analysis deliberately avoid any negative connotations. Second, this kind of data does not directly probe into personality, but results in information that can provide insight into both social competence and temperament. Third, the information thus obtained is useful in a number of ways for determining interventions and in consulting with parents. Finally, the real important question is: Can such valuable data be obtained without the use of sociometries? At present it appears that sociometries provide data of a unique value and strength. Although the ethics question is not completely answered, it does appear that if sociometric data are used without negative references, they are not identical to personality testing, and they can provide good feedback information to all concerned. Moreover, they appear to provide a unique contribution to the identification of temperament level and intensity of arousal, reactions to stress, and emotional characteristics. Most uniquely they provide some important keys to intervention. Despite the power of sociometries, a serious continuing problem with their use is the need to assess all children in a given environment. Given the reluctance of public schools to do any testing other than achievement and intelligence testing, sociometries or any test based in part on sociometry must be used contingent with parental permission. Obtaining this permission is so difficult that in practical terms it rules out the use of the technique. Is there an answer to this difficult problem? The writer believes that the tremendous value of obtaining measures of social competence, skill development, and temperament dimensions far outweighs this technical handicap. If schools wish to obtain a clear indication of the social status, social competency, and temperament inclinations of children, a sociometric battery could simply be mandated as a part of the testing program. If the goals of the assessment, the potential use of the technique in early identification, and preventive strategies of intervention can be openly discussed with informed parent groups such as the PTA and teachers, it would seem that assessment of this type is at least as important as the assessment of intelligence and achievement. This then becomes a leadership initiative for school psychologists.

108

BARCLAY

A second problem is the generalized decline in the diagnostic use of group tests by school psychologists. Because of the emphasis on individual testing and behavioral assessment, for the most part the reservoir of group tests that are or could be used in the school setting to facilitate programmatic planning and intervention by school psychologists are ignored. In part this may be due to the fact that school psychologists tend to view group tests as possessing lower reliability and validity than individual tests. But it is more likely that they do not consider how the testing program of a district ought to be implemented into a systems approach to the early prevention of psychological problems. To be sure, the time constraints on school psychologists make it difficult to become involved in a systems approach to the assessment of individual differences, but if such a goal is part of the school's agenda, and particularly if prevention is viewed as an important goal for all children, then the use of group data as a preliminary screening vehicle is imperative. As is well known, achievement and intelligence tests are very frequently administered in most schools, but virtually no schools include other measures of emotional and temperament adjustment. Given the falling achievement levels of American children, the epidemic problems in drug usage, delinquency, dropouts, broken homes and other factors that effect learning, then the use of group tests and interpretation by computer methods is a resource that school psychologists ought to consider seriously (Barclay, 1983c ). Peer nominations are clearly one set of multiple observations that can and should be used for amassing the evidence needed for clear intervention strategies. Peer data provide an analysis of the psychological support system available to a child that cannot be obtained in any other manner, and this information is vital to determining whether a child needs intervention planning. Insofar as peer nominations also relate to temperament diagnoses they represent a doubly important means for prevention as well as remediation. THE FUTURE OF SOCIOMETRIC ASSESSMENT

From the foregoing analysis, it would appear that sociometric assessment provides a competency and temperament evaluation of individual differences that is different in kind from the traditional self-report or observer evaluation by teacher or parents. This type of assessment utilized with self-report, teacher judgment, and achievement/intelligence data provides a comprehensive evaluation of individual differences not only on selected individuals, but entire classrooms and schools. 3 3 Although the BCAS requires the inclusion of all children in a classroom, work has been done on an experimental version entitled "template" that requires only a structured interview with the teacher and/or parent plus a student input. This form predicts reasonably well the final output of the BCAS factor scores and can be used with individuals (Barclay, 1988b ).

3.

SOCIOMETRY, TEMPERAMENT, AND SCHOOL PSYCHOLOGY

109

The use of sociometries in future school psychology assessment is highly dubious if the focus of school psychology remains fixed almost exclusively on handicapped or exceptional children. There are such deficits in many of these children that one does not need the confirmation of sociometries or temperament analysis. If the focus changes in school psychology to be concerned with national and even international issues of increasing achievement or decreasing obstacles to learning such as drug usage then the technique is highly valuable. However, the time constraints of psychological practice in the schools simply rule out the use of the older paper-pencil and hand-scored approaches. Computer technology can and must be used for such analysis as Barclay (1983c) and Langeheine (1978) have demonstrated. In addition, outcomes of such testing must provide quick indices of reference for identifying children who are beginning to show problems. It is with this group that interventions can be most effective, before habit patterns have reached such a chronic stage of problems that extreme crisis measures are needed. Finally, it is possible to convert both sociometric and achievement type data into school and district analyses to determine where the highest priorities may be (Barclay, 1974b). In this manner critical foci can be determined within a district for the deployment of whatever teaching, learning, and other resources are needed. To do this will require more technological improvement in assessment, and the cooperation of schools in district-wide research. It is timely to do this. Summary This chapter has reviewed the nature of sociometric choice and its relationship to school psychology practice. The conclusions are: (a) sociometric choices covary and are related to a number of personality group measures and individually-administered instruments; (b) sociometric choices are strongly related to ratings by social workers, psychologists, military personnel, industrial and educational ratings of efficiency, and competence; (c) sociometric choices are moderately related to the structure of intelligence and achievement tests; (d) there are positive correlations between measures of motivation and selfconcept or self-competency and to socioeconomic status; (e) sociometric choices are strongly related to behavioral observations of children in the classroom; (f) sociometric ratings taken with other assessment inputs can provide a computerized assessment of individual differences with many implications for strategies of intervention; and (g) the use of computerized sociometric judgments joined to academic scores

110

BARCLAY

can provide an overall estimate of the personal and social status of every child in the elementary classroom. This provides an unparalleled opportunity for school psychologists to mrtlate both curricular and personal-social strategies of intervention with both school personnel and parents.

REFERENCES Allport, G. W. (1937). Personality: A psychological interpretation. New York: Holt. Amidon, E. (1963). The isolate in children's groups: changing his sociometric status. In J. M. Seidman (Ed.), Educating for mental health. New York: Thomas Y. Crowell. Bandura, A. (1978). The self system in reciprocal determinism, American Psychologist, 33(4), 344-358. Barbe, W. B. (1954). Peer relationships of children of different intelligence levels. School and Society, 80, 60-62. Barclay, J. R. (1956). Unpublished data, Redford Union School District, Detroit, Michigan. Barclay, J. R. (1964). Studies in sociometry and teacher ratings: A diagnostic and predictive combination for school psychology, lithographed. Pocatello: Idaho State University. Barclay, J. R. (1966a). Sociometry: Rationale and technique for effecting behavior change in the elementary school. Personnel & Guidance Journal, 44, 10,1067-1076. Barclay, J. R. (1966b). Sociometric choices and teacher ratings as predictors of school dropout. Journal of School Psychology, 4,(2), 40-44. Barclay, J. R. (1966c). Interest patterns associated with measures of social desirability: Some implications for dropouts and the culturally disadvantaged. Personnel & Guidance Journal, 45 ,(1), 56-60. Barclay, J. R. (1966d). Variability in sociometric scores and teachers' ratings as related to teacher age and sex. Journal of School Psychology, 5,(1), 52-58. Barclay, J. R. (1967). Effecting behavior change in the elementary classroom: an exploratory study. Journal of Counseling Psychology, 4,(3), 240-247. Barclay, J. R. (1972). The Barclay classroom climate inventory: a research manual and studies. Lexington, KY: Educational Skills Development, Inc. Barclay, J. R. (1974a). A preliminary evaluation of Elyria's experiment in affective education. Unpublished report, University of Kentucky. Barclay, J. R. (1974b). System-wide analysis of social interaction and affective problems in schools. In P. 0. Davidson, F. M. Clark, & L.A. Hamerlynck (Eds.), Evaluation of behavioral programs in community, residential and school settings, Champaign, IL.: Research Press. Barclay, J. R. (1974c). What teachers, parents students and professionals say about the Barclay Classroom Climate Inventory, unpublished paper, University of Kentucky. Barclay, J. R. (1983a). Manual Barclay Classroom Assessment System Los Angeles: Western Psychological Services. Barclay, J. R. (1983b). A meta-analysis of temperament-treatment interactions with alternative learning and counseling treatments, Developmental Review, 3 (November), 410-443. Barclay, J. R. (1983c). Moving toward a technology of prevention: A model and some tentative findings, School Psychology Review, /2,(3), 228-239. Barclay, J. R. (1987). The Strelau temperament inventory as a broad classification system. Archives of Clinical Neuropsychology, 2, 307-327. Barclay, J. R. (1988a). Temperament as a typology, unpublished paper, University of Kentucky.

3.

