CHILDREN AT RISK

AN EVALUATION OF
FACTORS CONTRIBUTING TO
CHILD ABUSE AND NEGLECT
CHILDREN AT RISK
AN EVALUATION OF
FACTORS CONTRIBUTING TO
CHILD ABUSE AND NEGLECT

Edited by
ROBERT T. AMMERMAN
Western Pennsylvania School for Blind Children
Pil/sburgh, Pennsylvania

and
MICHEL HERSEN
University of Pil/sburgh
School of Medicine
Pillsburgh, Pennsylvania

Springer Science+Business Media, LLC

 LIbrary of Congress CatalogIng-In-PublIcatIon Data

Children at risk : an evaluatIon of factors contributing to child

abuse and neglect I edited by Robert T. Ammerman and Michel Hersen.
p• em .
Includes bibliographical references .
Includes Index .

ISBN 978-1-4419-3214-3 ISBN 978-1-4757-2088-4 (eBook)

DOl 10.1007/978-1-4757-2088-4

1. Abused children--Unlted States . 2. Child abuse--United States.

3. Child abuse --Unlted States--Preventlon. I. Ammerman, Robert T.
II. Hersen, MIchel .
[DNLM : 1. ChIld Abuse--epidemlology . 2. RIsk Factors. HA 320
C536517)
HV741.C536135 1990
362 .7 '6 '0973--dc20
DNLM/DLC
for LIbrary of Congress 90-7228
CIP

© 1990 Springer Science +Business Media New York

Originally published by Plenum Press, New York in 1990.
Softcover reprint of the hardcover Ist edition 1990

All rights reserved

No part of this book may be reproduced , stored in a retrieval system, or transmitted
in any form or by any means, electronic, mechanical, photocopying, microfilming,
recording, or otherwise, without written permission from the Publisher
To Caroline and Helen
 CONTRIBUTORS

ROBERT T. AMMERMAN, Western Pennsylvania School for Blind Chil-

dren, Pittsburgh, Pennsylvania 15213
JAY BELSKY, Department of Individual and Family Studies, College of
Health and Human Development, Pennsylvania State University,
University Park, Pennsylvania 16802
BEVERLY A. BUSH, Department of Psychology, University of Maryland-
Baltimore County, Catonsville, Maryland 21228
JUDITH A. COHEN, Department of Psychiatry, Western Psychiatric In-
stitute and Clinic, University of Pittsburgh School of Medicine,
Pittsburgh, Pennsylvania 15213
HOWARD DUBOWITZ, Department of Pediatrics, University of Maryland
School of Medicine, Baltimore, Maryland 21201
DAVID C. FACTOR, TRE-ADD Program, Thistletown Regional Center for
Children and Adolescents, Rexdale, Ontario, Canada M9V 4L8
JAMES GARBARINO, Erikson Institute for Advanced Study in Child Devel-
opment, Chicago, Illinois 60610
Roy C. HERRENKOHL, Center for Social Research, Lehigh University,
Bethlehem, Pennsylvania 18015
MICHEL HERSEN, Department of Psychiatry, Western Psychiatric In-
stitute and Clinic, University of Pittsburgh School of Medicine,
Pittsburgh, Pennsylvania 15213
JEFFREY A. KELLY, Division of Psychology, University of Mississippi
Medical Center, Jackson, Mississippi 39216

vii
viii CONTRIBUTORS

JOHN R. LUTZKER, Department of Psychology, University of Judaism, Los

Angeles, California 90077
ANTHONY P. MANNARINO, Department of Psychiatry, Western Psychi-
atric Institute and Clinic, University of Pittsburgh School of Medi-
cine, Pittsburgh, Pennsylvania 15213
GARY B. MELTON, Department of Psychology, University of Nebraska-
Lincoln, Lincoln, Nebraska 68588-0308
MAXINE R. NEWMAN, Department of Psychology, University of Judaism,
Los Angeles, California 90077
RANDY K. OTTO, Department of Law and Mental Health, Florida Mental
Health Institute, University of South Florida, Tampa, Florida 33612-
3899
RAYMOND H. STARR, JR., Department of Psychology, University of
Maryland-Baltimore County, Catonsville, Maryland 21228
JOAN I. VONDRA, Department of Psychology in Education, University of
Pittsburgh, Pittsburgh, Pennsylvania 15260
DAVID A. WOLFE, Department of Psychology, University of Western On-
tario, London, Ontario, Canada N6A 5B8
LISE M. YOUNGBLADE, Department of Individual and Family Studies,
College of Health and Human Development, Pennsylvania State
University, University Park, Pennsylvania 16802
 PREFACE

During the past decade, a dramatic increase in research and clinical

interest has risen in child abuse and neglect. This recent growth in
awareness is due at least partly to the alarming statistics documenting the
incidence of child maltreatment. Almost one million children are re-
ported to be abused and neglected each year, and many experts believe
that th is figure underestimates the true incidence. Indeed, recent surveys
suggest that almost 1.5 million children are the targets of domestic vio-
lence every year. A significant proportion of these children die as a func-
tion of thi s maltreatment, whereas the remainder suffer a variety of short-
and long-term deleterious medical and psychosocial con sequences . Ch ild
maltreatment is a universal problem that has precipitated a mobilization
of effort from a vari ety of disciplines, including psychology, medicine,
ps ychiatry, social work, sociology, and criminology.
Particular attention has been directed toward the prevention and
treatment of child abuse and neglect. Such endeavors require the screen-
ing of large groups in order to identify families that are at high-risk for
engaging in such behavior. Delineating those characteristics that differ-
entiate high- from low-risk families and children is one of the obvious
priorities for researchers and clinicians in the future. This book, there-
fore, carefully considers the status of research on risk factors of abuse
and neglect in children. Adduced data undoubtedly will have practical
value for subsequent intervention efforts. As research in child abuse and
neglect approaches its fourth decade of modern investigation, it is es-
pecially timely to assess the progress made in this field and provide
guidelines for future empirical work.
Children at Risk is divided into five parts. In Part I (Introduction), an
overview of risk research is presented, and guidelines for continued

ix
x PREFACE

empirical investigation are outlined. In Part II (General Issues), critical

areas of child maltreatment are examined, including the epidemiology of
abuse and neglect, legal issues, research practices and methodological
approaches, and the psychosocial sequelae of maltreatment of child vic-
tims. Part III (Risk Factors Associated with Child Abuse and Neglect)
includes evaluations of the three primary contributing factors to the
etiology of maltreatment: sociological and ecological variables, parental
characteristics, and child characteristics. Important new developments
in the prevention and treatment of abuse and neglect are covered in Part
IV (Prevention and Treatment). Finally, Part V (Conclusions) provides
the agenda for future research and clinical endeavors in combatting
child maltreatment.
A number of people devoted considerable time and effort in assist-
ing us in the compilation of this volume, and their patience and dedica-
tion to the project are greatly appreciated. First, we thank the contrib-
utors who have graciously reviewed their respective areas of expertise.
Second, we greatly appreciate the technical assistance provided by Mary
Anne Frederick, Mary [o Horgan, [enifer McKelvey, Louise E. Moore,
and Mary H . Newell. Third, we extend our gratitude to J. Lawrence
Aber III for his helpful comments on our original outline for the book.
Finally, we are indebted to the support and invaluable assistance pro-
vided by our editor, Eliot Werner, who proved himself a model of
tolerance.

Robert T. Ammerman
Michel Hersen
Pittsburgh , Pennsylvania
 CONTENTS

PART I. INTRODUCTION

Chapter 1
Research in Child Abuse and Neglect: Current Status and an
Agenda for the Future. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Robert T. Ammerman and Michel Hersen
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
The Problem of Definition 6
Consequences of Maltreatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
The Search for Risk Factors 10
Treatment and Prevention 14
Summary 16
References 16

PART II. GENERAL ISSUES

Chapter 2
The Epidemiology of Child Maltreatment 23
Raymond H. Starr, [r., Howard Dubowitz, and Beverly A. Bush
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Basic Epidemiological Concepts 24

xi
xii CONTENTS

The Definition of Child Maltreatment . . . . . . . . . . . . . . . . . . . . . . . . 24

Physical Abuse 25
Physical Neglect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 26
Psychological Abuse and Psychological Neglect 26
Sexual Abuse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 27
Summary 27
The Incidence and Prevalence of Child Maltreatment . . . . . . . . .. 28
Analyses of Reported Maltreatment . . . . . . . . . . . . . . . . . . . . . . . . 28
The National Incidence Surveys 34
Surveys of the General Public . . . . . . . . . . . . . .. 38
Summary and Implications 47
References 50

Chapter 3
Trends in Legislation and Case Law on Child Abuse
and Neglect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Randy K. Otto and Gary B. Melton
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Justification for State Intervention 56
Parens Patriae Power 56
Orientation and Scope of Intervention 56
Criminal and Civil Adjudication of Child Abuse
and Neglect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
Legislative Responses 63
The First Wave of Legislation: Child Abuse
Reporting Statutes 63
Child Abuse Prevention and Treatment Act of 1974 66
Other Federal Legislation 68
Children's Trust Funds 69
Procedural and Evidentiary Reforms Related to Adjudication 70
The Impact of Testifying upon Child Witnesses 71
Children's Competency to Testify . . . . . . . . . . . . . . . . . . . . . . . . . . 72
Hearsay Exceptions 73
Closed Courtrooms 74
Special Courtroom Procedures 75
Psychological Testimony about Abuser and Victim Profiles 76
Discussion 77
Summary 78
References 78
CONTENTS xiii

Chapter 4
Research Directions Related to Child Abuse and Neglect . .. . 85
Roy C. Herrenkohl
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Research Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 86
Incidence and Prevalence of Abuse and Neglect 86
Research Directions 87
Causes of Abuse and Neglect 89
Research Directions 92
Consequences of Abuse and Neglect 92
Physical Abuse 93
Emotional Abuse 94
Sexual Abuse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Child Neglect 94
Research Directions 95
Treatment of Abuse and Neglect 96
Research Directions 98
Prevention of Abuse and Neglect 100
Research Directions 100
Research Results and Policy Considerations . . . . . . . . . . . . . . . . . .. 101
Intergenerational Transmission . . . . . . . . . . . . . . . . . . . . . . . . . . .. 101
Antisocial Behavior 101
Adoption 102
Family Dissolution 102
Summary , . .. . . . . .. . . . . . . . . . ... ... . . . . . . .. . .. 104
References 105
Chapter 5
Social and Emotional Consequences of Child Maltreatment .. 109
Lise M . Youngblade and Jay Belsky
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
Issues in the Study of Child Abuse and Neglect 111
Overview 114
Social and Emotional Consequences during Infancy
and Toddlerhood 114
Toddler-Peer Relations 121
Social and Emotional Consequences during Childhood 126
Family Interaction in Maltreating Households 127
xiv CONTENTS

Behavioral Functioning of Maltreated Children 128

Social and Emotional Consequences during Adolescence
and Adulthood 134
Consequences in Adolescence . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 134
Intergenerational Transmission . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
Summary 138
References 140

PART III. RISK FACTORS ASSOCIATED WITH CHILD ABUSE AND NEGLECT

Chapter 6
Sociological and Ecological Factors 149
Joan I. Vondra
Introduction 149
The Multiple Determinants of Parenting 151
A Model of Influences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
Factors within the Family . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
Childrearing History 151
Adult Psychological Resources. . . . . . . . . . . . . . . . . . . . . . . . . . . .. 154
The Marital Relationship . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Child Characteristics 158
Factors Outside the Family . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 160
Social Network Support 160
Socioeconomic Considerations 161
The Sociocultural Milieu . . . . . . . . . . . . . . . . . . .. 162
Summary 163
References 164
Chapter 7
Parental Psychopathology and High-Risk Children 171
David C. Factor and David A . Wolfe
Introduction 171
Child-Rearing Patterns and Their Suspected Influence
on Child Development 172
Parental Disorders and Child Adjustment: An Overview
of the Literature 174
Parental Immaturity 175
Parental Criminality 176
CONTENTS xv

Chronic Parental Illness and Family Functioning 176

Children of Parents with Psychiatric Illnesses 177
Parental Depression and Its Influence on Child Development 180
Behavioral Problems among Children of Depressed Parents .. 182
Cognitive and Affective Disturbances among Children of
Depressed Parents 184
Parent/Child Interactions in Families with a
Depressed Parent 186
Parental Psychopathology and Child Maltreatment 191
Summary 193
References 195
Chapter 8
Predisposing Child Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
Robert T. Ammerman
Introduction 199
Conceptual Models of Child Abuse and Neglect 200
Comments 202
Early Childhood Risk Factors 203
Comments 207
Coercive Interactions in the Development and Maintenance
of Abuse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
Comments 210
Handicap as a Risk Factor for Maltreatment 211
Incidence of Maltreatment in Handicapped Populations 213
Comments 215
Summary 215
References 217

PART IV. PREVENTION AND TREATMENT

Chapter 9
Prevention Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
Maxine R. Newman and John R. Lutzker
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
Prevention Levels : . . . . . . . . . . . . . . . .. 226
Tertiary Pre vention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
xvi CONTENTS

Secondary Prevention 227

Primary Prevention 228
Historical Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
Child Abuse before the Battered Child Syndrome 229
The Battered Child Syndrome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230
After Kempe 231
Prevention in Child Abuse and Neglect 231
Prevention Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 238
Recommendations 242
Summary 244
References 244
Chapter 10
Treating the Abused Child . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 249
Anthony P. Mannarino and Judith A. Cohen
Introduction 249
Definitional and Methodological Issues 250
Definitional Problems 250
Methodological Problems 251
Treating the Physically Abused Child 251
Review of Treatment Studies of Child Victims 252
New Directions in the Clinical Treatment of Abuse Victims . . 254
Treating the Sexually Abused Child 256
General Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256
Review of Treatment Studies of Child Victims 258
New Directions in the Treatment of Sexually
Abused Children 264
Summary 265
References 266
Chapter 11
Treating the Child Abuser 269
Jeffrey A. Kelly
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 269
Multifactorial Models of Child Abuse: Implications
for Treatment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
Interventions to Improve Parenting Skills 271
Parenting Skills Interventions with Individual Clinical Cases 272
Group Comparison Studies of Parent-Training Interventions 274
CONTENTS xvii

Interventions That Address Parent Coping 276

Research Critique 280
Research Needs on Multifactorial Treatments 282
Treatment for Special Abuser Populations 283
Summary 283
References 284

PART V. CONCLUSIONS

Chapter 12
Future Directions 291
James Garbarino
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 291
Expanding Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292
Polarization of Family Experiences 293
Proliferating Linkage 294
Community Responsibility for Children 295
Psychological Maltreatment 297
Summary 297
References 297

Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299

Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311

 PART I

INTRODUCTION
 CHAPTER 1

RESEARCH IN CHILD ABUSE

AND NEGLECT
CURRENT STATUS AND AN AGENDA
FOR THE FUTURE

ROBERT T. AMMERMAN AND MICHEL HERSEN

INTRODUCTION
It is common to decry the relative paucity of empirical research in newly
emerging fields, and the area of child maltreatment is no exception. Yet,
during the past decade, we have seen an explosion of investigative
activity that has thrust research in child abuse and neglect to higher
levels of understanding and sophistication. Indeed, the growth in
knowledge has been so extensive that recent endeavors have brought
the field to something of a watershed. Therefore, it is timely to assess
the gains that have been achieved and review the changes in investiga-
tive practices and theoretical formulations that have thus far charac-
terized the field . From a research perspective, a transition has occurred
from unidimensional design approaches to multivariate strategies (e.g.,

ROBERT T. AMMERMAN • Western Penn sylvania School for Blind Children , Pitt sburgh,
Pennsylvania 15213. MICHEL HERSEN • Department of Psychiatry, Western Psychiatric
Institute and Clinic, Univer sity of Pittsburgh School of Med icine, Pittsburgh, Pennsylvania
15213.

3
4 ROBERT T. AMMERMAN and MICHEL HERSEN

Gaines, Sandgrund, Green, & Power, 1978)that more adequately reflect

the complex nature of child maltreatment. Similarly, current theoretical
views reject single-factor explanations of maltreatment in favor of more
intricate models describing the reciprocal interplay between causative
variables in the development and maintenance of abuse and neglect
(e.g., Starr, 1988; Wolfe, 1987). Although this shift in focus of attention
from simplistic conceptualizations to complicated multicomponent
models began in the late 1970s (Belsky, 1980)it is now almost universally
acknowledged (Starr, 1988).
The dramatic growth in understanding the etiology and conse-
quences of child abuse and neglect can be charted by the seminal articles
and reviews that mark the progression of the field in its first 25 years
(Belsky, 1980; Elmer & Gregg, 1967; Friedrich & Boriskin, 1976; Gelles,
1973; Kempe, Silverman, Steele, Droegemueller, & Silver, 1962; Parke &
Collmer, 1975; Spinetta & Rigler, 1972). In fact, each of these contribu-
tions is as relevant today as when they were first published, because
they highlight and underscore the critical methodological and concep-
tual issues faced by researchers of child maltreatment. Thus, although
the field has advanced tremendously since the appearance of these
works, their impact on current empirical efforts is considerable.
Kempe and coworkers (1962) were the first to describe the symp-
toms and characteristics of maltreated children and deserve credit for
rekindling modern-day interest in this area. In addition to launching a
field of scientific inquiry, they elucidated one of the most controversial
problems in child abuse and neglect-how to define the phenomenon
objectively-and relied primarily on physical evidence to identify abuse
and neglect. The evidence included multiple fractures, bruises, scars,
abrasions, and frequent accidents of suspicious origin. Since that time,
however, a significant expansion has taken place in the definition of
child maltreatment to comprise severe physical punishment without
permanent injury, placing the child in dangerous situations, and emo-
tional abuse (Brassard, Germain, & Hart, 1987). Although the broaden-
ing of criteria constituting child abuse and neglect has directed attention
toward previously ignored maltreated populations, it also has led to
methodological dilemmas and legal policy confusions (Besharov, 1982).
The pervasive deleterious consequences of maltreatment were first
outlined by Elmer and Gregg (1967). Their short- and long-term evalua-
tions of maltreatment (Elmer, 1977) documented the severe negative
sequelae resulting from abuse and neglect in such areas of functioning
as intelligence, academic achievement, social and cognitive develop-
ment, and emotional adjustment. Although methodological limitations
mitigate conclusions drawn from these studies, their efforts stimulated
extensive research in the consequences of child abuse and neglect (see
RESEARCH: CURRENT STATUS AND THE FUTURE 5

Ammerman, Cassisi, Hersen, & Van Hasselt, 1986; Friedrich & Ein-
bender, 1983).
The literature in the early and mid-1970s was dominated by the
search for causative variables in maltreatment. Kempe et al. (1962)specu-
lated that parental psychopathology was the most important determi-
nant in the etiology of abuse and neglect. Subsequently, in their review,
Spinetta and Rigler (1972) concluded that psychopathology and psychi-
atric disorder in parents were crucial factors in the development of abuse
and neglect. In turn, this review resulted in a burgeoning of more well-
controlled empirical studies examining the characteristics of abusive par-
ents. However, findings from these investigations failed to support
Spinneta and Rigler's (1972)initial conclusions. Rather, although current
data indicate that most abusive parents exhibit deficits in a variety of
areas of functioning, they rarely suffer from severe psychiatric distur-
bances (Wolfe, 1985).
On the other hand, Gelles (1973) proposed that societal factors are
the primary causes of abuse and neglect. Within this framework, it is
posited that maltreatment stems from stress engendered by poverty,
educational disadvantage, and cultural sanctioning of physical punish-
ment. Moreover, Gelles (1973) addressed the intergenerational transmis-
sion of child abuse and outlined factors that led many abused children to
grow up to become abusive parents. This model remains a major contri-
bution to the understanding of the causes of child abuse and neglect,
and has received compelling empirical support (e.g., Garbarino, 1976).
However, it is clear that societal elements alone are neither necessary nor
sufficient to bring about child maltreatment. Evidence for this derives
from the fact that most families experiencing economic hardships do not
engage in abuse and neglect. Similarly, child maltreatment occurs at all
levels of socioeconomic status. Furthermore, the extent of intergenera-
tional transmission of maltreatment does not appear to be as extensive
as was previously posited (Kaufman & Zigler, 1987).
The next significant theoretical development in the field was Parke
and Collmer's (1975) treatise hypothesizing that maltreatment was best
understood within the context of the parent-child relationship. Specifi-
cally, they contended that maltreatment was most likely to occur in
particular situations (e.g ., parent-child conflict) in which both parent
(e.g ., poor behavior management skills) and child (e.g ., oppositionality)
characteristics contribute to abuse. Thus, the interaction of parent, child,
and situational variables combine and interact to result in domestic vio-
lence. A large body of data has accrued supporting the social-situational
approach (see Wolfe, 1987) and forming the basis of subsequent models
of maltreatment.
Parke and ColImer's (1975) model suggests that the child can playa
6 ROBERT T. AMMERMAN and MICHEL HERSEN

role in the development and maintenance of abuse. In their review of

the literature up to that time, Friedrich and Boriskin (1976) speculated
that certain child characteristics increased the risk of maltreatment. Con-
tributing child risk factors consisted of prematurity, low birthweight,
mental retardation, being unwanted, and the presence of a physical
handicap. Other researchers (e.g., deLissovoy, 1979) broadened this
supposition to include attributes that made the child more difficult to
manage and, thus, "abuse-provoking." These factors include non-
compliance, oppositionality, acting out, and hyperactivity. The impor-
tance of child variables in determining maltreatment risk has since be-
come a matter of dispute. Although some authors suggest that certain
selected child characteristics may contribute to the development of mal-
treatment (Ammerman, Van Hasselt, & Hersen, 1988), others reject this
association (Starr, Dietrich, Fischhoff, Ceresnie, & Zweier, 1984).
Belsky (1980) provided one of the first integrations of causative
factors into a multidimensional model. His ecological formulation de-
scribes the mutual influences of individual, situational, and societal var-
iables in determining abuse and neglect. By identifying multiple ele-
ments of causative determination, he takes into account the finding that
no single variable fully explains the development of maltreatment. Fur-
thermore, this model reveals several levels of dysfunction that can be
targeted for remedial intervention or prevention. Although subsequent
formulations also use a multidimensional approach (e.g ., Starr, 1988),
Belsky's (1980) model can be credited with stimulating increased and
more sophisticated theoretical and empirical efforts in this area.
The above articles do not constitute an exhaustive list of important
contributions to the field of child abuse and neglect, but rather highlight
salient progressions in the development of theory and research. More-
over, they underscore four critical issues that continue to challenge in-
vestigators: (1) the problem of objectively defining abuse and neglect,
(2) determining the consequences and effects of maltreatment on chil-
dren, (3) identifying causes and risk factors for abuse and neglect, and
(4) using etiological models to develop effective treatments and preven-
tion programs. The remainder of this chapter will review the current
status of knowledge regarding these issues. In addition, future direc-
tions that empirical efforts might take in these areas will be presented.

THE PROBLEM OF DEFINITION

The study of child abuse and neglect has been impeded by the
failure to establish a consensus regarding an objective operational defi-
RESEARCH: CURRENT STATUS AND THE FUTURE 7

nition of what constitutes maltreatment. Unfortunately, the diverse and,

at times, competing needs of the fields involved in child abuse and
neglect (e.g., criminology, public health, psychology, psychiatry, so-
ciology) preclude reaching an agreement on acceptable criteria. Further-
more, by its very nature, child maltreatment is an ambiguous and elu-
sive construct that defies precise explication.
Starr (1988) stated that there are four elements involved in defining
child maltreatment: (1) the intentionality of the act, (2) the impact of the
act on the child, (3) value judgments about the act, and (4) the cultural
and societal standard upon which the act is evaluated. The fluctuating
nature of these mediating factors differentially affects attempts to arrive
at a universally acknowledged definition. Moreover, this task is further
confounded by the private and socially undesirable nature of abuse and
neglect. Information gathered typically derives from self-report or tan-
gential evidence, both of which are susceptible to error, distortion, and
confabulation.
Initially, the "battered child syndrome" (Kempe et al., 1962) was
identified via objective evidence secondary to physical injury (i.e.,
burns, multiple fractures, bruises). Likewise, severe cases of neglect
typically led to pervasive and observable negative consequences, such
as poor hygiene, failure to thrive, and even death. Most instances of
maltreatment, however, are not so severe as to result in such tangible
consequences. In fact, 88% of reported cases of physical abuse involve
minor physical injuries that usually do not result in permanent damage
(American Humane Association, 1984). In addition, most instances of
maltreatment consist of neglect which, in its less severe form, is largely
dependent on the judgment of the professional involved. Thus, in the
majority of cases, physical evidence will be sufficiently unclear or absent
as to cast doubt upon the occurrence of abuse or neglect.
Psychological abuse also complicates efforts to define maltreatment
clearly. Psychological or emotional maltreatment comprises repeated
verbal assaults and manipulations that can lead to lowered self-esteem
in the child (Garbarino & Vondra, 1987). Although victims of psychologi-
cal abuse are often in need of therapeutic intervention, the construct has
proven to be difficult to isolate and define succinctly.
Problems in defining child maltreatment affect researchers and ser-
vice providers alike . Researchers select their samples based upon criteria
that differ from study to study. The criteria used in these investigations
include parents who have engaged in substantiated incidents of mal-
treatment, those who are referred to child protective service agencies
because of suspicions regarding maltreatment, and those who are
judged "at risk, " based upon various standards determined by case-
8 ROBERT T. AMMERMAN and MICHEL HERSEN

workers and/or other clinicians. As a result, comparability across studies

is limited. On the other hand, lawyers and criminal justice workers are
more likely to emphasize physical evidence in determining the existence
of maltreatment, as compared to the more vague definitions often used
by social science researchers. These discrepancies have impacted upon
attempts to reform legal guidelines regarding abuse and neglect (Mac-
Murray & Carson, in press) as well as research efforts.
In summary, arriving at a universally accepted operational defini-
tion of maltreatment has proven to be a dilemma that is virtually insur-
mountable. The primary reasons for this finding include the private
nature of abuse and neglect resulting in tangential and unclear indica-
tors, and the reliance on fluctuating community-based standards and
judgments regarding maltreatment. From a research perspective, how-
ever, all is not lost. Comparability and generalizability of studies is en-
hanced with clear descriptions of subject characteristics and recruitment
methods. Although vague descriptions of methods have been the rule
rather than the exception in child abuse and neglect research (Plotkin,
Azar, Twentyman, & Perri, 1981), recent investigations provide more
detailed information. Particularly noteworthy are the research efforts of
Lutzker and his colleagues, whose subject descriptions are exemplary in
their thoroughness and clarity (e.g., Barone, Green, & Lutzker, 1986;
Lutzker, Megson, Webb, & Dachman, 1985).

CONSEQUENCES OF MALTREATMENT
Elmer and Gregg's (1967) seminal investigation suggested that the
psychosocial effects of maltreatment on children are severe and per-
vasive. Since that time, extensive research has been conducted which
has examined the sequelae of abuse and neglect. These studies have
documented a variety of deficits and dysfunctions in maltreated chil-
dren, including insecure attachment formation (Egeland & Sroufe,
1981), depression (Kazdin, Moser, Colbus, & Bell, 1985), anxiety (Green,
1978), conduct disturbance (McCord, 1983), poor peer relations (Bousha
& Twentyman, 1984), academic underachievement (Morgan, 1979), and
intellectual deficits (Hoffman-Plotkin, & Twentyman, 1984). In general,
maltreated children are heterogeneous in terms of type and in severity
of psychopathology. No syndrome has been identified that is specific to
abused or neglected children, and no symptom is common to all victims
of domestic assault (with the possible exception of insecure mother-
infant attachment in maltreated infants). Although it is evident that the
RESEARCH : CURRENT STATUS AND THE FUTURE 9

sequelae of maltreatment are deleterious, predicting their effects is im-

possible at this time (Ammerman et al., 1986).
The varied clinical presentations of child victims of maltreatment
are the result of several factors . First, the topographical characteristics of
maltreatment differ from case to case; for example, some children only
experience abuse or neglect, while others are abused and neglected.
Other features of maltreatment that differentially mediate its effects con-
sist of type of assault, severity of mistreatment, frequency of abuse, the
situational context in which the assault takes place, the age of onset, and
the length of maltreatment. Thus, it is possible that prolonged and fre-
quent occurrences of relatively minor abuse are more highly associated
with negative outcomes than a single-incident, severe assault. Although
this is a widely held belief, there is little empirical research that examines
the differential effects of varied forms of maltreatment.
A second determinant of the consequences of maltreatment is the
biopsychological status of the child prior to abuse and/or neglect. Fac-
tors that enhance resiliency will mitigate negative sequelae and protect
the child from long-term dysfunction. In similar fashion, increased vul-
nerability will maximize the likelihood that the maltreatment will have a
detrimental effect. Resiliency and vulnerability, in turn, are determined
by genetic factors, learning history, and ecological influences. The ef-
fects of maltreatment also are moderated by the developmental level of
the child. Disruption and delay during critical developmental periods in
mastering tasks involving social, affective, and cognitive skills and abili-
ties will profoundly affect subsequent psychosocial growth. Indeed, re-
cent findings confirm that children who are maltreated during infancy
experience pervasive problems in a variety of developmental areas in
later years (Egeland, Sroufe, & Erickson, 1983).
Although it is generally accepted that the features of maltreatment
in combination with child characteristics will profoundly affect outcome,
few investigations take such factors into account in evaluating the conse-
quences of abuse and neglect. The reasons being (1) such data are ex-
tremely difficult to collect in posthoc studies, given their overreliance on
error-prone parent reports, and (2) even when this information is avail-
able, it is very difficult to form homogeneous subgroups within subject
samples for statistical comparison given recruitment limitations. How-
ever, researchers can address this problem in other ways. First, more
detailed information about maltreatment characteristics should be col-
lected and reported in the subject descriptions. Such instruments as the
Child Abuse and Neglect Interview Schedule (CANIS) (Ammerman,
Hersen, & Van Hasselt, 1987), a semistructured interview assessment of
10 ROBERf T. AMMERMAN and MICHEL HERSEN

parents, can facilitate the gathering of detailed aspects of maltreatment.

Second, broad distinctions between subgroups of maltreated samples
can be made. For example, it was not until recently that researchers
routinely differentiated abused from neglected children. Also, Kazdin
and his colleagues (1985) contrasted abused children who had experi-
enced past abuse, current abuse, and past and current abuse, whereas
Kinard (1980, 1982) examined the effects of abuse characteristics on self-
concept. Such studies as these are necessary if we are to isolate the
relationship between types of maltreatment and their subsequent conse-
quences.
Another problem in delineating the effects of abuse and neglect is
the separation of maltreatment per se from other competing explanations
of psychosocial dysfunction. Abuse and neglect rarely occur indepen-
dently of other forms of family and individual psychopathology. Indeed,
most theorists view maltreatment as a symptom of more global family
dysfunction (Kelly, 1983;Starr, 1988; Wolfe, 1987). Therefore, identifying
the effects of maltreatment independent of prolonged disturbances in
family function is extremely difficult. A related concern is to determine
cause and effect in maltreatment. Because much of the research in child
abuse is retrospective, it is unclear to what extent psychopathology
precedes or is caused by maltreatment. The most acceptable solution to
these problems is the use of prospective longitudinal research designs.
Such experimental strategies allow for the observation of the unfolding
process of maltreatment and have yielded fruitful results when applied
(Egeland & Brunnquell, 1979; Egeland et al., 1983).

THE SEARCH FOR RISK FACTORS

From the early stages of research in child abuse and neglect, it was
realized that delineating risk factors for maltreatment is of paramount
importance. Identifying characteristics associated with risk is crucial in
(1) the recognition and screening of maltreated children, and in (2) tar-
geting populations in need of preventative intervention before abuse
and neglect take place. In addition, the search for risk factors is intri-
cately linked with the understanding of the etiology of child maltreat-
ment.
As previously mentioned, initial formulations of risk focused on
parental psychopathology. According to this approach, severe psychi-
atric disturbance (e.g., psychosis, antisocial personality) in perpetrators
RESEARCH: CURRENT STATUS AND THE FUTURE 11

was thought to be the primary cause of maltreatment. Although recent

evidence strongly disputes the role of severe parental psychopathology
in maltreatment (see Wolfe, 1985), attention continues to be directed
toward parental symptoms as risk factors . Indeed, a large proportion of
research in child maltreatment has emphasized the study of parental
characteristics. In general, these investigations have revealed a variety
of features that distinguish maltreating from adequate care parents, in-
cluding low-frustration tolerance, inappropriate expression of anger, so-
cial isolation, impaired parenting skills, unrealistic expectations of chil-
dren, and a sense of incompetence in parenting (Wolfe, 1987). Research
on parental contributions to maltreatment have produced useful results.
Indeed, because maltreatment typically comprises the commission
and/or omission of specific acts on the part of the parent or other adults,
it is evident that parents should be the primary focus of attention in risk
research.
However, several important limitations exist in the information
gathered to date on this topic. First, although child abuse and neglect
are committed by mothers, fathers, and non-parental figures, and oc-
curs in all socioeconomic groups, research efforts have almost ex-
clusively examined mothers from low SES backgrounds (Ammerman,
1989; Fantuzzo & Twentyman, 1986; Wolfe, 1987). This restricted range
of investigation and sampling bias precludes a full understanding of the
role of parental characteristics in child maltreatment. In particular, there
is an urgent need to examine fathers implicated in maltreatment, given
that they are involved in a significant proportion of substantiated re-
ported cases and are more likely to inflict serious injury on maltreatment
victims. A second concern is the relative paucity of prospective longitu-
dinal research in the study of maltreatment. The bulk of empirical efforts
in this area consist of post hoc comparisons of maltreating and non-
maltreating parents. Thus, although it is clear that many abusive parents
have poor parenting skills, it is not clear to what extent parents with
these characteristics eventually become abusive and/or neglectful them-
selves. Such a distinction is critical for the development of prevention
programs, because it is appropriate to target parents with poor parent-
ing skills as a high-risk group only if a large percentage of them are likely
to become maltreating. An important exception to this shortcoming in
the literature is the work of Egeland and his colleagues (Egeland,
Breitenbucher, & Rosenberg, 1980). Their prospective longitudinal re-
search has done much to further our understanding of factors involved
in the development of maltreatment. For example, Egeland et al. (1980)
identified stress as an important determinant of maltreatment, es-
12 ROBERT T. AMMERMAN and MICHEL HERSEN

pecially in mothers who lack parenting skills and knowledge of children.

Their empirical efforts, however, await replication and expansion.
Although child abuse and neglect occur at all socioeconomic levels,
it has been noted for some time that a disproportionate percentage of
maltreated children come from lower socioeconomic status (SES) fami-
lies. Indeed, it seems logical that violence may emerge when individuals
are confronted with severe and prolonged stress and hardship in the
form of economic disadvantage, poverty, limited educational oppor-
tunity, and unemployment. The focus on societal risk factors has pro-
vided much insight into the antecedents of maltreatment. Indeed, Gar-
barino (1977) concludes from his research that such stressors in combi-
nation with social isolation lead to maltreatment. Despite the compelling
evidence for the salience of social factors in maltreatment, such variables
are of limited utility in determining risk status. As with parenting char-
acteristics in abuse and neglect, the majority of economically disadvan-
taged families do not engage in maltreatment. Thus, viewing this popula-
tion as being at high risk for abuse or neglect would result in a large
number of inaccurately labeled families. Societal factors alone, therefore,
are inadequate risk markers despite their prominent role in the origins of
maltreatment.
The employment of child characteristics as risk factors has long
been a source of speculation (Friedrich & Boriskin, 1976). In particular,
samples of abused and neglected children contain a disproportionate
amount of children with (1) prematurity, (2) low birthweight, (3) mental
retardation, and (4) physical and/or sensory handicaps (see Ammerman
et al., 1988). In addition, difficult to manage children (e.g., those with
Attention Deficit Hyperactivity Disorder) have been posited to be at
high risk for maltreatment. The processes through which children might
playa role in the development of abuse is unclear, although disruptions
in mother-infant attachment (Ainsworth, 1980), increased stress engen-
dered by difficult to manage children (Ammerman et al., 1988), physio-
logical arousal (Frodi, 1981), and heightened vulnerability to maltreat-
ment (Morgan, 1987) have been proposed as possible explanations.
Unfortunately, the majority of investigations of this topic utilize retro-
spective designs that have yielded unclear findings (see Starr et al.,
1984). Moreover, the few prospective and longitudinal studies that have
been conducted do not indicate that child characteristics are significantly
involved in the development of abuse (Egeland & Brunnquell, 1979).
There is convincing evidence, however, that oppositional and defiant
children are involved in the exacerbation of parent-child conflict that
can lead to an abusive incident (Loeber, Felton , & Reid, 1984). Also,
some authors propose that certain populations of children (e.g., se-
RESEARCH: CURRENT STATUS AND THE FUTURE 13

verely disabled, aggressive and oppositional) may be at heightened risk

for abuse (Ammerman, Hersen, Van Hasselt, McGonigle, & Lubetsky,
1989), although such hypotheses await further empirical examination.
In summary, the search for risk factors in maltreatment have em-
phasized the contributions of parental, societal, and child charac-
teristics. Although each of these provides a partial explanation of the
etiology of maltreatment, and therefore suggests appropriate markers
for risk, no one variable is sufficiently sensitive or specific to be used in
the reliable identification of high-risk groups. Because of limited re-
sources, it is virtually impossible to employ such broad criteria in accu-
rately targeting populations for preventative interventions.
In response to the above problems in delineating risk factors for
maltreatment, a number of authors have proposed multicomponent
models of abuse and neglect (Belsky, 1980; Starr, 1988). These formula-
tions acknowledge the varied causative paths that lead to the develop-
ment of maltreatment. Although highlighting the importance of consid-
ering multiple levels of influence in etiology, these conceptualizations
do not elucidate how contributing factors combine to bring about high
risk of abuse and neglect. Understanding the processes involved in
domestic violence, however, is crucial to the discovery of reliable risk
markers.
Recent theoretical efforts have addressed issues of process in child
maltreatment. Indeed, Burgess and Drapier (1988) view the interactional
elements in family violence as critical in differentiating abusive from
nonabusive problem families . In addition, of particular value is Wolfe's
(1987) Transitional Model, in which abuse is conceptualized as growing
out of three tiers of escalating conflict. Specifically, these are the three
stages through which families pass in the development of violent do-
mestic conflict: (1) Reduced Tolerance of Stress and Disinhibition of Ag-
gression, (2) Poor Management of Acute Crises and Provocation, and (3)
Habitual Patterns of Arousal and Aggression with Family Members.
Within each stage there are destabilizing factors that increase the like-
lihood of aggression, and compensatory factors that mitigate the escala-
tion of conflict. A variety of combinations of destabilizing and compen-
satory factors are possible that serve to promote or inhibit the probability
of abuse. This model is the first to propose the specific elements that can
contribute to an increase or lessening of risk for abuse. It also delineates
how diverse interactions of variables can combine to form multiple path-
ways to maltreatment. Most importantly, the utility of the Transitional
Model can be examined empirically. Future progress in the field will
largely depend upon formulations such as this one, and their subse-
quent evaluation.
14 ROBERT T. AMMERMAN and MICHEL HERSEN

TREATMENT AND PREVENTION

One of the most significant developments to emerge in the past
decade is a growing literature on the treatment and prevention of abuse
and neglect. The need for remedial programs for maltreating families
was recognized early, and a good deal of attention has since focused on
perpetrators (see Ammerman, 1989). More recently, a number of authors
have identified the implementation and evaluation of prevention pro-
grams as the single most important objective in the area of child abuse
and neglect (Starr, 1988).
At this point, the primary form of intervention with maltreating
families is medical and legal. As many children first come to the atten-
tion of medical professionals because of injuries stemming from mal-
treatment, physicians and nurses are called upon to provide emergency
services and refer families to appropriate social agencies. Child protec-
tive services carry out preliminary legal interventions and determine the
child's disposition based upon safety concerns. However, such medical
and legal actions are predominantly reactive and crisis oriented. The
long-term functioning of maltreated children and their families depends.
upon remedial programs designed to prevent the recurrence of maltreat-
ment and enhance overall family adjustment.
Initial treatment efforts with perpetrators involved psychodynamic
interventions based upon the premise that maltreating parents suffered
from severe psychiatric disorders (see Kelly, 1983). Other programs, de-
rived from findings implicating social factors and stress as causes of
maltreatment, consisted of providing abusive and neglectful parents
with in-home counselors to provide support and guidance in parenting.
To a large degree, these approaches have received little empirical evalua-
tion, although some evidence supports the use of in-home staff in pre-
venting further maltreatment (see Rosenberg & Reppucci, 1985).
The greatest gains in the development of effective treatments have
come from behavior therapy. These interventions grow out of the ap-
plication of social learning theory to maltreatment and the expanding
literature documenting a variety of psychosocial skills deficits in abusive
and neglectful parents (see Ammerman, 1989). Behavioral interventions
are competency-based and emphasize the acquisition of skills needed for
effective parenting (Kelly, 1983). Typical treatment components include
training in anger control, child management skills, stress reduction, and
(in the case of neglect) home safety skills. Most importantly, these thera-
peutic approaches have been subjected to scientific scrutiny. On the
whole, treatment outcome studies have demonstrated the short-term
efficacy of these approaches to reduce recurrence of maltreatment and
RESEARCH: CURRENT STATUS AND THE FUTURE 15

enhance overall family functioning. Several questions, however, need to

be addressed in future clinical research investigations. First, it is unclear
if treatment gains are maintained for longer than one year. Although
recidivism data indicate a moderate decrease in maltreatment recurrence
in treated relative to nontreated abusive and neglectful families (Lutzker
& Rice, 1987), long-term follow-ups of more than 1 year for individual
families have yet to be conducted. And second, research in this area has
not evaluated if behavior therapy has broader beneficial effect beyond
the specific areas targeted for change. Because maltreatment is acknowl-
edged to be a reflection of more pervasive family dysfunction, it is
imperative that future investigations conduct more comprehensive as-
sessments in order to elucidate the full effects of behavioral
interventions.
A relatively recent development is the treatment of child victims of
abuse and maltreatment. This population has largely been ignored in
the empirical literature, although anecdotal suggestions for treatment
abound (Walker, Bonner, & Kaufman, 1988). Fantuzzo and his col-
leagues (Fantuzzo, [urecic, Stovall, Hightower, Goins, & Schachtel,
1988), however, have demonstrated the usefulness of peer prompts to
improve the social functioning of withdrawn abused children. Other
areas of child dysfunction have not yet been investigated. Continued
efforts in this direction are needed if we are to provide optimal care to
maltreated children.
Finally, as previously mentioned, the creation and examination of
preventative programs are critical for the next decade of child abuse
research. Lutzker has led the way in this area with Project 12-Ways, a
multicomponent secondary prevention program designed to help mal-
treating families and prevent recurrence of abuse and neglect (Lutzker,
1984). Short-term efficacy of Project 12-Ways has been documented
(Lutzker, Campbell & Watson-Perczel, 1984), and long-term prevention
appears to be quite promising (Lutzker & Rice, 1987). Moreover, his
efforts, have stimulated similar programs with populations thought to
be at high-risk for maltreatment (Ammerman, 1988).
Others have strongly advocated the use of primary prevention ap-
proaches to reach a larger population at risk for engaging in maltreat-
ment (Starr, 1988). Even though such efforts are the major priorities for
the future, empirical data regarding their efficacy are unavailable at this
time . Moreover, as the debate continues over appropriate risk factors for
maltreatment, it will be necessary to identify those populations that are
most in need for preventative intervention. Wolfe (1987) suggests a mul-
tilevel prevention approach in which subpopulations at risk are recog-
nized and given appropriate programs relevant to their specific areas of
16 ROBERT T. AMMERMAN and MICHEL HERSEN

need. Thus, parents experiencing frequent and severe conflicts with

their children may require group or individual training in behavior man-
agement skills. Socially isolated parents, on the other hand, may war-
rant community outreach or educational programs. According to this
strategy, resources are allocated based upon level of risk and area of
need. Given the heterogeneity of maltreating families, it is most likely
that prevention programs emphasizing varied target populations and
levels of intervention will have the greatest likelihood of success. Such
efforts, however, await empirical investigation.

SUMMARY
It is encouraging that both the public and the research community
have recently taken an active interest in child abuse and neglect. The past
25 years have seen tremendous growth in scientific investigation, and this
has led to a complementary increase in our understanding of the anteced-
ents and consequences of maltreatment. Although plagued by methodo-
logical impediments, research in child abuse and neglect has elucidated
the contributing factors in etiology. In addition, promising developments
in the treatment of perpetrators and child victims have emerged. The
future holds promise, providing that researchers attend to the multivari-
ate nature of maltreatment, and that they will direct their efforts toward
the development, implementation, and empirical evaluation of compre-
hensive assessment, treatment, and prevention strategies.

ACKNOWLEDGMENTS

Preparation of this chapter was facilitated in part by grant No.

G008720109 from the National Institute on Disabilities and Rehabilita-
tion Research, U. S. Department of Education, and a grant from the Vira
I. Heinz Endowment. However, the opinions reflected herein do not
necessarily reflect the position of policy of the U. S. Department of
Education or the Vira I. Heinz Endowment, and no official endorsement
should be inferred. The authors wish to thank Mary [o Horgan for her
assistance in preparation of the manuscript.

REFERENCES
Ainsworth, M. D. (1980). Attachment and child abuse. In G. Gerber, C. Ross, & E. Zigler
(Eds .), Childabuse: An agenda foraction (pp. 35-47). New York:Oxford University Press.
RESEARCH: CURRENT STATUS AND THE FUTURE 17

Am erican Humane Association . (1984). Highlights of official child neglect and abuse reporting
1982. Den ver, CO : Author.
Ammerm an , R. T. (1989). Child abuse and neglect. In M. Hersen (Ed .), Innovations in child
behavior therapy (p p. 353-394). New Yor k: Springer.
Ammerman, R. T. (1988). Preve ntio n of mother-child problem s in families with young
mul tihandicapped children . International Journalof Rehabilitation Research, 11,416-417.
Amme rma n, R. T., Cassisi, J. E., Hersen , M. , & Van Hasselt, V. B. (1986). Con sequences of
ph ysical abus e and neglect in children . Clinical Psychology Review, 6, 291-310.
Amme rma n, R. T., Hersen, M., & Van Hasselt, V. B. (1987). The Child Abuse and Neglect
Interview Schedule(CANIS). Unpublish ed manuscript, Western Pennsylvania School for
Blind Children , Pittsburgh , Pennsylvani a.
Ammerman, R. T., Van Hasselt, V. B., & Hersen, M. (1988). Maltr eatment in handicapped
children: A critical review. Journal of Family Violence, 3, 53-72.
Ammerman, R. T., Hersen , M., Van Hasselt, V. B., McGon igle, J. J., & Lubetsky, M. (1989).
Abu se and neglect in psychiatrically hospitalized multihandicapped children. Child
Abuse and Neglect, 13, 335-343.
Barone, V. J., Green, B. v., & Lutzker, J. R. (1986). Home safety with families being treated
for child abuse and ne glect. Behavior Modification , 10, 93-114.
Belsky, J. (1980). Child maltreatment: An ecological integration. American Psychologist, 35,
320- 335.
Besharov, D. J. (1982). Toward bett er research on child abu se and neglect : Makin g defini-
tional issues an explicit methodological concern. Child Abuse and Neglect, 5, 383-390.
Bou sha, D. M., & Twen tyman, C. T. (1984). Mother-child in teractional sty le in abuse,
neglect , and con trol gro ups : Naturalistic observations in th e home. Child Development,
93, 196-114.
Brassard, M. R., Germain, R., & Hart , S. N. (Eds .). (1987). Psychological maltreatment of
children and youth. New York: Pergam on Press.
Burgess, R. L., & Draper, P. (1988). A biosocial theo ry of family violence: The role of
natural selection, ecological instability, and coer cive in terpe rso nal contingencies. In L.
Ohl in & M. H . Ton ry (Eds .), Crime and justice-an annual review of research: Family
violence. Chicago : Univers ity of Chicago Press.
del.issovoy, V. (1979). Towa rd the definition of "abuse pro voking child ." Child Abuse and
Neglect, 3, 341- 350.
Egeland, B., Breitenbucher, M., & Rosenberg, D. (1980). Prosp ective study of significance
of etiology of child abu se. Journal of Consulting and Clinical Psychology, 48, 195-205.
Egeland, B., & Brunnquell, D. (1979). An at-risk approach to th e study of child abuse .
Journal of the American Academy of Child Psychiatry, 18, 219-236.
Egeland, B., & Sroufe, L. A. (1981). Att achment and early maltreatm ent. Child Develop-
ment, 52, 44-52.
Egeland, B., Sroufe, L. A., & Erickson , M. (1983). The developmental conse quences of
different patterns of maltreatment. Child Abuse and Neglect, 7, 459-469.
Elmer, E. (1977). Fragile families, troubled children: Theaftermath of infant trauma . Pittsburgh :
University of Pitt sburgh Press.
Elmer, E., & Gregg, G. S. (1967). Developmental characteristics of abu sed children. Pedi-
atrics, 40, 596-602 .
Fantuzzo, J. W., [urecic, L., Stovall, A., Hightower, A. D., Goins , C,; & Schachtel, D.
(1988). Effects of adult and peer social initiations on the social behavior of withdrawn,
malt reated presch ool children . Journal of Consulting and Clinical Psychology, 56, 34-39.
Fantuzzo, J. W., & Twen tyman, C. T. (1986). Ch ild abu se and psychotherap y research:
18 ROBERT T. AMMERMAN and MICHEL HERSEN

Merging social concerns and empirical investigation . Professional Psychology: Research

and Practice, 17, 375-380 .
Friedrich, W. N., & Boriskin, J. A. (1976). The role of the child in abu se: A review of the
literature. American Journal of Orthopsychiatry, 46, 580-590 .
Friedrich, W. N., & Einbender, A. J. (1983). The abused child : A psychological review.
Journal of Clinical Child Psychology, 12, 244-256.
Frodi, A. M. (1981). Contribution of infant characteristics to child abuse. American Journal of
Mental Deficiency, 85, 341-349 .
Gaines, R., Sandgrund, A. , Green, A. H., & Power, E. (1978). Etiological factors in child
maltreatment: A multivariate study of abusing, neglecting, and normal mothers. Jour-
nal of Abnormal Psychology, 87, 531-540 .
Garbarino, J. (1976). A preliminary study of some ecological correlates of child abuse: The
impact of socioeconomic stress on mothers. Child Development, 47, 178-185 .
Garbarino, J. (1977). The human ecology of child maltreatment: A conceptual model for
research. Journal of Marriage and the Family, 39, 721-735 .
Garbarino, J., & Vondra, J. (1987). Psychological maltreatment: Issues and perspectives. In
M. R. Brassard, R. Germain, & S. N . Hart (Eds .), Psychological maltreatment of children
and youth (pp. 25-44). New York: Pergamon Press .
Gelles, R. J. (1973). Child abuse as psychopathology: A sociological critique and reformula-
tion . American Journal of Orthopsychiatry, 43, 611-621.
Green, A. H . (1978). Psychopathology of abused children. Journal of the American Academyof
Child Psychiatry, 17, 92-103.
Hoffman-Plotkin, D., & Twentyman, C. T. (1984). A multimodel assessment of behavioral
and cogn itive deficits in abused and neglected preschoolers. Child Development, 55,
794-802.
Kaufman, J., & Zigler, E. (1987). Do abusive children become abusive parents? American
Journal of Orthopsychiatry, 57, 186-192 .
Kazdin, A. E., Moser, J., Colbus, D., & Bell, R. (1985). Depressive symptoms among
physically abused and psychiatrically disturbed children. Journal of Consulting and
Clinical Psychology, 94, 298-307 .
Kelly, J. A. (1983). Treating childabusive families : Intervention based on skills-training principles.
New York: Plenum Press.
Kempe , C. H., Silverman, F. N ., Steele , B. F., Droegemueller, W., & Silver, H . K. (1962).
The battered child syndrome. Journal of the American Medical Association, 181, 105-112.
Kinard , E. M. (1980). Emotional development in physically abused children. American
Journal of Orthopsychiatry, 50, 686-696.
Kinard, E. M. (1982). Experiencing child abuse. Effects on emotional adjustment. American
Journal of Orthopsychiatry, 52, 82-91.
Loeber, R., Felton, D. K., & Reid, J. (1984). A social learning approach to the reduction of
coercive processes in child abusive families : A molecular analysis. Advances in Behavior
Research and Therapy, 6, 29-45 .
Lutzker, J. R. (1984). Project 12-Ways: Treating child abuse and neglect from an eco-
behavioral perspective. In R. F. Dangel & R. A. Polster (Eds .), Parent training: Founda-
tions of research and practice (pp . 260-297). New York: Guilford Press.
Lutzker, J. R. Campbell, R. V., & Watson-Perczel, M. (1984). Using the case study method
to treat several problems in a family indicated for child neglect. Education and Treatment
of Children, 7, 315-333 .
Lutzker, J. R., Megson, D. A. , Webb, M. E., & Dachman, R. S. (1985). Validating and
training adult-child interaction skills to professionals and to parents indicated for child
abuse and neglect. Journal of Child and Adolescent Psychotherapy, 2, 91-104.
RESEARCH: CURRENT STATUS AND THE FUTURE 19

Lutzker, J. R., & Rice, J. M. (1987). Using recidivism data to evaluate Project 12-Ways: An
ecobehavioral approach to the treatment and prevention of child abuse and neglect.
Journal of Family Violence, 2, 283-290.
MacMurray, B. K., & Carson, B. A. (in press). Legal issues in violence toward children. In
R. T. Ammerman & M. Hersen (Eds.), Case studies in familyviolence. New York: Plenum
Press.
McCord, J. (1983). A 4O-year perspective on effects of child abuse and neglect. Child Abuse
and Neglect, 7, 265-270.
Morgan, S. R. (1979). Psycho-educational profile of emotionally disturbed abused children.
Journal of Clinical Child Psychology, 8, 3-6.
Morgan, S. R. (1987). Abuse and neglect of handicapped children. Boston: Little, Brown.
Parke, R. D., & Collrner, C. W. (1975). Child abuse: An interdisciplinary analysis. In E. M.
Hetherington (Ed.), Review of childdevelopment research (Vol. 5, pp. 509-590). Chicago:
University of Chicago Press.
Plotkin, R. c., Azar, S., Twentyman, C. T., & Perri, M. G. (1981). A critical evaluation of
the research methodology employed in the investigation of causative factors of child
abuse and neglect. Child Abuse and Neglect, 5, 449-455.
Rosenberg, M. S., & Reppucci, N . D. (1985). Primary prevention of child abuse. Journal of
Consulting and Clinical Psychology, 53, 576-585.
Spinetta, J. J., & Rigler, D. (1972). The child abusing parent: A psychological review.
Psychological Bulletin, 77, 296-304.
Starr, R. H., Jr. (1988). Physical abuse of children. In V. B. Van Hasselt, R. L. Morrison, A.
S. Bellack, & M . Hersen (Eds .), Handbook of family violence (pp. 119-155). New York:
Plenum Press.
Starr, R. H., Dietrich, K. N. , Fischhoff, J., Ceresnie, S., & Zweier, D. (1984). The contribu-
tion of handicapping conditions to child abuse. Topics in Early Childhood Special Educa-
tion, 4, 55-69.
Walker, C. E., Bonner, B. L., & Kaufman, K. L. (1988). The physically and sexually abused
child: Evaluation and treatment. New York: Pergamon Press.
Wolfe, D. A. (1985). Child abusive parents: An empirical review and analysis. Psychological
Bulletin, 97, 462-482.
Wolfe, D. A. (1987). Childabuse: Implications for child development and psychopathology. New-
bury Park, CA: Sage Publications.
 PART II

GENERAL ISSUES
 CHAPTER 2

THE EPIDEMIOLOGY OF
CHILD MALTREATMENT
RAYMOND H. STARR, JR., HOWARD DUBOWITZ, AND
BEVERLY A. BUSH

INTRODUCTION
Articles about child maltreatment appear daily in almost every metro-
politan newspaper. It seems that children are being injured, molested,
and even killed at an alarming rate. In an attempt to understand child
maltreatment, the first question to ask is what do we really know about
the extent of the problem? Answering that question is the purpose of
this chapter.
The first part of this chapter deals with the critical need to define
child maltreatment. Epidemiology cannot be considered independently
of the definition of differing types of abuse and neglect. The next section
of the chapter summarizes the results of three different types of study:
(1) analyses of reported maltreatment cases, (2) examinations of both
reported and unreported cases known to professionals, and (3) surveys
of maltreatment among the general public. The implications of these
data are discussed in the final section of the chapter.

RAYMOND H . STARR, JR. AND BEVERLY A. BUSH· Department of Psychology, University

of Maryland-Baltimore County, Catonsville, Maryland 21228. HOWARD DUBOWITZ
• Department of Pediatrics, Uni versity of Maryland School of Medicine, Baltimore, Mary-
land 21201.

23
24 RAYMOND H. STARR, JR., et al.

BASIC EPIDEMIOLOGICAL CONCEPTS

Epidemiology is "the study of the distribution and determinants of
diseases and injuries in human populations" (Mausner & Kramer, 1985,
p. 1). Thus, the field deals not only with issues of the incidence (the rate
at which new cases occur in the population) and prevalence (the number
of cases in a designated population at a given time or over a specific time
period), but also with the risk factors that predispose particular persons
or subgroups of the population to developing the condition of concern
(Last, 1983). The purpose of epidemiological studies is to provide knowl-
edge about disease and injury patterns that will aid in their treatment
and prevention. This chapter focuses on current knowledge concerning
the incidence and prevalence of maltreatment, problems in their deter-
mination, and what is known about broad, demographic risk factors.
Discussion of more specific, individual risk factors will be limited be-
cause they are outlined in detail elsewhere in this volume .

THE DEFINITION OF CHILD MALTREATMENT

A fundamental starting point for examining the etiology of a prob-
lem is to define it. Unless a problem is defined, it will be impossible to
determine its extent. Four factors are involved in defining maltreatment:
(1) the intentionality of the act, (2) the effect of the act on the child, (3)
the value judgment society makes about the act, and (4) the standard
used to make the judgment (Garbarino & Gilliam, 1980).
A lack of definitional clarity in maltreatment research complicates
the task of understanding its epidemiology. The most common defini-
tions of different types of maltreatment are those specified in the Child
Abuse Prevention and Treatment Act (National Center on Child Abuse
and Neglect [NCCAN], 1988). Although most states have adopted these
definitions, some modify them, making the direct comparison of re-
ported cases from state to state difficult. Thus, there are many compet-
ing definitions of child maltreatment, none of which is universally
accepted.
This is so because child abuse, in its various manifestations, resists
easy definition. For example, Swedish law bans "all forms of physical
punishment and other injurious or humiliating treatment of children"
(Radda Barnen, 1980, p. 7) by parents. Most American parents would
find such a law unduly restrictive while Swedish parents would see
many routine acts of U.S. parents as abusive. Definitional differences
occur not only between but within cultures (Korbin, 1987), further com-
plicating the issue.
EPIDEMIOLOGY OF MALTREATMENT 25

Not only are there different definitions of child maltreatment, but

these often vary across professions, researchers, states, and agencies
(Giovannoni & Becerra, 1979). Some of these differences were examined
by Gelles (1982) and by Giovannoni and Becerra (1979) in studies in
which members of various professional disciplines involved in child
protection were asked to designate whether a number of acts were or
were not maltreatment. The studies concluded that professionals do not
use a consistent set of criteria in defining and reporting child maltreat-
ment. However, Gelles concluded that judgments of intentionality were
particularly important, because intentional acts are more likely to be
reported. In spite of this finding, it must be remembered that intent is
particularly difficult to determine. For example, how can intention be
evaluated when significant parental psychopathology is a factor in a case
of abuse?
Legal definitions of maltreatment typically are vague. Thus, the
specific behaviors that constitute maltreatment often are not clearly
stated; and such undefined terms as mental suffering and unfit abound
(Giovannoni & Becerra, 1979). Some experts believe that definitional
vagueness is desirable because it allows social service workers to consid-
er specific, individual details of a particular case . Others think vague-
ness can lead to inconsistent case handling and due process violations
(Valentine, Acuff, Freeman, & Andreas, 1984). At best, a definition can
never be more than a guide. Each case of maltreatment is complex and,
regardless of comprehensiveness, a definition cannot provide a simple
answer to every case.
Because of the primacy of definitional issues, the remainder of this
section consists of a brief presentation of specific issues. Rather than
propose a uniform set of definitions, specific definitional criteria will be
included in the discussion of individual studies of the epidemiology of
maltreatment.
Child maltreatment is usually categorized into three types of abuse
and two types of neglect: physical, psychological, and sexual abuse; and
physical and psychological neglect. The major distinction between
abuse and neglect is that the former typically involves an act of commis-
sion whereas neglect is the result of an omission.

PHYSICAL ABUSE

Physical abuse occurs when a child is injured by a parent or other

caregiver. Beyond this criterion other aspects of definitions are typically
vague. Additional factors that may playa role in defining physical abuse
include parental approaches to discipline, intent to injure, the effect of
an act on the child, and the vulnerability of the child. Some of these
26 RAYMOND H . STARR, JR., et al.

variables, such as intent, are difficult-if not impossible-to assess. This

increases the vagueness of definitions in which intent is a factor. A key
variable that must be examined is whether the intent was to cause pain,
in which case an act may not be considered physically abusive, or to
physically injure, in which case abuse is more likely. In addition, in cases
of repeated injury where parental psychopathology is likely to be pres-
ent, it also is difficult to infer intent because of the psychopathology.
Thus, the criterion of intent to injure, combined with other family cir-
cumstances, makes the definition of physical abuse complex.
As was indicated earlier, definitions of abuse vary between different
professions. For example, a pediatrician may see spanking an infant as
undesirable and possibly abusive and will talk with the parents about
the negative effects of corporal punishment; a protective services worker
might have more stringent criteria requiring bruises or other injuries to
validate a report of physical abuse; and a prosecutor might work only
with even narrower criteria . In addition to varying between the profes-
sions, working definitions also differ considerably within members of
the same profession. It is just this sort of definitional variability that
complicates the study of the epidemiology of child maltreatment.

PHYSICAL NEGLECT

Physical neglect is more difficult to define than physical abuse.

Specific, child-centered criteria for diagnosing a case of neglect usually
are missing. However, the more inadequate a parent's caregiving is, the
more likely it is that parents and professionals will agree that it is phys-
ically neglectful (Polansky & Williams, 1978). For example, there is likely
to be agreement that a child who has not been fed and whose diaper has
not been changed for a day has indeed been neglected. In other, more
ambiguous cases professionals have difficulty deciding if the injury to a
child is the result of an act of omission or commission. For example, do
we classify as physically abused or as physically neglected those chil-
dren who are born with fetal alcohol syndrome or addicted to narcotics?
Furthermore, homeless children represent still another gray area (Du-
bowitz, 1987).

PSYCHOLOGICAL ABUSE AND PSYCHOLOGICAL NEGLECT

Psychological maltreatment is a more abstract concept than physical

maltreatment. Evidence often is intangible and more difficult to attribute
to a specific parental behavior. Definitional variability is common. Some
investigators think that any distinction between psychological abuse
and psychological neglect is artificial (Garbarino, Guttman, & Seeley,
EPIDEMIOLOGY OF MALTREATMENT 27

1986). Garbarino et aI. believe that there is significant damage to a child's

psyche in both cases and, therefore, argue for the primacy of the psyche
in all types of maltreatment. However, the physical and not the psycho-
logical consequences of a parental action or inaction usually cause soci-
ety to label the act as abusive or neglectful. Although the psychological
damage typically lasts longer than the physical injury, all the varying
effects of maltreatment must be considered. In contrast to Garbarino and
coworkers, Whiting (1976) distinguished between psychological abuse
and neglect. For Whiting, psychological abuse is present when parents
cause a child to become emotionally disturbed; psychological neglect
occurs when they refuse to allow their emotionally disturbed child to
receive treatment.
Protective service agencies typically become involved in only the
most severe cases of psychological maltreatment because of the difficulty
in proving that parental acts cause a child to develop patterns of dis -
turbed behavior (Dubowitz, 1987). Typically, psychological maltreatment
is classified as abuse or neglect only when it is extreme, or when other
forms of maltreatment are also present.

SEXUAL ABUSE

Sexual abuse is also difficult to define. This difficulty is partly the

result of societal attitudes concerning sexuality and partly due to the fact
that sexual abuse can involve family members or extrafamilial contacts.
Additionally, there is a lack of consensus on what acts are sexually
abusive. There need not be obvious physical injury, physical contact, or
psychological harm for some definitions to classify a child as sexually
abused. One study evaluated the role of various factors as determinants
of sexual abuse and concluded that, in declining order, the most impor-
tant factors leading those surveyed to see an act as sexual abuse were (1)
the perpetrator's age, (2) the nature of the act, (3) whether the child
consented to the act, (4) the age of the victim, (5) the sex and relatedness
of victim and perpetrator, and (6) the consequences of the act for the
child (Finkelhor, 1979).

SUMMARY

It can be concluded that "a myth of shared meaning surrounds the

general area of child maltreatment and the specific area of child sexual
abuse" (Haugaard & Reppucci, 1988, p. 29). Different professions use
various definitions. Also, there is no consistency in the working defini-
tions used within professions. For example, researchers use definitions
in their studies that vary from those used in legal jurisdictions. These
28 RAYMOND H . STARR, JR., et at.

and other aforementioned problems impede the task of accurately deter-

mining the epidemiology of child maltreatment.

THE INCIDENCE AND PREVALENCE

OF CHILD MALTREATMENT
The epidemiology of child maltreatment has been examined in a
number of ways. At the broadest level there are studies and surveys
tabulating reported cases (American Humane Association, 1984; Ameri-
can Association for Protecting Children [AAPC], 1985, 1986, 1987, 1988;
National Center on Child Abuse and Neglect [NCCAN], 1981, 1988; A.
Russell & Trainor, 1984). More narrowly, there are numerous studies
indicating the incidence of specific types of abuse (Gelles, 1978; Gil,
1973; Kinsey, Pomeroy, Martin, & Gebhard, 1953; Straus & Gelles, 1986;
D. E. H. Russell, 1983, 1984).
It is helpful to consider the samples included in epidemiological
studies as falling on a five-point continuum (NCCAN, 1988). At the first
level there are cases reported to child protective services (CPS). At a
second level there are maltreated children who have been detected by or
referred to other agencies with investigative powers (e.g., police, public
health departments) but who are not officially clas sified as abused or
neglected. Cases of maltreatment known to noninvestigatory agencies
(e.g. , hospitals, mental health centers, schools) and not officially re-
ported constitute the third level. Although these children are maltreat-
ed, professionals often do not file a report for a number of reasons-
such as a belief that they are better able to help the famil y than CPS
workers. At the next level are cases in which a lay person recognizes
maltreatment but does not report it. Finally, at the fifth level, there are
maltreated children who are not recognized as abused or neglected by
anyone.

ANALYSES OF REPORTED MALTREATMENT

Reported cases of child maltreatment, * those at the first level of the

NCCAN (1988) categorization, are an important measure of incidence .

"Strictly speaking, we cannot discuss incidence and prevalence with regard to reported
cases . Incidence figures may be inflated by reports of maltreatment of children already
known to child protective services personnel who do not represent new cases. Similarly,
pr evalence cannot be determined based on reports, only on the total number of cases
currently considered maltreated. For consisten cy, we use the term s rate and incidence in
anal yzing reporting practices in this chapter.
EPIDEMIOLOGY OF MALTREATMENT 29

The American Association for Protecting Children (AAPC) and its par-
ent agency, the American Humane Association, have been funded since
1974 to prepare annual summaries of child maltreatment reports submit-
ted to child protection agencies. The data submitted to the AAPC in-
clude the total number of reports, the data source, and the charac-
teristics of the reporting system. In addition, many state and local
agencies provide case level data.
The use of reported cases to examine child maltreatment incidence
has inherent problems. Foremost among these is reporting bias . For
example, poor and minority families are more likely to be reported as
maltreating (Hampton & Newberger, 1985; O'Toole, Turbett, & Nalepka,
1983; Pelton, 1977) due, in part, to their increased contact with social
service providers. Further evidence suggests that only about 40% of
maltreatment cases are reported to CPS (NCCAN, 1988). Other studies
have examined reporting practices in more detail (e.g., Adams, Barone,
& Tooman, 1982; Gelles, 1982; Knudsen, 1988; Morris, Johnson, &
Clasen, 1985; Newberger, 1983).
Results of the most recent AAPC data-for 1986-indicate that over
two million children (2,086,000) were reported as maltreated to protec-
tive services agencies, a rate of 33 per 1,000 children (AAPC, 1988).
Reports were filed on 1,335,000 families with a mean of 1.6 children per
family reported. These data indicate an 8% increase in reported cases
from 1985 and a 212% increase in the past decade, an average increase of
13% per year. However, these AAPC data need to be interpreted with
caution, because the figures may be overestimates due to the inclusion
of duplicate cases if more than one maltreatment report was filed for a
child in a given calendar year. Alternatively, they may be underesti-
mates because reports on Native Americans are not included by some
states. Regardless of the accuracy of these data, most investigators do
agree that reported cases are only the "tip of the child abuse iceberg."
The AAPC (1988) has also analyzed available data by type of mal-
treatment in those cases in which an investigation indicated that mal-
treatment did indeed occur a. Fluke, personal communication, Novem-
ber 22, 1988) (see Table 1). Most, but not all, of these indicated cases are
substantiated. The AAPC differentiates between indicated and substan-
tiated cases with the former representing a slightly larger class. How-
ever, only some states differentiate between indicated and substantiated
cases.
The AAPC data are based on extrapolations from those states that
provide computerized records of maltreatment reports and that used
similar definitional criteria for the various forms of maltreatment. Phys-
ical neglect (deprivation of necessities) comprised a majority of cases.
30 RAYMOND H. STARR, JR., et aI.

TABLE 1. Summary Profiles for Indicated Maltreatment (1986) a

Maltreatment type
All Phy sical Sexual Psychological
maltreatment abuse Fatalities abuse Neglect maltreatment

Percentage of all 27.6% 15.7% 54.9 % 8.3 %

cases
Rate/l,OOO 12.4 3.5 2.9 6.8 1.1
Child
Age (years) 7.3 8.0 2.8 9.2 6.2 7.9
Sex (male) 46% 51% 54% 23% 52% 48%
Race (white) 67% 68% 53% 77% 63% 77%
Perpetrator
Parent 81% 82% 76% 42% 92% 90%
Other relative 7% 6% 4% 23% 3% 3%
Age (years) 32 32 27 32 31 33
Sex (male) 47% 50% 44% 82% 30% 42%
Caretaker
Single female 25% 25% 24% 24% 51% 34%
Unemployed 35% 29% 26% 42% 33%

«Adapted from AAPC (1988). Indicated cases are tho se where abuse is deemed to have occurred as a
a.
result of inve stigation. Most , but not all, indicated cases are substantiated Fluke, personal com-
munication, November 22, 1988).

Slightly more than a quarter of indicated cases were due to physical

injury, with major injury (e.g., poisoning, fracture, or brain damage)
comprising 3% of cases, minor injury (e.g., bruises, cuts, or shaking) in
14%, and unspecified injuries 11%. According to the data, about one in
six maltreated children was sexually abused. Less than one in ten (8%)
was psychologically maltreated or experienced some other form of mal-
treatment. Some children were classified as having more than one form
of maltreatment. Compared to the years from 1976through 1982 (Russell
& Trainor, 1984), the current data indicate increases in reports of phys-
ical injury and sexual abuse and a decline in reports of physical neglect
and psychological maltreatment.
Data on the victims of maltreatment and their families have also
been examined. The figures differ by type of maltreatment (see Table 1).
Mean child age is youngest for fatalities , where a blow of a certain
intensity is likely to lead to greater physical injury, and oldest for sexual
abuse. In general, there was a tendency for boys to be more maltreated
for all maltreatment types except for sexual abuse where girls predomi-
nated. Black children were relatively more likely to be fatally injured and
less likely to be sexually abused. Parents were most likely to perpetrate
EPIDEMIOLOGY OF MALTREATMENT 31

neglect and least likely to sexually abuse their child. However, other
relatives were involved in almost a quarter of the cases of child sexual
abuse. Males were typically the perpetrator of sexual abuse. Females
were more likely to neglect, psychologically maltreat or murder a child
due to, at least in part, their greater contact with children. Single females
particularly tended to be more neglectful and psychologically maltreat-
ing. Unemployed, single females were at the highest risk for neglect.
This latter finding is not surprising. Such women are at risk for in-
creased reporting because of greater stress, poverty, and depression; not
to mention the bias introduced by negative professional stereotypes and
greater monitoring by social services agencies.
Other reporting trends were also found . Young children from black
families were disproportionately more likely to be reported as abused or
neglected (AAPC, 1988). Overall, 43% of reports in 1986 were for chil-
dren less than 6 years old, with a mean child age of 7.2 years, compared
to a national mean of 8.6 years for all children. Whites were underrepre-
sented in reported cases. They constituted 81% of all U.S. children but
only 66% of maltreatment reports. These reporting trends have existed
since the first analyses of reports in 1976 (AAPC, 1988).
There are still more significant trends in the AAPC (1988) report.
With the increase in recent years in the reporting of sexual abuse there
has been a corresponding decrease in the percentage of all maltreatment
reports involving boys. In 1986, 48% of all reports were for boys, a
decline from 50% ten years earlier. Males were increasingly less likely to
be the caretaker of the reported child (39%) but were disproportionately
more likely to be the perpetrator of maltreatment (44%). The decline in
male caregivers is positively correlated with changes in the percentage
of reports for single-parent, female-headed families (32% in 1986). Al-
most half of all reports (49%) in 1986 were for families who received
public assistance, although they comprise only 12% of families in the
United States. A more detailed analysis of confirmed maltreatment re-
ports from one state's central maltreatment registry supports these find -
ings (Rosenthal, 1988).
Relying on reported, rather than substantiated or indicated, cases to
examine the incidence of child abuse is questionable because allegations
frequently are not substantiated upon CPS investigation. Data for 26
states showed that between 40% and 42% of reports were held to be
valid by the reporting CPS agencies (AAPC, 1988). Using the 40% sub-
stantiation rate and applying it to the 1986 reporting rate of 3311,000, a
substantiated maltreatment rate of approximately 13/1,000 results,
which is similar to the rate of indicated cases of 12/1,000 (see Table 1).
Substantiation rates varied from state to state in 1986 with a range of
32 RAYMOND H. STARR, JR., et at.

23% to 64%, suggesting not only wide variability in definitions but also a
relationship between case validation and number of reports. Nationally,
74% of states have substantiation policies. As might be surmised, these
differ from state to state. Moreover, policies are used in a uniform way in
only one half of the states that have them (Trainor, cited in Russell &
Trainor, 1984). Other factors further influence substantiation rates. For
example, overburdened caseworkers are less likely to classify a given
case as one of maltreatment than are workers with lighter caseloads
(NCCAN, 1981).
Secondary analyses of the American Humane Association and the
AAPC data bases for 1980 and 1983 have also been performed (Max-
imus, Inc ., 1986a,b,c). These support the above conclusion that between
40% and 45% of child abuse and neglect cases are substantiated (Max-
imus, Inc ., 1986a). If the assumption is made that about 42% of abuse
and neglect cases are substantiated, then slightly more than 737,000
children were classified as maltreated in 1986.
Another type of reported case concerns children who are fatally
abused or neglected. Because fatalities are more likely to be officially
reported, data concerning their incidence has been used to analyze tem-
poral changes in child maltreatment. Results of one recent survey indi-
cate that the number of fatalities increased from a projected 899 in 1985
to 1,181 the next year. There was a slight decline to 1,132 in the following
year, 1987 (Daro & Mitchel, 1988). National data were extrapolated from
reports of fatalities from between 34 and 39 states, the number varying
each year.
Again, the counting of fatalities is not so simple as it may seem at
first glance (Mitchel, 1987), because many fatalities due to physical abuse
are misclassified as accidental deaths or sudden infant death syndrome.
The actual incidence of fatalities may therefore be much higher than the
suggested estimates of up to 5,000 a year (Christoffel, Liu, & Stamler,
1981; Mitchel, 1987). In general, studies have found few demographic or
case differences between families in which a child is maltreated but not
fatally injured and those in which the maltreatment is deadly (Mitchell,
1987). Fatalities that are due to maltreatment are a serious, and in all
likelihood an increasing, social problem. It is hoped that better pro-
cedures for securing accurate fatality data will be developed in the near
future so that the exact magnitude can be determined more accurately.
Research also has focused on reporting among more specific sam-
ples . Results of such studies complement the findings of larger, more
comprehensive investigations (Pelton, 1981). In one such study, report-
ing patterns for a single Indiana county over a 20 year period (1965-
1984) were examined (Knudsen, 1988). Demographic characteristics
EPIDEMIOLOGY OF MALTREATMENT 33

were not significantly different from national data for poverty level,
education, occupational status, median age, and sex ratio. The overall
rate of reporting and the percentage of substantiated reports increased
over time and was greater for children less than six years of age. Knud-
sen considers this increase to be the result of a broadening of the defini-
tion of and to an increase in the actual incidence of maltreatment, and
not just the result of increased reporting. The rate of maltreatment sub-
stantiation also increased over time, which would be the case if less
severe forms of maltreatment were being reported. Reports of less se-
vere maltreatment were seen as less likely to be substantiated. In 1984,
the incidence rate for substantiation on a first report of suspected mal-
treatment was more than 10 per 1,000 children. In the last year of his
study, 38% of all reports were for abuse and 62% for neglect. Most
reports were for lack of supervision (38%), followed by physical abuse
(20%), physical neglect and sexual abuse (14% each), other forms of
neglect (10%), and psychological abuse (4%).
A detailed analysis of reported cases in one state, over an 8-year
period, provides further information (Rosenthal, 1988). Although con-
firming other evidence showing that girls were more likely to be sexually
abused than boys, Rosenthal also concluded that boys were more likely
to be severely injured than girls . However, some age differences were
present. For example, when the age of the child was considered, girls
from 13 to 17 years old were more likely to be maltreated, even when
sexual abuse was removed from the analyzed data. In addition, males
were more likely to be the perpetrator with older children, and females
the perpetrator with you nger children. This latter finding is interpreted
as evidence that male teens, who are likely to fight back when struck,
are less likely to be abused and that, when abuse does occur, it is proba-
bly perpetrated by a relatively stronger adult male.
Still other analyses have examined what have been termed the "eco-
logical correlates" of maltreatment (see a review by Zuravin, 1989). Eco-
logical studies, based on the theoretical views of Bronfenbrenner (1977)
and Bronfenbrenner, Moen, and Garbarino (1984), emphasize the rela-
tion of community and environmental characteristics to maltreatment.
In one study, Zuravin (1989) examined the ecology of child maltreatment
in an urban area using reported cases of abuse and neglect as the depen-
dent variable and neighborhood characteristics for individual census
tracts as the independent variable. Eliminating duplicate reports on fam-
ilies, she found that the average abuse incidence for 1983 and 1984 was
23/1,000 families with children. The corresponding figure for neglect
was 26/1,000. The strongest correlates of both abuse and neglect were
low income and the rate of vacant housing in the neighborhood. AI-
34 RAYMOND H . STARR, JR., et al.

though this correlation does not imply a cause and effect relationship, it
does seem likely that poverty and maltreatment are related (Pelton,
1977). The work of Zuravin and others (e.g., Garbarino & Crouter, 1978;
Spearly & Lauderdale, 1983) suggests that adopting an ecological per-
spective should provide information of value in understanding the epi-
demiology of child maltreatment.
In summary, relying on analyses of reported cases to provide reli-
able and valid information on the incidence of maltreatment presents
problems-and foremost, the securing of accurate reports. Unfortunate-
ly, at present it is safe to conclude that relying on substantiated reports
yields an underestimate of the true incidence of abuse and neglect.
Investigators are currently evaluating other approaches to examining
the incidence and prevalence of child maltreatment.

THE NATIONAL INCIDENCE SURVEYS

Two national incidence surveys of professionals and their reporting

practices have been done to clarify incidence issues (NCCAN, 1981,
1988) at reporting Levels 1 (cases known to protective services) through
3 (cases known to professionals in major, noninvestigatory agencies).
The first study involved a probability sample in 26 counties in 10 states
of "community professionals" (NCCAN, 1981). The sample included
CPS staff, and school, hospital, police, and juvenile services personnel
who would be likely to have contact with maltreating families (NCCAN,
1981). Data were collected during 1979 and 1980. The second study,
using a similar sample from 29 counties, was done in 1986 (NCCAN,
1988).
Both studies evaluated the occurrence of six major types and a
number of subcategories of maltreatment: physical abuse (two sub-
types), sexual abuse (six subtypes), and psychological abuse (eight sub-
types); and physical neglect (seven subtypes), psychological neglect
(five subtypes), and educational neglect (three subtypes). Each type of
maltreatment was clearly defined in both studies. The second study
used both the original and a revised, expanded definition (NCCAN,
1988). This allowed for the direct comparison of the 1979-1980 and the
1986 data, using the original definition of demonstrable harm to the
child. The use of the revised definition with the 1986 data allowed for
the inclusion in the incidence data of children who were both en-
dangered and harmed, and for a wider variety of potential perpetrators
for some forms of maltreatment. The revised 1986 definition thus
yielded higher incidence and prevalence figures than did the original
1979-1980 definition.
EPIDEMIOLOGY OF MALTREATMENT 35

Study data analyses consisted of (1) an assessment of countability,

(2) unduplication, and (3) the weighting and estimation of incidence
figures. All case reports were evaluated as to whether they were counta-
ble as cases of maltreatment according to study definitions. Intercoder
reliability for countability was 86% for both sets of definitional criteria.
Data were also reviewed for duplications because of the possibility that
the same case might be known to more than one reporting source. In the
case of duplication, a case was assigned to the highest appropriate level
on the five-level model discussed earlier. Estimates of national incidence
were obtained for each type of maltreatment using a complex weighting
procedure. The results of the two studies are summarized in Table 2.
The National Incidence Studies are not without problems, as is the
case with all incidence studies that have been done to date. Data con-
cerning the incidence of sexual abuse present the major difficulty
(Finkelhor 1984; Finkelhor & Hotaling, 1984). Finkelhor and Hotaling
question the results of the first National Incidence Study indicating that
a disproportionately high percentage of sexual abuse cases had been
officially reported. Fewer cases were listed by individuals who knew
about a case of sexual abuse, but did not file a report. Finkelhor and
Hotaling (1984) cite additional evidence suggesting a low percentage of
officially reported cases and conclude that definitional difficulties are
also present. For example, a mother who allowed a child to be sexually
abused but who did not play an active role in the actual abuse was still
counted as a perpetrator. But, if each parent perpetrated a different type
of maltreatment, both were listed as perpetrators of maltreatment.
These issues complicate the task of determining what parental charac-

TABLE 2. Summary Results from the 1979-1980 and 1986

National Incidence Studies-

Original definitions
Maltreatment Percentage of change 1986
category 1979-1980 1986 1979-1980 to 1986 Definition

Total abuse 336,600 (5.3) 580,400 (9.2) +72 (+74) 675,000 (10.7)
Physical 199,100 (3.1) 311,200 (4.9) +56 (+58) 358,300 (5.7)
Sexual 42,900 (0.7) 138,000 (2.2) +222 (+214) 155,900 (2.5)
Psychological 132,700 (2.1) 174,400 (2.8) +31 (+33) 211,100 (3.4)
Total neglect 315,400 (4.9) 498,000 (7.9) +58 (+61) 1,003,600 (15.9)
Physical 103,600 (1.6) 182,100 (2.9) +76 (+81) 571,600 (9.1)
Psychological 56,900 (0.9) 52,200 (0.8) -8 (-11) 223,100 (3.5)
Educational 174,000 (2.7) 291,100 (4.6) +67 (+70) 292,100 (4.6)

«Adapted from NCCAN (1988). Numbers in parentheses are for incidence (rate/I ,OOO children) and for
change in incidence.
36 RAYMOND H. STARR, JR., et al.

teristics are associated with different forms of abuse or neglect. Finally,

only intrafamilial sexual abuse was considered, resulting in a definition
that differs considerably from that used in other studies which examine
extrafamilial sexual abuse as well .
Overall, the maltreatment incidence in 1986 was approximately
16/1,000 (NCCAN, 1988), which represents slightly more than 1 million
children. When the revised definition of maltreatment is examined these
figures increase to 25/1,000 or in excess of 1.5 million children. Using the
original definition, slightly more than half of all reported children were
abused (56%) and slightly less than half were neglected (48%). However,
the revised definition indicated relatively more neglect (63%) than abuse
(43%). When the 1986 data are compared with 1979-1980 data, the re-
sults suggest that recognized maltreatment increased significantly, due
largely to a 74% increase in the incidence of abuse. Neglect rates did not
change significantly over the 6-year period. Hence, most of the overall
increase in incidence was due to increased physical and sexual abuse
and not to increased neglect. The greatest part of this increase was for
those cases involving moderate injury, which increased 89%.
It is important to remember that these data do not represent the
actual number of maltreated children and probably are underestimates.
First, only 40% of the 1986 cases had been referred to CPS (46% using
the revised definition). Second, abuse cases at Level 4 (cases known to
other agencies and individuals) and Level 5 (undetected cases) of the
five-part model are not included in the data from either survey. Thus,
there is no estimate of the number of cases known only to such person-
nel as private physicians and mental health workers.
Three factors may account for the increase in physical and sexual
abuse between 1979-1980 and 1986. First, the actual incidence could
have increased with more children being maltreated. Second, profes-
sionals may be more likely to report abuse or neglect cases they see .
Third, the results may be due to methodological differences in the two
surveys. The authors propose that the second explanation is the more
likely (NCCAN, 1988). Thus, much of the increase in reported cases in
1986 was due to children who had moderate injuries (72% of the total)
followed by serious injuries (15%), probable injuries (12%), and fatalities
(0.1%). The incidence of severe injuries did not increase Significantly. If
the actual incidence of maltreatment had increased, there should have
been a significant rather than a nonsignificant increase in these cases. It
is assumed that cases of severe injury were equally likely to have been
reported at both survey times, given the importance of intervention
when such injuries are present. However, moderate rather than severe
cases are the ones that professionals, who are increasingly attuned to
EPIDEMIOLOGY OF MALTREATMENT 37

the detection of maltreatment, would be more likely to note, regardless

of whether an official report of maltreatment was or was not filed. The
second explanation also is supported by the finding of a more than 200%
increase in the incidence of sexual abuse. Sexual abuse became a major
child welfare issue during the period between the two surveys. It is
unlikely that there was a threefold increase in the actual number of
sexual assaults on children over a 6-year period.
A detailed examination of the six methodological changes intro-
duced in the 1986 survey suggests that they account for, at most, a small
portion of the increased incidence of abuse. It is more likely that profes-
sionals' greater awareness of and ability to identify maltreatment ac-
count for the increasing incidence of maltreatment. Also, it is plausible,
but somewhat less likely, that the actual incidence of abuse has in-
creased. Regardless of the interplay of all these factors, it is unlikely that
the incidence of maltreatment declined across the 6-year period of the
two studies, a finding that is important in clarifying the results of sur-
veys by Straus and Gelles (e.g., Gelles, 1978; Gelles & Straus, 1987, 1988;
Straus & Gelles, 1986; Straus, Gelles, & Steinmetz, 1980) discussed
below.
National Incidence Study data also provide information about the
types of children who were maltreated (NCCAN, 1988). The major find-
ings revealed in this study were that
• Females were more likely to be abused (13/1,000 compared to
8/1,000 for males), mostly because of their increased susceptibility
to sexual abuse (4/1,000 vs. 1/1,000)
• The incidence of child abuse increased with child age, particularly
for physical abuse
• Impoverished children were much more likely to be maltreated or
injured-children from families earning less than $15,000 a year
were more than five times as likely to be maltreated and more
than seven times as likely to be seriously injured or impaired
• Family size was unrelated to maltreatment using original defini-
tions; with the revised definition family size was positively associ-
ated with abuse and neglect
• Race, ethnicity, and county metropolitan status were not related
to the incidence of maltreatment
More detailed analyses of the first survey data have been done. One
such analysis examined the national incidence data for hospitals, institu-
tions at Level 3 of the five-level maltreatment awareness model (Hamp-
ton & Newberger, 1985). The subset of cases identified as maltreated by
the hospitals surveyed showed several detection trends that differed
38 RAYMOND H . STARR, JR., et al.

from those found for the incidence study as a whole. The children
detected by hospitals, compared with cases reported by other agencies,
were more likely to: (1) live in urban areas (66% vs . 42%), (2) be younger,
(3) have younger parents, (4) be black (25% vs. 16%), and (5) have been
physically abused. Multivariate analyses of the data for cases that were
and were not reported to CPS indicated four variables that were the key
predictors of hospital-based reporting: (1) type of maltreatment, (2) fam-
ily income, (3) maternal role in maltreatment, and (4) race or ethnicity.
Psychologically abused children from white families with above average
incomes, in which the mother was deemed responsible for the maltreat-
ment, were least likely to be reported. Physically abused children from
lower income, minority families, in which the alleged perpetrator was
not the mother, were more likely to be reported. These results support
the argument that a report is more likely to be made when there is a
demographic difference between the reporter and the maltreating family
(O'Toole et al., 1983).
Study results also clarify some of the issues involved in interpreting
the analyses of CPS reports discussed above (AAPC, 1985, 1986, 1987,
1988; American Humane Association, 1984; Russell & Trainor, 1984). Of
particular importance is the elimination of the counting of duplicate
reports in the NCCAN incidence studies. National Incidence Study data
suggest that the increased recognition of maltreatment cases by sources
other than CPS workers has not been reflected in an increased incidence
of confirmed CPS cases . There are two likely explanations for this find-
ing. First, the surveyed professionals may fail to report cases to CPS.
Second, CPS may fail to confirm as legitimate maltreatment those cases
that are reported. If the former is correct, efforts need to be made to
increase the recognition of maltreatment and to convince professionals
of the value of reporting it. If the latter view is correct, efforts need to be
made to secure more investigative resources for CPS agencies. In reality,
it is likely that both explanations contribute to this discrepancy.

SURVEYS OF THE GENERAL PUBLIC

Many studies have examined the epidemiology of child maltreat-

ment by surveying either a random sample of the general public or
special, at-risk populations. These studies provide data concerning
Level 4 of the five-level model of maltreatment awareness. They repre-
sent the broadest base for formulating estimates of maltreatment inci-
dence and prevalence. These studies have typically examined only one
type of maltreatment rather than being comprehensive, as was the case
for the studies in the earlier subsections of this chapter (AAPC, 1985,
EPIDEMIOLOGY OF MALTREATMENT 39

1986, 1987; American Humane Association, 1984; NCCAN, 1981, 1988;

Russell & Trainor, 1984).

Physical Abuse
Some of the earliest estimates of the incidence of physical abuse
were based on surveys of nationally representative samples. Gil (1973)
conducted a survey in 1965 in which 1,520 adults were asked about their
personal knowledge of families who had injured "a child, not by acci-
dent, but in anger or deliberately" (p. 49). The results showed that 3% of
the sample knew of at least one such incident. Allowing for error vari-
ance, between 2.5 and 4.1 million children were injured in the year prior
to the survey-a figure that seems high even today. However, Gil's
methodology has been criticized (Light, 1973). When Light controlled for
errors in the Gil analysis, the incidence of physical abuse decreased to
500,000 children a year.
Straus, Gelles, and their colleagues conducted national surveys of
the incidence of physical abuse in the general population in 1975 and in
1985 (Gelles, 1978; Gelles & Straus, 1987, 1988; Straus et al., 1980; Straus
& Gelles, 1986). Physical abuse was defined as the parental use of certain
violent acts toward a child as measured using the Conflict Tactics Scale
(Straus, 1979). This questionnaire asks family members about how they
resolve conflicts. There were slight differences in the scales used in the
two surveys. The original, 1975 scale asked about throwing objects;
pushing, grabbing, or shoving; slapping or spanking; kicking, biting, or
hitting with a fist; hitting or trying to hit with an object; "beating up";
and threatening with or using a knife or gun (Straus & Gelles , 1986).
Questions concerning scalding or burning were added in 1985.
Child abuse was defined as kicking, biting, punching, "beating up,"
and threatening with or using a knife or a gun (Gelles & Straus, 1987).
Other violent acts-including hitting with an object and threatening
with a weapon-were excluded. They were considered as variations of
normal discipline, rather than abusive, or were deemed unlikely to lead
to actual injury.
For the 1985 study, telephone interviews were conducted with a
nationally representative sample of 1,428 families containing a male-
female couple or one adult over 18 years old where there was at least
one child from 3 to 17 years old . Overall, 84% of contacted families
agreed to participate in the survey. Data are available for only two-
parent families. The 1975 study was conducted using a different method
(Gelles, 1978; Straus & Gelles, 1986). First, in-person rather than tele-
phone surveys were conducted. Second, families were given the alter-
40 RAYMOND H . STARR, JR., et al.

native of answering "never" with regard to the frequency with which a

given act occurred in their home. In 1985, they were not directly told
they could answer "never." Third, only two-parent families were inter-
viewed. In 1975, 65% of approached families agreed to be interviewed,
yielding a final sample of 1,146 families with a 3- to 17-year-old in the
home.
The results of the surveys indicate that there was a significant de-
cline in the rate with which parents reported using three forms of vio-
lence between 1975 and 1985: throwing something declined from 54 per
1,000 children to 27/1,000; kicking, biting, and hitting with a fist de-
clined from 32/1,000 to 13/1,000; and hitting or trying to hit with an
object declined from 134/1,000 to 97/1,000 . The rate of acts that could be
considered potentially abusive also declined significantly from 36/1,000
to 19/1,000 (Gelles & Straus, 1987).
The key question is whether these data indicate a real decline in the
incidence of child abuse. Straus and Gelles (1986) consider several possi-
ble explanations for the disparate findings, including (1) differences in
methodology, (2) an increasing reluctance to report family violence to an
outsider, and (3) a real decline in family violence. With regard to meth-
odology, Straus and Gelles cite several references for studies of the dif-
ferences between in-person and telephone interviews, all of which note
no significant differences in the results obtained using the two methods.
Alternatively, as public knowledge about child abuse and recognition of
its seriousness increases, people may be more reluctant to admit to
experiencing family violence. The authors argue that, if this were the
case, more families would have refused to participate in 1985 than did in
1975. However, it is difficult to evaluate this possibility because of the
different survey techniques used in the two studies.
Factors cited that support an actual decline in the incidence of child
abuse include structural changes in families, improved economic condi-
tions, and the availability of prevention and treatment programs (Gelles
& Straus, 1987, 1988). First, changes in family structure, including an
increasing age at the time of first parenthood and declining family size,
lead to less stress and, potentially result in decreased violence. Second,
intact families were less likely to experience economic stressors in 1985
than they were ten years earlier. Finally, the increased availability of and
publicity about treatment programs may mean that parents try to get
help with childrearing problems before resorting to abusive violence.
It is important to recognize that the true incidence of physical abuse
is probably higher than that reported by the surveys. First, many indi-
viduals are unlikely to admit to being violent. Second, the range of
EPIDEMIOLOGY OF MALTREATMENT 41

violent acts included in the survey was quite limited. Many other acts,
such as burning and poisoning, result in physical abuse. For example, 5
in 1/000 families in the 1985 survey reported burning or scalding their
child (Gelles & Straus, 1987). This finding suggests that the rate of child
abuse is probably higher than the reported 19/1/000. Third, the surveys
included onl y children over 2 years of age despite the fact that younger
children were more likely to be physically abused than older ones (ap-
proximately 7% for 3- to 4-year-olds, and 4% for older children) (Gelles,
1978). The presence of an inverse relationship between child age and
physical abuse is supported by other studies. For example, 79% of phys-
ical abuse reports in Arizona and 85% in Louisiana were for children less
than four years old (Maximus, 1986b). The latter figures, in turn, are
probably overestimates. Young children are more subject to scrutiny by
professionals and are more likely to represent unconfirmed cases of
physical abuse (Iason, Andereck, Marks, & Tyler, 1982). A fourth reason
the Gelles and Straus findings are likely to be underestimates of physical
abuse is that results are available for intact families even though only
single-parent families are under greater stress and, as such, are more
likely to be violent. In a reanalysis of National Incidence Study data,
Miller (1984) found that teenage mothers, who are likely to be single
parents, also were more likely to physically abuse their children than
were older mothers. Furthermore, retrospective evidence suggests that
single parents are more punitive toward their children (Sack, Mason, &
Higgins, 1985). Thus, the Gelles and Straus data only suggest the inci-
dence of physical abuse in the general population.
The Gelles and Straus (1987) data indicate that approximately 1.5
million children were subjected to potentially abusive violence in 1975
compared to about 750/000 children in 1985. Moreover, they suggest that
the decline in incidence is most probably because of some combination
of attitudinal and behavioral change rather than differences in survey
methodology. Only further research can yield a conclusive answer about
the validity of these alternative explanations.
The results of the 1975 survey do reveal something about the types
of families that direct high levels of violence toward children. Examining
the Conflict Tactics Scale variables that were used to determine child
abuse (Straus & Gelles, 1986)/ approximately 4% of fathers and 6% of
mothers were abusive (Gelles, 1978). This heightened level of maternal
abuse may be due to the greater amount of time mothers spend with
their children, to the greater degree to which children interfere with
maternal daily activities (Straus et al., 1980)/ or to other factors related to
family functioning. Boys were more likely to be subjected to physical
42 RAYMOND H. STARR, JR., et al.

abuse than were girls (approximately 6% vs . 3%) (Gelles, 1978). Al-

though the exact reason for this difference is debatable, it does confirm
the findings of the AAPC survey (1988) discussed earlier.
It is unfortunate that the data analyses of the Gelles and Straus
family surveys did not focus more specifically on acts that they label
physically abusive. Instead, most of their analyses are for total violent
acts (including, for example, throwing objects, spanking, slapping, and
hitting) rather than just abusive ones. Thus, when total violence toward
children, including abusive violence, is considered, a number of demo-
graphic differences were found (Straus et al., 1980). For example, there
was more overall violence in the midwest and west; in large cities;
among non-Jews, younger parents, and parents with either some high
school or who graduated from high school; in lower income and blue
collar occupations, and in families where the husband worked part-time
or was unemployed. No differences were found between black and
white families .

Sexual Abuse
The preceding sections have noted the dramatic increase in public
awareness concerning sexual abuse over the past decade. Given this
recognition, it is not surprising that many studies have revealed its
prevalence, including general population surveys and studies of special
populations that are believed to have a high likelihood of sexual abuse
(e.g., Gruber & Jones , 1983; Silbert & Pines, 1981).
It is important to remember that all the data in this section are for
the prevalence of sexual abuse during childhood as measured retro-
spectively in adult samples. These data indicate how many adults were
sexually abused during childhood. The rates reported are thus consider-
ably higher than would be the case if the incidence per 1,000 cases per
year (the measure used for all other data in this chapter) were used as a
measure of problem magnitude.
There is considerable variation in study methodologies. Studies use
dissimilar definitions, consider childhood to end at various ages, have
different criteria about the inclusion of cases where there was consent to
the maltreatment, use diverse question framing techniques, and inter-
view samples with different characteristics using a variety of interview
techniques.
As was indicated earlier, definitions are a major factor influencing
the outcome of incidence and prevalence studies. Key variables are
whether physical contact was involved, the age of the victim, the age
differential between the victim and the perpetrator, and whether the
EPIDEMIOLOGY OF MALTREATMENT 43

perpetrator was a family member. These definitional factors make com-

parisons across studies particularly difficult because prevalence figures
typically are lower when narrower definitions are used (Haugaard &
Reppucci, 1988). In one study of the relationship between definitional
restrictiveness and prevalence, the recalculation of data from a survey of
black and white urban-area women (Wyatt, 1985), using a more re-
strictive definition, led to a 14% drop in the prevalence of sexual abuse
(Wyatt & Peters, 1986a).
Given the current interest in sexual abuse it is surprising that only
two national, representative sample surveys have been conducted in the
United States (Kinsey et al., 1953; "22% in Survey," 1985). Kinsey and his
colleagues found that 22% of women had experienced some sexual ac-
tivity during childhood, with contact occurring in less than half of these
cases (9%). The Los Angeles Times survey interviewed a national, random
sample of 2,627 adults and asked 100 questions concerning sexual abuse
("22 % in Survey," 1985).The results were that 27% of women and 16% of
men reported sexual abuse as a child, with 55% of all victims experienc-
ing sexual intercourse. Less than a quarter (23%) of abusers were rela-
tives, and the modal age at the time of the abuse was 10 years. Unfortu-
nately, specific details concerning aspects of the survey, such as the
maximum victim and minimum perpetrator age and the frequency of
contact versus noncontact abuse, have not been published, limiting the
comparability of these data with those from other studies.
The only other national incidence data come from Canada (Commit-
tee on Sexual Offenses Against Children and Youth, 1984, cited in
Haugaard & Reppucci, 1988) where 2,135 men and women over 17 years
old were surveyed. Overall, 28% of the women and 10% of the men had
experienced some form of sexual abuse as children. These results in-
clude cases in which the abuse was perpetrated by a peer (40% of cases)
and in which no contact was involved (half of all cases). Thus, contact
abuse had occurred for, at the most, 14% of women and 5% of men-a
prevalence not very different from the Kinsey et al. (1953) findings for
women.
Other estimates of prevalence at the fourth level of maltreatment
recognition are provided by the results of a number of more limited
surveys of varying groups including college students (Finkelhor, 1979;
Fritz, Stoll, & Wagner, 1981; Fromuth, 1986; Haugaard, cited in
Haugaard & Reppucci, 1988; Risin & Koss, 1987;Sedney & Brooks, 1984),
and random sample surveys in specific geographic areas (Finkelhor,
1984; Kercher & McShane, 1984; D. E. H. Russell, 1983, 1984; Wyatt,
1985). These studies are summarized in Table 3.
The use of differing methodologies complicates comparisons among
 TABLE 3. Summary Studies of Sexual Abuse Prevalence

Geographic surveys
Finkelhor Kercher & McShane Russell Wyatt
(1984) (1984) (1983, 1984) (1985)

Sample size 700 2,000 930 248

Location Urban Boston Texas San Francisco Los Angeles
Response rate 74% 53% 64% 73%a
Male /female Both Both Female Female
Method Questionnaire Questionnaire Interview Interview
Maximum victim age 15 ? 17 17
Minimum perpetrator age None None None None
Age differential required Yes No No No
Contact required No No No No
Abuse wanted /unwanted Unwanted ? Either Either
Women abused as child 15% 11% 54% 62%
Men abused as child 6% 3%
College and university student surveys
Finkelhor Fritz et al. Fromuth Haugaard Risin & Koss Sedney & Brooks
(1979) (1981) (1986) (1987) (1987) (1984)

Sample size 796 952 482 1,089 2,972 301

Response rate 92 % Unknown Unknown 61% 98% Unknown
Male /female Both Both Female Both Male Female
Maximum victim age 16 Prepuberty 16 16 13 Unknown
Minimum perpetrator age None Postadolescent 16 16 None Unknown
Age d ifferential required Yes Yes Yes Yes Yes No
Contact required No Yes No Yes No No
Abuse wanted/unwanted Either Either Either Unwanted Either Either
Women abused as child 19 % 8% 22% 12% - 16%
Men abused as child 9% 5% - 5% 7%

«This figure may be 55% depending on how response rate is mea sur ed (Haugaard & Reppucci. 1988).
EPIDEMIOLOGY OF MALTREATMENT 45

these studies. Thus, as indicated in Table 3, some studies examined only

incidents involving contact, whereas others did not. In addition, the
maximum age of the victim, minimum perpetrator age and whether the
perpetrator had to be older than the victim or could be a peer also
differed from study to study. Further complexities are introduced when
the issue of whether the sexual contact was wanted or unwanted is
added. Some studies (e.g ., Finkelhor, 1979; Fritz et al., 1981; Fromuth,
1986) count both types of act as abuse; other studies analyze only un-
wanted contact (e.g., Committee on Sexual Offenses Against Children
and Youth, 1984, cited in Haugaard & Reppucci, 1988); and stiII others
vary for intrafamilial and extrafamilial abuse (D. E. H. Russell, 1983,
1984) or for children of varying ages (Wyatt, 1985).
These studies also have examined other key aspects of the preva-
lence of sexual abuse, including (1) abuse involving direct contact be-
tween victim and abuser, and (2) intrafamilial versus extrafamilial abuse.
D. E. H. Russell (1983) found that 38% of her sample reported contact
abuse before they were 18 years old . For less than 5% of the women
surveyed, the perpetrator was a parent, was a family member for 16%,
and was unrelated for 31% of her sample. If a more restrictive definition
is used and only abuse in girls 13 years old or younger is considered,
these prevalence figures decline to 28% of her sample experiencing
abuse before 14 years of age, with 12% experiencing intrafamilial and
20% extrafamilial abuse. These data also indicated that, for intrafamilial
abuse, stepfathers were much more likely to abuse (17% of her sample
who had a stepfather present) than were biological fathers (2%); and
that stepfathers, when they did abuse, perpetrated more severe forms of
abuse (D. E. H. Russell , 1984).
Wyatt's (1985) reported prevalence of sexual abuse that involved
contact was higher than Russell's-45% prior to age 18. Wyatt found
that 76% of all sexual abuse (both contact and noncontact) was commit-
ted by nonrelatives, less than 2% by fathers, about 6% by nonrelated
males in a father role, and 14% by other male relatives. Finkelhor (1984)
reported that 80% of the abuse experienced by the subjects in his ran-
dom survey involved contact, resulting in a female contact abuse preva-
lence of 12% and a male rate of 5%. A relative was the perpetrator in 32%
of all the cases. In addition, it should be noted that he used a somewhat
narrower definition of sexual abuse than Russell or Wyatt-an act was
counted as abusive only if the person self-defined it as sexual abuse.
Data for college students come mainly from middle-class adoles-
cents and young adults-a less representative sample than is the case
for general population surveys. When the results of student surveys, all
of which used questionnaires, are compared with those from the two
46 RAYMOND H . STARR, JR., et al.

more general questionnaire studies (Finkelhor, 1984; Kercher & Me-

Shane, 1984), there is considerable overlap in results for the various
sample types. The prevalence range for general population surveys is
11% to 15% for women and 3% to 6% for men. Comparable data for
college students are 8% to 22% for women and 5% to 9% for men (see
Table 3). This variability is not surprising considering the different defi-
nitions and methodologies used in the various studies. Indeed, one
study of male college students evaluated the effects of different defini-
tions and found that the prevalence rate ranged from 4% to 24% de-
pending on the definition of sexual abuse used (Fromuth & Burkhart,
1987).
There is even greater variation in the prevalence of sexual abuse
when the data for both questionnaire and interview studies are consid-
ered. The range for women is from 62% (Wyatt, 1985) to 8% (Fritz et al.,
1981), while it is 9% (Finkelhor, 1979) to 3% (Kercher & McShane, 1984)
for men. A number of factors account for these differences. In addition
to the effects of differing definitions discussed by Fromuth and Burkhart
(1987), methodological differences are also important. In an analysis of
four representative studies (Finkelhor, 1979, 1984; D. E. H. Russell,
1983, 1984; Wyatt, 1985), Wyatt and Peters (1986b) examined the role of
different data collection procedures. The prevalence rate was much
higher in the two interview studies (D. E. H . Russell, 1983, 1984; Wyatt,
1985) and lower in the questionnaire studies (Finkelhor, 1979, 1984).
Indeed, the results of the two interview studies in Table 3 are quite
similar and those of the eight questionnaire studies are relatively similar,
particularly for those including a male sample.
A second important methodological variable discussed by Wyatt
and Peters (1986a,b) was the way in which information was gathered,
particularly how questions about prior sexual abuse were framed. They
propose that studies that asked more detailed questions about a variety
of forms of sexual abuse yielded greater reporting of such maltreatment.
The confounding of these two variables limits our ability to determine
which is more important. In addition, published information concerning
four of the other six studies summarized in Table 3 (Fritz et al., 1981;
Fromuth, 1986; Haugaard, 1987, cited in Haugaard & Reppucci, 1988;
Sedney & Brooks, 1984) is not sufficient to determine the question fram-
ing procedure used . The remaining studies (Kercher & McShane, 1984;
Risin & Koss, 1987) used a narrow framing procedure and the results are
close to those of the Finkelhor (1984) geographic survey.
Other factors appear to be unrelated to the prevalence of sexual
abuse (Wyatt & Peters, 1986b). These include whether random sampling
or subject self-selection was used, the geographic area of the study, and
EPIDEMIOLOGY OF MALTREATMENT 47

the subject's race and education level. Taking all of this into considera-
tion, Wyatt and Peters (1986b) conclude that not only is more research
needed to clarify the exact extent of sexual abuse but that future studies
should use in-person interviews so that the issue of sexual abuse can be
embedded within a study described as having a broader purpose. They
also suggest (1) matching subject and interviewer race, (2) using care-
fully trained project personnel, (3) ensuring interview confidentiality,
and (4) paying subjects for the time and expense incurred.
What do these studies tell us about the prevalence of sexual abuse
in addition to the fact that study design plays an important role in
determining study outcome? A major conclusion, unfortunately, is that
the prevalence of child sexual victimization is much higher than would
have been predicted a decade ago . Although we cannot cite one preva-
lence figure as giving a single, valid indicator of the extent of child
sexual abuse, we can state with certainty that females are at significantly
greater risk. However, it is important to remember that, relying on the
high prevalence figures cited by D. E. H. Russell (1983) and Wyatt
(1985), less than 5% of women have been sexually abused by a parent or
parent-figure. Since these studies examined prevalence rates and did
not include males, the actual annual incidence of sexual abuse for the
overall child population will be significantly lower. In spite of this cave-
at, sexual abuse is a major problem and most cases never reach the
attention of authorities.

Psychological Maltreatment and Physical Neglect

It is surprising that there is so little information about what proba-
bly are the most common forms of child maltreatment. No research has
been done studying psychological maltreatment or physical neglect at
Level 4 of maltreatment. Although it is true that surveys at this level
would be difficult to conduct, they would yield valuable information
about how children are being cared for by their parents.

SUMMARY AND IMPLICATIONS

What is the answer to the question with which this chapter began:
"What do we really know about the extent of the child maltreatment
problem?" One obvious answer is that the determination of the "exact"
incidence and prevalence of child abuse is a very complex process. We
have discussed a number of studies that use different samples, defini-
tions, and research methodologies. The best that can be done at present
48 RAYMOND H . STARR, JR., et al.

is to compare the incidence and prevalence data from studies of the first
four levels of maltreatment: (1) cases known to protective services, (2)
those that have been reported to other investigative agencies, (3) mal-
treated children who have been detected by other professionals and
agencies, and (4) cases in which other individuals, including per-
petrators and victims, know of the maltreatment.
A second answer to the above question is that much is known about
the epidemiology of child maltreatment. Indeed, there is much data,
much more than we have been able to discuss in this chapter. The
problem is how to interpret and systematize the available findings . Table
4 summarizes data for the different maltreatment levels by the type of
maltreatment.
The data in Table 4 indicate a wide variation in estimates of the
incidence and prevalence of child maltreatment. The two data sets that
are most similar are those for reported cases (AAPC, 1988) and for the
National Incidence Study of cases that have either been reported or are
known to agencies or professionals (NCCAN, 1988). Although it would
be expected that the National Incidence Study data would show a high-
er incidence of maltreatment-because a broader sample of cases were
included-this is not consistently the case. Thus, the data for the Na-
tional Incidence Study indicate a lower incidence of sexual abuse than
the AAPC data suggest. But the AAPC data are inflated because of the

TABLE 4. Summary of Incidence Data

Total Physical Sexual Psychological

maltreatment abuse abuse Neglect maltreatment"

Reported cases 12.4 b 3.5 2.9 6.8 1.1

(AAPC, 1988)
National incidence 16.3 (25.2)c 4.9 (5.7) 2.2 (2.5) 7.5 (13.7)d 3.6 (6.9)e
study (NCCAN,
1988)
Cases known to in- 191 1O%-58% g No data No data
dividuals

«Includes both psychological abuse and psychological neglect.

bRate/1000.
<The first figure is incidence using the original (NCCAN, 1981) maltreatment definitions ; data in
parentheses are for the revised definition (NCCAN, 1988).
dData are for the sum of the incidence of physical and educational neglect. Figures include an unknown
amount of duplicate cases where both were present.
' Data are for the sum of the incidence of psychological abuse and neglect. Figures include an unknown
amount of duplicate cases where both were present.
IData from Gelles and Straus (1987, 1988)and Straus and Gelles (1986).
sl.ower figure is mean percentage for surveys of men and women reported in Table 3. Higher figure is
mean for women only from detailed interview studies (Russell, 1983, 1984; Wyatt, 1985).
EPIDEMIOLOGY OF MALTREATMENT 49

counting of duplicate cases (AAPC, 1988). Other differences in the re-

ported incidence figures for the two studies may be due to the broader
sample used in the National Incidence Study or to methodological dif-
ferences (AAPC, 1988). It is likely that to at least some unknown extent
the data from these two studies indicate a real discrepancy between the
rate of indicated reported cases evaluated by the AAPC studies (1988)
and both the reported and the known, but not reported cases assessed
in the National Incidence Study (NCCAN, 1988).
The data presented in this chapter also suggest that the vast major-
ity of cases of child maltreatment are never officially recognized. Inci-
dence and prevalence rates for the two areas in which studies have been
done (physical and sexual abuse) are much higher when surveys of the
general public are conducted than when they are reported and/or when
professionally known cases are considered. Many of the acts included in
the Straus and Gelles study (Gelles & Straus, 1987, 1988; Straus & Gelles,
1986) have only the potential to injure. In addition, if injury does occur,
the damage to the child may not be severe enough to require medical
attention, or the case may not be considered physical abuse even if care
is needed. Similarly, most incidents of sexual abuse are never brought to
official attention. The Los Angeles Times survey concluded that sexual
abuse victims report abuse to the police in only 3% of cases ("22 % in
Survey," 1985). Indeed, less than half of victims (42%) told anyone about
the abuse and, of these, only 30% said their telling resulted in "any
effective action" (p. 34). Data on police reporting are similar to those
Russell (1983) found in her survey, 2% of cases of intrafamilial and 6% of
extrafamilial abuse were reported to the police .
There are important factors that should be considered in interpret-
ing the data presented in this chapter and in planning future research
efforts. First, it is important to remember that epidemiology is the study
not only of the incidence and prevalence of a problem, but also of the
risk factors involved. With the exception of some discussion of demo-
graphic risk factors, this chapter has focused on the incidence and prev-
alence of maltreatment. Other risk variables are considered in Chapters
1, 6, 7, and 8.
Second, it is necessary to reiterate the key role that definitional
issues play in determining the incidence and prevalence of child mal-
treatment. The studies reviewed in this chapter consider a wide spec-
trum of acts as abusive, ranging from fatalities at one extreme to inde-
cent exposure at another. Reviewing and sorting out all of the important
definitional variables is a difficult and complex task and is beyond the
scope of this chapter. Child maltreatment is a complex phenomenon that
is defined in many different ways. Each definition has implications not
50 RAYMOND H. STARR, JR., et al.

only for determining incidence and prevalence, but also for case finding,
treatment, and prevention.
Third, progress has been made since 1975 when Cohen and Sus-
sman stated "the only conclusion which can be made fairly is that infor-
mation indicating the incidence of child abuse in the United States sim-
ply does not exist" (cited in Gelles, 1978, p. 582). But, much still remains
to be done. For example, research groups could come together to agree
on standardized definitions to be used in epidemiological studies. If a
group liked a given definition better than the standard, agreed upon
one, they could incorporate both definitions in their study. This was
done in the most recent National Incidence Study to allow comparability
of the 1979-1980 and 1986 data (NCCAN, 1988). Furthermore, better
information is needed about the prevalence of neglect and psychological
maltreatment at Level 4 of the National Incidence Study categorization
of level of awareness (NCCAN, 1988). Ideally, a study could be con-
ducted using a national interview sample in which the presence of all
types of maltreatment is evaluated. Although planning and conducting
such a study would be difficult, it would be a major step forward in
obtaining a better idea of the actual dimension of the problem of child
maltreatment in all its forms.
Another way to increase knowledge about the epidemiology of
child maltreatment is to improve analyses of reported case data. The
American Association for Protecting Children receives annual funding
sufficient for only partial analysis of state report data. Annual data anal-
yses are not needed. It would be a better investment to have detailed
analysis of data every 5 years in order to obtain a fuller examination of
reporting and case substantiation trends-the very data policy makers
need to make informed decisions. It is unfortunate that existing data do
not permit making a firm conclusion about the magnitude of child mal-
treatment and whether our prevention and treatment efforts are having
a meaningful impact in reducing the extent of the problem. Perhaps
future research and analyses of reporting trends will provide the an-
swers to these as yet unanswered questions.

REFERENCES
Adams, W., Barone, N. , & Tooman, P. (1982). The dilemma of anonymous reporting in
child protective services . Child Welfare, 61, 3-14.
American Association for Protecting Children. (1985). Highlights of official child neglect and
abuse reporting 1983. Denver: American Humane Association .
EPIDEMIOLOGY OF MALTREATMENT 51

American Association for Protecting Children. (1986). Highlights of official child neglect and
abuse reporting 1984. Denver: American Humane Association.
American Association for Protecting Children. (1987). Highlights of official child neglect and
abuse reporting 1985. Denver: American Humane Association.
American Association for Protecting Children. (1988). Highlights of official child neglect and
abuse reporting 1986. Denver: American Humane Association.
American Humane Association. (1984). Highlights of official child neglect and abuse reporting
1982. Denver: Author.
Bronfenbrenner, U. (1977). Toward an experimental ecology of human development. Amer-
ican Psychologist, 32, 513-531.
Bronfenbrenner, U., Moen, P., & Garbarino, J. (1984). Child, family, and community. In R.
Parke (Ed.), Review of childdevelopment research (pp . 283-328). Chicago: University of
Chicago Press.
Christoffel, K. K., Liu, K., & Stamler, J. (1981). Epidemiology of fatal child abuse: Interna-
tional mortality data . Journal of Chronic Diseases, 34, 57-64.
Daro, D., & Mitchel, L. (1988). Child abuse fatalities remain high: The results of the 1987 annual
fifty statesurvey (Working Paper No .8). Chicago: National Committee for Prevention of
Child Abuse.
Dubowitz, H. (1987). Child maltreatment in the United States : Etiology, impact, and prevention.
Washington, DC: Office of Technology Assessment.
Finkelhor, D. (1979). Sexually victimized children . New York: Free Press.
Finkelhor, D. (1984). Child sexual abuse: New theory and research . New York: Free Press.
Finkelhor, D., & Hotaling, G. T. (1984). Sexual abuse in the National Incidence Study of
Child Abuse and Neglect: An appraisal. Child Abuse and Neglect, 8, 23-33 .
Fritz, G. S., Stoll, K., & Wagner, N. N. (1981). A comparison of males and females who
were sexually molested as children. Journal of Sex and Marital Therapy, 7, 54-59.
Fromuth, M. E. (1986). The relationship of childhood sexual abuse with later psychological
and sexual adjustment in a sample of college women. ChildAbuseand Neglect , 10, 5-15.
Fromuth, M. E., & Burkhart, B. R. (1987). Childhood sexual victimization among college
men: Definitional and methodological issues. Violence and Victims, 2, 241-253 .
Garbarino, J.• & Crouter, A. (1978). Defining the community context for parent-child
relations: The correlates of child maltreatment. Child Development, 49, 604-616.
Garbarino, J., & Gilliam, G. (1980). Understanding abusive families . Lexington, MA: Lex-
ington Books.
Garbarino, J., Guttman, E., & Seeley, J. W. (1986). The psychologically battered child: Strategies
for identification, assessment, and intervention. San Francisco: [ossey-Bass .
Gelles, R. J. (1978). Violence toward children in the United States. American Journal of
Orthopsychiatry, 48, 580-592 .
Gelles, R. J. (1982). Problems in defining and labeling child abuse. In R. H. Starr, Jr. (Ed.),
Child abuse prediction: Policy implications (pp . 1-30) . Cambridge, MA: Ballinger.
Gelles, R. J., & Straus, M. A. (1987). Is violence toward children increasing? A comparison
of 1975 and 1985 national survey rates . Journal of Interpersonal Violence, 2, 212-222 .
Gelles, R. J., & Straus, M. A. (1988). Intimate violence. New York: Simon & Schuster.
Gil, D. G. (1973). Violence against children: Physical childabuse in the UnitedStates. Cambridge:
Harvard University Press.
Giovannoni, J. M., & Becerra, R. M. (1979). Defining child abuse. New York: Free Press.
Gruber, K. J., & Jones, R. J. (1983). Identifying determinants of risk of sexual victimization
of youth: A multivariate approach. Child Abuse and Neglect, 7, 17-24.
Hampton, R. L., & Newberger, E. H. (1985). Child abuse incidence and reporting by
52 RAYMOND H. STARR, JR., et al.

hosp itals: Significance of severity, class, and race. American Journal of Public Health, 75,
56-60.
Hau gaa rd, J. J., & Reppucci , N . D. (1988). The sexual abuse of children: A comprehensiveguide
to current knowledge and intervention strategies. San Franci sco: Jossey-Bass.
Jason, J., Andereck, N. D., Mark s, S., & Tyler, C. w., Jr. (1982). Child abuse in Georgia: A
method to evalua te risk factors and rep orting bias. American Journalof Public Health, 72,
1353-1358.
Kercher, G. A., & McShane, M. (1984). The pre valence of child sexua l abuse victimization
in an adult sample of Texas residents. Child Abuse and Neglect, 8, 495-501.
Kinsey, A. C. ; Pomeroy, W. B., Martin, C. E., & Gebhard, P. H. (1953). Sexual behavior in the
human female. Philadelphia: W. B. Saunders.
Knudsen , D. D. (1988). Child maltreatment over two decade s: Change or continuity?
Violence and Victims, 3, 129-144.
Korbin, J. E. (1987). Child abu se and neglect: The cultural context. In R. E. Helfer & R. S.
Kempe (Eds.), The battered child (4th ed. , pp. 23-41) . Chicago : Univ ersity of Chicago
Press.
Last, J. (1983). A dictionary of epidemiology. New York: Oxford Univer sity Press.
Light, R. (1973). Abused and ne glected children in America : A stu dy of alternative pol-
icies. Harvard Educational Review, 43, 556-598.
Mau sn er, J. S., & Kramer, S. (1985). Epidemiology: An introductory text. Philadelphia: W. B.
Saunders.
Maximu s, Inc. (1986a). Child maltreatment: A secondary analysis of the American Humane
Association data base. Washington, DC: Author.
Maxirnus, Inc. (1986b). Child maltreatment: A secondary analysis of the American Humane
Association data base: Exhibits. Washington , DC: Au thor.
Maximu s, Inc. (1986c). Assessingalternative statistical estimation procedures for determining the
incidence of child maltreatment: Appendix A: Reporting, substantiation, and confirmation
rates, 1980 and 1983 (Final rep ort, Task Order U, BOA 105-84-8103, prep ared for the
Office of Program Development, Office of Human Developm ent Services). Wash-
ing ton, DC: Author.
Miller, S. H. (1984). The relation sh ip between ad olescent childbearin g and child maltreat-
ment. Child Welfare, 63, 553- 557.
Mitche l, L. (1987). Child abuse and neglect fatalities: A review of the problem and strategies for
reform (Working Paper No. 838). Chicago : Na tional Committee for Prevention of Child
Abus e.
Morri s, J. L., Johnson, C. F., & Clasen, R. W. (1985). To report or not rep ort: Physician's
attitudes toward d iscipline and child abu se . American Journal of Diseases of Children,
139, 194- 197.
National Center on Child Abu se and Neglect. (1981). Study findings: National study of the
incidence and severity of child abuseand neglect (DHHS Publication No . OHDS 81-30325).
Washington, DC: U. S. Government Printing Office.
National Center on Child Abu se and Neglect. (1988). Study findings: Study of national
incidence and prevalence of child abuse and neglect: 1988. Washington, DC: U. S. Depart-
ment of Health and Human Services.
Newberger, E. H. (1983). The help ing hand strikes again: Unintended cons equences of
child abuse reporting. Journal of Clinical Child Psychology, 12, 307-311.
O'Toole, R., Turbett, P., & Nalepka, C. (1983). Theories, prof essional knowledge, and
diagnosis of child abu se. In D. Fink elhor, R. J. Gelles, G. T. Hotalin g, & M. A. Strau s
(Eds .), The dark side of families : Current family violence research (pp. 349-362). Beverly
Hills, CA: Sage .
EPIDEMIOLOGY OF MALTREATMENT 53

Pelton, L. H. (1977). Child abuse and neglect: The myth of classlessness. American Journal of
Orthopsychiatry, 48, 608-617 .
Pelton , L. H. (1981). Introduction. In L. H. Pelton (Ed.), The social contextof child abuse and
neglect (pp . 11-21). New York: Human Sciences Press.
Polansky, N., & Williams, D. (1978). Class orientations to child neglect. Social Work, 23,
397-401.
Radda Barnen. (1980). The ombudsman and child maltreatment. Stockholm, Sweden: Radda
Barnen.
Risin, L. I. , & Koss, M. P. (1987). The sexual abuse of boys : Prevalence and descriptive
characteristics of childhood victimizations. Journal of Interpersonal Violence, 2, 309-323 .
Rosenthal, J. A. (1988). Patterns of reported child abu se and neglect. Child Abuse and
Neglect, 12, 263-271.
Russell , A., & Trainor, C. M. (1984). Trends in childabuse and neglect: A national perspective.
Denver: American Humane Association.
Russell, D. E. H. (1983). The incidence and prevalence of intrafamilial and extrafamilial
sexual abuse of female children. Child Abuse and Neglect, 1987, 133-146 .
Russell, D. E. H . (1984). The prevalence and seriousness of incestuous abuse: Stepfathers
vs. biological fathers. Child Abuse and Neglect, 8, 15-22 .
Sack, W. H., Mason, R., & Higgins, J. E. (1985). The single-parent family and abusive child
punishment. American Journalof Orthopsychiatry, 55, 252-259 .
Sedney, M. A., & Brooks, B. (1984). Factors associated with a history of childhood sexual
experience in a non-clinical female population. Journal of the American Academy of Child
Psychiatry, 23, 215-218 .
Silbert, M. H. , & Pines, A. M. (1981). Sexual child abuse as an antecedent to prostitution.
Child Abuse and Neglect, 5, 407-411.
Spearly, J. L., & Lauderdale, M. (1983). Community characteristics and ethnicity in the
prediction of child maltreatment rates . Child Abuse and Neglect, 7, 91-105 .
Starr, R. H., Jr. (1988). Physical abuse of children. In V. B. Van Hasselt, R. L. Morrison, A.
S. Bellack, & M. Hersen (Eds.), Handbook of family violence (pp. 119-155). New York:
Plenum Press .
Straus, M. A. (1979). Measuring intrafamily conflict and violence : The Conflict Tactics (CT)
Scale. Journal of Marriage and the Family, 41, 75-88.
Straus, M. A., & Gelles, R. J. (1986). Societal change and change in family violence from
1975 to 1985 as revealed by two national surveys. Journal of Marriage and the Family, 48,
465-479 .
Straus, M. A. , Gelles, R. J., & Steinmetz, S. K. (1980). Behind closed doors: Violence in the
American Family. Garden City, NY: Anchor Books.
Valentine, D. P., Acuff, D. 5., Freeman, M. L., & Andreas, T. (1984). Defining child
maltreatment: A multidisciplinary overview. Child Welfare, 63, 497-509 .
Whiting, L. (1976). Defining emotional neglect. Children Today, 5, 2-5.
Wyatt, G. E. (1985). The sexual abuse of Afro-American and white-American women in
childhood. Child Abuse and Neglect, 9, 507-519 .
Wyatt, G. E., & Peters, S. D. (1986a). Issues in the definition of child sexual abuse in
prevalence research. Child Abuse and Neglect, 10, 231-240.
Wyatt, G. E., & Peters, S. D. (1986b). Methodological considerations in research on the
prevalence of child sexual abuse. Child Abuse and Neglect , 10, 241-251.
Zuravin, S. J. (1989). The ecology of child abuse and neglect: Review of the literature and
presentation of data. Violence and Victims, 4, 101-120.
 CHAPTER 3

TRENDS IN LEGISLATION AND

CASE LAW ON CHILD ABUSE
AND NEGLECT
RANDY K. OTTO AND GARY B. MELTON

INTRODUCTION
The recognition of child abuse and neglect as a significan t social problem
in the United States is a relatively recent de velopment. Although most
states had passed specific child maltreatment laws by the earl y 1920s, it
was not until publication of a 1962 article describing the "battered-child
syndrome" (Kempe, Silverman, Steele, Droegenmuller, & Silverman,
1962) that legislators and health care professionals paid considerable
attention to the problem of child abuse and neglect. Since then, there
have been several waves of legislation and judicial activity that have
been nearly universal in American jurisdictions but that seldom have
had unequivocally positive effects.

RANDYK. O TTO • Department of Law and Mental Health , Florida Mental Health Institute,
University of South Florida, Tampa, Florida 33612-3899. GARY B. MELTON' Dep art-
ment of Psychology, Unive rsity of Nebraska-Lincoln, Lincoln, Nebraska 68588-0308.

55
56 RANDY K. OITO and GARY B. MELTON

JUSTIFICATION FOR STATE INTERVENTION

PARENS PATRIAE POWER

To evaluate the trends in law on child maltreatment, a useful start-
ing place is consideration of the broad philosophical and empirical foun-
dation of such policies . Without question, the state has a legitimate
interest in the welfare of its citizens. Accordingly, under its parens patriae
(sovereign as parent) power, the state may take actions to protect those
individuals who are considered unable to protect or care for themselves
(e.g ., minors, the mentally disabled). In some cases, the state's interest
as parens patriae is so compelling that it overrides even fundamental
rights, such as the right to family privacy (see, e.g ., Prince v. Massachu-
setts, 1944; cf. Roe v. Wade, 1973). Serious child maltreatment presents
just such a situation.
Although the state's parens patriae power is expansive, it is not
without boundaries (see, e.g ., Pierce v. Society of Sisters, 1925; Wisconsin v.
Yoder, 1972). For example, an involuntary intervention to protect child
welfare can be justified only if no action less intrusive on family privacy
would accommodate the state's compelling interest in the healthy so-
cialization of dependent children. Indeed, family privacy is so funda-
mental that it must be considered even after children have been re-
moved from the custody of their biological parents because of a demon-
strated lack of safety in the home (Melton & Thompson, 1987). In San-
tosley v. Kramer (1982), the Supreme Court concluded that

[t]he fundamental liberty interest of natural parents in the care, custody, and
management of their child does not evaporate simply because they have not
been model parents or have lost temporary custody of their child to the state.
Even when blood relationships are strained, parents retain a vital interest in
preventing the irretrievable destruction of their family life. (p. 753)

ORIENTATION AND SCOPE OF INTERVENTION

In the face of a need to balance critical needs to protect both child
welfare and family privacy, no consistent orientation exists among child
advocates about the nature of child maltreatment, the value of parental
autonomy, and the merits of various interventions to prevent harm to
children (Bourne & Newberger, 1977; Gelles, 1982; Melton, 1987a;
Melton, Petrila, Poythress, & Slobogin, 1987). Accord ingly, there is little
TRENDS IN LEGISLATION AND CASE LAW 57

agreement about either the specific circumstances that justify child pro-
tective jurisdiction or the level of coercive intervention that is desirable.
One school of commentators, emphasizing the deleterious effects of
abuse and neglect, argues that the fundamental value placed on family
privacy undermines the state's ability to protect children from abuse and
neglect at the hands of their caretakers (e.g ., Bourne & Newberger, 1977;
Feshbach & Feshbach, 1976; Garbarino, 1977, 1982; Garbarino, Gaboury,
Long, Grandjean, & Asp, 1982). This group perceives children as partic-
ularly vulnerable and in need of special protection by the state. "Family
privacy" is considered merely to shield abusive families from public
scrutiny, rather than protect them from unnecessary or unjustified state
intervention. Accordingly, this group advocates minimal restrictions on
the state in its attempts to identify and intervene in cases of suspected
abuse. Such "child savers" support adoption of low standards for invok-
ing state intervention and aggressive, high levels of intervention in cases
of suspected abuse or neglect.
By contrast, two other schools of thought share the belief that state
intrusion into family life is rarely advisable, even though they disagree
why families should be better insulated against state intervention in
cases of suspected abuse and neglect. One particularly influential family
law scholar, Michael Wald (1975, 1976, 1982; Wald, Carlsmith, & Leiber-
man, 1988) has relied primarily on utilitarian arguments to support fami-
ly integrity and privacy. Given the potential harm and lack of clear
benefit frequently associated with state intervention in cases of alleged
abuse (see, e.g., Children's Defense Fund, 1987; Clark Foundation, 1985;
Mnookin, 1973), Wald has advocated clearly defined, strictly limited
bases for state intervention. In short, he has concluded that more harm
usually will be done by intervention than would have occurred if the
state had ignored possible child maltreatment. Accordingly, Wald con-
tends that coercive intervention should occur only when there is clear
evidence that serious harm will result from inaction or a relatively unin-
trusive intervention.
An important step toward adoption of Wald's view carne with the
publication of the Juvenile Justice Standards Relating to Child Abuse
and Neglect (Institute of Judicial Administration/American Bar Associa-
tion [IJA/ABA], 1981). The Standards generally would limit state inter-
vention to "situations in which there are findings that a child has suf-
fered, or is at substantial risk of suffering, serious harm, and ...
intervention is necessary to protect the child from being endangered in
the future" (Melton et al., 1987, p. 311). Although the Standards Relating
58 RANDY K. OTIO and GARY B. MELTON

to Abuse and Neglect, unlike most of the other volumes of the Juvenile
Justice Standards, have not been adopted by the ABA as its official
policy, the Standards remain an important reference for critics of the
child welfare system.
A second school argues for family privacy and greater protection
from state intrusion on the grounds that such intervention, insofar as it
threatens children's perceptions of their "psychological parents," can
have extremely deleterious effects (Goldstein, Freud, & Solnit, 1973,
1979). Like Wald (1975, 1982), these commentators would set a high
threshold before state intervention could take place (indeed even higher
than Wald advocates), but they differ from Wald in that they support
aggressive intervention (even immediate termination of parental rights)
once serious abuse or neglect is substantiated. Goldstein et aI. believe
that the best interests of the child (or, to use their term, the "least
detrimental alternative") are best ensured by minimization of uncertain-
ty and unpredictability and promotion of children's belief in the om-
nipotence of their parents. Their approach involves forceful, immediate
measures, but only when it is clear that serious abuse or neglect has
occurred. Otherwise, Goldstein et al. advocate virtually unfettered def-
erence to parental autonomy.
With such a difference of opinion among experts in the fields of
child mental health and family law, it is not surprising that the approach
that states should adopt in cases of suspected abuse and neglect is
controversial. Unfortunately, there is no reason to believe that this lack
of consensus will be remedied in the immediate future.
Even if a consensus about the nature of child maltreatment and the
appropriate policy responses to it is lacking, it is clear that the general
direction since the 1960s has been toward more expansive concepts of
abuse and neglect, accompanied by correspondingly increased interven-
tion, despite the paradoxical evolution of constitutional law on family
privacy during the same period (Melton, 1987c). The impression held by
legislators and the public alike is that there is a need for broad standards
for state intervention. For example, the scope of maltreatment for legal
purposes has expanded to include emotional abuse and neglect, even
though application of such concepts in a manner that is not arbitrary or
discriminatory may be impossible (Melton & Thompson, 1987). Accom-
panying expanded definitions have been calls for greater use of criminal
sanctions, especially in regard to sexual abuse. In that respect, in recent
years, legislatures in every American jurisdiction have adopted pro-
cedural and evidentiary reforms designed to make prosecution easier
(Bulkley, 1985), despite questions of the efficacy of a "get-tough" strat-
TRENDS IN LEGISLATION AND CASE LAW 59

egy generally (Melton, 1987a) or of utility of the specific reforms, many

of which are of dubious constitutionality (Melton, 1987b; see Coyv. Iowa,
1988).

CRIMINAL AND CIVIL ADJUDICATION OF CHILD ABUSE

AND NEGLECT

Criminal Adjudication
Although criminal prosecution has been rare until recent years,
remedies for child abuse and neglect have been available in both the civil
and criminal arenas for a considerable period of time (Davidson & Horo-
witz, 1984). All 50 states provide potential criminal sanctions for child
maltreatment. As a matter of practice, though, criminal prosecution is
generally reserved for sexual and the most serious cases of physical
abuse. Data gathered by Midonick (1972) are illustrative, although some-
what out of date. Examining New York records, Midonick found that
fewer than 10% of all reports of child abuse found their way to family
court, and fewer than 10% of those cases adjudicated were referred to
criminal court.
A philosophical argument can be made that child abuse and neglect
should be adjudicated criminally because to do otherwise minimizes the
seriousness of the behavior. "Decriminalizing" what is otherwise crimi-
nal behavior is considered to send a subtle message to the public. And as
with other crimes, criminal adjudication is considered by some to pro-
mote specific and general deterrence (Chisolm, 1978; Davidson, Horo-
witz, Marvell, & Ketcham, 1981).
Advocates of criminal adjudication argue that it best ensures the
safety of children in serious cases by incapacitating the abuser when
necessary (Davidson, 1981). Even with the use of restraining orders, civil
adjudication cannot guarantee that the abuser will not have further con-
tact with the abused child (or other children).
Additionally, use of criminal prosecution is advocated because it is
considered to better ensure compliance with sanctions that may be im-
posed by the judge. Compliance with criminal court-ordered treatment
is thought by some to be more likely than compliance with civil court-
ordered treatment (Urzi, 1981).
Criminal adjudication is also advocated on the grounds that law-
enforcement officials will investigate allegations more aggressively if
they believe that a criminal prosecution will result (Urzi, 1981). Criminal
prosecution, because of the greater due process requirements and bur-
60 RANDY K. OTIO and GARY B. MELTON

den of proof, is also more protective of families' privacy rights than civil
adjudication.
Finally, recent work suggests that criminal adjudication may be able
to reduce recidivism in criminal behaviors traditionally considered to be
psychosocial problems. Like child abuse, spouse abuse has traditionally
been managed through noncriminal interventions (Costa, 1983). More
recently, however, states have begun to respond to spouse abuse using
the criminal justice system. This approach has met with some success
and may have implications for the management of child abuse.
In a demonstration project conducted in Minneapolis, Sherman and
Berk (1984) found that individuals arrested for assaulting their spouse
showed a somewhat lower recidivism rate than those who were either
counseled by, or simply separated from their spouse by the responding
police officer. Certainly, these preliminary findings are open to in-
terpretation but they do suggest that criminal adjudication may be indi-
cated with some types of behavior traditionally considered to be psycho-
social problems and amenable to treatment.
The parallel to child abuse and neglect is clear. Criminal sanctions
may be effective in reducing child abuse and neglect, at least with a
segment of the population. However, research comparing recidivism
rates among abusive and neglectful parents who have been processed
through the criminal justice and civil systems needs to be conducted.
It appears then that there are specific advantages associated with
criminal adjudication. These include the provision of considerable pro-
cedural protections to families accused of abuse, ability to manage and
control individuals who present a continuing, serious threat to their
victims and other children, and the potential impact of general and
specific deterrence by way of threatened imposition of criminal sanc-
tions.

Civil Adjudication
Although criminal prosecution has become more common in child
maltreatment, civil intervention long has predominated in such cases.
The first child protective agency was established in New York in 1875,
and by the mid-1920s all states had laws prohibiting child maltreatment
(Besharov, 1983).
Perhaps foremost, civil adjudication of abuse and neglect has been
based on the belief that these behaviors are psychosocial problems that
can be differentiated from criminal behavior (Fontana & Besharov, 1979;
Rosenberg, 1975; Urzi, 1981). Abusing one's child often is considered to
be a mental health problem in itself, or at least indicative of underlying
TRENDS IN LEGISLATION AND CASE LAW 61

mental health problems. Such a belief predominates among the public

even for sexual abuse (Finkelhor, 1984), the form of maltreatment by far
most likely to generate a criminal complaint. Therefore, it is argued,
intervention should be treatment-oriented, rather than punitive in
nature. A collateral argument is that rehabilitation and treatment may
not be accomplished, or are at least are compromised, when conducted
under the auspices of the criminal justice system (Dickens, 1978).
Also supporting civil intervention in cases of child abuse and ne-
glect is the belief that abusive and neglectful behaviors are particularly
responsive to treatment (Midonick, 1972). This argument is appealing if
maltreatment is indeed a product of parental pathology, inadequacy, or
skills deficits remediable by therapeutic intervention. Court-ordered
treatment or other coercive interventions pursuant to civil jurisdiction
could be used to promote the family court's goals of maintenance of
family integrity and prevention of further abuse.
In the same vein, civil intervention often is favored because it ac-
cords greater dispositional flexibility than do criminal sanctions. In civil
adjudication, no one caretaker need be labeled as the problem. Rather,
when appropriate, the family system or multiple caretakers may be
identified as dysfunctional and interventions instituted accordingly.
Critics of criminal sanctions argue that their usefulness as a means of
family change is severely compromised by their focus on single indivi-
duals.
Such utilitarian arguments cannot be considered in isolation. If civil
intervention is to be justified on the grounds that it facilitates appropri-
ate treatment, then the efficacy of treatments available to the family
court is relevant. Unfortunately, interventions aimed at decreasing
abusive and neglectful behaviors have not met with much success. Re-
cidivism rates in model demonstration projects for the treatment of
abusive and neglectful parents approach 33%, with the best estimate
being that approximately one third of treated clients show improvement
during treatment (Cohn, 1979; Melton et al., 1987; Rosenberg & Hunt,
1984; Wald, Carlsmith, & Leiberman, 1988). Such findings call into ques-
tion the justification of civil intervention on the ground that further
abuse and neglect may be prevented through treatment.
Civil intervention in the case of abusive and neglectful caretakers
also has been justified on the grounds that it is less likely to traumatize
the child than criminal adjudication. Criminal adjudication may create
further stress for a family already in crisis, thereby placing the child at
greater risk for abuse or neglect (Dickens, 1978; IJA/ABA, 1981). Al-
though there is little empirical support for such a claim (Melton, 1987b),
many commentators believe that children may be revictimized if they
62 RANDY K. 0Tf0 and GARY B. MELTON

are required to testify in a criminal court in which the defendant's rights

to a public trial, a jury trial, and confrontation of witnesses are preserved
(DeConey, 1975).
It is also claimed that interventions should be nonpunitive in order
to encourage self-reporting (Dickens, 1978). Knowing that they are sub-
ject to criminal penalties (including prison sentences and resulting sepa-
ration from their family), parents may be less likely to come forward and
admit to abuse or neglect (Fontana & Besharov, 1979). Such an argument
ignores the fact that even in noncriminal dispositions the potential sanc-
tions are severe and may serve to deter self-reporting. In civil adjudica-
tion, the possibility remains that the abusive caretaker or abused child
will be separated from the family. Thus, even under the current system,
whereby abuse cases are frequently adjudicated by civil courts, the large
majority of cases are not reported by the abusive caretaker or his or her
spouse.
Critics also argue against criminal prosecution, because it requires a
more rigorous standard of proof and accordingly may result in more
false negative errors and, therefore, less protection of children (Mid-
onick, 1972). Because of the need for stronger evidence in criminal than
in civil cases, criminal prosecutors also may be more likely to seek the
testimony of child victims, who correspondingly may be especially leery
of testifying because of the combination of procedural protections of the
rights of criminal defendants and because of the penalties that they
ultimately may endure (DeConey, 1975).
In summary, as compared to the criminal system, civil adjudication
increases the likelihood that intervention can be ordered (as a result of
lower burdens and less procedural rigor), provides the decisionmaker
with more latitude in identifying those who are responsible for the
abuse and are in need of treatment, and often offers a greater range of
possible interventions. It does so, however, at the cost of increasing the
likelihood of incorrect adjudications and unnecessary or overly intrusive
interventions.

Discussion
The debate about the relative merits of criminal and civil interven-
tion does not result in a clear answer. Each form of intervention has
advantages and disadvantages of varying importance in particular cases,
and both civil and criminal sanctions may be sought in some instances.
Some (e.g., Chisolm, 1978) have argued that the courts and child protec-
tive workers should choose the judicial intervention accordingly, with
the child's best interest being of paramount concern. Such a principle
TRENDS IN LEGISLATION AND CASE LAW 63

may be attractive to legal decision makers because of its consistency with

the historic individualized approach to children's cases. However, such
an approach is not without problems. As a practical matter, courts (and
the experts advising them) may be unable to differentiate "dangerous"
and "non-dangerous" caretakers (as suggested by Chisolm, 1978). More-
over, a policy of individualized justice increases the probability of dis-
crimination on the basis of ethnicity, social class, or other suspect classi-
fications . Perhaps most fundamentally, case-by-case decisions about the
path that adjudication will take beg the question of whether maltreating
parents deserve punishment, as a matter of justice.
Although the political climate has changed sufficiently that an un-
equivocal "model" answer no longer would be likely to emerge, the
Juvenile Justice Standards (IJA/ABA, 1981) provide a clear preference for
civil action because of the paramount interests of the child and the
presumed adverse effects of the criminal process on the victim:
Criminal prosecution for conduct that is the subject of a petition for court
jurisdiction filed pursuant to the se standards should be authorized only if the
court in which such petition has been filed certifies that such prosecution will
not unduly harm the interests of the child named in the petition. (IJAI ABA,
1981, p. 180)

Whatever the preferred balance between treatment and punish-

ment and between the interests of the victim in a rapid return to nor-
malcy and of society in seeking retribution, concepts of the nature of
children, abusers, and maltreatment itself have had an ongoing effect on
legal policy. As each new social construct has been "discovered," a wave
of legislation has followed .

LEGISLATIVE RESPONSES

THE FIRST WAVE OF LEGISLATION: CHILD ABUSE

REPORTING STATUTES

The first such wave followed publication of the seminal article on

the "battered child syndrome" by Kempe and his colleagues (Kempe,
Silverman, Steele, Droegenmueller, & Silver, 1962). With the discovery
of a purported medical syndrome, case-finding was the logical public
health response. Accordingly, between 1963 and 1967, alISO states and
the District of Columbia passed statutes requiring that health profes-
sionals report cases of suspected abuse (Brown, 1974).
Current child abuse statutes typically identify different types of
behavior defined as constituting abuse or neglect, specify the agencies
64 RANDY K. OTID and GARY B. MELTON

that are responsible for abuse investigation, name those individuals re-
quired to report abuse, and list penalties for failure to report. Addi-
tionally, these statutes generally contain clauses overriding professional
privileges (with the general exception of the attorney-client privilege)
and granting immunity from civil liability for good faith reporting.
Initially, only physicians were statutorily mandated to report cases
of suspected abuse, presumably because they were considered to be in a
unique position to identify such cases (Isaacson, 1975; Paulsen, 1967).
Gradually, however, legislators expanded reporting statutes to the point
that most child professionals and many laypersons are required to re-
port suspected abuse (see Myers, 1986, for a review of the 50 state
reporting statutes).
This first wave of legislation had a considerable impact on the
number of suspected abuse cases reported to child welfare and law-
enforcement authorities. Besharov (1983) noted that there was an
eightfold increase in the number of cases of suspected abuse reported
between 1966 (150,000) and 1980 (1.1 million).
Reporting statutes have certainly not been the panacea that some
might have hoped. The most common criticism of these statutes is that
they are vague and overbroad (Besharov, 1983, 1984; Davidson & Horo-
witz, 1984; Flicker, 1977; Melton, 1987a; Melton & Thompson, 1987).
Representative of states' abuse laws is New York's reporting statute,
which mandates that health care and law-enforcement professionals re-
port abuse when "they have reasonable cause to suspect" abuse (New
York Social Services Law Sec. 413, McKinney, 1988). Exactly what the
professional needs to see or believe in order to report abuse and thereby
comply with the reporting statute is not made clear in most statutes.
There may also be difficulty reaching a consensus about the behav-
ior that falls within the scope of child maltreatment. Research has shown
that there is considerable disagreement between professionals regarding
what constitutes sexually abusive behavior on the part of caretakers
(Atteberry-Bennett, 1987; Atteberry-Bennett & Reppucci, 1986). It can
even be difficult to reach agreement about the boundaries of physical
abuse (Giovannoni & Becerra, 1979). There is considerable disagreement
about the circumstances under which spanking and other, less severe
forms of corporal punishment should be viewed as abusive (Alvy, 1975).
The matter becomes still more complicated when neglect or "failure
to provide" is considered. All states consider parents' failure to provide
their children with basic care and necessities to be neglectful (Davidson
& Horowitz, 1984). Nonetheless, the scope of "basic care and neces-
sities" is not self-evident. Although some state statutes requiring par-
ents to provide adequate shelter, food, and care have been struck down
TRENDS IN LEGISLATION AND CASE LAW 65

as unconstitutionally vague, others have withstood judicial scrutiny. In

the latter cases, the courts have generally held that such broad language
is necessary to ensure the protection of children (Davidson & Horowitz,
1984).
Further complicating matters is the issue of intent. Should im-
poverished parents who are unable to provide basic necessities (e.g .,
food, clothing, a safe physical environment, and medical care) be treated
differently from parents who are financially able, but who decide not to
provide such necessities? Although there may be general agreement that
the latter case constitutes neglect and some form of coercive interven-
tion is indicated, whether and what type of state intervention is required
in the former case is less clear. Because the civil child protective system
is supposed to be nonpunitive, intervention is premised on conse-
quences of behavior rather than blameworthiness, but intrusion into the
family nonetheless seems unfair when the harm is not the parents' fault .
Perhaps most difficult to define are the concepts of emotional or
psychological abuse and neglect, which are included in the majority of
state reporting statutes. Not surprisingly, sections describing emotional
abuse and neglect lack specificity. Most statutes refer to infliction of
"mental injury," without a more precise definition (Melton & Corson,
1987). The nature of the triggering behavior by a parent or response by a
child is unclear. For example, should the state intervene when the par-
ent engages in an admittedly abusive behavior when there is no appar-
ent injury to the child, or is it the child's emotional adjustment that
should serve as the triggering mechanism? The analogue to the problem
of blameworthiness is even more acute in emotional maltreatment than
it is in physical neglect, because some intentional parental behavior
(e.g ., divorce) that is known often to have harmful psychological effects
on children seems intuitively to lie outside the boundaries of abuse
(Melton & Thompson, 1987). Other behaviors that may have predictable
consequences often viewed as negative for children are strongly linked
to social class . A purely outcome-based definition thus risks class-based
intrusions into family life.
Given these difficulties, allegations of emotional abuse are usually
joined with allegations of physical maltreatment or sexual abuse, and
they are rarely filed alone. Because of the difficulty drawing a causal
nexus between the parent's allegedly abusive behavior and the child's
psychological state, many argue that the child's mental injuries should
be both substantial and observable, and be clearly linked to parental
behavior (Besharov, 1984; Melton & Thompson, 1987).
Data regarding the proportion of abuse reports that are later sub-
stantiated support the conclusion that the reporting standards are, at a
66 RANDY K. OTIO and GARY B. MELTON

minimum, unclear and possibly overbroad. Currently, between 40% and

43% of abuse and neglect reports are substantiated, and approximately
25% of these require some type of court action (National Center for State
Courts, 1988). Thus, less than one half of current child abuse and neglect
reports are substantiated. The high rate of unsubstantiation probably
results from a number of factors, including defensive practice by profes-
sionals fearing repercussions for failure to report (but see Maney &
Wells, 1988, on underreporting by professionals), inadequate investiga-
tory resources in child protective services (CPS), and the intrinsic diffi-
culty of proving some forms of maltreatment of young children. Failure
to substantiate maltreatment may also result from professionals' confu-
sion because of vague and possibly overbroad standards.
One of the criticisms of such vague and apparently value-laden
standards is that they open the door to arbitrary, discriminatory applica-
tion of state power. Mandated reporters effectively can pick and choose
standards to apply, and thus may impose their own values on indi-
viduals of different cultural backgrounds and socioeconomic status
(Davidson & Horowitz, 1984; Dickens, 1978; Melton, 1987a; Melton et
al., 1987).
Attempts to prevent and manage child abuse and neglect through
state legislation clearly have not met with complete success. Although
some gains have been made (e.g., the number of abused children com-
ing to the attention of authorities), there have been associated costs
(e.g., a considerable number of unsubstantiated reports; interventions
that sometimes may be more disruptive than helpful). Moreover, profes-
sionals are confused about their roles and responsibilities. The mixed
picture probably has resulted in part from the often quick and essen-
tially unplanned manner in which state legislatures often have ap -
proached the difficult policy problems related to child maltreatment
(Lucht, 1975; Nelson, 1984).

CHILD ABUSE PREVENTION AND TREATMENT ACT OF 1974

The picture at the federal level is a little rosier. The federal govern-
ment has been a late and ambivalent participant in policy on child mal-
treatment. Until 1974, virtually all government activity on child mal-
treatment was at the state level. With passage of the federal Child Abuse
Prevention and Treatment Act (CAPTA) of 1974 (Public Law 93-247), the
federal government began to set the agenda for child abuse prevention
and treatment. Prior to enactment of CAPTA, federal support for abuse
and neglect programs was limited largely to Title IV-B of the Social
Security Act, which authorized general child welfare services. However,
TRENDS IN LEGISLATION AND CASE LAW 67

in 1973, the year before passage of Public Law 93-247, only $507,000 of
Title IV-B monies went toward child abuse activities nationally (House of
Representatives Report No. 685, 1974).
Noting that state efforts largely went toward reporting and that
there was inadequate follow-up, the Child Abuse Prevention and Treat-
ment Act attempted to establish a research agenda and prevention and
treatment programming to be coordinated through the newly estab-
lished National Center on Child Abuse and Neglect (NCCAN). The act
authorized the director of NCCAN to compile, analyze, and publish
research results regarding child abuse; develop and maintain an infor-
mation clearinghouse; compile and publish training materials; provide
technical assistance (through grants and contracts) to assist public or
private nonprofit organizations in the planning, development, and im-
plementation of prevention and treatment programs; and conduct re-
search regarding causes and prevention of child abuse.
To be eligible for CAPTA funding, states had to meet certain criteria,
most of which were related to reporting and investigation procedures.
In order for states to receive CAPTA funding they had to have a system
in place for abuse reporting and investigation, provide criminal and civil
immunity for good-faith reporting, ensure that abuse reports and rec-
ords were kept confidential, and provide guardians ad litem for all chil-
dren. CAPTA also required states to maintain their funding of abuse
programming and extend preferential treatment to parental organiza-
tions organized to prevent and treat child abuse.
Presumably in response to some of the criticisms presented above,
there was discussion of the difficulty in defining such terms as neglect
and mental or emotional injury in the House report. However, the bill, as
passed, offered the following definition: "This section defines the term
'child abuse and neglect' as the physical or mental injury, sexual abuse,
negligent treatment, or maltreatment of a child under the age of eighteen
by a person who is responsible for the child's welfare" (House of Repre-
sentatives Report No. 685, 1974; emphasis added).
The Child Abuse Prevention and Treatment Act of 1974 has been
amended several times so that its scope is now somewhat broader. A
number of important amendments were passed in 1984. Public Law
98-457 extended the definition of abuse and.neglect to include the sexual
exploitation and the withholding 'of medically Indicated treatment (the
"Baby Doe" amendment), and expanded CAPTA so that abuse and ne-
glect occurring in residential and out-of-home placements was also cov-
ered by the act .
An amendment passed in 1986 (Public Law 99-401) directed the
Secretary of Health and Human Services to establish demonstration
68 RANDY K. OTIO and GARY B. MELTON

projects to provide in-home and out-of-home temporary nonmedical

child care for handicapped children and children with chronic and ter-
minal illnesses, and to provide crisis nurseries for children at risk for
abuse and neglect.
The Children's Justice and Assistance Act (Public Law 99-401) was a
1986 CAPTA amendment which encouraged states to enact reforms de-
signed to improve legal and administrative proceedings in child sexual
abuse cases. Such reforms, it was hoped, would protect children from
trauma associated with involvement in the abuse investigation and pros-
ecution, and thereby improve the chances of successful prosecutions.
The amendment earmarked 24 million dollars for fiscal years 1987 and
1988 (Senate Report No . 123, 1986).
As a funding statute, Public Law 93-247 has had its greatest impact
in terms of developing state child abuse reporting and investigation
programs. And these reporting requirements appear to have some effect
on the identification of abused and neglected children. Studies suggest
that mental health professionals have gradually become more knowl-
edgeable of, and increased their compliance with, reporting require-
ments (d. Kalichman & Craig, in press; Kalichman, Craig , & Follingstad,
1987; Swoboda, Elwork, Sales, & Levine, 1979). Still, a considerable
number of professionals fail to comply with statutory requirements and
report suspected abuse or neglect (Maney & Wells, 1988). There has also
been a considerable increase in the number of reports made by nonman-
dated reporters (Melton, 1987b).

OTHER FEDERAL LEGISLATION

In addition to the Child Abuse Prevention and Treatment Act and

its amendments, other measures have been enacted by the federal gov-
ernment that are designed to minimize child abuse and neglect, a
number of which are reviewed below.
The Parental Kidnapping Prevention Act of 1980 attempts to elimi-
nate some of the legal and practical difficulties occurring in the case of
parental kidnapping. The act orders states to honor and enforce any
child custody determinations made in the courts of other states and
provides assistance to the custodial parent in the case of abduction by
the noncustodial parent, including use of state and federal locator
services.
The Missing Children Assistance Act of 1984 was enacted as a par-
tial response to the problem of runaway and missing children. The act
provided funds for establishment of a toll free hotline to report informa-
tion about missing children and other projects aimed at locating children
TRENDS IN LEGISLATION AND CASE LAW 69

or providing services to missing children and their families. The statute

also provided funding for research related to missing children.
The Child Sexual Abuse and Pornography Act of 1986 provides for
federal prosecution of persons engaged in child pornography, including
parents who permit their children to engage in such activities. In addi-
tion to fines and imprisonment, the statute also allows for the confisca-
tion of property used in conjunction with the crime. The statute identi-
fies federal district courts as the forums in which civil suits alleging
damages to minors engaged in pornography are to be heard, and it sets
the minimum award in such cases at $50,000.
As the above review suggests, much of the emphasis at the federal
level has remained on reporting, investigation, and prosecution. In con-
trast to this focus, Children's Trust Funds, which are state mechanisms,
place a greater emphasis on prevention and treatment.

CHILDREN'S TRUST FUNDS

Since 1980, advocates for abused and neglected children have estab-
lished Children's Trust Funds in 44 states (National Committee for the
Prevention of Child Abuse [NCPCA], undated). These funds establish a
permanent funding mechanism for child abuse and neglect prevention
and treatment programs at the community level.
Children's Trust Funds are designed to create continuing funding
mechanisms that promote prevention programming for child abuse and
neglect at the community level (NCPCA, undated). Revenues are gener-
ated from various methods, including: surcharges on marriage licenses
and birth certificates, checkoffs on state income tax forms, increases in
divorce filing costs, increased fees for death certificates, and heirloom
birth certificates. This approach allows for the funding of traditionally
neglected prevention programs and insulates them from budget cuts
when state monies become scarce.
The governing body of the trust fund creates a public/private part-
nership. In some states, existing bodies that include representatives
from state agencies, the legislature, and the private sector have adminis-
tered the fund. In other states, public advisory councils consisting of
public and private sector representatives have been created.
Trust fund boards oversee program development and implementa-
tion, hire program staff, review program proposals, and disburse funds.
The major thrust of trust fund programs to date has been in promoting
and funding primary and secondary prevention programs at the com-
munity level. These include, but are not limited to pre- and postnatal
support programs for first-time and recent parents (especially "at risk"
70 RANDY K. orro and GARY B. MELTON

families with unhealthy newborns), parent education and training

groups, parent self-help and neighborhood support groups, family sup-
port services (e.g., crisis babysitters, crisis nurseries, crisis counseling),
child care programs, treatment programs for abused and neglected chil-
dren, and community education on child abuse prevention (NCPCA,
undated).
Although the full impact of trust fund programs may not be deter-
mined at this point in time , they show promise in that they are nontradi-
tional programs that identify "at risk" populations and attempt to pro-
mote cost-effective prevention and treatment programs.

PROCEDURAL AND EVIDENTIARY REFORMS RELATED

TO ADJUDICATION
Although nontraditional in approach, the trust funds are consistent
with the historic preventive role of the civil child protection system. The
initial federal initiatives can be seen in the same light, in that they were
intended to uncover maltreatment in order to prevent further harm to
children. The most recent wave of legislation in the states (which, as we
have already noted, is paralleled in recent federal legislation) appears to
be oriented more, though, toward increasing the likelihood of punish-
ment of offenders against children, especially in sexual abuse cases.
Rather than focusing on protection of children from future harm,
state legislatures and, indirectly, the Congress (through, for example,
the Children's Justice Act) have been active in recent years in establish-
ment of procedural and evidentiary reforms intended to increase the
likelihood of successful prosecution of alleged abusers. Most states have
enacted new laws providing for special qualification of child victims as
witnesses, special hearsay exceptions, exclusion of spectators during
testimony by child victims, videotaped depositions, closed circuit televi-
sion, and special courtrooms (see Bulkley, 1985; Eatman & Bulkley, 1986;
Whitcomb, 1985).
The underlying assumption of these reforms is that standard crimi-
nal procedure unduly traumatizes the child victim and therefore com-
promises the ability of the criminal system to identify and sanction those
individuals guilty of abuse. All of the proposed reforms are designed to
facilitate acceptance of children's statements, often even when not made
in court, in order to increase the number of successful child abuse
prosecutions.
However noble the intentions of the framers of legislation creating
special procedures and hearsay exceptions in abuse cases, such reforms
TRENDS IN LEGISLATION AND CASE LAW 71

have been controversial and often may have increased children's limbo in
the criminal justice system by opening the door to appeals. Considerable
appellate litigation has resulted from many of the procedural and evi-
dentiary reforms enacted in recent years, because they typically threaten
the defendant's sixth-Amendment rights to a public trial and confronta-
tion of witnesses, his fourteenth amendment right to due process,
and/or the public's first amendment right to access (through the press)
to the trial process (Melton, 1987a).