SOCIOMETRY, TEMPERAMENT, AND SCHOOL PSYCHOLOGY

111

Barclay, J. R. (1988b). The computer as a diagnostician for interventions: Implications for school psychology, manuscript in preparation, University of Kentucky. Barclay, J. R., Barclay, L. K., & Stilwell, W. E. (1972). Influence of paternal occupation on social interaction measures in elementary school children, Journal of Vocational Behavior, 2, 433-446. Barclay, J. R., Covert, R. M., Scott, T. W., & Stilwell, W. E. (1975). Some effects of schooling: A three-year follow-up of a title III project. Lexington, KY: Educational Skills Development. Barclay, J. R., & Kehle, T. J. ( 1979). The impact of handicapped students on other students in the classroom. Journal of Research and Development in Education, 12, 80-91. Barclay, J. R., Stilwell, W. E., Santoro, D. A., & Clark, C. M. (1972). Behavioral and achievement correlates of social interaction variables in the elementary classroom. Unpublished manuscript, University of Kentucky. Barclay, J. R., & Wu, W. (1980). Classroom climates in Taiwanese and American elementary classrooms: A cross-cultural study. Contemporary Educational Psychology, 5, 65-82. Barclay, J. R., Phillips, G., & Jones, T. (1983). Developing a predictive index of giftedness, Measurement and Evaluation in Guidance, 16,(1), 25-35. Barclay, L. K., & Barclay, J. R. (1965). Measured indices of perceptual distortion and impulsivity as related to sociometric scores and teacher ratings, Psychology in the Schools, 2,(4), 372-375. Barclay, L. K., & Barclay, J. R. (1986). PACE: A computer-based instrument for assessing and developing learning skills. Champaign, IL: Metritech, Inc. Bedoian, V. H. (1953). Mental health analysis of socially over-accepted, socially under-accepted, overage and under-age pupils in the sixth grade, Journal of Educational Psychology, 44, 366371. Bennett, N. (1976). Teaching styles and pupil progress. London: Open Books. Bjerstedt, A. (1956). The interpretation of sociometric status scores in the classroom, Nordisk Psychologi, 8, 1-14. Bonney, M. E. (1943a). The constancy of sociometric scores and their relationship to teacher judgments of social success and to personality self ratings. Sociometry, 6, 409-424. Bonney, M. E. ( 1943b). The relative stability of social, intellectual, and academic status in grades II to IV and the interrelationships between these various forms of growth. Journal of Educational Psychology, 34, 88-102. Bonney, M. E., & Powell, J. (1953). Differences in social behavior between sociometrically high and sociometrically low children. Journal of Educational Research, 46, 481-495. Bullock, M. J., lronsmith, M., & Poteat, G. M. (1988). Sociometric techniques with young children: A review of psychometrics and classification schemes. School Psychology Review, 17, 289-303. Buss, A. H., & Plomin, R. ( 1975). A temperamental theory of personality development. New York: Wiley-Interscience. Cairns, R. B. (1983). Sociometry, psychometry and social structure: A commentary on six recent studies of popular, rejected, and neglected children. Merrill-Palmer Quarterly, 29, No. 4, 429438. Cohen, D. W. (1983). The relationship of temperament clusters to causal attribution of academic success and perceived self competence in fourth, fifth and sixth grade children. Unpublished doctoral dissertation, University of Kentucky. Coie, J. D., & Dodge, K. A. (1983). Continuities and changes in children's social status: A fiveyear longitudinal study. Merrill-Palmer Quarterly, 29, 261-282. Coleman, J. S. (1961). The adolescent society. New York: The Free Press of Glencoe. Davis, D. (1967). The validity and reliability of a sociometric device, Unpublished master's thesis, Idaho State University. Dineen, M.A., & Garry, R. (1963). Effect of sociometric seating on classroom cleavage. In J. M. Seidman (Ed.), Educating for mental health. New York: Thomas Y. Crowell. Dodge, K. A., Schlundt, D. C., Schocken, 1., & Delugach, J. D. (1983). Social competence and

112

BARCLAY

children's sociometric status. The role of peer group entry strategies. Merrill-Palmer Quarterly, 29, 309-336. Eysenck, S. B. G. (1965). Junior Eysenck personality inventory (British Edition). London: University of London Press. Ferris, A. (1968). A descriptive analysis of the characteristics of elementary reticent males. Unpublished master's thesis, California State University at Hayward. Flanders, N. A., & Havumaki, S. (1963). The effect of teacher-pupil contacts involving praise on the sociometric choices of students. In J. M. Seidman (Ed.), Educating for mental health. New York: Thomas Y. Crowell. Forsythe, F., & Jackson, E. (1966). A comparison of upper elementary school childrens' responses on a vocational preference inventory and a semantic differential in relationship to a social desirability grid. Unpublished master's thesis, California State University at Hayward. Gallagher, J. J., & Crowder, T. (1957). The adjustment of gifted children in the regular classroom, Exceptional Children, 23, 306-312, 317-319. Goldsmith, H. H., Buss, A. H., Plomin, R., Rothbar, M., Thomas, A., Chess, S., Hinde, R. A., & McCall, R. B. (1987). Roundtable: What is temperament? Four approaches, Child Development 58, 505-529. Gray, J. A. (1964). Pavlov's typology. Oxford: Pergamon Press. Gray, J. A. (1972). Learning theory, the conceptual nervous system and personality. In V. D. Nebylitsyn and J. A. Gray (Eds.), biological bases of individual behavior. London: Academic Press. Gronlund, N. E. (1951). The accuracy of teacher judgments concerning the sociometric status of sixth grade students. Sociometry Monographs, No. 25. New York: Beacon House. Gronlund, N. E. (1955). Generality of sociometric status over criteria in the measurement of social acceptability. Elementary School Journal, 56, 173-176. Gronlund, N. E. (1959). Sociometry in the classroom. New York: Harper Bros. Guinouard, D. E., & Rychlak, J. F. (!962). Personality correlates of sociometric popularity in elementary school children. Personnel & Guidance Journal, 40, 438-442. Hart, J. W. (1980). An outline of basic postulates of sociometry, Group Psychotherapy, Psychodrama & Sociometry, 33, 63-70. Hart, J. W., & Nath, R. (1979). Sociometry in business and industry: New developments in historical perspective. Group Psychotherapy, Psychodrama & Sociometry, 32, 128-149. Hebb, D. 0. (1955). Drives and the C.N.S. (conceptual nervous system). Psychological Review, 62, 243-254. Heider, F. (1958). The psychology of interpersonal relations, New York: Wiley. Holland, J. L. (1964). Personal communication. Hunter, J. E. (1978). Dynamic sociometry, Mathematical Sociology, 6, 87-138. Hymel, S. (1983). Preschool children's peer relations: Issues in sociometric assessment. MerrillPalmer Quarterly, 29, 237-260. Jackson, D. (1969). Multimethod factor analysis in the evaluation of convergent and discriminant validity, Psychological Bulletin, 72, 30-49. Jennings, H. H. (1950). Leadership and isolation, New York: Longmans, Green. Kagan, J. (1989). Temperamental contributions to social behavior, APA award address, American Psychologist, 44, 668-674. Kehle, T. J., & Guidubaldi, J. (1978). Effect of EMR placement models on affective and social development. Psychology in the Schools, 15, 275-282. Kennedy, D. (1969). Sociometric assessment, a validity study. Paper presented at the 1969 American Personnel and Guidance Association convention, Las Vegas. Kerr, M. (1945). A study of social acceptability, Elementary School Journal, 45, 257-265. Kounin, J. S. (1970). Discipline and group management in classrooms, New York: Holt, Rinehart & Winston.

3.

SOCIOMETRY, TEMPERAMENT, AND SCHOOL PSYCHOLOGY

113

Kounin, J. S., Gump, P. V. (1961). The comparative influence of punitive and nonpunitive teachers upon children's concepts of school misconduct. Journal of Educational Psychology, 52, 44-49. Kounin, J. S., Gump, P. V., & Ryan, J. J. (1961). Explorations in classroom management. Journal of Teacher Education, I2, 235-247. Kranzler, G. D., Mayer, G. R., Dyer, C. 0., & Munger, P. F. (1966). Counseling with elementary school children: An experimental study. Personnel & Guidance Journal, 44, 944-949. Kuhlen, R. G., & Collister, E. G. (1952). Sociometric status of sixth and ninth graders who fail to finish high school Educational and Psychological Measurement, I2, 632-637. Ladd, G. W. (1983). Social networks of popular, average, and rejected children in school settings. Merrill-Palmer Quarterly, 29, 283-307. Langeheine, R. (1978). Computer aided data analysis in sociometry, Educational & Psychological Measurement, 38, 189-191. Lerner, R. M., Palermo, M., Spiron, A., & Nesselroade, J. R. (1982). Assessing the dimensions of temperament individuality across the life-span. The Dimensions of Temperament Survey (DOTS), Child Development, 53, 149-159. Lewin, K. (1935). A dynamic theory of personality: Selected readings. Trans. D. K. Adams & K. E. Zener. New York: McGraw-Hill. Lewin, K., & Grabbe, P. (1945). Conduct, knowledge and acceptance of new values. Journal of Social Issues, I,(3), 56-64. Lindzey, G., & Borgatta, E. F. (1954). Sociometric measurement. In G. Lindzey (Ed.), Handbook of social psychology. Cambridge, MA: Addison-Wesley. Martin, R. P. (1984). The temperament assessment battery: Interim manual. Athens, GA: Developmental Metrics. Martin, R. P., & Holbrook, J. ( 1985). Relationship of temperament characteristics to the academic achievement of first-grade children. Journal of Psychoeducational Assessment, 3, 131-140. Martin, R. P., Paget, K., & Nagle, R. (1983). Relationships between temperament and classroom behavior, teacher attitudes, and academic achievement. Journal of Psychoeducational Assessment, I, 370-386. Martin, R. P., Drew, K. D., Gaddis, L. R., & Moseley, M. (1988). Prediction of elementary school achievement from preschool temperament: Three studies. School Psychology Review, I7, 125137. McConnell, S. R., & Odom, S. L. (1986). Sociometries: Peer-referenced measures and the assessment of social competence. In P. S. Strain, M. J. Guralnick, & H. M. Walker (Eds.), children's social behavior: development, assessment and modification. New York: Academic Press. McCord, R. R., & Wakefield, J. A., Jr. (1981). Arithmetic achievement as a function of introversion-extraversion and teacher-presented reward and punishment. Personality & Individual Differences, I, 145-152. Meacham, M. L., & Trione, V. (1967). The role of the school psychologist in the community school. In J. F. Magary (Ed.), School psychological services in theory and practice, a handbook. Englewood Cliffs, NJ: Prentice-Hall. Miller, R. V. (1956). Social status and socio-empathic differences among mentally superior, mentally typical, and mentally retarded children, Exceptional Children, 23, 114-119. Moreno, J. L. (1934). Who shall survive? Washington, DC: Nervous and Mental Disease Publishing Co. Moreno, J. L., & Moreno, Z. T. ( 1976). A sociometric view of recent history: The rise and fall of leadership, Group Psychotherapy, Psychodrama and Sociometry, 29, 63-69. Morse, W. C. (1965). Intervention techniques for the classroom teacher of the emotionally disturbed. Presented to the First Annual Conference on the Education of Emotionally Disturbed Children. Mouton, J. S., Blake, R. R., & Fruchter, B. (1955a). The reliability of sociometric measures, Sociometry, 18, 7-48.