THE IMPACT OF TESTIFYING UPON CHILD WITNESSES

Perhaps the most fundamental complaint about the recent wave of

legislation is that changes are unneeded and, if needed, may be ineffec-
tive or even counterproductive. This point may seem odd to many ob-
servers, who conclude intuitively that child victims are further trau-
matized by their participation in the legal process, particularly testifying
at trial. Little is actually known, however, about the impact that testify-
ing has on children (Haugaard & Reppucci, 1988; Melton, 1987a).
In the absence of empirical support, it is premature to conclude that
child victims' participation in the criminal process will prove particularly
traumatizing. That participation in the criminal process can have posi-
tive effects on adult victims' psychological functioning is well-accepted
(see, e.g. , Kelly, 1987; Kilpatrick & Otto, 1987), and the same may be true
for child victims. Staff from two federally funded demonstration pro-
jects for the evaluation and treatment of sexual abuse (the Harborview
Medical Center, Seattle, and the Children's Hospital National Medical
Center, Washington, DC) have published accounts noting the positive
impact that participation in the criminal process can have for child vic-
tims. Berliner and Barbieri (1984) claim that testifying can be therapeutic
for child victims insofar as the experience empowers them and teaches
them that their complaints are taken seriously. Rogers (1982) noted that
participation in the criminal process provides child victims with a so-
cially sanctioned opportunity for retribution and also may serve as offi-
cial acknowledgment that the victims were not responsible for their
victimization. Preliminary data reported by Runyan and his colleagues
(King, Hunter, & Runyan, 1988; Runyan, Everson, Edelsohn, Hunter, &
Coulter, 1988) suggest that child victims' adjustment may improve as a
function of testifying.
The social psychological literature examining procedural justice and
satisfaction lends further credence to these data. Work by Thibaut and
Walker (1978) suggests that adults perceive the adversarial method to be
the fairest and most satisfactory form of inquiry. At least among older
72 RANDY K. OTTO and GARY B. MELTON

children, then, participation in the adversarial process, even if it is

stressful at the time, is likely to heighten their sense of control and
perception of justice (Melton, 1983c; Melton & Limber, 1990; Melton &
Lind, 1982).
Although it cannot be expected that all child victims will benefit
from participation in the criminal process, the theoretical perspectives
and data presented above challenge the presumption that participation
in criminal proceedings (including testifying) will necessarily traumatize
the child victim. Thus, reforms aimed at insulating child victims from
the effects of participating in the criminal process may be unnecessary
and are likely to be found unconstitutionally overbroad.

CHILDREN'S COMPETENCY TO TESTIFY

At common law, children under the age of 10 were presumed in-

competent to testify and the competence of older children was rebutta-
ble (Melton et al., 1987). In evaluating a child's (and adult's) competency
to testify, the judge typically examines the person's ability to differenti-
ate the truth from falsehoods, the ability to understand the duty to tell
the truth, and the consequences for not doing so. Also evaluated are the
witness's ability to avoid suggestion and to have formed a just or valid
impression of the facts at the time of the incident in question (Melton,
1987c).
A considerable amount of research examining the competency of
child witnesses has been conducted in recent years (see Goodman, 1984;
Goodman & Helgeson, 1985; Haugaard & Reppucci, 1988; Melton et al.,
1987, for reviews). For the most part, the research literature suggests
that the general presumption of children's incompetency is unfounded.
In response to these findings, and in order to facilitate introduction
of children's testimony in abuse cases, states have begun to eliminate
presumptions of children's incompetency to testify (Melton, 1987b;
Whitcomb, 1985). These efforts have been reinforced by the Federal
Rules of Evidence (Rule 601) that presume all witnesses to be competent.
Although the federal rules only apply to federal proceedings, they serve
as model rules which are followed by a large number of states. Thus,
whereas the traditional common law system presumed child witnesses
to be incompetent to testify, the Federal Rules of Evidence (and the rules
in an increasing number of states) presume child witnesses-to be compe-
tent. In either case the presumption in effect (i.e., competency or incom-
petency) can be challenged.
TRENDS IN LEGISLATION AND CASE LAW 73

HEARSAY EXCEPTIONS

A number of states have developed special hearsay exceptions for

children who are victims/witnesses in sexual abuse cases (Berliner,
1985). These exceptions are designed to allow introduction of out-of-
court statements that would otherwise be excluded as hearsay.
Generally, hearsay is not admitted into evidence because it is con-
sidered less reliable than court-based testimony subject to cross-exam-
ination. Admission of hearsay evidence is considered to violate the de-
fendant's Sixth Amendment right to confront his or her accusers.
However, the bar against hearsay testimony in criminal proceedings is
not absolute. The Supreme Court has held that hearsay may be admitted
against criminal defendants when the prosecuting witness is unavailable
and there are "indicia of reliability"-as determined by traditional hear-
say exceptions (e.g ., complaint of rape exception, statements of medical
or physical condition, excited utterances) or other specific "guarantees
of trustworthiness" (Ohio v. Roberts, 1980).
Under the current system, two tacks may be taken in trying to admit
children's out-of-court statements. First, prosecutors may try to fit chil-
dren's statements regarding abuse into present exceptions such as the
excited utterance or medical statement exceptions. Alternatively, some
have advocated that another exception be made specifically for child
sexual abuse victims. Graham (1985), for example, proposes that the
hearsay statements of child victims be allowed (1) if the child also testi-
fies at trial, or (2) if the child is unavailable and the statement is sup-
ported by corroborative evidence of some type. Of course, unavailability
because of incompetency should serve to bar admission of testimony on
the grounds that, on its face, it lacks indicia of reliability. Unavailability
resulting from the anticipated trauma associated with testifying may not
be assumed (see, e.g., Connecticut v. Iarzbek; 1987) and would have to be
proven in the particular case .
Among the traditional exceptions, the excited utterance exception
allows introduction of out-of-court statements offered immediately fol-
lowing, and in response to the incident in question on the basis that they
are highly reliable. Because they frequently are given a considerable
amount of time after the alleged act, children's accounts to investigators
usually do not appear to meet the general requirement of the excited
utterance exception. Some courts have applied this common law excep-
tion liberally (Bulkley, 1981), however, and admitted children's hearsay
testimony under it (e.g. , Commonwealth v. Adams, 1987).
The medical statement exception allows introduction of out-of-court
74 RANDY K. orro and GARY B. MELTON

statements made to health care professionals during the course of treat-

ment, based upon the assumption that because individuals are unlikely
to deceive health care professionals who treat them, their statements are
especially trustworthy. Prosecutors may also seek admission of chil-
dren's allegations gathered by psychologists or physicians but here too,
such action does not appear consonant with the intent of the medical
exception which requires that the declarant's motive in making the state-
ment to be for the purpose of treatment (United States v. Iron Shell, 1980).
Again, however, some courts have admitted children's statements made
to health care professionals on the basis of the medical exception (e.g.,
In re Freiberger, 1986). The admissibility of 1990 statements made by
child victims to health care professionals will be addressed by the Su-
preme Court in Idaho v. Wright, which the court had yet to hear at the
time this chapter went to press.
Perhaps the most appropriate hearsay exception through which
children's out-of-court statements should be admitted is the general ex-
ception noted in Rule 803 of the Federal Rules of Evidence . Section 24 of
Rule 803 provides for the introduction of out-of-court statements which
do not qualify under any specific exceptions. To be admitted under this
exception the statement must have circumstantial guarantees of trust-
worthiness; be offered as evidence of a material fact; and be more pro-
bative than any other evidence which is available to the presenting
party. Additionally, it must be determined that the general interest of
justice is best served by introduction of the statement.

CLOSED COURTROOMS

In an attempt to minimize potential trauma experienced by the child

victim/witness in criminal proceedings, a number of states have passed
statutes closing courtrooms to spectators and the press during the
child's testimony (Whitcomb, 1985). Such measures are considered by
some to violate the defendant's Sixth Amendment right to a public trial
and the public's First Amendment right of access to trials through the
press. The Supreme Court examined the constitutionality of closed
courtrooms in Globe Newspaper Company v. Superior Court (1982). In Globe,
the Court ruled that a Massachusetts statute directing judges to bar the
press from all trials during the testimony of child sexual assault vic-
tims/witnesses was overbroad and in violation of the First Amendment.
However, the Court indicated that the courtroom could be closed in
some instances. Noting that the state has a compelling interest in pro-
tecting child victims from trauma, the Court ruled that, "the trial court
TRENDS IN LEGISLATION AND CASE LAW 75

can determine on a case-by-case basis whether closure is necessary to

protect the welfare of a minor victim" (p. 608).

SPECIAL COURTROOM PROCEDURES

In attempting to increase the likelihood of successful prosecution of

child abusers, child advocates also have pressed for reforms in trial and
courtroom procedure. In Kentucky v. Stincer (1987), the Supreme Court
ruled that the state may bar the defendant from children's competency
hearings in some sexual abuse cases. In reaching its decision, the Court
held that barring the defendant from the hearing did not violate his
Sixth or Fourteenth Amendment rights because it did not interfere with
his opportunity to cross-examine the witnesses effectively.
The recent Supreme Court decision of Coy v. Iowa (1988) suggests,
however, that implementation of special procedures designed to protect
child witnesses in abuse cases will be examined carefully by the Court.
At issue in Coy was an Iowa statute permitting use of a screen which
could be placed between defendants and child victims/witnesses while
they testified. The screen, designed to protect child witnesses from the
trauma associated with testifying in front of alleged assailants, blocked
the witnesses' view of the defendant, but allowed the defendant to see
and hear the witnesses as they testified.
The Court interpreted the meaning of the confrontation clause
strictly, holding that it guarantees the defendant a "face-to-face meeting
with witnesses appearing before the trier of fact" (p. 2800). Similar to its
decision in Globe, the Court ruled that the legislatively-imposed pre-
sumption of trauma was not sufficient. Such an exception, ruled the
Court, is not "firmly rooted" in our system of jurisprudence, and there
were no individualized findings that the witnesses in this case were in
need of special protection.
Together, the Court's decisions in Globe Newspapers and Coy suggest
that although special procedures may be used in cases where it has been
shown that the specific witness is at risk for significant trauma without
such measures, their mandatory imposition will be barred when a con-
stitutionally protected interest is infringed. How the court may treat
similar procedures, such as the presentation of videotaped testimony
and the use of closed circuit television, was addressed in the concurring
opinion of Justice O'Connor in Coy in which she was joined by Justice
White. After emphasizing the majority's suggestion that the screen
might be justified when the witness was shown to be at special risk,
O'Connor went on to discuss the opinion in terms of its applicability to
other procedures such as testimony via one- and two-way closed circuit
76 RANDY K. arm and GARY B. MELTON

television. Noting that, at the time, 25 states provided for such closed
circuit procedures and 33 allowed the presentation of videotaped testi-
mony of child victims, O'Connor wrote that
nothing in today's decision necessarily dooms such efforts by state legisla-
tures to protect child witnesses. Initially, many such procedures raise no
substantial confrontation clause problem since they involve testimony in the
presence of the defendant. (p. 2804)
O'Connor's concurrence in Coy suggests that the Court may not be
as hostile to special procedures in the future . It is possible that O'Con-
nor, White, and Kennedy (who took no part in the decision) will join the
Coy dissenters (Blackmun and Rehnquist) to form a majority more ac-
cepting of special procedures designed to protect witnesses, providing
that they are not mandatory and that they allow for "face-to-face" con-
frontation between the defendant and witness. The Supreme Court will
address this issue more specifically in Craig v. Maryland, which the court
had yet to hear at the time this chapter went to press.

PSYCHOLOGICAL TESTIMONY ABOUT ABUSER

AND VICTIM PROFILES

In the adjudicatory phase of abuse and neglect cases, the state or

defense may seek to introduce expert opinion regarding the psychologi-
cal characteristics of abusers. Through the introduction of this testi-
mony, the state seeks to create the inference that the defendant is like
people who abuse children and is, therefore, guilty of the allegations. In
a similar vein, the defense might seek to introduce such testimony on
the basis that the defendant is not like those ind ividuals who abuse
children in order to combat allegations of abuse or neglect. Because such
testimony is generally treated as character evidence, it is inadmissible
according to Federal Rule of Evidence 404 unless the defendant offers
such evidence first.
Mental health professionals may also be asked to testify about the
characteristics of abused or neglected children in order to support or
diminish the child victim's allegations. The least objectionable testimony
of this sort involves the expert explaining to the court aspects of the
child's behavior that might be confusing (Haugaard & Reppucci, 1988;
Melton & Limber, 1990). Thus, the expert might explain why some child
victims fail to report abuse immediately or why they sometimes recant
allegations during the investigation. Whether or not such testimony is
helpful depends largely upon the knowledge of the lay public (Melton &
Limber, 1990).
More problematic is expert testimony regarding "sex abuse syn-
dromes." Testimony about the behavior patterns of abused children is
TRENDS IN LEGISLATION AND CASE LAW 77

presented in order to create the inference that the child has or has not
been abused, depending on whether he or she displays the relevant
behaviors. Although such testimony is usually offered by the prosecu-
tion in order to buttress allegations of abuse, the defense may some-
times attempt to refute abuse allegations by showing that the child does
not fit the abused child profile .
The majority of appellate courts considering this issue have ap-
proved of the admission of opinions about the nature of the abuse
syndrome, providing that the expert does not offer an opinion about the
credibility of the alleged victim. Melton and Limber (1990) argue against
admission of such testimony on numerous grounds. First, they note that
there is little empirical support for a "sex abuse syndrome," and the
purported syndromes are based mainly on clinical intuition. Moreover,
many abuse victims display no symptoms at all, and many of the
"symptoms" of child abuse are common to other clinical populations
(Browne & Finkelhor, 1986; Haugaard & Reppucci, 1988).
Secondly, Melton and Limber (1990) note that introduction of such
testimony may mislead the decisionmaker. Even if it could be deter-
mined through introduction of expert testimony that the child was
abused, it does not prove that the child was abused by the defendant.
The decisionmaker may be predisposed, though, to hold the defendant
responsible for the abuse.
Finally, Melton and Limber (1990) argue against admission of such
testimony on the grounds that it, in some ways, puts the victim "on
trial." Such a line of inquiry may result in intrusions into the victim's
privacy through extensive evaluations by both the prosecution and
defense.

DISCUSSION

The development and implementation of procedural and evidenti-

ary reforms designed to facilitate (and sometimes minimize) children's
participation in the criminal process tracks the development of statutory
reforms aimed at increasing reports of abuse and neglect. Both sets of
reforms were developed with relatively little planning or experimenta-
tion and implemented without their efficacy first being demonstrated.
If children are able to testify under standard conditions, it is coun-
terproductive to introduce new procedures that are of questionable util-
ity and constitutional validity (Melton, 1987a). The use of unfounded
procedures only provides the defense with firm ground for appeal, the
end result being continued involvement by the child witness with no
real end or closure in sight. Because of the limitations of evidentiary and
procedural reform, the most useful techniques for involving child vic-
78 RANDY K. OTIO and GARY B. MELTON

tims in the criminal process may be a product of the increased resource-

fulness of mental health professionals working with children (Melton,
1987b; Melton & Limber, 1990; Whitcomb, 1985).
Mental health professionals skilled in working with children should
be able to facilitate communication between the child witness and legal
professionals, instruct legal professionals on how to interact with the
child, and provide support for children engaged in the criminal process.
Because of their expertise in family dynamics and discussing personal
matters, child specialists may also be helpful as investigators, especially
in the disposition phase, providing the court with information it might
not otherwise have (Melton et al., 1987). These and other activities can
facilitate the child's participation in the adjudicative process and help
ensure that the child is not unduly traumatized.

SUMMARY
The legislature and judiciary have paid increasing attention to the
problems of child abuse and neglect over the past 25 years. Numerous
reforms have been implemented in an attempt to increase reporting of
suspected cases and to facilitate prevention and treatment programs.
Additionally, judicial reforms have been instituted in order to facilitate
adjudication of sexual abuse and more serious cases of physical abuse.
These efforts have produced mixed results for a variety of reasons.
Because no consistent orientation exists among child advocates regard-
ing the nature of child maltreatment, reforms have been based on emo-
tional responses to the problem of abuse as much as empirical analysis
or a theoretical basis . Many reforms have been instituted rather hastily,
with little foresight and planning. Accordingly, we are left with pro-
cedures that are of questionable utility, some of which are also of ques-
tionable constitutionality.
The opportunities presented to psychologists are numerous, given
the current state of affairs. Most importantly, psychologists must ad-
dress definitional issues and study the effects of different policies and
procedures that are now in use . Additionally, psychologists should ap-
ply their expertise and assist child victims and judicial system, while at
the same time acknowledging their limitations.

REFERENCES
Alvy, K. T. (1975). Preventing child abuse. American Psychologist, 30, 921-928.
Atteberry-Bennett, [. (1987). Child sexual abuse : Definitions and interventions of parents
TRENDS IN LEGISLATION AND CASE LAW 79

and professionals. Unpublished doctoral dissertation. University of Virginia, Institute

of Clinical Psychology, Charlottesville, Virginia .
Atteberry-Bennett, J., & Reppucci, N. D. (1986, August). What does childsexual abuse mean?
Paper presented at the meeting of the American Psychological Association, Wash-
ington, DC.
Berliner, L. (1985). The child witness: The progress and the emerging limitations. In J.
BulkIey (Ed.), Papers from a national conference on legal reforms in child sexualabuse cases
(pp. 93-108). Washington, DC: American Bar Association.
Berliner, L., & Barbieri, M. K. (1984). The testimony of the child victim of sexual assault.
Journal of Social Issues, 40(2), 127-137.
Besharov, D. J. (1983). Child abuse: Past progress, present problems, and future directions.
Family LAw Quarterly, 17, 151-172.
Besharov, D. J. (1984). Protecting abused and neglected children: Can law and social work
help? In W. Holder & K. Hayes (Eds.), Malpractice and liability in childprotective services
(pp. 29-48). Longmont, CA: Bookmakers Guild .
Bourne, R., & Newberger, E. H. (1977). Family autonomy or coercive intervention? Ambi-
guity and conflict in the proposed standards for child abuse and neglect. Boston
University LAw Review, 57, 670-706 .
Brown, R. H. (1974). Child abuse: Attempts to solve the problem by reporting laws . Women
LAwyers Journal, 60, 73-78.
Browne, A., & Finkelhor, D. L. (1986). The impact of child sexual abuse: A review of the
research. Psychological Bulletin, 99, 66-77.
Bulkley, J. (1981). Evidentiary theories for admitting a child's out-of-court statement of
sexual abuse at trial. In J. Bulkley (Ed.), Child sexual abuse and the law (pp . 153-161).
Washington, DC: American Bar Association, National Legal Resource Center for Child
Advocacy and Protection.
Bulkley, J. (1985). State legislative reform efforts and suggested future policy directions to improve
legal intervention in childsexual abuse cases. Washington, DC: American Bar Association,
National Legal Resource Center for Child Advocacy and Protection.
Child Abuse Prevention and Treatment Act, 42 U.S.C.S. § 5101-5115 (1979, Cum. Supp.
1988).
Child Sexual Abuse and Pornography Act, 18 U.s.C.S. § 2251-2255 (1979, Cum. Supp.
1988).
Children's Defense Fund. (1987). A children's defense budget: An analysis of the President's FY
1988 budget and children. Washington, DC: Author.
Chisolm, B. A. (1978). Questions of social policy-A Canadian perspective. In J. M.
Eekelaar & S. N. Katz (Eds.), Family violence: An international and interdisciplinary study
(pp. 318-328). Toronto: Butterworths.
Clark Foundation. (1985). Keeping families together: Thecase forfamily preservation. New York:
Author.
Cohn, A. H. (1979). Essential elements of successful child abuse and neglect treatment.
Child Abuse and Neglect , 3, 491-496 .
Commonwealth v. Adams, 503 N.E.2d . 1315 (Mass . App. Ct. 1987).
Connecticut v. [arzbek, 204 Conn. 683, 529 A.2d. 1245 (1987).
Costa, J. J. (1983). Abuse of women: Legislation, reporting and prevention . Lexington, MA:
Lexington Books.
Costa, J. J., & Nelson, G. K. (1978). Childabuse and neglect: Legislation, reporting, and preven-
tion. Lexington, MA: D. C. Heath.
Coy v. Iowa , _ _ U.S. _ _ r 108 S. Ct. 2798 (1988).
Davidson, H. A. (1981). Sexual abuse of children: Effective utilization of the legal systems.
Excerpted in H. A. Davidson, R. M. Horowitz, T. B. Marvell, & O. W. Ketcham (Eds.),
80 RANDY K. OITO and GARY B. MELTON

Child abuse and neglect litigation: A manual for judges (pp . 145-146). Wash ington , DC:
Department of Health and Human Services.
Davidson , H . A., & Horowitz , R. M. (1984). Protection of children from family maltreat-
ment. In R. M. Horowitz & H . A. David son (Eds.), Legal rights of children (pp. 262-312).
Colorado Springs: Shepard's/McGraw-Hill.
David son, H . A., Horowitz, R. M., Marvell, T. B., & Ketcham, O. W. (1981). Childabuseand
neglect litigation: A manual for judges . Washington, DC: Department of Health and
Human Service s.
DeConey, J. J. (1975). The legal pro cess: A positive force in the interest of child ren . In
American Humane Association (Ed .), Fourth nationalsymposium on child abuse (pp . 62-
69). Den ver : Author.
Dickens, B. M. (1978). Legal resp on ses to child abuse. In J. M. Eekelaar & S. N . Kat z (Eds. ),
Fam ily violence: An international and interdisciplinary study (pp. 338- 362). Toronto:
Butt erworths.
Eatman, R., & Bulkley, J. (1986). Protecting child victim/witnesses: Sample laws and materials.
Washington, DC: American Bar Asso ciation .
Feshbach, N. D., & Feshbach, S. (1976). Punishmen t: Parent rites versus children's rights.
In G. P. Koocher (Ed.), Children's rights and the mental health professions (pp. 149-170).
New York: Wiley.
Finkelhor, D. (1984). Child sexual abuse: New theory and research. New York: Free Press.
Flicker, B. D. (1977). Standards for juvenile justice: A summaryand analysis. Cambridge , MA :
Ballinger.
Fontana , V. J., & Besharov, D. J. (1979). The maltreated child (4th ed .). Sp rin gfield , IL:
Ch arl es C Thomas .
Fraser, B. G. (1979). A glance at the pa st, a gaze at the pr esent, a glimps e of the future: A
critical analysis of th e deve lopment of child abus e rep orting statu tes . Chicago-Kent Law
Review, 54 , 641-686.
Garbarino, J. (1977). The pri ce of privacy in the social dynamics of child abuse . Child
Welfare, 56, 565-575.
Garbarino, J. (1982). Children and families in the social environment. New York: Aldine .
Garbarin o, J., Gaboury, M. T., Long, F., Grandjean, P., & Asp, E. (1982). Wh o own s the
children? An ecological perspective on public poli cy affecting children. In G . B.
Melt on (Ed .), Legal policies affecting child and youth services (p p. 43-63). New York:
Haworth.
Gelles, R. J. (1982). Probl em s in defining and labeling child abuse . In R. H . Starr (Ed .),
Child abuse prediction: Policy implications (p p. 1- 30). Cambri dge, MA : Ballin ger.
Giovannon i, J. M., & Becerra, R. M. (1979). Defining child abuse. Ne w York: Free Pre ss .
Globe Newspaper Co . v. Superior Court, 457 Ll.S. 596 (1982).
Goldstein, J., Freud, A. , & Solnit, A. J. (1973). Beyond the best interests of the child. New York:
Free Pre ss .
Goldste in, J., Freud, A. , & Solnit, A. J. (1979). Before the best interests of the child. New York:
Free Press .
Goodman, G. S. (1984). The child witness: Conclusion s and future direction s for resea rch
and legal practice. Journalof Social Issues, 40, 157-175.
Goodman , G. S., & Helgeson , V. S. (1985). Child sexual assault: Ch ildren's memory and
the law. In J. Bulkley (Ed. ), Papers from a nationalpolicy conferencein legal reforms in child
sexual abuse cases (p p. 41-60). Wash ington, DC: Am erican Bar Association , Na tional
Legal Resource Center for Child Advocacy an d Protection .
Graham, M. H . (1985). Child sex abuse persecutions: Hearsay and confront atio n clause
iss ues. In J. Bulkley (Ed .), Papers from a national policy conference on legal reforms in child
sexual abuse cases (pp. 159-208). Washington, DC: Am erican Bar Association .
TRENDS IN LEGISLATION AND CASE LAW 81

Haugaard, J. J., & Reppucci, N. D. (1988). Thesexual abuse of children. San Francisco: [ossey-
Bass.
Hou se of Representatives Report No. 685, 93d Cong ., reprinted in 1974 U.S. Code Congres-
sional & Administrative New s 2663.
Idaho v. Wright, _ _ U.S. _ _, 110 S. Ct. 883 (1990).
In re Freiberger, 153 Mich. App. 251, 395 N .W.2d 300 (Mich. Ct. App. 1986).
Institute of Judicial Administration/American Bar Association (1981). Standards relating to
abuse and neglect. Cambridge, MA: Ballinger.
Isaacson, L. B. (1975). Child abu se reporting statutes: The case for holding ph ysicians
civilly liable for failure to report. San Diego LAw Review, 12, 743-777 .
Kalichman, S. E., & Craig , M. E. (in pre ss). Victims of incestuous abuse: Mental health
professionals' attitudes and tendency to report. Victimology: An International Journal.
Kalichrnan , S. E., Craig, M. E., & Follingstad , D. R. (1987, August). Factors contributing to
mental health professionals' reporting of child abuse. Paper presented at the meeting of the
America n Psychologica l Association, New York.
Kelly, D. P. (1987). Victims . Wayne LAw Review, 37, 69-86.
Kempe, C. H., Silverma n, F. N ., Steele, B. F., Droegenmueller, w., & Silver, H. K. (1962).
The battered child syndrome. Journal of the American Medical Association, 181, 17-24.
Kentucky v. Stincer, 107 S. Ct. 2658 (1987).
Kilpatrick, D. G., & Otto, R. K. (1987). Constitutiona lly guaranteed participation in crimi-
nal proceedings for victims: Potential impact on psychological functioning. Wayne LAw
Review, 37, 7-28.
King, M. P., Hunter, W. M., & Runyan , D. K. (1988). Going to cour t: The experience of
child victims of intrafamilial sexual abuse. Journal of Health Politics, Policy, and LAw, 13,
705-721.
Lucht, C. L. (1975). Providing a legislative base for reporting child abu se . In American
Humane Association (Ed.), Fourth national symposium on child abuse (pp . 49-60) . Den-
ver: Author.
Maney, A., & Wells, S. (Eds.). (1988). Professional responsibilities in protecting children: A public
health approach to child sexual abuse. New York: Praeger.
Maryland v. Craig, u.s . ,100 S. Ct. 834 (1990).
Melton, G. B. (1982). Children's right s: Where are the children? American Journal of Ortho-
psychiatry, 52, 530-538 .
Melton, G. B. (1983a). Toward "personhood" for adolescents: Autonomy and privacy as
values in public policy. American Psychologist, 38, 99-103 .
Melton, G. B. (1983b). Decision making by children: Psychologica l risks and benefits. In G.
B. Melton, G. P. Koocher, & M. J. Saks (Eds.), Children's competence to consent (pp . 21-
40). New York: Plenum Press.
Melton, G. B. (1987a). Special legal prob lems in the protection of handicapped child ren
from parental maltreatment. In J. Garbarino, P. E. Brookhouser, & K. J. Authier (Eds .),
Special children-special risks: The maltreatment of children with disabilities (pp . 179- 193).
New York: Aldine de Gruyter.
Melton, G. B. (1987b). Children's testimony in cases of alleged sexua l abuse. In M.
Wolraich & D. K. Routh (Eds.), Advances in developmental and behavioral pediatrics (pp.
179-203). Green wich, CT: JAI Press.
Melton, G. B. (1987c). Law and random event s: The sta te of child mental heal th policy.
International Journal of LAw and Psychiatry, 10, 81-90.
Melton, G. B., & Corson, J. (1987). Psychological mal trea tment and the schools: Problems
of law and professional responsibility. School Psychology Review, 16, 188-194.
Melton, G. B., & Thompson, R. A. (1987). Legislative approaches to psychological mal-
treatment: A social policy anal ysis. In M. R. Brassard , R. Germain, & S. N. Hart (Eds .),
82 RANDY K. OTIO and GARY B. MELTON

Psychological maltreatment of children and youth (pp . 203-216). New York: Pergamon
Pre ss.
Melton, G. B., & Limber, S. (1990). Psychologists' involvement in case s of child maltreat-
ment: Limits of role and expertise. American Psychologist, 44, 1225-1233 .
Melton, G. B., Petrila, J., Poythress, N . G., & Slobogin, C. (1987). Psychologicalevaluations
for the courts: A handbook for mental health professionals and lawyers. New York: Guilford
Press.
Midonick, M. L. (1972). Children, parents and the courts: Juvenile delinquency, ungovernability
and neglect. New York: Practis ing Law Institute.
Missing Children Assistance Act 42 U.S.c. § 5771-5777 (1982, Supp. 1986).
Mnookin, R. H. (1973). Foster care : In wh ose best int ere st? Harvard Educational Review, 43,
599-638.
Muehleman , T., & Kimmons, C. (1981). Psych ologists' views on child abuse rep orting,
con fidentiality, life, and the law: An explorat ory study. Professional Psychology, 12, 631-
638.
Myer s, J. E. B. (1986). A survey of child abuse and neglect reporting statutes. Journal of
Juvenile Lnw, 10, 1-72.
National Center for State Courts. (1988). The changing face of child abuse registries.
National Center for State Courts Report, 15(7), 2.
National Committee for th e Prevention of Child Abuse (undated). Children's trust funds.
Chica go: Author.
Nelson , B. J. (1984). Making an issue of child abuse: Political agenda setting for social problems.
Ch icago: Univ ersity of Ch icago Pre ss.
Ne w York Social Services Law Sec. 413 (McKin ney, 1988).
Ohio v. Roberts, 448 U.S. 56 (1980).
Parental Kidnapping Prevention Act 28 U.S.C.S § 1730A (1979, Cum. Supp. 1988).
Paulsen, M. G. (1967). Ch ild abuse reporting laws: The sha pe of th e legislation . Columbia
Lnw Review, 67, 1-49.
Pierce v. Society of Sisters, 268 U.s. 510, 45 S.Ct. 571 (1925).
Prince v. Massachusetts, 321 U.S. 158 (1944).
Roe v. Wade, 410 u.s. 113 (1973).
Rogers, C. M. (1982). Child sexual abuse and the court: Preliminary find ings. In J. Conte &
D. A. Shore (Eds .), Social work and child sexual abuse (pp. 145-1 53). New York: Haworth
Pre ss.
Rosenberg , A. H. (1975). The law and child abuse . In N. B. Ebeling & D. A. Hill (Ed s.),
Child abuse: Intervention and treatment (pp . 161-170). Acton , MA: Publi shing Sciences
Group.
Rosenberg, M. S., & Hunt, R. D. (1984). Child maltreatment: Legal and mental health
issues. In N . D. Reppucci, L. A. Weithom, E. P. Mulve y, & J. Monahan (Eds .), Chil-
dren, mental health, and the law (p p. 79-101) . Beverly Hills: Sage .
Runyan, D., Everson, M., Edelsohn, G., Hunter, W., & Coulter, M. L. (1988). Impact of
legal intervention on sexually abused children. Journal of Pediatrics, 113, 647-653.
Santosky v. Kramer, 455 U.S. 745 (1982).
Senate Rep ort No . 123, 99th Congo reprinted in 1986 Ll.S, Code Congressional & Adm in-
istrative New s 1967.
Sherman, L. w., & Berk, R. A. (1984). The Minnesota domestic violence experiment. Wash-
ington, DC: Police Foundation.
Swoboda, J. S., Elwork, A., Sales, B. D., & Levine, D. (1979). Knowledge of and com-
pliance with privileged communication and child-abuse reporting laws. Professional
Psychology, 10, 448-457.
TRENDS IN LEGISLATION AND CASE LAW 83

Thibaut, J., & Walker, L. (1978). A theory of procedure. California Law Review, 66, 541-566.
Urzi, T. (1981). Cooperative approaches to child protection: A communityguide. Excerpted in H .
A. Davidson, R. M. Horowitz, T. B. Marvell , & O. W. Ketcham (Eds .), Childabuse and
neglect litigation: A manual for judges. Wash ington, DC: Department of Health and
Human Services .
United States v. Iron Shell, 633 F.2d. 77 (8th Cir. 1980).
Wald, M. (1975). State interventions on behalf of "neglected" children: A search for real-
istic standards. Stanford Law Review, 27, 985-1040 .
Wald, M. S. (1976). State intervention on behalf of "neglected" children: Standards for
removal from their homes, monitoring the status of children in foster care, and termi-
nation of parental rights. Stanford Law Review, 28, 625-706 .
Wald, M. S. (1982). State intervention in behalf of neglected children: A proposed legal
response. Child Abuse and Neglect, 6, 3-45.
Wald, M. S., Carlsmith, J. M., & Leiberman, P. H. (1988). Protecting abused and neglected
children. Stanford, CA: Stanford University Press.
Whitcomb, D. (1985). Assisting child victims in the courts: The practical side of legislative
reform. In J. Bulkley (Ed.), Papers from a national policy conference on legal reforms in child
sexual abuse cases (pp. 13-29). Washington, DC: American Bar Association, National
Legal Resource Center for Child Advocacy and Protection.
Wisconsin v, Yoder, 406 U.S. 205 (1972).
 CHAPTER 4

RESEARCH DIRECTIONS
RELATED TO CHILD
ABUSE AND NEGLECT
Roy C. HERRENKOHL

INTRODUCTION
Future directions for research on child abuse and neglect depend on
goals for the general area and the role to be played by research in
achieving these goals. The goals proposed here are to provide: (1) treat-
ment to abusive families to reduce or remove the likelihood of recur-
rence, and (2) services to abused children to ameliorate the conse-
quences of abuse and to develop prevention strategies to reduce its
incidence. The role of research in meeting these objectives is to obtain
information about the incidence and prevalence of abuse, to examine its
causes, and to evaluate the effectiveness of treatment and prevention.
Policymakers, the third party to the challenges posed by child
abuse, use results from research and the experience of service providers
to inform their decisions concerning the provision of resources for treat-
ment, prevention, and research. After two decades of research, the
nature and extent of the problem in a descriptive sense is clearer, but

Roy C. HERRENKOHL • Center for Social Research, Lehigh University, Bethlehem, Penn-
sylvania 18015.

85
86 ROY C. HERRENKOHL

key questions for policy (i.e., the causes of abuse and neglect, effective
treatment and prevention strategies) remain largely unclear. The inade-
quacy of relevant findings means that policymakers must develop and
implement plans based on informed judgment rather than more objec-
tive evidence.

RESEARCH ISSUES
Five research issues concerning child abuse are considered below:
incidence/prevalence, causes, consequences, treatment, and preven-
tion. These are interrelated in that what is learned about one often has
implications for one or more of the others. There also are important
distinctions, depending on which of the four types of abuse (physical,
emotional, sexual, or neglect) is involved. These often are mentioned
together but are seldom studied or analyzed concurrently. In practice
they sometimes occur singly or in combination (E. C. Herrenkohl & R.
C. Herrenkohl, 1981). This poses a problem of research strategy. Should
research focus on each type of maltreatment separately or on types in
combination? The answer may differ depending on the issue under
study.
The discussion that follows examines the current state of and possi-
ble future directions for research in the aforementioned five areas. Only
illustrations from the different types of abuse are offered. For a more
comprehensive review of specific projects, the reader is referred to
Finkelhor, Hotaling, and Yllo (1988).

INCIDENCE AND PREVALENCE OF ABUSE AND NEGLECT

Information on incidence and prevalence indicates the extent of the

problem in society. Accurate data on incidence and prevalence were
sought from the early days of research on child abuse . For example, Gil
(1970) examined the prevalence of physical abuse (cf. Nagi, 1977), while
a comprehensive prevalence study was more recently reported by
Straus, Gelles, and Steinmetz (1980). Russell (1983), on the other hand,
conducted a prevalence study of sexual abuse. The prevalence of com-
ponents of neglect have been studied (e.g., poor nutrition: Bassetti,
1974), but a thorough study of the prevalence of neglect has not been
conducted, although estimates have been offered (Nagi, 1977; Polansky,
Chalmers, Buttenweiser, & Williams, 1981).
The accuracy of incidence/prevalence data for different types of
abuse suffers from imprecise conceptualization and inadequate opera-
RESEARCH DIRECTIONS 87

tional definitions (Starr, 1988). For example, definitions of physical abuse

may specify "intent" to harm, but the occurrence of nonaccidental injury
often is taken as sufficient evidence of such intent (Starr, 1988). Emotional
abuse also has proved very difficult to define. Only the most extreme
cases of emotional cruelty are labeled as "abuse" (Hart, Germain, &
Brassard, 1987; Garbarino, Guttmann, & Seeley, 1986). Furthermore,
strategies for assessing the occurrence of sexual abuse are only beginning
to be developed (Conerly, 1986). Specific forms of neglect have been
studied (e.g., malnutrition, medical neglect), but the combination of
these conditions has not. The Polansky measure of neglect (Polansky et
al., 1981), the Childhood Level of Living Scale (CLL), reflects poverty as
well as neglect. For example, a negative response to the CLL Scale item,
"mother plans at least one meal consisting of two courses," may indicate
neglect or it may indicate poverty. A procedure is needed by which to
disentangle neglect from poverty. Such a procedure would distinguish
the neglectful quality of the parent's relationship to the child from child
care associated with parental level of education and income.
Adequate measures of abuse are the key to obtaining accurate inci-
dence and prevalence data. Such measures can, in tum, serve as the
dependent variables for studies of the causes of abuse and as the inde-
pendent variables in studies of the consequences of abuse. The inci-
dence and prevalence of abuse also are the criteria against which the
effectiveness of treatment and prevention strategies are judged.

RESEARCH DIRECTIONS

Generally, a thorough catalogue of manifestations of each type of

abuse, including instances that are borderline, is needed. Such a cata-
logue could provide the basis for developing an index, the reliability and
validity of which could then be assessed. Without information on psy-
chometric properties, the accuracy of statistics such as correlations is
suspect, and the validity of differentiations between abused and non-
abused children is questionable.
Six broad problems confront the development of better measures.
First, there is a need for better specification of the differences between:
(1) physical abuse and physical discipline, (2) emotional abuse and emo-
tionally oriented discipline, and (3) appropriate touching and sexually
abusive touching. In addition, a more comprehensive conception of ne-
glect would bring together the range of features known to comprise
neglect.
Second, research to develop more adequate operational measures
would follow from clearer definitions. For example, there are opera-
88 ROY C. HERRENKOHL

tional definitions of physical abuse that emphasize: (1) reported inci-

dents, (2) specific incidents recorded in case records (R. C. Herrenkohl,
E. C. Herrenkohl, Egolf, & Seech, 1979), (3) incidents reported as part of
discipline (R. C. Herrenkohl & E. C. Herrenkohl, 1988a), and (4) inci-
dents reported as part of family conflict (Straus et aI., 1980). Analyses are
needed of the psychometric properties of each approach and an assess-
ment of differences and similarities between strategies. The aim would
be to develop a more widely agreed on assessment strategy for each type
of abuse.
Third, it also is important to consider what represents an incident of
maltreatment. For example, Herrenkohl et al. (1979) report that if formal
citations for abuse are considered, the rate of recurrence was approx-
imately 25% for a two county area in Pennsylvania. For the same two
counties, if incidents recorded in case records are used the rate of recur-
rence is 55% to 66%. On the other hand, during the same period the
official state rate was 8% to 12%. Thus, if abusive incidents reported on a
discipline practices measure are used as the criterion, repeated abuse is
more frequent and more widespread than child welfare records would
indicate (Herrenkohl & Herrenkohl, 1988b).
Fourth, how are valid cases to be identified? Current methods de-
pend on mandated reporters, hot lines, and self-reports, among others.
More adequate screening procedures are needed, although their use
raises a variety of ethical issues. For example, more accurate information
is likely to depend to some degree on self-reports. Adults may be willing
to indicate the type of discipline they have used or other reactions they
have had to a child's behavior. If such reports are made to persons
required by law to make formal notification of abuse, the self-reports can
be incriminating (Geffner, Rosenbaum & Hughes, 1988; Finkelhor et al.,
1988). Children also can report abusive treatment they have received.
Children mature enough to offer such information are old enough to feel
responsibility for creating trouble for other family members. Further-
more, a dilemma is raised by identifying more cases while lacking the
resources needed to provide treatment. Some states (e.g ., Pennsylvania)
have narrowed their definition of abuse over the last decade, so as to
provide treatment only to more serious cases . Under such circumstances
efforts to identify new cases may not be paralleled by increased re-
sources to provide treatment for them.
A fifth issue concerns the prevalence of combinations of maltreat-
ment types. There is little information on how frequently different com-
binations of abuse occur. Herrenkohl et al. (1981) report that 33% of
children experience more than one type of abuse. Moreover, the occur-
RESEARCH DIRECTIONS 89

renee of and the relationship between the different types of abuse have
not been examined.
Finally, how is the information on incidence and prevalence to be
coordinated so that services can be provided to abusive families, pol-
icymakers can learn the extent of the problem, and the information can
be used in research? This problem points to the need for better coordina-
tion and cooperation among service providers, policymakers, and
researchers.

CAUSES OF ABUSE AND NEGLECT

Speculation concerning causes of abuse was raised early in the de-
velopment of research on maltreatment (Steele, 1976). Clearer identifica-
tion of causes would add to our basic understanding of parenting. It also
would give a focus to treatment and prevention because, in order to be
effective, treatment and prevention strategies must reduce or remove
factors that are assumed to play a causative role in child abuse and
neglect.
There have been many causal models of maltreatment (Parke &
Collmer, 1975). Some propose that abusive parents have a personality
disorder or are mentally ill (McCleer, 1988; Spinetta & Rigler, 1972).
Others suggest that abusive parenting styles are learned (Parke & Col-
lmer, 1975; Wolfe, 1987), or that there is an intergenerational transmis-
sion of abuse (E. C. Herrenkohl et al., 1983; Egeland, [acobvitz, & Pa-
patola, 1987). Egeland and Sroufe (1981) examined disruptions in the
attachment of parent and child as a possible causative explanation. Oth-
ers see abusive parents as having inappropriate expectations of their
children (Twentyman, Rohrbeck, & Amish, 1984). Still others suggest
that abuse is a result of frustration because of stress that leads to exces-
sively aggressive parenting (Elmer, 1979). Some explanations implicate
the child in the etiology of abuse; for example, the difficult (handicapped
or premature) child has been noted as an elicitor of abuse (de Lissovoy,
1979). Others (Kadushin & Martin, 1981; R. C. Herrenkohl et al., 1983)
have suggested that the child's normal behavior can elicit abusive reac-
tions. Societal norms and attitudes that tolerate or condone domestic
violence can further influence the occurrence of maltreatment (Straus et
al., 1980).
Current explanations for the occurrence of abuse can broadly be
characterized as a set of conceptually unrelated propositions or hypoth-
eses; and although most are supported by some empirical data, the
90 ROY C. HERRENKOHL

consistency and utility of such evidence varies considerably. An early

example of such coordination and review was by Spinetta and Rigler
(1972) who found many inconsistencies in evidence concerning the hy-
pothesis that personality abnormalities were the cause of physical
abuse. A more recent example is the work of Kaufman and Zigler (1987)
who examined evidence concerning the intergenerational transmission
hypothesis-that abused children will themselves become abusive par-
ents. These authors report that the combined evidence does not lend
extensive support to this hypothesis.
Increased efforts are needed to examine etiologic formulations of
abuse. These should go beyond simply cataloguing evidence and seek to
determine the validity of an hypothesis, taking into consideration varia-
tions in the formulation of the hypothesis, methodological strengths and
weaknesses of studies which test the hypothesis and variations of the
hypothesis that may not have been examined. For example, the proposal
that abusive parents have unrealistic expectations for their children's
behavior was originally tested by several researchers who questioned
parents about developmental norms (e.g., the age at which a child will
walk or talk). Differences between abusive and non-abusive parents
were not found . There does, however, seem to be validity to the hypoth-
esis, if parental expectations are conceptualized as tolerance for chil-
dren's misbehavior or what parents consider misbehavior. Work by Ka-
dushin and Martin (1981), Twentyman et al. (1984), and R. C. Herrenkohl
et al. (1983), each from somewhat different perspectives, lend support to
the view that abusive parents have inappropriate expectations of their
children's behavior as compared to nonabusive parents.
Research is also needed to develop and/or revise current theories or
models about the causes of abuse. A valid theory about the occurrence
of abuse would involve specification of relevant constructs and their
interrelationships. Such a theory also could indicate the constructs or
relationships between constructs that must be modified so that recur-
rence could be avoided or, if modified before the initial occurrence of
abuse, so that abuse could be prevented. For example, if simply the
overall level of stress on a family were considered the cause of abuse,
then casework services to reduce the stress would be the appropriate
strategy. If, however, the more specific stresses of child rearing and
social isolation were theorized to be responsible for abuse, then services
directed at reducing those particular stresses (i.e., the provision of child
care services, training in child rearing skills, and participation in a par-
ent support group) would be expected to reduce the stresses and pre-
vent occurrence/recurrence of maltreatment.
The role of social class in general and poverty in particular in the
RESEARCH DIRECTIONS 91

development of abuse has been addressed infrequently (Pelton, 1981).

Those who have examined this topic have generally found a relationship
between social status and the occurrence of abuse (Elmer, 1981; Straus et
al., 1980; Wolfe, 1987). However, the causal implications of this rela-
tionship have received little attention.
Many believe that each type of abuse is multidetermined, that is, no
single factor will expla in its occurrence. Thus, there are several multifac-
tor models. For example, Gelles (1973) described a model in which a
network of influences, such as stress, social isolation, parents' child-
rearing experience, among other factors was hypothesized to contribute
to abuse. On the other hand, Garbarino (1976, 1981) proposed an "eco -
logical model" in which situational factors are given more attention.
Belsky (1984) suggested that abusive parenting is considered an extreme
on a continuum of general parenting, the quality of which is influenced
by a set of factors within the family and community. Wolfe (1987), in
contrast, advocated a transitional model in which parent-child interac-
tions change over time into aversive interactions. It should be noted,
however, that the complexity of multiconstruct models may require sin-
gle-construct empirical tests until more is known about each construct.
Existing models are inadequate for several reasons. One is that
constructs in the models are not well-defined. Such constructs as "par-
enting" are not sufficiently explicit to indicate which aspects of inade-
quate parenting have the more detrimental effect on the child and how
different aspects of parenting are interrelated. Because of the inexplicit-
ness of these models, development of operational definitions is difficult.
Another difficulty is that there is little differentiation between the causes
of one type of abuse and the causes of another and suggested causes of
one type of abuse may overlap the proposed causes of a second type.
An additional strategy in constructing more comprehensive models
of abuse involves identifying risk factors that are indicators of the poten-
tial for abuse. Finkelhor et al. (1988) considers this topic one of the better
defined research areas. One attraction of this approach is that risk fac-
tors, such as being a single-parent family, the child's being handicapped,
or having long or multiple separations from parents, are somewhat sim-
pler concepts than stress or social isolation. Consequently, such factors
can be identified more readily. Several researchers (Starr, 1982) have
examined risk factors (Milner & Wimberlay, 1979; Schneider 1982). Prob-
lems beset this approach, however. First, accurate identification of risk
factors depends on knowing the "universe" of abusive families. How-
ever, the universe of abusive families is unclear. The number of cases
continues to increase, and it is likely that even those cases that have
been identified are not representative of all cases of abuse. Second,
92 ROY C. HERRENKOHL

without precise information on the characteristics of all abusive families,

identification of risk factors will be based on the characteristics of fami-
lies known to be abusive. Any biases in those characteristics will be
reflected in the identified risk factors . On the one hand, if among identi-
fied abusive families there is a disproportionately large number of pover-
ty level families as compared to all abusive families, living in poverty
will be identified as a significant risk factor. On the other hand, examina-
tion of the characteristics of identified abusive families may fail to pin-
point a characteristic which, in the actual universe, is more salient. A
further problem is specifying the level of a factor that constitutes risk .
For some possible risk factors, such as prematurity, there are "rules of
thumb" about what represents risk. For others, however, such as the
level of harsh treatment by a parent of a child, there is no consensus as
to what represents risk . The tendency for risk models to predict larger
numbers of high risk families than are observed to become abusive is a
negative consequence of such sources of bias .

RESEARCH DIRECTIONS

Several steps are needed concerning research on causes. First, prob-

lems of conceptualization and measurement of causal constructs are
serious. Causal effects cannot be identified unless such constructs are
measured reliably and with validity. To develop these measures requires
considerable time and skill. Second, there is a need to review and coor-
dinate existing information concerning each causative factor in an effort
to narrow the number of possibilities. Those causal explanations that
offer the most promise will merit testing in greater depth. Since it is
likely that abuse is multidetermined, the smaller the set of likely causal
constructs, the more readily they can be tested. Third, given the as-
sumption that variables contributing to abuse are multidetermined, a
model or theory is needed that specifies how this multidetermined state
operates. This is best developed when more in-depth information on
specific causes is available . Care needs to be taken to consider the fact
that many abusive families display numerous additional problems, and
that a significant number have experienced more than one type of
abuse.

CONSEQUENCES OF ABUSE AND NEGLECT

Research on the consequences of abuse can be divided roughly into
two parts. One concerns the immediate, often medical, sequelae and the
RESEARCH DIRECTIONS 93

physical trauma of the abuse. The other concerns the longer term, gener-
ally psychosocial, consequences of maltreatment.
The medical effects of physical abuse and neglect have been the
focus of many studies reported primarily in medical journals. These
studies should be catalogued and evaluated for use by those who must
determine the etiology of childhood injuries. Such a cataloguing might
also provide the basis for indexing injuries with reference to severity
which, in turn, could be related to potential for longer term conse-
quences.
Several researchers (e.g., Gray & Kempe, 1976) have described
physically abused children's psychological state soon after they were
identified as having been abused. Such information can be used to index
the degree of psychological trauma manifested by the child at the time
the abuse was identified. For example, the short-term consequences of
sexual abuse have been described (Wolfe, Wolfe, & Best, 1988) in largely
anecdotal terms. Similarly, neglected children's physical and psycholog-
ical status have been reported by several authors (Polansky et al., 1981;
Oates, 1986). Such reports should be coordinated and evaluated. In the
case of emotional abuse, however, the short-term consequences have
not been differentiated from longer term effects .
The longer term consequences of abuse have been studied by a
small group of researchers (e.g ., Cicchetti & Rizley, 1981; Egeland &
Sroufe, 1981; Elmer, 1977; E. C. Herrenkohl & R. C. Herrenkohl, 1981;
Lynch & Roberts, 1982; Martin, 1976; Oates, 1986). To varying degrees,
their studies have considered physical abuse, emotional abuse, and ne-
glect. On the other hand, studies of the longer term consequences of
sexual abuse have only recently been conducted (Wolfe et al., 1988).
Most of this research is longitudinal in nature, although it differs consid-
erably in terms of the time period covered. The reader is referred to
more comprehensive reviews of these investigations by Augoustinos
(1987) and Toro (1982).

PHYSICAL ABUSE

Much of the research on consequences of maltreatment has focused

on physical abuse. For the roughly 25% (E. C. Herrenkohl & R. C.
Herrenkohl, 1981) of abused children who have severe physical injury,
the physical damage may result in long-term physical handicaps that, in
turn, can lead to problems of social-emotional adjustment (Elmer, 1977;
Lynch & Roberts, 1982). For the remaining 75%, the physical injury may
be less severe, (i.e., bruises and abrasions), but there still may be longer
term developmental difficulties (Martin, 1976). For example, severity of
94 ROY C. HERRENKOHL

physical abuse can be manifested in the form of injuries that result in

lifetime handicaps, injuries that are severe but heal leaving no long-term
handicap, and injuries that are not severe. But emotional abuse leaves
no observable marks and also can differ in degree. Parents who threaten
a young child with abandonment may inflict more serious consequences
than parents who tell the same thing to an older child who can judge the
likelihood of that happening. Furthermore, less severe but repeated
physical or emotional abuse may inflict more physical and/or emotional
damage to the child than a single abusive incident.

EMOTIONAL ABUSE

The longer term consequences of emotional abuse are less docu-

mented because of difficulties in identifying and measuring emotional
abuse. Furthermore, emotional abuse occurs as a separate form of abuse
[e.g., threats to abandon or kill a child, or telling a child repeatedly he or
she is "stupid" and "worthless" (Navarre, 1987)] or in conjunction with
other forms of maltreatment. For example, the effects of emotional abuse
are manifested in the helplessness and worthlessness often experienced
by physically abused children (Hyman, 1987), the sense of violation and
shame found in sexually abused children (Brassard & McNeil, 1987), or
the lack of environmental stimulation and support for normal develop-
ment found by neglected children (Schakel, 1987). The consequences of
these different types of emotional abuse, however, have not been em-
pirically examined.

SEXUAL ABUSE

The longer term consequences of sexual abuse are believed to be

primarily in the areas of trust and heterosexual relationships (Wolfe et
al., 1988). Research on the impact of sexual abuse is currently more in the
form of case studies than of empirically controlled studies (Lusk & Wa-
terman, 1986; Tong & Oates, 1987). Consequently, it is difficult to sepa-
rate the impact of the abuse from the impact of other factors at work in
the sexually abusive family context.

CHILD NEGLECT

Research directly on the consequences of child neglect is sparse .

Polansky (see Polansky et al., 1981)has studied the neglectful family but
has focused more on the characteristics of neglectful parents than on the
RESEARCH DIRECTIONS 95

longer term consequences of neglect in children. Polansky does provide

evidence, however, that the cognitive development of neglected chil-
dren is more retarded than is that of nonneglected children. Oates (1986)
examined the longer term consequences of neglect, including an exam-
ination of nonorganic failure-to-thrive children. Results from his study
found failure-to-thrive children to be significantly lower in social matu-
rity, language development, and verbal ability, and to have more per-
sonality abnormalities than a comparison group. Research on areas that
comprise child neglect, such as inadequate nutrition (Martin, 1973) and
social-emotional understimulation (Schakel, 1987), clearly indicated the
negative impact of neglectful parenting on a child's development. These
latter studies demonstrate the effects of a particular inadequacy (e.g.,
malnutrition) on a child, but have not been coordinated in such a man-
ner as to elucidate their combined impact on a neglected child's develop-
ment.

RESEARCH DIRECTIONS

Several factors render examination of the consequences of abuse a

complex and difficult task. First, the reliable and valid measurement of
each type of maltreatment is a major problem here as well. Furthermore,
improved conceptualization and measurement of consequences are
needed. Measures such as that developed by Achenbach (Achenbach &
McConaughy, 1987) have proven useful. This instrument assesses cog-
nitive, educational, emotional, social, and physical development, has
strong psychometric properties, and has been used with a variety of
child and adolescent populations that can provide helpful comparisons.
Second, more adequate formulations of how each type of abuse
exerts its effect on the child are needed. This is important for knowing
how and when to direct treatment for the consequences of abuse. For
example, the act of discipline that injures the child is embedded in an
ongoing adult-child relationship. It is possible that the damage to the
child's development is determined more by the negative quality of the
ongoing interaction than by the abusive act per se. Although a number of
studies examine the parent-child interaction of abusive and nonabusive
families (e.g ., Burgess & Conger, 1978; E. C. Herrenkohl, R. C. Her-
renkohl, Toedter, & Yanushefski, 1984), none investigate the impact of
the combination of discipline and on-going interactions on the child's
psychosocial development.
Third, if the various types of abuse have different types of sequelae,
it will be necessary to develop hypotheses about consequences for each
96 ROY C. HERRENKOHL

form of abuse. Further, because some children experience more than one
type of abuse, a model or models of how the four types interrelate in
various combinations to effect the child are needed.
One strategy to examine how abuse influences psychosocial func-
tioning is to compare abused children who do well developmentally
with those who do not. For example, Garmezy (1983) has used this
approach to determine why some children who are developmentally at
risk do well and others, similarly at risk, do not. It may prove helpful in
identifying features in an abusive family or characteristics of an abused
child that serve to buffer the child against the impact of abuse (Mrazek &
Mrazek, 1987). For example, the abused child who receives nurturance
from a parent or parent surrogate or who has sufficient insight to per-
ceive the abusive behavior as due to the perpetrator's problems may not
succumb to the most damaging consequences of abuse.
Finally, abuse is related to such factors as poverty, marital discord,
and social isolation, and each of these may have an effect on the child's
development similar to those stemming from abuse. How to differenti-
ate the contribution of abuse from the effects of other factors with which
abuse covaries is an additional research problem that is particularly
problematic, because factors, such as poverty, may have a more far-
reaching effect than, say, physical abuse, especially, if the physical inju -
ries are endured in the less-severe range.

TREATMENT OF ABUSE AND NEGLECT

The treatment services available for abusive and neglectful families
are varied. Reports of treatment approaches generally consider one type
of abuse. For example, Starr (1988) summarized treatment forms for
physical abuse; Garbarino et al. (1986) considered intervention strategies
applicable to emotional abuse; Wolfe et al. (1988) described treatment
approaches to sexual abuse; and Polansky et al. (1981) have examined
treatment strategies for neglectful families . Sudia (1981) has taken a
more comprehensive approach. Noting the multiple problems of many
abusive families, she suggests that there is a need to classify families
both in terms of the types of problems requiring treatment and in terms
of the anticipated duration of interventions before actually providing
services.
Another perspective is to consider the recipient of treatment ser-
vices. The goal of treatment of a perpetrator of abuse or neglect is to
prevent recurrence. Frequency of recurrence for perpetrators can range
from no recurrence after a first incident to many incidents after a first
RESEARCH DIRECTIONS 97

incident (R. C. Herrenkohl et al., 1979). There is little evidence, however,

regarding how effective treatment is when provided to perpetrators.
Evidence that does exist suggests that individualized services for cooper-
ative female perpetrators are effective, although there is relatively little
evidence pertaining to male perpetrators (Ammerman, 1990). Further-
more, there is little evidence to indicate how frequently and under what
circumstances perpetrators cooperate with treatment (jones, 1987). It is
not known how effective different treatment strategies are with per-
petrators of different rates of recurrence. It also is unclear whether treat-
ment strategies that are effective with one type of abuse (e.g ., physical),
also are effective with a second type (e.g., neglect). There are indications
that different types of abuse are based on different dynamics (E. C.
Herrenkohl et al., 1983), which may mean that different treatment strat-
egies are needed.
Treatment can also be implemented with family members as a
group. The family provides a context for the abuse and its members are
affected by its occurrence. There is speculation of how a family provides
a context for abuse (e.g ., Justice & Justice, 1976). However, the way in
which the family system contributes to abuse has not been empirically
examined.
A third focus of treatment is to reduce or prevent the consequences
of abuse for the abused child . Although it is not altogether clear what
the consequences are and how they occur, a specific direction for treat-
ment is also unclear. Furthermore, it is unclear whether the child who
observes violence, but is not a victim, is as much in need of treatment as
the child who is directly maltreated. There is some research on the
short-term effects of viewing violence toward a parent (e.g ., Hershorn &
Rosenbaum, 1985; Hotaling & Sugarman, 1986). Geffner et al. (1988),
who discuss research needs related to this topic, concluded that "very
few conclusions can be drawn regarding child witnesses of parental
violence" and that "investigators .. . are probably decades away from
being able to specify what type of treatment, under what circumstances,
for which types of clients, is most effective" (p. 475-476).
A variety of treatment services are provided to maltreated children
and their families (e.g., casework, parent education, individual thera-
py). Each form of service has an implicit conceptualization of how to
prevent recurrence. It assumes a cause or cluster of causes to which the
service is addressed, using the assumption that reducing or removing
the presumed cause will prevent recurrence. For example, casework to
assist families to reduce stresses assumes that stress is a cause of abuse.
Family therapy to improve relations between family members assumes
that conflicts or distorted relationships between family members con-
98 ROY C. HERRENKOHL

tribute to abuse. Counseling to assist an individual in coping more ade-

quately with psychological conflict related to child rearing assumes that
such conflict is the cause. A support group to help reduce social isolation
or parenting programs to improve parenting skills assumes that these
are the causes. The questions that follow from such assumptions are first
whether a particular form of treatment can be effective in reducing or
removing a presumed cause and, second, whether having removed or
reduced the presumed cause recurrence is reduced or stopped.
If the different types of abuse have different etiologies, then treat-
ment strategies that address each of the causes are needed in families
involved with more than one type of abuse. The implication of this
possibility is that complicated (and costly) treatment strategies may be
needed for some families.
When treatment concerns consequences to the child, it is essential
that we understand the effects of each type of abuse so that treatment
can be directed to the areas of need. For example, the physically abused
child, in addition to needing treatment for physical injury, likewise may
feel vulnerable and worthless. Group therapy may help such problems,
as it may for the emotionally abused or sexually abused child . Treatment
for the neglected child may involve medical care, education remediation
and other services known to counter the effects of cognitive, social and
emotional deprivation. However, assessment of the effectiveness of
strategies to reduce the impact of maltreatment of children has not been
undertaken.
In some instances, recurrence is prevented because the perpetra-
tor(s) has no access to the child . The perpetrator may voluntarily leave
the home, or in those instances of abuse judged to be most severe,
children are placed in foster care. This results in the added trauma for
the child of separation from family (Goldstein, Freud, & Solnit, 1973).
Foster placement introduces the child to a system that has both advan-
tages and disadvantages (Wald, 1976). Although foster placement is
widely used, the circumstances under which it is most and least bene-
ficial to the child have not been fully examined (Wald, Carlsmith, Leider-
man, French, & Smith, 1985). It also is unclear how foster care may
interact with the psychosocial consequences of the abuse, or how cir-
cumstances in the foster care system, such as moves from family to
family, may influence the child's development.

RESEARCH DIRECTIONS

During the past two decades a number of service/treatment projects

have been undertaken to demonstrate the efficacy of the one or a com-
RESEARCH DIRECTIONS 99

bination of treatment services designed to protect against recurrence.

Some of these demonstrations have included evaluations to determine
the effectiveness of the service/treatment program (Cohn, 1979a,b;
Cohn & Daro, 1987). Some evaluations produced useful results, whereas
others did not. Reasons for the lack of useful results were not always
evident, although there are some indications. First, the difficulties and
complexities of treating abusive individuals and their families have not
always been acknowledged. This results in rather unrealistic goals for
experimental programs. Second, coordination between service delivery
and evaluation staffs in planning and conducting the evaluation was
inadequate. There often has been too little joint planning by service
providers, researchers, and policymakers with service providers some-
times feeling that they have little to say about the evaluation. As a result,
service providers may distrust evaluation specialists and resist the eval-
uation process, hindering completion of crucial portions of evaluation
activities.
Third, clear documentation of the treatment strategy was not devel-
oped. This would indicate precisely the treatment components used,
how and in what amounts the treatment was provided, and the staff
skills needed to carry out the program. Without such information, rep-
lication of treatment procedures is not possible.
Fourth, more comprehensive models are needed to indicate how (or
why) the phenomenon to be treated (i.e., recurrence, occurrence, or
consequences of abuse) developed. An indication is also required as to
why occurrence/recurrence can be expected to respond to the treatment
program. Such models make it possible to focus treatment in ways that
optimize the potential for obtaining the desired results and, where out-
comes suggest the need for adjustment, to alter the treatment strategy in
a positive way.
Fifth, as noted above, evaluation methods often were not suffi-
ciently rigorous. Time pressures led to inadequate measurement pro-
cedures in that reliability and validity characteristics were unknown or
determined only after data collection. Research designs often involved
serious threats to validity (Cook & Campbell, 1979). Consequently,
when results failed to support the effectiveness of a treatment strategy
or program, it was unclear whether the reason was the ineffectiveness of
the treatment or the inadequacy of the evaluation procedures.
Sixth, cumbersome evaluation procedures were sometimes used
placing sizable demands on overworked service delivery staff. As a re-
sult, service staff, who were already skeptical of the evaluation under-
taking, carried a major portion of the evaluation effort.
Finally, a more comprehensive plan for evaluating treatment effec-
100 ROY C. HERRENKOHL

tiveness is needed. In the past, there has been a tendency to depend on

one or two large demonstration programs when several smaller projects
might have proved more productive. Planning should take into consid-
eration the potential for replicating the treatment procedures and the
evaluation since no single evaluation is likely to provide definitive
answers.

PREVENTION OF ABUSE AND NEGLECT

Prevention refers to strategies intended to inhibit or deter the occur-
rence of something undesirable. There have been various suggestions
for preventing abuse (Wolfe, 1987). In a formal sense, there are three
types of abuse prevention (Goldston, 1977). One type aims at a general
population in an effort to reduce or prevent the occurrence of abuse. For
example, proposals have been made that the secondary educational sys-
tem be utilized to teach high school students parenting skills. To imple-
ment such a program would require a clear conception of how such a
parenting program would work and how it would exert its influence in
such a way that would prevent the occurrence of abuse. A second type is
prevention aimed at families or children where there is a likelihood of
abuse in an effort to prevent its occurrence. An example here is preven-
tion that focuses on identifying children most at risk for abuse and then
developing strategies to prevent the risk from actualizing into occur-
rences of abuse (Wolfe, 1987). A third approach, discussed in the pre-
vious section, can be considered prevention in that it aims to prevent
recurrence of maltreatment.

RESEARCH DIRECTIONS

The difficulties faced by efforts to prevent abuse derive directly from

the problems that have been described. For prevention to have a poten-
tial for success, the process by which abuse occurs must be understood
well enough to suggest how an effective prevention intervention would
be undertaken (Alvy, 1975). This means that from among the wide vari-
ety of causes of abuse a smaller number should be apparent as prime
candidates for preventive intervention.
The intervention strategy should be spelled out so that its goals are
defined and the procedures for achieving those goals are clear. Evalua-
tion then has a dual task. First, to determine if the goal of altering the
behavior or circumstances presumed to lead to abuse was actually
achieved. For example, did individuals really improve their parenting
RESEARCH DIRECTIONS 101

skills? Second, if there is evidence that the presumed cause was altered,
was there evidence that the incidence of abuse was reduced among
those who were the target population of the intervention (Rosenberg &
Reppucci, 1985)?
Evaluation research has developed rapidly although it has had a
relatively small impact on the child abuse/neglect area . The lack of a
community of child abuse/neglect evaluators who evaluate child abuse
preventive intervention programs and counsel each other on their eval-
uation activities hinders development of more adequate evaluations of
prevention programs. Communities of researchers exist around the
other issues considered in this discussion and these groups have played
an important role recently. So few evaluations of preventive interven-
tions have been made that each starts largely on its own and tends to
operate independently. Campbell (1987), considering evaluations of pre-
ventive interventions in the area of mental health, notes that one well-
trained scientist or team producing one research report does not result
in understanding the effective uses of an intervention. Rather, progress
toward such understanding is achieved by a scientific community that
stays in close communication on a shared puzzle and that promotes
competitive replication and criticism of each other's evaluative activities.

RESEARCH RESULTS AND POLICY CONSIDERATIONS

Thus far, our focus has been on five broad research issues that
pertain to child abuse and neglect. Specific questions remain to which
answers are needed by policymakers.

INTERGENERATIONAL TRANSMISSION

Is the potential for being abusive passed from generation to genera-

tion? Current evidence suggests that some parents who are abused as
children become abusive as parents, whereas others do not. What fac-
tors enter into an abused child's becoming an abusive parent? Should
having been abused as a child be considered, for example, in custody
decisions in which judgments about a parent's capacity to care for a child
must be made?

ANTISOCIAL BERAVIOR

Does abuse lead to antisocial or violent behavior? Evidence related

to this question is retrospective and suggests that there is a link, but it
102 ROY C. HERRENKOHL

appears to be conditional (Lane & Davis, 1987). What conditions lead to

an abused child's becoming antisocial or violent? Should having been
abused as a child be considered as a mitigating circumstance in judg-
ments about penalties for antisocial or violent behavior in juveniles or
adults?

ADOPTION

For some instances of abuse it is necessary to terminate parental

rights. What considerations should be taken into account when deciding
to terminate parental rights? How much should potential adoptive par-
ents know about the abuse experienced by a child whom they are seek-
ing to adopt? What help is needed by such children, even after adop-
tion, to resolve the consequences of abuse?

FAMILY DISSOLUTION

Particularly with revelations of sexual abuse, a family may be dis-

solved. What is the effect of such a rupture on the child? Are there
considerations that suggest whether a child is better off with the per-
petrator leaving the family or remaining in the family and receiving
treatment?
Answers to such questions are of interest to society in general. They
are of particular concern to policymakers who must set guidelines relat-
ed to them. Results from research on abuse generally provide equivocal
answers and, as a result, lack utility for policymakers. One reason some-
times cited is that research, particularly longitudinal studies, is too time-
consuming to obtain results; even though this response begs the ques-
tion. The results, even after the prescribed wait, may not be sufficiently
definitive to be of use. Although the time needed to obtain results is a
problem, inadequacies of results are the result of more fundamental
issues.

Unrealistic Expectations

There is intense pressure on service providers and policymakers to

do something about the apparent rise in the incidence of abuse. The
pressure on policymakers, as evidenced by the need for short-term re-
sults that answer policy questions, has played into some researchers'
tendency to oversell potential results in an effort to obtain funding.
RESEARCH DIRECTIONS 103

Fragmentation of the Research Process

Research on child abuse is conducted in a number of different disci-
plines and subdisciplines, many of which have distinctive perspectives
and research methods. Such diversity is important for progress but also
creates a degree of fragmentation that requires concerted effort to tran-
scend. Without continuing coordination in the form of discussion and
information exchange between disciplines, service providers, policy-
makers, and researchers, the context for obtaining the needed informa-
tion will not exist, nor will the questions to which the research is ad-
dressed be those that are relevant to service and policy concerns.

Inadequate Theories
As has been suggested above, our theories and models are inade-
quate to the demand. The pressure for quick results has tended to steer
researchers away from working on conceptual issues and model build-
ing that are fundamental to obtaining more definitive results.

Measurement Procedures
A serious need exists for more adequate measurement procedures.
For example, conceptualization of the different types of abuse suggests a
continuous variable ranging from mild to severe to abusive discipline.
Measures most often reflect categorical indicators that are of question-
able reliability and validity. In addition, continuous operational defini-
tions are currently being developed.

Adequacy of Designs
Many of the general questions asked regarding research on abuse
explicitly or implicitly concern causality. It is generally accepted that
unequivocal evidence about causality cannot be obtained (Cook &
Campbell, 1979), given the limitations of applied research. The primary
aim of research, then, is to come as close as possible to achieve this . In
improving research designs the objective is to reduce the number of
alternative explanations or, as Cook and Campbell (1979) label them,
"plausible rival hypotheses." Much is known about improving quasiex-
perimental designs that has yet to be applied in this area of research. For
example, quasiexperimental designs are subject to selection bias; that is,
a nonequivalence of experimental and control or comparison groups on
some dimension in addition to the experimental influences under study.
104 ROY C. HERRENKOHL

Consequently, an observed effect could be due to the experimental influ-

ences or to the additional dimension. In research on abuse, socio-
economic status (SES) can be such a source of selection bias .

Sample Selection
Research to date has tended to use experimental group samples
based upon convenience (e.g., hospital patients, clinic patients) and to
select control or comparison groups in much the same way. Such sam-
ples are important for developing new research areas. As the research
becomes more precise, however, issues of generalizability of results in-
crease in importance.

Statistical Procedures
The statistical procedures used in many existing studies are some-
times inadequate to the questions being addressed. Again, this is to be
expected in the early stages of developing a research area. However, the
models that depict the causes of abuse are likely to be complex, and
statistical procedures are currently being tested that are more appropri-
ate to the statistical demands these models create. To date, few studies
have used multivariate statistical techniques, even though these are like-
ly to be used with greater frequency in the future. The models proposed
to explain the occurrence of abuse or the consequences of abuse are
multidetermined and involve a variety of constructs (e.g., Belsky, 1984;
Gelles , 1973; Wolfe, 1987). Realistic tests of these models can only be
accomplished by using such multivariate procedures as multiple regres-
sion, multivariate analysis of variance, factor analysis, or structural
equation (LISREL) modeling.

SUMMARY
The preceding discussion has considered current status and future
research directions for five issues in the area of child abuse and neglect:
incidence/prevalence, causes, consequences, treatment, and preven-
tion. Each issue has a number of problems that require increased atten-
tion by the research and clinical community. In particular, major prob-
lems to be addressed include the need for more explicit theories and
hypotheses, improved measurements and design, and more adequate
sampling and statistical procedures. A renewed focus on improved re-
search approaches will significantly enhance our understanding of the
development, impact, and treatment of child abuse and neglect.
RESEARCH DIRECTIONS 105

A CKNOWLEDGMENTS

Preparation of this chapter was supported in part by Grant NO.

MH41109 from the National Institute of Mental Health, Program on
Anti-Social and Violent Behavior.

REFEREN CES
Achenbach, T. M., & McConaughy, S. H . (1987). Empirically based assessment of child and
adolescent psychopathology. Newbury Park, CA: Sage .
Alvy, K. T. (1975). Preventing child abu se . American Psychologist, 30, 921-928 .
Ammerman, R. T. (1990). Child abuse and neg lect. In M. Hersen (Ed.), Innovations in child
behavior therapy. New York: Springer.
Augoustinos, M. (1987). Deve lopmental effects of child abuse: Recent find ing s. ChildAbuse
and Neglect, 11, 15-27.
Bassetti, M. (1974). N utrition . In K . E. Barnard & H . B. Douglas (Eds.), Childhealth assess-
ment (pp. 59-73). (Part I, USDHEW, Publication No. HRA 75-30) . Bethesda, MD :
Department of Health, Education, and Welfare.
Belsky, J. (1984). Determinants of parenting: A proce ss model. Child Development, 55, 83-
96.
Brassard, M. R., & McNeil , L. E. (1987). Child sexual abuse. In H. R. Brassard, R. Germain,
& S. N . Hart (Eds .), Psychological maltreatment of children and youth (pp . 69-88). New
York: Pergamon Pres s.
Burgess, R. L., & Conger, R. D. (1978). Family in teraction in abu sive, neg lected and
normal families. Child Development, 49, 1163-11 73.
Campbell, D. T. (1987). Guidelines for monitoring the scien tific competence of preventive,
intervention research centers: An exercise in the sociology of scien tific validity. Knowl-
edge, Creation, Diffusion, Utilization, 8, 389- 430.
Cicchetti, D., & Rizley, R. (1981). Developmental perspectives on the etiology, intergenera -
tional transmission and sequelae of child maltreatment. In R. Rizley & D. Cicchetti
(Eds.), Developmental perspectives on child maltreatment (pp . 31-55). San Francisco:
je ssey-Bass.
Cohn, A. H . (1979a). An evaluation of thr ee demonstration child abuse and negl ect treat-
ment programs. Journal of the American Academy of Child Psychiatry, 18, 283-291.
Cohn, A. H. (1979b). Essential components of successful child abu se and neglect treat-
ment. Child Abuse and Neglect, 3, 491-496.
Cohn, A. H., & Dare , D. (1987). Is treatment too late: What ten years of evaluation
research tells us . Child Abuse and Neglect, 11, 433-442 .
Con erly, S. (1986). Asse ssment of suspected child sexua l abuse . In K. Macfarlane, J.
Waterma n, S. Conerly, L. Dam on , M. Durfee, & S. Long (Eds.), Sexualabuse of young
children (pp. 30-51). New York: Guilford Press .
Cook , T. D., & Campbell, D. T. (1979). Quasi-experimentation: Design and analysis issues for
field settings. Boston : Houghton Mifflin.
de Lissovoy, V. (1979). Toward the definition of "abuse provoking child ." Child Abuse and
Neglect, 3, 341-350 .
Egeland, B., & Sroufe, A. (1981). Developmental sequelae of maltreatment in infancy. In R.
Rizley & D. Cicchetti (Eds .), Developmental perspectives on child maltreatment (pp. 77-92) .
San Francisco: [os se y-Bass,
106 ROY C. HERRENKOHL

Egeland, B., [acobvitz, D., & Papatola, K. (1987). Intergenerational continuity of abuse. In
R. J. Gelles & J. B. Lancaster (Eds.), Child abuse and neglect: Biosocial dimensions (pp . 255-
276). New York: A1dine DeGruyter.
Elmer, E. (1977). Fragile families, troubled children. Pittsburgh: University of Pittsburgh Press .
Elmer, E. (1979). Child abuse and family stress. Journal of Social Issues, 35, 60-71.
Elmer, E. (1981). Traumatized children, chronic illness and poverty. In L. Pelton (Ed .), The
social context of child abuse and neglect (pp . 185-227). New York: Human Sciences Press.
Finkelhor, D., Hotaling, G. T., & Yllo, K. (1988). Stopping family violence: Research priorities
for the coming decade. Newbury Park, CA: Sage .
Garbarino, J. (1976). A preliminary study of some ecological correlates of child abuse: The
impact of socio-economical stress on mothers. Child Development, 47, 178-185 .
Garbarino, J. (1981). An ecological approach to child maltreatment. In L. H. Pelton (Ed .),
The social context of child abuse and neglect (pp . 228-267). New York: Human Sciences
Press.
Garbarino, J., Guttmann, E., & Seeley, J. W. (1986). The psychologically battered child. San
Francisco : Jossey-Bass.
Garmezy, N . (1983). Stressors of childhood. In N . Garmezy & M. Rutter (Eds.), Stress,
coping and development in children (pp . 43-84). New York: McGraw-Hili .
Geffner, R , Rosenbaum, A., & Hughes, H. (1988). Research issues concerning family
violence. In V. B. Van Hasselt, R. L. Morrison, A. S. Bellack, & M. Hersen (Eds .),
Handbook of family violence (pp . 457-481). New York: Plenum Press.
Gelles, R. J. (1973). Child abuse as psychopathology: A sociological critique and reformula-
tion . American Journal of Orthopsychiatry, 43, 611-621.
Gil, D. G. (1970). Violence against children: Physical abuse in theUnited States. Cambridge, MA:
Harvard University Pres s.
Goldstein, J., Freud, A., & Solnit, A. (1973). Beyond the best interests of the child. New York:
Free Press.
Gold ston, S. E. (1977). Defining primary prevention. In G. W. Albee & J. M. Joffee (Eds .),
Primary prevention of psychopathology, Vol. I (pp . 18-23). Hanover, NH : University Press
of New England.
Gray, J., & Kempe, R. (1976). The abused child at time of injury. In H. Martin (Ed .), The
abused child (pp. 57-65) . Cambridge, MA: Ballinger.
Hart, S. N ., Germain, R B., & Brassard, M. R (1987). The challenge: To better understand
and combat psychological maltreatment of children and youth. In M. R Brassard, R
Germain, & S. N . Hart (Eds.), Psychological maltreatment ofchildren andyouth (pp . 3-24).
New York: Pergamon Press.
Herrenkohl, E. C.; & Herrenkohl, R. C. (1981). Some antecedents and developmental
consequences of child maltreatment. In R. Rizley & D. Cicchetti (Eds .), Developmental
perspectives on child maltreatment (pp . 57-76). San Francisco: jossey-Bass,
Herrenkohl, E. C,; Herrenkohl, R. C,; & Toedter, L. (1983). Perspectives on the in-
tergenerational transmission of abuse. In D. Finkelhor, R. J. Gelles, G. T. Hotaling, &
M. A. Straus (Eds .), The dark side of families (pp . 305-316). Beverly Hills, CA: Sage .
Herrenkohl, E. c.. Herrenkohl, R. c.. Toedter, L. , & Yanushefski, A. M. (1984). Parent-
child interactions in abusive and non-abusive families . Journal of theAmerican Academy
of Child Psychiatry, 23, 641-648 .
Herrenkohl, R. C,; & Herrenkohl, E. C. (1988a). Assessing the occurrence of physical and
emotional maltreatment. Unpublished manuscript, Lehigh University, Bethlehem, PA.
Herrenkohl, R. C.; & Herrenkohl, E. C. (1988b). Identifying theoccurrence of maltreatment in
comparison groups . Unpublished manuscript, Lehigh University, Bethlehem, PA.
Herrenkohl, R. c., Herrenkohl, E. c., Egolf, B., & Seech, M. (1979). The repetition of child
abuse: How frequently does it occur? Child Abuseand Neglect , 3, 67-72.
RESEARCH DIRECTIONS 107

Herrenkohl, R. c., Herrenkohl, E. C,; & Egolf, B. (1983). Circumstances surrounding the
occurrence of child maltreatment. Journal of Consulting and Clinical Psychology, 51, 424-
431.
Hershorn, M., & Rosenbaum, A. (1985). Children of marital violence . American Journal of
Orthopsychiatry, 55, 260-266.
Hotaling, G., & Sugarman, D. (1986). An analysis of risk markers in husband and wife
violence : The current state of knowledge. Violence and Victims, 1, 101-124.
Hyman, I. A. (1987). Psychological correlates of corporal punishment. In M. R. Brassard,
R. Germain, & S. N. Hart (Eds.), Psychological maltreatment of children and youth (pp. 59-
68). New York: Pergamon Press .
Jones, D. (1987). The untreatable family. Child Abuse and Neglect, 11, 409-420.
Justice, B., & Justice, R. (1976). The abusing family. New York: Human Sciences Press .
Kadushin, A., & Martin, J. (1981). Childabuse: An interactional event. New York: Columbia
University Press.
Kaufman, J., & Zigler, E. (1987). Do abused children become abusive parents? American
Journal of Orthopsychiatry, 57, 186-192.
Lane, T. W., & Davis, G. E. (1987). Child maltreatment and juvenile delinquency: Does a
relationship exist: In J. D. Burchard & S. N. Burchard (Eds.), Prevention of delinquent
behavior. Newbury Park, CA: Sage.
Lusk, R., & Waterman, J. (1986). Effects of sexual abuse on children. In K. MacFarlane, J.
Waterman, S. Conerly, L. Damon, M. Durfee, & S. Long (Eds.), Sexualabuse of young
children (pp. 101-118). New York: Guilford Press .
Lynch, M. A., & Roberts, J. (1982). The consequences of abuse. London: Academic Press.
Martin, H. P. (1973). Nutrition: Its relationship to children's physical, mental and emo-
tional development. American Journal of Clinical Nutrition, 26, 766-775.
Martin, H. P. (1976). The abused child. Cambridge, MA: Ballinger.
McCleer, S. V. (1988). Psychoanalytic perspectives on family violence . In V. B. Van Hassett,
R. L. Morrison, A. S. Bellack, & M. Hersen (Eds.), Handbook of family violence (pp. 11-
30). New York: Plenum Press .
Milner, J. S., & Wimberley, R. C. (1979). An inventory for the identification of child abuse.
Journal of Clinical Psychology, 35, 95-100.
Mrazek, P., & Mrazek, D. (1987). Resilience in child maltreatment victims : A conceptual
exploration. Child Abuse and Neglect, 11, 357-366.
Nagi, S. (1977). Child maltreatment in the United States. New York: Columbia University
Press .
Navarre, E. L. (1987). Psychological maltreatment: The core component of child abuse. In
M. R. Brassard , R. Germain, & S. N. Hart (Eds.), Psychological maltreatment of children
and youth (pp. 45-56). New York: Pergamon Press .
Oates, K. (1986). Childabuse and neglect, whathappens eventually? New York:Brunner/Maze!.
Parke, R. D., & Collmer, C. (1975). Child abuse: An interdisciplinary analysis. Chicago: Univer-
sity of Chicago Press.
Pelton, L. H . (1981). Child abuse and neglect : The myth of classlessness. In L. H. Pelton
(Ed.), Thesocial contextof childabuse and neglect (pp. 23-38). New York:Human Sciences
Press.
Polansky, N. A., Chalmers, M. A., Buttenweiser, E., & Williams, D. P. (1981). Damaged
parents: An anatomyof child neglect. Chicago: University of Chicago Press.
Rosenberg, M. S., & Reppucci, N. D. (1985). Primary prevention of child abuse. Journal of
Consulting and Clinical Psychology, 53, 576-585.
Russell, D. (1983). The incidence and prevalence of intra familial and extrafamilial sexual
abuse of female children. Child Abuse and Neglect , 7, 133-146.
Schakel, J. A. (1987). Emotional neglect and stimulus deprivation. In M. R. Brassard, R.
108 ROY C. HERRENKOHL

Germain, & S. N . Hart (Eds .), Psychological maltreatment of children and youth (pp . 100-
109). New York: Pergamon Press.
Schneider, C. (1982). The Michigan Screening Profile of Parenting. In R. H. Starr (Ed .),
Child abuse prediction: Policy implications (pp. 157-174). Cambridge, MA: Ballinger.
Spinetta, J. J., & Rigler, D. (1972). The child abusing parent: A psychological review.
Psychological Bulletin, 77, 296-304.
Starr, R. H . (1982). Child abuse prediction: Policy implications. Cambridge, MA: Ballinger.
Starr, R. H. (1988). Physical abuse of children. In V. B. Van Hasselt, R. L. Morrison, A. S.
Bellack, & M. Hersen (Eds .), Handbook of family violence (pp . 119-155). New York:
Plenum Pres s.
Steele, B. F. (1976). Violence within the family. In R. E. Helfer & C. H. Kempe (Eds .), Child
abuse and neglect: The family and the community (pp . 3-23). Cambridge, MA: Ballinger.
Sudia, C. E. (1981). What services do abusive and neglecting families need? In L. H. Pelton
(Ed.), The social context of child abuse and neglect (pp . 268-290). New York: Human
Sciences Press.
Straus, M., Gelles, R., & Steinmetz, S. (1980). Behind closed doors: Violence in the American
family. Garden City, NY: Anchor Press.
Tong, L., & Oates, K. (1987). Personality development following sexual abuse. ChildAbuse
and Neglect, 11, 371-383.
Toro, P. A. (1982). Developmental effects of child abuse : A review. ChildAbuse and Neglect,
6,423-431.
Twentyman, C. T., Rohrbeck, C. A., & Am ish, P. L. (1984). A cognitive-behavioral ap-
proach to child abuse: Implications for treatment. In S. Saunders, A. M. Anderson, C.
A. Hart, & G. M. Rubenstein (Eds.), Violent individuals and families (pp. 87-211).
Springfield: Charles C Thomas.
Wald, M. S. (1976). State in terven tion on behalf of "neglected" children: Standards for
removal of children from their homes, monitoring the status of children in foster care,
and termination of parental rights. Stanford Law Review, 28, 623-706.
Wald, M. S., Carlsmith, J. M., Leiderman, P. H ., French, R. de S., & Smith, C. (1985).
Protecting abused/neglected children: A comparison of homeand foster placement. Palo Alto ,
CA: Stanford Center for the Study of Youth Development.
Wolfe, D. A. (1987). Childabuse: Implications for child development and psychopathology. New-
bury Park, CA: Sage .
Wolfe, D. A., Zak , L., Wilson, S., & Jaffe, P. (1986). Child witnesses to violence between
parents: Critical issues in behavioral and social adjustment. Journal of Abnormal Child
Psychology, 14, 95-104.
Wolfe, D. A., Wolfe, V. V., & Best, C. L. (1988). Child victims of sexual abuse. In V. B. Van
Ha sselt, R. L. Morrison, A. S. Bellack, & M. Hersen (Eds.), Handbook of family violence
(pp. 157-185). New York: Plenum Press.
 CHAPTER 5

SOCIAL AND EMOTIONAL

CONSEQUENCES OF
CHILD MALTREATMENT
LISE M. YOUNGBLADE AND JAY BELSKY

INTRODUCTION
More than a quarter of a century ago, Kempe and his colleagues
(Kempe, Silverman, Steele, & Droegemueller, 1962) alerted th e medical
and academic communities to the "battered child syndrome ." Ever
since, research on, and concern for, child maltreatment has proliferated .
Although substantial concern has been directed toward the victims of
abuse and neglect, most research has focused upon the perpetrators (see
Belsky, 1978, 1980, and Parke & Collmer, 1975, for reviews). There are
compelling reasons why etiology rather than consequences of child mal-
treatment have been the principle focus of empirical inquiry, perhaps
the most obvious of which is priorities. The first task of those concerned
with child abuse and neglect is to stop it from occurring again or to
prevent it from happening in the first place . In order for either remedia-
tion or prevention efforts to succeed, understanding of etiology is
essential.