114

BARCLAY

Mouton, J. S., Blake, R. R., & Fruchter, B. (1955b). The validity of sociometric responses, Sociometry, 18, 181-206. Northway, M. L. (1952). A primer of sociometry, Toronto: University of Toronto Press. Ratner, C., Weissberg, R., & Caplan, M. (1986, August). Ethical considerations in sociometric testing: The reactions of preadolescent subjects. Paper presented at the APA National Convention, Washington, DC. Red!, F., & Wineman, D. (1952). Controls from within: Techniques for the treatment of the aggressive child. Glencoe, IL: The Free Press. Royce, J. R. (1973). The conceptual framework for a multi-factor theory of individuality. In J. R. Royce (Ed.), Multivariate analysis and psychological theory. New York: Academic Press. Rubin, K. H., & Daniels-Beirness, T. (1983). Concurrent and predictive correlates of sociometric status in kindergarten and grade one children. Merrill-Palmer Quarterly, 29, 337-351. Sears, J., & Barclay, J. R. ( 1989). Exploratory study with chronic drug abusers, Lexington: University of Kentucky. Shapiro, E. S. ( 1987). Intervention research methodology in school psychology. School Psychology Review, 16,(32), 290-305. Smith, D. P. (1951). Teachers manual for sociometry. Lithographed paper, School of Education, University of Michigan. Smith, G. M. (1967). Usefulness of peer ratings of personality in educational research. (ERIC Document, ED 012 494). Stickel, N. (1968). A descriptive analysis of the characteristics of elementary reticent females. Unpublished master's thesis, California State University at Hayward. Stilwell, W. E., & Barclay, J. R. (1976). Effects of education through developmental guidance services: a one-year study. Stuttgart, AR.: Stuttgart School District. (ERIC Document No.ED. 133 077). Stilwell, W. E., & Barclay, J. R. (1977). Effects of an affective-social education program over two years. Stuttgart, AR. (ERIC Document No.ED. 143 425). Stilwell, W. E., & Barclay, J. R. (1979). Effects of affective education interventions in the elementary school, Psychology in the Schools, 16, 80-87. Strelau, J. (1983). Temperament, personality, activity. London: Academic Press. Strelau, J. (1988, November). The regulative theory of temperament. Lecture presented at the University of Kentucky. Stroup, A. L., & Manderscheid, R. W. (1977). CPI and 16 PF second-order factor congruence. Journal of Clinical Psychology, 33,(4), 1023-1026. Tapp, G. S., & Barclay, J. R. (1974). Convergent and discriminant validity of the Barclay Classroom Climate Inventory, Educational and Psychological Measurement, 34, 439-447. Thomas, A., & Chess, S. (1957). Temperament and development. New York: Brunner/Mazel. Tindall, R. H. (1955). Relationships among indices of adjustment status. Educational and Psychological Measurement, 15, 152-162. Wainwright, J. C. (1980). A framework for the observation of movements and sounds, Group Psychotherapy, Psychodrama and Sociometry, 33, 6-24. Windle, M., Hooker, K., Lenerz, K., East, P. L., Lerner, J. V., & Lerner, R. M. (1986). Temperament, perceived competence and depression in early and late adolescents, Developmental Psychology, 22,(3), 384-392.

4

Preparation of School Psychologists in Behavioral Consultation Service Delivery

Thomas R. Kratochwill

University of WisconsinMadison

Susan M. Sheridan University of Utah

Pamela Carrington Rotto

University of WisconsinMadison

Diane Salmon

National Louis UniversityEvanston

Practicing school psychologists have reported that consultation generally has been a high priority in service delivery approaches in schools and other applied settings (e.g., Gutkin, Singer, & Brown, 1980; Lesiak & Lounsbury, 1977; Meacham & Peckam, 1978). Although many practitioners now receive didactic training, few school psychology programs may actually provide a field practicum or other supervised experience in consultation. 1 Formal applied training in consultation is important for several professional and ethical reasons. In the last decade, there has been growing concern over the limitations of traditional psychoeducational services provided to children in school settings. For example, studies have demonstrated that psychologists using traditional assessment practices have major difficulties reliably identifying children for the provision of special educational services (e.g., Ysseldyke, Thurlow, Graden, Wesson, Algozzine, & Deno, 1983). Once referred, there is a high probability that students tested will be placed in special education, and there continues to be low relationship between the actual assessment protocol and the design, implementation, and monitoring of intervention programs. Thus, there has been growing interest in consultation practice as part of the "regular education initiative" (REI) which calls for radical reform in services to mildly handicapped children, some of !Meyers, Wurtz, and Flanagan (1981) surveyed national school psychology training programs and reported that among those responding, 55% of the doctoral and 32% of the subdoctoral programs offered a didactic course in consultation. However, 17% of the doctoral and 42% of the subdoctoral programs had no courses with significant focus on consultation. Furthermore, of those programs that offered consultation training, only 27% were reported to incorporate a field practicum component. Although this study was conducted over a decade ago, the picture has likely not changed dramatically.

115

116

KRATOCHWILL, SHERIDAN, ROTTO, SALMON

whom receive services in regular education settings. Recommendations for the inclusion of consultation as a part of the intervention process in the regular education setting supports the growing interest of training in this area (Cancelli & Lange, 1990). Although it is beyond the scope of our chapter to provide an overview of concerns pertaining to the REI, the interested reader is referred to Kauffman (1991 ), and a mini-series edited by Rosenfield (1990) that reviews issues in this area. Training in consultation often is seen as a viable option for provision of services and as a strategy for linking assessment directly to prevention and intervention. Specifically, numerous intervention strategies have been presented for implementation within a consultation service delivery role (see Graden, Zins, & Curtis, 1988; Price, Cowen, Lorion, & Ramos-McKay, 1988). The important issue here is that consultation as a process and intervention might be held to the same scrutiny as traditional service delivery approaches. For example, standardization of the protocols, developing the psychometric features of the interview process (i.e., reliability, validity), and establishing the treatment validity of the approach represent a spectrum of standards that behavioral consultation researchers should have to address (Bergan & Kratochwill, 1990). Competencybased training represents the foundation for conducting consultation with integrity and hopefully, addressing standards for practice. Specific training in consultation and intervention strategies has also taken on ethical importance within the context of children's right to effective treatment. Thus, adequate preparation in consultation services is considered important in preserving children's right to effective services. Individuals with specific competencies are potentially more capable of delivering high quality psychological services. These are but a few of the major concerns that have prompted more careful consideration of consultation training research, and which have facilitated growing interest in methods to train preservice and inservice school psychology consultants effectively. The purpose of our chapter is to provide an overview of the training literature in consultation, with a specific focus on training behavioral consultants. Broad coverage of training consultants, clinicians, or counselors is presented in other sources and the interested reader is referred to these writings (e.g., Ford, 1979; Friend & Cook, 1988; Gallessich, McDermott-Long, & Jennings, 1986; Idol, 1988; Kurpius & Lewis, 1988; Rosenfield & Gravois, in preparation). The chapter includes an overview of research in various areas of consultant training, and advantages and limitations of this work within a theoretical and practical context. Finally, future directions are presented for research in the field of behavioral consultation training. CURRENT RESEARCH IN TRAINING CONSULTANTS

Although we do not focus extensively on issues pertaining to terminology, consultation is defined as an indirect form of service delivery in which the consultant

4.

PREPARATION OF SCHOOL PSYCHOLOGISTS

117

works through a consultee or mediator to effect some change in the client, usually a child or adolescent. A variety of individuals may serve in the consultant, consultee, and client roles (see Bergan & Kratochwill, 1990 and Milne, 1986, for examples). Like other forms of consultation, behavioral consultation has as its single most definitive characteristic, indirect service delivery (see Gutkin & Curtis, 1990 for a review of the core characteristics of consultation). We view behavioral consultation as an indirect service delivery model in three categories (Vemberg & Reppucci, 1986). The first and most familiar, is behavioral case consultation, in which the consultant interacts with mediators or consultees who treat the client(s). The most procedurally developed model of case consultation is that presented by Bergan and Kratochwill ( 1990). Behavioral system consultation involves an analysis of the process and structure of a social system (e.g., school) using behavioral technology with the goal of improvement of the system. Although perhaps less familiar to readers, the systems approach presented by Maher ( 1981) represents an analogous sequence for system consultation. ~'The key distinction between behavioral system consultation and behavioral case consultation lies in the focus of the process and structure of a system per se in the former versus a focus on a particular individual's (or group of individuals') relation to a system in the latter" (Vemberg & Reppucci, 1986, p. 59). The third form of behavioral consultation involves behavioral technology training, and is used to increase skill levels in mediators or consultees who interact with clients. In school psychology the most common form of this format is parent and/or teacher workshops, parent training, etc. The technology training may or may not involve a specific case consultation follow-up, although the integration of service would typically be preferred. In research and usually in practice, these three formats may overlap, or one form of consultation may evolve from the other. For example, in the conduct of individual case consultation, the consultant may find that system issues must be addressed prior to effective services at the case level. Technology training consultation may further be necessary for the overall successful impact of services. The purpose for elaborating briefly these behavioral consultation formats is to illustrate the range and scope, and potential complexity involved in training consultants. As noted before, growing interest in developing effective ways to train students and school-based practitioners in consultation has prompted research focused on effective ways to teach specific consultation skills. These skills have been variously labeled "problem-solving skills," "interview skills," and "behavioral consultation skills." A number of training formats have emerged that can be used for establishing competencies in consultation and related areas of functioning. Four major domains are discussed which include conventional graduate preservice program training, workshop inservice training, competencybased preservice and inservice training (including micro-consultation, behavioral consultation training, and self-instructional training), and mentor training.