LI SE M. YOUNGBLAD E AND JAY BELSKY· Department of Ind ividual and Family Studies,
College of Health and Human Development, Penns ylvania State University, University
Park, Penn sylvania 16802.

109
110 LISE M. YOUNGBLADE and JAY BELSKY

Despite this humanitarian imperative, there are several important

reasons for empirical inquiry into the developmental consequences of
child maltreatment to be a central focus of research on this topic. The
first, ironically enough, derives directly from concern with etiology. Be-
cause both clinical and empirical evidence link a history of maltreatment
in one's childhood to its subsequent perpetration as an adult (e.g., Bur-
gess & Youngblade, 1988; Egeland, [acobvitz, & Papatola, 1987; Kaufman
& Zigler, 1987), understanding the effects of child abuse and neglect on
the developing child should illuminate processes by which it is in-
tergenerationally transmitted.
A second reason for studying developmental consequences in addi-
tion to illuminating etiology is to gain insight into basic processes of
human development. As Rutter has stated so cogently:
Just as knowledge of normal development carries important lessons for those
wishing to unravel disease mechanisms, so the investigation of abnormality
may shed light on the course of normal development. This is because a focus
on the unusual may be crucial for pulling apart elements that ordinarily go
together (Rutter, 1982, p. 106; see also Cicchetti, 1984; Cicchetti & Sroufe,
1976).

Our third and final reason for reviewing in this chapter what is
known about the developmental effects of being abused or neglected as
a child is motivated in part by the results of an investigation reported
just a decade ago, indicating that the effects of child maltreatment ap-
pear to be indistinguishable from those of economic deprivation more
generally. In a study that challenged the field, Elmer (1977) compared
three groups of eight-year-olds, two of which has been identified as
victims of abuse or of accidents in their first year of life at time of
admission to Children's Hospital in Pittsburgh. Broad-based assess-
ments of 17 children from each group, plus an additional 17 control
subjects not previously studied, revealed few differences between
abused, accident, and control children-a finding that was true whether
one looked at physical health, language development, self-concept, in-
tellectual standing, school performance, or self-control. What was most
noteworthy was how widespread deficits were in all three samples of
children.
The results of the Elmer (1977) study called into question the mean-
ing of virtually all investigations up to that point in time, most of which
had been conducted by clinicians not blind to the child's rearing history
and without reliance upon appropriate comparison groups (see Aber &
Cicchetti, 1984, for review and critique). Thus, Elmer's findings raised
the very real possibility that the effects of child abuse might be indis-
tinguishable from that of rearing in an economically deprived family and
SOCIAL AND EMOTIONAL CONSEQUENCES 111

community setting, contexts which are known to foster maltreatment

(Pelton, 1978). In order to determine the extent to which conclusions
drawn from Elmer's work have been substantiated in the more rigorous
empirical work that has followed publication of her study, it is our plan
to review available research on the effects of child maltreatment on
socioemotional development. Before doing so, however, several general
conceptual, methodological, and theoretical issues must be considered.

ISSUES IN THE STUDY OF CHILD ABUSE AND NEGLECT

What is Child Abuse?

Perhaps the most serious conceptual issue to ponder is what, in
fact, is meant by the term childabuse. Certainly, this term captures a wide
range of behavior, including acts of commission (e.g., physical abuse,
sexual abuse, emotional and psychological abuse) and acts of omission
(e.g. , physical neglect, emotional neglect) (Giovannoni & Becerra, 1979).
Moreover, the continuum of child abuse subsumes a wide range of
intensity, from repeated slaps or harsh spankings, to one or several
blows with an instrument, to vicious verbal attacks, to cigarette burns,
and so on. The picture becomes even more clouded when we consider
the length of time since the abuse, the duration of abuse, and the devel-
opmental period in which the onset of abuse occurred (see Cicchetti &
Barnett, in press).
Unfortunately, the majority of studies in the literature fail to opera-
tionally define what is meant by child abuse, and do not consider these
various qualifiers in their conceptualization and measurement of abuse,
relying instead on descriptors such as "officially documented cases of
abuse." In all fairness, such an approach reflects the fact that precise
documentation of exactly what leads a family to be labeled as abusive or
neglectful is not equally clear across cases in the records of child welfare
agencies and because, in the ecology of family violence, multiple types
of abuse often co-occur (e.g., Cicchetti & Rizley, 1981; Egeland & Sroufe,
1981a). Because the predominant form of abuse perpetrated or the exact
mixture of problematical childrearing patterns is not always clearly artic-
ulated in case records or research reports, comparison and synthesis
across multiple studies can be quite difficult. Grouping together all fami-
lies that have maltreated their children can lead to confusion and ob-
scure very real differences in etiology, sequelae, cross-generational
transmission patterns, and treatment response for different types of
maltreatment (Rizley & Cicchetti, 1980; cited in Cicchetti & Rizley, 1981).
Because the lack of conceptual consistency and clarity in the liter-
112 LISE M . YOUNGBLADE and JAY BELSKY

ature, we are forced, for the purposes of this review, to employ a broad
definition of the term child abuse to refer to acts of commission, re-
gardless of duration, intensity or frequency, and "child neglect" to refer
to acts of omission, regardless of duration, frequency or intensity.
Where operational definitions are more refined in the research liter-
ature, we will draw attention to distinctions made between different
types of abuse. However, virtually no study enables us to draw distinc-
tions between groups in terms of duration, frequency, or intensity.

Methodological Issues
In terms of research methodology, there are three issues that must
be considered. The first, already alluded to, revolves around the use of
comparison/control groups. Although virtually all the research to be
cited involves contrast groups of some kind, by no means are all groups
(e.g., Gaensbauer, 1982) matched as carefully as were Elmer's (1977).
Second, we must consider the developmental sensitivity of the mea-
sures used. As we will describe shortly, in infancy the measure used
most widely to assess socioemotional "health" involves the assessment
of infant-mother attachment security. Even though this assessment has
shown adequate reliability and validity properties for use with 12- and
18-month-old infants in "normative" samples (see Waters, 1978), in
some of the studies to be reported it has been used to assess maltreated
infants ranging from 11 to 24 months, and with minor (Lamb,
Gaensbauer, Malkin, & Schulz, 1985) and major (Lewis & Schaeffer,
1981) procedural changes. On the other hand, children of older ages are
typically observed in unstructured situations interacting with persons
other than their parents-often their peers. And third, although a
number of reports emanate from longitudinal research investigations of
child abuse in infancy and early childhood-most notably the Min-
nesota Mother-Child Project (Egeland & Sroufe, 1981a,b) and the Har-
vard Child Maltreatment Project (e.g., Cicchetti, Carlson, Braunwald, &
Aber, 1987; Cicchetti & Braunwald, 1984; Gersten, Coster, Schneider-
Rosen, Carlson, & Cicchetti, 1986; Schneider-Rosen, Braunwald, Car-
lson, & Cicchetti, 1985; Schneider-Rosen & Cicchetti, 1984)-the major-
ity of research, in infancy, preschool, and mid- to late childhood, in-
volves cross-sectional research designs (see Aber & Cicchetti, 1984, for
review). All these methodological inconsistencies limit the comparisons
that can be made and inferences that can be drawn from this body of
research.
SOCIAL AND EMOTIONAL CONSEQUENCES 113

Theoretical Issues
A good deal of research on the developmental effects of child mal-
treatment can be regarded as atheoretical, in that it is guided by little
more than the common-sense notion that maltreatment is aversive and
thereby bad for children, and thus that children who have been sub-
jected to it should function more poorly than age-mates reared in more
considerate ways. Even so, much of the research can be cast in terms of
one of two predominant schools of thought-attachment theory (Ains-
worth & Wittig, 1969; Bowlby, 1969, 1980; Sroufe, 1977) and social learn-
ing theory (Bandura, 1977; Bijou & Baer, 1961; Patterson, 1982). The role
that attachment theory plays in guiding empirical inquiry is most evi-
dent in the work done on the youngest children who are hypothesized
to establish mistrusting or insecure affective bonds with abusive and
neglecting parents (e.g., Egeland & Sroufe, 1981a,b). The contribution of
social learning theory is most evident in studies of preschoolers and
older children, particularly in the study of aggressive behavior (e.g .,
Reid, Taplin, & Loeber, 1981), because it is hypothesized that children
who have been physically abused will be aggressive because they are
imitating the behavior they have been subjected to, reinforced for,
and/or are reproducing behavior patterns that they have seen rewarded.
Social learning theory and attachment theory are in many ways
distinct, in that the former, by tradition, has focused principally upon
overt behavior and the role of imitation and rewards and punishment in
the generation and maintenance of behavior patterns, whereas the latter
has been concerned particularly with affective bonds that influence how
individuals view themselves, others, and relationships more generally.
With the emergence of a more cognitively oriented social learning ap-
proach to behavioral development (Bandura, 1977), it is clear that de-
spite different language systems for explaining human development
processes, the two theoretical approaches have much in common
(Youngblade, Burgess, & Belsky, 1988). Noteworthy perhaps is the as-
sumption that ways of relating to others result from interpersonal expe-
rience and that such experiences not only shape what one does but also
what one attends to in the social arena and how social experience is
interpreted. Thus, both see the individual as an active agent who, as a
result of a social experience, develops expectations that guide interper-
sonal activity and shape the processing of interpersonal experience. Al-
though it is true that attachment theorists place more emphasis upon
how social experience fosters "internal working models" or affective -
cognitive processes that affect social functioning than on social skills per
114 LISE M. YOUNGBLADE and JAY BELSKY

se, whereas the reverse tends to be true of social learning accounts of

social development, it seems to us that differences are more of emphasis
and willingness to attribute an inner self to the individual than they are
of some fundamental disagreement about behavioral development. Be-
yond the actual assessment of attachment security per se, it is not at all
clear what differential predictions the two theoretical approaches would
lead to when it comes to considering the behavioral consequences of
child abuse and neglect. Indeed, virtually all the differences in the func-
tioning of children who have and have not been subject to maltreatment
are consistent with basic tenets of each theoretical orientation.

OVERVIEW

For the remainder of this chapter, we will survey the empirical

research that has emerged since the publication of Elmer's (1977) investi-
gation in order to examine the social and emotional consequences of
child abuse and neglect. We will review this literature with respect to the
developmental period of the dependent variable under investigation.
First, we will examine the relationship of child abuse and neglect to
children's socioaffective functioning in infancy and toddlerhood, as re-
flected in the security of infant-mother attachments and toddler-peer
interactions. Following, we will consider the relationship of child mal-
treatment to social interactions in the preschool and elementary school
years, both with parents and with peers. We will end with a discussion
of consequences for the intergenerational transmission of abusive pat-
terns of parenting in adulthood, before drawing some general conclu-
sions.

SOCIAL AND EMOTIONAL CONSEQUENCES DURING

INFANCY AND TODDLERHOOD
Most investigations of the socioemotional consequences of child
maltreatment on parent-child relations during the first two years of life
rely upon the Strange Situation as a means of assessing the security of
infant-parent attachment security. The Strange Situation (Ainsworth,
Blehar, Waters, & Wall, 1978; Ainsworth & Wittig, 1969)is an experimen-
tal laboratory procedure in which the infant's behavior is studied in
response to a series of separations and reunions with a stranger and a
parent. Irrespective of whether or not they are overtly distressed by the
procedure, infants judged to be secure in their relationship with their
SOCIAL AND EMarIONAL CONSEQUENCES 115

parent (Group "B" infants) greet the parent in an unambiguous man-

ner-by either smiling and vocalizing across a distance or approaching
the parent and establishing physical contact. The attachment bond is
judged to be insecure when the infant's response to the parent is one of
pointed avoidance, either by aborting an approach or averting gaze or
ignoring, or when the infant cannot find comfort in the parent's pres-
ence and directs angry resistant behavior at the parent upon reunion
(pushing off, kicking to be put down, refusing a toy). Although develop-
mentalists continue to debate the relative reliability, validity, and utility
of this methodology (Lamb, Thompson, Gardner, Charnov, & Estes,
1984), there is widespread agreement among many that despite its lim-
itations it is one of the most sensitive procedures that is available for
assessing individual differences in infant socioemotional development
(Belsky & Nezworski, 1988; Bretherton & Waters, 1985). Indeed, as shall
become evident, some of the best evidence of the validity of the pro-
cedure is to be found in its capacity to distinguish infants and toddlers
who vary in terms of their exposure to abuse and/or neglect.
Beyond the general expectation that maltreated children should be
more at risk for developing insecure attachments, it has been theorized
that because of their exposure to emotionally unavailable parents, and
particularly their experience of physical rejection, children who have
been physically abused should be at particular risk for establishing inse-
cure-avoidant (Group "A" infants) relationships (Ainsworth et al., 1978;
Egeland & Sroufe, 1981b). Infants and toddlers subjected to neglect, on
the other hand, some suggest, should be particularly at risk for develop-
ing insecure-resistant attachments (Group "C" infants). The reasoning
here is that the infant's failure to find comfort in the hands of the parent
and the anger that is expressed in the Strange Situation by the insecure-
resistant child is a function of the unresponsive or intermittently/ incon-
sistently responsive care that he or she has received (Egeland & Sroufe,
1981b). Consistent with this line of theorizing is evidence that non-
maltreated infants who are classified as insecure-resistant often have
experienced maternal care that is less contingent upon or responsive to
their own behavior, particularly their distress cries, than is the care
provided by mothers of secure infants (e.g., Belsky, Rovine, & Taylor,
1984; Isabella, Belsky, & Von Eye, 1989). Whereas the insecure-resistant
infant's behavior in the Strange Situation is considered to reflect the
frustration of not having one's needs consistently cared for, that of the
insecure-avoidant infant is presumed to reflect an unwillingness to dis-
play and share feelings with the parent and doubt about the parent's
willingness or ability to meet the child's needs for physical contact and
comfort when distressed. Having outlined the general expectations of
116 LISE M . YOUNGBLADE and JAY BELSKY

attachment theory with regard to child abuse and neglect, we will pro-
ceed to summarize the available evidence.
At the most global level, there is general consensus from an em-
pirical standpoint that maltreatment is associated with elevated rates of
insecure infant-mother attachments (Cicchetti & Braunwald, 1984; Crit-
tenden, 1985, 1988; Egeland & Sroufe, 1981a,b; Gordon & Jameson, 1979;
Lamb et al., 1985; Lyons-Ruth, Connell, Zoll, & Stahl, 1987; Schneider-
Rosen, Braunwald, Carlson, & Cicchetti, 1985; Schneider-Rosen & Cic-
chetti, 1984). The strength of this association is most evident when data
obtained from independent samples in separate studies are compiled
and subjected to statistical analysis at the aggregate level (see Table 1).
Although the data presented in Table 1 derive from investigations that
are cross-sectional and longitudinal in design, from studies using stan-
dardized or modified Strange Situations, and from research on children
of varying ages (see Table 2 for description of study characteristics), it is
clear that young children who have been maltreated are far more likely
to be classified as insecure in their attachments to their mothers than are
agemates from economically similar backgrounds who have not been
maltreated. Indeed, this is true even when the data are examined in
terms of distinct age groups (12 months, Table 1-2; 18 months, Table 1-3;
> 18 months, Table 1-4; and mixed ages, Table 1-5).
When it comes to addressing the issue of specificity, that is, whether
particular forms of maltreatment are differentially associated with differ-
ent patterns of attachment, some of the available evidence is consistent
with the proposition that infants subject to physical abuse should be
more at risk for developing insecure-avoidant attachments, whereas
those subject to neglect should be at heightened risk of developing
insecure-resistant attachments. Egeland and Sroufe (1981a,b), for exam-
ple, found that this prediction holds when studying 12-month-olds, and
Crittenden (1985, 1988), who studied young children of various ages,
reported abuse to be associated with avoidance and neglect with re-
sistance. Inconsistent with the specificity proposition, however, are
Egeland's and Sroufe's (1981b) data indicating that by 18 months of age
both abused and neglected children are most likely to be classified as
insecure-avoidant, a pattern also consistent with the cross -sectional
findings of Schneider-Rosen et al. (1985). Exactly why such developmen-
tal changes take place in the expression of insecurity in the Strange
Situation in the case of maltreated children remains unclear.
As it turns out, there is ever increasing evidence that both abused
and neglected children display elevated levels of resistance and of avoid -
ance (Crittenden, 1985, 1988; Lyons-Ruth et al., 1987), a finding which
has led to the emergence of a new attachment classification labeled Ale
(see also Carlson, Cicchetti, Barnett, & Braunwald, 1989). In fact, Crit-
SOCIAL AND EMOTIONAL CONSEQU ENCES 117

TABLE 1. The Relation between Maltreatment and Attachmentv-"

1. All ages; all studies comb ined-

Maltreatment Control Total

Insecure 328 134 462

(e = 247) (e = 215)
X2 [1) = 112.12, P < .0000
Secure 175 304 479
(e = 256) (e = 223)

503 438 941

2. At 12 months"
Maltreatment Control Total

Insecure 85 57 142
(e = 61) (e = 81)
X2 [1) = 33.01, P < .0000
Secure 37 106 143
(e = 61) (e = 82)

122 163 285

3. At 18 months'
Maltreatment Control Total

Insec ure 60 40 100

(e = 41) (e = 59)
X2 [1) = 25.49, P < .0000
Secur e 39 102 141
(e = 58) (e = 83)
99 142 241

4. Older than 18 months!

Maltreatment Control Total

Insecure 29 12 41
(e = 20) (e = 21)
x2 (1) = 22.56, P < .0000
Secure 14 32 46
(e = 23) (e = 23)

43 44 87

(continu ed)
118 LISE M. YOUNGBLADE and JAY BELSKY

TABLE 1. (Continu ed)

5. Varying ages in samples

Maltreatment Control Total

Insecu re 154 25 179

(e = 123) (e = 33)
X2 [1] = 67.25, P < .0000
Secure 42 64 106
(e = 73) (e = 33)
106 89 285

"In some cases, the same subjects are repor ted at more than one time period , because of the fact that
they were mea sured at mult iple ages.
bWithin each cell, data are tabled such that the top value reflects observed/actual frequenc y and the
bottom value (in paren theses) is the expected value .
<Egeland & Sroufe (1981a,b); Lyon s-Ruth et al . (1987); Schn eider-Rosen et al . (1985); Schneider-Rosen &
Cicche tti (1984); Carlson, Braunwald , & Cicchetti (1984); Crittenden (1985, 1988); Lamb et al . (1985);
Gordo n & Jame son (1979).
dEgeland & Sroufe (1981a,b ); Lyons -Ru th et al. (1985); Schn eider-Rosen et al . (1985).
' Egeland & Srou fe (1981a,b); Schn eider-Rosen et al . (1985).
/Schneider-Rosen & Cicchetti (1984)-19 mos.: Schneider -Rosen et al. (1985)-24 mos. (Note: A sepa rate
sys tem to score 24 mos. olds was dev eloped and validated .)
s Carlson et al . (1984); Crittende n (1985, 1988); Gordon & Jameson (1977); Lamb et al . (1985).

tenden (1988) discovered that without the new classification many mal-
treated infants were classified, apparently falsely, as secure. Such find -
ings raise questions as to whether the data presented in Table 1 might be
even more revealing if all studies included this classification category. In
fact, the absence of such a category may explain why Lyons-Ruth et al.
(1987) failed to find an association between attachment classification and
maltreatment, even though they discerned elevated levels of resistance
and avoidance among infants who were maltreated.
The co-mingling of these two expressions of insecurity in the re-
union episodes of the Strange Situation is quite unusual in view of
recent find ings indicating that avoidance and resistance tend to charac-
terize two distinct ends of a behavioral continuum and may even have
their origins, to some extent, in temperamental characteristics of the
infant (Belsky & Rovine, 1987; Frodi & Thompson, 1985). In view of the
possibility that some infants may be inclined to express their insecurity
in one form or another (i.e., resistance or avoidance) because of some
temperamental or affective proclivity, the behavior of maltreated infants
suggests that they may be so distressed and disorganized by their rear-
ing experience that they actually run the gamut of affective expression in
the search for a pattern of relating that will prove more acceptable to the
parent. Although some infants might be predisposed to become avoid-
 TABLE 2 . Study Characteristics

Maltreated sub jects Contrast group

Age Age Attachment
Reseachers (months) N Maltrea tment N (months) Matched on assessm ent Design

Carlson, Braunwald , 13-25 29 Abu sed /neglected 16 13-25 SES Standard Stran ge Cross-sectional
& Cicchetti (1984) Situation
Crittenden (1985) 2-24; 17 Abused 13 2- 24; SES Standard Strange Cross-sectional
i = 13.7 21 Neglected i = 13.7 Situation
22 Problematic
Crittenden (1985) 2-48; 22 Abused 29 2- 48; SES Standard Strange Cross-sectional
i = 24 31 Abused/neglected i = 24 Situ ation
20 Neglected
22 Marginally maltreated
Egeland & Sroufe 12, 18 19 Verbally abused 85 12, 18 SES Standard Strange Longitudinal
(1981a) 12, 18 19 Psychologically abu sed Situation
12, 18 24 Neglected
12, 18 24 Physically abu sed
Egeland & Sroufe 12, 18 33 Abu sed /neglected 33 12, 18 SES; but received Standard Strange Longitudinal
(1981b) "excellent" care Situation
Gaensbauer (1982) 12-19 12 Abused/n eglected ; low 20 12 Middle-class sample Noncla ssified Cross-sectional
SES 20 15
20 18
Gord on & Jameson 12-19 12 Non organic failure to 12 12-19 SES; ho sp ital expe- Mod ified Strange Cross-section al
(1979) thr ive rience Situ ation

(continued)
 TABLE 2. (Continu ed)

Maltreated subjects Contrast group

Age Age Attachment
Reseachers (months) N Maltreatment N (months) Matched on assessment Design

Lamb et aI. (1985) 8-32; 32 Abu sed /neglected 32 8-32; SES Modified Strange Cros s-sectional
i = 18.4 i = 18.7 Situation
Lyon s-Ruth et aI. 12 10 Abused/neglected 28 12 SES Standard Strange Cross -sectional
(1987) 12 18 Non -maltreated high- Situation; and
risk Rating Scales
Schneider-Rosen & 19 18 Abused/neglected 19 19 SES Standard Strange Cross-sectional
Cicchetti (1984) Situa tion
Schneider-Rosen et 12 17 Abused/ne glected 18 12 SES Stan dard Strange Cros s-sectiona l
aI. (1985) 18 24 Abused/n eglected 24 18 SES Situation
24 25 Abused/ne glected 25 24 SES
Sub sample:
12, 18 10 Abused/n eglected 14 12,18 SES Standard Stran ge Longitudinal
12, 18 16 Abu sed /neglected 16 12, 18 SES Situation
SOCIAL AND EMOTIONAL CONSEQUENCES 121

ant or resistant in their attachment relationship in response to care that

is insensitive but not abusive or neglectful, seriously neglected or
abused infants may simply be forced to abandon behavioral predisposi-
tions in the search for a safer form of expression. Unfortunately, it does
not appear that much success is achieved.
The findings reviewed regarding the association between attach-
ment security and child maltreatment clearly indicate that some degree
of specificity characterizes the relation between attachment and mal-
treatment. Consequently, when evidence from all available studies is
aggregated to assess the specificity hypothesis, a reasonable degree of
empirical support emerges for the propositions that avoidance should be
associated with abuse, neglect with resistance, and the combination of
avoidance and resistance with maltreatment. As the data displayed in
Table 3 indicate, the A/C classification is virtually restricted to children
who have been maltreated, particularly abused and abused/neglected
children. Infants and toddlers who have been neglected and abused/
neglected are over represented among children with insecure-resistant
classifications. Finally, although abused and abused/neglected children
are over-represented among children classified as insecure-avoidant, so
too are neglected children.
In summary, then, a strong association exists between child mal-
treatment and attachment insecurity. Not only does this theoretically
anticipated association emerge across studies and across ages, but to a
certain extent specificity of association is also evident in the data. In
view of these findings, as well as on the basis of both attachment and
social learning theory, there is reason to expect that the young maltreat-
ed child's behavior with other social agents should show evidence of
disturbance.

TODDLER-PEER RELATIONS

Central to attachment theory is the assumption that children's at-

tachment relationships should contribute to their interpersonal rela-
tionships outside the family because internal working models, or affec-
tive-cognitive processes derived from interactional experience with par-
ent, shape social relations with others. Perhaps the most compelling
evidence consistent with this proposition is research showing that tod-
dlers and preschoolers with varying attachment histories (as assessed in
the Strange Situation) behave differently toward peers and teachers
(Arend, Gove, & Sroufe, 1979; Lieberman, 1977; Sroufe, 1983). Both
social learning theory, with its emphasis on social skills and the gener-
alization of behavior patterns, and attachment theory, then, lead to the
122 LISE M. YOUNGBLADE and JAY BELSKY

TABLE 3 . The Relation between Type of Maltreatment

and Quality of Attachment-v

1. All ages; all studies comb in ed -

Type of maltreatment

Abused / Marginally
Classification Abu sed neglect ed Neglected maltreated Contr ol Total

A 43 62 43 18 74 240
(e = 27) (e = 46) (e = 32) (e = 16) (e = 119)
B 22 40 31 24 283 400
(e = 46) (e = 76) (e = 53) (e = 27) (e = 198)
C 12 34 30 7 45 128
(e = 15) (e = 24) (e = 17) (e = 9) (e = 63)
A/C 18 22 6 7 8 61
(e = 7) (e = 12) (e = 8) (e = 4) (e = 30)
Total 95 158 110 56 410 829
X2 (12) = 176.29, P < .0000

2. At 12 months"
Type of maltreatment

Abused / Marginally
Classification Abused neglected Neglected maltreated Control Total

A 16 17 5 4 32 74
(e = 7) (e = 11) (e = 9) (e = 5) (e = 42)
B 8 10 11 8 106 143
(e = 14) (e = 22) (e = 17) (e = 9) (e = 82)

C 4 12 17 1 19 53
(e = 5) (e = 8) (e = 6) (e = 3) (e = 30)
A/C 0 4 0 5 6 15
(e = 2) (e = 2) (e = 2) (e = 1) (e = 9)

Total 28 43 33 18 163 285

X2 (12) = 88.27, P < .0000
SOCIAL AND EMOTIONAL CONSEQUENCES 123

TABLE 3. (Cont inu ed)

3. At 18 months'
Type of maltreatment

Abused /
Classificati on Abused neglected Neglected Control Total

A 12 19 12 22 65
(e = 8) (e = 11) (e = 8) (e = 38)
B 12 12 15 102 141
(e = 16) (e = 24) (e = 18) (e = 83)

C 4 10 3 18 35
(e = 4) (e = 6) (e = 4) (e = 21)

Total 28 41 30 142 241

X2 [6] = 31.76, P < .0000

4. Older than 18 months!

Type of maltreatment

Abused /
Classificati on ne glected Control Total

A 18 5 23
(e = 11) (e = 12)

B 14 32 46
(e = 23) (e = 23)
C 11 7 18
(e = 9) (e = 9)

Tot al 43 44 87
X2 [2] = 16.45, P < .0000

5. Varying ages in sam ples

Type of maltreatment

Abused/ Mar ginally

Classification Abused neglected Neglected maltreated Control Total

A 15 8 26 14 15 78
(e = 14) (e = 11) (e = 17) (e = 14) (e = 22)

(continued)
124 LISE M . YOUN GBLAD E and JAY BELSKY

TABLE 3. (Continued)

B 2 4 5 16 43 70
(e = 13) (e = 10) (e = 15) (e = 12) (e = 20)

C 4 1 10 6 1 22
(e = 4) (e = 3) (e = 5) (e = 4) (e = 6)
A /C 18 18 6 2 2 46
(e = 8) (e = 7) (e = 10) (e = 8) (e = 13)
Total 39 31 47 38 61 216
X2 [121 = 111. 94, p < . ‫סס‬oo

«In some cases, the same subjects are reported at more than one time period, du e to the fact that they
were measured at multiple ages .
bWithin each cell, data are tabled such that the top value reflects observe d /actual frequ ency and the
bott om value (in parentheses) is the expected valu e.
<Crittenden (1985, 1988); Egeland & Sroufe (1981a, 1981b); Lamb et at. (1985); Lyon s-Ruth et at. (1987);
Schn eid er-Rosen & Cicchett i (1984); Schneid er-Rosen et at. (1985).
dEgelan d & Sroufe (1981b); Lyons-Ruth et at. (1987); Schn eider-Rosen et al. (1985).
' Egeland & Srou fe (1981b); Schn eider-Rosen et at. (1985).
/Schneider-Rosen & Cicchetti (1984)- 19 mos.; Schneide r-Rosen et at. (1985)-24 mos. (No te: A separate
sys tem was develop ed and validated to score 24 mos. olds .)
s Crittend en (1985; 1988); Lamb et at. (1985) .

expectation that children mistreated by their parents should have diffi-

culties in their interactions with their peers. Particularly noteworthy,
therefore, is a study by Lewis and Schaefer (1981) that fails to document
any differences in the social interactions of abused and non-maltreated
children aged 8 to 32 months old in a day-care center.
Indeed, on the basis of their find ings, these inve stigators were led
to question the core assumptions of attachment and social learning theo-
ry by concluding that parent-child and peer relationship systems are
not so much interconnected as they are autonomous. Although there
can be little doubt that not all aspects of relations with peers (or with any
others, for that matter) are derivative of earlier parent-child rela-
tionships, it would seem rather precarious to conclude that no linkage
exists. Not only are there serious problems with embracing null find-
ings, but the limits of the Lewis and Schaeffer study, particularly the
very short observation periods and the sensitivity of the data obtained,
should lead more cautious scientists to question the quality of the data
collected and, thus, thi s particular investigation's capability of assessing
the functioning of maltreating toddlers in interaction with age-mates.
Such caution would seem especially appropriate in view of other
evidence link ing abuse /neglect with problematic behavior with peers.
Most noteworthy are three studies of toddlers that show quite clearly, in
SOCIAL AND EMOTIONAL CONSEQUENCES 125

the aggregate, that maltreated children in interaction with age-mates are

more aggressive, less prosocial, and more disturbed in their responses
to others' distress, than are children who have not been abused or ne-
glected in the family. Two investigations were based upon detailed nar-
rative observations of 10 abused and 10 control children in four day-care
centers, two of which served battered children and two of which served
children of families under stress (George & Main, 1979; Main & George,
1985). Conceptualizing child abuse as representing one extreme along a
continuum of rejecting maternal behaviors, George and Main (1979)
hypothesized that abused children's behavior toward peers and child-
care workers would be similar to that displayed in the Strange Situation
by children who had been rejected but not abused; thus, abused tod-
dlers were expected to display avoidant, approach-avoidant, and ag-
gressive behavior. A third investigation (Howes & Eldredge, 1985),
using a somewhat different observational methodology of 5 physically
abused, 4 neglected, and 9 control children, observed during free and
structured play, generally replicated the results of the first two studies.
In the George and Main (1979) investigation, abused children were
much more likely to avoid other children and caregivers, almost always
exhibited approach-avoidance behaviors in response to prosocial initia-
tions, and spontaneously assaulted other children and caregivers signifi-
cantly more often than the non-maltreated children. In the Howes and
Eldredge (1985) study, maltreated children responded to aggression
with either aggression or resistance, whereas non-maltreated children
generally responded with distress; in response to prosocial behaviors,
maltreated children evinced resistance whereas non-maltreated children
displayed friendly behavior. Perhaps most interesting, though, was the
discovery in both studies that maltreated children responded to dis-
tressed peers with aggressive behaviors, whereas control children tend-
ed to respond with concern, empathy, and/or sadness. In the Main and
George (1985) study, eight of the nine abused toddlers responded with
fear, physical attack, nonphysical aggression, or diffuse anger to distress
incidents, whereas only one nonabused child reacted in this way.
Because the maltreated and non-maltreated samples of children in
the Main and George (1985) study were observed in separate settings,
and because it was not possible to control the frequency or intensity of
distress which children in either investigation witnessed in their day-
care program, some caution is called for in interpreting the findings.
Nevertheless, evidence does suggest that exposure to the distress of
others, rather than evoking the sympathy or concern that one might
expect, may elicit anger and hostility in young children who, presum-
ably, have experienced much distress in their own lives. Perhaps the
126 LISE M. YOUNGBLADE and JAY BELSKY

negative affect expressed by an age-mate proves contagious to the mal-

treated child and striking out against its source represents his or her
only way of coping or otherwise achieving some semblance of affective
self-regulation. Conceivably such a response is just what the child has
experienced at the hands of an insensitive parent. In any event, it seems
plausible that the behavior displayed by children in the studies dis-
cussed above may be one psychological mechanism through which child
maltreatment is intergenerationally transmitted.
Considered together, these three investigations (George & Main,
1979; Howes & Eldredge, 1985; Main & George, 1985) provide support
for the contention that there is an association between parent-child and
child-peer relations. Not only are maltreated toddlers more likely to be
aggressive-in response to prosocial encounters and to displays of dis-
tress-but they are also more likely to avoid interpersonal contacts with
familiar persons who have not mistreated them. There would seem to be
little doubt for concluding, then, at least with respect to the infancy-
toddler years, that child abuse and neglect foster insecure infant-parent
attachments and, perhaps thereby, what can only be regarded as dys-
functional peer relationships.

SOCIAL AND EMOTIONAL CONSEQUENCES

DURING CHILDHOOD
In view of the evidence summarized up to this point, indicating that
maltreatment can have profound effects on the infant-parent rela-
tionship, as well as upon the toddler's emerging peer relationships,
there is reason to anticipate difficulties, particularly in social rela-
tionships, as maltreated children grow older. The general absence of
longitudinal investigations makes it impossible, however, to determine
whether differences between older age children who have and have not
been maltreated are a function of earlier care or their concurrent experi-
ence in the family. This reality makes it necessary, before considering the
functioning of preschool and school-aged children who have been mal-
treated, to examine the ongoing experiences which such children have
in their families. Consideration of parent-child interaction patterns in
particular should alert us to the fact that we should not attribute any
dysfunctional behavior of abused and/or neglected children simply to a
particular episode of parental dysfunction, but rather to ongoing, daily
patterns of interaction in the family. After reviewing what is known
about the daily experiences of maltreated children, attention will be
SOCIAL AND EMOTIONAL CONSEQUENCES 127

turned to their own behavioral development and psychological func-

tioning during the preschool and elementary school years.

FAMILY INTERACTION IN MALTREATING HOUSEHOLDS

It is easiest to understand the experiences of abused/neglected chil-

dren in their families and to gain insight into their development by first
considering Patterson's groundbreaking work on dysfunctional family
interaction processes in households with deviant boys (Patterson, 1976,
1982), as there is all too much consistency between the evidence respon-
sible for his coercion theory of antisocial behavior and the experiences of
abused children. Central to the coercion model of inept family manage-
ment processes, derived as it is from social learning theory, is the notion
that parents, by being inattentive, erratic, and, thereby, noncontingent
in responding to the child's behavior, essentially-and inadvertently-
teach the child that if he engages in aversive activity or responds in a
sufficiently aversive manner, he will succeed in terminating parental
demands (Patterson & Reid, 1973, 1984). Over time, given a parental
failure to effectively discipline, a setting is generated marked by escalat-
ing aversive interchanges, in which the participants, both parents and
children, increase their use of hostile control techniques, including ver-
bal and physical assault (Patterson, 1986).
Consistent with this model is evidence that irritable parent disci-
pline practices and child coercion observed in the home predict more
generalized antisocial behavior, as reported by parents, teachers, peers,
and the child himself (Patterson, 1986; Patterson, Dishion, & Bank, 1984;
Patterson, Reid, & Dishion, in press). Other investigators, too, find that
harsh, erratic, power assertive, and inconsistent parent discipline prac-
tices precede aggressive, delinquent, and violent behavior in adoles-
cence (Loeber & Dishion, 1984; Olweus, 1980). Thus, according to this
model, parental failures in family management skills produce a child
who is antisocial and very likely lacking in social survival skills in such
areas as work, relationships, and academics (Patterson, 1986).
Upon compositing findings from multiple studies comparing paren-
tal attitudes and behavior of abusive and nonabusive parents, it becomes
readily apparent that Patterson's description of relationship dysfunction
is all too evident in maltreating households. For example, maltreating
parents respond to their children in a functionally noncontingent manner
(Dumas & Wahler, 1985; Patterson, 1979; Wahler, Rogers, Collins, &
Dumas, 1984), display substantially more aversive behavior in com-
parison with not just nonabusive parents, but even in comparison to
128 LISE M. YOUNGBLADE and JAY BELSKY

parents who experience child-management problems with their children

(Reid, 1984;see also Bousha & Twentyman, 1984),and acknowledge using
more punitive disciplinary practices and fewer reason-based ones (Trick-
ett & Kuczynski, 1986). They also experience more anger and conflict in
the family (Trickett & Sussman, 1988), display less approval (Herrenkohl
& Herrenkohl, 1981) and otherwise positive behavior toward their chil-
dren (Burgess & Conger, 1978; Oldershaw, Waters, & Hall, 1986), and no
doubt as a consequence view children and child-related activities less
positively than do non-maltreating parents (Disbrow, Doeer, & Caulfield,
1977). Thus, it will come as no surprise to attachment or social-learning
theorists that abusive parents are noticeably ineffective when it comes to
child management (Reid et al., 1981)and feel more detached in relation to
their offspring (E. C. Herrenkohl & R. C. Herrenkohl, 1981).

BEHAVIORAL FUNCTIONING OF MALTREATED CHILDREN

The family experiences of maltreated children, when considered in

light of both attachment and social learning theory, should affect not
only the ways in which children behave toward parents and age-mates,
but also how they feel about themselves.

Self-Concept
Central to attachment theory is the notion that the child learns more
than just social skills or ways of behaving as a function of the way he or
she is cared for. In fact, from the standpoint of attachment theory, behav-
ior itself is derivative of the child's internal working model, that is, the
child's self-image and views of relationships and the world. From what
we have seen, it can be anticipated that maltreated children should feel
less positively about themselves than do other children. Evidence from a
number of sources provides consistent support for this contention in
showing that maltreated children evince deficits in self-esteem (Kauf-
man & Cicchetti, 1989; Oates, Forrest, & Peacock, 1985), self-adjustment
(Perry et al., 1983; Straker & Jacobson, 1981), and emotional development
more generally (Kinard, 1980).
In an investigation of 37 6- to 14-year-olds admitted to a hospital
with a diagnosis of abuse and 37 matched controls, Oates, Forrest, and
Peacock (1985) found that the maltreated children scored significantly
lower on a measure of self-concept and, in addition, were less ambitious
than nonabused children with respect to occupational goals. Also, they
viewed themselves as having fewer friends . Even though Kinard's
SOCIAL AND EMOTIONAL CONSEQUENCES 129

(1980) study of 30 legally verified, physically abused children and 30

matched controls failed to discern any significant differences with re-
spect to self-esteem, it did find that abused 5 to 12-year-olds were less
trusting of others, just as the findings reviewed earlier pertaining to
child maltreatment and attachment security would lead one to antici-
pate. Despite the fact that it is difficult to reconcile the differences across
these two studies on measures of self-esteem, particularly because in the
investigation in which an effect of maltreatment on self-concept was
discerned (Oates et al., 1985) more time had passed since the abuse
report was filed (5.5 years, on average) than in the study in which the
differences between abused and control children on self-esteem were
not reliable (Kinard, 1980: 1 year), it should be noted that in both investi-
gations maltreated children did score lower than control children. Simi-
larly, in two investigations that relied upon parental as opposed to child
report, school-aged children who had been abused were rated by their
mothers as having poorer self-concepts (Perry et al., 1983) and lower self-
esteem (Kaufman & Cicchetti, 1989) than nonabused children. Notewor-
thy, too, is the fact that the children in the Perry et al. study were rated
by mothers as more poorly adjusted to school, as having more nonnor-
mal behaviors, and as possessing significantly fewer social and commu-
nication skills.
Further evidence of emotional and motivational impairment on the
part of maltreated children is provided by Aber and Allen (1987; Aber,
1984) in the Harvard Child Maltreatment Project (Cicchetti & Rizley,
1981). Their comparison of 91 maltreated children, 70 children from
AFDC families , and 30 children from middle-class households, all be-
tween 4 and 8 years of age, revealed that abused children at preschool
and schoolage evinced more dependency, less curiosity and, like the
aforementioned Perry et al. (1983) investigation, poorer cognitive func-
tioning. Especially important is that these differences remained even
after SES background factors were controlled.
The fact, however, that in the Perry et al. (1983) and in the Aber and
Allen (1987) investigations, samples that varied as a function of rearing
experience on measures related to the self-system differed also in mea-
sured intelligence raises the possibility that it is intellectual deficits that
are responsible for many of the apparent effects of maltreatment on self-
concept and adjustment that have been chronicled (Frodi & Smetana,
1984). Although the results of one investigation is consistent with this
reasoning, in showing that group differences in the ability to identify
and discriminate other people's emotions disappeared with IQ con-
trolled (Frodi & Smetana, 1984), results of a related inquiry showed that,
even with IQ controlled, between-group differences remained signifi-
130 LISE M. YOUNGBLADE and JAY BELSKY

cant in labeling feelings and decentering from one's perspective (Bahar-

al, Waterman, & Martin, 1981).
Differences in intelligence, brain damage, and psychosis could not
account for perhaps the most disturbing findings in the literature related
to self-concept and maltreatment, those pertaining to self-destructive
behavior. Upon comparing a clinical sample of 60 school -aged abused
children with 30 neglected and 30 control children, Green (1978) found
that nearly half of the abused children engaged in acts such as self-
mutilation and suicide attempts in response to actual or threatened sep-
aration or abandonment from parents or caretakers. Even though such
self-destructive behavior should not be regarded as routinely displayed
by maltreated children, it does underscore the extent to which abusive
care can undermine the integrity and even self-preservative function of
the self system.

Relations with Parents and Adults: Aggression/Noncompliance

One of the major difficulties in trying to identify the developmental
consequences of child maltreatment involves distinguishing cause from
consequence. To what extent might it be the case that the patterns of
functioning found to distinguish maltreated from nonmaltreated chil-
dren actually serve to evoke, or at least maintain their parents' "strat-
egies" of childrearing? The difficulties of interpreting the data are per-
haps most apparent when we consider the way maltreated children
behave, and are perceived by their parents to function, in the family.
Repeatedly, it has been observed that maltreated children manifest a
greater number and frequency of behavior problems of the type that are
classified as of the externalizing (as opposed to internalizing) variety,
including disobedience, tantrums and aggression directed toward other
family members (Aragona & Eyberg, 1981; Kaufman & Cicchetti, in
press; E. C. Herrenkohl & R. C. Herrenkohl, 1981; Hoffman-Plotkin &
Twentyman, 1984; Oldershaw et al., 1986; Reid et al. , 1981; Reidy, 1977;
Trickett & Kuczynski, 1986; Wolfe & Mosk, 1983). Four separate investi-
gations of parent-child interaction, using direct behavioral observation
data collected in the home (Burgess & Conger, 1978; E. C. Herrenkohl &
R. C. Herrenkohl, 1981; Reid et al., 1981) or home simulations (Older-
shaw et al., 1986), reveal that children in abusive families exhibit higher
rates of aversive behavior to parents and siblings than do nonmaltreated
children. Specifically, abused children emit more threatening demands
and physically negative behaviors (Reid et al., 1981), are more non-
SOCIAL AND EMOTIONAL CONSEQUENCES 131

compliant and aggressive (E. C. Herrenkohl & R. C. Herrenkohl, 1981;

Oldershaw et al., 1986), and display fewer positive behaviors (Burgess &
Conger, 1978; Oldershaw et ai., 1986) than do nonabused children. In
one study in which behavioral differences did not emerge, questions can
be raised about the length of the observation period and the extremely
small sample size (Mash, Johnston, & Kovitz, 1983).
It is not just observational studies that highlight externalizing-type
behavior on the part of maltreated children. Investigations that rely
upon parent reports of behavior problems reveal that the neglected
(Aragona & Eyberg, 1981) and the abused children (Mash et al., 1983) are
often rated by their mothers as more problematic than other children.
Perhaps the most informative parental report study is that designed by
Trickett and Kuczynski (1986) in which the abusive and the matched
control parents were trained to observe and immediately record natu-
rally occurring incidents of discipline in the home for 5 consecutive
days. The evidence so obtained revealed abused children to be more
noncompliant and aversive than control children. On the basis of both
the observational and parent report data, then, it is difficult to believe
that the aggressive, noncompliant and otherwise acting-out/externaliz-
ing patterns of behavior discerned, even if initially caused by maltreat-
ment, do not function to maintain such dysfunctional care.

Peer Relations
The research conducted during the preschool and early childhood
years, like that already summarized during the toddler years, under-
scores the apparent effect of maltreatment on relationships with age
mates. In one of the earliest relevant studies, Reidy (1977) compared 20
physically abused, 16 neglected, and 22 matched control children in a
multimethod assessment of aggressive child characteristics. In play,
abused children showed significantly more aggressive behavior than
neglected or control children, who rarely exhibited aggressive behavior.
Teachers' ratings characterized both abused and neglected children as
more aggressive than controls, though not significantly different from
each other. And, finally, abused children displayed significantly more
fantasy aggression on a projective test than either the neglected or con-
trol children. It must be noted that Straker and Jacobson (1981) could not
replicate the association between maltreatment and fantasy aggression
in their study of 19 abused and 19 control children between 5 and 10
years of age in South Africa.
132 LISE M. YOUNGBLADE and JAY BELSKY

E. C. Herrenkohl and R. C. Herrenkohl (1981), employing a multi-

method approach, also observed that maltreated children (in day-care
settings) were more aggressive toward their peers than were children in
3 control groups (all of whom were nonmaltreated, but whose families
were receiving either welfare, Head Start services, or day-care pro-
grams). A similar day-care-based observational study of physically
abused, neglected, and control children, aged 3- to 6-years-old revealed
that, overall, neglected children exhibited less prosocial behavior and
abused children more aggressive behavior than control children (Hoff-
man-Plotkin & Twentyman, 1984). Aggressive behaviors were found to
occur chiefly in response to difficult tasks or to interfering behavior by
peers, suggesting a lower tolerance of frustration.
A follow-up study of the children participating in the Harvard Child
Maltreatment Program which assessed 70 maltreated and 67 demograph-
ically similar matched comparison school-aged children in a daycamp
setting also discerned the now expected relation between maltreatment
and aggression, th is time using counselor and peer ratings (Kaufman &
Cicchetti, 1989). What makes this investigation particularly noteworthy is
not only the innovative setting that permitted extensive observations and
the large, well-matched sample, but also the attention paid to type of
maltreatment. When a variety of subgroups were compared, it was the
physically abused children who proved most aggressive.
As it turns out, it is not just acting-out, aggressive, or externalizing
behavior that differentiates maltreated and nonmaltreated children in
their interactions with their peers. The other standard form in which
developmental psychopathology is often manifested, namely internaliz-
ing behavior disorders, is hinted at in several investigations. In the
aforementioned South African study, Jacobson and Straker (1982) found
that their sample of 19 severely physically abused children, aged 5- to
10-years-old engaged in less social interaction and displayed less plea-
sure, concentration, and imagination in play during 5-min, triadic in-
teractions with 2 matched nonabused children than did these control
children. Similarly, observations of maltreated children in the summer
camp set up by the Harvard Child Maltreatment Project revealed, ac-
cording to counselor ratings, that maltreated children were more with-
drawn than other children (Kaufman & Cicchetti, 1989). The fact that
some of these children also received higher aggression ratings (see
above; see also Fagot, Hagan, Youngblade, & Potter, 1989) alerts us to
the fact that internalizing and externalizing disorders should not be
regarded as mutually exclusive.
The findings summarized earlier pertaining to the way in which
SOCIAL AND EMOTIONAL CONSEQUENCES 133

toddlers respond to peers who become distressed, along with the results
of studies across childhood that now consistently link maltreatment and
aggression, lead to the hypothesis that maltreated children do not sim-
ply have difficulty with self-control but, perhaps, are limited in a variety
of affective and perceptual arenas that are likely to influence their peer
interactions. Evidence that this is indeed the case comes from several
investigations highlighting deficits in their understanding of the view-
points of others and in their sensitivity to the affect displayed by others.
Consider in this regard Straker and Jacobson's (1981) finding that abused
5- to lO-year-olds were less empathic than matched controls, Frodi and
Smetana's (1984) discovery that abused preschoolers were less able to
identify and discriminate other people's emotions from picture stories
(though group differences disappeared with IQ controlled), and Baharal
et aI.,'s (1981) report that (even with IQ controlled) abused children were
less able than carefully matched controls to label feelings accurately, to
cognitively decenter, and to understand complex social roles.

Summary
During the preschool and school-age period there is evidence, just
as there was in the case of infants and toddlers, that child abuse in
particular is rather consistently, though not universally, associated with
problems in parent-child and peer relations as well as in attitudes and
feelings about the self. As we have seen, the dysfunctions evident
among maltreated children in all too many studies highlight both exter-
nalizing and internalizing behavior disorders. And as we have noted,
there can be little doubt that although such patterns of behaving are a
result of the quality of care the children have received, they also serve to
maintain problematic social experiences with parents, teachers/ coun-
selors, and agemates. From the perspective of social-learning theory it
would seem that the children have learned maladaptive wa ys of behav-
ing that function to elicit responses that maintain their problematic be-
havioral proclivities. From the standpoint of attachment theory, we can
speak of behavior patterns that serve to evoke from others responses
that confirm internal working models of the self as bad and unworthy of
love. To be sure, we need to acknowledge again that these interpretive
frames are by no means mutually exclusive and indeed are rather com-
plimentary. Even though one stresses social skills and behavior and the
other affective-cognitive processes, both see the maltreated child as a
product of his experience who actively contributes and apparently un-
dermines his continuing behavioral and psychological development.
134 LISE M. YOUNGBLADE and JAY BELSKY

SOCIAL AND EMOTIONAL CONSEQUENCES DURING

ADOLESCENCE AND ADULTHOOD
Given the evidence presented thus far regarding the dysfunctional
social and emotional consequences of abuse and neglect with respect to
child-parent and child-peer relationships during the infancy, preschool
and childhood years, we are faced with the obvious question concerning
the long-term impact of such experiences beyond these developmental
periods. Such impact is hard to document, again given the general lack
of prospective, longitudinal research, particularly with respect to dis-
cerning whether any long-term consequences are the result of the abuse
per se, interactive processes characteristic of abusive families, or proba-
bilistic life course trajectories set in motion via the circumstances
abused/ neglected children find themselves in, both in and outside of
the family (Burgess & Youngblade, 1990). In spite of these concerns,
there is evidence, nonetheless, to suggest maltreatment can affect ado-
lescent functioning (Garbarino, Sebes, & Schellenbach, 1984)and subse-
quent parenting, or what is routinely termed "the intergenerational
transmission" of child abuse (Egeland et al., 1987; Hunter & Kilstrom,
1979).

CONSEQUENCES IN ADOLESCENCE

Several reports indicate that a large portion of parental abuse vic-

tims are adolescents. For example, results from the National Study of
the Incidence and Severity of Child Abuse and Neglect (Burgdorff, 1982)
show that 47% of reported cases of maltreatment involve adolescents
between the ages of 12 and 17. Likewise, in a nationwide survey of 2,143
families, Straus, Gelles, and Steinmetz (1980) found that 54% of the 10-
to 14-year-olds and 36% of the 15- to 17-year-olds they surveyed had
experienced some form of maltreatment. Even so, very little research has
examined the problem of adolescent maltreatment, particularly in terms
of socioemotional consequences. Slightly more attention has been paid
to addressing reasons, processes, and conditions leading to the per-
petration and continuation of adolescent abuse (see Burgess & Richard-
son, 1984, and Garbarino & Gilliam, 1980). Nevertheless, one investiga-
tion aimed at examining families at risk for destructive parent-child
relations in adolescence provides some insight into socioemotional con-
sequences for adolescents.
In this study, Garbarino, Sebes, and Schellenbach (1984) assessed 62
clinically referred families (referred on the basis of adjustment problems
of the adolescent), containing at least one child between the ages of 10 and
SOCIAL AND EMOTIONAL CONSEQUENCES 135

16, on a battery of questionnaire, interview, and observation instruments.

Twenty-seven families were identified, on the basis of a self-report ques-
tionnaire tapping attitudes toward and the likelihood of appropriate
parental responses to adolescents' actions, as being "high-risk" for hav-
ing an abusive relationship, whereas 35 families were identified as "low-
risk." Using a choice of vignettes to provide an overall conclusion about
the family, based on talking with them for several hours, interviewers
rated 70% of the high-risk families, and 26% of the low-risk families as
abusive (Garbarino, Shellenbach, & Sebes, 1986). The "high-risk" group
of families tended to be more chaotic and enmeshed, to include more
stepparents (notably, all 8 stepparents in the "high-risk" group were,
reportedly, abusive), to be more punishing and less supportive, and to be
more stressed by life changes, than the "low-risk" group of families .
Adolescents in the "high-risk" families were characterized by signifi-
cantly more developmental problems (both internalizing and externaliz-
ing) and the number of such problems correlated significantly with the
risk for destructive parent-child relations.
Admittedly, although this investigation can be faulted for its re-
liance solely on a clinical sample, without a matched control group, and
on grounds of generalizability with respect to documentation and val-
idation of abusive processes or incidences, the results are strikingly
consistent with our earlier description of the characteristics of interac-
tions in abusive families and problematic developmental outcomes for
maltreated children. Of course, this is an area that merits future investi-
gation.

INTERGENERATIONAL TRANSMISSION

Significantly more attention has been devoted to the most per-

nicious long-term consequence of a history of maltreatment-the subse-
quent maltreatment of one's own children. The assertion that abusive
parents were maltreated as children not only pervades common knowl-
edge, but is widely reported in child-abuse publications, seemingly
without disagreement (Blumberg, 1974; Curtis, 1963; Galdston, 1965;
Gibbens & Walker, 1956; Helfer, 1980; Kempe et al., 1962; Silver, Dublin,
& Lourie, 1969; Steele & Pollock, 1968). For example, Steele (1983) stated
recently that "with few exceptions, parents or other caretakers who
maltreat babies were themselves neglected (with or without physical
abuse) in their own earliest years" (p. 235). Actually, however, these
commonly cited studies are largely clinical and retrospective in design,
involve selected case histories, are limited by the use of small, nonrepre-
sentative samples without comparison subjects, routinely lack defini-
136 LISE M. YOUNGBLADE and JAY BELSKY

tional criteria for "history of abuse" and "current abuse," and rely upon
observers who were not blind to the subjects' maltreatment status (Kauf-
man & Zigler, 1987). Importantly, because these investigations typically
do not employ parents who were maltreated but are now providing
adequate care to their own children, they tend to overestimate the inci-
dence of intergenerational transmission (Kaufman & Zigler, 1987).
Nevertheless, it is also true that more recent, better-designed, pro-
spective investigations provide support for the link between a history of
maltreatment and subsequent maltreatment of one's own children. For
example, in an investigation of 282 economically at-risk parents of new-
borns admitted to an intensive care nursery, 49 parents reported a histo-
ry of abuse and/or neglect at the initial interview. One year later, 10 of
these babies were confirmed as being abused of neglected; nine of the
abusing parents had a history of childhood maltreatment (Hunter et al.,
1978). In an unrelated study, Egeland et al. (1987) solicited information
from nearly 200 impoverished, predominantly single-parent mothers
regarding their childhood histories and current disciplinary practices.
Using a broad definition of abuse (definite plus borderline or suspected
cases), Egeland et al. reported a 70% rate of intergenerational transmis-
sion. Using a more conservative estimate (i.e. , reported cases only), they
found that 34% of the parents who had been abused but only 3% of the
parents who had been emotionally supported were mistreating their
children. These data are consistent with results from investigations
using less extreme samples (see Belsky & Pensky, 1988, for review). For
example, in their follow-up of English girls who had been institutionally
reared as children, Rutter and his colleagues found that, in comparison
to family-reared girls from the same neighborhood, the ex-care girls
were much more likely to show insensitivity to their 2- to 4-year-old
children, were more prone to exhibit irritability and use frequent spank-
ing and, as a consequence, were far more likely to be categorized as poor
parents (Dowdney, Skuse, Rutter, Quinton, & Mrazek, 1985; Quinton &
Rutter, 1985; Quinton, Rutter, & Liddle, 1984; Rutter & Quinton, 1984).
On the other hand, in each of the studies just reported, there also is
evidence that a significant number of parents, at least during the time of
the particular investigation, broke the intergenerational cycle. For exam-
ple, in the Hunter et al. (1978) investigation, 40 out of 49 parents, at least
during infancy, did not mistreat their infants despite their own histories
of maltreatment. How might this have happened? Interestingly, these
nonabusing parents, as well as the nonrepeaters in the Egeland et al.
(1987) sample, reported having more extensive social supports, and
were less likely to have been abused by both their parents as children
(see also Knutsen, Mehm, & Burger, 1984), were more apt to report a
SOCIAL AND EMafIONAL CONSEQUENCES 137

supportive relationship with one parent while growing up/ and were
more openly angry and better able to give a detailed coherent account of
their earlier abuse. Additionally, Egeland et al. (1987) found that involve-
ment with a supportive spouse or boyfriend, fewer current stressful life
events, and a conscious resolve not to repeat a history of abuse charac-
terized the nonabusing mothers.
Again, these buffering effects are consistent with results from stud-
ies using less extreme samples. Rutter and his colleagues, for example,
found that ex-care women who spoke warmly of their spouse and/or
indicated confiding in him were far more likely to be rated as good
parents and far less likely to be rated as poor parents, leading to the
conclusion that "the spouse's good qualities exerted a powerful ame-
liorating effect" on the parental functioning of women known to be at
risk as a result of their developmental history (Quinton, Rutter, & Lid-
dle/ 1984/ p. 115). Similarly, in a study of teenage mothers, Crockenberg
(1987) discovered that those who reported a history of parental rejection
but experienced good partner support were significantly less likely to be
angry and punitive toward their toddlers than those who received com-
parable care as a child but received limited partner assistance.
In summary, it appears that even though limitations of the data base
are widely acknowledged, most reviewers agree that a history of mal-
treatment in one's own childhood places the person at increased risk of
mistreating his or her own offspring (Belsky, 1978; Belsky & Pensky, 1988;
Burgess & Youngblade, 1988; Parke & ColImer, 1975). Importantly, how-
ever/ intergenerational transmission is not inevitable, even if potentially
likely. In a recent review, in fact, Kaufman and Zigler (1987) estimated the
transmission rate to be around 30%/ a figure that underscores the asser-
tion that the focus for researchers and practitioners should not simply be
whether transmission across generations occurs but the conditions under
which one might expect continuity (Burgess & Youngblade, 1988) or
"lawful discontinuity" (Belsky & Pensky, 1988). Although few studies
chronicle the factors that enable individuals to escape the intergenera-
tional cycle (but see Egeland, Jacobvitz, & Sroufe, 1988; Hunter & Kils-
trom, 1979)/ it is generally acknowledged that it is principally in interac-
tion with other etiological factors (e.g., child temperament, marital
quality, social support) that the risk associated with child-rearing history
is or is not "realized" (Belsky & Pensky, 1988; Cicchetti & Rizley, 1981).
Thus, we might conceptualize the risk of perpetuating the abused-abus-
ing cycle as being akin to the latent vulnerability of a brittle bone. In and of
itself, the property of being brittle will not cause the bone to break, but to
the extent pressure is put on the bone, the prospect of breakage increases.
Likewise, to the extent "pressure" (e.g ., no social or spousal support, low
138 LISE M. YOUNGBLADE and JAY BELSKY

income) is put on the parent who experienced abuse as a child, might the
parent succumb to maltreating his or her own child .

SUMMARY
We began this chapter by considering Elmer's (1977) findings indi-
cating that the developmental functioning of impoverished children
who were maltreated was indistinguishable from that of non-maltreated
children from equally impoverished households. Such results clearly
brought into question the assumed negative impact of child abuse and
neglect on child development. Our review of the evidence that has
become available since the publication of the Elmer study, framed as it
was in terms of attachment and social-learning theory, leads us to con-
clude that there are indeed serious socioemotional consequences of
being maltreated in childhood above and beyond those that emanate
from growing up in an economically disadvantaged household. Al-
though clearly not inevitable, we consistently found a history of abuse
and/or neglect to be linked to negative consequences.
More specifically, three relatively coherent and interconnected pat-
terns of socioemotional effects emerged. First, from infancy through
adolescence, we found maltreatment to be accompanied by dysfunc-
tional parent-child relations, marked by the increased likelihood of
forming an insecure attachment in infancy, coercive interpersonal ex-
changes in the childhood years, and chaotic, punishing and enmeshed
family life during adolescence. Second, the effects of child maltreatment
were not limited to familial relations, as there was repeated indication
that maltreatment is associated with dysfunctional peer relations. Sever-
al studies reveal maltreated children to be more aggressive, less pro-
social, and more disturbed in interaction with age mates than are com-
parison children. Particularly noteworthy was the discovery that in
response to displays of distress-as well as prosocial overtures-mal-
treated toddlers were more likely to be aggressive. Moreover, they were
more likely to avoid interpersonal contacts with familiar persons who
have not mistreated them (e.g., preschool teachers) . Third, data from
multiple investigations also indicated that abused and neglected chil-
dren tend to have lower self-esteem and to display significantly more
internalizing and externalizing behavior problems than non-maltreated
children. Most importantly, although such patterns of behaving are al-
most certainly a result of the quality of care the children have received,
they also serve to maintain problematic social experiences with parents,
teachers, counselors, and agemates (see also Johnson & Morse, 1968).
SOCIAL AND EMOTIONAL CONSEQUENCES 139

Not only does the experience of maltreatment have potentially pro-

found effects on the individual, but as we have seen, the effects of such a
history can manifest themselves in the next generation, insofar as that
individual is at risk of mistreating his or her offspring. Such intergenera-
tional transmission is certainly congruent with the argument that cog-
nitive-affective information about self, others, and relationships, as well
as behavioral skills relevant to those domains, derived from experience
in the family, is "not only internalized but carried forward to new rela-
tionships" (Sroufe & Fleeson, 1986, p. 61). It is by no means the case,
however, that all maltreated children grow up to mistreat their own
children. When discontinuity characterizes the developmental process,
it appears that some compensatory relationship experiences have taken
place with spouses, schoolmates, or some nonparental adult, which
presumably enhanced the individual's feelings of worthiness while, at
the same time, providing behavioral models of consideration and caring
(Belsky & Pensky, 1988; Burgess & Youngblade, 1988).
This last set of findings should not be taken to mean that one should
not be concerned about the long-term developmental consequences of
child abuse and neglect. Even if, as Kaufman and Zigler (1987) estimate,
onl y 30% of maltreated individuals grow up and abuse and/or neglect
their own offspring, one should wonder whether parental behavior is
influenced in less extreme, yet still adverse ways by a history of mal-
treatment in one's own childhood. What the findings on intergenera-
tional discontinuity do indicate, however, is that even an experience as
disconcerting as child maltreatment need not inevitably doom a child to
a life of psychological and behavioral dysfunction. If therapeutic experi-
ences can be provided which instill a sense of trust, of self-worth, while
providing behavioral skills to deal with others, particularly in affectively
charged situations, then there is every reason to expect that the effects
of child maltreatment can be ameliorated. For a fortunate few, and for
reasons that remain unclear, such experiences are provided in the course
of growing up and mating. For others it is likely that concerted efforts by
a caring community will need to be made to structure such growth-
promoting opportunities into their lives . In light of the widespread psy-
chological and behavioral deficits that maltreated children are at height-
ened risk for displaying, there is every reason for such intervention
efforts to be instituted at the earliest possible time .

ACKNOWLEDGMENTS

Work on this chapter was supported by a grant from the National

Institute of Child Health and Human Development (R01HD15496) and
140 LISE M. YOUNGBLADE and JAY BELSKY

by an NIMH Research Scientist Development Award (K02MH00486) to

the second author. Authors' address: College of Health and Human
Development, The Pennsylvania State University, University Park, PA
16802.

REFERENCES
Aber, J. L. (1984, August). Environmental influences on the socialemotional and cognitive devel-
opment of maltreated children. Paper presented at the 92nd Annual Meeting of the
American Psychological Association, Toronto.
Aber, J. L., & Allen , J. P. (1987). Effects of maltreatment on young children's socioerno-
tional development: An attachment theory perspective. Developmental Psychology, 23,
406-414.
Aber, J. L., & Cicchetti, D. (1984). The socio-emotional development of maltreated chil-
dren: An empirical and theoretical analysis. In H . Fitzgerald, B. Lester, & M. Yogman
(Eds .), Theory and research in behavioral pediatrics (Vol. 2, pp. 147-205). New York:
Plenum Press.
Ain sworth, M. D. 5., Blehar, M. C. ; Waters , E., & WalI, S. (1978). Patterns of attachment.
Hillsdale, NJ: Lawrence Erlbaum.
Ain sworth, M. D. 5., & Wittig, B. A. (1969). Attachment and exploratory behavior of one-
year-olds in a strange situation . In B. M. Foss (Ed.), Determinantsof infant behavior (Vol.
4, pp. 113-136) . London: Methuen.
Aragona, J. A., & Eyberg, S. M. (1981). Neglected children: Mother's report of child
behavior problems and observed verbal behavior. Child Development, 52, 596-602.
Arend, R, Gove, F., & Sroufe, L. A. (1979). Continuity of individual adaption from infancy
to kindergarten: A predictive study of ego resiliency and curiosity in preschoolers.
Child Development, 50, 950-959.
Beharal, R , Waterman, J., & Martin, H . (1981). The social-cognitive development of
abused children. Journal of Consulting and Clinical Psychology, 49, 508-516.
Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice HalI.
Belsky, J. (1978). Three theoretical models of child abuse: A critical review. ChildAbuse and
Neglect, 2, 27-49.
Belsky, J. (1980). Child maltreatment: An ecological in tegration . American Psychologist, 35,
320-335.
Belsky, J., & Nezworski, T. (Eds .). (1988). Clinical implications of attachment. Hillsdale, NJ:
Lawrence Erlbaum.
Belsky, J., & Pensky, E. (1988). Developmental history, personality and family rela-
tionships: Toward an emergent family system. In R Hinde & J. Stevenson-Hinde
(Eds .), The interrelation offamily relationships (pp. 193-217). London: Cambridge Univer-
sity Press.
Belsky, J., & Rovine , M. (1987). Temperament and attachment security in the Strange
Situation: An empirical rapproachment. Child Development, 58, 787-795.
Belsky, J., Rovine, M., & Taylor, D. (1984). The Pennsylvania infant and family develop-
ment project III: The origins of individual differences in infant-mother attachment:
Maternal and infant contributions. Child Development, 55, 718-728.
Bijou, S. w., & Baer, D. M. (1961). Childdeuelopmeni, Vol. 1: A systematic and empirical theory.
New York: Appleton-Century-Crofts.
SOCIAL AND EMOTIONAL CONSEQUENCES 141

Blumberg, M. L. (1974). Psychopathology of the abusive parent. American Journal of Psycho-

pathology, 28, 21-29.
Bowlby, J. B. (1969). Attachmentand loss, Vol. 1: Attachment. New York: Basic Books.
Bowlby, J. (1980). Attachmentand loss, Vol. 3: Loss, sadness and depression . New York: Basic
Books.
Bousha, D. M., & Twentyman, C. 1. (1984). Mother-child interactional style in abuse,
neglect, and control groups: Naturalistic observations in the home. Journal of Abnormal
Psychology, 93, 106-114.
Bretherton, 1., & Waters, E. (Eds.), (1985). Growing points in attachment theory and research .
Monographs of the Society for Research in Child Development 49 (6, Serial No. 209).
Burgdorff, K. (1982, December). Recognition and reporting of childmaltreatment: Findings from
the national study of the incidence and severity of childabuse and neglect. Prepared for the
National Center on Child Abuse and Neglect, Washington, D.C.
Burgess, R. L., & Conger, R. (1978). Family interaction in abusive, neglectful, and normal
families . Child Development, 49, 1163-1173 .
Burgess, R. L., & Richardson, R. A. (1984). Child abuse during adolescence. In R. M.
Lerner & N. L. Galambos (Eds .), Experiencing adolescents: A sourcebook for parents, teach-
ers, and teens (pp. 119-151). New York: Garland Publishing.
Burgess, R. L., & Youngblade, L. M. (1989). Social incompetence and the intergenerational
transmission of abusive parental-behavior. In R. J. Gelles, G. Hotaling, D. Finkelhor,
& M. Strauss (Eds.), New directions in family violence research (pp. 38-60). Beverly Hills:
Sage .
Carlson, V., Braunwald, K. G., & Cicchetti, D. (1984, August). Maternal history, family
environment and history of childmaltreatment: Their effects on securityof attachment. Paper
presented at the 92nd Annual Meeting of the American Psychological Association,
Toronto, Ontario, Canada.
Carlson, v., Cicchetti, D., Barnett, D., & Braunwald, K. (1989). Finding order in disor-
ganization: Lessons from research on maltreated infants' attachments to their care-
givers. In D. Cicchetti & V. Carlson (Eds.), Childmaltreatment: Theory and research on the
causes and consequences of child maltreatment. New York: Cambridge University Press.
Cicchetti, D. (1984). The emergence of developmental psychopathology. ChildDevelopment,
55,1-7.
Cicchetti, D., & Barnett, D. (in press). Toward the development of a scientific nosology of
child maltreatment. In D. Cicchetti & W. Grove (Eds.), Thinkingclearly aboutpsychology.
New York: Cambridge University Press.
Cicchetti, D., & Braunwald, K. (1984). An organizational approach to the study of emo-
tional development in maltreated infants. Infant Mental Health Journal, 5, 172-183.
Cicchetti, D., Carlson, V., Braunwald, K. G., & Aber, J. L. (1987). The sequelae of child
maltreatment. In R. Gelles & J. Lancaster (Eds .), Child abuse and neglect: Biosocial dimen-
sions. New York: Aldine Gruyter.
Cicchetti, D., & Rizley, R. (1981).Developmental perspectives on the etiology, intergenera-
tiona I transmission and sequelae of child maltreatment. New Directions for Child Devel-
opment, 11, 31-55.
Cicchetti, D., & Sroufe, L. A. (1976). The relationship between affective and cognitive
development in Down's Syndrome infants. Child Development, 47, 920-929.
Crittenden, P. M. (1985). Maltreated infants: Vulnerability and resilience. Journal of Child
Psychology and Psychiatry, 26, 85-96.
Crittenden, P. M. (1988). Relationships at risk . In J. Belsky & T. Nezworski (Eds.), Clinical
implications of attachment (pp . 136-174). Hillsdale, NJ: Lawrence Erlbaum.
142 LISE M. YOUNGBLADE and JAY BELSKY

Crockenberg, S. (1987). Predictors and correlates of anger toward and punitive control of
toddlers by adolescent mothers. Child Development, 58, 964-975.
Curtis, G. (1963). Violence breeds violence-perhaps. American Journal of Psychiatry, 120,
386-387.
Disbrow, M. A., & Doeer, H., & Caulfield, C. (1977). Measuring the components of par-
ents' potential for child abuse and neglect. Child Abuse and Neglect, 1, 279-296.
Dowdney, L., Skuse, D., Rutter, M., Quinton, D., & Mrazek, D. (1985). The nature and
qualities of parenting provided by women raised in institutions. Journal of Child Psy-
chology and Psychiatry, 26, 599-625.
Dumas, J. E., & Wahler, R. G. (1985). Indiscriminate mothering as a contextual factor in
aggressive-oppositional child behavior. Journal of Abnormal Child Psychology, 13, 1-18.
Egeland, B., [acobvitz, D., & Papatola, K. (1987). Intergenerational continuity of abuse. In
R. Gelles & J. Lancaster (Eds .), Child abuse and neglect: Biosocial dimensions (pp. 255-
276). New York: Aldine Gruyter.
Egeland, B., [acobvitz, D., & Sroufe, L. A. (1988). Breaking the cycle of abuse. Child
Deoelomeni, 59, 1080-1088 .
Egeland, B., & Sroufe, L. A. (1981a). Developmental sequelae of maltreatment in infancy.
New Directions for Child Development, 11, 77-92.
Egeland, B., & Sroufe, L. A. (1981b). Attachment and early maltreatment. Child Develop-
ment, 52, 44-52.
Elmer, E. (1977). A follow-up study of traumatized children. Pediatrics, 59, 273-279.
Fagot, B. I., Hagan, R., Youngblade, L. M., & Potter, L. (1989). A comparison of the play
behaviors of sexually abused, physically abused, and normal preschool children. Top-
ics in Early Childhood Special Education, 9(2), 88-100.
Frod i, A., & Smetana, J. (1984). Abused, neglected, and nonmaltreated preschoolers' abil-
ity to discriminate emotions in others: The effects of IQ . Child Abuse and Neglect, 8,
459-465.
Frodi, A., & Thompson, R. (1985). Infants' affective responses in the Strange Situation:
Effects of prematurity and of quality of attachment. ChildDevelopment, 56, 1280-1291.
Gaensbauer, T. J. (1982). Regulation of emotional expression in infants from two contrast-
ing caretaking environments. Journal of the American Academy of Child Psychiatry, 21,
163-171.
Galdston, R. (1965). Observations on children who have been physically abused and their
parents. American Journal of Psychiatry, 122, 440-443.
Garbarino, J., & Gilliam, G. (1980). Understanding abusive families . Lexington, MA: Heath
Publishing. .
Garbarino, J., Sebes , J., & Schellenbach, C. (1984). Families at risk for destructive parent-
child relations in adolescence. Child Development, 55, 174-183.
Garbarino, J., Schellenbach, c., & Sebes, J. (1986). Troubled youth, troubled families . New
York: Aldine.
George, c., & Main, M. (1979). Social interactions of young abused children: Approach,
avoidance, and aggression. Child Development, 50, 306-318 .
Gersten, M., Coster, W., Schneider-Rosen, K., Carlson, V., & Cicchetti, D. (1986). A socio-
emotional basis of communicative functioning: Quality of attachment, language de-
velopment, and early maltreatment. In M. E. Lamb, A. L. Brown, & B. Rogoff (Eds .),
Advances in developmental psychology (Vol. 4, pp. 105-151). Hillsdale, NJ: Lawrence
Erlbaum.
Gibbens, T. E. N., & Walker, A. (1956). Violent cruelty to children. British Journal of Delin-
quency, 6, 260-277.
Giovannoni, J., & Becerra, R. (1979). Defining childabuse. New York: Free Press.
SOCIAL AND EMOTIONAL CONSEQUENCES 143

Gordon, A. H., & Jameson, J. C. (1979). Infant-mother attachment in patients with non-
organic failure to thrive syndrome. Journal of the American Academy of Child Psychiatry,
18, 251-259 .
Green, A. H. (1978). Psychopathology of abu sed children. Journal of the American Academy of
Child Psychiatry, 17, 92-103 .
Helfer, R E. (1980). Developmental deficits which limit interpersonal skills. In C. H.
Kempe & R E. Helfer (Eds.), Thebattered child (3rd ed ., pp. 36-48). Chicago: University
of Chicago Press.
Herrenkohl, E. C,; & Herrenkohl, R C. (1981). Some antecedents and developmental
consequences of child maltreatment. New Directions for Child Development, 11, 57-76.
Hoffman-Plotkin, D., & Twentyman, C. T. (1984). A multi-modal assessment of behavioral
and cognitive deficits in abu sed and neglected preschoolers. Child Development, 55,
794-802 .
Howes, c., & Eldredge, R (1985). Respon ses of abused, neglected and non-maltreated
children to the behaviors of their peers. Journal of Applied Developmental Psychology, 6,
261-270.
Hunter, R., & Kilstrorn, N. (1979). Breaking th e cycle in abusive families . American Journal
of Psychiatry, 136, 1320-1322.
Isabella, R, Belsky, J., & Von Eye, A. (1989). Origin of infant-mother attachment: Examina-
tion of interaction synchrony during the infant's first year . Developmental Psychology,
25(1), 12-21.
Jacobson, R. S., & Straker, G. (1982). Peer group interaction of physically abused children.
Child Abuse and Neglect, 6, 261-270 .
Johnson, B., & Morse , H. A. (1968). Injured children and their parents. Children, 15, 147-
152.
Kaufman, J., & Cicchetti , D. (1989). The effects of maltreatment on school-aged children's
socioemotional development: Assessments in a day-camp setting. Developmental Psy-
chology, 25(4), 516-524 .
Kaufman, J., & Zigler, E. (1987). Do abused children become abusive parents? American
Journal of Orthopsychiatry, 57, 186-197 .
Kempe , C.; Silverman, F., Steele, 5., & Droegemueller, W. (1962). The battered child
synd rome. Journal of the American Medical Association, 181, 17-24.
Kinard , E. M. (1980). Emotional development in physically abused children. American
Journal of Orthopsychiatry, 50, 686-696.
Knutsen, J. F., Mehm, J. G., & Burger, A. M. (1984). Is violence in the family passed from
generation to generation? Unpublished manuscript, University of Iowa , Iowa City.
Lamb, M. E., Gaensbauer, T. J., Malkin, C. M., & Schulz , L. (1985). The effects of child
abuse and neglect on security of infant-adult attachment. Infant Behavior and Develop-
ment, 8, 35-45.
Lamb, M. E., Thompson, R A., Gardner, W. P., Charnov, E. L., & Estes , D. (1984).
Security of infantile attachment as assessed in the Strange Situation: Its study and
biological interpretation. The Behavioral and Brain Sciences, 7, 127-171.
Lewis, M., & Schaeffer, S. (1981). Peer behavior and mother-infant interaction in maltreat-
ed children. In M. Lewis & L. A. Rosenblum (Eds.), Theuncommon child. (pp. 193-223).
New York: Plenum Press .
Lieberman, A. (1977). Preschoolers' competence with a peer: Relationships with attach-
ment and peer experience. Child Development, 48, 1277-1287.
Loeber, R., & Dishion, T. J. (1984). Boys who fight at home and school : Family conditions
influencing cross-setting consistency. Journal of Consulting and Clinical Psychology, 52,
759-768.
144 LISE M. YOUNGBLADE and JAY BELSKY

Lyons-Ruth, K, Connell, D. B., Zoll, D., & Stahl, J. (1987). Infants at social risk: Relations
among infant maltreatment, maternal behavior, and infant attachment behavior. De-
velopmental Psychology, 23, 223-232.
Main, M., & George, C. (1985). Responses of abused and disadvantaged toddlers to dis-
tress in age-mates: A study in the day-care setting. Developmental Psychology, 21, 407-
412.
Mash, E. J., Johnston, C,; & Kovitz, K (1983). A comparison of the mother-child interac-
tions of physically abused and non-abused children during play and task situations.
Journal of Clinical Child Psychology, 12, 337-341.
Oates, R. K , Forrest, D., & Peacock, A. (1985). Self-esteem of abused children. ChildAbuse
and Neglect, 9, 159-163.
Oldershaw, 1., Waters, G., & Hall, D. K (1986). Control strategies and noncompliance in
abusive mother-child dyads: An observational study. Child Development, 57, 722-732.
Olweus, D. (1980). Familial and temperamental determinants of aggressive behavior in
adolescent boys : A causal analysis. Developmental Psychology, 16, 644-660.
Parke, R., & Collmer, C. (1975). Child abuse: An interdisciplinary review. In E. M.
Hetherington (Ed.), Reviewof childdevelopment research (Vo!. 5, pp . 509-590). Chicago :
University of Chicago Press .
Patterson, G. R. (1976). The aggressive child : Victim and architect of a coercive system. In
E. Mash, 1. Hamerlynck, & 1. Handy (Eds.), Behavior modification andfamilies (pp. 267-
316). New York: Brunner/Maze!.
Patterson, G. R. (1979). A performance theory for coercive family interaction. In R. B.
Cairns (Ed.), Theanalysis of social interactions: Methods, issues, and illustrations (pp. 117-
162). Hillsdale, NJ: Lawrence Erlbaum .
Patterson, G. R. (1982). Coercive family process. Eugene, OR: Castalia .
Patterson, G. R. (1986). The contribution of siblings to training for fighting: A microsocial
analysis. In D. Olweus, J. Block, & M. Radke-Yarrow (Eds.), Development of antisocial
and prosocial behavior: Research, theories, and issues (pp. 235-261). Orlando, FL: Academic
Press.
Patterson, G. R., Dishion, T. J., & Bank, 1. (1984). Family interaction: A process model of
deviancy training. Aggressive Behavior, 10, 253-267.
Patterson, G. R., & Reid, J. B. (1973). Intervention for families of aggressive boys : A
replication study. Behaviour Research and Therapy, 11, 383-394.
Patterson, G. R., & Reid, J. B. (1984). Social interactional processes within the family: The
study of moment-by-moment family transaction in which human social development
is embedded. Journal of Applied Developmental Psychology, 5, 237-262.
Patterson, G. R., Reid, J. B., & Dishion, T. J. (in press) . Antisocial boys. Eugene, OR:
Castalia.
Pelton, 1. (1978). Child abuse and neglect: The myth of classlessness. American Journal of
Orthopsychiatry, 48, 608-617.
Perry, M. A., Doran, 1. D., & Wells, E. A. (1983). Developmental and behavioral charac-
teristics of the physically abused child . Journal of Clinical and ChildPsychology, 12, 320-
324 .
Quinton, D., & Rutter, M. (1985). Parenting behaviors of mothers raised "in care." In A.
Nichol (Ed.), Longitudinal studies in child psychology and psychiatry (pp. 157-210). Lon-
don: Wiley.
Quinton, D., Rutter, M., & Liddle, C. (1984). Institutional rearing, parenting difficulties,
and marital support. Psychological Medicine, 14, 107-124.
Reid, J. B. (1984). Socio-interactional patterns in families of abused and nonabused chil-
SOCIAL AND EMOfIONAL CONSEQUENCES 145

dren. In C Zahn-Waxler, M. Cummings, & M. Radke-Yarrow (Eds.), Social and biolog-

ical origins of altruism and aggression. Cambridge: Cambridge University Press.
Reid, J. B., Taplin, P. 5., & Loeber, R. (1981). A social-interactional approach to the treat-
ment of abusive families . In R. Stuart (Ed.), Violent behavior: Social learning approaches to
prediction, management and treatment (pp. 83-1(0). New York: Brunner/Maze!.
Reidy, T. J. (1977). The aggressive characteristics of abused and neglected children. Journal
of Clinical Psychology, 33, 1140-1145.
Rizley, R., & Cicchetti, D. (1980). Heterogeneity in the etiology, type, sequelae, and treatment of
child maltreatment. Unpublished manuscript, Harvard University, Cambridge.
Rutter, M. (1982). Epidemiological-longitudinal approaches to the study of development.
In W. Collins (Ed.), Minnesota symposium of child development: Vol. 15. The concept of
development (pp . 105-144). Hillsdale, NJ: Lawrence Erlbaum .
Rutter, M., & Quinton, D. (1984). Long-term follow-up of women institutionalized in
childhood: Factors promoting good functioning in adult life. British Journal of Develop-
ment and Psychology, 2, 191-204.
Schneider-Rosen, K., Braunwald, K. G., Carlson, V., & Cicchetti, D. (1985). Current per-
spectives in attachment theory: Illustration from the study of maltreated infants. In I.
Bretherton & E. Waters (Eds.), Growing points in attachment theory and research: Mono-
graphs of the Society for Research in Child Development (pp . 194-210). Chicago: University
of Chicago Press.
Schneider-Rosen, K., & Cicchetti, D. (1984). The relationship between affect and cognition
in maltreated infants: Quality of attachment and the development of visual self-recog-
nition. Child Development, 55, 648-658.
Silver, L. B., Dublin, C C, & Lourie, R. S. (1969). Does violence breed violence? Contribu-
tions from a study of the child abu se syndrome. American Journal of Psychiatry, 126,
404-407.
Sroufe , L. A. (1977). Wariness of strangers and the study of infant development. Child
Development, 48, 731-746.
Sroufe, L. A. (1983). Infant-caregiver attachment and patterns of adaptation in preschool:
The roots of maladaptation and competence. In M. Perlmutter (Ed.), Minnesota sym-
posium in child psychology (Vol. 16, pp. 41-83). Hillsdale, NJ: Lawrence Erlbaum .
Sroufe, L. A. , & Fleeson, J. (1986). Attachment and the construction of relationships. In W.
Hartup & Z. Rubin (Eds.), Relationships and development. Hillsdale, NJ: Lawrence
Erlbaum .
Steele, B. F. (1983). Child abuse and neglect. In J. D. Call, E. Galenson, & R. L. Tyson
(Eds .), Frontiers of infant psychiatry. New York: Basic Books.
Steele, B. F., & Pollock, C (1968). A psychiatric study of parents who abuse infants and
small children. In R. Helfer & C Kempe (Eds .), Thebattered childsyndrome (pp . 89-131).
Chicago: University of Chicago Press.
Straker, G., & Jacobson, R. S. (1981). Aggression, emotional maladjustment and empathy
in the abused child . Developmental Psychology, 17, 762-765.
Straus, M. A., Gelles, R. J., & Steinmetz, S. K. (1980). Behind closed doors: Violence in the
American family. Garden City, NY: Anchor Press.
Trickett, P. K., & Kuczynski, L. (1986). Children's misbehaviors and parental discipline
strategies in abusive and nonabusive families. Developmental Psychology, 22, 115-123.
Trickett, P. K., & Sussman, E. J. (1988). Parental perceptions of child-rearing practices in
physically abusive and nonabusive families. Developmental Psychology, 24, 270-276.
Wahler, R. G., Rogers, D., Collins, B., & Dumas, J. E. (1984, November). Maintenance factors
in abusive mother-child interactions: The compliance and uncertainty hypotheses. Paper pre-
146 LISE M. YOUNGBLADE and JAY BELSKY

sented at the 18th Annual Convention of the Association for Advancement of Behav-
ior Therapy, Philadelphia.
Waters, E. (1978). The stability of individual differences in infant-mother attachment. Child
Development, 49, 483-494.
Wolfe, D. A., & Mosk , M. D. (1983). Behavioral comparisons of children from abusive and
distressed families . Journal of Consulting and Clinical Psychology, 51, 702-708.
Youngblade, L. M., Burgess, R. L., & Belsky, J. (1988, May). Are social skillsenough? A multi-
level view of the ecology of peer rejection. Paper presented at the 5th Biennial Waterloo
Conference on Child Development, Waterloo, Ontario.
 PART III

RISK FACTORS ASSOCIATED

WITH CHILD ABUSE
AND NEGLECT
 CHAPTER 6

SOCIOLOGICAL AND
ECOLOGICAL FACTORS
JOAN 1. VONDRA

INTRODUCTION
Understanding how child maltreatment evolves-whether it is charac-
terized by violent confrontations between parent and child, inattention
to a child's needs for adequate nutrition or supervision, chronic emo-
tional belittlement and/or withdrawal of affection, or some combination
of these-is inherently a task of integration. Knowledge from develop-
mental psychology about what children need for healthy psychological
development, knowledge from clinical psychology about the origins and
manifestations of child and/or adult psychopathology, knowledge from
family disciplines about the dynamics that underlie day-to-day family
functioning and crisis situations, and knowledge from sociology about
social and economic forces that foster or undermine the well-being of the
family, each make a vital contribution to our understanding of both
normal and dysfunctional parenting. Thus, the study of parental care
and its effects on child development is, or should be, by its very nature

JOAN 1. VONDRA· Department of Psychology in Education, University of Pittsburgh , Pitts-

burgh, Pennsylvania 15260.

149
150 JOAN 1. VONDRA

interdisciplinary in its perspective. Without integrating these comple-

mentary sources of information, we are left with an incomplete portrait
of parenting-its origins, its expression, and its impact on subsequent
generations of children and parents. Indeed, nowhere is this more ap-
parent than in the case of child maltreatment.
Emerging understanding of both the necessary and sufficient causes
of maltreatment points to broad and interacting sets of factors that con-
tribute to individual instances of abusive behavior or, more commonly, a
pattern of chronic emotional maltreatment and neglect of the physical
and/or psychological needs of the developing infant, child, or adoles-
cent. These factors arise from within and outside the family and, more
often than not, converge to create a family situation characterized by
both extreme need and an inability to develop or maintain the external
supports that could help bolster this very fragile system. Indeed, if
research has accomplished little else, it has demonstrated again and
again that recurrent maltreatment is not the outcome of any single fac-
tor-whether parental psychopathology or the experience of maltreat-
ment in childhood, child temperamental or behavioral deviance, marital
conflict or violence, economic hardship and job stress, inadequate and
ineffective social supports, or sociocultural mores that encourage puni-
tive, authoritarian parenting.
Nevertheless, by examining each factor in turn, it becomes clearer
how these multiple factors interact and coalesce to produce circum-
stances that are ripe for abusive and/or neglectful parenting. In this
sense, we are discussing an "ecology" of child maltreatment-an inte-
grative approach to the etiology of maltreatment that focuses on the
interaction of both individual and environmental characteristics.
Given this perspective, it stands to reason that any compelling in-
tervention model must incorporate this multifactor model in its selection
and organization of services for families . Interventions that fail to ad-
dress the multiple problems confronting these families are unlikely to
provide for the full range of services necessary to help these families to
get back on their feet and in control of their lives . It will become in-
creasingly apparent throughout this chapter that there are as many com-
binations of inadequate resources and dysfunctioning individuals as
there are families, but that the patterns of need, distress, and distur-
bance repeat themselves over and over, and offer a consistent message
about the nature and scope of the interventions we are challenged to
provide. By examining these problems with care, we are better equipped
to suggest strategies for effective, long-term change.
SOCIOLOGICAL AND ECOLOGICAL FACTORS 151

THE MULTIPLE DETERMINANTS OF PARENTING

A MODEL OF INFLUENCES

In the decade since Bronfenbrenner (1977) proposed his integrative

"ecological" model of influences on development, much progress has
been made linking work from different fields and broadening theoretical
perspectives in order to formulate comprehensive models of the deter-
minants of parenting. Models describing normative samples (Belsky,
1984; Elder, Caspi, & Downey, 1986; Elder, Liker, & Cross, 1984; Samer-
off, 1975; Sameroff & Chandler, 1975) and maltreating samples (Belsky &
Vondra, 1989; Cicchetti & Rizley, 1981; Engfer & Schneewind, 1982) chal-
lenge simplistic notions of main effects and unidirectional influence.
Not only do multiple factors operate cumulatively and interactively, but
many may have multiple paths of influence and may serve to modify or
ameliorate the impact of other contributing factors.
Although it is always possible to focus on a single source of influ-
ence and to measure its relation to parenting variables, the truth is that
every factor is embedded in an entire network of influences, and each
factor's relation to parenting is, in part, a consequence of its relations
with many other relevant influences. Ignoring these relations descrip-
tively, or failing to examine or control for them methodologically, leads
to misleading overemphasis of the importance of individual factors. Cer-
tain conditions or circumstances may play a central role in influencing
other resources or risk conditions. Nevertheless, it is the cumulative
balance of stress and supports (Belsky, 1984) or "protective" and "vul-
nerability" factors (Cicchetti & Rizley, 1981) that determines individual
differences in parental care.

FACTORS WITHIN THE FAMILY

CHILDREARING HISTORY

Parents bring with them a long history of experiences within their

family of origin and in interaction with peers, teachers, and other adults
and children loosely comprising their social network. Attachment theo-
ry (Ainsworth, Blehar, Waters, & Wall, 1978; Bretherton, 1985; Sroufe
& Fleeson, 1986) provides persuasive argument for the enduring influ-
ence of the quality of relationship established and maintained with
152 JOAN I. VONDRA

one's primary attachment figure(s). According to attachment theory,

infants construct an "internal working model" of relationships based
on the quality of their primary attachment relationship. When this
model is derived from and maintained over time in the context of a trou-
bled attachment relationship, the working model a child-and later
adult-applies to subsequent relationships is unlikely to foster secure,
mutually satisfying, enduring interpersonal relations (see Belsky & Nez-
worski, 1988).
The validity of this argument is evident in research on college stu-
dents and adults (Hazan & Shaver, 1987;Kobak & Sceery, 1988)that links
self-described attachment status or retrospective reports of attachment
history with current psychological adjustment, social functioning, and
attitudes about relationships. Individuals evidencing insecure attach-
ment relationships from childhood are described by peers as less flexible
and resourceful, more anxious, and more hostile . They express less
positive expectations of love relationships and maintain their "most
important love relationship" over a shorter period of time. They also
describe themselves as more lonely, having more distress symptoms,
and having less supportive peer and family relationships. In other
words, cumulative attachment experiences begun early in life would
appear to exercise considerable influence over the quality and percep-
tion of close interpersonal relationships established over the course of
later childhood and adulthood.
Other data in support of the existence and impact of attachment
"working models" are emerging from recent intergenerational attach-
ment studies (see Ricks, 1985). There is growing evidence that mothers
who experience unresolved anger and distress over troubled childhood
relations with their own parents tend to have "insecure" relationships
with their infants and young children. Fraiberg (1980; Fraiberg, Adelson,
& Shapiro, 1975) provided clinical case studies illustrating this same
pattern among maltreating mothers. In her casework, Fraiberg found
that abusive and neglectful mothers could not mobilize resources in
support of their infants' development until their own history of attach-
ment disturbances was acknowledged and addressed. The same obser-
vation has been reported by Greenspan and his colleagues (Greenspan,
Wieder, Nover, Lieberman, Lourie, & Robinson, 1987) in their interven-
tion work with maltreating mothers.
Within the maltreating family, interpersonal difficulties and/or dis-
turbances characterize multiple family members as well as multiple fam-
ily relationships. A pattern of troubled relationships, without neces-
sarily involving overt physical abuse or neglect, has often been
SOCIOLOGICAL AND ECOLOGICAL FACTORS 153

established long before maltreatment evolves in the family system. Ka-

ufman and Zigler (1990) reported an incidence rate for physical maltreat-
ment of 33% among parents similarly maltreated in childhood. At the
same time, there is ample evidence that a high proportion of parents
who maltreat and/or who have a child removed from their care have
experienced disturbances and disruptions in relations with their own
parents, without necessarily having suffered the identical form of mal-
treatment they themselves perpetrate (Altemeier, O'Connor, Vietze ,
Sandler, & Sherrod, 1982; Engfer & Schneewind, 1982; Kotelchuck, 1982;
Newberger, Reed, Daniels, Hyde, & Kotelchuck, 1977; Rutter, Quinton,
& Liddle, 1983).
When considering the broad spectrum of troubled parent-child re-
lations, maltreating adults appear to share a common history charac-
terized by insecure, unstable, and/or pathological relations with their
parents. Given what we know from attachment research, it is significant
that these impoverished relations are typically reflected in their other
intimate relationships as well. For example, recent data indicate that the
vast majority of maltreated infants exhibit insecure attachment relations
with their caregivers (Cicchetti & Olsen, 1990; Crittenden & Ainsworth,
1989). Studies described later in the chapter will highlight deficiencies in
relations of maltreating parents with partners and peers as well. Taken
together, these findings strongly suggest that attachment issues form
the crux of the maltreating family's interpersonal problems. Attachment
issues can apparently predispose vulnerable parents to maltreat their
own children, who then carry forward the socioemotional legacies of
another generation of disturbed attachment relationships.
This is not to say that every child who experiences even severe
attachment disruptions or disturbances will maltreat or even have trou-
bled relations with his or her own children. Retrospective prediction to
maltreatment is invariably more accurate than is prospective prediction.
A childhood history of pathogenic attachment relations can, for exam-
ple, serve as the basis for rejecting harsh parental attitudes and practices.
Indeed, data suggest that open acknowledgment and criticism of painful
attachment experiences may be the "decoupling" mechanism that halts
the progression of disturbed parent-child relations (including official
"maltreatment") across the generations (Main & Goldwyn, 1984). Con-
sidered from another standpoint, however, defensively positive or ide-
alistic descriptions of childhood attachment experiences on the part of
parents maltreated in childhood may preclude discerning real group
differences between maltreating and comparison parents. Methodologi-
cally, this implies careful attention to the measures used to assess retro-
154 JOAN 1. VONDRA

spectively the quality of childhood experiences (see Main & Goldwyn,

1985).
Despite variations in its expression from one generation to the next,
then, the disturbed attachment relations and emotional suffering that
underlie all maltreatment are, in all probability, passed down from par-
ents to children. The mechanism of transmission may be dysfunctional
working models of relationships, reactive personality characteristics
(e.g., low self-esteem, frustration tolerance, and/or impulse control),
modeling and internalization of parenting roles and behaviors, or a com-
bination of all three. In any event, the underlying emotional messages of
rejection and worthlessness based in disturbed parent-child rela-
tionships seem to be the common factor in child maltreatment across
families and generations. It has already been pointed out that the legacy
of emotional maltreatment may be especially difficult to overcome (Gar-
barino & Vondra, 1987; Hart & Brassard, 1987). Attachment theory pro-
vides the theoretical underpinning for this observation. Physical mal-
treatment may not be passed down in the majority of families, but
attachment disturbances and emotional maltreatment may very well be.

ADULT PSYCHOLOGICAL RESOURCES

With this psychological background, it is understandable why the

maltreating parent has typically been characterized in terms of psycho-
logical disorders and deficits (Altemeier, Vietze, Sherrod, Sandler,
Falsey, & O'Connor, 1979; Brunnquell, Crichton, & Egeland, 1981; Es-
troff, Herrera, Gaines, Shaffer, Gould, & Green, 1984; Spinetta, 1978).
Gilbreath and Cicchetti (1990; see also Kaplan, Pelcovitz, Salzinger, &
Ganeles, 1983), for example, report a rate of clinically defined depression
among maltreating mothers (47%) that is twice that of low-income com-
parison mothers (22%). The deficits exhibited by depressed mothers in
the care of their infants and children have been well documented (Bill-
ings & Moos, 1983; Coletta, 1983; Field, 1984; Hops, Sherman, & Biglan,
in press; Tronick & Field, 1986).
In fact, qualitative differences in parenting have been linked to a
wide variety of individual and personality characteristics, including age,
emotional maturity, ego-strength, and mental health (see Belsky, 1984;
Belsky & Vondra, 1989). In each case, studies can be cited reporting
poorer status or functioning of maltreating families in these domains
(e.g ., Brunnquell et al., 1981; Melnick & Hurley, 1969). On the average,
maltreating parents tend to be younger when they have their children,
more simplistic and egocentric in their thinking about themselves and
SOCIOLOGICAL AND ECOLOGICAL FACTORS 155

their children, less mature in their emotional expression and regulation,

and more likely to exhibit symptoms of mental distress or illness.
Depression, negative affectivity (Watson & Clark, 1984), poor ego-
control and ego-resiliency (Block & Block, 1980), low self-esteem, and a
host of other impairments in ego functioning may well have their roots
in the early family environment. With or without the contribution of
genetic endowment, the kind of troubled parent-child relations de -
scribed earlier among maltreating families no doubt increases greatly the
degree of subsequent adult psychopathology. Data are already ac-
cumulating that demonstrate relations among both high-risk (Erickson,
Sroufe, & Egeland, 1985) and nonclinical populations (Lewis, Feiring ,
McGuffog, & [askir, 1984; Sroufe, 1983) between insecurity of early at-
tachment and both psychological deficits and behavior problems in
childhood. Thus, childrearing history-through the channel of attach-
ment security-has indirect implications for child maltreatment via
adult psychological resources and functioning.
Dubow, Huesmann, and Eron (1987) provide an important direct
link in this regard by relating childhood experiences of rejection and
authoritarian punishment to subsequent ego development in adult-
hood. Their data, like those of Rohner and Rohner (1980) in their cross-
cultural study of parental rejection and those of Elder et al. (1986)in their
intergenerational study of families during the Depression, help support
a model of effects whereby childhood experiences shape adult psycho-
logical resources which, in turn, influence differences in parental care .
To the extent, then, that childhood experiences undermine subsequent
adult psychological well-being, they play an additional, albeit indirect,
role in shaping parental behavior.
Here, then, is an example of one source of influence-childrearing
history-that affects parenting outcomes via multiple pathways. On the
one hand, childhood experiences of parental care are hypothesized to
have a direct impact on the quality of relations established with one's
own children. On the other hand, experiences of early care influence
parenting outcomes indirectly by promoting or undermining parental
psychological resources. To ignore either factor-childrearing history or
parental personality-is to leave the picture incomplete and the inter-
vention model potentially weakened.

THE MARITAL RELATIONSHIP

Another critical pathway by which childrearing history and parental

personality-and factors outside the family as well-have an impact on
parenting is through the presence/absence and quality of relationships
156 JOAN I. VONDRA

established with friends, neighbors, and, of particular relevance to this

section, spouse or partner. Both psychological background and current
functioning are strongly implicated in determining the quality of and
satisfaction with social network support, including that provided by the
spouse(s) or partner(s) one selects.
As the evidence reviewed here and reported in diverse fields of
study indicates, childhood experiences translate into differences in so-
cial skills and relationship "working models" that, no doubt, contribute
to the quality of intimate relationships established in adulthood. By the
age of two to four years, for example, it is already possible to distinguish
maltreated children by abnormalities and deficits in their social interac-
tions with both peers and adults (George & Main, 1979, 1980; Gaens-
bauer & Sands, 1979; Hoffman-Plotkin & Twentyman, 1984). Social defi-
cits are cited in studies of older maltreated children as well (Howes &
Espinosa, 1985; Kaufman & Cicchetti, 1990).
These deficits, established in childhood, help account for consistent
findings demonstrating a poverty of both intimate and extrafamilial sup-
portive social relationships among maltreating parents (Crittenden,
1985). Social isolation is perhaps the single most common finding of
studies comparing maltreating families with low-income comparisons
(Egeland & Brunnquell, 1979; Kotelchuck, 1982; Rosario, Salzinger, Feld-
man, & Hammer, 1987; Starr, 1982; see also Garbarino & Gilliam , 1980).
Crittenden's (1985) work suggests, however, that social isolation may be
more characteristic of neglectful parents, whereas social conflict-arising
from enmeshed, asymmetric relationships-may be more characteristic
of abusive parents. In any case, maltreating mothers are more likely
than their low-income comparisons to be unmarried at the time of study
(Egeland & Brunnquell, 1979; Kotelchuck, 1982; Rosario et al., 1987),
suggesting that stable, long-term support from an intimate partnership
may be lacking.
Further inferences about the quality of relationships established by
maltreating parents may be drawn from the observed association be-
tween interspousal violence and child maltreatment (Rosenbaum &
O'Leary, 1981; Straus, Gelles, & Steinmetz, 1980). Violence between
partners and violence between parent and child are likely to co-occur.
From the standpoint of indirect child-rearing effects, it is interesting to
note the correlation Gwartney-Gibbs, Stockard, and Bohmer (1987)
found among undergraduates between self-reports of witnessing par-
ents in violent confrontations and reports of aggressing against and
sustaining violence from dating partners. Whether or not these young
adults also sustained violence from their parents, there is evidence link-
ing violence in the family of origin to violence in earl y dating rela-
tionships (see also Kalmuss, 1984).
SOCIOLOGICAL AND ECOLOGICAL FACTORS 157

Cummings, Zahn-Waxler, and Radke-Yarrow (1981) have begun

elucidating one probable pathway of effects in their work on young
children's responses to the expression of anger by family members and
strangers. Observing physical aggression in the context of anger epi-
sodes and/or frequent interparental fighting seems especially to elicit
emotional arousal and affective responses among young children and to
sensitize these children to conflict. In the researchers' view, such experi-
ences act as "socioenvironmental stressors" that could, when they in-
volve individuals of importance to the child, threaten his or her sense of
social security. These results and interpretation offer potential explana-
tory power about the link between violence in childhood and in later
adulthood that is distinct from the conceptual positions of both attach-
ment and observational learning theory.
In conjunction with data presented earlier on the effects of early
maltreatment, it is probable that troubled, emotionally abusive rela-
tionships in childhood (perhaps observed as well as experienced) jeopar-
dize the quality of later adult relationships, particularly the critical mar-
ital relationship. Correlational data reported by Crockenberg (1987) are
consistent with this hypothesis. Mothers in her low-SES sample who
perceived their own mothers to be less accepting reported receiving less
current social support both from them and from their own partners.
These mothers were especially likely to appear angry and punitive to-
ward their 2-year olds. Similarly, in their study of middle-class mothers
in Germany, Engfer and Gavranidou (1985, 1988) found that maternal
developmental history predicted marital satisfaction, which itself was
associated with both maternal sensitivity to the ir infants and later use of
extreme physical punishment.
The central importance of conjugal relationships as a source of so-
cial support for parents is by now well-established. Intimate support of
this kind is associated with less maternal depression (Bromet & Cornely,
1984; Colletta, 1983), greater psychological well-being (Levitt, Weber, &
Clark, 1986), more positive parenting attitudes and more developmen-
tally supportive parenting (Cotterell, 1986; Crnic, Greenberg, Ragozin,
Robinson, & Basham, 1983; Crnic, Greenberg, Robinson, & Ragozin,
1984; Engfer & Gavranidou, 1985; Goldberg & Easterbrooks, 1984;
Weinraub & Wolf, 1983).
Pianta, Egeland, and Hyatt (1986), in particular, noted that when
disadvantaged mothers engaged in unstable, intimate relationships dur-
ing their children's first four years, initially maternal sensitivity and,
ultimately child functioning, was jeopardized. The same was not true of
low-income mothers who were able to maintain stable relationships with
their partner. Evidence, then, that maltreating parents are especially
likely to be deficient in social skills and supportive social relationships is
158 JOAN I. VONDRA

very much in keeping with the hypothesized role of intimate social

support in the parenting model. Whatever strengths these adults may
possess as parents-likely to be meager from the outset-their inability
to garner the support of a stable, intimate partner will undermine their
functioning as caregivers.

CHILD CHARACTERISTICS

The role that the child plays in eliciting patterns of parental care is
accorded special attention in the case of child maltreatment. Neverthe-
less, a growing body of research exists that demonstrates relations be-
tween child factors and parental care within the general population as
well . Factors during infancy, such as gender, prematurity, congenital
handicaps, and temperament, have all been cited as correlates of the
quality of care received. With age, differences in care are increasingly
related to the cognitive status and to observable behaviors of the child .
However, the bidirectional nature of these relations is apparent even at
birth. Parents shape their children's behavior and development which,
in turn, influences subsequent caregiving, and so on .
One finding within the child maltreatment literature is that mal-
treated infants tend to be ill more often (Sherrod, O'Connor, Vietze, &
Altemeier, 1984) and that their mothers experience somewhat more
pregnancy and birth complications (Egeland & Brunnquell, 1979).This is
tempered by the fact, however, that abusive mothers have been found to
express more negative feelings about their pregnancy, to be delivering
more unplanned and unwanted infants, and to have made fewer prepa-
rations for the infant in terms of prenatal and infant care than low-
income comparison mothers (Altemeier et al., 1982, 1984; Egeland &
Brunnquell, 1979). The extent to which these latter findings also reflect
unstable and/or conflicted relations with the father of the child has not
been examined.
Similarly, although neglected infants have been observed to with-
draw socially and affectively in response to their social environment
(Fraiberg, 1980; Gaensbauer & Sands, 1979; Provence & Lipton, 1962),
maltreating mothers have been found to respond adversely to vid-
eotapes of smiling or crying infants, and to exhibit greater hostility and
intrusiveness, and less responsivity, in their interactions with their in-
fants during the first year of life (Crittenden, 1981; Dietrich, Starr, &
Kaplan, 1980; Frodi & Lamb, 1980; Lyons-Ruth, Connell, Zoll, & Stahl,
1987). Using prospective observational data from six infants later identi-
fied as victims of extreme physical punishment, Engfer and Gavranidou
SOCIOLOGICAL AND ECOLOGICAL FACTORS 159

(1988) were able to document that, as newborns, these infants did not
differ in irritability or social responsiveness from demographically
matched controls. As early as 8 months, however, they found these
youngsters to be significantly unhappier and more negative in mood
and, by 33 months, less cooperative and compliant than comparison
children. Importantly, the mothers of these children appeared less sen-
sitive in their interactions even on the maternity ward, and later ap-
peared more negative, angry, and coercive with their children. Further-
more, they described their youngsters as more difficult at every age of
assessment, ratings found to correlate more with characteristics of the
mothers themselves than with observed infant behavior. There is cer-
tainly reason to suspect that the majority of behavioral differences ob-
served in maltreated children have their foundation in the poor care
and/or disturbed relationships the children share with their caregivers.
The same may be noted for older children. Maltreating parents
report that their school-aged children exhibit a clinical rate of behavior
problems, and investigators observe that maltreated preschoolers en-
gage in less "positive self-directed activity" in the home, exhibit low
self-esteem, ego-control, persistence, and compliance in the laboratory,
more behavior problems, dependency, and psychopathology in pre-
school, and perform more poorly on intelligence tests than low-income
comparisons (Aragona & Eyberg, 1981; Barahal, Waterman, & Martin,
1981; DiLalla & Crittenden, 1987; Egeland, Sroufe, & Erickson, 1983;
Estroff et al., 1984; Green, 1978; Sandgrund, Gaines, & Green, 1974).
This is countered by findings that abusive parents are less positive and
more negative and hostile, and that neglecting parents are less positive,
more critical, and more controlling in social interactions with their chil-
dren (Aragona & Eyberg, 1981; Burgess, Anderson, Schellenbach, &
Conger, 1981; DiLalla & Crittenden, 1987; Mash, Johnston, & Kovitz,
1983). In addition, maltreating parents provide, in general, a poorer
quality home environment as a context for development than do low-
income controls (Rosario et al., 1987; Trickett & Susman, 1988).
Obviously, there is a problem with untangling cause and effect in
these observations. Starr, Deitrich, Fischhoff, Ceresnie, and Zweier
(1984) discuss this same issue in reference to the finding that maltreated
children are more likely to be physically or mentally handicapped. Char-
acteristics that make maltreated infants and children less appealing and
more difficult to care for very likely evolve, at least in part, from the
quality of care received . . . even in the prenatal period. By the time
they reach preschool and school age, maltreated youngsters are actively
contributing to a destructive cycle of aversive interactions (Burgess &
Conger, 1978; Trickett & Kuczynski, 1986). Trying to assign responsibil-
160 JOAN 1. VONDRA

ity either to parents or children for the problematic relationship at this

point becomes a meaningless endeavor.

FACTORS OUTSIDE THE FAMILY

SOCIAL NETWORK SUPPORT

The significance and probable etiology of maltreating parents' social

isolation and/or interpersonal difficulties have already been pointed out.
Dissatisfaction with social network support, less contact with friends
and more exclusive contact with relatives who are likely to be needy
themselves, and greater reliance on institutional support agencies are
common patterns in at-risk and/or actively maltreating families (Collet-
ta, 1983; Crittenden, 1985; Rosario et al., 1987; Starr, 1982). Egeland and
his colleagues (Egeland, Breitenbucher, & Rosenberg, 1980) put it this
way:
These [mothers who provide inadequate care] were reported to be easily
frustrated and annoyed and quick to respond to their frustration in a hostile
and aggressive fashion. They tended to annoy and alienate their families and
friends rather than using these relationships to help in dealing with problems
and crises . (p. 203)

The impact on children of parental social network deficits is at least

twofold. On the one hand, inadequate social support undermines the
functioning of parents, both as individuals and as caregivers. Parents
need the emotional support and instrumental assistance furnished by
social networks for their own day-to-day functioning and feelings of
well-being (Colletta, 1983; Levitt et al., 1986; Whittaker & Garbarino,
1983). These needs are compounded when childcare demands add to the
burden of meeting personal and familial needs. Parents who report dis-
satisfaction with the support provided by friends, neighbors, and rela-
tives tend to be dissatisfied with their caregiving role, to engage in less
optimal parent-child interaction, and to provide a poorer quality home
environment for child development (Cotterell, 1986; Crnic et aI., 1983,
1984; Stevens, 1988; Weinraub & Wolf, 1983).
At the same time, relations that are conflicted, enmeshed, or other-
wise self-limiting, affect children directly via the observations and expe-
riences they furnish (Cochran & Brassard, 1979). Garbarino (Garbarino
& Gilliam, 1980; Garbarino & Sherman, 1980) further pointed out that
the social networks maltreating families establish typically fail to provide
the social control functions what would normally set limits on extremes
SOCIOLOGICAL AND ECOLOGICAL FACTORS 161

of parental behavior. Thus, these impoverished social relations offer

neither positive standards or role models nor any lifeline or safety net to
these families in need. Combined with the fact that maltreating families
are especially likely to be stressed by extreme poverty, unstable job and
marital arrangements, family addictions and psychopathology, and trau-
matic life events (Egeland et al., 1980; Greenspan etal., 1987; Kotelchuck,
1982), these social support deficits take on added significance.

SOCIOECONOMIC CONSIDERAnONS

That the majority of chronically maltreating families fall within the

lowest social echelons is no coincidence. Economic, sociocultural, and
interpersonal factors act jointly in these families to create a situation of
severe economic stress, hardship, and dependency that has been cited
as the single greatest threat to adequate family functioning (Gil, 1970;
Siegal, 1982; see also Garbarino & Gilliam, 1980).
Decades of research support the link between low socioeconomic
status and styles of child-rearing that emphasize authoritarian control,
encouragement of conformity, and punitive disciplinary techniques, all
of which increase the probability of child maltreatment. Engfer and
Gavranidou (1988) note three general clusters of attitudes toward child-
rearing that are common among maltreating families : (1) a quest for
emotional gratification from one's own children, (2) impatience and
helplessness in the face of negative child behaviors, and (3) rigid au-
thoritarianism. The first is typical of a history of attachment distur-
bances, the second of an insecure parent-child relationship, and the
third of lower-class families in general.
Lower-class mothers of below-average education are consistently
found to provide home environments and to employ child-rearing strat-
egies that are associated with poorer developmental functioning and
lower educational achievement among children (e.g., Bradley & Cald-
well, 1984; Bronfenbrenner, 1958; Gecas, 1979; Gottfried & Gottfried,
1984; Kohn, 1975; 1977). Children from such backgrounds are at in-
creased risk for poor school performance and low educational attain-
ment (Garbarino & Asp, 1981), critical links to subsequent occupational
status and economic resources.
Parental job characteristics and work conditions have also been re-
lated to styles of parenting that are, themselves, likely to contribute to
the perpetuation of low-status employment across generations (Mor-
timer, 1974, 1976; Mortimer & Kumka, 1982; Piotrkowski & Katz, 1982;
see also Bronfenbrenner & Crouter, 1982). In particular, when parents
hold jobs that emphasize subordination to authority and offer little self-
162 JOAN I. VONDRA

direction, they tend to emphasize authoritarian values of compliance,

conformity, and physical punishment with their children at home.
For parents who experienced troubled relationships during child-
hood, the problems attendant with low SES may be exacerbated by
personal and interpersonal deficits that make it difficult to maintain
good job relations and consistent work performance. Job loss and eco-
nomic instability, circumstances likely to befall maltreating families (Sie-
gal, 1982; Steinberg, Catalano, & Dooley, 1981), may be especially detri-
mental to family functioning (Elder et al., 1984; Elder, Nguyen, & Caspi,
1985; Moen, Kain, & Elder, 1981). Increased irritability, arbitrary disci-
pline, conflict, and physical punishment are common parental re-
sponses to these economic stressors.
When economic difficulties are combined with impoverished social
relations, and both evolve in the context of a parent vulnerable to de-
pression or chemical dependency (from a history of troubled relations
with parents), and without positive parental role models, the prognosis
for parenting is poor. Thus, we may view sociological risk conditions as
both a cause and outcome of child maltreatment. Socioeconomic status
influences childrearing practices that help shape adult psychological and
sociological circumstances which, in turn, contribute to the quality of
parental care in the next generation. As Siegal (1982) noted, "Perhaps
there is no better single example of the importance of studying the
socioeconomic conditions underlying parent-child relations than child
abuse" (p. 16). Child maltreatment and family social and economic im-
poverishment go hand in hand (Garbarino & Sherman, 1980; Straus et
al., 1980; Vondra, 1986).

THE SOCIOCULTURAL MILIEU

Many have argued persuasively that maltreatment of children at the

rates observed in this country must, furthermore, be predicated on so-
ciocultural mores that condone the use of physical force and the view of
children as private property (Garbarino, Stocking, & Associates, 1980;
Gil, 1970; Talbot, 1976; Zigler, 1981). The notion that parents have the
right to rear children as they see fit, in the privacy of their home, is a
deeply rooted tradition in American history. Public scrutiny and, worse,
public intervention into private lives are almost universally frowned
upon. However, the combination of this zealous defense of family pri-
vacy with the belief that children are the property of their parents, opens
the way for child-rearing practices that victimize children (Garbarino,
1977b). In the context of lower-class attitudes about the nature of chil-
dren and child-rearing, the probability that childcare will drift into child
SOCIOLOGICAL AND ECOLOGICAL FACTORS 163

maltreatment increases proportionately. With social isolation, economic

hardship, parental psychological deficits, spousal conflict, and child be-
havior problems factored into the equation, whatever social forces exist
in support of children are unlikely to provide them with adequate pro-
tection, in light of "private property" and "physical force" principles of
childrearing. Ignoring "macrosystem" forces such as these, which are
filtered down to families from larger social and cultural institutions,
denies a significant and pervasive ecological contribution to the quality
of parenting and of family life (Bronfenbrenner, 1977; Garbarino, 1977a).

SUMMARY
Given the multiple and interacting factors that contribute to inade-
quate and abusive parental practices, it is apparent that anyone parent-
ing outcome may be the result of very different patterns of resources and
stressors. A single, adolescent, black mother raising two children in her
mother's small, inner-city apartment-already shared by married and
unmarried siblings-may display some of the same forms of maltreat-
ment as the wife of an abusive husband, enlisted in the military, who
moves his wife and four children from base to base in rapid succession.
In each case, different ecological factors are likely to be contributing to
family dynamics. In both cases, however, the balance of supportive
resources to undermining stressors is insufficient to sustain adequate
parental care.
Crockenberg (1987) found that it was only when low-SES mothers
felt rejected by their own mothers in childhood and reported a lack of
support from their partners that they acted in an angry, punitive way
toward their toddlers. Carroll (1977) reported that it was the combination
of low family warmth and high parental punishment during childhood
that predicted family violence in adult clinic patients. Egeland, Breiten-
bucher, and Rosenberg (1980) noted that high life event stress predicted
child maltreatment only when it was experienced by mothers who exhib-
ited a cluster of personality deficits indicating poor coping skills . And
Sameroff and Chandler (1975) have demonstrated that prematurity only
predicts developmental delay and/or disability when child-rearing oc-
curs in the context of family social and economic impoverishment.
Belsky (1984; Belsky & Vondra, 1989) conceptualized this balance in
terms of "stresses and supports," describing parenting as a "buffered"
system. Cicchetti and Rizley (1981) referred to "potentiating and com-
pensatory factors" of short- and long-term duration that jointly contrib-
ute to parenting outcomes. Both argue that maltreatment is not the
164 JOAN 1. VONDRA

outcome of any single condition or circumstance. Rather, it evolves

when the factors working to ameliorate the impact of risk conditions are
insufficient to overcome the cumulative effects of those stressors. In fact,
Belsky assembles data in support of his argument that the quality of
parental care is directly proportional to the relative balance of stresses to
supports. As stresses increase, parenting quality is increasingly jeopar-
dized, to the point where inadequate care becomes active maltreatment.
The ecology of child maltreatment, in other words, may be different
only quantitatively-not qualitatively-from the ecology of adequate
and superior childcare. In theory, then, prevention, intervention, and
enhancement services for parents should be sensitive to many of the
same ecological issues, because they operate for all parents-as risk
conditions for some, and as supportive resources for others. In any case,
these services should all be implemented in recognition of the full spec-
trum of stresses and supports that will be operating in each individual
case .
Providing quality parental care is a great enough challenge in and of
itself. Providing such care without having experienced it oneself, with-
out the self-confidence, self-esteem, and adaptability such a role de-
mands, without a loving spouse to share the burden, without the
knowledge and skills to overcome difficult child characteristics, without
supportive family and friends to help out when the path grows bumpy,
without the educational and economic resources to ensure financial sta -
bility, and in the context of a society that sanctions violent solutions to
conflict, "ownership" of children, and privacy before safety, requires
more than can be expected of any individual. Child maltreatment, in
this sense, is an overdetermined outcome. So many factors conspire
against adequate parenting at this point that the only question left un-
answered is how the system failed to intervene earlier. This is the ques-
tion that risk research is now seeking to answer.

REFERENCES
Ainsworth, M. D. S., Blehar, M. C,; Waters, E., & Wall, S. (1978). Patterns of attachment.
Hillsdale, NJ: Lawrence Erlbaum .
Altemeier, W. A., Vietze, P. M., Sherrod, K. A., Sandler, H. M., Falsey, S., & O'Connor, S.
M. (1979). Prediction of child maltreatment during pregnancy. Journal of the American
Academyof Child Psychiatry, 18, 205-218.
Alteme ier, W. A. , O'Connor, S., Vietze, P. M., Sandler, H. M., & Sherrod, K. B. (1982).
Antecedents of child abu se . Journal of Pediatrics, 100, 823-829.
Aragona, J. A., & Eyberg, S. M. (1981). Neglected children: Mothers' report of child
behavior problems and observ ed verbal behavior. Child Development, 52, 596-602.
SOCIOLOGICAL AND ECOLOGICAL FACTORS 165

Barahal, R. M., Watennan, J., & Martin, H . P. (1981). The social cognitive development of
abused children. Journal of Consulting and Clinical Psychology, 49, 508-516.
Belsky, J. (1984). The determinants of parenting: A proc ess model. Child Development, 55,
83-96.
Belsky, J., & Nezworski, T. (Eds.) . (1988). Clinical implications of attachment. Hillsdale, NJ:
Lawrence Erlbaum.
Belsky, J., & Vondra, J. (1989). Lessons from child abuse: The determinants of parenting. In
D. Cicchetti & V. Carlson (Eds.), Child maltreatment: Research and theory on the conse-
quences of abuse and neglect (pp. 153-202). New York: Cambridge University Press.
Billings, A. G., & Moos, R. H. (1983). Comparison of children of depressed and non-
depressed parents: A social-environmental perspective. Journal of Abnormal Child Psy-
chology, 11, 463-486.
Block, J. H., & Block, J. (1980). The role of ego-control and ego-resiliency in the organiza-
tion of behavior. In W. A. Collins (Ed.), Minnesota Symposia on Child Psychology (Vol.
13). Hillsdale, NJ: Lawrence Erlbaum.
Bradley, R. H ., & Caldwell, B. M. (1984). 174 children: A study of the relationship between
home environment and cognitive development during the first five years. In A. W.
Gottfried (Ed.), Homeenvironment and early cognitive development. New York: Academic
Press.
Bretherton, I. (1985). Attachment theory: Retrospect and prospect. Monographs of the Society
for Research in Child Development, 50 (1-2, Serial No . 209).
Bromet, E. J., & Comely, P. J. (1984). Correlates of depression in mothers of young chil-
dren. Journal of the American Academy of Child Psychiatry, 23, 335-342.
Bronfenbrenner, U. (1958). Socialization and social class through time and space. In E. E.
Maccoby, T. M. Newcomb, & E. L. Hartley (Eds.), Readings in social psychology. New
York: Henry Holt .
Bronfenbrenner, U. (1977). Toward an experimental ecology of human development. Amer-
ican Psychologist, 32, 513-530.
Bronfenbrenner, U., & Crouter, A. C. (1982). Work and family through time and space. In
S. B. Kammennan & C. D. Hayes (Eds.), Families that work: Children in a changing world.
Washington, DC: Nati onal Academy Press, 1982.
Brunnquell, D., Crichton, L., & Egeland, B. (1981). Maternal personality and attitude in
disturbances of child rearing. American Journal of Orthopsychiatry, 51, 680-691 .
Burgess, R. L., & Conger, R. D. (1978). Family interaction in abusive, neglectful, and
normal families . Child Development, 49, 1163-1173.
Burgess, R. L., Anderson, E. S., Schellenbach, C. J., & Conger, R. D. (1981). A social
interactional approach to the study of abusive families . In J. P. Vincent (Ed.), Advances
in family intervention, assessment, and theory: An annual compilation of research (Vol. 2).
Greenwich, CT: JAI Press.
Carroll, J. C. (1977). The intergenerational transmission of family violence: The long-term
effects of aggressive behavior. In Aggressive Behavior (Vol. 3, pp . 289-299). Alan R. Liss,
Inc.
Cicchetti, D., & Olsen, K. (1990). The developmental psychopathology of child maltreat-
ment. In M. Lewis & S. Miller (Eds .), Handbook of developmental psychopathology. New
York: Plenum Press.
Cicchetti, D., & Rizley, R. (1981). Developmental perspectives on the etiology, intergenera-
tional transmission, and sequelae of child maltreatment. In R. Rizley & D. Cicchetti
(Eds .), Developmental perspectives on child maltreatment. San Francisco: Iossey-Bass .
Cochran, M., & Brassard, J. (1979). Child development and personal social networks. Child
Development, 50, 601-616.
166 JOAN I. VONDRA

Colletta, N . D. (1983). At risk for depression: A study of young mothers. Journal of Genetic
Psychology, 142, 301-310.
Cotterell, J. L. (1986). Work and community influences on the quality of childrearing. Child
Development, 57, 362-374.
Crittenden, P. M. (1981). Abusing, neglecting, problematic, and adequate dyads: Differ-
entiating by patterns of interaction. Merrill-Palmer Quarterly, 27, 201-218.
Crittenden, P. M. (1985). Social networks, quality of childrearing, and child development.
Child Development, 56, 1299-1313.
Crittenden, P. M., & Ainsworth, M. D. S. (1989). Attachment and child abuse. In D.
Cicchetti & V. Carlson (Eds .), Childmaltreatment: Research and theory on the consequences
of abuse and neglect (pp . 432-463). New York: Cambridge University Press.
Crnic, K. A., Greenberg, M. T., Ragozin, A. S., Robinson, N. M., & Basham , R B. (1983).
Effects of stress and social support on mothers and premature and full-term infants.
Child Development, 54, 209-217.
Crnic, K A., Greenberg, M. T., Robinson, N . M. , & Ragozin, A. S. (1984). Maternal stress
and social sup port: Effects on the mother-infant relationship from birth to eighteen
months. American Journal of Orthopsychiatry, 54, 224-234.
Crockenberg, S. (1987). Predictors and correlates of anger toward and punitive control of
toddlers by adolescent mothers. Child Development, 58, 964-975.
Cummings, E. M., Zahn-Waxler, C,; & Radke-Yarrow, M. (1981). Young children's re-
sponses to expressions of anger and affection by others in the family. Child Develop-
ment, 52, 1274-1282.
Dietrich, K N., Starr, R H., & Kaplan, M. G. (1980). Maternal stimulation and care of
abused infants. In T. M. Field, S. Goldberg, D. Stern, & A. M. Sostek (Eds .), High-risk
infants and children: Adult and peer interactions. New York: Academic Press.
DiLalla, D. L., & Crittenden, P. M. (1987). A factor analytic study of maltreated children's home
behavior. Research presented at the Biennial Meeting of the Society for Research in
Ch ild Development, Baltimore , April.
Dubow, E. F., Huesmann, L. R, & Eron , L. D. (1987). Childhood correlates of adult ego
development. Child Development, 58, 859-869.
Egeland, B., & Brunnquell, D. (1979). An at-risk approach to the study of child abuse:
Some preliminary findings. Journal of the American Academy of ChildPsychiatry, 18, 219-
235.
Egeland, B., Breitenbucher, M., & Rosenberg, D. (1980). Prospective study of the signifi-
cance of life stress in the etiology of child abuse. Journal of Consulting and Clinical
Psychology, 48, 195-205.
Egeland, B., Sroufe , L. A., & Erickson , M. (1983). The developmental consequence of
different patterns of maltreatment. Child Abuse and Neglect, 7, 459-469.
Elder, G. H ., [r., Liker, J. K, & Cross, C. E. (1984). Parent-child behavior in hard times and
the life course: A multi-generational perspective. In P. Baltes & B. Brim (Eds.), Lifespan
development and behavior (Vol. 6). New York: Academic Press .
Elder, G. H ., [r., Nguyen, T. V., & Caspi, A. (1985). Linking family hardship to children's
lives. Child Development, 56, 361-375.
Elder, G. H ., [r., Caspi , A., & Downey, G. (1986). Problem behavior and family rela-
tionships: A multigenerational analysis. In A. Sorensen, F. Weinert, & L. Sherrod
(Eds .), Human development and the life course. Hillsdale, NJ: Lawrence Erlbaum.
Engfer, A., & Gavranidou, M. (1985, January). Antecedents and consequences of observed
maternal sensitivity: A longitudinal study. Paper presented at the International Sym-
posium on Psychobiology and Early Development, Free University, Berlin, West
Germany.
SOCIOLOGICAL AND ECOLOGICAL FACTORS 167

Engfer, A., & Cavranidou, M. (1988, June) . Prospective identification of violent mother-child
relationships . Paper presented at the Third European Conference on Developmental
Research, Budapest.
Engfer, A., & Schneewind, K. A. (1982). Cause s and consequences of harsh parental
punishment. Child Abuse and Neglect, 6, 129-139.
Erickson, M. F., Sroufe, L. A., & Egeland, B. (1985). The relationship between quality of
attachment and behavior problems in preschool in a high-risk sample. Monographs of
the Society for Research in Child Development, 50 (Serial No . 209).
Estroff, T. W., Herrera, C,; Gaines, R., Shaffer, D., Gould, M., & Green, A. H. (1984).
Maternal psychopathology and perception of child behavior in psychiatrically referred
and child maltreatment families . Journal of the American Academy of ChildPsychiatry, 23,
649-652 .
Field, T. M. (1984). Follow-up developmental status of infants hospitalized for nonorganic
failure to thrive. Journal of Pediatric Psychology, 9, 241-256 .
Fraiberg, S. (1980). Clinical studies in infant mental health: The first year. New York: Basic
Books.
Fraiberg . S., Adelson, E., & Shapiro, V. (1975). Ghosts in the nursery: A psychoanalytic
approach to the problems of impaired infant-mother relationships. Journal of the Ameri-
can Academy of Child Psychiatry, 14, 387-421.
Frodi, A. M., & Lamb, M. E. (1980). Child abusers' responses to infant smiles and cries.
Child Development, 51, 238-241.
Gaensbauer, T. [., & Sands, S. K. (1979). Distorted affective communications in abused/
neglected infants and their potential impact on caretakers. Journal of the American
Academy of Child Psychiatry, 18, 236-250.
Garbarino, ]. (1977a). The human ecology of child maltreatment: A conceptual model for
research. Journal of Marriage and the Family, 39, 721-727 .
Garbarino, ]. (1977b). The price of privacy in the social dynamics of child abuse. Child
Welfare, 56, 565-575.
Garbarino, ]., & Asp, E. (1981). Successful schools and competent students. Lexington, MA:
Lexington Books.
Garbarino, L, & Gilliam, G. (1980). Understanding abusive families. Lexington, MA: D.C.
Heath.
Garbarino, ]., & Sherman, D. (1980). High-risk neighborhoods and high-risk families : The
human ecology of child maltreatment. Child Development, 51, 188-198 .
Garbarino, j., Stocking, S. H., & Associates (1980). Protecting children from abuseand neglect:
Developing and maintaining effective support systems for families . San Francisco, CA:
]ossey-Bass .
Garbarino, j., & Vondra,]. (1987). Psychological maltreatment of children and youth. In M.
A. Brassard, R. Germain, & S. N. Hart (Eds.), The psychological maltreatment of children
and youth (pp . 25-44) . New York: Pergamon Press .
Gecas, V. (1979). The influence of social class on socialization . In W. R. Burr, R. Hill, F. I.
Nye, & I. L. Reiss (Eds .), Contemporary theories about thefamily (Vol. 1). New York: Free
Press.
George, C,; & Main , M. (1979). Social interactions of young abused children: Approach,
avoidance, and aggression. Child Development, 50, 306-318 .
George, c., & Main, M. (1980). Abused children: Their rejection of peers and caregivers. In
T. M. Field (Ed.), High-risk infants and children: Adult and peer interactions. New York:
Academic Press .
Gil, D. (1970). Violence against children: Physical childabuse in the United States. Cambridge,
MA: Harvard University Press.
168 JOAN 1. VONDRA

Gilbreath, B., & Cicchetti, D. (1990). Psychopathology in maltreating mothers. Manuscript

submitted for publication.
Goldberg, W. A., & Easterbrooks, M. A. (1984). Role of marital quality in toddler develop-
ment. Developmental Psychology, 20, 504-514 .
Gottfried, A. W., & Gottfried, A. E. (1984). Home environment and mental development in
young children. In A. W. Gottfried (Ed .), Home environment and early mental develop-
ment: Longitudinal research . New York: Academic Press .
Green, A. (1978). Psychopathology of abused children. Journal of the American Academy of
Child Psychiatry, 17, 92-103.
Greenspan, S. 1., Wieder, S., Nover, R., Lieberman, A., Lourie, R., & Robinson, M. G.
(Eds.) (1987). Infants in multirisk families . Madi son, WI: International Universities
Press.
Gwartney-Gibbs, P. A., Stockard, J., & Bohmer, S. (1987). Learning courtship aggression:
The influence of parents, peers, and personal experiences. Family Relations, 36, 276-
282.
Hart, S. N., & Brassard, M. R. (1987). A major threat to children's mental health: Psycho-
logical maltreatment. American Psychologist, 42, 160-165.
Hazan, c., & Shaver, P. (1987). Romantic love conceptualized as an attachment process.
Journal of Personality and Social Psychology, 52, 511-524.
Hoffman-Plotkin, D., & Twentyman, C. T. (1984). A multimodal assessment of behavioral
and cognitive deficits in abused and neglected preschoolers. Child Development, 55,
794-802.
Hops, H ., Sherman, L., & Biglan, A. (1990). Maternal depression, marital discord, and
children's behavior: A developmental perspective. In G. R. Patterson (Ed .), Depression
and aggression: Two facets of family interactions. Hillsdale, NJ: Lawrence Erlbaum.
Howes, C,; & Espinosa, M. P. (1985). The consequences of child abuse for the formation of
relationships with peers. Child Abuseand Neglect , 9, 397-404.
Kalmuss, D. (1984). The intergenerational transmission of marital aggression. Journal of
Marriage and the Family, 46, 11-19.
Kaplan, S. J., Pelcovitz, D., Salzinger, S., & Caneles, D. (1983). Psychopathology of parents
of abused and neglected children and adolescents. Journal of the American Academy of
Child Psychiatry, 22, 238-244.
Kaufman, J., & Cicchetti, D. (1989). The effects of maltreatment on school-aged children's
socioemotional development: Assessments in a day camp setting. Developmental Psy-
chology, 25, 516-524.
Kaufman, J., & Zigler, E. (1990). The intergenerational transmission of child abuse and the
prospect of predicting future abusers. In D. Cicchetti & V. Carlson (Eds.), Child mal-
treatment: Research and theory on the consequences of child abuse and neglect (pp. 129-150).
Cambridge, MA: Harvard University Press.
Kobak, R. R., & Sceery, A. (1988). Attachment in late adolescence: Working models, atfect
regulation, and representations of self and others. Child Development, 59, 135-146.
Kohn, M. L. (1975). Social class and parent-child relationships : An interp retation. In U.
Bronfenbrenner & M. Mahoney (Eds .), Influences on human development. Hinsdale, IL:
Dryden Press.
Kohn, M. L. (1977). Class and conformity: A study in values. Chicago, IL: University of
Chicago Press.
Kotelchuck, M. (1982). Child abuse and neglect: Prediction and misclassification. In R. H .
Starr, Jr. (Ed.), Childabuse prediction. Cambridge, MA: Ballinger.
Levitt, M. J., Weber, R. A., & Clark, M. C. (1986). Social network relationships as sources
of maternal support and well-being. Developmental Psychology, 22, 310-316.
SOCIOLOGICAL AND ECOLOGICAL FACTORS 169

Lewis, M., Feiring, C.; McGuffog, C,; & jaskir, J. (1984). Predicting psychopathology in six-
year-olds from early social relations. Child Development, 55, 123-136.
Lyons-Ruth, KI ., Connell, D. B., Zoll, D., & Stahl, J. (1987). Infants at social risk: Relations
among infant maltreatment, maternal behavior, and infant attachment behavior. De-
velopmental Psychology, 23, 223-232.
Main, M., & Goldwyn, R. (1984). Predicting rejection of her infant from mother's represen-
tation of her own experience: Implications for the abused-abusing intergenerational
cycle. Child Abuse and Neglect, 8, 203-217.
Main, M., & Goldwyn, R. (1985). Adult attachment classification system. Unpublished manu-
script, University of California, Berkeley.
Mash, E. J., Johnston, C.; & Kovitz, K. (1983). A comparison of the mother-child interac-
tions of physically abused and non-abused children during play and task situations.
Journal of Clinical Child Psychology, 12, 337-346.
Melnick, B., & Hurley, J. (1969). Distinctive personality attributes of child-abusing moth-
ers . Journal of Consulting and Clinical Psychology, 33, 746-749.
Moen, P., Kain, E. L., & Elder, G. H. , Jr. (1981). Economic conditions andfamilylife: Contempo-
rary and historical perspectives. Paper presented for the National Academy of Sciences
Assembly of Behavioral and Social Sciences, Committee on Child Development Re-
search and Public Policy, December.
Mortimer, J. T. (1974). Patterns of intergenerational occupational movements: A smallest
space analysis. American Journal of Sociology, 79, 1278-1299.
Mortimer, J. T. (1976). Social class, work, and the family : Some implications of the father's
career for familial relationships and sons' career decisions. Journal of Marriage and the
Family, 38, 241-256.
Mortimer, J. T., & Kumka, D. (1982). A further examination of the "occupational linkage
hypothesis." The Sociological Quarterly, 23, 3-16.
Newberger, E. H ., Reed, R. B., Daniels, J. H., Hyde, J. N ., & Kotelchuck, M. (1977).
Pediatric social illness: Toward an etiological classification. Pediatrics, 60, 175-185.
Pianta, R. C.; Egeland, B., & Hyatt, A. (1986). Maternal relationship history as an indicator
of developmental risk . American Journal of Orthopsychiatry, 56, 385-398.
Piotrkowski, C. S., & Katz, M. H . (1982). Indirect socialization of children: The effects of
mothers' jobs on academic behaviors. Child Development, 53, 1520-1529.
Provence, S., & Lipton, R. (1962). Infants in institutions. New York: International Univer-
sities Press.
Ricks, M. H. (1985). The social transmission of parental behavior: Attachment across gener-
ations. Monographs of the Society for Research in Child Development, 50 (1-2, Serial No .
209).
Rohner, R. P., & Rohner, E. C. (1980). Antecedents and consequences of parental rejection:
A theory of emotional abuse. Child Abuseand Neglect, 4, 189-198.
Rosario, M., Salzinger, S., Feldman, R., & Hammer, M. (1987, April) . Home environments of
physically abused andcontrol school-age children . Research presented at the Biennial Meet-
ing of the Society for Research in Child Development, Baltimore.
Rosenbaum, A., & O'Leary, D. (1981). Marital violence: Characteristics of abusive couples.
Journal of Consulting and Clinical Psychology, 49, 63-71.
Rutter, M., Quinton, D., & Liddle, C. (1983). Parenting in two generations: Looking
backwards and looking forwards . In N . Madge (Ed.), Families at risk (pp . 60-98).
London: Heinemann.
Sameroff, A. J. (1975). Early influences on development: Fact or fancy? Merrill-Palmer
Quarterly, 21, 267-294.
Sameroff, A. J. & Chandler, M. (1975). Reproductive risk and the continuum of caretaking
170 JOAN 1. VONDRA

casualty. In F. Horowitz, E. Hetherington, & G. Siegel (Eds.), Reviewof childdevelop-

ment research (Vol. 4). Chicago, IL: University of Chicago Press.
Sandgrund, A., Gaines, R. W., & Green, A. H. (1974). Child abuse and mental retardation:
A problem of cause and effect. American Journal of Mental Deficiency, 79, 327-330.
Sherrod, K. B., O'Connor, S., Vietze, P. M., & Alterneier, W. A. (1984). Child health and
maltreatment. Child Development, 55, 1174-1183.
Siegal, M. (1982). Economic deprivation and the quality of parent-child relations. In Fair-
ness in children. New York: Academic Press.
Spinetta, J. J. (1978). Parental per sonality factors in child abuse. Journal of Consulting and
Clinical Psychology, 46, 1409-1414.
Sroufe , L. A. (1983). Infant-caregiver attachment and patterns of adaptation in preschool:
The roots of maladaptation and competence. In M. Perlmutter (Ed.), Minnesota Sym-
posia in Child Psychology (Vol. 16). Hillsdale, NJ: Lawrence Erlbaum .
Sroufe , L. A., & Pleeson, J. (in press). Attachment and the construction of relationships. In
W. Hartup & K. Rubin (Eds.), The natureand development of relationships. Hillsdale, NJ:
Lawrence Erlbaum.
Starr, R. H ., Jr. (1982). A research-based approach to the prediction of child abuse. In R. H .
Starr, Jr. (Ed.), Child abuse prediction. Cambridge, MA: Ballinger.
Starr, R. H. , [r., Dietrich, K. N. , Fischhoff, J., Ceresnie, S., & Zweier, D. (1984). The
contribution of handicapping conditions to child abuse. Topics in Early Childhood Special
Education, 4, 55-69.
Steinberg, L., Catalano, R., & Dooley, D. (1981). Economic antecedents of child abuse and
neglect. Child Development, 52, 975-985.
Stevens, J. H . (1988). Social support, locus of control, and parenting in three low-income
groups of mothers: Black teenagers, black adults, and white adults. ChildDevelopment,
59, 635-642.
Strau s, M. A., Gelles, R. J., & Steinmetz, S. K. (1980). Behind closed doors: Violence in the
American family. New York: Anchor Press .
Talbot, N. E. (Ed.). (1976). Raising children in modern America: What parents and society should
be doing for their children. Boston: Little, Brown.
Trickett, P. K., & Kuczynski, L. (1986). Children's misbehaviors and parental discipline
strategies in abusive and nonabusive families. Developmental Psychology, 22, 115-123.
Trickett, P. K., & Susman, E. J. (1988). Parental perceptions of childrearing practices in
physically abu sive and nonabusive families. Developmental Psychology, 24, 270-276.
Tronick, E. Z., & Field, T. M. (Eds.) (1986). Maternal depression and infant disturbance. San
Francisco: Iossey-Bass.
Vondra , J. 1. (1986). Socioeconomic considerations: The family economy. In J. Garbarino &
Associates, Troubled youth, troubled families. New York: Aldine .
Watson, D., & Clark, L. A. (1984). Negative affectivity: The disposition to experience
aversive emotional states. Psychological Bulletin, 96, 465-490.
Weinraub , M., & Wolf, B. M. (1983). Effects of stress and social supports on mother-child
interaction in single- and two-parent families. Child Development, 54, 1297-1311.
Whittaker, J. K. & Garbarino, J. (1983). Social support networks: Informal helping in the human
services. New York: Aldine .
Zigler, E. (1981). Controlling child abuse: Do we have the knowledge or the will? In G.
Gerbner, K. Ross, & E. Zigler (Eds .), Child abuse: An agenda for action. New York:
Oxford University Press .
 CHAPTER 7

PARENTAL PSYCHOPATHOLOGY
AND HIGH-RISK CHILDREN
DAVID C. FACTOR AND DAVID A. WOLFE

INTRODUCTION
Interest in the relationship between parental behavior and children's
developmental outcome has long been a fundamental concern of social
scientists. This is a fact regardless of the direction of a person's basic
developmental orientation-genetic or environmental-because the
two orientations place significant weight on parental characteristics and
influence. Furthermore, it is now widely recognized that parents and
children influence each other in a reciprocal fashion (e.g ., Bell & Harper,
1977), rather than in a unidirectional, parent-to-child fashion as was
originally assumed by many investigators. Understanding of the man-
ner in which parental psychological characteristics can affect the devel-
oping child's emotional and behavioral adjustment has grown immense-
ly over the past decade, and these recent findings form the foundation
for the discussion in this chapter.
Numerous explanations have been given as to why children of psy-
chologically disturbed parents may be affected by their parents' psycho-
pathology. These arguments range from the heritability of specific disor-

DAVID C. FACTOR· TRE-ADD Program, Thistledown Regional Center for Children and
Adolescents, Rexdale, Ontario, Canada M9V 4L8. DAVID A. WOLFE· Department of
Psychology, University of Western Ontario, London, Ontario, Canada N6A 5B8.

171
172 DAVID C. FACTOR and DAVID A. WOLFE

ders and genetic predispositions to the environmental contingencies

and learning opportunities made available throughout the child's devel-
opment. Because of the significance to children's mental and early inter-
vention programming, our attention is primarily on the recent findings
from the study of subsamples of parents who have been diagnosed with
a specific psychiatric disturbance (e.g., affective disorder) or an identi-
fiable adjustment problem (e.g., child abuse). We assume that children
may be affected by their parent's psychopathology on the basis of factors
that may directly and indirectly impair or limit the course of a child's
normal development. These factors include, for example, the well-docu-
mented effects of parental modeling of inappropriate behavior, the ef-
fects of the individual's pathology on the marital relationship and family
functioning, and the direct and indirect effects resulting from inconsis-
tent or inappropriate child-rearing methods and unpredictability in
mood or behavior.
Our assumption is that children of parents who are suffering from
psychological disturbances or who exhibit extreme behavior in the pres-
ence of the child are vulnerable and at-risk of developing mental health
problems. Such problems are caused, at least in part, to the important
influences of modeling, marital conflict, family dysfunction, and inap-
propriate child -rearing methods. Accordingly, various forms of parental
pathology can be viewed as significant "risk factors" that contribute to a
higher probability for the development of a disorder (Masten & Gar-
mezy, 1985). In fact, longitudinal research has shown that mental illness
and substance abuse in parents increase the risk of impairment in their
offspring, and children who accumulate four or more risk factors fre-
quently exhibit learning and behavioral problems, delinquency, and/or
psychiatric disturbance (Rutter, 1979; Werner, 1989).

CHILD-REARING PATTERNS AND THEIR SUSPECTED

INFLUENCE ON CHILD DEVELOPMENT
Social learning theory provides a sound and widely supported the-
oretical basis for explaining the preponderance of correlational data link-
ing disorders in parents to increased developmental risk in their
offspring. This theory, which accomodates behavioral, emotional, cog-
nitive, and physiological mechanisms of behavior acquisition and main-
tenance, places a heavy emphasis on children's abilities to learn a con-
tingency, or relationship, between their actions and consequences. Once
children learn the association between their behavior and the conse-
quences for that behavior, they form a cognitive understanding of the
PARENTAL PSYCHOPATHOLOGY AND HIGH-RISK CHILDREN 173

probability, or expectation, of such consequences in choosing to engage

in the behavior in the future.
Therefore, according to social learning theory, children of disturbed
parents may be at-risk of developmental deviation or delay, at least in
part, because of their exposure to inappropriate learning opportunities
(e.g., witnessing seemingly positive consequences for parental violence
or criminal behavior), the lack of consistent, appropriate consequences
for normal variations in child behavior (e.g., harsh parental punishment
for minor transgressions, curiosity, and distorted expectations of them-
selves and others. Because a parent or caregiver exerts the major influ-
ence on the developing child's sense of self and of others, awareness of
moral values, and the learning of basic social and self-control skills (to
name but a few), deviant socialization practices can find a number of
possible routes by which to disrupt the child's ongoing development.
One of the more prominent routes or mechanisms by which paren-
tal behavior is believed to influence child development directly is
through variations in family socialization practices, or "parenting
styles." Variations in socialization practices occur normally in relation to
child, family, and situational events. Yet, the erroneous assumption is
often made that there are two "types" of parents: those who naturally
possess the motivation to raise their children in a positive, supportive
fashion, and those who presumably do not possess this desire or ability
(a conclusion often based on a small, but significant, aspect of their
child-rearing, such as abuse, neglect, or rejection). Parents who fall into
this latter group are labeled "abnormal" or "disturbed," and a false
dichotomy is easily formed between "good" and "bad" parenting. Alter-
natively, parenting practices can be viewed along a continuum, whereby
one extreme represents practices that are severe and potentially harmful
to the child, and at the other extreme are methods that promote many
aspects of the child's development. Viewing parental behavior toward
the child along a continuum, rather than a dichotomy, helps to draw
attention to the fact that even "disturbed" or "abusive" parents may
highly resemble "typical" child-rearing practices except in terms of their
degree or frequency of engaging in some of the more extreme methods
(Wolfe, 1987).
The mechanisms by which parenting style may influence child de-
velopment are represented by the intersection of two fundamental di-
mensions of parenting: demandingness and responsiveness. Deman-
dingness is described as the amount or degree of control the parent
attempts to exert over the child, whereas responsiveness is defined as
the frequency of interactions with the child (both positive and negative)
that are child-centered versus parent-centered (Maccoby & Martin,
174 DAVID C. FACTOR and DAVID A. WOLFE

1983). Throughout the discussion of parental disorders that follows,

regardless of the specific topology of the disorder (i.e., depression, crim-
inal behavior, immaturity), a common element that emerges is the extent
to which each disorder is associated with an impairment of the parent's
age-appropriate demands on the child (i.e., demands for mature behav-
ior, independence, and clear communication) and the parent's sen-
sitivity or responsiveness to the child's capabilities.
As a point of reference, the most healthy, appropriate parenting
style is one in which parents are both demanding and child-centered in
their responsiveness, a style that is referred to as authoritative. This style
is considered to be most effective in terms of desirable developmental
outcome and in reducing parent-child conflict, because authoritative
parents place age-appropriate demands on their children, are more con-
sistent in their discipline, and rely on a wide choice of techniques to
teach their children (Baumrind, 1971). In contrast to this healthy parent-
ing style, three other styles can be identified on the basis of the degree of
demandingness versus responsiveness that the parent displays toward
the child, each of which has been associated with less desirable develop-
mental outcomes for the child (Maccoby & Martin, 1983). For example,
parents who are demanding of their child but who are rejecting or
unresponsive to their child's needs have been labeled "authoritarian";
those who demand little of their children yet are extremely child-cen-
tered are referred to as "indulgent"; and those who are both undemand-
ing as well as unresponsive to their children are described as "neglect-
ing." Thus, the mechanisms whereby parental disorders influence child
development can be understood in terms of the extent to which a given
disorder affects parents' demandingness and responsiveness towards their
child. The resultant disruption in the balance of these two important
dimensions is subsequently reflected in parents' socialization practices,
often with negative consequences to their child's development. In the
following discussion of parental disorders and their relationship to
negative developmental changes, we highlight the advances that have
been made in recent years in our understanding of these processes.

PARENTAL DISORDERS AND CHILD ADJUSTMENT:

AN OVERVIEW OF THE LITERATURE
Based on the above argument linking children's developmental risk
status to parental and family functioning, some discussion is warranted
of several of the more well-known situational conditions that have been
associated with children's adjustment. In reviewing these relationships,
PARENTAL PSYCHOPATHOLOGY AND HIGH-RISK CHILDREN 175

it should be kept in mind that none of these identified conditions or

situations has been found to be fully or even largely responsible for a
given child's developmental outcome; rather each serves as a "marker"
variable that may indicate a significant factor that has been shown to
influence the course of development. Because of the immense literature
in this area, only psychological phenomena affecting developmental
progress will be treated (interested readers may wish to pursue other
studies dealing with biological, genetic, and sociological factors related
to parental adjustment and child development). Accordingly, after an
overview of the more significant psychological factors, a more in-depth
look at children of depressed parents will illustrate the major implica-
tions that are currently emerging from this field.

PARENTAL IMMATURITY

Although not a specific psychiatric disorder per se, parental imma-

turity (i.e., a young adult's development of independent living skills,
cognitive problem-solving, or emotional development) has long been
suspected to play a role in the developmental status of the offspring.
Specifically, children of adolescent parents have been described as being
more at-risk for developmental and behavioral disorders than children
of adult parents, because of the inability of the adolescent parent to meet
the demanding needs, consistently or effectively, of a young infant or a
child. Investigations of this relationship between adolescent parents and
developmental risk status in their offspring have confirmed that these
children display more cognitive, emotional, and ph ysical problems than
do children of more mature parents (Phipps-Yonas, 1980). In particular,
infants of adolescent parents have more physical health problems and
higher mortality rates when mothers were younger than 15 years of age.
Moreover, there is a high (negative) correlation between the rate of child
abuse and the age of the mother when she gave birth to her first child
(Wolfe, 1987). Important situational factors, such as income and housing
stability, education, social and economic resources, conjugal violence,
and many others, playa role in impairing the child-rearing effectiveness
of the adolescent parent, which together lead to a significant increase in
the developmental risk to the infant (Schwartz, 1979).
Although maturity of the adolescent mother has been the primary
focus of attention to date, more effort must be directed as well toward an
improved understanding of the role of the father/partner in contributing
to developmental risk. Male immaturity in relation to child-rearing re-
sponsibility is commonly expressed as a pattern of poor job history,
violence (toward partner as well as in the community), substance abuse,
176 DAVID C. FACTOR and DAVID A. WOLFE

financial irresponsibility, and minimal contact or involvement with the

infant or toddler (Scott, Field, & Robertson, 1981). Although it is difficult
to establish an underlying causal mechanism for such patterns among
very young male and female parents, the notion of developmental im-
maturity (perhaps associated with previous lack of proper childcare and
family life in their own families of origin) seems to describe adequately
the existing relationship between adolescent parents and developmental
risk.

PARENTAL CRIMINALITY

Criminal involvement on the part of one or both parents has also

been identified as a significant developmental risk factor, although the
mechanisms by which such patterns are transmitted are poorly under-
stood. A recent study provides a clear example of data that support this
relationship between increased risk of antisocial acts in children and the
criminal history of the father, in particular. Kandel et al. (1988) examined
the sons of fathers who had engaged in criminal behavior that was
deemed to be "severely sanctioned" or the sons who were free from any
"registration from criminal behavior." These authors developed four
groups of offspring based on their classification of the fathers: (1) high-
risk, severely sanctioned; (2) high-risk, no registration; (3) low-risk, se-
verely sanctioned; and (4) low-risk, no registration. By matching these
samples on socioeconomic status, several hypotheses were confirmed.
Most significantly, sons with severely sanctioned fathers had an ele-
vated risk of criminal involvement that was 5.6 times greater than it was
for the low-risk sons. Interestingly, serious criminal offenders in the
youth sample evidenced lower IQ scores than did subjects who did not
engage in criminal behavior, although this finding was true only for
high-risk subjects. In fact, boys in the high-risk group who did not
engage in any criminal activity had relatively higher intelligence test
scores .

CHRONIC PARENTAL ILLNESS AND FAMILY FUNCTIONING

Chronic illness on the part of one or both parents has also been
associated with greater risk of developmental impairment. Blackford
(1988) argued that such children have a greater incidence of depression
and psychosomatic disorders, and that antisocial behavior and poor
school performance are common associated features. Interestingly,
females of chronically ill parents appear to be less noticeably affected by
this situation than are males, a finding that Blackford (1988) and Rutter
(1971) explained in terms of the development of increased sensitivity to
PARENTAL PSYCHOPATHOLOGY AND HIGH-RISK CHILDREN 177

others and the greater resistance to stress among girls than among boys.
Although it is accepted that chronic parental illness is linked with unde-
sirable developmental outcomes in the children, several authors do dif-
ferentiate between loss of a parent through natural events (i.e., death,
divorce) and the diminution of child-rearing responsibilities, as evi-
denced by chronic illness and other impairments affecting family func-
tioning; that is, a loss of the parent does not necessarily lead to impair-
ments in the child's development, but that developmental outcome can
be predicted more or less on the basis of the presence of other compen-
satory factors in the child's family, such as the support of family mem-
bers, availability of the other parent, and the absence of conflict and
discord in the child's presence (Emery, 1982; Hetherington & Martin,
1986).
Along these same lines, several researchers have looked more gen-
erally at family functioning and its relationship to children's develop-
ment. Drawing from Olson, Sprenkle, and Russell's (1979) work with
family cohesion and adaptability, Smets and Hartup (1988) explored the
family systems in which child behavior problems seem to emerge. "Co-
hesion" is defined as the "connectedness of relationships within the
family or the extent to which family members are 'bonded' to one an-
other," whereas adaptability refers to "the capacity of the family system
to change its power structure, role relations, and rules in response to
situational and developmental stress" (p. 239). According to family sys-
tems theory, a ''balanced'' family system is described as one in which
members are both moderately cohesive and adaptable, rather than fall-
ing at an extreme of one or both dimensions. In their review of previous
studies, Smets and Hartup (1988) highlighted the finding that families of
juvenile offenders often score more frequently in the extreme regions of
family functioning (i.e., showing either too high or too low cohesion
and/or adaptability). Similar results are found among families who have
referred their children to clinics because of problematic child behavior.
These researchers concluded, on the basis of previous work as well as on
the basis of their own study involving 120 clinic families, that extreme
forms of family functioning tend to be associated with abnormal child
behavior. These findings concur, as well, with the studies by Minuchin
(1974; Minuchin, Rossman, & Baker, 1978) involving psychosomatic
families.