118

KRATOCHWILL, SHERIDAN, ROTTO, SALMON

Conventional Preservice Program Training Students in graduate preparation programs obviously represent an important focus in consultation training research. Typically, graduate students in school psychology programs are exposed to formal didactic training in consultation at some time during their schooling. This practice is supported by the National Association of School Psychologists (NASP) which recommends training and establishing competencies in consultation (Fagan, 1990). However, as already noted, few programs may actually incorporate opportunities for supervised applied practice of consultation skills through a field practicum or clinical training component (Meyers, Wurtz, & Flanagan, 1981). Thus, it is likely that few preservice school psychologists actually receive applied training to supplement classroom instruction even though this type of training is important to practice. Some initial research efforts have been directed at evaluating conventional student training programs which have integrated traditional classroom instruction with opportunities for direct applications. Carlson and Tombari ( 1986) developed and evaluated components of a multilevel consultation training sequence with field practicum matched to goals and previous skill development across levels. Table 4.1 presents the four levels of training, corresponding field experiences, and learning goals associated with each level. Their model is based on the assumptions that (a) the three major models of consultation in school psychology (i.e., behavioral, mental health, and organizational) must be included in training, (b) an experiential component of

TABLE 4.1 Multilevel Consultation Training Model Level Entry

Course(s)

Field Experiences

Learning Goals

Individual Mental Individual case studies Competence in child Testing psychological Behaviora!Assessment assessment Affective Assessment School Consultation: Theory and Practice

Student teacher consultation

Cognitive differentiation and practice in diverse consultation models

2

School Consultation Practicum

Consultation in school setting

Application of models to practice setting: skills development

3

Advanced School Consultation

Arranged to meet individual program of study

Expansion of skills to additional settings or depth within school setting

From Carlson and Tombari (1986). Multilevel school consultation training: Preliminary program evaluation. Professional School Psychology, 1, 89-104. Repcoduced by permission.

4.

PREPARATION OF SCHOOL PSYCHOLOGISTS

119

training is necessary, and (c) a field practicum component incorporating the three models is necessary to promote differentiation and flexible practice. Specifically, this model includes four levels of training and field experience: Entry:

Prior to consultation coursework, students complete coursework in client-centered, behavioral, psychoeducational, and affective assessments. Level 1: During the consultation course, students practice diverse consultation roles by consulting with student teachers and a student teacher supervisor. Level 2: Following the consultation course, students can complete and 8-12 hour per week field placement in a public or private school setting. Level 3: For students specializing in consultation, an advanced field experience is individually arranged to develop skills across settings or to expand skills within the public schools to include broader organizational concerns such as program development and evaluation. (p. 90)

Carlson and Tombari ( 1986) evaluated Level 1 training in terms of feasibility, utility, and optimal conditions of training. The subjects involved in the evaluation were two classes of consultation students with six students in each class and their corresponding consultees. The consultants were school psychology students in their second through fourth years of doctoral training in the University of Texas at Austin. The authors evaluated the outcome of the project on several dimensions. Specifically, three questionnaires were developed to evaluate the consultation training program: Consultant Self-Assessment Form, Consultation Evaluation Form, and Supervisory Problem Evaluation Form. 2 The Consultant Self-Assessment Form is a 30-item Likert scale assessing consultants self-perceptions regarding their skill and knowledge development as a function of the consultation training. For example, students are asked how much they improved in their ability to develop and maintain a collaborative relationship with the consultee; clarify issues; formulate intervention plans; and deliver in-service workshops. Regarding knowledge development, students are asked how much they increased their understanding of ethical issues in consultation, and of various models and stages of consultation. The Consultant Evaluation Form is a 28-item Likert scale assessing student-teachers perceptions of the consultant and the consultation experience. Examples of items include "The consultant is able to summarize and facilitate problem-solving" and "As a result of consultation, I find myself trying some of my new ideas." The Supervisor Evaluation Form was designed to provide information on program redesign and included questions such as, Were you given adequate information about the program?, With what aspect of the 2 The Consultant Self-Assessment Form, Consultation Evaluation Form, and Supervisory Problem Evaluation Form are available from Cindy Carlson, Department of Educational Psychology, University of Texas-Austin.

120

KRATOCHWILL, SHERIDAN, ROTTO, SALMON

program have you been most satisfied?, Least satisfied?, If available again, would you want to participate?, What suggestions do you have for increasing the effectiveness of the program? The results of the training program evaluation yielded favorable ratings from school psychology students, student-teachers, and student-teacher supervisors. The overall training effectiveness as measured by the Consultant Self-Assessment Scale suggested that the students perceived gains in knowledge and skills across the models of consultation in Level 1. Furthermore, a matched field experience reportedly enhanced student mastery of consultation skills at a satisfactory level. Consultees also evaluated the consultant skills favorably, but individual responses were more conservative regarding gains due to consultation. However, consultees did report a high degree of program satisfaction. Studentteacher supervisors also indicated general program satisfaction and rated the consultants as most helpful in reducing student-teacher stress in the classroom environment. Self-report data collected from student consultants, student teachers, and student teaching supervisors indicated that the consultation training sequence and field experiences were beneficial for providing consultants with an opportunity to apply knowledge and practice skills learned through the training program. However, limitations of this program included the lack of individual child outcome data for each case consultation, and the absence of data to document changes in child behavior as a result of teacher implementation of intervention programs in the applied settings. Moreover, the research design involving posttest only and no comparison group militates against firm conclusions. In a more recent investigation, Curtis and Zins (1988) evaluated the effectiveness of a consultation training program that emphasized didactic methods, simulation exercises, and videotape analyses, as well as instructor feedback on consultant acquisition of consultation skills. The subjects in this investigation were 14 graduate students in counseling and school psychology who were enrolled in a consultation course. The outcome measures of the investigation involved an analysis of a videotape simulated consultation using the Consultation Verbal Analysis System (CVAS). The CVAS is based on the Consultation Analysis Record (CAR) and analyzes "thought units" (a statement or phrase that expresses a complete thought) in terms of message control, message process, and message content. Subjects in this study were assigned to either an experimental or control group. The training for both groups was held constant, but members of the experimental group received individualized feedback from an instructor based on an analysis of videotaped consultation simulations. Curtis and Zins (1988) reported that the training approach increased acquisition of consultation skills across both conditions, but the addition of individualized instructor feedback did not result in enhanced performance across all of the variables in the experimental group. Instructor feedback had a significant effect on Behavioral Specificity, but not on Questioning or Problem Solution.

4.

PREPARATION OF SCHOOL PSYCHOLOGISTS

121

Workshop lnservice Training Another method for training consultants in consultation skills involves workshops of varying lengths and formats in which consultation strategies are presented. This training approach is typically used at many professional conferences wherein workshop participants receive brief training in consultation techniques. Although this type of training is used often, there is very little empirical research to verify its efficacy. In one of the earlier reports of workshop training, Goodwin, Garvey, and Barclay (1971) evaluated a training procedure referred to as micro-consultation, 3 for teaching behavioral interview skills to school psychologists. Specifically, an 8-week summer workshop was conducted with 30 experienced school psychologists. The training procedure used a micro-consultation format that included discrimination learning, feedback, shaping, and imitation modeling. The psychologists were trained in interview techniques that involved selecting a target behavior for change, identifying environmental events sustaining target behaviors, planning a strategy for change, and evaluating the program. Thirty experimental trainees were compared to a control group that received videotape exposure and a no contact comparison group. The authors used an interview rating form that included 20 items (see Table 4.2). Trained graduate student coders reviewed and rated the interviews at baseline, at termination of the 8-week training program, and 2 months later in the subjects' home school districts. The actual micro-consultation procedure used in the project involved the following 5 steps: I. After reviewing child-specific referral information on the child to be discussed, the trainee conducted a 10-minute interview with a teacher volunteer in which the trainee attempted to make use of the particular skill being learned. For example, if assessment skills were presented, the trainee attempted to identify the environmental contingencies controlling the behavior and ask questions about events which the child found rewarding. 2. Along with five other group members, the trainee watched a videotape replay of his or her performance. Each member completed a rating of this performance against a list of criterion behaviors, and then discussed with the group aspects of the interview where they had attained criterion levels, and those where they fell short of the skill being learned. 3. Alternative responses for a particular interaction were discussed and roleplayed by faculty or a group member when a trainee faced difficulty in acquiring a particular skill. 3Micro-consultation or micro-counseling (MC) as it is often called, is reviewed in detail by Ford (!979). He concluded that MC has proved to be an effective training package generally.

TABLE 4.2

Interview Structure

1. Explains the procedures for using the behavioral analysis with respect to responsibilities, roles, assessment, etc.

2. Assists teacher in deciding on a target problem behavior.

3. Establishes and maintains rapport by clarifying, restating, expressing interest, etc. Environmental Assessment 4. Behavioral definition: asks for definitions of target behaior in terms of frequency and duration.

5. Antecedents: asks about conditions under which behavior occurs; when, where, what, etc. 6. Consequents: asks about conditions following the target behavior; when where, who, etc. 7. Reinforcers: asks about events serving as positive and negative reinforcers for this pupil.

B. Intervening variables: asks about inner states, such as attitudes, personality structure, feelings, concerns.

g_ Demographic data: asks for information regarding test results, family background, education history, etc.

10. Dynamic interpretation: sugests a psychodynamic interpretation of pupil's behavior. 11. Behavioral interpretation: suggests interpretation based on principles of learning. 12. Actions taken: asks about steps taken to correct the problem with respect to conferences, other referrals, etc.

13. Summary: summarizes information provided or decisions made for clarification, decision making, etc.

14. Behavioral data: suggests obtaining more information about the behavioral contingencies, for example, observing in the class, playground, or home, asking for teacher observations. 15. Dynamic data: suggests obtaining more information relative to the child's attitude, personality via testing, depth interview, etc. 16. Nonspecific data: suggests obtaining more information without specific purpose, for example, "111 arrange to see the parent about that.· 17. Behavioral change: suggests a specific course of action in coping with behavioral responses. 18. Dynamic change: suggests changes directed toward increasing self-understanding and improving attitudes or other mediating processes. 19. Nonspecific changes: suggests nonspecific changes, for example, "We'll see what can be done to help the child improve." 20. Behavioral rehearsal: models a desired response and uses role playing to promote acquisition, for example, explaining a strategy to a child.