CHILDREN OF PARENTS WITH PSYCHIATRIC ILLNESSES

In further exploring the relationships between parental disorders

and developmental deviations among their offspring, there is a vast
literature that focuses on specific psychiatric diagnoses. Although ad-
178 DAVID C. FACTOR and DAVID A. WOLFE

mittedly incomplete, Goldstein (1988) summarized the views of much of

the field when he stated that "the best risk marker for most mental
disorders is still the rather crude index of being an offspring of a parent
with that disorder" (p. 285). This familial link was explored in one of the
earliest studies conducted by Rutter (1966). Following the rationale that
children are affected by their parents' psychopathology on the basis of
the disturbance such disorders may create in family functioning as well
as in child behavior, Rutter's study demonstrated that children with
psychiatrically disturbed parents were more likely (than control chil-
dren) to suffer from a diagnosable disorder. Similar findings were re-
ported by Cooper, Leach, Storer, and Tonge (1977), in which the rate of
psychiatric disorders among children of adult patients was 45%, com-
pared to 26% in the control group of children from organically ill
parents.
Additional support for the additive influence of generic parental
psychopathology follows from a study by Cantwell and Baker (1984), in
which they compared subsamples of children with two healthy parents,
one psychiatrically ill parent, and two parents with psychiatric illness.
Their results indicated that when children had two disturbed parents,
there was a significantly greater probability that they would suffer devel-
opmental problems, in comparison to both of the other groups. In-
terestingly, no significant differences emerged between the two subsam-
ples in which one or both parents were healthy; however, children with
one or two ill parents experienced more psychosocial stress (e.g. ,
changes in schools, residence, income) than did the children of healthy
parents, which may partially account for the mechanisms whereby pa-
rental behavior disrupts child development. Furthermore, more behav-
ior problems were found among those children having two psychi-
atrically disturbed parents if one of the parents was diagnosed as having
an antisocial disorder (compared to any other diagnosis). Similar to the
findings presented above on children of criminal parents, Cantwell and
Baker (1984) stressed the significance of the relationship between devel-
opmental deviation and the presence of antisocial behavior in one or
both parents.
The sex of the parent may also play a role in the expression of
developmental problems in the offspring. Schore (1988) reported that in
families in which only one parent was psychiatrically disturbed, 47% of
the children had received psychiatric diagnoses when that parent was
the father, compared to 66% when the diagnosed parent was the moth-
er. Thus, with the exception of the diagnosis of antisocial behavior, the
mother's psychopathology was suspected to be even more influential
than that of the father.
PARENTAL PSYCHOPATHOLOGY AND HIGH-RISK CHILDREN 179

Although few studies have been conducted to date, the emerging

literature on children of anxiety-disordered parents merits some discus-
sion. Generally, these studies have focused on parents who exhibited
signs of panic disorder, generalized anxiety disorder, obsessive-com-
pulsive disorder, agoraphobia, or some specific phobic pattern. In an
exemplary study of this issue, Turner, Beidel, and Costello (1987) inves-
tigated 59 children between the ages of 7 and 12 years. Sixteen of these
children had a parent with a known anxiety disorder (agoraphobic or
obsessive-compulsive), 14 had a parent with a dysthymic or depressive
disorder, and 13 came from families with nonpsychiatrically ill parents.
The remaining 16 children were from families with no known psychi-
atric illness. Based on the Child Assessment Schedule, children who had
parents with anxiety disorders were significantly different from the nor-
mal control group on 9 of the 12 comparison areas: they had more
difficulties at school and in making friends, more specific fears and
worries, and experienced more symptoms of depression, anxiety, and
somatic complaints. However, only two differences emerged between
the children of anxiety-disordered parents and children of depressed
parents, with the former stating more problems at school and spending
more time in solitary activities than the children of depressed parents.
In addition, the above-named study found that 44% (7 out of 16) of
the children of anxiety-disordered parents met the criteria for a psychi-
atric diagnosis on the basis of test results, compared to 22% of the
children of depressed parents, 8% of the first control group, and none of
the normal control group. Hence, children of anxiety-disordered parents
were found to be more than two times as likely to have a DSM-III
disorder as the children of dysthymic parents, and were twice as likely
to show signs of an anxiety disorder. Similar findings, using other mea-
sures and samples, have confirmed a link between the presence of anx-
iety disorders in the parents and psychiatric disturbance (and symptoms
of anxiety) in the offspring (Silverman, Cerny, Nelles, & Burke, 1988;
Sylvester, Hyde, & Reichler, 1987).
In commenting on these findings relating various psychiatric prob-
lems in the parents to developmental abnormalities in the offspring, we
should note that the data are not conclusive or explanatory. More partic-
ularly, the data do not support the notion of specific, diagnostically
relevant problems being replicated in the offspring. The findings cer-
tainly support the belief that having two psychiatrically ill parents
means a greater likelihood that the child will receive a psychiatric diag-
nosis, and a somewhat diminished (but still clinically significant) like-
lihood of receiving a diagnosis if only one parent suffers from a distur-
bance. Yet, there is little evidence to demonstrate that having a parent
180 DAVID C. FACTOR and DAVID A. WOLFE

with a particular diagnosis implies that the child will have a similar or
predictable diagnostic pattern. Most of the literature to date exploring
specific psychiatric diagnostic categories has focused on children of ei-
ther thought-disordered parents (i.e., schizophrenic adults) or children
of parents with major affective disorders (i.e., uni- or bipolar depres-
sion). To illustrate the psychological processes involved, as well as the
developmental implications for the child, we now turn to the expansive
literature on children of depressed parents.

PARENTAL DEPRESSION AND ITS INFLUENCE

ON CHILD DEVELOPMENT
Adult depression is the most common psychiatric disorder, affect-
ing approximately 10% of the population. According to Anderson
(1987), 20% of women and 11% of men will experience a serious de-
pressive episode at some time in their lives . Because a depressive disor-
der usually affects self-esteem and behavior as well as mood, its impact
on family functioning in general, and childrearing in particular, has long
been suspected to be pervasive and severe (see Barnett & Gotlib, 1988,
for further discussion of adult depression).
Reviews of the literature examining the suspected link between
adult depression and childhood disorders began to emerge in the 1970s,
and have generally confirmed the existence of a correlational rela-
tionship. However, several additional findings have simultaneously
emerged from these studies, which have helped to clarify the nature and
extent of this relationship.
One of the first reviews of this literature was reported by Orvaschel,
Weissman, and Kidd (1980). Reviewing the major studies conducted at
the time, these investigators noted that the majority of children of de-
pressed parents evinced developmental abnormalities, such as depres-
sion, interpersonal problems, acting out behavior, and school and atten-
tional difficulties. Although the authors felt assured that having a
depressed parent led to greater risk of developmental deviation, they
also recognized that adequate control groups had rarely been used in
these early studies and that methods of assessment were quite varied
and inexact.
Shortly after this review was published, Sameroff, Seifer, and Zax
(1982) reported several significant findings from their longitudinal work
with children at-risk for schizophrenia and major psychiatric disorders.
Comparing infants and toddlers of a "neurotic depression" group with
normal controls, these researchers noted that at all ages studied (pre-
PARENTAL PSYCHOPATHOLOGY AND HIGH-RISK CHILDREN 181

natal through 30 months) the depressed mothers appeared more anx-

ious and lower in social competence, and their newborns had the lowest
obstetric status. At 4 months of age, depressed mothers were showing
less involvement with their newborns, although these differences were
less pronounced by 1 year. By the time the children were 30 months old,
they were found to have acquired fewer adaptive behaviors in the home,
and their mothers reported them to be "less cooperative with family
members, less cooperative with others, more bizarre, more depressed,
and more often engaged in imaginary play" (p. 43).
Orvaschel (1983) concluded that, in the aggregate, these early stud-
ies strongly suggest a causal relationship between parents with depres-
sion and problems exhibited by their children, citing problematic com-
munication, less affection, and maladaptive coping techniques learned
from their parents. Another review to emerge at the same time exam-
ined 24 studies of children who were considered to be at-risk because of
the presence of an affective disorder in one or both parents (Beardslee,
Bemporad, Keller, & K1erman, 1983). Similar to the findings of Or-
vaschel (1983), these researchers pointed out that children with de-
pressed parents did not necessarily exhibit symptoms of childhood de-
pression, but rather showed a wide array of difficulties, such as neurotic
or behavioral disturbances, attentional problems, drug abuse, and con-
flict with the law.
Thus, researchers in this area noted early on that depression in the
offspring of depressed parents is not necessarily a common or even
probable observed diagnosis, but that such children appear to display a
variety of sym ptoms indicative of developmental de viation or delay.
Unfortunately, the prognosis of such impairments is not known, be-
cause this area is too new to have produced information on long-term
effects among targeted samples.
Commenting on these preliminary conclusions, Coyne, Kahn, and
Gotlib (1987) suggested that the expression of developmental deviation
in the offspring may be accounted for to a large extent on the basis of
disturbance in childrearing methods associated with depression. Rather
than a unilateral, causal relationship between parental depression and
childhood disturbance, however, these researchers pointed to the major
role depression plays in the marital relationship and in similar aspects of
family functioning. Furthermore, they noted that most of the data to
emerge from studies in this area are generalizable to mothers only, since
the influence of depressed males in the family has not been as thor-
oughly investigated. Finally, Coyne et al. (1987) pointed out that few
observations of the daily interactions between depressed parents and
their children have been conducted, which might help to explain the
182 DAVID C. FACTOR and DAVID A. WOLFE

mechanisms whereby parental depression and its associated behavioral

manifestation may influence ongoing child development.
In the review of recent studies in this area that follows, the liter-
ature is divided into three subsections reflecting the major directions
taken in the field to understand the relationship between parental de-
pression and development. The first subsection explores in greater de-
tail the behavioral problems reported among this child population.
These studies are followed by a close look at the cognitive and affective
disturbances exhibited among children of depressed parents. The final
subsection presents findings regarding the observed parent-child in-
teractions in such families, in an attempt to outline the major processes
that may explain the negative developmental sequelae that have been
observed to date.

BEHAVIORAL PROBLEMS AMONG CHILDREN

OF DEPRESSED PARENTS

To aid in clarification and specificity, two distinctions are typically

made in the literature on adult depression: (1) the presence of a unipolar
versus bipolar (i.e., manic/depressive episodes) form of mood disorder,
and (2) whether or not the adult is receiving inpatient versus outpatient
treatment for the disorder, because the presence of one or the other form
or severity of depressive symptomatology is predictive of the course and
prognosis of the disorder. Accordingly, Kashani, Burk, Horowitz, and
Reid (1985) were interested in comparing 41 children (from 19 different
families) who had parents with unipolar disorder with 9 children (from 5
families), whose parents suffered from bipolar depression. All parents
were inpatients at the time the study was conducted, with a ratio of 3
mothers to every one father in the study. Among these two small sam-
ples of children (who ranged in age from 7 to 17 years, with a mean age
of 12.4), the researchers found that children in the unipolar sample
reported significantly more symptoms of somatization and separation
anxiety than did children in the bipolar group. In contrast, children in
the bipolar sample reported significantly more symptoms of alcohol
use/abuse than the comparison sample. Despite these important dif-
ferences, however, it should be noted that the researchers failed to find
very many significant differences between the two at-risk groups overall
on formal measures of child adjustment (e.g., diagnostic interviews and
parent-report questionnaires), a finding that was repeated in a similar
study reported by Hammen, Adrian, Gordon, Burge, Jaenicke, and
Hiroto (1987).
PARENTAL PSYCHOPATHOLOGY AND HIGH-RISK CHILDREN 183

Zahn-Waxler et at. (1988) reported on a unique investigation in

which they were able to follow the development of seven 5- and 6-year-
old male children who each had a depressed parent (4 mothers and 3
fathers suffered from bipolar depression, and 5 of their spouses were
diagnosed as having a unipolar depression disorder). The researchers
noted that from a very early age , these children were likely to display
insecure attachment with their caregivers and to have problems in reg-
ulation of affect. In addition, as infants and small children, these sub-
jects had difficulties in the appropriate expression of prosocial behavior
and the control of aggression. When compared to 12 male controls of the
same age, the children of depressed parents displayed a higher frequen-
cy and severity of problematic behaviors, a finding that was sustained at
least over the preschool period in which these children were followed .
Using an older sample, Lee and Gotlib (1989) compared 7- to 13-
year-old children of unipolar depressed mothers (N = 20), children of
nondepressed, psychiatric patients (N = 13), children of nondepressed
medical patients (N = 8), and children of nondepressed, nonpatient
mothers in the community (N = 30). Based on the Child Behavior Check-
list, the researchers reported that the highest proportion of scores in the
clinically disturbed range (both internalizing and externalizing prob-
lems) were found among children of depressed mothers (i.e. , two-thirds
of these children had elevated scores falling in the clinical range on this
instrument). Based on clinician ratings, children of both ps ychiatric pa-
tient subgroups exhibited more symptomatology and poorer adjustment
overall when compared to the other two groups of children. Interesting-
ly, the authors noted that both groups of children of ps ychiatric patients
were behaving at a level comparable to children who have been referred
to outpatient clinics for behavior problems.
These recent findings on the adjustment problems among children
of depressed parents illustrate some of the general disturbances in be-
havior that have been observed, and generally support the view that
parental depression is associated with greater adjustment problems
among their offspring. However, mention should be made that a partic-
ular pattern or diagnostic cluster of symptoms has not been identified
which corresponds wholly or in major part to the parental diagnosis of
depression. Similar to other situational and familial factors (such as child
maltreatment) associated with developmental risk (e.g ., Wolfe, 1987),
the presence of parental depression appears to disrupt the child's nor-
mal, ongoing development in a pervasive manner that cannot be pre-
dicted or described in a unidimensional fashion . Rather, parental de-
pression, and perhaps other forms of psychiatric disturbance as well,
interferes with the child's normal development of behavioral, cognitive,
184 DAVID C. FACTOR and DAVID A. WOLFE

and affective abilities, and such interference carries with it an unpredict-

able developmental course. Further support for such interference and
negative influence is found in the following studies on children's cog-
nitive delays or impairments associated with major depression in the
caregiver.

COGNITIVE AND AFFECTIVE DISTURBANCES AMONG

CHILDREN OF DEPRESSED PARENTS

In addition to the increased risk of behavior problems, investigators

have been concerned with other developmental impairments that might
result from parental psychopathology, but that might be less detectable
or disruptive to the untrained observer. Worland, Weeks, Janes, and
Strock (1984) were among the first to note the preponderance of cog-
nitive and affective symptomatology among children of disturbed par-
ents. These researchers studied 158 6- to 12-year-old children who had
parents who were diagnosed as either schizophrenic, manic-depressive,
or suffering from chronic physical illness, along with a nondiagnosed
control group. Using path analysis to determine the best fit to their data,
they looked at such variables as children's intelligence, classroom behav-
ior, achievement in school, and emotional functioning. The effect of
parental depression seemed to be most pronounced in terms of its influ-
ence on cognitive development, followed by impairments in emotional-
behavioral adjustment; that is, children's classroom behavior was medi-
ated by intelligence and by academic achievement-cognitive variables
that are believed to be most vulnerable among children of depressed
parents.
The greater preponderance of developmental psychopathology
among children of depressed parents is further confirmed in studies
looking at the frequency of various psychiatric diagnoses among this
population. Weissman et al. (1984), in an investigation of school-aged
children (ages 6-18), found that the rate of psychiatric diagnosis among
the children of depressed parents was three times greater than that of
the normal sample. Major depression accounted for approximately 13%
of the diagnoses, and was the most frequently occurring single diag-
nosis among the sample of 44 children of parents with major depression
and 89 children of parents with mild depression. Attention deficit disor-
der and separation anxiety each accounted for 10% of the diagnoses, and
multiple diagnoses in these children often appeared. Orvaschel, Walsh-
Allis, and Ye (1988) found further support for the relationship between
parental depression and cognitive/affective disturbance among a sample
PARENTAL PSYCHOPATHOLOGY AND HIGH-RISK CHILDREN 185

of 61 children of depressed parents and 40 children of nondepressed

parents. Their data revealed that 41% of the children of depressed par-
ents met the criteria for at least one psychiatric disorder during the
course of their childhood, in comparison to 15% of the normative sam-
ple . In addition, the high-risk target group received more outpatient
treatment services, and were described as receiving significantly more
diagnoses for affective disorders and attention deficit disorder than the
comparison group. Not only did they find higher rates of these disorders
among the proband's children, the results indicated that these problems
were more severe and long-term.
A growing consensus among researchers does support a small but
significant continuity between parents and their children in terms of
symptoms of affective disturbance. As noted previously, research has
not supported the simple notion that children of depressed parents
become depressed themselves, yet this is one of the more prominent
developmental outcomes that is gaining recognition in recent studies.
For example, Hammen et al. (1987a) and Hammen, Gordon, Burge,
Adrian, Jaenicke, & Hiroto (1987b) looked at frequency of diagnoses
among school-aged children of depressed mothers (including both uni-
polar and bipolar subsamples) in addition to medical and normal com-
parison groups. They discovered that rates of affective disturbances
among the children of depressed parents differed only from the normal,
and not the medical, controls. However, it is revealing to note that
slightly less than one half of the children in the unipolar group (9 out of
19) had high rates of depression, whereas only one quarter (3 out of 12)
of the bipolar group exhibited such problems. The researchers con-
cluded that although a complicated risk path is involved, children with
parents who suffer from affective disturbances are at a significantly
higher risk of receiving a psychiatric diagnosis and impairment in social
and cognitive functioning. Similar results and conclusions are reported
by Klein, Clark, Dansky, and Margolis (1988), with the added finding of
a higher rate of dysthymia in the female offspring of patients with uni-
polar depression than in the offspring of parents with medical problems
or those who did not evidence any history of psychiatric or medical
disorders.
One major study did not find support for this relationship between
maternal depression and affective disorders among child/adolescent off-
spring, although the findings raise important questions for future re-
search in this area. Forehand et al. (1988) studied young adolescents
(mean age = 13.5 years) who were assessed along with their parents on
two measures of depression involving self-report and behavioral rat-
ings. Over time, they found that maternal depression self-ratings were
186 DAVID C. FACTOR and DAVID A. WOLFE

highly correlated, yet this was not found for the adolescents on either
measure. However, they discovered that maternal depression increased
in association with an increase in marital conflict, a finding that mirrors
the literature on wife battering and maternal effectiveness (e.g., Wolfe,
Jaffe, Wilson, & Zak, 1985). That is, marital conflict and violence are
suspected to influence negatively maternal effectiveness and depressive
symptomatology directly, and these impairments in turn have an indi-
rect impact on parenting abilities. Thus, the study by Forehand et al.
(1988) has several implications for the type of measures used to assess
parental as well as adolescent depressive symptoms, and for the assess-
ment of critical situational events in the family, such as marital discord
and violence.
In summarizing these studies on behavioral, cognitive, and affec-
tive problems among children of depressed parents, the common find-
ing should be emphasized that there is no simple relationship between
parental depression (maternal or paternal) and affective disorders in the
offspring. Instead, studies are discovering a much more complicated
pathway whereby children may become at a greater risk for a wide array
of psychiatric problems when either or both parents suffers from depres-
sion. Based on the few studies conducted to date, differences in the
behavior of children of unipolar versus bipolar depressed parents do not
appear to be of clinical significance; rather, both subtypes appear to
elevate the probability of receiving a psychiatric diagnosis during child-
hood or adolescence. Furthermore, the vast majority of studies to date
have focused on maternal depression, and thus the role of paternal
psychiatric illness is poorly understood.
We now turn to a closer look at the interactions between depressed
parents and their children in order to see how their child-rearing meth-
ods and communications to their children may be responsible for devel-
opmental impairments.

PARENT/CHILD INTERACTIONS IN FAMILIES

WITH A DEPRESSED PARENT

The characteristics of adult depression (i.e., feelings of help-

lessness, uselessness, being unable to function effectively, poor con-
centration, interpersonal disinterest), when combined with the
responsibilities and demands of parenthood (e.g ., feeding schedules,
interrupted sleep, constant surveillance, disciplining the child) pose
considerable odds against the likelihood of conflict-free, positive par-
ent-child interactions. Furthermore, as noted by Weissman (1979), de-
PARENTAL PSYCHOPATHOLOGY AND HIGH-RISK CHILDREN 187

pression is more common in women, who also carry the bulk of the
child-rearing responsibility in most families . Although the directionality
issue cannot be easily resolved (i.e., do difficult infants bring about
greater maternal depression, or do depressed mothers give rise to more
difficult infants?), several important aspects of the daily interactions
between depressed mothers and their infants or young children provide
some clarity to the issue of how parental depression plays a pathogenic
role in child development.
Weissman and her colleagues (1979; Weissman, Paykel, & Klerman,
1972; Weissman et al., 1987) concluded from their observations of in-
teractions between depressed mothers and their offspring that these
children were deprived of normal involvement with their parents. For
example, during play these parents were unenthusiastic and provided
little involvement or guidance for the child; as the children grew older,
the parents continued this pattern by showing a lack of interest in the
child's school activities, social events, or peers. Decreased involvement
on the part of the parent took other forms as well, such as paying little
attention to the child's physical health or appearance. The researchers
further noted that the children were not as often encouraged to discuss
their feelings or to discuss their daily activities, compared to comparison
children from nondepressed families. Not only were the parent/child
interactions marked by greater disinterest and less involvement, they
observed that acutely depressed parents behaved in a more hostile fash-
ion toward their children, a finding that has important implications as
well for the child's development of a sense of self and self-mastery.
Overall, these researchers interpreted their findings to highlight the four
primary areas of parental dysfunction that can be identified in families
with one or more depressed parents: involvement and disinterest, com-
munication, affection, and hostility. All four of these areas (when
phrased in positive terms) are deemed critical to healthy child develop-
ment (Maccoby & Martin, 1983), and therefore the mechanisms by
which parental depression plays a role in developmental psycho-
pathology are becoming more apparent.
Additional studies have focused on the critical period of attachment
between mother and infant during the first months of life, and have again
revealed a pathogenic pattern among depressed parents of low involve-
ment and responsivity toward their offspring. Cohn and Tronick (1983)
used an experimental design to manipulate the presence or absence of
maternal depressed mood while interacting with their 3-month-old in-
fants (12 female and 12 male). Mothers in the study were instructed to
interact normally or to interact with a simulated "depressed expression"
with their infants. Results indicated that those mothers assigned to the
188 DAVID C. FACTOR and DAVID A. WOLFE

"depressed" condition were significantly more undercontrolling and less

elaborate compared to mothers in the normal condition. More important-
ly, condition effects revealed that infants in the "depressed" condition
showed higher rates of wariness and protest (50% of the time), and less
brief positive reactions. In contrast, in the normal condition, infants
rarely exhibited protest or wary behavior and showed higher frequencies
of brief positive displays. In commenting on these data, the researchers
pointed out that "the sequencing of infant affect states is clearly related to
the quality of maternal expression" (p. 190).
Radke-Yarrow, Cummings, Kuczynski, and Chapman (1985) also
examined patterns of attachment, involving 2- and 3-year-old children
and their mothers. Fourteen children had bipolar depressive mothers, 42
children had mothers diagnosed as major unipolar depression, 12 moth-
ers were diagnosed as minor depression, and 31 were nondepressed.
The results indicated that there was no difference between children of
nondepressed mothers and children of mothers with minor depression
in terms of the frequency of insecure attachments. However, there was a
significantly greater incidence of insecure attachments in the major af-
fective disorder group in comparison to normals. The researchers in-
terpret these findings (in conjunction with additional findings not re-
viewed here) to suggest that maternal depression decreases the likeli-
hood of secure attachment between the mother and child. Among the
mothers with one of the forms of a major effective disorder, 55% were
observed to have an insecure attachment with their child (this figure
increased to 79% among mothers with a diagnosis of a bipolar depres-
sion, and was 47% and 29% among unipolar and nondepressed moth-
ers, respectively). Once again, these results are quite clear in underscor-
ing the limited range of affect among depressed mothers and the effects
of such a pattern on the cognitive and emotional development of her
child.
Changes in communication patterns under stressful and non-
stressful conditions was the focus of attention in a study of 3-year-olds
reported by Breznitz and Sherman (1987). These investigators studied 32
mother/child dyads in which the mother either suffered from periods of
depression or had no psychiatric history or symptomatology. Not sur-
prisingly, they found that the nondepressed mothers spoke significantly
more often to their young children in comparison to depressed mothers
during everyday, nonstressful circumstances. However, when placed in
a stressful situation, depressed mothers increased the amount of speech
they produced whereas nondepressed mothers had a slight decrease in
their amount of speech. Both groups decreased their response latency
during the stressful condition, but again the depressed mothers showed
PARENTAL PSYCHOPATHOLOGY AND HIGH-RISK CHILDREN 189

an even greater reduction in the amount of time it took for them to

respond to their child. The investigators suggested on the basis of these
findings that the two groups of mothers may handle stress quite differ-
ently. They interpreted the faster reaction time and increased speech
productivity of the depressed mothers as an exaggerated or overreaction
to the situation, which ma y convey anxiety or fear to the child. Conse-
quently, maternal depression can be linked to developmental repercus-
sions in the offspring that are due not only to the content of the interac-
tion as discussed previously, but perhaps due as well to the patterning
and delivery of such interactions.
The manner in which depressed mothers interact with and view the
behavior of their older children has also been the focus of considerable
inquiry in recent years . Investigating the reports of depressed (N = 46)
and nondepressed (N = 49) mothers of the 3- to 8-year-old children,
Webster-Stratton and Hammond (1988) found that depressed mothers
rated their children significantly higher in the areas of externalizing,
internalizing, and depression scales of the Child Behavior Checklist.
Depressed mothers also reported greater stress in the parenting role as
measured by the Parenting Stress Index, noting higher scores in such
areas as attachment, depression, role restriction, low sense of compe-
tence, and socialization and health. Surprisingly, home observations of
parent-child interaction did not reveal differences between the groups
on five behavioral measures (although a trend for more critical state-
ments among depressed mothers was noted). However, the results in
total (including additional findings not reviewed herein) support the
premise that the children of depressed mothers are perceived by the
parent as being more disruptive and disturbed than they may be . This
finding parallels the child abuse literature, in which abusive mothers
have been found to perceive their children in a more negative light than
do more objective observers (Wolfe, 1987), and is a reminder that the
problems of children of disturbed parents cannot always be inferred or
denied on the basis of parental report. As stated by Christensen, Phil-
lips, Glasgow, and Johnson (1983), the perception of child behavior prob-
lems may be more related to marital discord and parental negative be-
havior rather than to the actual behavior of the child.
Further insights can be gained from our study of children of de-
pressed parents by comparing these findings to those involving abusive
parents. A logical connection exists between these two clinical popula-
tions, given the preponderance of negative interactions, low ratio of
positive attention or reinforcement of the child, and reliance on critical
commands (and punishment) to teach and control the child, found in
both groups. However, as noted elsewhere (Wolfe, 1985), abusive par-
190 DAVID C. FACTOR and DAVID A. WOLFE

ents are defined more specifically in terms of their singular behavior

toward their children-extreme physical actions-rather than by the
presence of a psychiatric condition or a syndrome of identifiable impair-
ments, such as major affective disorder. Thus, child abuse is an event
(more so than a psychiatric disorder) that is often associated with other
prime examples of authoritarian parenting style, whereas depression
can be more accurately described as the existence of a condition or
cluster of symptoms that leads to impairments in the child-care role.
These distinctions notwithstanding, it is useful to compare the two
populations of abusive and depressed parents to develop a consensus as
to the similarities and differences in parent-child interactions. In an
unusual study that involved subsamples of normal, abusive, and de -
pressed mothers, Susman et al. (1985) found important differences in
methods of child-rearing. Abusive mothers, in particular, were higher
on self-reported ratings of authoritarianism, anxiety induction, guilt in-
duction, and discipline inconsistency, and were lower on rationale guid-
ance and enjoyment of the parenting role. Furthermore, abusive moth-
ers were unlikely to encourage independence or openness to a new
experience, tended to be very protective and to worry about the child,
and to show less open expression of affect toward the child. Interesting-
ly, this study revealed fewer deviations from the norm among the de-
pressed mothers, with the exception of discipline inconsistency and
need for control. However, depressed mothers were similar to the
abusive mothers on 38% of the 21 child-rearing factors studied. Extrapo-
lating from these findings, the authors suggested that the greater
number and variety of child-rearing difficulties reported by the abusive
mothers is indicative of their more specific deficits in the parenting
domain, whereas the problems experienced by depressed mothers may
interfere with child-rearing in a more general manner; that is, depend-
ing on mood or situational factors, depressed mothers may respond to
their children in a more or less hostile and rejecting fashion (see also
Lahey, Conger, Atkeson, & Treiber, 1984).
In general, the literature reviewed in this section on depressed
parents underscores the relationship between parental affective disorder
and developmental risk throughout the periods of infancy, childhood,
and adolescence. Not only is there a direct association between the
existence of a psychiatric diagnosis in the offspring and the presence of
parental depression, but there is also evidence of the mechanisms by
which parental depression interferes with the developing parent-child
relationship. Depressed mothers, in particular, are observed to be less
interactive and more critical of their children, a finding that mirrors to
some extent the results of observational studies with abusive parents.
PARENTAL PSYCHOPATHOLOGY AND HIGH-RISK CHILDREN 191

Moreover, studies have demonstrated that, in some depressed families,

marital conflict accounts for more of the variance in child behavior prob-
lems than does the presence of parental depression. In turn, we also
find that symptoms of maternal depression show a positive correlation
with marital conflict, a finding that warrants greater attention if efforts at
early detection and intervention of child behavior problems are to be
maximally effective.

PARENTAL PSYCHOPATHOLOGY AND

CHILD MALTREATMENT
When child maltreatment first began to receive the attention of
researchers and practitioners over 25 years ago, it was considered to be a
manifestation of severe parental psychopathology. Only psychiatrically
disturbed individuals, it was reasoned, could show the lack of control
and concern that was exhibited in documented cases of child abuse and
neglect. Inspired by this early viewpoint, two decades of research stud-
ies have focused on developing an accurate understanding of the role
that parental psychopathology might play in the expression of child
abuse and neglect. Because the vast majority of parents are capable of
dealing with difficult child behavior and stressful circumstances without
resorting to physical violence or neglect, researchers have suspected
that maltreating parents must lack some form of inner control, have
experienced early trauma in their own families of origin, or suffer from a
major thought disorder that limits their recognition of the consequences
of their actions (see Green, 1978; Spinetta & Rigler, 1972). Although
many of these disturbances have been identified among samples of mal-
treating parents, no distinctive psychological profile or pattern has been
documented that supports the view that parental psychopathology is at
the root of child maltreatment.
In order to understand the role of parental adjustment in the ex-
pression of abusive and neglectful behavior toward their offspring, it is
necessary to clarify the context and nature of such behavior. In more
recent studies, child maltreatment has been viewed as the product of an
interaction between parental functioning and situational demands
(rather than being limited primarily to parental psychopathology), Al-
though the role of parental behavior still remains crucial in such a re-
definition, the significance of parental disturbance has become less dra-
matic and less conspicuous than was originally assumed on the basis of
their actions.
Early information on the behavior of abusive parents was derived
192 DAVID C. FACTOR and DAVID A. WOLFE

primarily from clinical case studies, which provided a rich source of

descriptive knowledge about the characteristics of such individuals. In
the behavioral dimension, these case studies described abusive parents
as chronically aggressive, isolated from family and friends, rigid and
domineering, impulsive, and experiencing marital difficulties. Similarly,
the cognitive-emotional functioning of these parents was reported to be
marked by low frustration tolerance, emotional immaturity, role rever-
sal, deficits in empathy and self-esteem, high expectations for their
child's behavior, and problems in the expression and control of anger
(see reviews by Parke & Collmer, 1975; Spinetta & Rigler, 1972; Wolfe,
1985, 1987).
These early clinical case studies became more sophisticated in the
early 1970s, as more researchers entered the field and began to include
matched control families to provide a basis for comparison on these
psychological dimensions. The concept of an identifiable personality
disorder or disturbance that could account for abusive behavior was
contested by the social learning emphasis on person-situation interac-
tion, which resulted in different methodological approaches to under-
standing this phenomenon. In contrast to personality dimensions, the
emphasis shifted toward the daily interactions of family members (es-
peciall y parent-child interactions) that might explain the escalation in
intensity and severity that defines an abusive episode. In addition, self-
report devices were developed that targeted the unrealistic expectations
of some parents and the type and extent of physical and emotional
symptoms experienced, which resulted in several significant contrasts
between identified abusive parents and nonabusive parents.
The empirical findings that emerged from these more recent studies
revealed many important similarities as well as differences in com-
parison to the earlier reports. In particular, several behavioral differences
between abusers and nonabusers were reaffirmed, such as low frustra-
tion tolerance, social isolation, and impaired child-rearing skills. More-
over, several cognitive-emotional differences were supported, such as
the tendency to demonstrate unrealistic expectations of their children,
to report that their child's behavior is extremely stressful to them, and to
describe themselves as being inadequate or incompetent in the parent-
ing role (Wolfe, 1987).
A major description of abusive parents that was not confirmed or
supported by these more recent studies had to do with personality
disorders or ps ychopathology as the common element shared by abu-
sive parents. Psychiatric descriptions, such as emotional immaturity,
impulsivity, and low self-esteem have not received a great deal of con-
sensus, perhaps because of the difficulty in defining and measuring
PARENTAL PSYCHOPATHOLOGY AND HIGH-RISK CHILDREN 193

these constructs. Thus, the notion of a distinctive personality profile or

cluster of symptoms describing abusive parents can be modified to re-
flect the interactive nature of this phenomenon. Certain predisposi-
tional characteristics of some parents may place them at greater risk for
abuse, because of such behavior patterns as inflexibility and maladaptive
responses to certain situational contexts, especially those situations in-
volving a difficult child, handling stressful situations, and solving fami-
ly-related problems. Unfortunately, our knowledge of child neglect is far
behind that of physical abuse. However, it appears from preliminary
studies that neglect may involve an even greater degree of parental
psychopathology than abuse which, coupled with situational events,
leads to an avoidance response to stressful child behavior rather than to
aggression (Wolfe, 1985). Whether these two forms of child maltreat-
ment represent different manifestations of the same disorder or whether
they are the two most identifiable patterns of parenting dysfunction (out
of perhaps many more) remains to be investigated.

SUMMARY
We have reviewed a number of areas of parental psychopathology
that have implications for child development. Because our attention was
focused on the major psychosocial factors associated with parental disor-
ders, we drew from the expanding literature on children of depressed
parents to highlight the suspected processes involved in the transmis-
sion or transaction of developmental psychopathology in offspring. De-
spite the growing knowledge in this area of parental psychopathology,
we again remind the reader that the studies covered in this review were
limited primarily to situational and psychological variables. Interested
readers will have to find elsewhere additional theoretical explanations,
methodological procedures, and causal or correlational relationships as-
sociated with genetic, biological, and psychiatric fields of study.
Other important psychosocial variables that were not the focus of
the present review merit some mention in closing. Parental alcoholism,
for example, is known to interfere with normal child-rearing and the
development of a healthy parent-child relationship. Consequently, it
comes as no surprise that alcoholism in parents has been found to be
associated with adjustment problems in children, especially males
(Adler & Raphael, 1983; West & Prinz, 1987). Alcoholic parents appear to
fit a pattern of inconsistent and unpredictable childcare that has been
similarly discovered in many of the studies of parental disorders re-
viewed. In addition, we need to look more carefully at the cultural,
194 DAVID C. FACTOR and DAVID A. WOLFE

familial, and social support factors that play a role in mediating the
negative impact of poverty and disadvantage of children, because chil-
dren from lower socioeconomic backgrounds tend to be more at risk for
school problems (especially among minority children; Felner, Gillespie,
& Smith, 1985).
The findings presented throughout this chapter support a social
learning explanation of developmental psychopathology, which predicts
a disruption or alteration in development as a function of significant fac-
tors affecting learning opportunities, of which parental depression is but
one. Rather than assuming a one-to-one correspondence between pa-
rental psychopathology and developmental outcome, social learning
theory suggests that events that have a significant influence on the
child's learning environment can lead to changes or deviations in coping
responses, expectations, problem-solving skills, and related develop-
mental events. In this manner, parental depression represents one of the
more visible and dramatic circumstances that can change the course of
normal child development, much the same as do child abuse (Wolfe,
1987)/ parental divorce (Hetherington & Martin, 1986)/ wife battering
(Wolfe et al., 1985)/ and parental criminality (Lewis, Balsla, Shanok, &
Snell, 1976)/ to name only a few. Consistency and predictability of child
care may be the common threads that link many of these deviant forms
of parenting practices to the wide variety of negative developmental
outcomes (Wahler & Dumas, 1987).
Several methodological considerations emerge from the current lit-
erature in this area that justify further attention. Perhaps because of the
inexact nature of our diagnostic systems, the criteria for determining
and agreeing upon the type of problem(s) exhibited by some parents are
often unclear. Furthermore, the choice of assessment instruments and
procedures typically varies from study to study, and therefore it is diffi-
cult to draw comparative conclusions. One major direction that has been
undertaken in this regard is a better understanding of the manner in
which certain pathological conditions in parents affect their ratings of
child behavior (see, for example, Jensen, Traylor, Xenakis, & Davis,
1988; Schaughency & Lahey, 1985). Most importantly, we need more
longitudinal research on this topic in order to draw more firm conclu-
sions about the long-term effects of parental psychopathology and the
course and stability of such changes. Finally, we point to the growing
recognition that a significant number of children survive without any
detectable harm, even in very unhealthy environments (Beardslee &
Podorefski, 1988)/ a reality that may offer assistance in understanding
the plight of children of disturbed parents.
PARENTAL PSYCHOPATHOLOGY AND HIGH-RISK CHILDREN 195

ACKNOWLEDGMENTS

Preparation of this chapter was supported in part by the Institute

for the Prevention of Child Abuse (Toronto, Canada), and by a grant to
David A. Wolfe from the Social Sciences and Humanities Research
Council of Canada. We wish to thank the Charlestown Residential
School, Adrienne Perry, Nancy Freeman, and Diane Factor for their as-
sistance and support.

REFERENCES
Adler, R., & Raphael, B. (1983). Review : Children of alcoholics. Australian and New Zealand
Journal of Psychiatry, 17, 3-8.
Anderson, C (1987). Depression and families . Journal of Psychotherapy and the Family, 3,33-
48.
Barnett, P. A., & Gotlib , I. H . (1988). Psychosocial functioning and depression : Dis-
tinguishing among antecedents, concomitants, and consequences. Psychological Bul-
letin, 104, 97-126.
Baumrind, D. (1971). Current patterns of parental authority. Developmental Psychology
Monographs, 4 (1, pt. 2).
Beardslee, W. R., & Podorefsky, D. (1988). Resilient adolescents whose parents have se-
rious affective and other psychiatric disorders : Importance of self-understanding and
relationships. American Journal of Psychiatry, 145, 63-69.
Beard slee, W. R., Bemporad, j., Keller, M. B., & Klerman, G. L. (1983). Children of parents
with major affective disorder: A review. The American Journal of Psychiatry, 140, 825-
832.
Beardslee, W. R., Keller, M. B., & Klerman, G. L. (1985). Children of parents with affective
disorder. International Journal of Family Psychiatry, 6, 283-299.
Bell, R. Q. , & Harper, L. (1977). Child effects on adults. Hillsdale, N]: Lawrence Erlbaum.
Blackford, K. A. (1988). The children of chronically ill parents. Journal of Psychosocial Nurs-
ing, 26, 33-36.
Breznitz, Z., & Sherman, T. (1987). Speech patterning of natural discourse of well and
depressed mothers and their young children. Child Development, 58, 395-400.
Cantwell, D. P., & Baker, L. (1984). Parental mental illness and psychiatric disorders in "at
risk" children. Journal of Clinical Psychiatry, 45, 503-507.
Christensen, A., Phillips , S., Glasgow, R. E., & Johnson, S. M. (1983). Parental charac-
teristics and interactional dy sfunction in families with child behavior problems: A
preliminary investigation. Journal of Abnormal Child Psychology, 11, 153-166.
Cohn, ]. E , & Tronick, E. Z. (1983). Three-month-old infants' reaction to simulated mater-
nal depression. Child Development, 54, 185-193.
Cooper, S. E , Leach, C, Storer, D., Tonge, W. L. (1977). The children of psychiatric
pa tients: Clinical findings. British Journal of Psychiatry, 131, 514-522.
Coyne, ]. C, Kahn, ]., Gotlib , I. H . (1987). Depression. In T. Jacob (Ed.), Family interaction
and psychopathology: Theories, methods, and findings (pp . 509-533). New York: Plenum
Press.
196 DAVID C. FACTOR and DAVID A. WOLFE

Emery, R. (1982). Interparental conflict and the children of discord and divorce. Psychologi-
cal Bulletin, 92, 310-330.
Feiner, R D., Gillespie, J. F., & Smith, R. (1985). Risk and vulnerability in childhood: A
reappraisal. Journal of Clinical Child Psychology, 14, 2-4.
Forehand, R, Brody, G., Slotkin, J., Fauber, R, McCombs, A., & Long, N. (1988). Young
adolescent and maternal depression: Assessment, interrelations, and family predic-
tors . Journal of Consulting and Clinical Psychology, 56, 422-426.
Gold stein, M. J. (1988). The family and psychopathology. Annual Review of Psychology, 39,
283-299.
Green, A. H . (1978). Child abuse . In B. B. Wolman, J. Egan, & A. Ross (Eds .), Handbook of
treatment of mentaldisorders in childhood andadolescence (pp. 430-455) . Englewood Cliffs,
NJ: Prentice-Hall.
Hammen, C ,; Adrian, C.; Gordon, D., Burge, D., Jaenicke, C,; & Hiroto, D. (1987a).
Children of depressed mothers: Maternal strain and symptom predictors of dy sfunc-
tion. Journal of Abnormal Psychology, 96, 190-198.
Hammen, C, ; Gordon, D., Burge, D., Adrian, c., Jaenicke, C.; & Hiroto, D. (1987b).
Maternal affective disorders, illness, and stress: Risk for children's psychopathology.
American Journal of Psychiatry, 144, 736-741.
Hetherington, E. M., & Martin, B. (1986). Family factors and psychopathology in children.
In H. C. Quay, & J. S. Werry (Eds.), Psychopathological disorders ofchildhood (3rd ed ., pp.
332-390). New York: John Wiley.
Jensen, P. S., Traylor, J., Xenakis, S. N ., & Davis, H . (1988). Childhood psychopathology
rating scales and interrater agreement: I. Parents' gender and psychiatric symptoms.
Journal of the American Academy of Childand Adolescent Psychiatry. , 27, 422-450.
Kandel, E., Mednick, S. A., Kirkegaard-Sorensen, L., Hutchings, B., Knop, J., Rosenberg,
R., & Schulsinger, F. (1988). IQ as a protective factor for subjects at high risk for
antisocial behavior. Journal of Consulting and Clinical Psychology, 56, 224-226.
Kashani, J. H ., Burk, J. P., Horwitz, B., & Reid, J. C. (1985). Differential effect of subtype of
parental major affective disorder on children. Psychiatry Research, 15, 195-204.
Klein, D. N ., Clark, D. C,; Dansky, L., & Margolis, E. T. (1988). Dysthymia in the offspring
of parents with primary unipolar affective disorder. Journal of Abnormal Psychology, 97,
265-274.
Lahey, B. B., Conger, R. D., Atkeson, B. M., & Treiber, F. A. (1984). Parenting behavior
and emotional status of physically abusive mothers. Journal of Consulting and Clinical
Psychology, 52, 1062-1071.
Lee, C. M., & Gotlib, I. H. (1989). Clinical status and emotional adjustment of children of
depressed mothers. American Journal of Psychiatry, 146, 478-483 .
Lewis, D.O., Balla, D., Shanok, S., & Snell , L. (1976). Delinquency, parental psycho-
pathology, and paternal criminality: Clinical and epidemiological findings. Journal of
the American Academy of Child Psychiatry, 15, 665-678.
Maccoby, E. E., & Martin, J. A. (1983). Socialization in the context of the family : Parent-
child interaction. In P. H . Mussen (Ed.), Handbook of child psychology (4th ed .). New
York: Wiley.
Masten, A. S., & Garmezy, N. (1985). Risk, vulnerability, and protective factors in develop-
mental psychopathology. In B. B. Lahey, & A. E. Kazdin (Eds.), Advances in clinical
child psychology (Vol. 8, pp. 1-52). New York: Plenum Press.
Minuchin, S. (1974). Families and family therapy. Cambridge: Harvard University Press.
Minuchin, S., Rosman, B. L., & Baker, L. (1978). Psychosomatic families . Cambridge: Har-
vard University Press.
Olson, D. H ., Sprenkle, D. H ., & Russell, C. S. (1979). Circumplex model of marital and
PARENTAL PSYCHOPATHOLOGY AND HIGH-RISK CHILDREN 197

family systems: I. Cohesion and adaptability dimensions, family types, and clinical
applications. Family Process, 18, 3-28.
Orvaschel, H. (1983). Maternal depression and child dysfunction: Children at risk . In B. B.
Lahey, & A. E. Kazd in (Eds.), Advances in clinical childpsychology (Vol. 6, pp. 169-197).
New York: Plenum Press.
Orvaschel, H., Weissman, M. M., & Kidd, K. K. (1980). Children and depression. Journal of
Affective Disorders, 2, 1-16.
Orvaschel, H., Walsh-Allis , G., & Ye, W. (1988). Psychopathology in children of parents
with recurrent depression. Journal of Abnormal Child Psychology, 16, 17-28.
Parke, R. D., & Collmer, C. W. (1975). Child abuse: An interdisciplinary analysis. In E. M.
Hetherington (Ed.), Reviewof child development research (Vol. 5, pp. 509-590). Chicago:
University of Chicago Press.
Phipps-Yonas, S. (1980). Teenage pregnancy and motherhood: A review of the literature.
American Journal of Orthopsychiatry, 50, 403-430 .
Radke-Yarrow, M., Cummings, E. M., Kuczynski, L. , & Chapman, M. (1985). Patterns of
attachment in two- and three-year-olds in normal families and families with parental
depression. Child Development, 56, 884-893.
Rutter, M. (1966). Children of sick parents: An environmental and psychiatric study.
Institute of Psychiatry, Maudsley Monographs No 16. London: Oxford University
Press.
Rutter, M. (1971). Parent-child separation: Psychological effects on the children. Journal of
Child Psychiatry, 12, 233-260.
Rutter, M. (1979). Protective factors in children's response to stress and disadvantage. In
M. W. Kent & J. E. Rolf (Eds.), Primary prevention of psychopathology: Vol III. Social
competence in children . Hanover, NH : University Press of New England.
Sameroff, A. J., Seiffer, R., & Zax, M. (1982). Early development of children at risk for
emotional disorder Monographs of the Society for Research in Child Development, 47(7
Serial No . 199).
Schaughency, E. A., & Lahey, B. B. (1985). Mothers' and fathers' perceptions of child
deviance: Roles of child behavior, parental depression, and marital satisfaction. Journal
of Consulting and Clinical Psychology, 53, 718-723.
Schore, E. L. (1988). Families, family roles and psychological diagnoses in primary care.
Developmental and Behavioral Pediatrics, 9, 327-332.
Schwartz, B. A. (1979). Adolescent parents. In J. D. Nosphitz, I. N. Berling, & L. A. Stone
(Eds.), Basic handbook of child psychiatry: Prevention and current issues (Vol. 4, pp . 360-
363). New York: Basic Books.
Scott, K. G., Field, T., & Robertson, E. (1981). Teenage parents and their offspring. New York:
Grune & Stratton.
Silverman, W. K., Cerny, J. A., Nelles , W. B., & Burke, A. E. (1988). Behavior problems in
children of parents with anxiety disorders. Journal of the American Academy of Childand
Adolescent Psychiatry, 27, 779-784.
Smets, A. C,; & Hartup. W. w., (1988). Systems and symptoms: Family cohesion/ adapt-
ability and childhood behavior problems. Journal of Abnormal ChildPsychology, 16, 233-
246.
Spinetta, J. J. & Rigler, D. (1972). The child abusing parent: A psychological review. Psycho-
logical Bulletin . 77, 296-304.
Susman, E. J., Trickett, P. K., Iannotti, R. I. , Hollenback, B. E., & Zahn-Waxler, C. (1985).
Child-rearing patterns in depressed, abusive, and normal mothers. American Journal of
Orthopsychiatry, 55, 237-251.
Sylvester, C. E., Hyde, T. S., & Reichler, R. J. (1987). The diagnostic interview for children
198 DAVID C. FACTOR and DAVID A. WOLFE

and personality inventory for child ren in studies of children at risk for anx iety disor-
ders or depression . Journal of the American Academy ofChildand Adolescent Psychiatry, 26,
668-675.
Turner, S. M., Beidel, D. C.; & Costello, A. (1987). Psychopathology in the offspring of
anxiety disorders patients. Journal of Consulting and Clinical Psychology, 55, 229-235.
Wahler, R. G., & Dumas, J. E. (1987). Family factors in childhood psychology: Towards a
coercion-neglect model. In T. Jacob (Eds.), Family interaction and psychopathology: Theo-
ries, methods, and findings (pp . 581-627). New York: Plenum Pres s.
Webster-Stratton, C.; & Hammond, M. (1988). Ma ternal depression and its relation sh ip to
life stress, perceptions of child be hav ior problems, parenting behaviors, and child
conduct problems. Journal of Abnormal Child Psychology, 16, 299-315.
Weissman, M. M. (1979). Depressed pa rents and their children: Implications for preven-
tion . In J. D. No shpitz, I. N ., Berlin, & L. A. Stone (Eds .), Basic handbook of child
psychiatry: Prevention and current issues (Vol. 4, pp . 292-299). Ne w York: Basic Books .
Weissman, M. M., Gammon G. D., Joh n, K., Merikangas, K. R., Warner, V., Pru soff, B. A.,
& Sho lornskas, D. (1987). Children of depressed parents . Archives of General Psychiatry,
44 847- 853 .
Weissman, M. M., Paykel , E. S., & Klerman, G. L. (1972). The depressed woman as a
mother. Social Psychiatry, 7, 98-108.
Werner, E. E. (1989). Hig h-ri sk children in young adulthood: A longitudinal stu dy from
birth to 32 years . American Journal of Orthopsychiatry, 59, 72-81.
West, M. 0 ., & Prinz, R. J. (1987). Parental alcoholism and childhood psychopathology.
Psychological Bulletin, 102, 204-21 8.
Wolfe, D. A. (1985). Child-abusive parents: An empirical review and anal ysis . Psychological
Bulletin, 97, 462-482.
Wolfe, D. A. (1987). Child abuse: Implications for child development and psychopathology. New-
bury Park, CA : Sage .
Wolfe, D. A., Jaffe, P. J., Wilson, S. K., & Zak, L. (1985). Children of battered women: The
relation of child behavior to family violence and maternal stre ss. Journal of Consulting
and Clinical Psychology, 53, 657-665.
Worland, J., Weeks , D. G., Janes, c. L., & Stro ck, B. D. (1984). Intelligence , classroo m
behavior, and acad emic ach ievement in children at hig h and low risk for psycho-
pathology: A structural eq uation analysis . Journal of Abnormal ChildPsychology, 12,437-
454.
Za hn-Waxler, c., Mayfi eld , A., Radke-Yarrow, M., McKnew, D. H. , Cytryn, L. , & Dave n-
port, Y. B. (1988). A follow-up investigation of offspring of parents with bipolar disor-
der. American Journal of Psychiatry, 145, 506-509.
 CHAPTER 8

PREDISPOSING CHILD FACTORS

ROBERT T. AMMERMAN

INTRODUCTION
Since the original description of the Battered Child Syndrome by Kempe
and his colleagues (Kempe, Silverman, Steele, Droegemueller, & Silver,
1962), there has been a tremendous growth in professional and media
interest in child abuse and neglect. Evidence of expanded awareness is
found in the plethora of specialty journals (e.g., Journal of Family Violence,
Child Abuse and Neglect) and books (MacFarlane, Waterman, Conerly,
Damon, Durfee, & Long, 1986; Oates, 1982; Wolfe, 1987) devoted to this
topic. This increased focus of attention is at least partly attributable to
the dramatic rise in reported cases of child maltreatment in recent years.
For example, there has been an 8% increase in reported cases of child
abuse and neglect from 1985 to 1986 (American Association for Protect-
ing Children, 1988). Although such an elevation in reports is partly
related to greater professional awareness of child maltreatment, most
experts agree that child abuse and neglect currently represent significant
social problems (see Chapter 2 in this volume).
Because of the deleterious physical and psychological consequences
associated with child maltreatment (see Ammerman, Cassisi, Hersen, &
Van Hasselt, 1986; Friedrich & Einbender, 1983), recent investigators

ROBERT T. AMMERMAN· Western Pennsylvania School for Blind Children, Pittsburgh,

Pennsylvania 15213.

199
200 ROBERT T. AMMERMAN

have directed their efforts toward the prevention of abuse and neglect
(e.g., Lutzker & Rice, 1984). An important component of such efforts is
to identify children who are at high risk for maltreatment in an effort to
curtail the processes leading to abuse or neglect (Parke & Collmer, 1975).
At-risk populations have been targeted, using a variety of factors
thought to be implicated in the etiology of abuse and neglect, including
family demographics, socioeconomic status, parental psychopathology,
parent history of abuse, and substance abuse disorders (see Starr, 1988).
In addition, some theorists have suggested that certain child charac-
teristics may playa role in the development and maintenance of abuse
(deLissovoy, 1979). Thus, based upon this premise, it is possible that
these child characteristics can be subsequently used in screening for
potential child maltreatment or selecting at-risk populations for preven-
tative interventions.
This chapter will review the evidence that child factors contribute to
the etiology and maintenance of physical abuse and neglect. First, con-
ceptual models that describe the process by which children can be in-
volved in the development of maltreatment will be presented. Second,
early childhood characteristics posited to be risk factors for abuse or
neglect will be reviewed. Third, interactional studies describing the co-
ercive relationship between abusive parents and their children will be
discussed. These studies elucidate the ways in which child factors may
contribute to the maintenance of abuse. Fourth, the role of child hand-
icapping conditions in increasing risk for maltreatment will be consid-
ered. Finally, the utility of using child factors in assessing risk will be
examined, and future directions that research might take are outlined.
Child contributions to the etiology of sexual abuse will not be covered,
and the reason for this is that it is widely acknowledged that the dynam-
ics involved in sexual abuse differ greatly from those observed in phys-
ical abuse and neglect (see MacFarlane et al., 1986). Moreover, there is
little support for the contention that children have any causative role in
sexual abuse.

CONCEPTUAL MODELS OF CHILD ABUSE

AND NEGLECT
Traditionally, three models have been used to explain the develop-
ment and continuation of child abuse and neglect. These models em-
phasize parental psychopathology, sociological factors, and the social
context of the parent-child relationship, respectively. The Psycho-
pathology Model attributes child abuse and neglect to parental psychiatric
PREDISPOSING CHILD FACTORS 201

disturbance or mental illness. According to this formulation, child

abusers can be described as psychotic, impulsive, or sadistic. Also,
physical abuse or neglect is seen as a manifestation of psychodynamic
dysfunction or personality disturbance in the parent (Spinetta & Rigler,
1972). Although this model received early support in the literature, more
recent investigations have failed to document severe psychopathology
in child abusers (see Wolfe, 1985). Although parents who engage in
maltreatment clearly exhibit maladjustment in a variety of areas of func-
tioning (see Chapter 7 in this volume), their abusive behavior is rarely a
direct product of specific psychiatric disorders.
The Social-Cultural Model of child maltreatment emphasizes social
and cultural forces in the formation of child abuse and neglect. Within
this framework, domestic violence is viewed as a response to stress
engendered by unemployment, economic hardship, and educational
disadvantage (Gelles, 1973). Moreover, cultural sanctioning of physical
punishment to resolve family conflict further adds to the likelihood of
abuse (Maurer, 1974). Empirical support for this model comes from mul-
tivariate studies that identify sociological and demographic factors as
good predictors of maltreatment (Gaines, Sandgrund, Green, & Power,
1978; Garbarino, 1976). For example, Garbarino (1976) found that so-
cioeconomic factors accounted for 36% of the variance in his sample of
child abuse reports, whereas no other variable exhibited such explanato-
ry power. Critics, however, have argued that social-cultural elements do
not explain the process through which child maltreatment develops
(Wolfe, 1987). Furthermore, the majority of socioeconomically disadvan-
taged parents do not abuse or neglect their children, thus demonstrating
that this component alone does not fully account for the etiology of
maltreatment (Egeland, Breitenbucher, & Rosenberg, 1980).
The Social-Interactional Model of child abuse and neglect focuses on
the relationship between parent and child and the social context in
which the maltreatment occurs (Parke & Collmer, 1975). This approach
proposes that the unique characteristics of both the parent and the child
interact to bring about maltreatment in particular conflict situations. For
example, a parent with poor child-management skills, excessive expec-
tations regarding child behavior, and a past history of being abused may
be more likely to engage in physical abuse toward a difficult to manage,
noncompliant child during times of stress (e.g., recent job loss, eco-
nomic hardship). Therefore, the dynamic interchanges between parent
and child in conjunction with situational variables are viewed as critical
to the etiology and maintenance of abusive behavior. This model clearly
suggests that child characteristics can contribute to the occurrence of
abuse.
202 ROBERT T. AMMERMAN

More recently, theorists have constructed comprehensive multi-

casual models that attempt to combine the aforementioned concep-
tualizations and account for the complex interplay between etiological
factors (Belsky, 1980; Burgess & Draper, 1988; Starr, 1988). Although
these thorough models delineate the many levels of causative influence,
they do not further our understanding of how such elements interact to
bring about maltreatment. Wolfe (1987), on the other hand, has pro-
posed a formulation that seeks to describe the ways in which individual
and situational variables interact to inhibit or promote the likelihood of a
family engaging in abuse. The Transitional Model views abuse as a prod-
uct of the gradual escalation of power assertive parenting practices .
Through the interaction of Destabilizing Factors (that increase the risk of
abuse), and Compensatory Factors (that decrease the risk of abuse),
parents pass through three stages reflecting increased likelihood of vio-
lence directed toward the child. The three stages are: (1) Reduced Toler-
ance for Stress and Disinhibition of Aggression, (2) Poor Management of
Acute Crises and Provocation, and (3) Habitual Patterns of Arousal and
Aggression with Family Members. Destabilizing factors within each
stage include such variables as weak preparation for parenting (Stage I),
conditioned emotional arousal (Stage 2), and the child's increase in prob-
lem behavior (Stage 3). Compensatory factors consist of such elements
as socioeconomic stability (Stage I), improvement in child behavior
(Stage 2), and use of community restraints and services (Stage 3). The
important contributions of the Transitional Model are that it (1) describes
child abuse as a gradually unfolding interactive process rather than an
isolated phenomenon, (2) identifies both high-risk elements and protec-
tive factors critical to the occurrence of maltreatment, and (3) under-
scores the now widely accepted notion that a combination of risk factors,
and not just those existing in isolation, are necessary for the develop-
ment of abuse (see Ammerman, 1989; Starr, 1988). In addition, as with
the Social-Situational Model, the Transitional Model clearly identifies
child characteristics as a possible contributor to the development of child
abuse. Further empirical study is required, however, to examine if, in
fact, families do progress through these stages before engaging in
maltreatment.

COMMENTS

The Social-Interactional Model and the Transitional Model of abuse

elucidate the processes whereby child characteristics can contribute to
the etiology of abuse. Specifically, children with severe behavior prob-
lems or those who are difficult to manage may add to the risk of mal-
PREDISPOSING CHILD FACTORS 203

treatment. However, it is important to emphasize that these charac-

teristics alone are insufficient to bring about abuse. Rather, risk is
heightened when child factors combine with preexisting elements (e.g. ,
high levels of stress, poor coping skills, acceptance of physical punish-
ment as a disciplinary technique) in the development of maltreatment.

EARLY CHILDHOOD RISK FACTORS

Initial formulations of the role of the child in the etiology of mal-
treatment focused on infant characteristics (see Friedrich & Boriskin,
1976). Specifically, these included such factors as prematurity, low birth-
weight, difficult infant temperament, or failure to develop secure moth-
er-infant attachment. It was argued that the stress caused by the afore-
mentioned conditions strained the parent-infant (especially the moth-
er-infant) relationship that, in turn, led to abuse and neglect.
The impetus for implicating prematurity and low birthweight in the
development of child maltreatment stemmed from the disproportionate
occurrence of these conditions in abused and neglected samples. Based
upon retrospective designs, numerous reports documented the over-
representation of premature and low birthweight babies in maltreating
families (Elmer & Gregg, 1967; Fontana, 1973; Klaus & Kennell, 1970;
Klein & Stern, 1971). For example, Klein and Stern (1971) found that 23%
of their sample of abused infants had low birthweight, whereas Elmer
and Gregg (1967) noted that one third of their abused subjects received
this diagnosis. There are two proposed mechanisms through which pre-
maturity and low birthweight are posited to attribute to the develop-
ment of maltreatment. First, frequent parent-infant separations second-
ary to complications associated with prematurity lead to an erosion of
the attachment relationship or "bonding failure" between mother and
infant (Klaus & Kennell, 1970). And second, the stress related to raising a
difficult infant increases the overall risk of child maltreatment.
Preliminary studies of early separation as a contributor to maltreat-
ment provided evidence for this association (Klaus & Kennell, 1970;
Lynch, 1975; Lynch & Roberts, 1977). For example, Lynch (1975) re-
ported that, in a study of 25 abused children and their siblings, maltreat-
ed children were more likely to have experienced birth complications or
medical difficulties leading to postbirth hospitalizations than siblings.
Likewise, Hawkins and Duncan (1985) reported a high incidence of
chronic physical illnesses in a sample of substantiated child abuse cases.
On the other hand, Sherrod, O'Connor, Vietze, and Altemeier (1984)
failed to find differential patterns of early separation because of infant
204 ROBERT T. AMMERMAN

illness in abused, neglected, and non-maltreated children. Although

abused children were found to experience more illnesses than their
peers, these were most often attributed to the consequences of abuse.
Egeland and Vaughn (1981) rejected much of the research implicat-
ing the role of early separation in the etiology of maltreatment. They
argued that (1) none of these investigations provides a direct measure of
the strength of mother-infant attachment, and (2) that the retrospective
design approaches utilized by these studies leaves numerous competing
explanations for obtained findings (e.g., prenatal parental characteristics
of the mother such as not obtaining proper medical care, poor nutrition,
or substance abuse, may result in the birth complications and pre-
maturity reported in these samples). In response to the methodological
limitations of past investigations, Egeland and his colleagues (Egeland &
Brunnquell, 1979; Egeland & Vaughn, 1981) conducted a prospective
study of 267 mothers at high-risk for child maltreatment. Subjects were
from low socioeconomic status (SES) backgrounds and were recruited
during the last trimester of pregnancy. Of this sample, 32 infants were
identified as receiving inadequate home care, whereas 33 were found to
receive optimal care . Examination of birth complications and early sepa-
rations revealed no differences between optimal and nonoptimal care
groups (Egeland & Vaughn, 1981). Specifically, groups were statistically
equivalent in terms of prematurity, birthweight, number of days spent
in the hospital, delivery complications, and newborn medical problems.
The authors emphasized the relative superiority of the prospective de-
sign when compared to a retrospective approach and they stated that
"looking backward in time always provides a cause, but the inferred
linearity is misleading. It may be the case . . . that a number of abused
children are premature, but . . . the vast majority of premature children
are not subsequently abused" (p 82).
Although there are conflicting data regarding the utility of using
prematurity or low birthweight as a risk factor for child maltreatment,
Frodi (1981) has elucidated the circumstances under which prematurity
can contribute to maltreatment. She points out that premature infants
display a variety of characteristics that are perceived as aversive and
may precipitate an abusive response from caretakers. These consist of
unattractive physical features (e.g., small size, developmentally re-
tarded in growth) and a high-pitched, arhythmic cry.
Through a series of investigations, it has been demonstrated that
many parents experience negative emotional and physiological arousal
upon hearing the cry of a premature infant. For example, in one investi-
gation (Frodi et al., 1978) parents were shown videotapes of premature
and full-term infants engaged in crying. Cries of the premature infant
PREDISPOSING CHILD FACTORS 205

were dubbed onto the full-term baby's video tape, and vice versa, in
order to control for the influence of physical characteristics on parental
responding. Findings indicated increased emotional and physiological
arousal in response to the premature baby's cry as contrasted with that
of the full-term baby. Furthermore, parents reported less willingness to
interact with the premature than with the full-term infant. Further ex-
amination revealed that parents react negatively to infants labeled as
"premature" or "difficult" regardless of actual birth status (Frodi, Lamb,
Leavitt, & Donovan, 1978). Frodi and Lamb (1980) extended their re-
search on emotional and physiological responses to premature infants to
child abusers. When compared to nonabusive parents, child abusers
evidenced more pronounced increases in autonomic and emotional
arousal when presented with the cries of a premature infant. In addi-
tion, abusive parents showed similar patterns of response to a smiling
infant. Thus, abusive parents appear to view as aversive almost any
social contact with a premature infant.
Although these studies describe the mechanisms through which
prematurity may contribute to abuse, two methodological limitations
prevent drawing firm conclusions from the data. First, the analogue
nature of the aforementioned investigations do not demonstrate that
such processes are found in the natural environment. And second, it is
unclear from these findings that the negative behaviors displayed by
premature infants precede and subsequently elicit abuse from care-
takers. Rather, such aversive characteristics may develop as a function of
abuse, although they also may serve to elicit further maltreatment in the
future (Frodi, 1981).
Similar difficulty delineating cause and effect is encountered in the
study of attachment formation in maltreated infants. Attachment is the
affective and social bond between mother and infant formed via the
unique contributions of parent and child (Ainsworth, Blehar, Waters, &
Wall, 1978). Attachment is a qualitative construct that can be categorized
as secure or insecure (Sroufe & Waters, 1977). Behavioral deficits in
attachment-promoting behaviors exhibited by the mother and/or the
infant can lead to insecure attachment. Ainsworth (1980), a pioneer in
the empirical examination of attachment formation, has proposed that
insecure attachment related to parental characteristics (e.g ., unrespon-
siveness, inadequate caretaking) or aversive child behaviors (e.g., fre-
quent crying, difficult to calm) can lead to maltreatment. However, no
data are available showing that disruption in attachment leads to abuse
or neglect, but numerous studies have shown disproportionate occur-
rence of insecure attachment in maltreated infants and their mothers
(see Cicchetti, 1987), although it is most likely that such disruptions are a
206 ROBERT T. AMMERMAN

consequence rather than a cause of maltreatment (see Chapter 5 in this

volume). Well-controlled prospective research is needed to examine
more fully the possibility that insecure mother-infant attachment can
lead directly to maltreatment.
Gaensbauer and Sands (1979) outlined the temperamental charac-
teristics displayed by abused infants that may contribute to maltreat-
ment. Based on observations of mother-infant interactions, they deline-
ated the following disruptions in infant social and emotional responses
that impede the formation of secure mother-infant attachment: affective
withdrawal, lack of pleasure, inconsistency and unpredictability in affec-
tive communications, shallowness of affect communications, ambiguity
or ambivalence in affective expression, and negative affect messages.
The authors acknowledged the po ssibility that such infant behavioral
disturbances may be a function of abusive caretaking rather than a
cause, but point out the critical interdependence of the mother-infant
relationship. They stated that
disturbances in affective communication probably grow out of desynchro-
nou s, un satisfying interactions with caretakers beginn ing very early in the
child's life. Such desynchrony may result from con stitutional factors in the
child . . . or from inade quacies in the parents, or both . . . once estab lish ed ,
such characteristics take on a life of their own and actively work up on the
en vironment, including the caret akers who may have been instrumental in
th eir initial de velopment. (p. 248)

Crittenden (1985) empirically examined deviant infant behavior in

abusive, neglecting, problematic, and adequate care mother-infant
dyads. In general, maltreated infants were found to display behavior
patterns that were more difficult to manage and more disagreeable than
their non-maltreated counterparts. In a second experiment, Crittenden
(1985) provided maltreating mothers with a program designed to en -
hance their sensitivity and responsivity to their infants. Results indi-
cated that when these mothers showed gains in sensitivity, their infant's
behavior subsequently improved as well. No changes in infant behavior
were noted in families in which the abusive mother did not benefit from
treatment. These findings provide compelling evidence for the relative
primacy of mother behaviors over infant temperament in the rela-
tionship, and cast doubt on the infant's role in the development of
maltreatment.
Egeland and Brunnquell (1979) provided the only prospective re-
search data evaluating the role of child temperament in causing abuse.
In their large-scale of high-risk mothers identified during pregnancy,
Egeland and Brunnquell (1979) compared adequate care and inadequate
PREDISPOSING CHILD FACTORS 207

care mothers and their infants using a variety of assessments, including

measures of parenting attitudes, infant temperament, infant observa-
tions, and mother-infant observations. A discriminant analysis revealed
that several infant characteristics distinguished adequate and inade-
quate care mothers. These consisted of infant orientation, irritability,
and consolability. However, such other factors as maternal hostility and
negative reactions to pregnancy, were more predictive of group classifi-
cation than these infant behaviors. Thus, although infant variables' be-
haviors have some predictive power in distinguishing adequate from
inadequate care mothers, other influences appear to be more salient in
the development of maltreatment.
Relatively few empirical efforts have been directed toward the
causative contribution of behavior problems in toddlers and preschool
children to maltreatment. Despite its prominence in certain theoretical
formulations, a paucity of data exists examining this relationship. There
is no doubt that most maltreated children display extensive and varied
behavior problems (see Ammerman et al., 1986). But it is typically as-
sumed that such psychopathologies are sequelae of abuse and neglect
rather than causes of maltreatment. Also, it has been shown that child
misbehavior can precipitate a specific abusive incident (Kadushin &
Martin, 1981). Once again, this process is thought to take place in fami-
lies already predisposed to abuse. Thus, although acting out, non-
compliance, and oppositionality can be viewed as "abuse-provoking
behaviors" (deLissovoy, 1979; Rusch, Hall, & Griffin, 1986), it is unclear
to what extent such problems bring about an abusive relationship as
opposed to maintaining or exacerbating previously extant maltreatment.

COMMENTS

Findings regarding the contribution of early childhood charac-

teristics to maltreatment are equivocal. Retrospective studies underscore
the high rate prematurity, low birthweight, early separations, and dis-
rupted attachment in abused and neglected infants. Prospective studies,
which offer a methodologically stronger experimental approach, provide
little support for the hypothesis that difficult children are a major cause
of abuse or neglect. The critical question is one of cause and effect: Are
child characteristics contributors to or sequelae of maltreatment? Pro-
spective designs are the only acceptable empirical approach to examin-
ing this issue and, to date, such studies have not indicated a major role
for child characteristics in the etiology of abuse and neglect. However, it
is documented that certain early childhood behavioral factors can be
208 ROBERT T. AMMERMAN

highly aversive, particularly to abusive parents (Frodi, 1981). According

to the Transitional Model of maltreatment (Wolfe, 1987), childhood fac-
tors can lead to abuse under specific conditions related to the presence or
absence of destabilizing and compensatory influences. Therefore, early
childhood characteristics by themselves may be insufficient to elicit mal-
treatment, although within particular contexts (that have yet to be em-
pirically identified) they might lead to abuse or neglect.

COERCIVE INTERACTIONS IN THE DEVELOPMENT

AND MAINTENANCE OF ABUSE
Although findings linking child characteristics and maltreatment are
equivocal, a growing body of literature indicates that children can con-
tribute to an escalation of conflict that may lead to physical abuse. Indeed,
recent theorists have identified interactions in the parent-child dyad as
one of the most critical components in explaining the development and
maintenance of abuse (Burgess & Draper, 1988). According to this frame-
work, individual features of parent and child are less important than the
interaction between the two. Thus, Wolfe (1987) rejected single-factor
etiologic models of maltreatment and argued that "child abuse can best be
explained as the result ofaninteraction between the parent and child within
a system that seldom provides alternative solutions .. . or clear cut
restraints . . ." (p. 51).
Numerous empirical efforts have been directed toward examining
the nature of the interactions between maltreating parents and their
children (see Wolfe & St. Pierre, 1989). These studies clearly document
dysfunctional interaction patterns. In general, findings indicate that,
although abusive parent-child dyads rarely exhibit higher levels of
negative interactions, there is a paucity of reciprocal positive behaviors.
Thus, parental responding is characterized by a lack of positive social
interactions, low delivery of positive reinforcement, and restricted or
negative affect . Likewise, abused children often are withdrawn, ag-
gressive, and rarely initiate positive peer and adult contacts (Bousha &
Twentyman, 1984; Schindler & Arkowitz, 1986).
Findings from the aforementioned studies, however, do not eluci-
date the coercive and interdependent processes of escalating conflict in
maltreating parents and their children. Rather, it is possible that child
misbehavior is a direct consequence of their parent's abusive and con-
trolling management strategies. On the other hand, investigations em-
ploying sequential analytic techniques have documented the interac-
PREDISPOSING CHILD FACTORS 209

tional nature of parent-child conflicts that leads to a mutual escalation

of conflict and violence. For example, Oldershaw, Walters, and Hall
(1986) compared the interactions of 10 abusive and 10 nonabusive moth-
er-child pairs. Use of control strategies and child compliance were ob-
served in a 40-min laboratory observational assessment during which
mothers and their children simulated selected home activities. Results
indicated that abusive mothers were more likely to use power assertive
control strategies (e.g., threats, negative demand, disapproval) rather
than more positively oriented approaches (e.g. , reasoning, cooperation,
approval) when contrasted with nonabusive mothers. In addition,
abused children exhibited higher levels of disobedience and non-
compliance than their nonabused peers. Sequential analyses further elu-
cidated the tendency of abusive mothers to respond to child non-
compliance with negative control strategies, thus worsening overall
parent-child conflict.
In another study, Trickett and Kuczynski (1986) examined disciplin-
ary practices in abusive and nonabusive mothers of children aged 4-10
years . Specifically, parents completed a Parent Daily Report of child
transgressions and parental responses. Findings showed that abused
children exhibited more behavior problems than nonabused children,
and that abusive parents were more likely to use punitive disciplinary
practices when compared to nonabusive parents. Moreover, abusive
parents resorted to punitive strategies regardless of the severity of the
child's misbehavior, whereas their nonabusive counterparts onl y em-
ployed punitive techniques for more serious misbehavior. These results
underscore the coercive interaction that leads to the escalation of par-
ent-child conflict in abusive families .
Stringer and LaGreca (1985) examined the relationship between co-
ercive processes and child abuse in a hospital clinic sample of mothers.
The authors hypothesized that, according to coercion theory (Patterson,
1981)/ potential for engaging in child abuse is related to an external locus
of control and child behavior problems. Subjects were administered the
Child Abuse Potential Inventory (CAP) (Milner, 1986)/ and measures of
locus of control and child misbehavior. Findings revealed a positive
relationship between elevated scores on the CAP and perceptions of
control by others in mothers of male children. In addition, perceptions
of control by others and CAP scores were associated with conduct and
anxious-withdrawn behavior problems in children. Similar results were
not noted in mothers of female children. However, this discrepancy is
not surprising in that the coercive process appears to be more pro-
nounced in families of male rather than female children. Although this
210 ROBERT T. AMMERMAN

investigation did not use mothers engaging in documented abuse or

directly measure mother-child interactions, it does elucidate factors that
contribute to the development of coercive dynamics in families.
Reid, Patterson, and Loeber (1982) provided one of the most com-
prehensive descriptions of coercive processes in abusive families. Reid et
al. (1982) argued that the occurrence of child abuse is related to two
factors: (1) the frequency of aversive interactions between parent and
child that might lead to an abusive incident, and (2) the parent's ability
to terminate quickly such parent-child confrontations. Observational
studies have confirmed that abusive families engage in frequent and
prolonged confrontations relative to nonabusive families. For example,
Reid (1981, cited in Reid et al., 1982) conducted home observations of
distressed, distressed but not abusive, and abusive families. In general,
there was a high positive correlation between rate of child opposi-
tional/noncompliant behavior and parents engaging is threats or phys-
ically punitive behaviors across all groups. This association was most
pronounced, however, in abusive families. Patterson (1981) further dem-
onstrated that the child's acting out and parent's aversive behavior co-
vary within families, as well. Finally, Reid, Taplin, and Loeber (1981)
reported that nonabusive parents failed to terminate confrontations 14%
of the time, whereas abusive parents were unsuccessful 35% of the time.
Abusive mothers, in particular, demonstrated difficulty in discontinuing
aversive interactions, failing to end them 53% of the time.

COMMENTS

There is compelling evidence that coercive parent-child interactions

are a major component in the etiology and continuation of child abuse.
Studies in this area are noteworthy because they (1) typically utilize
well-controlled experimental designs, and (2) examine interactional and
sequential elements of abusive family dynamics rather than using a
more simplistic unidimensional formulation. The coercive approach to
understanding child abuse clearly implicates the child as a contributor to
abuse. However, findings in this area do not suggest that the child is
necessarily an instigator of abuse. Rather, it is more likely that severe
behavioral disturbance develops as a product of parental dysfunction.
Once in place, however, the pattern of escalating conflict between par-
ent and child assumes an independent existence, seemingly resistant to
change from either family member. At this point, the child plays an
active role in maintaining the abusive relationship, and outside interven-
tion is required to break the maladaptive cycle.
PREDISPOSING CHILD FACTORS 211

HANDICAP AS A RISK FACTOR FOR MALTREATMENT

A number of recent reviews have suggested that handicapped chil-
dren may be at heightened risk for abuse and neglect (Ammerman, Van
Hasselt, & Hersen, 1988b; ]audes & Diamond, 1985; White, Benedict,
Wulff, & Kelley, 1987; Zirpoli, 1986). These authors proposed that, be-
cause many handicapped or disabled children exhibit characteristics im-
plicated in maltreatment in nonhandicapped populations, they should
be viewed as being particularly susceptible to maltreatment. In general,
it is hypothesized that increased risk is manifested via three conditions:
(1) disruptions in mother-infant attachment, (2) greater stress experi-
enced by caretakers engendered by the behavior problems exhibited by
many handicapped children, and (3) heightened vulnerability of chil-
dren with disabilities to maltreatment.
There are six factors that may serve to disrupt the formation of
mother-infant attachment in many children with disabling conditions
(see Ammerman et al., 1988b). First, there is evidence to suggest that
some parents exhibit negative reactions upon the birth of a handicapped
child (Gath, 1977; Waisbren, 1980). Typical responses include disbelief,
shock, guilt, and disappointment (Emde & Brown, 1978). Although
most parents adequately adjust to this occurrence (Drotar, Baskiewicz,
Irvin, Kennell, & Klaus , 1975), some do not, and they may subsequently
experience further disturbance in the development of the parent-child
relationship. Second, some investigators have reported maternal lone-
liness, depression, and withdrawal following the birth of a handicapped
infant (Emde & Brown, 1978; Lambert & West, 1980). The clinical signifi-
cance of these symptoms, however, and their impact on parent-child
functioning, is unclear. Third, there are reports in the clinical literature
documenting hostile feelings directed by parents toward their handi-
capped children (Bauman & Yoder, 1966; Shaffer, 1964). Although em-
pirical support for this factor is lacking, it is possible that such hostility
can lead to maltreatment in handicapped children.
Fourth, several investigations have reported disruptions in the for-
mation of secure mother-infant attachment in handicapped populations
(Cicchetti & Serafica, 1981; Emde & Brown, 1978; Fox, 1988; Stone &
Chesney, 1978; Wasserman, 1986; Wasserman, Lennon, Allen, & Shil-
ansky, 1987). Such disturbances are partly a function of the behavioral
deficits (e.g., hypo or hypertonicity, unresponsiveness, lack of eye con-
tact) displayed by many disabled children, particularly those with severe
impairments. Moreover, mothers may misinterpret these behavioral def-
icits and lack of responsiveness as "disinterest" on the part of the infant.
In turn, this may result in a reciprocal decrease in attachment-promoting
212 ROBERT T. AMMERMAN

behaviors exhibited by the mother, thus further eroding the attachment

bond (Wasserman, Allen, & Solomon, 1985). A fifth negative influence
on the parent-child relationship often found in handicapped popula-
tions is the frequent early separations between the child and his or her
family because of hospitalizations secondary to medical complications at
birth. Because of several retrospective studies indicating a relationship
between early separation and maltreatment (e.g., Lynch, 1975), this
might be an additional risk factor for handicapped children. Finally, a
sixth aspect affecting the parent-child relationship in handicapped pop-
ulations are unrealistic and inaccurate expectations held by some par-
ents regarding their disabled child's development and abilities . Exces-
sive expectations and parental denial (Warren, 1977) may result in
profound disappointment and frustration as the child repeatedly fails to
meet anticipated goals (Scott, 1969). This may, in turn, increase the
probability of abusive behavior.
Numerous investigators have documented the role of stress, and
particularly the lack of adequate coping skills, in the development of
maltreatment (Browne, 1986; Egeland et al., 1980; Straus, 1980). Specifi-
cally, it is proposed that stress leads to increased frustration that can lead
to physical abuse. In particular, prolonged exposure to stress appears to
be more related to abuse than isolated stressful incident (Rohner &
Rohner, 1978). This issue is particularly critical for families with a handi-
capped child . Indeed, the introduction of a handicapped child into the
home raises parental stress levels because of increased care require-
ments, greater financial obligations, and difficult-to-manage behavior
problems (see Gallagher, Beckman, & Cross, 1983). In fact, the behavior
problems evinced by many handicapped children, and especially those
with multiple handicapping conditions (Van Hasselt, Ammerman, &
Sisson, in press), can be very aversive and difficult to manage. Some
examples of these include: rocking, eye poking, hand flapping, disrup-
tiveness, aggression, and screaming. These problems often require con-
sistent application of behavior management techniques in order to bring
them under control. Moreover, most parents require specific training to
use these techniques in an effective manner (Sisson, Van Hasselt,
Hersen, & Aurand, 1988). Thus, the chronicity and pervasiveness of the
more severely handicapped children's dysfunctions would seem to place
them at high risk for abuse (Ammerman, Lubetsky, Hersen, & Van
Hasselt, 1988a).
The final risk factor associated with maltreatment in the handi-
capped is vulnerability. Infants are often overrepresented in child abuse
reports partly because they are more vulnerable to abuse (Gelles, 1978).
This fact may be an artifact, however, in that infants also are more likely
PREDISPOSING CHILD FACTORS 213

to be injured by maltreatment when contrasted with older children, and

therefore are more likely to come to the attention of authorities. Handi-
capped children, particularly those with severe disabilities, also can be
categorized as vulnerable. Depending upon the extent and severity of
their handicap, these children may be unable to defend themselves
(Morgan, 1987). Furthermore, children with cognitive or language defi-
cits may be "easy targets" given their inability to report incidents of
maltreatment. Detection of abuse and neglect in severely handicapped
children may be further confounded by the occurrence of bruises and
abrasions resulting from accidents related to their impairment (e.g., fall-
ing down, bumping into objects). At times, distinguishing between acci-
dental and abuse-related injury may be difficult. Also, many disabled
children are more prone to neglect because of their increased care re-
quirements and medical needs (Ammerman et al., 1988b).