From Goodwin, Garvey, and Barclay (1971). Microconsultation and behavior analysis: A method of training psychologists as behavioral consultants. Journal of Consulting and Clinical Psychology, 37, 355-363. Reproduced by perm;ss;on.

122

4.

PREPARATION OF SCHOOL PSYCHOLOGISTS

123

4. The trainee next conducted a shorter, 5-minute taped interaction and attempted to bring his or her performance to criterion levels. 5. The second videotape was replaced and further evaluative comments were offered by faculty and group members. Because of the intent and nature of the micro-consultation process, the trainee participated in no more than two I V2 hour sessions per day. ( p. 359) The authors reported that posttest measures of interview techniques at 8-week and 2-month follow-up assessment showed positive skill acquisition overall in comparison to the other groups. Although the authors reported successful outcomes for interview techniques, they did not report changes in individual client behavior on actual consultation cases (see discussion below). In a recent report, McDougall, Reschly, and Corkery (1988) evaluated the effectiveness of a 1-day inservice workshop designed to teach behavioral consultation skills. In the project, 16 school psychologists submitted audiotapes depicting prereferral interviews with teachers before and after the workshops. The audiotapes were analyzed using the CAR on the problem identification interview (see later discussion of this system). The authors found four problem identification interview (PII) subcategories increased significantly following the workshop training. Fewer Background Environment statements were made in the PII than in the baseline interview; Observation statements increased from baseline to PII; and the percentage of consultant statements increased and that of consultee statements decreased from baseline to PII. However, only six PII objectives were significantly changed, indicating a need for a more broad-based training which focuses on objectives rather than discrete verbal segments of interviewing. Although this study is one of the few to evaluate empirically workshop-type training, a major shortcoming, like the earlier Goodwin et a!. ( 1971) study, is the absence of individual client outcome data. Moreover, in this investigation only the PII was evaluated, and the additional components of problem analysis and treatment evaluation interviews were not examined. Competency-based Preservice and lnservice Training

Another major domain of training methodology in special education and school psychology involves competency-based approaches (Idol, 1988; Reilly, Barclay, & Culbertson, 1977). In a typical competency-based training format, the trainer identifies specific objectives and trains these within the consultation context.

Micro-Consultation Training. In an early consultation training project Goodwin and Coates ( 1972) evaluated a micro-consultation format with 70 preservice school psychology graduate students at San Jose State University. The micro-consultation format (Goodwin eta!., 1971) included the following components: (a) an introduction to behavioral analysis (film and videotape models); (b)

124

KRATOCHWILL, SHERIDAN, ROTTO, SALMON

specification of consultant behaviors conducive to teacher change (e.g., establishing rapport; target behavior identification, data collection, and assessment strategy; treatment design and implementation, and a videotape and live models of appropriate consultation skills); (c) practice in the steps of behavioral consultation in laboratory settings through role playing and videotape feedback; (d) supervised casework; and (e) the completion of a final report on the outcome of the project. (Detailed discussion of the training format was presented in the previous section on workshop training.) Outcome measures on the project were selected to assess dependent variables across consultant, teacher, and pupils. Consultants were given the Learning Inventory (Goodwin & Coates, 1972), a 25-item, author-constructed questionnaire to assess knowledge of social learning theory. They also used the School Psychologist Practice Inventory, a 16-item four-choice questionnaire to assess behavioral solutions to academic problems. The Behavior Analysis Rating Scale (Goodwin & Coates, 1972) is a criterion-referenced instrument for evaluating the extent to which the trainees completed the behavior modification procedures according to preestablished criteria. The Behavior Interview Rating Scale was used to determine the frequency with which specified behavior responses were emitted during the actual consultation interviews. Teacher behavior change was assessed by the Teacher-Pupil Interaction Scale (Goodwin & Coates, 1972). This instrument is a classroom observation strategy yielding four categories of teacher behavior including instructing, rewarding, ignoring, and reprimanding and four categories of pupil behavior including attending, scanning, social interaction, and disruptive behavior. This procedure involves time sampling of both teacher and pupil behavior. Finally, client change was assessed by the TeacherPupil Interaction Scale and the Behavior Rate Tabulation Chart (i.e., assessing specific target behavior change) (Goodwin & Coates, 1973). The latter instrument is an observational format that a teacher uses to record daily frequencies of specific target behaviors for students. The authors found significant gains across all measures for the consultants, teachers, and pupils and concluded that the micro-consultation training format produced changes in psychologists' behavior with accompanying change in student performance. Specifically, consultants increased their knowledge of social learning theory and preference for behavioral solutions to problems. Likewise, 74% of consultants met the criteria for target selection on the Behavior Analysis Rating Scale, 82% for strategy planning, 100% for environmental assessment, and 100% for data collection. Gains also occurred on the Behavioral Interview Rating Scale. Teacher behavior (instructing, reinforcing, ignoring, reprimanding) did not change from pre to post, but following consultation, increases in teacher instructing and reinforcing occurred more when students were attending, and generalized to other students. Student behavior changed, with noted increased percentages of attending behavior and decreased "scanning," "socializing," and "disruptive behavior." Moreover, there was at least 50% improvement in behavior of referred students.

4.

PREPARATION OF SCHOOL PSYCHOLOGISTS

125

Behavioral Consultation Training. Some competency-based training approaches have been used by Kratochwill and his associates to train behavioral consultation interview skills. The process of behavioral consultation is presented in four general stages including problem identification, problem analysis, treatment implementation, and treatment evaluation (Bergan, 1977; Bergan & Kratochwill, 1990). With the exception of the treatment implementation phase, the stages of behavioral consultation are procedurally operationalized through interviews. The problem identification interview involves a specification of problems to be targeted during consultation. Problem analysis focuses on exploration of the problem through the evaluation of baseline data, identification of variables that might facilitate a solution to the problem, and development of a specific plan to implement during the treatment phase. Treatment implementation involves the introduction of a plan designed during problem analysis. Finally, treatment evaluation is undertaken in a formal interview to determine the extent to which the intervention plan is successful. Competency-based training in behavioral consultation has involved identification of specific interview objectives to maximize a consultant's success at identifying, analyzing, and evaluating a problem and its associated intervention. The training of consultants typically involves standardized formats to facilitate the learning of discrete verbalization skills. The model for competency-based training in behavioral consultation was outlined by Kratochwill and Bergan (1978) and is based on earlier research on consultant skill and efficiency in the implementation and outcomes of consultation (Bergan & Tombari, 1975). In their early research, Bergan and Tombari (1975) examined the predictive power of consultant efficiency, skill in applying psychological principles, and interviewing skills as related to the phases of consultation. The results of the study suggested that the prediction variables exerted a maximum impact on the problem identification phase. The multiple correlation for the regression of problem identification on consultant variables was .639, for plan implementation on problem identification and consultant variables it was . 775, and for problem solution on plan implementation, problem identification and consultant variables was .977. Consultants lacking in interviewing skills failed to identify target problems and consequently did not reach the stage of plan development and implementation. The consultants who were successful in identifying problems typically were able to solve these problems. Thus, one important implication from the study was that consultants must be trained in the verbal processes used during consultation. Some of this training has been conceptualized within a competency-based framework. Objectives for the various interviews in behavioral consultation have been identified within the context of two major assessment technologies. First, a strategy that allows classification of the messages occurring during consultation classifies verbal interchanges in terms of (a) source, (b) content, (c) process, and (d) control. The vehicle for analyzing these specific consultant and consultee behaviors is called the Consultation Analysis Record (CAR). Basically, coding

126

KRATOCHWILL, SHERIDAN, ROTIO, SALMON TABLE 4.3 Message Classification Categories and Subcategories

Categories Subcategories

Message Source

Message Content

Message Process

Message Control

Consultant

Background environment

Specification

Elicitor

Consultee

Behavior setting

Evaluation

Emitter

Behavior

Inference

Individual characteristics

Summarization

Observation

Validation

Plan Other From Bergan and Kratochwill (1990). Behavioral consultation and therapy. New York: Plenum Press.

Reproduced by pennission.

with the CAR requires that units of observation be determined and numbered on typed transcripts. The four categories of the CAR and their associated subcategories are presented in Table 4.3. Figure 4.1 depicts the CAR and shows how all four message classification categories can be coded. It is beyond the scope of this chapter to provide a detailed overview of each of these categories (see Bergan & Kratochwill, 1990). However, verbal behaviors occurring during the interview can be categorized in terms of the messages in the table. Moreover, there is a growing body of psychometric evidence that attests to reliability and validity of the CAR (see Gresham, 1984; Gresham & Davis, 1988). A second format for determining consultant competencies involves a list of criterion objectives for each of the consultation interviews. Training objectives for the problem identification, problem analysis, and treatment evaluation interviews have been developed and accompany a manual presented by Kratochwill and Bergan ( 1990). The objectives consist of required verbal behaviors to be elicited by the consultant for successful completion of each interview (see Table 4.4). A series of studies have been directed toward competency-based training in behavioral consultation. In an early study in this series, Brown, Kratochwill, and FIG. 4.1. Consultation-analysis record form. From Ber~,Jan and Tombari (1975). The analysis of verbal interactions occurring during consultation. Journal of School Psychology, 13, 212. Reprinted by permission of Human Sciences Press, 72 Fifth Avenue, New York, New York 10011. Copyright (1975).

CONSULTANT ________________________________ CASENUMBER ____________ CONSUL TEE _______________________________ INTERVIEW TYPE ___________ PAGE __________________ CONSULTATION-ANALYSIS RECORD

Messige Source

Messige Content

c

.S!

1:111

~

3... c

0

u

c ~

3.,. c 0

u

.:

t; "Cc c

"'0 eg .c._ ·;; ~

;,E

~-'-' > "'C a:~

...

"'

·~

0 ·;;

;; ·; :l~ "01.1

·-"' >• - .. -="' -="' "C.= ::c .=u

= ~

~

;; :l ;;

~

"'

.&J

0

:a ~

t

"' 0-=

c

Ci:

.i!:

z

~

c

c

.S!