INCIDENCE OF MALTREATMENT IN
HANDICAPPED POPULATIONS

Although several investigations have reported a disproportionate

number of handicapped children in abused and neglected samples,
findings vary considerably. A wide range of incidence levels (4% to 70%)
has been found (Birrell & Birrell, 1968, Crittenden, 1985; Gil, 1970;
Hawkins & Duncan, 1985; E. C. Herrenkohl & R. C. Herrenkohl, 1979;
Iowa Department of Social Services, 1977;Johnson & Morse, 1968; Light-
cap, Kurland, & Burgess, 1982; Sandgrund, Gaines, & Green, 1974;
Starr, 1982). Discrepancies in results are related to differences in criteria
for what constitutes a handicap, and difficulties in identifying abuse and
neglect in handicapped children. Many of the studies employ unclear
and incomplete definitions of disabling conditions, whereas others pro-
vide no such information. Thus, samples include children with minor
physical anomalies as well as those with more severely handicapping
sensory and orthopedic handicaps. Another impediment to accurate
epidemiological investigation of the occurrence of maltreatment in these
populations is the problem of recognizing and substantiating assault. As
previously mentioned, many more severely handicapped children are
unable to understand or report the occurrence of physical and/or sexual
abuse. In addition, some physically handicapped children suffer contu-
sions and abrasions related to their handicap rather than abuse per se.
Although a large body of research has accrued suggesting greater
risk for abuse and neglect in handicapped children, a few investigators
have questioned the relationship between having a handicapping condi-
214 ROBERT T. AMMERMAN

tion and subsequent maltreatment. At issue, in general, is the child's

role in the etiology of maltreatment (Egeland & Vaughn, 1981), and, in
particular, the increased risk for abuse in handicapped children (Starr,
Dietrich, Fischhoff, Ceresnie, & Zweier, 1984). Paradoxically, it has been
suggested that the more severely disabled child is at decreased risk for
assault (Martin & Beezley, 1974; Steele, 1980). Starr et al. (1984) and
Steele (1980) base their conclusions on the methodological weaknesses
(i.e., retrospective designs) of studies linking certain child charac-
teristics (e.g., prematurity) with maltreatment. Also, Martin and
Beezley (1974) cite their clinical experience that handicapped children
are not likely targets for maltreatment. They propose that, for the par-
ent, a handicapped child's misbehavior is clearly related to his or her
impairment, and not intent. This understanding, in turn, mitigates the
frustration that may lead to abuse. Children with conditions that are less
evident to the parent (e.g., Attention Deficit Hyperactivity Disorder), it
is hypothosized, are more likely to be at risk for maltreatment (Martin &
Beezley, 1974). However, perceptions of intent appear to have little rela-
tion to the occurrence of abusive incidents (Rosenberg & Reppucci, 1983).
Rather, abusive parents are more likely to respond negatively to child
problem behaviors that are resistant to intervention regardless of inten-
tionality (e.g. , crying). In addition, although incidence rates of abuse are
higher in hyperactive children relative to the general population (Hef-
fron, Martin, Welsh, Perry, & Moore, 1987), they are not significantly
different from other clinic-referred or psychiatrically disturbed children.
There are several methodological and conceptual problems with the
aforementioned positions. First, very few of the studies on child charac-
teristics and their causal role in abuse include handicapped children in
their samples. Second, investigations that do invol ve handicapped chil-
dren primarily examine those with more minor physical anomalies (e.g.,
extra digits, cleft palate) rather than those with more severely debilitat-
ing handicapping conditions. Third, almost no multihandicapped chil-
dren are represented in these efforts. Because it is hypothesized that the
form and magnitude of behavior problems exhibited by many multi-
handicapped children (e.g ., aggression, stereotypic behaviors) play an
important role in the development of maltreatment, it is of paramount
importance that such children be carefully evaluated before conclusions
about their risk for abuse and neglect are reached (Ammerman et al.,
1988a).
Some investigations, however, have assessed the incidence of mal-
treatment in samples of handicapped children. In a retrospective analy-
sis, Diamond and ]audes (1983) reviewed medical charts of 86 children
and adolescents with cerebral palsy. Their results indicated that 9% had
PREDISPOSING CHILD FACTORS 215

been maltreated following the diagnosis of cerebral palsy, and an addi-

tional 14% were labeled as "at risk" for maltreatment using numerous
criteria thought to be related to maltreatment risk. In a more recent
effort, Ammerman, Hersen, Van Hasselt, McGonigle, and Lubetsky
(1989) examined abuse and neglect in psychiatrically hospitalized multi-
handicapped children. These children were selected for study because
of their extensive handicaps and, in most cases, severe behavioral dys-
function, that would place them at especially high-risk for abuse and
neglect. A retrospective analysis of 150 charts revealed that 39% had
been victims or warranted high suspicion of past and/or current mal-
treatment. Physical abuse was the most common form, occurring in 69%
of the subgroup that was maltreated. This was followed by neglect
(45%) and sexual abuse (36%). Of particular significance was the severity
of maltreatment. For example, 66% of the sexual abuse cases involved
penetration. Also, 40% of the sexually abused children were assaulted
by multiple perpetrators. Finally, 52% of the subgroup of maltreated
children experienced more than one form of maltreatment (e.g ., abuse
and neglect), many of them concurrently.

COMMENTS

The role of the handicapped child in the etiology of maltreatment is,

at best, unclear. Researchers are equally divided in their confidence that
disability is (Morgan, 1987) or is not (Starr et aI., 1984) a risk factor for
abuse and neglect. Unfortunately, this area has not received the re-
search attention required to resolve this issue. An overreliance on retro-
spective designs prevents the examination of a causal relationship be-
tween handicapping conditions and maltreatment. In addition, much of
the prospective research fails to examine more severely handicapped
and multihandicapped children. Thus, inferences drawn from these
data may be inapplicable to multihandicapped populations. As with the
association between other child characteristics and maltreatment, it is
unlikely that handicaps in isolation cause abuse and neglect (Ammer-
man et al., 1988b). Rather, presence of a handicap in combination with
other factors raises the overall risk for maltreatment.

SUMMARY
This chapter began with the premise that child characteristics can
contribute to the etiology of physical abuse and neglect. The framework
for this hypothesis is drawn from recent theoretical formulations (the
216 ROBERT T. AMMERMA N

Social-Interactional Model and the Transitional Model) that describe the

processes through which certain child factors increase maltreatment
risk. The impetus for research in this area comes from the pressing need
to identify populations at risk for abuse and neglect so that they can
receive preventive interventions before maltreatment ultimately devel-
ops .
Current findings do not provide strong support for the role of the
child in abuse and neglect. Prospective studies have identified several
behavioral characteristics that can contribute to maltreatment (Egeland
& Brunnquell, 1979), but the explanatory power of these variables is
minor relative to other causative factors. However, a variety of well-
controlled studies have documented that child noncompliance and op-
positionality can serve to maintain a preexisting abusive relationship
(Reid et al., 1982). In this instance, the child may have little to do with the
development of abuse, although he or she may contribute to the occur-
rence of subsequent abusive incidents. Finally, several authors have pro-
posed that presence of a severe handicap can uniquely increase the
overall likelihood of maltreatment (Ammerman et al., 1986; Morgan,
1987). Although preliminary studies document high rates of abuse and
neglect in these populations (Ammerman et al., 1989; Diamond & [au-
des, 1983), few direct empirical investigations have been conducted in
this area.
A major problem afflicting the study of child characteristics is that
child maltreatment is a multidetermined phenomenon that defies uni-
tary causal explanation. By its very nature, child abuse grows out of an
interaction between two individuals: the perpetrator and the child. The
most promising area of future research is to identify characteristics of
the interaction or parent-child relationship rather than individual fac-
tors as predictors of maltreatment. The important question then be-
comes under what conditions do which child characteristics lead to mal-
treatment? What combination of variables (related to the child, perpetra-
tor, or situation) heighten risk for abuse? Such a multivariate empirical
approach is critical at this juncture in the search for childhood risk
factors.

ACKNOWLEDGMENTS

Preparation of this chapter was facilitated in part by grant No .

G008720109 from the National Institute on Disabilities and Rehabilitation
Research, V.S. Department of Education, and a grant from the Vira I.
Heinz Endowment. However, the opinions reflected herein do not neces-
sarily reflect the position of policy of the U.S. Department of Education or
PREDISPOSING CHILD FACTORS 217

the Vira I. Heinz Endowment and no official endorsement should be

inferred. The author wishes to thank Mary [o Horgan for her assistance in
preparation of the manuscript.

REFERENCES
Ainsworth, M. D. (1980). Attachment and child abuse. In G. Gerber, C. Ross, & E. Zigler
(Eds .), Childabuse: An agenda foraction (pp. 35-47). New York: Oxford University Press.
Ainsworth, M. D. 5., Blehar, M. C.; Waters , E., & Wall, S. (1978). Patterns of attachment: A
psychological study of the strange situation. Hillsdale, NJ: Lawrence Erlbaum.
American Association for Protecting Children. (1988). Highlights of official child neglect and
abuse reporting 1986 . Denver: American Humane Association.
Ammerman, R. T. (1989). Child abuse and neglect. In M. Hersen (Ed.), Innovations in child
behavior therapy (pp. 353-394). New York: Springer.
Ammerman, R. T., Cassisi, J. E., Hersen, M., & Van Hasselt, V. B. (1986). Consequences of
physical abuse and neglect in children. Clinical Psychology Review, 6, 291-310.
Ammerman, R T., Lubetsky, M. J., Hersen, M. , & Van Hasselt, V. B. (1988a). Maltreatment
of children and adolescents with multiple handicaps: Five case examples. Journal of the
Multihandicapped Person, 1, 129-139.
Ammerman, R. T., Van Hasselt, V. B., & Hersen, M. (1988b). Maltreatment of handicapped
children: A critical review. Journal of Family Violence, 3, 53-72.
Ammerman, R T., Hersen, M. , Van Ha sselt, V. B., McGonigle, J. J., & Lubetsky, M. J.
(1989). Abuse and neglect in ps ychiatrically hospitalized multihandicapped children.
Child Abuse and Neglect, 13, 335-343.
Bauman, M. K., & Yoder, N. M. (1966). Adjustment to blindness-Reviewed. Springfield, IL:
Charles C Thomas.
Belsky, J. (1980). Child maltreatment: An ecological integration. American Psychologist, 35,
320-335.
Birrell, R , & Birrell, J. (1968). The maltreatment syndrome in children: A hospital survey.
Medical Journal of Australia, 2, 1023-1029 .
Bousha, D. M., & Twentyman, C. T. (1984). Mother-child interactional style in abuse,
neglect, and control groups: Naturalistic observations in the home. ChildDevelopment,
93, 106-114.
Browne, D. H . (1986). The role of stress in the commission of subsequent acts of child
abuse and neglect. Journal of Family Violence, 1, 289-297.
Burgess, R L., & Draper, P. (1988). A biosocial theory of family violence: The role of
natural selection, ecological instability, and coercive interpersonal contingencies. In L.
Ohlin & M. H. Tonry (Eds .), Criminal and justice-An annual review of research: Family
violence. Chicago: University of Chicago Press.
Cicchetti, D. (1987). Developmental psychopathology in infancy: Illustration from the
study of maltreated youngsters. Journal of Consulting and Clinical Psychology, 55, 837-
845.
Cicchetti, D. , & Serafica, F. C. (1981). Interplay among behavioral systems: Illustrations
from the study of attachment, affiliation , and wariness in young children with Down's
Syndrome, Developmental Psychology, 17, 36-49.
Crittenden, P. M. (1985). Maltreated infants: Vulnerability and resilience. Journal of Child
Psychology and Psychiatry, 26, 85-96.
218 ROBERT T. AMMERMAN

deLissovoy, V. (1979). Toward the definition of "abuse provoking child." Child Abuse and
Neglect, 3, 341-350.
Diamond, L. J., & [audes, P. K. (1983). Child abuse in a cerebral-palsied population.
Developmental Medicine and Child Neurology, 25, 169-174.
Drotar, D., Baskiewicz, A., Irvin , N. , Kennell, J., & Klaus, M. (1975). The adaptation of
parents to the birth of an infant with congenital malformation: A hypothetical model,
Pediatrics, 56, 710-717.
Egeland, B., & Brunnquell, D. (1979). An at-risk approach to the study of child abu se.
Journal of the American Academy of Child Psychiatry, 18, 219-236.
Egeland, B., & Vaughn, B. (1981). Failure of "bond formation" as a cause of abuse, neglect,
and maltreatment. American Journal of Orthopsychiatry, 51, 78-84.
Egeland, B., Breitenbucher, M., & Rosenberg, D. (1980). Prospective study of significance
of etiology of child abuse. Journal of Consulting and Clinical Psychology, 48, 195-205.
Elmer, E., & Gregg, G. S. (1967). Developmental characteristics of abused children. Pedi-
atrics, 40, 596-602 .
Emde, R. N., & Brown, C. (1978). Adaptation to the birth of a Down 's Syndrome infant.
Grieving and maternal attachment. Journal of the American Academy of ChildPsychiatry,
17, 299-323.
Fontana, V. J. (1973). The diagnosis of the maltreatment syndrome in children. Pediatrics,
51, 780-782.
Fox, N. A. (1988). The effect of prematurity and postnatal iIlness on the mother-infant
interaction: A study of a high-risk dyad. In E. J. Anthony & C. Chiland (Eds.), Thechild
in his family: Perilous development: Child raising and identity formation under stress (pp.
211-222). New York: Wiley.
Friedrich, W. N., & Benskin , J. A. (1976). The role of the child in abuse: A review of the
literature. American Journal of Orthopsychiatry, 46, 580-590.
Friedrich, W. N., & Einbender, A. J. (1983). The abu sed child : A psychological review.
Journal of Clinical Child Psychology, 12, 244-256.
Frodi, A. M. (1981). Contribution of infant characteristics to child abuse. AmericanJournal of
Mental Deficiency, 85, 341-349.
Frodi, A. M., & Lamb, M. E. (1980). Child abusers' responses to infant smiles and cries.
Child Development, 51, 238-241.
Frodi, A. M., Lamb, M. E., Leavitt, L., & Donovan, W. (1978). Fathers' and mothers'
responses to infant smiles and cries . Infant Behaviorand Development, 1, 187-198.
Frodi, A. M., Lamb, M., Leavitt, L., Donovan, W., Neff, c., & Sherry, D. (1978). Fathers'
and mothers' responses to the appearance and cries of premature and normal infants.
Developmental Psychology, 14, 490-498.
Gaensbauer, T. J., & Sands, K. (1979). Distorted affective communications in abus-
ed /neglected infants and their potential impact on caretakers. Journal and the American
Academy of Child Psychiatry, 18, 236-250.
Gaines, R., Sandgrund, A., Green, A. H., & Power, E. (1978). Etiological factors in child
maltreatment: A multivariate study of abusing, neglecting, and normal mothers. Jour-
nal of Abnormal Psychology, 87, 531-540.
Gallagher, J. J., Beckman, P., & Cross, A. H . (1983). Families of handicapped children:
Sources of stress and its amelioration. Exceptional Children, 50, 10-19 .
Garbarino, J. (1976). A preliminary study of some ecological correlates of child abuse: The
impact of socioeconomic stress on mothers. Child Development, 47, 178-185.
Gath, A. (1977). The impact of an abnormal child upon the parents. British Journal of
Psychiatry, 130, 405-410.
PREDISPOSING CHILD FACTORS 219

Gelles, R. J. (1973). Child abuse as psychopathology: A sociological critique and reformula-

tion . American Journal of Orthopsychiatry, 43, 611-621.
Gelles, R. J. (1978). Violence toward children in the United State s. American Journal of
Orthopsychiatry, 48, 580-592.
Gil, D. (1970). Violenceagainst children: Physical child abuse. Cambridge: Harvard University
Press.
Hawkins, W. E., & Duncan, D. F. (1985). Children's illnesses as risk factors for child abuse.
Psychological Reports, 56, 638.
Heffron, W. M., Martin, C. A., Welsh, R. J., Perry, P., & Moore, C. K. (1987). Hyperactivity
and child abuse. Canadian Journal of Psychiatry, 32, 384-386.
Herrenkohl, E. C,; & Herrenkohl, R. C. (1979). A comparison of abused children and the ir
nonabused siblings. Journal of the American Academy of Child and Adolescent Psychiatry,
18, 260-269.
Iowa Department of Social Services . (1977). Statistical data on childabuse cases. Report Series
A-4.
[audes, P. K. , & Diamond, L. J. (1985). The handicapped child and child abuse. ChildAbuse
and Neglect , 9, 341-347.
Johnson, B., & Morse, H. (1968). Injured children and their parents. Children, 15, 147-152.
Kadushin, A., & Martin, J. A. (1981). Childabuse: An interactional event. New York: Colum-
bia University Press.
Kempe, C. H ., Silverman, F. N ., Steele, B. F., Droegemueller, vt., & Silver, H. K. (1962).
The battered child syndrome. Journal of American Model Association, 181, 105-112.
Klaus, M., & Kennell, J. (1970). Mothers separated from their newborn infan ts. Pediatric
Clinics of North America, 17, 1015-1037 .
Klein, M., & Stern, L. (1971). Low birth weight and the battered child syndrome. American
Journal of Diseases in Childhood, 122, 15-18.
Lambert, R., & West, M. (1980). Parenting styles and the depressive syndrome in con-
genitally blind individuals. Journal of Visual Impairment and Blindness, 74, 333-337.
Lightcap, J. L., Kurland, J. A., & Burgess, R. L. (1982). Child abuse: A test of some
predictions from evolutionary theory. Ethology and Sociobiology, 3, 61-67.
Lutzker, J. R., & Rice, J. M. (1984). Project 12-Ways: Measu ring outcome of a large in-h ome
service for treatment and prevention of child abuse and neglect. Child Abuse and
Neglect, 8, 519-524.
Lynch, M. A. (1975). Ill-health and child abuse. Lancet, 2, 317-319.
Lynch , M. A., & Roberts, J. (1977). Predicting child abuse: Signs of bonding failure in the
maternity hospital. British Medical Journal, 1, 624-626 .
Macfarlane, K., Waterman, J., Conerly, S., Damon, L., Durfee, M., & Long , S. (Eds.) .
(1986) . Sexualabuse of young children. New York: Guilford Press.
Martin, H. P., Beezley, P. (1974). Prevention and the consequences of child abuse. Journal of
Operational Psychiatry, 6, 68-77.
Maurer, A. (1974). Corporal punishment. American Psychologist, 29, 614-626.
Milner, J. S. (1986). The child abuse potential inventory manual (2nd ed .). Webster, NC :
Psytec .
Morgan, S. R. (1987). Abuse and neglect of handicapped children. Boston : Little, Brown .
Oates, K. (Ed .). (1982). Child abuse: A community concern . New York: Brunner/Mazel.
Oldershaw, L., Walters, G. C.; & Hall, D. K. (1986). Control strategies and noncompliance
in abusive mother-child dyads: An observational study. Child Development, 57, 722-
732.
Parke, R. D., & Collmer, C. W. (1975). Child abuse: An interdisciplinary analysis: In E. M.
220 ROBERT T. AMMERMAN

Hetherington (Ed.), Review of childdevelopment research (Vo!. 5, pp . 509-590). Chicago:

University of Chicago Press.
Patterson, G. R. (1981). Families of antisocial children: An interactional approach. Eugene, OR:
Castalia Publ ishing.
Reid, J. B., Taplin, P., & Loeber, R. (1981). A social interactional approach to the treatment
of abusive families . In R. B. Stuart (Ed .), Violent behavior: Social learning approaches to
prediction, management, and treatment (pp . 83-101). New York: Brunner/Maze!.
Reid , J. B., Patterson, G. R., & Loeber, R. (1982). The abused child : Victim, instigator, or
innocent bystander? Nebraska Symposium on Motivation, 29, 47-68.
Rohner, R. P., & Rohner, E. C. (1978). A multivariate model for the study of parental acceptance-
rejection and childabuse. Storrs, CT: University of Connecticut.
Rosenberg, M. S., & Reppucci, N. D. (1983). Abusive mothers: Perceptions of their own
and the ir children's behavior. Journal of Consulting and Clinical Psychology, 51, 674-682.
Rusch, R. G., Hall, J. c., & Griffin, H . C. (1986). Abuse-provoking characteristics of
institutionalized mentally retarded individuals. American Journal of Mental Deficiency,
90, 618-624 .
Sandgrund, A., Gaines, R. , & Green, A. H. (1974). Child abuse and mental retardation: A
problem of cause and effect. American Journal of Mental Deficiency, 79, 327-330.
Schindler, F., & Arkowitz, H. (1986). The asses sment of mother-child interactions in phys-
ically abusive and nonabusive families. Journal of Family Violence, 1, 247-258.
Scott, R. A. (1969). The socialization of blind children. In D. Goslin (Ed.), Handbook of
socialization theory and research. Chicago: Rand McNally.
Shaffer, H . R. (1964). The "too-cohesive" family: A form of group pathology. International
Journal of Social Psychiatry, 10, 266-275 .
Sherrod, K. B., O'Connor, S., Vietze, P. M. , & Altemeier, W. A. (1984). Child health and
maltreatment. Child Development, 55, 1174-1183.
Sisson, L. A., Van Hasselt, V. B., Hersen, M., & Avrand, J. (1988). Tripartite behavioral
intervention to reduce stereotypic and disruptive behaviors in young multihandi-
capped children. Behavior Therapy, 19, 503-526 .
Spinetta, J. J., & Rigler, D. (1972). The child abusing parent: A psychological review.
Psychological BulIetin, 77, 296-304.
Sroufe, L. A., & Waters, E. (1977). Attachment as an organizational construct. Child Devel-
opment, 48, 1184-1199.
Starr, R. H ., [r., (1982). A research-based approach to the prediction of child abuse. In R.
H. Starr, Jr. (Ed.), Child abuse prediction: Policy implications. Cambridge, MA: Ballinger.
Starr, R. H ., Jr. (1988). Physical abuse of children. In V. B. Van Hasselt, R. L. Morrison, A.
S. Bellack, & M. Hersen (Eds.), Handbook of family violence (pp . 119-155). New York:
Plenum Press .
Starr, R. H ., [r., Dietrich, K. N ., Fischhoff, J., Ceresnie, S., & Zweier, D. (1984). The
contribution of handicapping conditions to child abuse. Topics in Early Childhood Special
Education, 4, 55-69.
Steele, B. (1980). Psychodynamic factors in child abuse. In C. H. Kempe & R. E. Helfer
(Eds .), The battered child (3rd ed., pp. 49-85). Chicago : University of Chicago.
Stone, N. W., & Chesney, B. H . (1978). Attachment behaviors in handicapped infants.
Mental Retardation, 16, 8-12.
Straus, M. A. (1980). Stress and child abuse. In. C. H . Kempe and R. E. Helfer (Eds .), The
battered child (3rd ed. , pp. 86-102). Chicago: University of Chicago Press .
Stringer, S. A. , & LaGreca, A. M. (1985). Correlates of child abuse potentia!. Journal of
Abnormal Child Psychology, 13, 217-226 .
PREDISPOSING CHILD FACTORS 221

Trickett, P. K., & Kuczynski, L. (1986). Children's misbehaviors and parental discipline
strategies in abusive and nonabusive families . Developmental Psychology, 22, 115-123.
Van Ha sselt, V. B., Ammerman, R. T., & Sisson, L. (in press). Physical disability. In A. S.
Bellack, M. Hersen, & A. E. Kazdin (Eds.), International handbook of behavior modification
and therapy (2nd ed .), New York: Plenum Pre ss .
Waisbren, S. E. (1980). Parents' reactions after the birth of a developmentally disabled
child. American Journal of Mental Deficiency, 84, 345-351.
Warren, D. A. (1977). Blindness and earlychildhood development . New York: American Foun-
dation for the Blind .
Wasserman, G. A. (1986). Affective expression in normal and physically handicapped
infants: Situational and developmental effects . Journal of the American Academy of Child
Psychiatry, 25, 393- 399.
Wasserman, G. A., Allen, R., & Solomon, C. R. (1985). At-risk toddlers and their mothers:
The special case of physical handicaps. Child Development, 56, 73-83.
Wasserman, G. A., Lennon, M. C.; Allen, R. & Shilansky, M. (1987). Contributors to
attachment in normal and physically handicapped infants. Journal of the American
Academy of Child and Adolescent Psychiatry, 26, 9-15.
White, R., Benedict, M. I., Wulff, L., & Kelley, M. (1987). Physical disabilities as risk factor s
for child maltreatment: A selected review. American Journal of Orthopsychiatry, 57, 93-
101.
Wolfe, D. A. (1985). Child abusive parents: An empirical review and analysis. Psychological
Bulletin, 97, 462-482.
Wolfe, D. A. (1987). Childabuse: Implications for childdevelopment and psychopathology. New-
bury Park, CA: Sage.
Wolfe, D. A., & St. Pierre, J. (1989). Child abuse and neglect. In T. H . Ollendick & M.
Hersen (Eds .), Handbook of child psychopathology, (2nd ed ., pp. 377-398). New York:
Plenum Press .
Zirpoli, T. J. (1986). Child abuse and children with handicaps. Remedial and Special Educa-
tion, 7, 39-48.
 PART IV

PREVENTION AND TREATMENT

 CHAPTER 9

PREVENTION PROGRAMS
MAXINE R. NEWMAN AND JOHN R. LUTZKER

INTRODUCTION
A discussion of the prevention of child abuse or neglect, as an y discus-
sion of prevention, requires definition. In a seminal book Principles of
Preventive Psychiatry (1964), Caplan defined three types of prevention,
some of which were already in use by psychiatrists and psychologists as
early as the mid-1960s: primary prevention, secondary prevention, and
tertiary prevention. In this chapter, we will discuss these three levels of
prevention generically, give a brief historical overview of past efforts to
identify and prevent child abuse or neglect, de fine prevention in the
area of child abuse or neglect, specificall y when the intervention is
intended to: prevent further abuse or prevent first abuse, con sider some
programs designed to prevent further abuse and others which focus on
intervening before abuse occurs, and provide some recommendations
for future prevention programs.

MAXINE R. N EWMAN AND JOHN R. LUTZKER • Department of Psychology, University of

Judaism , Los Angeles, Californ ia 90077.

225
226 MAXINE R. NEWMAN and JOHN R. LUTZKER

PREVENTION LEVELS

TERTIARY PREVENTION

Tertiary prevention has been described as a process that has the

community-at-Iarge as a client (Rappaport, 1977). The aim of tertiary
prevention is to lower the rate of emotional or behavioral dysfunction in
a given community. This label, then, would cover large-scale community
programs directed toward the rehabilitation of individuals already diag-
nosed as suffering from a mental, emotional, or behavioral problem. The
goals of tertiary programs are to reduce the severity and duration of
these disorders. Caplan (1964) argued that the terms rehabilitation and
tertiary prevention are not synonymous. For Caplan, the term rehabilita-
tion refers to work that is done with an individual client , whereas terti-
ary prevention implies programs that intervene in an entire community.
Thus, Caplan's definition covers large-scale service programs that effec-
tively reach the total population of identified mental patients within a
community. For example, the Joint Commission on Mental Health and
Illness (1961) and the Joint Commission on Mental Health of Children
(1970) were involved in tertiary prevention when they recommended
removing as few patients as possible from their homes and their com-
munities (Rappaport, 1977). The purpose of these decisions was to stop
the process that Cumming and Cumming (1957) had labeled "closing
the ranks," that is, the process whereby individuals living in a system, in
which a member is removed, "close ranks" and another individual takes
the identified patient's role in the system. The longer the duration of
removal, the more difficult it is for the identified patient to resume a role
within the system because it has been filled. Therefore, tertiary preven-
tion deals with a policy of creating local community facilities for day-
and-night care, including hospitals in which the duration of removal
from the home or community is minimal in the belief that the shorter the
removal, the less likely the chance of long-term problems. These tertiary
prevention programs include an assumption that there will be a reduc-
tion in such problems across the community as a result of the interven-
tion stance.
Further, tertiary prevention has implications for the quality of hos-
pital or institutional care, should such care be necessary. Following the
goals of tertiary prevention, an institution would provide much more
than custodial care. When the stated or unstated goal is to get the pa-
tient back into the community as soon as possible, then rehabilitation
programs, "milieu" or socioenvironmental treatment, token economies
PREVENTION PROGRAMS 227

designed to teach adequate social functioning, use of "significant oth-

ers" as volunteers, use of nonprofessional mental health workers in
hospital and in the community, halfway houses, and vocational training
programs are offered, all of which are examples of tertiary prevention
(Rappaport, 1977).

SECONDARY PREVENTION

Although tertiary prevention is directed at large-scale community-

or institution-based programs geared toward keeping diagnosed pa-
tients within their homes or communities and providing the necessary
care and training to accomplish this end, secondary prevention has as its
goal the early identification of and intervention into problems before they
become major mental illnesses. In this context, Rappaport (1977) defined
secondary prevention programs as those that are intended to reduce the
rate of a given disorder by lowering its prevalence in an identified high-
risk population. Thus, in secondary prevention the emphasis is on early
identification of the problem and the risk factors in the life cycle of an
individual. For example, secondary prevention programs are often
aimed at children in order to provide an "enhanced opportunity to cope
with developmental tasks" (Rappaport, 1977, p. 64). The goal of such
programs is to prevent problems that may develop and thereby reduce
the incidence and the prevalence of mental dysfunction. Although sec-
ondary prevention is most often geared toward the detection of risk
factors in children, it can also mean such detection in adults who are at
risk before their problems develop into severe disorders .
Caplan (1964) suggested that sharpening diagnostic tools and en-
couraging early referral are strategies that can be effective in early detec-
tion of risk factors . For example, Rappaport (1977) noted that during
World War II the United States Army used the Minnesota Multiphasic
Personality Inventory (MMPI) as a prescreening device to detect pos-
sible problems in inductees before they were sent to training or combat
situations. Although he emphasized the shortcomings of such large-
scale testing and the sacrifices in accuracy that may occur, he, neverthe-
less , recommended them to maximize efficiency and to get a global
identification of problem groups. Cowen, Dorr, Clarfield, Kreling , Me-
Williams, Pokracki, Pratt, Terrell, and Wilson (1973) developed a brief
ll-item scale that is designed so that teachers can record the symptom
frequency of dysfunctional classroom behaviors among elementary
school children that is another example of sharpening diagnostic tools to
228 MAXINE R. NEWMAN and JOHN R. LUTZKER

detect risk factors before problems actually develop. An example of

encouraging early referral to treatment if a problem is suspected is the
use of public education through the mass media (Rappaport, 1977). Rap-
paport (1977) further suggested consultation with physicians, psychol-
ogists, teachers, and clergy, and recommended that so-called walk-in
clinics should be part of every hospital's services . Next, he prescribed
gearing the diagnosis to specific treatment steps rather than separating
diagnosis from treatment planning. Caplan (1964) urged that no diag-
nosis or labeling be done unless there is prompt and effective treatment
available. He pointed out that labeling maladaptive behaviors may, in
and of itself, create problems for the individual who is thus labeled.
Finally, Rappaport (1977) cautioned professionals to withhold diagnosis
and intervention in those types of cases in which "spontaneous recov-
ery" is likely. Encouraging the collection of such data is a way of making
community mental health facilities more accountable (Rappaport, 1977).

PRIMARY PREVENTION

Caplan (1964) referred to primary prevention as a "community con-

cept"; that is, it does not have as its goal the reduction of risk to any
given individual. Rather, primary prevention programs seek to reduce
the rate of new cases of disorders in a population over time by coun-
teracting adverse influences before they have a chance to produce dys-
function. Primary prevention programs seek to lower the risk for a
whole population. In this situation, some members of the population
may become ill, but their numbers will be reduced in the population as a
whole.
Such a definition perse involves mental health professionals in areas
outside their customary practice. For example, to meet primary preven-
tion goals a professional might become involved in political or social
action. Further, the professional, in this instance, is encouraged to look
at those factors that lead to health as well as those that lead to illness.
And illness, in the context of primary prevention, is viewed as having a
much wider scope than is usual, thereby inspiring professionals to inter-
vene in areas that include societal problems or problems of living that
might not, in the usual course of their work, fall into their province
(Caplan, 1964). Rappaport (1977) concluded that primary prevention
programs must focus on identifying environmental factors that influ-
ence a community. Some of these factors may be harmful and may
therefore decrease a community's ability to cope with the stresses of life;
others may be helpful and increase the coping ability of the community.
Therefore, the goals of primary prevention programs are to produce
PREVENTION PROGRAMS 229

optimum outcomes for a community, to discover what negatively influ-

ences a community, and provide treatment and/or social-political action
to lower negative circumstances, to increase positive forces, and to re-
duce the incidence and prevalence of dysfunction in a community.

HISTORICAL OVERVIEW

CHILD ABUSE BEFORE THE BATTERED CHILD SYNDROME

During the Hellenistic period, the Greek, Soranus, advocated infan-

ticide based upon a list of factors similar to the now-known risk factors
for child abuse and neglect. In A .D. 900, Rhazes, a Persian, speculated
that in cases of "prominence of the umbilicus" and "hernia of children,"
the child may have been intentionally injured. Writings from seven-
teenth and eighteenth century sources indicate knowledge of willful
abuse of children by teachers and family members (Lynch, 1985). During
the nineteenth century, experts were divided between two positions in
cases of child maltreatment; some pleaded eloquently for medical recog-
nition of the problem and included calls to action in their writings,
whereas others, describing injuries and/or the results of neglect, were
content to attribute such phenomena to unknown causes. It was during
this same period that some physicians were instrumental in founding
the English Society for the Prevention of Cruelty to Children, which
brought to public and judicial attention 762 cases of child abuse and
succeeded in obtaining 120 convictions. During this period the link be-
tween alcoholism and child abuse was openly discussed (Lynch, 1985).
In the United States, the American Society for the Prevention of Cruelty
to Animals gave rise to the New York Society for the Prevention of
Cruelty to Children (Carstens, 1921). Schultz (1968) suggested that be-
cause the child protective movement grew out of the animal protective
movement, the initial concern was with physical abuse, and the goal
was to remove the child from the abusive environment. At the turn of
the century, a new view of child protection was articulated. Fostered
primarily by the Massachusetts Society for the Prevention of Cruelty to
Children, the emphasis turned from removal and punishment, which
were chiefly the province of the police, to interventions designed to
reduce environmental causes of abuse and neglect and to promote im-
proved conditions of family life, which were the province of humane
societies (Schultz, 1968). Carstens (1911-1912), the general secretary of
the Massachusetts Society for the Prevention of Cruelty to Children,
looked to social workers to provide programs dealing with infant mor-
230 MAXINE R. NEWMAN and JOHN R. LUTZKER

tality, birth registration, child labor, pauperism, dependency, and vene-

real infection-that is, a "complete social program." Along with the
Massachusetts society, other groups joined in the call for social pro-
grams to reduce poverty, disease, and dependency. After World War I,
the New York Society for the Prevention of Cruelty to Children joined its
more liberal colleagues by including the concepts of rehabilitation of
perpetrators and reconstruction of family life into its program (Carstens,
1921). During the Great Depression and the years of social programming
that followed, professionals involved in the protection of children no
longer focused on law enforcement, but rather on rehabilitative inter-
ventions combined with a commitment to extensive social action.
After World War II, Caffey (1946) wrote of the continued reluctance
of physicians to draw the "obvious conclusion" from the association of
subdural hematomas and fractures in children. Although many authors
writing from the 1930s through the 1950s still did not connect the charac-
teristic X-ray appearances of injuries in very young children as being
caused by parental abuse (Bakwin, 1956; Barmayer, Alderson, & Cox,
1951; Silverman, 1952; Snedecor, Knapp, & Wilson, 1935), others began
to discuss the possibility of children being willfully mistreated by par-
ents (Fisher, 1958), criminally neglected by parents (Kempe & Silver,
1959), and murdered by parents or caretakers (Adelson, 1969). By calling
attention to the resistance of physicians and social workers to acknowl-
edge the diagnosis (Elmer, 1960) and by describing a project in which
legal authorities and hospital staff cooperated as advocates to protect
children (Boardman, 1961), researchers in child abuse further set the
stage for Kempe's seminal paper in 1961.

THE BATTERED CHILD SYNDROME

In 1961, at the meeting of the American Academy of Pediatrics, C.

Henry Kempe and his associates at the Denver Medical Center first
proposed the diagnosis of child abuse and neglect in a symposium en-
titled, "The Battered Child Syndrome" (Kempe, Silverman, Steele,
Droegemuller, & Silver, 1962). Kempe and his colleagues clearly stated
that
the radiologic manifestations of trauma are specific and the metaphysical
lesions in particular occur in no other disease of which we are aware . (p. 23)

Although physicians had previously relied on such diagnoses as

scurvy, syphillis, osteogenisis imperfecta, infantile cortical hyperostosis,
paraplegia, and congenital indifference to pain, after Kempe et al. (1962)
it became difficult to avoid the obvious distinctions between the patho-
PREVENTION PROGRAMS 231

genic manifestations of these conditions and the traumas inflicted in

cases of child abuse and neglect (Lutzker & Newman, 1986). In this
historic presentation, Kempe and his colleagues further suggested man-
agement strategies, including the controversial issue of reporting such
cases to authorities. They further recommended strategies to communi-
cate therapeutically with perpetrators. Finally, they urged physicians
who had been disinclined to believe that parents could ever inflict such
pain on their children, to reevaluate the evidence, and accept their duty
to protect the child under their care .

AFTER KEMPE

Although it had become clear that the reporting of child abuse

and/or neglect was crucial before any treatment or intervention could
occur, there were differences between the Children's Bureau, the Ameri-
can Medical Association, and the American Humane Association over
who should report (physicians only, or others who came in professional
contact with children), to whom reports should be made (the police or a
protective agency), and what kinds of abuse and neglect were to be
reported (physical only or neglect and other forms of abuse) (Paulsen,
1968). However, by 1967, all 50 states had enacted some form of report-
ing law, and by 1973, a mandatory reporting law that required reports in
all cases of suspected abuse and that protected the mandated reporter
from reprisal for any action was in place in every state in the United
States (Lucht, 1975).
Over the years, the medical model of child abuse and neglect has
grown in acceptance, and child abuse and neglect have even been de-
scribed as an "infectious disease" (Child Protection Report, 1976). This
emphasis tends to neglect the social and environmental stressors that
were recognized in earlier years as contributing factors in child abuse
and neglect (Carstens, 1911-1912; Carstens, 1921; Schultz, 1968). Only
recently have health professionals begun to integrate the two points-of-
view and to recognize that child abuse and neglect are ecobehavioral
problems influenced by parent characteristics, child characteristics, and
socioenvironmental factors (Lutzker & Newman, 1986).

PREVENTION IN CHILD ABUSE AND NEGLECT

In the area of child abuse and/or neglect, prevention efforts may
focus on reducing or eliminating further abuse in families already identi-
fied as abusive or neglectful, or actually preventing abuse or neglect
232 MAXINE R. NEWMAN and JOHN R. LUTZKER

before the problem occurs. In the first instance, prevention is likely to be

tertiary or secondary; in the second, it will be primary or secondary. An
emphasis was found in the literature on prevention programs that focus
on sexual abuse of adolescents or programs dealing with either physical
or emotional abuse.
A five-part tertiary prevention strategy emerged from a working
conference on "Preventing Child Sexual Abuse: A Focus on the Potential
Perpetrator" (Cohn, 1986).The following 10 points comprise the rationale
for such a strategy: (1) child sexual abuse is a complex problem requiring
multiple prevention strategies; (2) knowledge about abuse and preven-
tion is not generally based on empirical findings; (3) because there is no
data-based profile of the sexual abuser, consequently, prevention efforts
cannot be geared to a specific group of potential perpetrators; (4) since
sexual abuse does not only occur after age 21, adolescents and even
younger children should also be targeted; (5) sexual abuse is not only an
issue of power, thus such areas as sexual ideas, beliefs, misconceptions,
and preferences must also be addressed; (6) there are no strongly voiced
taboos in our society against molesting children, and often the values and
messages transmitted through the media may even appear to condone
behaviors that might lead to sexual abuse; (7) children do not know how
to resist abuse; (8) children are sometimes placed in nonprotective en-
vironments; (9) there is no single law and no single profession that can
handle a problem so deeply embedded in our societal values as sexual
abuse; and (10) it is essential to increase public understanding of the
problem and get public support for prevention programs. Given these
assumptions, Cohn (1986) reported five prevention methods recom-
mended by the conference: (1) quality sex education programs for teens
and preteens; (2) training professionals and others who work with chil-
dren how to identify and help abused children; (3) providing education to
parents on attachment, bonding, appropriate and inappropriate touch-
ing, how to identify inappropriate behaviors in mates or others, how to
identify abuse and help their children; (4) ensuring that all institutions
and programs that serve children offer children training in self-awareness
and self-protection; and (5) changing media messages for adults and
adolescents to say : "Child sexual abuse is a crime; there is help out there;
abuse is a chronic problem unless you get help; children get hurt when
you sexually abuse them; children cannot consent to this kind of behav-
ior." The message to children should say: "It's okay to say no; it's not your
fault; reach out for help should this begin to happen to you; help is
available out there for you."
As we previously mentioned, goals of secondary prevention efforts
include the early identification of high-risk individuals or groups before
PREVENTION PROGRAMS 233

abuse occurs, the reduction of the rate of abuse in high-risk populations,

and the identification of risk factors for abuse and/or neglect. For exam-
ple, Hagenhoff, Lowe , Hovell, and Rugg (1987) recommended a social
learning theory approach to the prevention of teenage pregnancy.
Hagenhoff and her colleagues (1987) suggested that Bandura's (1977)
self-efficacy model may be useful in addressing this growing societal
problem. They noted the following factors that govern the use of con-
traceptives by teens: (1) accessibility; (2) social/cultural factors ; (3) family
factors; (4) social skills and peer influence; (5) life career goals; (6) knowl-
edge; and (7) expectations of personal mastery. Because such programs
would be directed to a high-risk population, they would be considered
secondary prevention interventions.
Another secondary prevention approach was described by Englert,
Marneffe, Soumenkoff, and Hubinont (1985) regarding the use of tubal
sterilization in a population at risk for abusing or neglecting their chil-
dren. In their study, Englert et al. (1985) noted the following precondi-
tions for tubal sterilization of regular maternity patients at the prenatal
clinic at Saint-Pierre Hospital in Brussels: (1) the woman must be at least
30 years old; (2) there must be two living children; (3) there must be a
stable communal background of at least 5 years; and (4) the couple must
give written consent. In a comparison of 19 sterilized high-risk women
and 18 regular maternity patients, Englert et al. (1985) found that the
high-risk women differed from the regular maternity population: (1)
they tended to be younger; (2) they tended not to be in a stable commu-
nal relationship; and (3) their children, if alive, were often either aban-
doned or in foster homes. Englert et al. (1985) concluded that there may
be some institutional abuse in such a secondary prevention policy. Tubal
sterilization is, "an aggressive act upon a woman's reproductive sys-
tem ." However, a physician has an ethical dilemma when serving the
mother and the unborn child . If the physician's ultimate goal is "famil y
health," then such ethical conflicts must be confronted.
Azar, Barnes, and Twentyman (1988) discussed secondary preven-
tion with a focus on the characteristics of the abused child, assessment
of the physically abused child, and treatment approaches for the abused
child. They noted that the physically abused child appears to show
problems in five areas: (1) neurological; (2) cognitive; (3) social skills; (4)
peer interaction; and (5) behavioral dysfunction. Azar et al. (1988) sug-
gested that little data-based literature currently exists regarding the as-
sessment of such children and recommend such strategies as (1) a re-
view of social service and school records to establish patterns. For
example, abuse occurring at the end of the month may indicate financial
difficulties in the child's family; abuse occurring after pay da y may point
234 MAXINE R. NEWMAN and JOHN R. LUTZKER

to some form of parental substance abuse. (2) Thorough medical and

neurological examinations, including such areas as speech, language,
and articulation problems. (3) Careful cognitive testing done in a man-
ner that is sensitive to the particular population. For example, an abused
child may become anxious with an adult psychologist in a closed room;
with the door open and extra time spent on establishing rapport and a
safe environment, results may change. (4) Collection of behavioral data
by asking parents or caregivers to complete standardized problem
checklists. With the exception of foster care placement, treatment has
seldom been focused directly on the abused child. Although Azar and
her colleagues (1988) suggested that multiple interventions, such as day
care, individual and group counseling, and parent training, may be
employed, there has not, to date, been extensive, empirical study of the
impact of behavioral interventions with abused and/or neglected chil-
dren. Studies that have been done include one that compared the effec-
tiveness of peer and teacher prompts on the socially withdrawn behavior
of 36 maltreated preschool children (Fantuzzo, Azar, & Twentyman,
1985) and another that examined the effectiveness of bibliotherapy (Par-
dek & Pardek, 1984).
Browne (1986) discussed the role of stress in the commission of
subsequent acts of child abuse and/or neglect. She found that stressful
situations, along with the seriousness of the first abuse, significantly
influenced the explanation of the commission of further abuse.
Family violence is another descriptor that has been found to be
associated with individuals who are at high-risk for abuse and neglect
(White, Snyder, Bourne, & Newberger, 1987). White and her colleagues
(1987) asserted that, although the focus in the child abuse and neglect
literature is often on the maltreatment of children, it may be more accu-
rate to conceptualize domestic violence as a family problem. For exam-
ple, children may be victimized by watching other family members
being attacked rather than being attacked themselves. They also pointed
out the difference between factors that may have clinical significance as
opposed to factors that may have research significance. As an example,
they suggested that researchers may argue that parental alcohol abuse
occurs only in a small percentage of child abuse and neglect cases,
whereas the clinician may see families in which parental alcohol abuse
may be a highly significant factor in the maltreatment of their children.
White, et al. (1987) also pointed out the differences between the rosy
picture of family life painted by the media and the possibility that family
members may actually cause harm to each other. The y cited the seminal
work by Straus, Gelles, and Steinmetz (1980) in their book, Behind Closed
Doors, which brought the incidence of family violence out of the societal
PREVENTION PROGRAMS 235

closet. They noted the Bittner and Newberger (1981) model for under-
standing child abuse in which predisposing factors to family violence
were described as (1) child-produced stresses (handicapped, retarded,
difficult, hyperactive, foster child); (2) family stresses (poverty, unem-
ployment, isolation, poor housing, relationship difficulties, parent-
child problems, inappropriate child-rearing style); and (3) parent-pro-
duced stresses (low self-esteem, abused as a child, depression, sub-
stance abuse, psychiatric disorder, ignorance of child-rearing, unrealistic
expectations). These factors are then influenced by a triggering situa-
tion, such as a discipline problem, substance abuse, an argument or
family conflict, or some acute environmental problem, followed by mal-
treatment by one family member toward another.
Identifying a similar population, Meredith, Abbott, and Adams
(1986) found that, as the use of physical violence among family members
increased, there were significant decreases in the family perception of
family strengths, marital satisfaction, and parental satisfaction. This
highlights the high risk for abuse or neglect in families who practice
physical violence toward each other.
Primary prevention methods, similar to tertiary prevention mea-
sures, have a community focus. Although as we have noted, tertiary
prevention is geared toward the already diagnosed situations of child
abuse and neglect, primary prevention has a broader goal: To reduce the
rate of new cases in a population over time by counterbalancing adverse
influences, and to reduce the risk of child abuse and neglect for the
entire community. Thus, primary prevention programs for child abuse
and neglect would include political/social action, examination of factors
that produce healthy family behaviors, as well as examination of factors
that lead to family dysfunction, exploration of problems of living, and,
the study of environmental factors that impact family life.
Paisley (1987) discussed legislative actions to prevent child abuse
and neglect enacted by states and, in particular, the legislation efforts in
North Carolina. She described the role of the school counselor and
noted that primary prevention programs should be geared toward re-
ducing the parent's unrealistic expectations of child and adolescent be-
haviors, caretaker responsibilities, long-term costs of inappropriate par-
ent skills, and family isolation. She recommended programs that en-
hance parent-child bonding, emotional ties, and improved communica-
tion skills. She further suggested programs that would help parents
increase coping skills, especially when under stress. Peer support, ac-
cess to social services and medical/health resources, and parental home-
management/child-management skills are also recommended.
Hodson and Skeen (1987) reviewed the research and theories of
236 MAXINE R. NEWMAN and JOHN R. LUTZKER

child sexual abuse and found that family life education could make a
significant contribution to the prevention and treatment of sexual abuse.
Thus, family life educators and other professionals who routinely work
with children and their families must be supplied with information on
the prevention and detection of childhood sexual abuse, including such
issues as (1) the theoretical explanations of childhood sexual abuse, (2)
the characteristics of abusers, (3) the characteristics of victims, (4) the
means of detecting abuse, and (5) the prevention and response interven-
tions.
In his presidential address for the International Society for Preven-
tion of Child Abuse and Neglect, Ferrier (1986) suggested that, even
though the need for tertiary prevention exists because "accidents always
occur," he urged that professionals agree that cases of abuse and neglect
should only be accidents and "rare ones at that (p. 280)." He noted the
difficulty in secondary prevention because though the identifying of
high-risk groups is attractive it may be, "fraught with the danger of self-
fulfilling prophecy (p. 281)." Ferrier (1986) stated that his "personal bias
is primary prevention." He recommended using education of the gener-
al public to create a change in current attitudes, values, and beliefs about
the use of violence as an acceptable method of child-rearing as well as
heightening societal awareness of the "devastating effect of rejection or
verbal abuse." He supported the position of the American Academy of
Pediatrics in its condemnation of the lack of censorship in television to
ban the most "viciously realistic forms of violence" presented for home
viewing. Finally, he suggested that the further study of those children
who survive abuse or neglect and overcome their circumstances com-
pared with those children who are unable to overcome the abuse or
neglect, would "boost our morale and lift our spirits."
Nelson (1984) analyzed four arenas of political action on behalf of
children: the U.S. Children's Bureau, the media, state legislators, and
Congress. She found that child abuse is a "consensual issue"; that is,
everyone agrees that child abuse is a problem. It also appears to be
relatively high on the agenda of those who make policy decisions. How-
ever, she noted that such decisions are often made from the heart rather
than on the basis of empirical information or data-based studies. Thus,
although made with the best of intentions, state response often unwit-
tingly offers the least effective response.
School sexual abuse prevention programs may produce some unin-
tended consequences and dilemmas (Trudell & Whatley, 1988). This pri-
mary prevention strategy often involves the use of classroom teachers
and other elementary school personnel as instructors or guides for stu-
dents. Although child sexual abuse is a complex issue that is likely to be
PREVENTION PROGRAMS 237

affected by the decrease of traditional societal controls over sexual be-

havior, varying sexual expectations, family isolation, an increase in di-
vorce rates, an increase in the number of families joined by remarriage,
differences in the way male and females are socialized, media and por-
nographic depictions of children as sexual objects, changing norms of
child obedience to adults, and low rates of identification, prosecution,
and conviction of perpetrators, school prevention programs frequently
focus on the prospective perpetrator's access to the child and neglect to
address such issues as the adult's attraction to a child, his or her lack of
internal controls, and the lack of external controls. Trudell and Whatley
(1988) argued that, because of the complexity of the problem and the
necessary oversimplification of a wide-based primary prevention pro-
gram, simply offering a prevention program in every classroom would
not lead to a significant decrease in child sexual abuse. Furthermore, the
emphasis on the child's behavior may encourage blaming the victim. In
such programs, parents have been informed that in 80% of the cases, the
victim could have stopped the assault by just saying no . Implied within
this message is that the child is powerful enough to stop the victimiza-
tion and therefore responsible for the consequences. Trudell and What-
ley (1988) also pointed out several dilemmas for classroom teachers in
this prevention approach: (1) the increased use of pre developed curricu-
lar materials may lead to deskillling teachers; (2) these materials are
often designed to be brief interventions because packages that involve a
minimum of extra teacher and student time are more marketable; and (3)
such programs are designed to avoid controversy, and since one of the
most controversial issues in the area of childhood sexual abuse is ex-
plicitness, these programs are often vague rather than clear. Also , many
teachers are reluctant to use such programs because of the mandatory
reporting laws, which, as some critics argue, obscure the ethical nature
of teacher decisions; assume that reporting will, necessarily, produce
positive outcomes; and, neglect the educational context of the classroom
in order to encourage the openness and trust necessary for child-to-
teacher disclosure. Trudell and Whatley (1988) concluded that broad-
based school programs to prevent childhood sexual abuse may not solve
the problem. Rather, they suggested the primary prevention strategy of
social action to make "fundamental changes in a society that allows and
even encourages child sexual abuse" (p. 111).
Rosenberg and Reppucci (1985) examined three categories of pri-
mary prevention in child abuse: (1) programs that helped families be-
come more competent; (2) programs geared toward the prevention of
first-time abuse; and (3) programs that focused on high-risk popula-
tions. They pointed out three methodological problems that appeared to
238 MAXINE R. NEWMAN and JOHN R. LUTZKER

cut across all such programs: (1) there were not enough appropriate
comparison groups; (2) outcome measures were poorly chosen; and (3)
there was a "failure to measure proximal programmatic objectives and
distal prevention goals." They concluded that although the programs
studied offered some exciting possibilities in the area of primary preven-
tion of child abuse and neglect, many have not proved that they could
actually accomplish this goal.

PREVENTION PROGRAMS
Tertiary prevention programs are directed toward an already diag-
nosed population and have, as their goals, the rehabilitation of the per-
petrator, keeping the family intact, and the prevention of further abuse .
Wolfe, Kaufman, Aragona, and Sandler (1981) developed an inter-
vention model for child abusers that consists of a series of step-by-step
procedures to teach abusive parents the skills required in managing
their children's activities, problem-solving and conflict resolution, anger
and impulse control, and building and maintaining positive social rela-
tionships outside the home. Wolfe and his colleagues (1981) presented a
treatment and assessment program and a series of evaluative studies.
The Kansas Child Abuse Prevention Trust Fund Program, which
was founded in 1980 (Poertner, 1987), offers programs to abusive parents
that include parenting education, public awareness (a primary preven-
tion program), home visits, and special programs for latchkey children
and children from homes in which spousal abuse has been reported.
Secondary prevention programs target an at-risk population and
have as their goals the prevention of first-time abuse or the prevention
of further abuse.
Wolfe, Edwards, Manion, and Koverola (1988) evaluated an early
intervention program for parents who were at-risk of abusing or neglect-
ing their children. The subjects were women, aged 16-25, who were
living on welfare, with young children aged 9-60 months. Even though
many of the women had male friends who visited them for different
lengths of time, only three were married or involved in permanent rela-
tionships. These women and their children were assigned to one of two
intervention groups: an information-only program offered by the child
protection agency, or a behavioral parent-training program in addition
to the other program. Although both groups showed improvement in
their child-rearing environments and child behaviors, only the women
who received the behavioral parent-training package showed significant
improvements in parenting risk and child-behaviors problems at post-
PREVENTION PROGRAMS 239

test and at a 3-month follow-up. Further, at a I-year follow-up case-

workers rated these women as having lower risks of maltreatment and
higher abilities to manage their children.
The Children's Hospital Program on Family Violence (White et al.,
1987) in Boston was developed to provide an interdisciplinary team-
treatment approach to the problem of child abuse and neglect. The
program adopted the term "Trauma X," rather than use what was then
referred to as the "battered child syndrome," as an expression of general
emphasis on violence and neglect directed to problems of family sys-
tems rather than attributes attributable to pathological parents. The
clinical work of the Trauma X team led to the creation of a family vio-
lence research center-The Family Development Center. The Children's
Hospital also provides a clinical training program and inservice teach-
ing. Funded by federal grants, this model hospital-based training pro-
gram on family violence provides an ongoing Family Violence Seminar
and a fellowship program for pediatricians and behavioral and social
scientists.
Project 12-Ways, supported by Title XX federal funds operates out
of the Southern Illinois University at Carbondale and evolved out of the
logic of ecologically oriented research. An ecobehavioral approach to the
treatment of child abuse and neglect (Lutzker, 1984; Lutzker, Frame, &
Rice, 1982; Lutzker & Rice, 1984; Lutzker, Wesch, & Rice, 1984; Lutzker
& Newman, 1986), Project 12-Ways focuses on the remediation of sever-
al of the known risk factors contributing to the problem, including pre-
mature birth and birth complications, mental retardation, physical
handicaps and developmental difficulties, behavioral problems, such as
whining, crying, hyperactivity, and other negative behaviors, unem-
ployment, large family, unstable marriage, stress-producing events, sin-
gle-parenting, and poverty. Project 12-Ways provides such services as
parent training, using contingency management and activity training
programs (Dachman, Halasz, Bickett, & Lutzker, 1984;Lutzker, Megson,
Webb, & Dachman, 1985); stress reduction, using progressive muscle
relaxation, biofeedback, or behavioral relaxation training (Campbell,
O'Brien, Bickett, Newman, & Lutzker, in press); self-control training,
such as weight or anger control; social support and basic skills for chil-
dren, such as personal hygiene (Lutzker, Campbell, & Watson-Perczel,
1984); activity training for parents and children, such as health mainte-
nance and nutrition, home cleanliness, and nutritious meal planning
and serving (Rosenfield-Schlichter, Sarber, Bueno, Greene, & Lutzker,
1983; Sarber, Halasz, Messmer, Bickett, & Lutzker, 1983; home safety
(Barone, Greene, & Lutzker, 1986; Tertinger, Greene, & Lutzker, 1984);
and job-finding, problem-solving, money management, and pre-
240 MAXINE R. NEWMAN and JOHN R. LUTZKER

natal/postnatal care for single parents (Lutzker, Lutzker, Braunling-

McMorrow, & Eddelman, 1987).
Project Ecosystems which is based in southern California, is a rep-
lication of Project 12-Ways and is directed at families with de velopmen-
tally disabled children who are at-risk for abuse or neglect. Lutzker,
Campbell, Newman, and Harrold, (1990) described the multiple in-
home services provided by Project Ecosystems in the first 6 months of its
existence. Goals include keeping the child at home wherever possible,
preventing incidents of child abuse or neglect, improving adaptive func-
tioning within families , and providing training and research oppor-
tunities to students in applied behavior analysis and human services.
Project Ecosystems maintains a caseload of 30-50 families within any
given fiscal year. Its objectives are that, in any fiscal year, no more than
three children being served will be placed in a more restrictive setting. A
further objective is that no more than three parents will be indicted for
abuse or neglect. Project Ecosystems provides the same kind of multi-
faceted interventions described in Project 12-Ways, including basic skill
training, problem-solving; job-finding; money management; nutrition,
behavioral pediatrics; hygiene; and, home-safety. Assessment is accom-
plished with parent self-report measures and child-observation mea-
sures. Clinical evaluation, single-case experiments, and program evalua-
tion are the three methods used to evaluate services.
Primary prevention programs are broad-based, with the community
at-large as their target. These programs are designed to prevent first-
ever abuse in the general population. The programs often use the public
media to communicate their message although not alwa ys .
A special comic book issue of Spiderman , which dealt with sexual
abuse, was presented to 73 2nd, 3rd, and 4th-grade children (Garbarino,
1987). This low-cost, primary prevention approach was effective in
teaching children and their parents how to report sexual abuse. Gar-
barino recommended enhancing the power of the comic book by pre-
senting it in the framework of an ongoing and comprehensive program
geared not only to children but to parents as well .
The Red Flag/Green Flag People coloring book and a film, Better Safe
than Sorry II, was given to 289 3rd and 4th-graders, 276 parents, and 13
teachers from two schools (Kolko, Moses, Litz, & Hughs, 1987), along
with a discussion of imaginary and actual scenarios involving inap-
propriate touching. Kolko and his colleagues (1987) evaluated this pro-
gram to promote awareness and prevent child sexual victimization by
comparing it to a group of 41 3rd and 4th-graders, 41 parents, and 2
teachers from a control school. Children in the experimental group
PREVENTION PROGRAMS 241

learned more about the differences between appropriate and inappropri-

ate touching, were more likely to report abuse, and were more able to
use learned skills than those in the control group. The program was also
effective in increasing parents' knowledge about the program and in
enhancing communication about abuse at home.
Wurtelle, Marrs, and Miller-Perrin (1987) compared a sexual abuse
prevention program that included participant modeling with a similar
program using symbolic modeling. Wurtele and her colleagues (1987)
found that the participant modeling program that taught self-protective
skills through modeling and active rehearsal provided superior efficacy
than the symbolic modeling program that taught the same skills, but
that involved observation of the experimenter rather than hands-on
practice.
Similarly, Stillwell, Lutzker, and Greene (1990) evaluated the Sexual
Abuse Prevention Program for Preschoolers (SAPPP), which added such
components as behavioral training, rehearsal, and evaluation to the for-
mat of a popular prevention program. Children participated in six
SAPPP lessons that included the topic introduction (for example, differ-
ent types of touch, how to say no and how to tell others), stories related
to the topic, behavioral rehearsal of a target behavior, and finally class
discussion. Stillwell and her colleagues (1990) found that all of the chil-
dren improved their scores on correct verbal responses from pretest to
posttest. However, none of the children showed improvement on be-
havioral demonstration. Stillwell et al. (1990) concluded that an impor-
tant goal of future prevention programs should be to train children to
demonstrate optimum prevention behaviors.
Conte, Rosen, Saperstein, and Shermack (1985) evaluated a 3-hour
program that taught common sexual abuse prevention concepts, such as
the difference between OK and not-OK touching. Using a repeated mea-
sures multivariate analysis of variance, they found that children in a
prevention training group significantly increased their knowledge of
prevention concept, as opposed to children in a control group who did
not. Furthermore, older children learned more than younger children,
and both groups had difficulty with concepts presented in the abstract
rather than in the concrete.
As noted, Rosenberg and Reppucci (1985) reviewed primary pre-
vention programs in the area of child abuse and neglect. They found
programs that addressed three major areas: competency enhancement,
preventing the onset of abusive behavior, and targetting high-risk
groups. Competency enhancement programs included the use of the-
ater and television to increase parenting skills (Inter-Act: Street Theater
242 MAXINE R. NEWMAN and JOHN R. LUTZKER

for Parents); a large-scale comprehensive program to decrease the preva-

lence of child abuse and improve children's school performance (Project
C.A.N. Prevent); and parent education groups to enhance parents' abili-
ties to cope with child-rearing problems resulting from cultural dif-
ferences (Pan Asian Parent Education Project). Programs to prevent the
onset of abusive behavior included media campaigns, information,
crisis, and referral networks, and projects designed to empower social
groups to provide support to families . Some projects that were reviewed
included a comprehensive community education and referral campaign
designed to strengthen formal and informal helping networks (Project
Network, Atlanta University); a multifaceted public awareness project
for rural counties (Primary Prevention Partnership); and telephone
hotlines (Parents Anonymous, Michigan's Warm Line, Connecticut's
Care-Line). In their review of primary prevention projects, Rosenberg
and Reppucci (1985) included programs that targetted high-risk groups.
We would consider these interventions secondary prevention; however,
in this instance we will include them here. Those reviewed are a service
for prospective mothers expecting a first child with emphasis on young,
single, and women of lower socioeconomic status (Prenatal/Early Infan-
cy Project), and another multileveled program, discussed earlier, with
an emphasis on home safety (Project 12-Ways).

RECOMMENDATIONS
The traditional mental health concept of prevention, as we have
shown, includes three seemingly distinct levels of interventions. How-
ever, Rosenberg and Reppucci (1985) argued that tertiary level programs
that have as their target population individuals whose disorders are
already identified are not, in the strictest sense, prevention programs at
all. On the other hand, we suggest that prevention of further abuse is a
valid perspective.
In searching the literature, we found very few large-scale communi-
ty programs that had as their goal the rehabilitation of all members of the
diagnosed population. We included Project 12-Ways (Lutzker & New-
man, 1986) in our section on secondary prevention because it addresses
many of the risk factors in child abuse and neglect; however, it is an
example of how a tertiary prevention program might succeed. For exam-
ple, Project 12-Ways is community-based; it provides in-home, in-situ
services with the goal of keeping the family intact, where feasible, in-
creasing the parenting and living skills of caregivers, decreasing dys-
functional child behaviors, and decreasing abuse and/or neglect.
PREVENTION PROGRAMS 243

Programs must also be cost effective. Lutzker and Newman (1986)

pointed out that Project 12-Ways provided services at approximately one
half the cost that would have been incurred if similar services had been
provided under the auspices of a mental health facility. Program evalua-
tion and empirical examination must support claims of success by any
prevention program. Project 12-Ways has an ongoing research, assess-
ment, and evaluation arm that provides consistent feedback about inter-
ventions (Lutzker & Newman, 1986).
In the area of secondary prevention, we note the many risk-factors
that impact on families and that may lead to abuse or neglect. There are
parent characteristics, child characteristics, and environmental charac-
teristics, which, when triggered by an event, may lead to abusive or
neglectful environments (White et al., 1987). Project 12-Ways (Lutzker &
Newman, 1986) and Project Ecosystems (Lutzker, Campbell, Newman
& Harrold, 1990) provide multilevel services directed at these three as-
pects of the problem. We recommend "ecologically orientated" interven-
tions and research; that is, we suggest that future secondary prevention
programs examine and address the many complex factors that seem
necessary and sufficient to produce child abuse and neglect.
At the same time, primary prevention is the simplest and the most
complex issue of our time . There appears to be money and enthusiasm
for large-scale, media-based campaigns to "just say no." However, in
the attempt to reach everyone, the message becomes vague rather than
concrete. Often, with the best will in the world, programs are watered
down to be palatable to the largest common denominator to the point of
being ineffective and possibly harmful. For example, if children are
taught that they have only to say no, then we are blaming the victims for
their victimization (Trudell & Whatley, 1988). Social action is another
aspect of primary prevention. For example, Zuravin (1986) found a con-
nection between residential density and urban child maltreatment. In
this instance, it is incumbent upon health care professionals to publish
such findings, not only for each other in professional journals, but to
make such research known to the general public. Hermalin, Melendez,
Kamarck, Kievans, Ballen, and Gordon (1979) noted the usefulness of
self-help groups as support networks for more formal health care deliv-
ery systems in enhancing primary prevention. Thus, it is the responsi-
bility of the health care professional to provide an environment suppor-
tive of the growth of self-help groups. Finally, our society has values and
a culture that supports such axioms as "spare the rod and spoil the
child"; we even admire such lone killers as Rambo, and we laugh at such
lines as Noel Coward's quip "women should be beaten regularly, like a
gong." In this atmosphere, which is hostile to nurturance, it is necessary
244 MAXINE R. NEWMAN and JOHN R. LUTZKER

for the health care professional to become an activist, to publically boy-

cott films, music, and books that promote violence to women and chil-
dren, and to protest policies that keep a major segment of our popula-
tion in despair.

SUMMARY
There are three levels of child abuse or neglect prevention: tertiary,
secondary, and primary. Tertiary prevention in the area of child abuse
and neglect refers to large-scale community programs, often institution-
based, whose goal is to prevent further abuse or neglect in already cited
families. Secondary prevention in the area of child abuse and neglect is
concerned with reaching those groups or individuals who appear to be
at risk for abuse and/or neglect, either because of parental charac-
teristics, child characteristics, or environmental characteristics. The goal
of such programs is to prevent further abuse or first-time abuse. Primary
prevention in the area of child abuse and neglect considers the entire
population as its target. Media campaigns, bibliotherapy, school-based
programs, films, and social and political action are components of these
interventions. Although much work has been done that has proven of
value in the area of prevention programs for child abuse and neglect,
there is still much more to be accomplished in the future.

REFERENCES
Adelson, L. (1969). Slaughter of the innocents. New EnglandJournal of Medicine, 264, 1345-
1349 .
American Humane Association. (1963). Child abuse: Preview of a nationwide survey.
Denver, CO: Author.
Azar, S. T., Barnes, K. T., & Twentyman, C. T. (1988). Developmental outcome in phys-
ically abused children: Consequences of parental abuse on the effects of a more gener-
al breakdown in caregiving behaviors. The Behavior Therapist, 11, 27-32.
Bakwin, H. (1956). Multiple skeletal lesions in young children due to trauma. American
Journal of Pediatrics, 49, 7-15.
Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice Hall.
Barmayer, G. H., Alderson, L. R., & Cox, W. R. (1951). Traumatic periostitis in young
children . American Journal of Pediatrics, 38, 184-190.
Barone, U. J., Greene, B. F., & Lutzker, J. R. (1986). Home safety with families being
treated for child abuse and neglect. Behavior Modification, 10, 93-114.
Bittner,S. , & Newberger, E. H. (1981). Pediatric understanding of child abu se and neglect.
Pediatrics in Review, 2, 197.
Boardman, H. E. (1961). A project to rescue children from inflicted injuries. Social Work
(Am . Ed.) 1, 43-51.
PREVENTION PROGRAMS 245

Browne, D. H. (1986). The role of stress in the commission of subsequent acts of child
abuse and neglect. Journal of Family Violence, 1, 289-297.
Caffey, J. (1946). Multiple fractures in the long bones of infants suffering from chronic
subdural hematoma . American Journal of Roentgendogy Radium Therapy and Nuclear Medi-
cine, 56, 163-174.
Campbell, R. V., Twardosz, S., Lutzker, J. R., & Cuvo, A. J. (1983, May). Comparison study
of affectionate behavior across status of abuse neglect and non-abuse neglect. Paper presented
at the annual convention of the Association for Behavior Analysis, Milwaukee,
Wiscon sin .
Campbell, R. V., O'Brien, S., Bickett, A. D., Newman, M. R., & Lutzker, J. R. (in press).
Migraine headaches and marriage and family happiness: An ecobehavioral approach.
In S. Diamond & M. Maliszewski (Eds.), Sexuality and headache: A medical-behavioral
overview. New York: International Universities Press.
Caplan, G. (1964). Principles of preventive psychiatry. New York: Basic Books.
Carstens, C. C. (1911-1912). The prevention of cruelty to children. Proceedings of the Acade-
my of Political Science, 2, 616-617.
Car stens, C. C. (1921). The development of social work for child protection . Annals of the
American Academy of Political and Social Science, 98, 137-141.
Child Protection Report. Vol. 2, January 14, 1976.
Cohn, A. A. (1986). Preventing adults from becoming sexual molesters. Child Abuse and
Neglect, 10, 559-562 .
Conte, J. R., Rosen, c., Saperstein, L., & Shermack, R. (1985). An evaluation of a program
to prevent the sexual victimization of young children. ChildAbuse and Neglect, 9, 319-
328.
Cowen, E. L., Dorr, D., Clarfield, S., Kreling, B., McWilliams, S. A. , Pokracki, D., Pratt,
M., Terrell, D. , & Wilson , A. (1973). The AML: A quick screening device for early
identification of school maladaption. American Journal of Community Psychology, 1, 12-
35.
Cumming, J., & Cumming, E. (1957). Closed ranks . Cambridge: Harvard University Press.
Dachman, R. S., Halasz, M. M., Bickett, A. D., & Lutzker, J. R. (1984). A home-based
ecobehavioral parent-training and generalization package with a neglectful mother.
Education and Treatment of Children, 7, 183-202.
Elmer, E. (1960). Abused young children seen in hospitals. Social Work,S, 98-102 .
Englert, Y., Marneffe, C.; Soumenkoff, G., & Hubinont, P. O. (1985). Institutional abuse:
Tubal sterilization in a population at risk of ill-treating their children. ChildAbuse and
Neglect, 9, 31-35.
Fantuzzo, J., Azar, S. T., & Twentyman, C. T. (1985, November). Child abuse and psycho-
therapy research:Mergingsocial concerns andempirical investigation. Paper presented at the
annual convention of the Association for Advancement of Behavior Therapy, Phila-
delphia, Pennsylvania.
Ferrier, P. E. (1986). Pres idential address: The International Society for Prevention of Child
Abuse and Neglect. Child Abuse and Neglect , 10, 279-281.
Fisher, S. H . (1958). Skeletal manifestations of parent ind uced trauma in infants and
children. Southern Medical Journal , 51, 956-960.
Gambrill, E. D. (1983). Behavioral intervention with child abuse and neglect. In M. Hersen,
R. M. Eisler, & P. M. Miller (Eds.), Progress in behavior modification: Vol. 15 (pp . 1-56).
Orlando, FL: Academic Press
Garbarino, J. (1987). Children's response to a sexual abuse prevention program: A study of
the Spiderman comic. Child Abuse and Neglect, 11, 143-148.
Ha genhoff, C.; Lowe , A., Hovell, M. F., & Rugg , D. (1987). Prevention of the teenage
246 MAXINE R. NEWMAN and JOHN R. LUTZKER

pregnancy epidemic: A social learning theory approach. Education and Treatment of

Children, 10, 67-83.
Helfer, R. B. (1982). A review of literature on the prevention of child abuse and neglect.
Child Abuse and Neglect, 6, 251-261.
Hermalin, J., Melendez, L. , Kamarck, T., Kievans, F., Ballen, E., & Gordon, M. (1979).
Enhancing primary prevention: The marriage of self-help groups and formal health
care delivery systems. Journal of Clinical Child Psychology, 8, 125-129 .
Hodson, D., & Skeen, P. (1987). Child sexual abuse: A review of research and theory with
implications for family life educations. Family Relations: Journal of Applied Family and
Child Studies, 36, 215-221.
Joint Commission on Mental Health and Illness. (1961). ActionforMentalHealth . New York:
Wiley.
Joint Commission on Mental Health of Children. (1970). Crisis in child mental health: Chal-
lenge for the 1970's. New York: Harper & Row.
Kempe, C. H ., & Silver, H. K. (1959). Problem of parental criminal neglect and severe
physical abuse of children. American Journal of Diseases of Children, 98, 529.
Kempe, C. H., Silverman, F. N ., Steele , B. F., Droegemueller, v«, & Silver, H. K. (1962).
The battered-child syndrome. Journal of the American Medical Association, 181, 105-112.
Kolko, D. J., Moses, J. T., Litz, J., & Hughs, J. (1987). Promoting awareness and prevention
of child sexual victimization using the Red Flag/Green Flag Program: An evaluation
with follow-up. Journal of Family Violence, 2, 11-35.
Lucht, C. (1975). Providing a legislative base for reporting child abuse. Fourth National
Symposium on Child Abuse. Denver, CO : American Humane Association.
Lutzker, J. R. (1984). Project 12-Ways: Treating child abuse and neglect from an eco-
behavioral perspective. In R. F. Dangel & R. A. Poloten (Eds.), Parent training: Formula-
tions of research and practice (pp. 260-291) . New York: Guilford Press.
Lutzk er, J. R., & Newman , M . R. (1986). Child abuse and neglect: Community problem,
community solutions. Education and Treatment of Children, 9, 344-354.
Lutzker, J. R., & Rice, J. M. (1984). Project 12-Ways: Measuring outcome of a large in-home
service for treatment and prevention of child abuse and neglect. Child Abuse and
Neglect, 8, 519-524.
Lutzker, J. R., Campbell, R. V., & Watson-Perczel, M. (1984). Utility of the case study
method in treatment of several problems of a neglected family. EducationandTreatment
of Children, 7, pp. 315-353.
Lutzker, J. R., Frame , R. E., & Rice, J. M. (1982). Project 12-Ways: An ecobehavioral
approach to treatment and prevention of child abuse and neglect. Education and Treat-
ment of Children, 5, 141-155.
Lutzker, J. R., Wesch, D., & Rice, J. M. (1984). A review of Project 12-Ways: An eco-
behavioral approach to treatment and prevention of child abuse and neglect. Advances
in Behaviour Research and Therapy, 6, 63-73.
Lutzker, J. R., Campbell, R., Newman, M. R., & Harrold, M. (1989). Ecobehavioral inter-
vention for abusive, negl ectful, and high-risk families : Project 12-Waj's and Project
Ecosystems. In G. H. S. Singer & L. K. Irvin (Eds.), Family support services: Emerging
pertnerships between families with severely handicapped individuals and communities (pp .
313-326). Baltimore, MD: Paul H . Brookes.
Lutzker, J. R., Megson, D. A., Webb, M. E., & Dachman, R. S. (1985). Validating and
training adult-child interaction skills to professionals and to parents indicated for child
abuse and neglect. Journal of Childand Adolescent Psychotherapy, 2, 91-104.
Lutzker, S. Z., Lutzker, J. R. , Braunling-McMorrow, D., & Eddelman, J. (1987). Promoting
PREVENTION PROGRAMS 247

to increase mother-baby stimulation with single mothers. Journal of ChildandAdolescent

Psychotherapy, 4, 3-12.
Lynch , M. A. (1985). Child abuse before Kempe : An historical review. Child Abuse and
Neglect, 9, 7-15.
Meredith, W. H. , Abbott, D. A., & Adams, S. L. (1986). Family violence: Its relation to
marital and parental satisfaction and family strengths. Journal of Family Violence, 1,
299-309.
Nelson, B. [. (1985). Making an issue of child abuse: Political agenda setting for social problems.
Chicago: University of Chicago Press.
Paisley, P. O. (1987). Prevention of child abuse and neglect: Legislative response. School
Counselor, 34, 226-228.
Pardek, J. T., & Pardek, ]. A. (1984). Treating abused children through bibliotherapy. Early
Child Development and Care, 16, 195-205.
Paulsen, M. (1968). The law and abused children. In R Helfer & C. H. Dempe (Eds .), The
battered child. Chicago: University of Chicago Pres s.
Poertner, J. (1987). The Kansas family and child trust fund : Five year report. Child Welfare,
66,3-12.
Rappaport, J. (1977). Community psychology: Values, research, and action. New York: Holt ,
Rinehart, & Winston.
Rosenberg, M. S., & Reppucci, N . D. (1985). Primary prevention of child abuse. Journal of
Consulting and Clinical Psychology, 55, 576-585.
Rosenfield-Schlichter, M. D. , Sarber, R E., Bueno, G., Greene, B. F., & Lutzker, J. R.
(1983). Maintaining accountability for ecobehavioral treatment of one aspect of child
neglect: Personal cleanliness. Education and Treatment of Children, 6, 152-164.
Sarber, R. E., Halasz, M. , Messmer, M. c., Bickett, A. D., & Lutzker, J. R (1983). Teaching
menu planning and grocer shopping skills to a mentally retarded mother. Mental
Retardation, 21, 101-106.
Schultz, W. J. (1968). The humanemovement in the United States, 1910-1922. New York: AMS
Pre ss.
Silverman, F. N. (1952). The roentgen manifestation of unrecognized skeletal trauma in
infants. American Journal of Roentgenology, Radium Therapy, and Nuclear Medicine, 69,
413-427.
Snedecor, S. T., Knapp, R E., & Wilson, H . B. (1935). Traumatic ossifying periostitis of the
newborn. Surgery, Gynecology, and Obstetrics, 61, 385-387.
Stillwell , S. L., Lutzker, J. R., & Greene, B. F. (1988). Evaluation of a sexual abuse preven-
tion program for preschoolers. Journal of Family Violence, 3, 269-280.
Straus, M. A., Gelles , R. S., & Steinmetz, S. (1980). Behind closed doors: Violence in the
American family. Garden City, NY: Doubleday.
Tertinger, D. A., Greene, B. F., & Lutzker, J. R (1984). Horne safety: Development and
validation of one component of an ecobehavioral treatment program for abused and
neglected children. Journal of Applied Behavior Analysis, 17, 159-177.
Trudell, B., & Whatley, M. H . (1988). School sexual abuse prevention: Assessing conse-
quences and dilemas. Child Abuse and Neglect, 12, 103-113.
United States Department of Health, Education, and Welfare. (1963). The abused child:
Principles and suggested language for legislation on reporting of the physically abused child.
Washington, DC: Ll.S, Government Printing Office.
Vasta, R (1982). Physical child abuse: A dual-component analysis. Developmental Review, 2,
125-149.
White, K. M., Snyder, J., Bourne, R, & Newberger, E. H . (1987). Treatingfamilyviolence in a
248 MAXINE R. NEWMAN and JOHN R. LUTZKER

Pediatric Hospital: A program of training, research, and seroice. U.S. Department of Health
and Human Seroices. Rockville, MD: Public Health Service, Alcohol, Drug Abu se, and
Mental Health Adm inistration, National Institute of Mental Health.
Willems, E. P. (1974). Behavioral technology and behavioral ecology. Journal of Applied
Behavior Analysis, 7, 151-165.
Wolfe, D. A. , Edwards, B., Manion, C,; & Koverola, C. (1988). Early intervention for
parents at risk of child abuse and neglect: A preliminary investigation. Journal of
Consulting and Clinical Psychology, 56, 40-47.
Wolfe, D., Kaufman, K., Aragona, J., & Sandler, J. (1981). The child management programfor
abusive parents: Procedures for developing a child abuse interoention program. Florida : Anna
Publi shing.
Wurtelle, S. K., Marrs, S. R., & Miller-Perrin , C. L. (1987). Practice makes per fect? The role
of participant modeling in sexual abu se prevention programs. Journal of Consulting and
Clinical Psychology, 599-602.
Zuravin, S. J. (1986). Residential density and urban child malt reatment: An aggregate
analysis. Journal of Family Violence, 1, 307-314 .
 CHAPTER 10

TREATING THE ABUSED CHILD

ANTHONY P. MANNARINO AND JUDITH A. COHEN

INTRODUCTION
In most clinical settings, treatment for child abuse rarel y focuses solely
on th e victim, which is inherently sensible since the perpetrator of abuse
must also be treated if the risk for re-abuse is to be sign ificantl y reduced.
Particularly in the area of physical abuse, treatment has largely focused
on abusive parents, whereas little has been written about the treatment
of child victims. (For a discussion of treatment of the child abuser, see
Chapter 11 in this volume.) More recently, however, our understanding
of how to treat abused children ha s increased as data have begun to
accumulate regarding the impact of abuse on victims.
This chapter will address a number of issues related to the treat-
ment of abused children. First, we will present a discussion of defini-
tion al issues, and methodological concerns that have impeded our
efforts to develop appropriate and effective treatment modalities fol-
lowed by discussions of treatment methods and clinical themes relevant
to physically and sexually abused children. We will not address the
concept of emotional abuse in this chapter because we believe that this
type of abuse is so unspecified and ill-defined that a section on treat-
ment devoted to it is simply not possible.