...

...> ;;...>

c

.S!

...~

Messige Control

Messige Process

:l

~

.i!: ·~

g:_

c

~

'-'

c

~

~

.=

.S!

...

~

-~ Q.

c

.S!

...

·~

"' E E :l

"' "'

.S!

c

-;;;

;;

... .S! :'S! :'...S! > ~

i!:

~

z

> ~

.i!: ·~

0

~·

.2 ~

:Q

""

...·e

2

3 4

5

6 7

8 9 10 11

12 13

14

15 16 17

18

19

20 21

22 23 24

25

127

128

KRATOCHWILL, SHERIDAN, ROTTO, SALMON TABLE 4.4 Training Objectives for Each Phase of Consultation

Problem Identification Interview

Problem Analysis Interview

Treatment Evaluation Interview

1. Opening salutation 2. General statements 3. Behavior specification a. Specify examples b. Specify priorities 4. Behavior setting a. Specify examples b. Specify priorities

1. Opening salutation 2. General statements 3. Behavior strength

1. Opening salutation 2. Outcome questions 3. Goal attainment questions

4. Behavior conditions a. Antecedent b. Sequential c. Consequent

4. Internal validity

5. Identify antecedents 6. Identify sequential conditions 7. Identify consequents 8. Summarize and validate

5. Summarize and validate 6. Interpretation

5. External validity 6. Plan continuation

7.Pian statement 8. Summarize and validate

7. Plan modification validation 8. Generalization and maintenance

9. Behavior strength 10. Summarize and validate 11. Tentative definition of goal 12. Assets question 13. Questions re: existing procedures 14. Summarize and validate 15. Directional statement re: data recording 16. Data collection procedures 17. Summarize and validate 18. Date to begin data collection 19. Establish data of next appointment 20. Closing salutation

9. Continuing data collection 10. Establish data of next appointment 11. Closing salutation

9. Follow-up assessment 10. Future interviews 11. Termination of consultation 12. Closing salutation

From Kratochwill, VanSomeren, and Sheridan (1989). Training behavioral consultants: A competencybased model of teach interview skills. Professional School Psychology, 4, 41-59. Reproduced by permission.

Bergan (1982) evaluated a program designed to teach consultation interview skills. In the project, four graduate students were trained in problem identification skills, and their performance was analyzed before and after training in analogue situations using hypothetical client problems. The components of the CAR for the problem identification interview were used to evaluate the outcomes of training. The training components included written outlines, a videotape model with prompts and explanations, and feedback sessions. Generally, the training program was found to be effective for all students and generalized across time with a 2-month follow-up assessment. In a follow-up study, Kratochwill, VanSomeren, and Sheridan (1989) developed a training package to teach consultation interview skills for each of the

4.

PREPARATION OF SCHOOL PSYCHOLOGISTS

129

stages of behavioral consultation. Specifically, skills were taught for the problem identification, problem analysis, and treatment evaluation interviews. The interview objectives for assessing outcome in this investigation were identical to those depicted in Table 4.4. In the first study, four students were exposed to a training manual and viewed videotaped models. In Experiment 2, the videotapes were not used, but all other components of the training remained intact. Experiment 3 used the training manual to teach the interview skills to two practicing school psychologists, who then used the interview guidelines and format with an actual case in their schools. Results indicated that the consultants demonstrated successful acquisition of interview skills. Specifically, the students in the project showed improvements from baseline to treatment; however, there was a decrease in performance demonstrated at the second post-training interview phase. Although the subjects were given some verbal feedback, they did not see coded results of their training interviews and did not have this information available to improve interviewing techniques. This format may have resulted in the observed effect of a decrease between the first and second post-training interviews. It also was speculated that self-evaluative techniques such as self-coding and reviewing audiotape interviews may be beneficial in facilitating cost-and-time efficient training. The data outcomes also showed some evidence of generalization from the problem identification interview to the problem analysis interview, at least in Experiments 1 and 2. Although it was not clear why this occurred, it is possible that some unique generalizable skills are acquired during problem identification and transferred to problem analysis. An important finding of this study is that when the outcomes of training were compared to analysis of performance on the CAR, verbalizations generally conformed to the appropriate categories as outlined initially by Bergan and Tombari (1975). Again, a major limitation of this investigation is the lack of extensive case study outcome data. Only one psychologist in the third experiment was able to provide outcome data for a client. More recently, Kratochwill, Rotto, Sheridan, and Salmon (1991) reported the results of three studies focused on training behavioral consultants to work with teachers of emotionally disturbed children. The study involved an analysis of a training package to teach consultation interview skills in three phases of behavioral consultation: (a) problem identification, (b) problem analysis, and (c) treatment evaluation. In Experiment I, five consultants were exposed to a training package consisting of a training manual and readings, videotape interview models, skill rehearsal, and supervision feedback. In Experiment 2, the package involved the addition of teaching relationship skills and instruction in system issues to five consultants. In Experiment 3, the package was replicated with four consultants and one received a mentorship training format. In each experiment the consultants were evaluated on the acquisition of interview skills, generalization of the skills to role play analogue consultees, and generalization to actual

130

KRATOCHWILL, SHERIDAN, ROTTO, SALMON

child cases referred for psychological services by teacher-consultees of emotionally disturbed children. Results of all three studies indicated that the training package and its variations were an effective means for providing consultation interview training. Results of specific case consultation child-client outcomes were more modest. Results of the three studies indicated that the training package was generally effective. The effects of the training on consultation skills was more clear in Experiment 1 and 3 due to the high level of skills demonstrated on baseline in Experiment 2. Self-Instructional Training. A third approach in competency-based training of consultation interview skills promotes the active involvement of the consultant in self-administered learning of discrete skills. Based, in part, on concerns over the cost of conventional competency-based models of training, researchers have developed a variety of self-instructional approaches that are used for consultation or, more specifically, interview assessment. In a series of studies on training clinicians in behavioral assessment, Miltenberger and his associates have evaluated the effectiveness of several different self-instructional formats. In an early study in this line, Miltenberger and Fuqua (1985) evaluated a self-instructional manual that incorporated written instructions, modeling, and feedback. There were 10 assessment questions that were taught in this investigation. Eight students were trained, four with the training manual and four with one-to-one instruction. Specific interviewing skills (e.g., asks for a general description of the problem; sets priority; asks for specification of problem behaviors, onset, dimensions, antecedents, consequences, and goals) were recorded in a simulated format across baseline and treatment, and the training outcomes were similar under the two different training conditions. In another study Miltenberger and Veltum (1988) evaluated a self-instructional package that incorporated written instructions and written and audio modeling for training 30 assessment interviewing skills. These skills included asking behavioral assessment questions, asking questions relevant to treatment selection, and offering professional courtesy responses prior to and following the interview. In the first study, four upper-level undergraduate psychology majors at North Dakota State University served as subjects. The subjects received training with instructions in audio and written modeling in a multiple baseline design format. The authors reported that the subjects increased interviewing performance substantially following the training program. The second study evaluated the written instructional material alone and involved six upper level undergraduate psychology majors who again were evaluated on the interview outcome measures. Following the baseline interviews the manual alone was implemented. Results suggested that the interviewing performance of four of the participants did not increase to 90% correct, and therefore, a modeling component was added. Following a subsequent feedback phase, all participants increased their performance to 90% or better. Results of both of the investigations demonstrate the effective-

4.

PREPARATION OF SCHOOL PSYCHOLOGISTS

131

ness of instructions and modeling for training interviewing skills and were consistent with previous research that examined self-instructional programs (e.g., Hirschenberger, McGuire, & Thomas, 1986; Miltenberger & Fuqua, 1985). In a third investigation in this area, Vel tum and Miltenberger (1989) examined the effectiveness of a self-instructional package for training both process- and outcome-related assessment interviewing skills. Similar to previous work in this area, the treatment package involved instructions in written modeling alone, a combination of instructions plus audiotape modeling, and self-directed feedback. The subjects in this project (10 graduate students in clinical psychology) were assessed on simulated assessment interviews conducted across baseline and training conditions with confederates who were portraying clients with clinical problems. The outcome data in this investigation involved 30 consultant responses identical to those 1988 study and various process responses. Social validation ratings, client ratings of the therapist, and therapist self-ratings were included as part of the outcome measures. Results of this investigation indicated that moderate levels of performance following baseline resulted in immediate and marked improvements for most of the participants. The authors noted that the training procedures were, therefore, effective on process- and outcome-related responses in the interview. These self-instructional formats represent a promising approach in consultation training given recognition of some limitations. First, the research contains analogue characteristics in that real clients with clinical problems were not included in the protocols. Second, although the "therapist-client" relationship is assessed on graduate students, in these investigations, inservice consultant-consultee interactions may present fundamentally different relationship characteristics, especially in school settings. Finally, only the initial interview is examined, and an interview linked to the entire process of consultation has not been incorporated in these investigations. Relatedly, an important issue is whether an interviewer is more capable of completing a functional analysis of a problem after training in this interview assessment protocol. The authors speculated that the ability to conduct a good interview may not necessarily be correlated with the ability to conduct an adequate functional analysis for the client. Thus, a question in future research is whether interview training should involve an assessment of functional analysis skills before and after training to determine if the therapist can complete this task effectively. Some of these limitations are not unique to self-instructional training formats, but must be addressed to establish the utility of the format.

Mentorship Training Mentorship training in consultation is based on the belief that learning to be a consultant is more than the acquisition of particular skills. The training process also involves constructing and organizing a new conception of oneself in relation

132

KRATOCHWILL, SHERIDAN, ROTTO, SALMON

to other professionals and the organizational processes of the school. According to a mentorship model, as students develop and practice consulting skills, they need to link them to higher-order conceptions of the possible consultant, consultee, and client roles, as well as an understanding of the school as a system. In this way, learning to be a consultant is also a process of acculturation into the profession. Mentorship, a long standing form of gaining knowledge and experience within a domain, offers an additional avenue for socializing beginning school psychologists into the consultant role. Initially, a student may simply shadow a mentor, observing his/her performance in various contexts of practice. In time, the student assumes more responsibility and ultimately enacts an independent professional role. Throughout this learning process, the mentor and student engage in dialogue that provides the student with information about the field, and feedback on his/her own performance. The mentor designs the dialogue in a fashion that scaffolds or supports the student's thinking, and promotes student reflection on his/her own problem-solving processes (Collins, Brown, & Newman, 1989; Schon, 1987). Mentorship currently is receiving a great deal of attention in teacher preparation literature (e.g., Galvez-Hjornevik, 1986; Gray & Gray, 1985; Little, 1990). Researchers recognize mentoring as a holistic form of teacher training that orients beginning teachers to the school culture as well as offering guidance on specific teaching practices. Moreover, mentorship is a means of renewal for experienced teachers. Through the eyes of beginning teachers, veterans have an opportunity to reflect upon and re-evaluate their practices. The holistic focus and the dual benefit to mentor and student are important advantages in the educational and social sciences, where problems are highly complex and new theories and practices develop rapidly. In this regard, mentorship also seems especially well suited for the preparation of school consultants. Indeed, a recent survey of school psychologists indicated that mentorship was a particularly valued, but often under-utilized aspect of professional training (Swerdlik & Bardon, 1988). Salmon, Penning, and Kratochwill (1990) described a model and an exploratory case study of mentorship in school consultation. The authors adopted a cognitive approach to mentorship in which the goal of mentoring is to help students re-organize their knowledge with respect to consultant role and begin to construct a personal theory of practice (Argryis & Schon, 1974; Collins et a!., 1989; Salmon & Lehrer, 1989, in press; Schon, 1983, 1987). In this regard, the student was encouraged to integrate content knowledge with clinical skills. Reflection on knowledge and action was the key to this process of conceptual growth and change. The supervisory dialogue facilitated this reflective process over time. The consultation mentorship process was structured into phases that represent a gradual shift in responsibility for a case from the mentor to the student. This

4.

PREPARATION OF SCHOOL PSYCHOLOGISTS

133

shift was evident by the individual who takes the lead in the consultation interview with the teacher, and the quality of the supervisory dialogue following the interview. Two levels of goals guided the mentor-student supervisory dialogues across the phases: case-level goals and principle-level goals. The focus of case level goals was to gain a better understanding of a specific consultation case by (a) developing a useful way of representing case variables (e.g., conditions when the problem does and does not occur), (b) planning experiments to address questions and hypotheses, and (c) reflecting upon and evaluating the results of these experiments. These goals paralleled an action research approach to problem solving (Argyris, Putnam, & McClain-Smith, 1985). In achieving these goals, the student constructed a working model of the case variables and a direction for a particular consultation case. At the principle level, the purpose of the dialogue was to help students elaborate their knowledge regarding specific cases by (a) integrating case knowledge with other theory and research, and (b) constructing general principles that help link events across cases. New knowledge and skill were consciously integrated with theoretical frameworks, and these theoretical frameworks were further elaborated through case experiences. In this way, the learning that occurred with regard to specific problems was more likely generalized beyond the case context in which it originally occurred (Collins, Foster, & Berler, 1986). Training consultants through mentorship offers several benefits. Students have an opportunity to build a conceptual model of the consultant's role prior to enacting it. In this regard, they have a conceptual road map to guide their own thought and action. The process helps to foster the integration of content knowl- · edge and its application to clinical situations. The dialogue with the mentor across the different mentoring phases allows students to play different roles within the learning process. In the initial phases, students have an opportunity to reflect upon key issues and variables within a case and act as a critic with respect to the actions taken. As students assume more responsibility as the consultant, the mentor offers support and encourages reflection on action. In effect, the mentor acts as a professional mirror helping students to monitor and integrate their professional knowledge and skill. Hence, students are likely to construct a well-organized flexible system of knowledge for enacting the consultant role. ISSUES AND FUTURE DIRECTIONS IN CONSULTATION TRAINING AND RESEARCH Given the great interest in consultation generally, and behavioral consultation specifically, one would expect a solid research base in the area of training behavioral consultants. Unfortunately, this is far from the case. From our review of this literature we must conclude that the research on training behavioral consultants has not kept pace with the growing research on training counselors

134

KRATOCHWILL, SHERIDAN, ROTTO, SALMON

and clinicians in more traditional forms of therapy. For example, just over a decade ago Ford ( 1979) found over l 00 empirical studies on training counselors and clinicians. Research on training behavioral consultants has occurred within four major domains including conventional graduate preservice program training, workshop inservice training, competency-based preservice and inservice training, and mentor training. Most of the research has occurred in the domains of micro-consultation, behavioral consultation training, and self-instructional training. Virtually no research has occurred in the area of workshop and mentor training. Several specific issues must be addressed in this body of research. In the following section we chart some conceptual and methodological issues for future research.

Teaching Interview Skills by an Interactive Computer/Video Researchers have made little use of current computer technology in their training research efforts. Some interesting work in consultation training has been reported by Idol (in press) and her associates (see Lloyd, 1984; Lloyd & Idol-Maestas, 1983). One promising direction for future work in interview training is use and development of interactive video technologies. The interactive video format has been defined as a program designed in segments in which the viewer responds to structured opportunities that influence sequence size and shape of the actual program. Although work in this area is not well-developed in terms of consultation training, some preliminary efforts have occurred in the actual training of individuals in consultation skills. It should be noted that a variety of interview formats already have been adapted to computers, and there are a number of interactive video technologies available for purposes of interview assessment activities (see Kratochwill, Doll, & Dickson, 1990). The major benefit of using the computer and video-disk is that these technologies can be adapted to present a wide range of training options. For example, the computer can provide text and graphic instruction as well as the usual branching routines. The video option can provide natural presentations of real interactions with either role-play or actual clients. One promising development in the use of teaching interviewing skills with interactive video was presented by Cummings, Hansen, and Sillings (1989). The purpose of their project was to develop and use two interactive video programs for teaching interview skills to school psychology students and practicing clinicians. These authors developed two programs based on a theoretical model of skill acquisition presented by Dreyfus and Dreyfus (1980, 1986). The interested reader is referred to their original article or to Cummings et al. (1989) for further information. In their first program ( "Intertalk"), the primary goal was to create an interactive video program for teaching interviewing skills. The focus was primarily on the concepts of establishing rapport, giving the respondent a concise

4.

PREPARATION OF SCHOOL PSYCHOLOGISTS

135

description of the interview purpose, and developing a collaborative relationship between the two parties. The program incorporated a number of features to facilitate learner involvement including dramatization, repetition, explanation, audio and video feedback. The second program ("SOBEX"; Systematic Observation Exercise) was developed from the lntertalk program. While initially serving as a data collection instrument, it was thought to accomplish a training function when used with more advanced interviewers. The program was designed to incorporate six features including split-screen video, interviewer comments, marking device (i.e., a format to mark a point in the interview with minimal disruption of the viewing experience), video transcript notetaking, and a tracking component designed to keep track of activities of users while they were working on an observation exercise. The efficacy and benefits of such a program include its interactive characteristics, abilities to reflect on skills, opportunities to compare performances, and sophisticated learning strategies for more advanced individuals. However, although the audhors described the program in two separate formats, they did not specify the interview skills that were taught or provide outcome data on consultant performance or client behavior.

Training Consultees in Consultation The traditional approach to consultation training has focused solely on consultant skill performance, without much attention to the skill level of the consultee in training outcomes. Consideration of this issue can be framed within the context of the behavioral consultation indirect service formats presented by Vern berg and Reppucci ( 1986). In traditional behavioral case consultation little or no training of the consultee occurs, although this training is potentially advantageous (Bergan & Kratochwill, 1990). In behavioral technology training the primary focus of consultation efforts is on training the consultee in discrete skills to better manage the client. Training the consultee may have some specific advantages in the training of the consultant. The consul tee may model effective knowledge and skill and therefore, the consultation relationship is educational and reciprocal. Client outcomes can also be investigated as a distal measure of training effectiveness, and are likely to be improved given good consultee skills. Are these potential benefits supported in the empirical literature? Only a few empirical studies have addressed this issue (Anderson, Kratochwill, & Bergan, 1986; Cleven & Gutkin, 1988; Kratochwill & Elliott, 1990). Anderson et al. (1986) evaluated the relative effectiveness of two teacher training packages on two analogue consultation dependent measures. The subjects in the study included elementary school teachers (N = 56) in an urban school district in Arizona. The two conditions included (a) training in classroom behavior modification and consultation verbal processes, and (b) training in consultative service delivery procedures and general multidisciplinary team processes (a control condition).

136

KRATOCHWILL, SHERIDAN, ROTIO, SALMON

Analogue consultation dependent measures included specific and general problem identification, problem analysis, and problem evaluation elicitors from the CAR. Subjects in each condition were evaluated on knowledge of behavior modification principles and concepts, and the frequency of specific categories of consultee verbal behaviors on the CAR. Results suggested that the behavior modification and consultation training package was effective in increasing teacher knowledge of behavioral procedures and in increasing the frequency of teacher verbalizations regarding overt child behaviors, behavior observation techniques, and intervention plans during the problem identification and problem analysis phases of analogue consultation. Anderson et al. also found that even after training consultees, specific consultant questions were important in eliciting consultee statements related to environmental conditions surrounding behavior. In contrast, the use of more general consultant verbal behaviors resulted in significantly more vague, unspecified, and irrelevant types of consultee verbalizations. In this study, the consultants previously trained in behavioral consultation interacted with the teachers trained in the two conditions. In this regard, there was no analysis of how teachers would perform with consultants who were not trained in behavioral consultation. A study could be designed to manipulate both consultant and consultee training. An obvious limitation of the Anderson et al. study is that it was implemented under analogue conditions and without measurement of child outcomes. To overcome the latter concerns, Kratochwill and Elliott ( 1990) currently are investigating the efficiency of behavioral consultation services under conditions of consultant and consultee training with measures of teacher and child outcomes. Cleven and Gutkin ( 1988) conducted a training study in which they examined "consultees" understanding and use of problem definition skills. In the investigation, they implemented a training procedure to increase consultees' understanding and use of problem definition skills during consultation interactions. The study involved 195 volunteer female undergraduate students who were assigned to one of three conditions. Condition 1 involved consultation with cognitive modeling, where consultees viewed videotaped consultation interactions (i.e., problem definition) in which the consultant's verbalizations frequently included overt references to problem-solving processes used by consultants. The consultees in the consultation without cognitive modeling group viewed similar interactions, but the consultant statements did not include overt references to a problem solving process. These two groups were compared to a control condition in which the consultees viewed irrelevant interactions between consultants and consultees. The outcome measure in the study included consultees' responses to the problem definition description and process questionnaires. The investigators reported that consultees who were exposed to consultation scenarios/examples that included cognitive modeling wrote better problem definitions and were better able to describe the process for defining children's prob-

TABLE 4.5 Social Validation of Task Analysis of Parent Communication Skills Preparation

Parent thanks the professional for taking the time to meet with him or her. Parent states that he or she is willing to actively participate in the meeting by providing information, feedback, and/or helping in the decision-making process. Parent states how he or she has prepared for the meeting by bringing relevant materials (e.g., "I have brought my child's report cards") and/or stating that he or she has given thought to his or her child's behavior. Parent brings to the meeting a list of the evaluators and their disciplines. Parent brings to the meeting materials necessary to record information. Complete Communication

Parent states a summary of what the professional has said at least at conclusion of the professional's report (i.e., summary must include same topic area as that of professional's report). Parent states his or her general observations of the child in the natural environment with regard to the topic at hand. Parent asks for feedback from his or her partner (if present). If there is a discrepancy between parent's and professional's observations, then parent requests clarification of the professional's statements of observation(s) just given. Clarification

Parent asks questions about what has been discussed, or states that he or she has no questions about (understands) information given. Parent states a summary of the professional's response to his or her question(s), (i.e., a summary must include same topic area) or states that he or she understands. Consensus

Parent compliments the professional, the evaluation, or the meeting and/or makes a statement to acknowledge appropriateness to some aspect of professional's suggestion or report. Parent states specific area(s) of agreement with the professional (i. e., uses "I" statements), or states there are no areas of agreement. Identification of Issues

Parent states area(s) of disagreement with the professional (i.e., using "I" statements) without stating that the professional is incorrect, or states that there are no areas of disagreement. If disagreement, parent states his or her understanding of the professional's concern for the child. If the parent was mistaken about the disagreement or the professional's statement(s), he or she admits mistake. Suggestion of Options

Parent states or requests the possible options based on areas of agreement and/or disagreement. Parent summarizes all options that have been presented. Parent asks for or states the advantages and disadvantages of each option listed. Parent states options in descending order of preference. If there is a disagreement regarding most preferred option, parent makes a statement allowing the professional to "own" the parent's most preferred option (e.g., using "you," "your"). If there is a disagreement, parent states the more positive spect of his or her chosen option.

Continued

137

138

KRATOCHWILL, SHERIDAN, ROTTO, SALMON TABLE 4.5 (continued) Decide on Action to Be Taken

Parent states or asks who will deliver services. Parent states or asks what services are to be delivered. Parent states or asks where services will be delivered. Parent states or asks when services will begin. Parent states or asks the time and day of week that services will be delivered. Parent states or asks how much the services will cost and/or if-insurance will cover cost. Parent states or asks how long the services will need to be provided. Parent states chosen option or states why option is not feasible. If necessary, parent states his or her next preferred option. Feedback and Acknowledgment Parent compliments the professional, the evaluation, and/or the meeting. Parent states or asks who will make the next contact. Parent states or asks when the next contact is to be made. Parent asks for or states how the contact person may be reached. From Kohr, Parrish, Neef, Driessen, and Hallinan (1988). Communication skills training for parents: Experimental and social validation. Journal of Applied Behavior Analysis, 21, 2 I -30. Reproduced by permission.

lems than those in either the consultation without cognitive modeling or control groups. Consultees in the consultation without cognitive modeling group performed better than those who participated in the control condition. The major implication of this study is that it may be important for school-based consultants to help teachers develop skills in problem definition during the process of consultation. Specifically, providing teacher consultees with a "cognitive road map" of the problem definition process may facilitate effective problem definitions. Whether the results of this study can be extended to graduate students in school psychology or practicing school psychologists engaged in consultation with actual teachers remains to be determined. Also, the issue of whether consultant skills would be enhanced as a function of consultee skill development and performance beyond direct consultant training, remains to be explored. An interesting study by Kohr, Parrish, Neef, Driessen, and Hallinan (1988) provides a template for training communication skills to parents who subsequently interacted with professionals. The study has interesting implications for the performance of the professional exposed to a parent trained in these skills. Eight parents were trained in socially validated skills with an instructional package (see Table 4.5). The authors found that the parents acquired the targeted skills during simulated conferences, and some skills generalized to actual conferences. Again, whether these skills in parents would influence the professional skill level of the consultant is not determined. Nevertheless, it is likely that a training approach that involves both a focus on consultant and consultee skills would facilitate overall training impact at both proximal and distal levels. It is

4.

PREPARATION OF SCHOOL PSYCHOLOGISTS

139

obvious that many of these skills would facilitate the process of parent and professional consultation. Conjoint Consultation

The majority of consultation training programs have been conducted with teachers serving as consultees, and therefore have been limited within a narrow training focus (Kratochwill, Sheridan, & VanSomeren, 1988). It is important that consultants in training receive experience in a wider range of settings, with diverse consultees. A focus on parent consultation and parent involvement is especially important in school psychology graduate programs, where homeschool partnerships and the relationship between school psychologists and parents have been identified as critical training issues (Bergan & Duley, 1981; Brown & Cardon, 1982; Guidabaldi, 1982). Indeed, a rather large body of research has focused on training parents in behavior modification (e.g., Forehand & McMahon, 1981; Patterson, Reid, Jones, & Conger, 1975). However, although some have referred to this work as "parent consultation" (Cobb & Medway, 1978; Ollendick & Cerney, 1981), most of these programs involve only a combination of parent training and direct child intervention and are best conceptualized as behavioral technology training. Thus, little is known about the practice of case-centered parent consultation and implications that this approach might have for training. In fact, none of the training studies that we reviewed focused on parent consultation. "Conjoint consultation" (with parents and teachers serving together as joint consultees) has received some support as an effective method of service-delivery for socially withdrawn children (Sheridan, Kratochwill, & Elliott, 1990). Conjoint consultation is defined as a systematic, indirect form of service delivery, in which parents and teachers are joined to work together to address the academic, social, or behavioral needs of an individual for whom both parties bear some responsibility. It is designed to engage parents and teachers in a collaborative problem-solving process with the assistance of a consultant. In conjoint consultation, there is a focus on the interacting systems operating within a child's life, and the interconnections between home and school. This framework has the potential to promote increased communication, reciprocity within and between systems, identification of cross-setting influences on a child's behavior, consistent and systematic behavioral programming, and generalization of treatment effects within and across settings (Sheridan et a!., 1990). Given the focus on the interacting systems within a child's life, conjoint consultation is conceptually and procedurally different from traditional teacherand parent-only consultation. Thus, several training questions in relation to this model of service-delivery arise. Are traditional training programs, which typically focus on consultation with one consultee who represents only one system in a child's life, adequate in preparing consultants to work with the two interact-

140

KRATOCHWILL, SHERIDAN, ROTTO, SALMON

ing systems (i.e., home and school) simultaneously? What specific competencies are necessary for consultants to interface effectively with parents and teachers in a conjoint fashion? Should this training occur after consultants demonstrate "competence" in working with one consultee individually, or would a conjoint approach in training facilitate skill development that is generalizable to a single consultee? These are important issues that will undoubtedly face trainers as they attempt to address parent-teacher consultation and home-school partnerships more directly in their training programs.

Consultation Supervision The role that supervision plays in consultation training remains largely unexplored. By "supervisor" we mean the individual conducting the training and/or involved in supervising the consultation activities of the consultant. Knoff ( 1985) did not uncover one study of supervision in the school psychology literature and although Ford ( 1979) reviewed considerable research on the role that relationship factors play in supervision, he was cautious in drawing any firm conclusions. Ford ( 1979) concluded that a facilitative supervisory approach is necessary in training counselors and clinicians, but that is not sufficient for effective training. The influence that supervisors have on training outcomes will vary along several dimensions of involvement in the consultation process. For example, the impact that supervisors have will be potentially greater if they are involved in case supervision than if they discontinue their influence following training in discrete consultation skills. Few researchers presented information on the role that the supervisor plays in training outcomes. Although Kratochwill et a!. (1991) included supervision as part of the training package (see Salmon & Sheridan, 1987 for a discussion), the specific influence of the supervisor cannot be determined. A review of various models and approaches to supervision is presented in Knoff ( 1985) and a discussion of one approach to supervision in behavioral consultation is presented in Kratochwill, Bergan, and Mace (1981). Building on the research initiatives presented by Knoff (1985, pp. 538-539), we can offer the following agendas for future research on the role of supervision in behavioral consultation training. First, a model of supervision specific to indirect service delivery should be developed to guide future research efforts in the area. The model presented by Knoff (1985) and reproduced in Fig. 4.2 will serve as an important stepping stone in this process. Second, specific supervisor behaviors (and characteristics) and their relationship to effective training outcomes should be determined. Although some hints on what behaviors might be useful in supervisors during consultation training are presented (Ford, 1979; Kratochwill et a!., 1981 ), there is no research in this area. Third, and related to point two, measures of supervisor behavior as related to the training process need to be explored. A potentially useful assessment model of consultation supervision involves the requirement of consultant

4.

PREPARATION OF SCHOOL PSYCHOLOGISTS Entry Level

Scltonl l,.'tJiclwlnnical

Internship

Status{~s:

Pnu.:ticum

'....._

\

141

Independent Practice

\

Supervisor

~'~~~'~~~~~~----

Superolsion Components

Skill----C