ANTHONY P. MANNARINO AND JUDITH A. COHEN· Department of Psychiatry, Western

Psych iat ric Institute and Clinic, Un iver sity of Pitt sburgh Sch ool of Medicine, Pitt sburgh,
Pennsylvania 15213.

249
250 ANTHONY P. MANNARINO and JUDITH A. COHEN

Furthermore, we contend that for a number of legitimate clinical

reasons, a discussion of the treatments applicable to sexually and phys-
ically abused children should be separated. First, whereas physical
abuse is largely perpetrated by parents, sexual abuse can be intra- or
extrafamilial. In addition, although physical abuse is clearly an act of
aggression toward a child , it is uncommon for sexual abuse to be accom-
panied by violence. Finally, little empirical evidence exists pertaining to
familial factors that contribute to physical or sexual abuse. Current data
do not suggest that family dynamics and related factors are necessarily
similar for these two types of abuse (Walker, Bonner, & Kaufmann,
1988). For these reasons, we will provide separate treatment of victims of
physical and sexual abuse. If the discussion of one type of abuse has
relevance to the other, there will be an attempt to highlight the areas of
overlap.

DEFINITIONAL AND METHODOLOGICAL ISSUES

DEFINITIONAL PROBLEMS

In order to develop and provide effective treatment, the problem to

be addressed needs to be clearly specified. This has not been the case in
the area of child abuse. Various types of abuse (e.g., physical, sexual,
emotional) are frequently lumped together and handled as if they are
one entity (Blythe, 1983). In some treatment studies, the nature of the
abuse is not even documented (Isaacs, 1982).
Even when attempts are made to separate physical from sexual
abuse, definitional problems persist. For example, what exactly con-
stitutes physical abuse? Must there be physical evidence of an injury?
Or, in the area of sexual abuse, how are these cases investigated and
what criteria are used in determining credibility? We readily acknowl-
edge that these are complex issues that are not easily resolved. How-
ever, more serious efforts should be made by researchers studying the
kinds of treatments potentially useful to abused children to specify clear-
ly how abuse is defined and what criteria are utilized in making this
determination.
In addition to definitional issues, characteristics of the treated popu-
lations have been infrequently specified in studies of child abuse (Smith,
Rachman, & Yule, 1984). It is very surprising that sometimes not even
the age range or sex of the abused group is mentioned. There are other
sample characteristics as well, such as race, socioeconomic level, and
intellectual status that rarely are addressed. All of these factors can have
TREATING THE ABUSED CHILD 251

a profound impact on the course of treatment and its outcome. Failure to

specify these variables has been a major limiting factor in treatment
studies involving abused children.

METHODOLOGICAL PROBLEMS

The study of potentially effective treatments for abused children

has also been beset by other types of severe methodological difficulties.
For example, control groups or alternative treatments have rarely been
employed (Smith et al., 1984). In addition, the specific nature of the
treatment program often is unclear (Blythe, 1983). Details of the inter-
vention are so seldom given that it is sometimes difficult to discern if
treatment is being provided to the victim, the alleged perpetrator, or to
both.
The kind of outcome data collected in studies of child abuse are
typically subjective and unsystematic (Isaacs, 1982). Subjective ratings
by the treating clinician are the norm. It is uncommon for empirically
validated questionnaires, rating scales, or semistructured interviews to
be used . Moreover, the lack of adequate follow-up has been another
methodological shortcoming.
These definitional problems and methodological issues have been
major stumbling blocks to progress in the general area of child abuse
and in the specific area of treatment for this population. Fortunately,
these issues are beginning to be addressed adequately by a number of
researchers. Particularly in the area of sexual abuse, more recent studies
(Cohen & Mannarino, 1988; Conte & Schuerman, 1987; Friedrich, Ur-
quiza, & Beilke, 1986) have been methodologically more sophisticated
and precise. It is hoped that this trend will continue in order to build a
foundation of solid data and outcome studies in this field.

TREATING THE PHYSICALLY ABUSED CHILD

In the area of child physical abuse, the focus of treatment has
largely been on the abusive parent. In a thorough review of the existing
literature, the authors found that in the vast majority of studies, there
was either no treatment provided for children or it was clearly secondary
to that being given to parents (Gabinet, 1983a,b; Green, Power, Stone-
brook, & Gaines, 1981; Lutzker & Rice, 1984; Pelcovitz, Kaplan, Sarnit,
Krieger, & Cornelius, 1984; Shelton, 1982; Trowell & Castle, 1981).
It is difficult to assess why child victims of physical abuse have so
frequently been ignored in the development of interventions. Typicall y,
252 ANTHONY P. MANNARINO and JUDITH A. COHEN

elimination of further abuse has been the major treatment goal. Accord-
ingly, there has been a natural tendency to gear treatment toward par-
ents who perpetrate abuse against their children-an approach that has
been somewhat short-sighted, though, because it fails to recognize that
children may continue to suffer from emotional problems caused by
earlier abuse.
Other reasons may exist why treatment has not focused on child
victims. As Williams (1980) pointed out, child physical abuse has com-
monly been perceived to be in the professional domain of the physician
and not the mental health practitioner. Accordingly, there has been a
major concern about physical injuries and not necessarily the emotional
trauma caused by the abuse. Moreover, from a societal-legal perspec-
tive, some pressure has been focussed on abusive parents who are per-
ceived as needing treatment or punishment.
Recently, an increased emphasis was seen in the emotional and
behavioral difficulties that children experience as a result of having been
physically abused. Several studies have pointed to anxiety, aggression,
and social skills deficits, as being significantly correlated with physical
abuse in children (Mask, Johnson, & Kovitz, 1983; Reid, Taplin, &
Loeber, 1981; Wolfe & Mosk, 1983). With this added focus on the prob-
lems experienced by victims, a more serious interest has developed in
the kinds of interventions that would be most appropriate and effective
for this population.
Although assessment is not the focus of this chapter, it is important
to underscore that any potentially effective treatment for physically
abused children must be based on careful assessment. Walker et aI.
(1988) have stressed the significance of a thorough evaluation in order to
develop a method of intervention that is tailored to the individual needs
of each child. This issue cannot be overemphasized. Appropriate clinical
decisions should be made only after each case has been comprehen-
sively evaluated. More informed judgments can then be formed regard-
ing what kinds of treatment would potentially be most effective for the
child, parent, or family.

REVIEW OF TREATMENT STUDIES OF CHILD VICTIMS

Few published studies are available that have specifically imple-

mented treatment programs for child physical abuse victims . The stud-
ies that do exist are not linked in any empirical or theoretical way and
will be reviewed briefly to provide the reader with some sense of what
has transpired in the field .
One of the earliest attempts to investigate the impact of treatment
TREATING THE ABUSED CHILD 253

with abuse victims was part of a national demonstration project (Cohn,

1979). In this study, treatment was provided to 70 abused children, aged
2-7 years, in three different centers. These children manifested a wide
array of behavioral problems. Based on the subjective reports of treating
clinicians, it was determined that victims made gains in receiving affec-
tion, interacting with adults, and in the area of self-image . Unfortunate-
ly, the nature of the treatment provided to these children was unclear. In
addition, the use of subjective, nonblind clinician ratings to assess treat-
ment progress was a serious methodological shortcoming. Despite these
criticisms, Cohn's study was the first attempt to examine whether treat-
ment of victims and not only of abusive parents may be potentially
efficacious .
Other treatment programs designed for child physical abuse victims
have been extremely diverse. Myers, Brandner, and Templin (1985) as-
sessed developmental gains in 53 preschoolers who participated in a
therapeutic preschool and whose parents were actively involved in a
parenting group. Results demonstrated that most children made modest
developmental progress. As with the Cohn (1979) investigation, the
nature of the intervention provided to the children in this study was
unspecified. Also there was no control group used or alternate treat-
ment employed.
A methodologically more sophisticated project was designed by
Heide and Richardson (1987) in which one half of a group of neglected
but not necessarily abused children (mean age = 3) was placed in a day
treatment program while the other half comprised waiting list controls.
Children remained in day treatment for an average of nearly 8 months.
The authors reported that treated children made significantly greater
gains than did controls in five areas of functioning: perceptual-motor,
cognition, gross motor, social/emotional, and language. The design and
results of this study are very encouraging. However, it is again quite
frustrating that the nature of the treatment provided was only minimally
reported.
Other types of treatment for physically abused children have been
described in the literature but not empirically evaluated. Huebner (1984)
discussed the potential usefulness of group counseling for middle
school abused and neglected children. The group intervention consisted
of techniques to improve problem-solving skills and self-esteem. In a
contrasting model, Frazier and Levine (1983) suggested "reattachment
therapy" for young abuse victims. Critical elements in this approach are
helping children to become attached to a therapist and then assisting
them to learn to elicit attachment behavior from their caretaker. An
enlightened concept advanced by these authors is that certain child
254 ANTHONY P. MANNARINO and JUDITH A. COHEN

behaviors can provoke a parent into more abusive behaviors. According-

ly, helping the child to change these behaviors can break down the
typical abuse cycle. Although this model has yet to be evaluated, it offers
an innovative, theoretically based approach to the treatment of abuse
victims. More details will be discussed in the next section of this chapter
regarding the notion of "provocative child behaviors" and how they
might serve as the focus of treatment of abused children.
This brief review has demonstrated that there are few published
studies of potentially effective treatments for physically abused chil-
dren. Moreover, most of the studies presented have not been guided by
a particular theoretical or conceptual framework and have typically been
characterized by serious methodological shortcomings.

NEW DIRECTIONS IN THE CLINICAL TREATMENT

OF ABUSE VICTIMS

Treatment for Abuse-Related Symptoms or Problems

One way to conceptualize physical abuse is as a traumatic event in
the life of a child that may have a wide range of outcomes, many of
which are deleterious. Unfortunately, our current knowledge of the im-
pact of physical abuse on children is quite limited. Recent studies (Mask
et al., 1983; Reid et al., 1981; Wolfe & Mosk , 1983) have suggested that
abused children experience significantly greater levels of anxiety, ag-
gression, social skills deficits , and other behavioral problems when con-
trasted with nonabused children. It is clearly premature to assume,
however, that all abused children will manifest similar difficulties. There
is little doubt that the effects of physical abuse will be a function of the
child's preabuse characteristics, including developmental status, rela-
tionship with the abusive parent, and other family factors, and that
there will be a range of outcomes. All abused children must have a
comprehensive assessment to determine what symptoms or problems
they presentin response to the abuse and how these problems can be
treated most effectively.
There are a number of clinical interventions that have been demon-
strated to be effective with children with specific emotional or behavioral
difficulties . Regrettably, this chapter does not have sufficient space to
review all of these techniques and, moreover, they have not been devel-
oped specifically for abused children. Nonetheless, a few that may have
special relevance for abused children will be briefly mentioned.
Problem-solving strategies have been found to be useful in treating
a variety of problems, including hyperactivity, aggressiveness, and im-
TREATING THE ABUSED CHILD 255

pulsivity (Kendall & Braswell, 1985). This model could be employed with
abused children who display any of these types of "externalizing" prob-
lems in response to physical abuse (see Spivack, Platt, & Shure, 1976,
and Meichenbaum, 1977, for specific clinical instructions on how to use
problem-solving techniques with children). In a similar light, systematic
desensitization and other relaxation techniques with documented effec-
tiveness could be used to treat anxiety and fears in children who have
been abused or neglected (Walker et al., 1988). Finally, social skills train-
ing (Michelson, Sugai, Wood, & Kazdin, 1983; Mannarino, Christy, Dur-
lak, & Magnussen, 1982) could be the treatment of choice for abused
children with interpersonal deficits .
None of the aforementioned clinical interventions have been sys-
tematically evaluated as to their potential utility with abused children.
Furthermore, empirical studies need to be conducted that examine how
the effectiveness of these clinical procedures is affected by the child's
developmental status, family situation, or other variables. Nonetheless,
the important point that we are trying to make is that clinical child
psychology and child psychiatry have a number of treatment techniques
(primarily behavioral or cognitive-behavioral) with documented clinical
efficacy and that clinicians working with abused children should not
hesitate to utilize these interventions as part of their overall treatment
strategy.

Treatment of "Provocative Child Behaviors"

Another perspective in the treatment of physically abused children
is more specifically abuse-related. It is well known among therapists
who treat abusive families that the children who are at highest risk for
abuse may engage in a number of "provocative" behaviors that increase
parental frustrations and aggressiveness, including inattentiveness,
restlessness, noncompliance, and temper outbursts. These behaviors
can contribute to an interactive pattern in which the child's behaviors
raise parental frustrations and elicit greater aggressiveness which, in
turn, reinforce the child's provocative behaviors. A self-perpetuating
cycle is thus created that increases the probability of physical abuse.
A potentially useful model to deal with the kind of interactive pat-
tern mentioned above is to help the abused or at-risk child change these
provocative behaviors. As discussed earlier, Frazier and Levine (1983)
focused on assisting the victim to eliminate "child produced stressors"
that contribute to the total abuse cycle. A number of useful interventions
have also been outlined by Walker et al. (1988), who suggested that anger
management techniques can be used to help abused children reduce the
256 ANTHONY P. MANNARINO and JUDITH A. COHEN

level of aversive behaviors displayed toward parents. If the child can

learn to reduce the number and intensity of angry outbursts and to
exhibit increased self-control, this may have a positive impact in the
form of a decreased likelihood of future abuse.
There are other ways in which children at risk for abuse can change
the behaviors that perpetuate this abusive pattern. For example, children
can learn how to obey commands promptly and how to earn rewards for
adaptive behaviors (Walker et al., 1988). Moreover, self-instructional
training (Kendall & Braswell, 1985) and other cognitive-behavioral strat-
egies can be employed to reduce inattentiveness, restlessness, and
impulsivity.
It is important to state that this discussion in no way implies that
abused children bear any responsibility for having been physically as-
saulted. The authors recognize fully that abusive parents must take com-
plete responsibility for their assaultive behaviors. Nonetheless, if chil-
dren at-risk for abuse can be helped to alter some of the behaviors that
contribute to the abusive cycle discussed earlier, then perhaps the like-
lihood of subsequent physical maltreatment can be lowered.
None of the strategies to alter provocative behaviors in abused chil-
dren have been empirically investigated. It is certainly possible that
some will prove to be fruitful whereas others will not. Perhaps the
greatest advantage of this perspective is that it is inherently preventive
in nature . Specifically, it addresses one subset of variables (i.e., child
provocative behaviors) that contribute to a pattern of abuse in families. If
interventions can be implemented that reduce these provocative behav-
iors, then future abuse can perhaps be prevented or at least the risk
significantly reduced.
This section on the treatment of physical abuse has focused almost
exclusively on clinical interventions with individual victims. Space does
not permit a discussion of other potentially useful techniques such as
family treatment (which is sometimes suggested as an appropriate inter-
vention for physically abusive families). Moreover, a thorough discus-
sion of parent training and other clinical procedures with physically
abusive parents can be found in Chapter 11 of this volume.

TREATING THE SEXUALLY ABUSED CHILD

GENERAL CONSIDERATIONS

There is a growing body of literature addressing techniques for

treating sexually abused children. As is the case with treatment studies
TREATING TH E ABUSED CHILD 257

of child physical abuse, almost all of these are based on clinical rather
than empirical information. It is important to recognize that no study
exists to date that examines empirically the efficacy of any of these
therapeutic interventions in a methodologically sound manner. Thus,
although many of the clinical approaches to sexuall y abused children
seem inherently logical, there is no objective evidence that they are
indeed effective . This should be born in mind throughout the following
discussion.
In designing treatment programs, it would seem essential to define
what precisely is to be treated. One of the most important issues in the
field of child sexual abuse has been determining just what kind of prob-
lems these children experience. For many years, no one examined this
question empirically, because it was assumed that, based on clinical
experience, psychodynamic theory, or methodologically weak retrospec-
tive studies, sexually abused children had such problems as depression,
anxiety, poor self-esteem, dysfunctional families, impaired ability to
trust others, and poor assertiveness skills. Although these ideas often
seemed to fit with theoretical concepts of child sexual abuse and repre-
sented the best information available at the time , recent empirical stud-
ies have challenged some of those assumptions.
Researchers have demonstrated almost uniformly that sexually
abused children do not necessarily display significant symptoms of de -
pression, anxiety, or low self-esteem as measured by standardized self-
report measures (Cohen & Mannarino, 1988;Tufts New England Medical
Center, 1984; Einbender & Friedrich, 1989). No studies have empirically
examined such issues as tru st or assertiveness in the se child ren, no doubt
in part because such variables are difficult to measure objectively. Em-
pirical studies of family functioning of sexuall y abused children are also
lacking. Even studies that have demonstrated significant ps ycho-
pathology in sexually abused children (as measured by parental ratings
on standardized instruments), have noted that about one half of the
abused subjects were not rated as having any significant pathology
(Friedrich, Urquiza, & Beilke, 1986; Tufts New England Medical Center,
1984). Thus, it is not always obvious what problems need to be treated in
children who have been sexually abused. Also there seems to be great
variability in the type and severity of symptoms experienced.
This is not surprising if sexual abuse is conceptualized as a life event
(or series of events) rather than as a discrete psychiatric syndrome. In
this way, it is similar to experiencing a divorce or a death, and its impact
can vary enormously depending on many factors, some of which have
yet to be determined. Sexual abuse is a very diverse phenomenon. It
may be intra- or extrafamilial, vary in type of abu se, frequency, and level
258 ANTHONY P. MANNARINO and JUDITH A. COHEN

of force used. The context in which it occurs may greatly affect the
impact it has on a child. Abuse may occur in the context of a supportive
cohesive family who believe the child, or in a chaotic dysfunctional
family who either blame the child for the abuse or disbelieve him or her
altogether. It may happen to a well-adjusted socially skilled child, or to a
child who exhibited multiple behavioral, developmental, or emotional
problems prior to the abuse. Some children may be removed from the
home and go through criminal court proceedings, whereas others will
have no involvement in the criminal or child protective systems. Certain
aspects of the child's temperament, such as general adaptability to stress
and cognitive style, may influence the impact of abuse as well.
Because of the above factors, it stands to reason that child sexual
abuse would have variable effects on children. Thus, in planning treat-
ment it is not helpful to conceptualize sexual abuse as a unitary clinical
syndrome with certain constant behavioral and emotional features . Each
child and family must be evaluated carefully and individually to deter-
mine what issues are relevant and what problems need to be addressed
in therapy, rather than categorizing the child as a "sexual abuse victim"
and planning treatment around that label.

REVIEW OF TREATMENT STUDIES OF CHILD VICTIMS

In the last ten years, increasing numbers of treatment models for

sexually abused children have appeared in the literature. Virtually all of
these have come from clinical programs, whose primary focus is treat-
ment rather than research. This is one reason why these models have
lacked systematic empirical data regarding their utility. These meth-
odological shortcomings do not necessarily imply that these treatment
models are ineffective. Rather they highlight the fact that research in this
field is in its infancy, and that more controlled outcome studies are
needed.

Individual Psychotherapy
The case histories of sexually abused children, which have ap-
peared in the psychoanalytic literature for over half a century, have been
of theoretical interest and may have provided insight into the dynamic
issues of some of these children. Unfortunately, because of their focus
on the analytic process, the general applicability of such studies has
been limited.
With the advent of the women's movement in the late 1960s, rape
TREATING THE ABUSED CHILD 259

crisis centers and women's shelters provided an alternative resource for

abused women and children. Many of the innovative individual and
group therapy techniques for sexually abused children were started in
these centers. As many of the staff in these settings were not profes-
sional child therapists, but rather former victims or feminist activists,
their treatment approaches were not rooted in any particular theoretical
framework. The increasing numbers of sexual abuse victims coming to
treatment, however, enabled individual therapists to identify issues that
had been overlooked by the traditional analytic model. Porter, Bleck,
and Sgroi (1982) summarized these newly recognized treatment issues:
the "damaged goods" syndrome (where victims feel irreversibly phys-
ically or otherwise damaged as a result of the abuse), guilt, fear, depres-
sion, low self-esteem, repressed anger, impaired ability to trust, blurred
intrafamilial role boundaries, pseudomaturity, and impaired self-mas-
tery and control. Although recent empirical data have challenged the
idea that these problems typify most sexually abused children, recogni-
tion of these as potential issues to be explored was a major advance at
that time.
Many therapists have relied on play therapy to treat young victims
based on the assumption that emotionally laden experiences are easier
for children to express symbolically through play than to talk about
directly. One study (Mitchum, 1987) described the use of developmental
play therapy with five 4-year-old children over the course of 10 weeks.
Although Mitchum stressed the importance of developing a new trust-
ing relationship with an adult partner, details of the therapeutic inter-
ventions used were vague. In large part this is due to the nature of
nondirective play therapy, which is generally unstructured and difficult
to describe.
Although there is a wealth of anecdotal information written about
individual therapy in sexually abused children, systematic descriptions
of these therapeutic intervention techniques are lacking.

Group Therapy
Many therapists have described group treatment approaches for
child sexual abuse victims. The idea of group therapy has been appeal-
ing not only for its relatively high cost-effectiveness, but also because of
the observation that many of these victims feel different from other
children after having been abused; consequently, group treatment al-
lows them to meet and interact with other abuse victims and feel more
"normal." None of the following descriptive studies included outcome
260 ANTHONY P. MANNARINO and JUDITH A. COHEN

measures, but they have expanded the kinds of approaches that could
be attempted with this population of children.
Lubell and Soong (1982) described group therapy with sexually
abused children who had been placed in foster care (presumably be-
cause the perpetrator remained at home). They noted the importance of
addressing the feelings of loss that these children experienced as a result
of removal from the parental home. They also focused on the sense of
isolation the children felt because abuse had made them feel "different"
from peers. Another issue addressed in this model was the children's
anger toward the perpetrator, their families, and the system. They
stressed the utility of having co-therapists lead these groups, not only
for the additional sense of support it provided the clients but also to give
support to the therapists.
Berliner and Ernst (1984) described a group program that high-
lighted issues of self-protection, victims' acknowledgment that abuse
had occurred, and appropriate attribution of responsibility to the per-
petrator. They used a variety of therapeutic activities, including art proj-
ects, educational exercises, and focusing on appropriate intragroup in-
teractions.
Sturkie (1983) provided a structured 8-week therapy group for la-
tency-aged children who had been sexually abused. In this format, one
major treatment theme was addressed during each session, including
believability (i.e, the importance of telling the truth about abuse until
someone believes you), guilt and responsibility, body integrity and pro-
tection, secrecy and sharing, anger, powerlessness, other life crises and
tasks, and court attendance . They stressed the value of using role play-
ing to model and practice appropriate behaviors and expression of
feelings.
Damon and Waterman (1986) designed a parallel group treatment
model for treating abused children (aged 8 and younger) and their moth-
ers in concurrent groups. They provided a clear and detailed description
of their therapeutic interventions. Thirteen modules were presented to
the children and parents, carefully coordinated, so that the mothers
would be sensitive to and prepared for what the children had addressed
each week in therapy. Some of the issues focused on included the right
to say no, the emphasis that private parts are private, whom to tell if
abuse occurs, fault and responsibility, anger and punishment, and sex
education. These authors offered a unique wealth of clinical materials,
such as stories and activities, that are particularly useful with younger
children, as well as many practical suggestions. In sum, Damon and
Waterman have provided an excellent descriptive account of an inno-
vative group therapy model for sexually abused children.
TREATING THE ABUSED CHILD 261

Family Therapy

Much of the writing in the field of child sexual abuse exclusively

addresses intrafamilial abuse. For many years, there was a widespread
assumption that most sexual abuse was incestuous, and that in these
cases the mother was collusive with the abuse. Recent demographic
studies have challenged both of these assumptions, because most re-
ported abuse has been found to occur outside of the nuclear family
(Finkelhor, 1979; Mannarino & Cohen, 1986). Furthermore, studies have
noted that the majority of mothers of incest victims report the abuse to
authorities, as soon as it is disclosed to them (Mannarino & Cohen,
1986), and choose to reject their mate in order to protect their children
(Myer, 1985). This challenges the theory that intrafamilial abuse is
largely a function of problematic mother-child relationships or that
mothers of abused children know about and tolerate the abuse.
On the other hand, there are clearly some cases where the mother
knows about and colludes with ongoing abuse. In addition, often family
members are in need of therapeutic interventions even in cases of extra -
familial abuse. Thus, it is once again essential to assess each child and
family, to clarify which members need treatment, and what issues are
pertinent in their particular situation.
Sgroi (1982) suggested a format for evaluating the type and degree
of family treatment required in child sexual abuse cases. She stated that
in all cases, five factors should be examined: (1) poor supervision of the
children, (2) poor choice of surrogate caretakers, (3) inappropriate sleep-
ing arrangements, (4) blurred role boundaries, and (5) sexual abuse by a
family member. She discussed varying treatment approaches to the fam-
ily depending on which of these factors is problematic. She also pointed
out the need for more family treatment programs and studies of treat-
ment outcome.
Furniss-Tilman (1983) proposed a treatment approach for inces-
tuous families, specifically those cases involving father-daughter incest.
The program is based on the author's clinical experience with 27 of these
families . In this program, the initial step is work on intergenerational
boundaries in the family system, including defining and strengthening
appropriate generational bonds and boundaries. Then the problems in
the mother-daughter dyad are addressed, including issues of competi-
tion and the mother's responsibility to protect the child. After those
problems are resolved, the relationship between the mother and father
is the focus of attention. When this relationship becomes more appropri-
ate, the father-daughter relationship is examined.
A treatment program described by Giarretto (1982) has received a
262 ANTHONY P. MANNARINO and JUDITH A. COHEN

great deal of attention, largely because of its very low (1%) reported
recidivism rate. This project primarily treats father-daughter incest
cases, and is based on work with over 4,000 such families. Treatment
consists of a joint effort involving professionals in the mental health,
criminal justice, and child protective service systems. There are self-help
components (Parents United, Daughters and Sons United) that provide
support and other ancillary services. Therapy focuses on mother-
daughter counseling to overcome their mutual alienation. A variety of
services are provided, including group, family, and individual therapy.
Giarretto's study is unusual in that it provides follow-up data. Ninety
percent of the victims were reunited with their families, and the re-
ported recidivism rate was less than 1%, a result that is very promising,
but it must be noted that these statistics do not necessarily reflect
healthy outcomes. Children returned home were not necessarily free of
psychopathology. Also, the lack of reported reabuse does not neces-
sarily imply that it did not recur. It is possible that many victims or
families, after having gone through extensive interventions, including
the child's removal from the home, may have been more hesitant to
report abuse a second time.
The major limitation of this study, however, is that 90% of the
offenders in this program took full responsibility for the sexual abuse.
This fact is unrepresentative of sexual abuse cases in general, where the
overwhelming majority of perpetrators deny either responsibility for the
abuse or that it happened at all. This suggests that Giaretto's program
treats a highly selected population, and raises questions of how applica-
ble it would be to the majority of incestuous families . Nevertheless, it is
an impressive program for many reasons, including the high level of
cooperation among agencies.
Zimmerman, Wolbert, Burgess, and Hartman (1987) described a
modified family group treatment method, used in cases in which multi-
ple children are abused by the same offender. This model makes use of
artwork within a group of intrafamilial peers, with a great deal of atten-
tion paid to the attributions the children form about the abuse. Specifi-
cally, the authors used attributional questions to examine the victims'
causal beliefs about the abuse. This model also used peer/family support
to prepare the children for court appearances.
In many families in which the nonabusive members respond
promptly and appropriately, the main issues may be in dealing with the
stress of the disclosure, possible feelings of guilt regarding the abuse,
the loss of the perpetrator from the family, and the subsequent legal
proceedings that occur with regard to custody, visitation, and criminal
charges. Porter et al. (1982) pointed out that ambivalent feelings are very
TREATING THE ABUSED CHILD 263

common in this situation, and that these need to be addressed in thera-

py with the family members who are affected . There is a lack of studies
addressing family treatment in cases of extrafamilial abuse. However, it
is generally accepted among therapists that family members experience
stress regarding sexual abuse and disclosure and may need education
and support concurrent with the victim's involvement in therapy.

Educational Interventions
Many programs have included victim and family education as part
of the treatment, including such interventions as teaching self-protective
skills, assertiveness training and the right to say no to intrusive behavior
(Berliner & Ernst, 1984; Damon & Waterman, 1986; Sturkie, 1983), differ-
entiating between appropriate and inappropriate touching (Damon &
Waterman, 1986; Sturkie, 1983), what to do if abuse occurs again
(Damon & Waterman, 1986), and education about legal procedures and
going to court (Sturkie, 1983; Zimmerman et al., 1987). Parents as well as
victims often need information about the complex criminal justice sys-
tem and support with regard to court appearances. The kinds of educa-
tional tools available have increased greatly. Coloring books about recog -
nizing and reporting sexual abuse, card and board games teaching how
to avoid potentially abusive situations, and children's books and vid-
eotapes about going to court are now commonplace in centers that treat
sexually abused children. Some of these are also used in sexual abuse
prevention programs. Although the efficacy of such interventions has
yet to be demonstrated, clinicians have frequently found these aids very
helpful.
But because of the complex systems involvement in many sexual
abuse cases, often the therapist is obliged to take on the task of coordi-
nating the various services available, including making a referral for a
physical examination (to rule out or treat possible sexually transmitted
diseases or traumatic genital injuries), educating the family about, and
possibly accompanying the child to, various legal proceedings (juvenile
or criminal court hearings for charges pressed against the perpetrator,
family court hearings to resolve custody and visitation issues), and re-
maining in close contact with child protective services workers to keep
mutually informed about progress and recommendations regarding the
child's situation. Many victims' centers provide advocates who assume
some of these responsibilities, allowing the therapist to concentrate on
treatment. However, any therapist working with sexually abused chil-
dren should have a thorough understanding of the systems involved
and be prepared to spend considerable time on case management and
264 ANTHONY P. MANNARINO and JUDITH A. COHEN

liaison activities. Often it seems that providing these services is as help-

ful to the child and family as is the actual therapy.

NEW DIRECTIONS IN THE TREATMENT OF SEXUALLY

ABUSED CHILDREN

Attribution Theory
Zimmerman et al. (1987) devoted a great deal of time in treatment to
the attributions children make about sexual abuse. Attributional style
has received increased attention recently among researchers. It appears
that this may be one of the important mediating factors in determining
how symptomatic a child may become following sexual abuse. At the
present time, however, there is no empirical evidence to support or
challenge this idea.
Theoretically, two related attributional factors may be involved. The
first pertains specifically to the abuse: To what does the child attribute
the abuse? Some children place full responsibility on the perpetrator,
whereas other victims feel the abuse was entirely their own fault. Many
children fall somewhere in between, believing that the abuse was basi-
cally the perpetrator's fault, but that some facets of the victim (such as
being handsome or pretty, being friendly, or being too weak to fight
back) also contributed in some degree to the abuse. The second factor is
the child's general attributional style: To what does the child attribute
typical life experiences, such as failing an exam or making a new friend?
Is the child's style to attribute such occurrences to aspects of himself or
herself, or to aspects of the outside world, or to some combination?
Future studies could focus on whether either of these attributional
factors affect significantly the development or avoidance of psycho-
pathology, and whether there is an attributional style that is optimal for
recovery from the abuse. If so, cognitive therapy approaches could be
utilized to alter the child's attributions in a way that could positively
affect outcome . Clearly, more research is needed before conclusions can
be drawn about the role of attributional style; however, it is a promising
area for future treatment designs.

Traumagenic Dynamics
Other possible directions for treatment have been suggested by
Finkelhor and Browne (1985) who discuss four concepts that are poten-
tial foci for therapy: traumatic sexualization, stigmatization, betrayal,
and powerlessness. They described these concepts as "traumagenic dy-
TREATING THE ABUSED CHILD 265

namics" that alter the child's cognitive and emotional orientation to the
world, and distort the child 's self-concept, world view, or affective abili-
ties. This conceptualization suggests that possible intervention strat-
egies could be designed to correct these cognitive and emotional distor-
tions.

Behavior Therapy
Berliner and Wheeler (1988) proposed a conceptualization that child
sexual abuse results in conditioned anxiety and socially learned mal-
adaptive responses. They suggested that effective therapy may involve
the use of established modalities, such as systematic desensitization,
relaxation training, and problem-solving training.
Finally, behavior modification programs may be useful in control-
ling many of the problematic symptoms displayed by some sexually
abused children. Aggressive behavior, sexually provocative or overt sex-
ual behavior, and enuresis are all potentially responsive to behavioral
interventions that could be implemented concurrently with other thera-
peutic strategies. Although this approach has been used by some clini-
cians/ it seems to be dismissed by many others who believe that focusing
on behavioral symptoms will obscure the underlying psychological is-
sues. It is not suggested that these psychological, abuse-related issues
are less important to address. However, it does seem critical to reduce
these types of maladaptive behaviors quickly, in order to prevent sec-
ondary problems, such as loss of peers through aggressive behavior, or
ostracism or re-abuse because of inappropriate sexual behavior.
The future direction of treatment for sexually abused children de-
pends to a large degree on the availability of empirical information. In
order to choose effective treatment approaches, there must be more
systematic data gathered with regard to the impact of abuse, and the
effect and outcome of well-defined treatment modalities.

SUMMARY

This chapter has attempted to accomplish a number of tasks. First,

the methodological issues and problems in conducting treatment out-
come research with abused children were addressed. Next, in separate
sections, existing treatment studies for physically and sexually abused
children were reviewed and a clinically oriented discussion of poten-
tially useful therapeutic techniques with each group was presented.
Emphasis throughout this chapter has been that there is little empirical
266 ANTHONY P. MANNARINO and JUDITH A. COHEN

documentation of the effectiveness of treatment modalities with abused

children and that further research is clearly needed to fill this void.

REFERENCES
Berliner, L., & Ernst, E. (1984). Group work with preadolescent sexual assault victim s. In 1.
R. Stuart & J. G. Greer (Eds .), Victims of sexual aggression: Treatment of children, women
and men (pp . 105-124). New York: Van Nostrand Reinhold.
Berliner, L., & Wheeler, J. R. (1988). Treating the effects of child sexual abuse on children.
Journal of Interpersonal Violence, 2, 415-434 .
Blythe, B. J. (1983). A critique of outcome evaluation in child abuse treatment. ChildWelfare,
62, 325-335.
Cohen, J. A., & Mannarino, A. P. (1988). Psychological symptoms in sexually abused girls.
Child Abuse and Neglect, 12, 571-577.
Cohn, A. H. (1979). An evaluation of three demonstration child abu se and neglect treat-
ment programs. Journal of the American Academy of Child Psychiatry, 18, 283-291.
Conte, J., & Schuerman, J. R. (1987). The effects of sexual abuse on children: A multidi-
mensional view. Journal of Interpersonal Violence, 2, 380-390 .
Damon, L., & Waterman, J. (1986). Parallel group treatment of children and their mothers.
In K. MacFarlane & J. Waterman (Eds .), Sexual abuse of young children: Evaluation and
treatment (pp . 244-298) . New York: Guilford Press.
Einbender, A. J., & Friedrich, W. N . (1989). The psychological functioning and behavior of
sexually abused girls . Journal of Consulting and Clinical Psychology, 57, 155-157 .
Finkelhor, D. (1979). Sexually victimized children. New York: free Press .
Finkelhor, D., & Browne, A. (1985). The traumatic impact of child sexual abu se: A concep-
tualization. American Journal of Orthopsychiatry, 55, 530-541.
Frazier, D., & Levine , E. (1983). Reattachment therapy: Intervention with the very young
abused child. Psychotherapy: Theory, research and practice, 20, 90-100.
Friedrich, W. N ., Urquiza, A. J., & Beilke, R. (1986). Behavior problems in sexually abu sed
young children. Journal of Pediatric Psychology, 11, 47-57.
Furniss-Tillman, T. (1983). Family proce ss in the treatment of intrafamilial child sexual
abuse. Journal of Family Therapy, 5, 263-278 .
Cabinet, L. (1983a). Child abuse treatment failures reveal need for redefinition of the
problem. Child Abuse and Neglect , 7, 395-402 .
Cabinet, L. (1983b). Shared parenting: A new paradigm for the treatment of child abuse.
Child Abuse and Neglect, 7, 403-411.
Giarretto, H . (1982). A comprehensive sexual abuse treatment program. Child Abuse and
Neglect, 6, 263-279 .
Green, A. H ., Power, E., Stonebrook, B., & Gaine s, R. (1981). Factors associated with
succe ssful and un successful intervention with child abuse in families . Child Abuse and
Neglect, 5, 45-52.
Heide, J., & Richardson, M. T. (1987). Maltreated children's developmental scores : Treat-
ment versus nontreatment. Child Abuse and Neglect , 11, 29-34 .
Huebner, E. S. (1984). A group treatment approach for abused middle school students.
Techniques, 1, 139-143.
Isaacs, C. D. (1982). Treatment of child abu se: A review of the behavioral interventions.
Journal of Applied Behavioral Analysis, 15, 273-294.
TREATING THE ABUSED CHILD 267

Kendall, P., & Braswell, L. (1985). Cognitive-behavioral therapy for impulsive children. New
York: Guilford Press .
Lubell, D., & Soon g, W. (1982). Group therapy with sexually abu sed adolescents. Canadian
Journal of Psychiatry, 27, 311-315 .
Lutzker, J. R., & Rice, J. M. (1984). Project 12-Ways: Measuring outcome of a large in-home
service for treatment and prevention of child abuse and neglect. Child Abuse and
Neglect, 8, 519-524 .
Mannarino, A. P., Christy, M., Durlak, J. A., & Magnussen, M. G. (1982). Evaluation of
social competence training in the schools. Journal of School Psychology, 20, 11-19.
Mannarino, A. P., & Cohen, J. A. (1986). A clinical-demographic study of sexually abused
children. Child Abuse and Neglect , 10, 17-28.
Mask, E., Johnson, C.; & Kovitz , K. (1983). A comparison of the mother-child interactions
of physically abused and non-abused children during play and task situations. Journal
of Clinical Child Psychology, 12, 337-346 .
Meichenbaum, D. (1977). Cognitive-behavior modification: An integrative approach. New York:
Plenum Press.
Michelson, L. , Sugai, D. P., Wood, R. P., & Kazdin , A. E. (1983). Social skillsassessment and
training with children: An empirically based handbook. New York: Plenum Press.
Mitchum, N. T. (1987). Developmental play therapy: A treatment approach for child vic-
tims of sexual molestation. Journal of Counseling and Development, 65, 320-321.
Myer, M. H . (1985). A new look at mothers of incest victims . Journal of Social Work and
Human Sexuality, 3, 47-58.
Myers, P. A., Brandner, A., & Templin, K. (1985). Developmental milestones in abused
children and their improvement with a family-oriented approach to the treatment of
child abuse. Child Abuse and Neglect, 9, 245-250.
Pelcovitz, D., Kaplan, S., Samit , C.; Krieger, R. , & Cornelius, D. (1984). Adolescent abuse:
Family structure and implications for treatment. Journal of the American Academy of
Child Psychiatry, 23, 85-90.
Porter, F. S., Bleck, L. c., & Sgroi, S. M. (1982). Treatment of the sexually abused child . In
S. M. Sgroi (Ed.), Handbook of clinical intervention in child sexual abuse (pp . 113-148).
Lexington, MA: Lexington Books.
Reid, J. B., Taplin, P. S., & Loeber, R. (1981). A social interactional approach to the treat-
ment of abu sive families. In R. B. Stua rt (Ed.), Violent behavior: Social learning approaches
to prediction , management, and treatment (pp . 83-101) . New York: Brunner/Maze!.
Sgroi, S. M. (1982). Family treatment of child sexual abuse. Journal of Social Work and Human
Sexuality, 1, 109-128 .
Shelton, P. R. (1982). Separation and treatment of child-abusing families . Family Therapy, 9,
53-60.
Smith, J. E., Rachman, S. J., & Yule, B. (1984). Non-accidental injury to children-III.
Behaviour Research and Therapy, 22, 367-383.
Spivack, G., Platt, J. J., & Shure, M. B. (1976). Theproblem-solving approach to adjustment. San
Francisco : Jossey-Bass.
Sturkie, K. (1983). Structured group treatment for sexually abused children. Health and
Social Work, 8, 299-308 .
Trowell, J., & Castle, R. L. (1981). Treating abused children. ChildAbuseand Neglect, 5,187-
192.
Tufts New England Medical Center, Division of Child Psychiatry (1984). Sexuallyexploited
children: Service and research project. Final report for the Office of Juvenile Justice and
Delinquency Prevention. Washington, DC: U.S. Department of Justice.
268 ANTHONY P. MANNARINO and JUDITH A. COHEN

Walker, C. E., Bonner, B. L., & Kaufman, K. L. (1988). The physically and sexually abused
child: Evaluation and treatment. New York: Pergamon Press .
Williams, G. J. (1980). Management and treatment of parental abu sive and neglect of
children. In G. J. Williams & J. Money (Eds .), Traumatic abuse and neglect of children at
home (pp . 483-487). Baltimore : Johns Hopkins University Press .
Wolfe, D., & Mosk, M. (1983). Behavioral comparisons of children from abu sive and
distressed families . Journal of Consulting and Clinical Psychology, 51, 702-708 .
Zimmerman, M. L., Wolbert, W. A., Burgess, A. w., & Hartman, C. R. (1987). Art and
group work: Interventions for multiple victim s of child molestation (Part II). Archivesof
Psychiatric Nursing, 1, 40-46.
 CHAPTER 11

TREATING THE CHILD ABUSER

JEFFREY A. KELLY

INTRODUCTION
Scientific, empirically ba sed approache s to the treatment of any problem
depend upon the adequacy of theoretical models concerning the cause of
the problem. As discussed in other chapters, for many years there was a
paucity of sou n d empirical data that could be used either to account for
the presence of child abusive patterns or to gu ide the de velopment of
effective interventions for child abus ers. Prior to the mid-1970 s, mo st
models of child abuse were unifactorial in nature and attempted to
predict the occurrence of child maltreatment in familie s from single
etiological causes, such as parent psychopathology or sociological disad-
van tage (see reviews by Belsky, 1980; Burgess, 1979; Parke & Collmer,
1975). Treatment interventions based on such models also tended to be
unifactorial, usually emphasizing treatment of a parent's postulated un-
derlying psychiatric disorder or the alleviation of socioeconomic distress .
However, most of these therapeutic interventions were unevaluated or
investigated only with uncontrolled and anecdotal outcome reports.
Over the past 10 years, theories of child abuse have become more
ecological and multifactorial, stressing the functional interplay between
a variety of child characteristics; parent characteristics, cognitive-behav-
ioral skills, and coping strategies; and environmental influences on the

JEFFREY A. KELLY· Division of Psychology, University of Mississippi Medical Center,

Jackson, Mississippi 39216.

269
270 JEFFREY A. KELLY

family (Burgess, 1979; Burgess & Richardson, 1984; Kelly, 1983; Parke &
Collmer, 1975; Wolfe, 1985). The emergence of these more complex so-
cial-interactional conceptual models of child abuse has given rise to new
approaches to the treatment of child abusive parents. Because outcomes
of therapy for child maltreating parents have been studied scientifically
for only a short period of time, this field is still in its earliest stages, and
many questions remain uninvestigated. However, sufficient information
has already been gained to guide the development of clinical research
and intervention for child abusive parents. Thus, this chapter will briefly
discuss conceptual and practical issues related to treatment for child
abusers, review empirical research on therapy outcomes for this popula-
tion, and consider topics important for further treatment research.

MULTIFACTORIAL MODELS OF CHILD ABUSE:

IMPLICATIONS FOR TREATMENT
If the occurrence of child maltreatment is deemed to be a conse-
quence of what potentially are multiple parent, child, and environmen-
tal factors (and the reciprocal influences between them), it is apparent
that there also are many different potential targets or objectives for
treatment intervention. For example, it has now been well-established
that child abusive parents relative to their nonabusive counterparts ex-
hibit deficits in child management skills (Burgess, 1979; Burgess & Con-
ger, 1978), emotional overreactivity to aversive cues of child behavior
(Disbrow, Doerr, & Caulfield, 1977; Frodi & Lamb, 1980;Wolfe, Fairbank,
Kelly, & Bradlyn, 1983), higher levels of conflict and violence with family
members other than the abused child (Burgess & Conger, 1978; Lahey,
Conger, Atkeson, & Treiber, 1984; Reid, Taplin, & Lorber, 1981), greater
mislabeling and misattributions concerning child behavior (Mash, John-
ston, & Kovitz, 1983), higher behavioral impulsiveness (Rohrbeck &
Twentyman, 1986), and deficient social problem-solving skills (Azar,
Robinson, Hekinans, & Twentyman, 1984). Such findings indicate that
treatment for child abusive parents might include parent training, anger
control and stress management training, marital and family therapy,
cognitive and reattributional therapy, and problem-solving training. Fur-
thermore, although psychiatric and sociological "single-factor" theories
of child abuse are criticized as overly narrow and simplistic, between 5%
and 10% of abusive parents do exhibit demonstrable and severe psycho-
pathology (Bell, 1973; Kempe, 1973; Kempe & Kempe, 1978), and so-
cioeconomic stress is implicated in a significant proportion of reported
abuse cases (Garbarino, 1976; Garbarino & Sherman, 1980; Giovannoni
TREATING THE CHILD ABUSER 271

& Billingsley, 1970). Multifactorial models of child abuse create multiple

potential targets for therapeutic intervention although it is not neces-
sarily the case that all abusive families have all these problems. A ques-
tion of practical clinical and research importance is that of determining
which factors contribute functionally to child maltreatment in a given
family so that parent treatment can address the most relevant problem
areas for that family (Kelly, 1983).
In addition to multiple problem areas related to abuse that can be
targeted in treatment, the term child abuse itself is very broad and sub-
sumes a number of different patterns of child maltreatment. Extreme
physical violence directed by an adult toward a child is the form of
maltreatment most often considered abuse and is the problem most
often studied in the child abuse literature. However, child maltreatment
because of neglect may occur even more often than cases of physically
violent abuse (Mayhall & Norgard, 1983), and recent studies suggest
that sexual abuse occurs at a rate and with psychological consequences
more widespread than previously believed (Finkelhor & Hotaling, 1984;
Kolko, 1987; Mrazek, 1983). Although physically abusive and neglectful
parents share similar patterns in certain treatment-relevant areas, such
as child management skill deficits (Burgess & Conger, 1978) and physio-
logical arousal to cues of child behavior (Disbrow et al., 1977), there is
little a priori evidence that physical abuse, sexual abuse, and child ne-
glect share the same causes or will respond to identical therapeutic
interventions. Treatment that is effective for certain child maltreating
parents may prove less effective or be ineffective for parents who malt-
reat their children in different wa ys and for different reasons.
If multifactorial and ecological perspectives have produced more
complex models of child abuse and family violence, they also set the
stage for more innovative and well-controlled studies of treatment out-
come. Prior to the mid-1970s, there was little, if any, empirical research
on the efficacy of treatment for child abusers. Since that time, uncon-
trolled descriptive and anecdotal reports have gradually given rise to
more scientifically sound evaluations. We will next turn our attention to
treatment interventions for abusive parents.

INTERVENTIONS TO IMPROVE PARENTING SKILLS

The rationale of child management training for abusive parents is
that violent parenting acts, which can result in child injury, may develop
because of the parent's failure to manage child behavior in a more appro-
priate, nonviolent manner. From this perspective, a parent's over-
272 JEFFREY A. KELLY

reliance on ineffective child control strategies and high rates of corporal

punishment increase child aversive behavior, escalate the intensity and
frequency of aversive interchanges between the parent and child, and
establish an ongoing "coercive cycle" between them (d. Patterson, 1977;
Patterson, Reid, Jones, & Conger, 1975). Use of physical punishment is
maintained probably because it can produce temporary suppression of
child's misbehavior, although organisms adapt to the intensity of fre-
quently administered punishment (Azrin, Holz, & Hake, 1963; Rey-
nolds, 1975) and, over time , escalating levels of punishment are needed
to produce the same degree of behavioral suppression. Ultimately, par-
ents who rely excessively on physically punitive controls may in fact
injure their children. This formulation is consistent with a substantial
body of observational research that has demonstrated the occurrence of
fewer positive parental behaviors, more frequent negative and aversive
behaviors, and lower levels of attentiveness to the child by abusive
parents than control parents, even during routine parent-child interac-
tions (Bousha & Twentyman, 1984; Burgess, 1979; Burgess & Conger,
1978). For these reasons, a number of investigators have employed par-
ent-training methods to alter the disciplinary style and parent-child
interaction patterns of abusive parents.

PARENTING SKILLS INTERVENTIONS WITH INDIVIDUAL

CLINICAL CASES

Mastria, Mastria, and Harkins (1979) conducted one of the earlier

case study parent-training interventions with a mother who physically
abused her 7-year-old child . Treatment in the study consisted of vid-
eotaping parent-child interactions that took place in a clinic playroom
and providing video feedback, therapist modeling of appropriate means
to handle child misbehavior, and homework assignments to practice the
same skills in vivo. Parent behaviors targeted for intervention included
the use of positive verbal reinforcement, ignoring inappropriate child
behavior, and problem-solving skills to identify nonviolent methods to
handle child misbehavior. Mastria et al. (1979) reported increases in ob-
served appropriate parenting skill behavior after treatment and at a 3-
month followup. Moreover, these changes were corroborated by parent,
child, and therapist reports of improved adjustment and coping by the
parent and her child . However, this early study did not employ experi-
mental controls and presented limited data on the specific nature of
change observed in parent-child interactions.
Sandler, Van Dercar, and Milhoan (1978) also conducted an early
TREATING THE CHILD ABUSER 273

investigation of parent-training effectiveness, but provided intervention

and conducted observational assessments within the homes of two
abusive families rather than in a clinic setting. In both cases, the abusive
parents were single mothers who were reported to have beaten their
children. Observations were conducted over seven to nine occasions at
mealtimes to establish pretreatment baselines for the occurrence of 29
different categories of parent and child behavior using the coding sys-
tem developed by Patterson, Ray, Shaw, and Cobb (1969). The parents
initially exhibited low rates of positive verbal and positive physical be-
havior, frequent negative commands, and minimal conversational talk
with their children. Intervention in this project consisted of child man-
agement technique reading assignments, therapist modeling of appro-
priate parenting skills, and tangible reinforcement for parent compliance
in the training. Treatment produced substantial increases in positive
verbal and positive physical parent behavior, increased conversational
talk between parent and child, decreased use of negative commands,
and more frequent affectionate laughter. Most changes were well main-
tained at 4-month follow-up, although the Sandler et al. (1978) study did
not employ an experimental design.
A similar intervention was also conducted by Wolfe and Sandler
(1981). In this study, each of three physically abusive parents received
training to improve interactions with their children, including the han-
dling of child misbehavior. As in the Sandler et al. (1978) investigation,
both parent-training and family observations took place in family
homes, and the parent-training intervention included child-manage-
ment reading assignments, therapist modeling of appropriate skills,
role-playing of problem situations, and feedback to the parent following
behavior rehearsal. Over the course of the lO-session intervention, sub-
stantial decreases were observed in aversive child behavior (such as
crying, noncompliance, and verbal or physical negativity) and aversive
parent behavior (including negative commands and yelling). Increases
in the proportion of child compliance to parental commands were also
associated with treatment, and all effects were maintained at long-term
follow-up. Treatment conditions in the Wolfe and Sandler (1981) study
were manipulated to determine whether contingency contracting with
the parent to utilize one new child-management technique at home each
week would produce greater skill change. Although the investigators
reported that this additional homework contracting was useful clinically,
it was not associated with improvement beyond that produced by in-
the-home parent training alone.
A somewhat different parent-training approach was employed in a
case reported by Wolfe, St. Lawrence, Graves, Brehony, Bradlyn, and
274 JEFFREY A. KELLY

Kelly (1982), with treatment conducted in a clinic playroom setting and

its impact assessed by observations of parent-child interaction behavior
made during "staged" tasks in both clinic playroom and home settings.
The staged tasks involved situations that were intended to elicit child
noncompliance (picking up and sorting a large number of toys) as well
as positive parent-child behavior (playing a game together). The parent
who received treatment in this case demonstration was a 29-year-old
single mother of two mentally retarded 9-year-old twin boys; welfare
authorities reported the children were difficult to control and that the
mother injured them during frequent spankings and whippings. Follow-
ing baseline playroom observations, the therapists providing treatment
coached and modeled appropriate means to handle child behavior, in-
cluding the use of verbal and physical touch reinforcement, selective
ignoring of minor misbehavior, and time out for more serious miscon-
duct. During practice parent-child interaction in each session, the moth-
er was prompted and herself reinforced by the observing therapists via a
"bug-in-the-ear" remote receiver that she wore for using trained skills.
Experimental control in the study was achieved by means of a multiple
baseline introduction of treatment attention first targeting the mother's
hostile physical and verbal behavior and then improving her positive
interaction skills. Reductions in hostile behavior and increases in posi-
tive reinforcing skills during clinic and at-home observations were found
and were maintained 2 months following intervention. There were no
further reports of child maltreatment by welfare caseworkers after the
completion of treatment. A very similar intervention was conducted by
Crimmins, Bradlyn, St. Lawrence, and Kelly (1984) with a parent who
was both abusive and neglectful toward her 4-year-old son with com-
parable results.

GROUP COMPARISON STUDIES OF

PARENT-TRAINING INTERVENTIONS

In an extension of their earlier work with individual clinical cases,

Wolfe, Sandler, and Kaufman (1981) offered child-management training
to a group of eight abusive parents referred by a child welfare agency.
Most of the parents were of low income, all had engaged in physical
abuse, and their children ranged in age from 2 to 10 years. A demo-
graphically-comparable set of eight other abusive parents served as a no-
treatment control group and completed assessment measures but re-
ceived no parent training.
All parents in the study completed the Eyberg Child Behavior In-
TREATING THE CHILD ABUSER 275

ventory (ECBI) (Eyberg & Ross, 1978) and were evaluated on global
measures of family functioning by social service caseworkers unin-
volved in the parent-training program. In addition, all families in the
study were observed in their homes during free (unstructured) interac-
tion periods, during tasks in which the parent taught the child a new
puzzle, and during tasks in which the parent was asked to elicit the
child's compliance in picking up toys. All observed interactions were
coded for the frequency of parent positive reinforcement techniques,
appropriate commands and prompts, and appropriate punishment.
Following these baseline assessments, parents in the experimental
condition attended a series of group sessions that provided instruction
in child development and management, behavioral principles applied to
parenting (such as positive reinforcement, time out, shaping, and ap-
propriate punishment), problem-solving of child-management difficul-
ties, and relaxation and self-management skills. In addition to the week-
ly groups, each family received individual in-home child-management
training tailored to specific problems of that family. At the conclusion of
the experimental group's intervention, all families were reassessed on
the same set of measures that had been used during baseline. Significant
improvement was found for experimental group parents relative to the
control group parents on skill behaviors during the observed parent-
child interaction tasks, and changes were well maintained at a 10-week
followup of five of the parents. Differential change was not found be-
tween the child-management training and control groups for ECBI
scores or caseworker ratings, perhaps because of the relatively small
sample sizes in the study. Based on a I-year follow-up inspection of
welfare department records, none of the families that received treatment
was suspected of further abuse.
The Wolfe et al. (1981) project, like most others reported in the
literature on child-management training for abusive parents, taught par-
enting skills based largely on operant principles. Brunk, Henggeler, and
Whelan (1987) compared the relative efficacy of operant-behavioral
child-management training with a multisystemic family therapy inter-
vention approach in a sample of 33 families with a history of child abuse
or neglect Parents in the child-management training intervention at-
tended an 8-session series of groups, modeled after the group treatment
used by Wolfe et al. (1981), which taught general behavioral parenting
skills and skills for handling specific child problems experienced by each
family. In-the-home training was not conducted with individual fami-
lies. Subjects in the multisystemic therapy condition received eight ses-
sions of family therapy, conducted with individual family units rather
than in groups; the intervention was based on family restructuring prin-
276 JEFFREY A. KELLY

ciples (d. Haley, 1976; Minuchin, 1974) rather than behavioral training.
Before and after intervention, all parents were assessed using symptom
and child behavior problem checklists, family environment and social
system self-report measures, and treatment satisfaction questionnaires.
In addition, observations were made of parent-child interactions during
a 10-minute, in-home talk in which the parent was asked to teach his or
her child block designs of increasing difficulty level. These interactions
were rated for verbal and nonverbal measures of parental control style
using the Schaffer and Crook (1979) coding system.
Brunk et al. (1987) found that the child-management training and
the multisystemic family therapy produced significant and comparable
reductions in symptoms of parents' emotional distress, reduced overall
family stress, and reduced severity of identified problems. The multi-
systemic family therapy intervention produced greater improvement
than did the child-management training on observational measures of
effectiveness during the parent-child interaction task. Parents who re-
ceived this treatment showed increased effectiveness in child control
skills and were more appropriately responsive to child behavior. Also,
some collateral improvements in child behavior were observed. How-
ever, these results must be viewed as preliminary because child-man-
agement training was conducted in groups, whereas the family therapy
intervention was provided to individual family units, no follow-up was
conducted, and a no-treatment control was not employed. Nonetheless,
the results of Brunk et al. (1987) suggest that attention to factors beyond
child-management skill alone may produce additional improvement in
family functioning.

INTERVENTIONS THAT ADDRESS PARENT COPING

The treatment programs described in the previous section, with the
exception of the Brunk et al. (1987) examination of family therapy and
the Wolfe, Kelly, and Drabman (1981) inclusion of a relaxation training
intervention component, entailed relatively direct and straightforward
applications of behavioral child-management training. Although most
empirical studies on child abuser treatment include such parent train-
ing, a number of researchers have incorporated other coping training
elements in their interventions.
Denicola and Sandler (1980) combined child-management training
with training in self-control, anger, and stress management skills for two
physically abusive mothers. The child-management treatment compo-
TREATING THE CHILD ABUSER 277

nent was conducted in standard fashion and included reading assign-

ments, verbal instruction, modeling, behavior rehearsal, and therapist
feedback concerning appropriate means to handle child behavior prob-
lems interactions with children. The self-control treatment element en-
tailed deep muscle relaxation training, self-instructional training to con-
trol arousal and anger, stress inoculation procedures, and problem-
solving practice in citations that would typically produce aggressive
behavior. The child-management and self-control elements were intro-
duced separately in a withdrawal experimental design, and each parent
received 12 treatment sessions. The introduction of intervention,
whether child management or self-control skills, produced reductions in
total aversive behavior for both the parent and the child during routine
interactions observed and coded in the home. Thus, both intervention
elements resulted in improved parent-child interaction patterns; in addi-
tion, each parent reported greater skill in handling child-related problems
and reduced anger and emotional upset.
In similar fashion, Scott, Baer, Christoff, and Kelly (1984) described
the treatment of a mother who physically abused her ll-year-old son
primarily at times when she was already angry and when the child then
misbehaved. Assessment revealed a recurrent pattern in which the par-
ent (1) became frustrated following interpersonal conflicts with other
adults which she usually handled passively and unsatisfactorily, (2) ex-
perienced increased feelings of anger and stress as a result of these
problems, and (3) was then confronted by noncompliant, tantrum, or
"talking back" problems exhibited by her child. Because episodes of
violence appeared to be triggered by these multiple antecedents, Scott
et al. (1984) intervened with a combined therapy of assertiveness train-
ing (targeted toward the mother's relationship frustrations), child-
management skills training (to teach alternative means of handling child
behavior problems), and problem-solving training (to address life situa-
tions that were reported by the parent to create stress and generalized
anger). These three treatment elements were introduced sequentially in
a multiple-baseline design, and effects of training were examined with
measures of role-played assertiveness skill, subjective anger ratings,
problem-solving skill, and skill behavior during observed interactions
between parent and child in the clinic and home settings. Over the
course of a 20-session intervention, improvements were found on all
measures. In general, change in the various skill areas occurred at pre-
dicted points during the intervention (e.g., improved assertiveness fol-
lowed assertiveness training and improved parent-child interaction be-
havior was found contingent upon training specific to child-
278 JEFFREY A. KELLY

management skills). A 4-month follow-up revealed maintenance of

change in all areas and no evidence of continued abuse or extreme
punishment.
Most of the interventions described to this point were conducted
with parents who were physically abusive or were both abusive and
neglectful. In contrast, Dawson, de Armas, McGrath, and Kelly (1986)
have reported on the treatment of three parents who endangered their
children because of neglect rather than violent behavior. Each parent
had a history of failing to meet child care needs properly, inadequately
supervising the children, and otherwise exhibiting poor judgment in
child-care responsibilities. All were referred by a child protective ser-
vices agency. Following clinical interviews, Dawson et al. (1986) chose to
focus treatment on parent cognitive problem-solving skills in areas relat-
ed to child care .
Assessment measures in this study consisted of a set of IS problem-
solving vignettes. The situations described in the vignettes involved
themes of leaving one's child in a potentially dangerous situation; han-
dling conflicts between one's own leisure, social, or work needs and
child-care responsibilities; establishing spending priorities; and han-
dling child health emergencies. Prior to intervention, each parent re-
sponded verbally when asked how she would handle the situation. All
responses were tape recorded and later rated for elements of effective
problem-solving skill (D'Zurilla & Goldfried, 1971; Shure & Spivak ,
1972). In addition, each family's caseworker completed more global mea-
sures evaluating perceived parent skill, judgment, and quality of family
functioning.
Treatment in the Dawson et al. (1986) project consisted of problem-
solving training introduced in a multiple-baseline design across parents.
Each parent received seven to nine individual treatment sessions that
focused on developing skills for effective problem solving in situations
related to child care. Training was accomplished using modeling, shap-
ing, problem-solving practice, feedback , and homework assignments.
The examples practiced in sessions involved genuine problems that each
parent had encountered in the past and had not handled successfully. In
addition, parents received instruction in child-development principles.
Analyses of parent performance on problem-solving assessment vi-
gnettes completed at the end of each training session revealed substan-
tial increases in skill following the introduction of treatment. These
changes were well-maintained throughout the intervention and through
a IS-month follow-up . In addition, there was evidence of skill gener-
alization to problems presented in novel vignettes, improvements in
TREATING THE CHILD ABUSER 279

global caseworker ratings of family functioning, and no further reports

of child maltreatment.
In the Dawson et al. (1986) study, cognitive problem-solving skills
were taught in order to improve parents' judgment in child-care situa-
tions. A more direct environment modification approach to improve
home safety condition has also been used by several investigators. Not-
ing that poor home safety is frequently cited by investigators of child
neglect and abuse cases, Tertinger, Greene, and Lutzker (1984) and Bar-
one, Greene, and Lutzker (1986) have employed in-the-home interven-
tions to train parents in ways to protect children from health and acci-
dental injury hazards. In the Tertinger et al. (1984) project, an observa-
tional inventory was used to rate the homes of abusive and neglectful
parents for hazards, accessible to children, that could produce child
injury by poisoning, suffocation, electrical, fire, and other means. In-
home counseling, which included instruction, modeling of hazard re-
duction steps, homework assignments, and feedback, was provided to
each parent by community therapists. Subsequent observational checks
revealed substantial reductions in the number of safety hazards in each
family's home through a 7-month follow-up period. Barone et aI. (1986)
addressed the same problem with three abusive/neglectful families but
employed audio-slide show instructional materials, a home safety re-
view manual, and safety accessories rather than individual counseling in
order to make the intervention more cost-efficient. Positive changes in
home safety, again based on in-home observations of potential risks,
were found for all three families .
In most of the studies reviewed thus far, relatively focused treat-
ments have been used and evaluated with relatively small numbers of
abusive parents. A different and larger scale treatment program has been
described by Lutzker and his colleagues (Lutzker, Frame, & Rice, 1982;
Lutzker & Rice, 1987; Lutzker & Rice, in press; Lutzker, Wesch, & Rice,
1984). Termed "Project 12-Ways," the program provides intervention
services to child abusive and neglectful parents in a number of areas,
including child-management training, stress management, assertiveness
training, self-control therapy, job-finding skills assistance, alcoholism
treatment, social support development, home safety improvement, and
money management. Between 50 and 100 families receive in-the-home
services in this program each year, and the intervention components
offered to a given family are based on a clinical assessment of the family's
needs and circumstances. Child-management training is the service most
often provided, although the majority of parents receive intervention in
other areas as well (Lutzker etal., 1982). Global evaluation of this multiple
280 JEFFREY A. KELLY

component treatment approach has relied primarily on comparisons

made between families that participate in the project with families from
the same catchment area that do not, using welfare agency records of
substantiated incidents of child abuse. These data indicate that program
participation results in less frequent child abuse or neglect incidents
during the year in which intervention is offered and less frequent reports
of multiple abuse or neglect incidents in subsequent years, although
some of the recidivism data are less clearcut (Lutzker & Rice, 1984, in
press; Lutzker et al., 1982). Outcome analyses are considered preliminary
by the investigators given nonrandom assignment of families to the
project and control groups and unknown demographic equivalence of
the groups. Nonetheless, the provision of multiple kinds of intervention
to child abusive parents depending upon individual family needs is a
logical and promising approach to abuser treatment.

RESEARCH CRITIQUE
Research on the effects of treatment for child abusive parents is
difficult for a variety of reasons. Acts of parental violence typically occur
in private. Although the physical consequences of abuse-child inju-
ry-can be detected in some cases, the vast majority of instances of
parental violence are neither detected nor directly observable. Parent
reports of violent behavior are susceptible to bias, inaccuracy, and distor-
tion. Lack of candor is especially possible when a parent is under inves-
tigation or is involved in judicial processes related to child maltreatment.
For these and other reasons, research on treatment outcome with
abusive parents has relied on "probe" assessments of skill or behavioral
competence in situations which are presumed to have a functional rela-
tionship to abuse. As we have seen, the most common paradigm em-
ployed in recent studies involves "sampling" parent and child behavior
in naturalistic or staged interaction tasks, either in the home or in a clinic
setting. To the extent that these parent-child interaction tasks approxi-
mate the real situations that give rise to family conflict and abuse, they
constitute a valid assessment mode. To the extent that improvements in
parent skill observed in these interactions following treatment then gen-
eralize to in vivo (and unobserved) interactions in the home that could
actually trigger violence, the impact of therapy is also substantiated.
It now seems clear that training parents in child-management skills
does produce positive change in parent-child interactional style during
observational assessments and that change maintains over time when
TREATING THE CHILD ABUSER 281

the same assessments are repeated. How adequately the effects of this
training generalize outside formal interaction assessment observation
tasks is largely unknown because data collection has relied primarily on
performance in these tasks . Documentation that no further known
abuse was reported to authorities over a follow-up period is a positive
but, at best, imprecise indicator of intervention outcome because indi-
vidual acts of maltreatment are unlikely to be reported to authorities.
Increased attention to multimodal outcome assessment is needed in the
child abuser treatment literature. Confidence that the effects of child-
management training for abusers lessen violence and improve family
functioning would be increased if changes in parent-child interaction
skill were systematically corroborated with (1) ongoing self-monitoring
of behavior and child-related problems made by the parent and perhaps
also by the child; (2) evaluations made by significant others who regular-
ly see family members such as teachers or relatives; (3) physical, emo-
tional , and behavioral characteristics of the child; (4) measures sensitive
to family stress and functioning; or (5) parent performance during
novel, challenging parent-child interaction tasks different from those
repeatedly practiced in training. Establishing change across several such
measures following treatment could serve to validate more efficiently
and corroborate the clinical impact of an intervention.
The issue of generality versus specificity in parent training focus
may also playa role in clinical outcomes when treating child abusers.
Some abusive parents appear to exhibit generalized skill deficiencies
when interacting with their children in everyday, routine situations
(Burgess, 1979; Burgess & Conger, 1978). However, there may also be
specific and idiosyncratic child-problem situations that carry a high
probability for violence or inappropriate handling within a given family.
For example, Wolfe, Kelly, and Drabman (1981) described a case in which
spankings and beatings were administered by a parent when her chil-
dren "dawdled" excessively in the morning and at bedtime. Treatment
in the case entailed an analysis of this particular child-management
problem and the development of a parent-training intervention specifi-
cally tailored to it. The impact of child-management approaches would
appear to be greatest when intervention is relevant to those specific
conflict areas, child problems, and parent skill deficits known to affect
the family being treated (Kelly, 1983). On the other hand, an overly
narrow focus in child-management training may equip a parent to han-
dle only a few isolated problem situations, but fail to grasp underlying
principles needed to appropriately deal with other problems that were
not specifically covered in training.
282 JEFFREY A. KELLY

RESEARCH NEEDS ON MULTIFACTORIAL

TREATMENTS

Noted earlier was the observation that multifactorial conceptual

models of child abuse imply the need for multifaceted treatment inter-
ventions targeted toward parent, child, social, and environmental vari-
ables functionally related to abuse in a given family. It is clear that child-
management/parent training has received the bulk of attention in the
empirical child abuser treatment literature. This is warranted given the
now substantial evidence that abusers often lack the skills necessary to
interact appropriately with their children and handle nonviolently child
behavior problems, and also given the promising results of interven-
tions that directly train abusive parents to use new child-management
methods.
On the other hand, and as several investigators have noted (Kelly,
1983; Lutzker & Rice, 1987; Lutzker et aI. , 1982), ecological models of
abuse suggest that abusing parents often have problems in areas outside
child-management knowledge and skill alone . If this is the case, and if
these other problems are also salient functional contributors to abusive
behavior, attention to them may be needed to produce the most potent
changes in family functioning. As one example, Wahler and his col-
leagues have found that parental "insularity," or isolation from positive
interactions with friends in a parent's social network, predicts outcome
and benefit following participation in general child-management train-
ing (Wahler, 1980; Wahler, Leske, & Rogers, 1979). Because abusive par-
ents tend to be socially isolated (Helfer, 1973; Parke & Collmer, 1975),
attention to this factor may well bear on therapy outcome. However,
research in the child abuser treatment area has not directly examined
this question to date. In similar fashion, some of the studies reviewed in
this chapter incorporated stress management, anger control, or prob-
lem-solving techniques together with child-management training in
their intervention packages (d. Denicola & Sandler, 1980; Scott et aI.,
1984; Wolfe et aI. , 1981). Although these combination treatments yielded
promising results, they are for the most part single-subject or small-
sample demonstrations. Larger-scale research is needed to evaluate how
(and perhaps for which abusive parents) the incorporation of anger-
arousal control, problem-solving training, and skills training to handle
other life stressors can amplify the effects of child-management inter-
vention alone or promote more successful implementation of new child
control methods. To date, there has been little or no research on these
questions. If multifactorial conceptual models of abuse are advanced
TREATING THE CHILD ABUSER 283

and prove useful, but if treatment focuses solely on child-management

training, the impact of treatment interventions may be lessened.

TREATMENT FOR SPECIAL ABUSER POPULATIONS

As noted earlier, the populations most widely studied in the child-
abuser treatment literature are parents who behave violently toward
their children and who injure their youngsters as a result of inappropri-
ate corporal discipline. In the majority of cases, research has focused on
abuse toward young children. Very little treatment attention has been
directed to parents who physically abuse their infants or, at the other
end of the age range, their older children or adolescent children even
though surveys indicate that these forms of family violence are both
common and, often, severe (Gelles, 1978). Many studies have also found
abuse to be disproportionately prevalent among parents of children with
developmental disabilities and physical handicaps (see reviews by Am-
merman, Van Hasselt, & Hersen, 1988; Kirkham, Schinke, Schilling,
Meltzer, & Norelius, 1986), although reports of treatment for these fami-
lies are rare. The development and evaluation of treatment programs for
parents who abuse their infants, older children, or handicapped chil-
dren are needed. Although many of the principles from the general
child abuse treatment literature are no doubt relevant to these popula-
tions as well, families with infants, older children, and handicapped
children may face different problems and experience stressors that re-
quire specialized attention in therapy. Finally, there is a striking paucity
of empirical research on treatment for sexually abusive parents (see
Kolko, 1987). Over the past several years, clinical researchers have start-
ed to evaluate methods to treat the problems experienced by child vic-
tims of sexual exploitation (Alter-Reid, Gibbs, Lachnmeyer, Sigal, &
Massoth, 1986; Browne & Finkelhor, 1986; Kolko, 1987). However, re-
search on methods to intervene with parents who perpetrate sexual
abuse is very limited. Clinical research in the area of pedophilia treat-
ment may prove relevant for some sexually abusive parents.

SUMMARY
Significant strides have recently been made in the development of
treatment approaches for child abusive parents. A number of outcome
studies now demonstrate the utility of child-management, parent-train-
ing, and other family intervention approaches in altering parents' re-
284 JEFFREY A. KELLY

liance on excessively corporal, aversive, and violent child-control strat-

egies. The effects of parent-training interventions have been confirmed
by objective change in observed parent-child interactional patterns in
clinic and, more importantly, in home settings.
Multifactorial models of child abuse stress the functional interplay
between multiple parent, child, and environmental factors, and the re-
ciprocal influences between them. Some case studies and group inter-
ventions have incorporated attention to such factors by training abusive
parents in skills for anger and stress coping, problem-solving with re-
spect to child-related and other life problem situations, and handling
social and economic stressors as well as child-management skills . These
approaches have produced positive outcomes and highlight the impor-
tance of addressing what may be multiple needs, skill deficits, pres-
sures, and behavioral-social problems that affect some abusive parents.
Research on the differential effectiveness of various therapy approaches
for child abusers is rare in the literature but needed. Furthermore, be-
cause the antecedents of violent behavior may vary across families, it
will be important to develop means to tailor treatment interventions
more efficiently to specific problems that affect parent-child interactions
and the family unit as a whole.

REFERENCES
Alter-Reid, K., Gibbs, M. S., Lachnm eyer, J. R., Sigal, J., & Massoth , N. A. (1986). Sexual
abuse of children: A review of the empirical finding s. Clinical PsychologyReview, 6, 249-
266.
Ammerman, R. T., Van Hasselt, V. B., & Hersen, M. (1988). Maltreatment of handicapped
children: A critical review. Journal of Fam ily Violence, 3, 53-72.
Azar, S. T., Robinson , D. R., Hekinans, E., & Twentyman, G. T. (1984). Unrealistic expecta-
tion s and problems-solving ability in maltreating and compari son mothers. Journal of
Consulting and Clinical Psychology, 52, 687-691.
Azrin, N. H., Holz, W. C,; & Hake, D. F. (1963). Fixed ratio punishment. Journal of the
Experimental Analysis of Behavior, 6, 141-148.
Barone, V. J., Greene, B. F., & Lutzker, J. R. (1986). Home safety with families being treated
for child abuse and neglect. Behavior Modification, 10, 93-114.
Bell, G. (1973). Parents who abuse their children. Canadian Psychiatric Association Journal, 18,
223-228.
Belsky, J. (1980). Child maltreatment: An ecological integration. American Psychologist, 35,
320-335 .
Bousha , D. M., & Twentyman, C. T. (1984). Mother-child interactional style in abuse,
neglect, and control groups: Naturalistic observations in the home. Journal of Abnormal
Psychology, 93, 106-114.
Brown e, A. , & Finkelhor, D. (1986). Impact of child sexual abu se: A review of the research.
Psychological Bulletin, 99, 66-77.
TREATING THE CHILD ABUSER 285

Brunk , M., Henggeler, S. W., & Whelan, J. P. (1987). Comparison of multisystemic therapy
and parent training in the brief treatment of child abuse and neglect. Journal of Consult-
ing and Clinical Psychology, 55, 171-178.
Burgess, R. L. (1979). Child abuse: A social interactional analysis. In B. B. Lahey & A. E.
Kazdin (Eds .), Advances in clinical child psychology (Vol. 2, pp. 142-172). New York:
Plenum Press.
Burges s, R. L., & Conger, R. (1978). Family interactions in abusive, neglectful, and normal
families. Child Development, 49, 1163-1173.
Burgess, R. L., & Richardson, R. A. (1984). Coercive interpersonal contingencies as deter-
minants of child abuse: Implications for treatment and prevention. In R. F. Dangel &
R. A. Polster (Eds.), Behavioral parent training: Issues in research and practice (pp . 239-
259). New York: Guilford Press .
Crimmins, D. B., Bradlyn, A. S., St. Lawrence, J. S., & Kelly, J. A. (1984). In-clinic training
to improve the parent-child interaction skills of a neglectful mother. Child Abuse and
Neglect, 8, 533-539.
Dawson, B., de Armas, A., McGrath, M. L., & Kelly, J. A. (1986). Cognitive problem-
solving training to improve the child-care judgment of child neglectful parents. Journal
of Family Violence, 1, 209-221.
Denicola, J., & Sandler, J. (1980). Training abusive parents in child management and self-
control skills. Behavior Therapy, 11, 263-270.
Disbrow, M. A., Doerr, H ., & Caulfield, C. (1977). Measuring the components of parents'
potential for child abuse and neglect. Child Abuse and Neglect, 1, 279-296.
D'Zurilla, T. J. , & Goldfried, M. R. (1971). Problem solving and behavior modification.
Journal of Abnormal Psychology, 78, 107-126.
Eyberg , S. M., & Ross, A. W. (1978). Assessment of child behavior problems: The valida-
tion of a new inventory. Journal of Clinical Child Psychology, 7, 113-116.
Finkelhor, D., & Hotaling, G. T. (1984). Sexual abuse in the national incidence study of
child abuse and neglect: An appraisal. Child Abuse Neglect, 8, 23-33.
Frodi, A. , & Lamb, M. E. (1980). Child abusers' responses to infant smiles and cries . Child
Development, 51, 238-241.
Garbarino, J. (1976). A preliminary study of some ecological correlates of child abuse: The
impact of socioeconomic stress on mothers. Child Development, 47, 178-185.
Garbarino, J., & Sherman, D. (1980). High risk neighborhoods and high risk families: The
human ecology of child maltreatment. Child Development, 51, 188-198.
Gelles, R. J. (1978). Violence toward children in the United States. American Journal of
Orthopsychiatry, 48, 580-592.
Giovannoni, J., & Billingsley, A. (1970). Child neglect among the poor: A study of parental
inadequacy in families of three ethnic groups. Child Welfare, 49, 196-204.
Haley, J. (1976). Problem solving therapy. San Francisco: [ossey-Bass .
Helfer, R. E. (1973). The etiology of child abuse. Pediatrics, 51, 777.
Kelly, J. A. (1983). Treating child-abusive families: Intervention based on skills-training principles.
New York: Plenum Press.
Kempe , C. H. (1973). A practical approach to protection of the abused child and rehabilita-
tion of the abusing parent. Pediatrics, 51, 804-812.
Kempe, R. S., & Kempe, C. H. (1978). Childabuse. Cambridge: Harvard University Press.
Kirkham, M. A., Schinke, S. P., Schilling II, R. F., Meltzer, N. J., & Norelius, K. L. (1986).
Cognitive-behavioral skills, social supports, and child abuse potential among mothers
of handicapped children. Journal of Family Violence, 1, 235-245.
Kolko, D. J. (1987). Treatment of child sexual abuse: Programs, progress, and prospects.
Journal of Family Violence, 2, 303-318.
286 JEFFREY A. KELLY

Lahey, B. B., Conger, R. D., Atkeson, B. M., & Treiber, F. A. (1984). Parenting behavior
and emotional status of physically abusive mothers. Journal of Consulting and Clinical
Psychology, 52, 1062-1071.
Lutzker, J. R., & Rice, J. M. (1984). Project 12-ways: Measuring outcome of a large-scale in-
home service for the treatment and prevention of child abuse and neglect. ChildAbuse
and Neglect 8, 519-524.
Lutzker, J. R., & Rice, J. M. (1987). Using recidivism data to evaluate project 12-ways: An
ecobehavioral approach to the treatment and prevention of child abuse and neglect.
Journal of Family Violence, 2, 283-290.
Lutzker, J. R., & Rice, J. M. (in press) . Project 12-ways: Measuring outcome of a large in-
home service program for the treatment and prevention of child abuse and neglect .
Child Abuse and Neglect.
Lutzker, J. R., Frame, R. E., & Rice, J. M. (1982). Project 12-ways: An ecobehav ioral
approach to the treatment and prevention of child abuse and neglect. Education and
Treatment of Children.B, 141-155 .
Lutzker, J. R., Wesch, D., & Rice, J. M. (1984). A review of project 12-ways: An eco-
behavioral approach to the treatment and prevention of child abuse and neglect .
Advances in Behavior Research and Therapy, 6, 63-73 .
Mash, E. J., Johnston, C.; & Kovitz, K. (1983). A comparison of the mother-child interac-
tions of physically abused and non-abused children during play and task situations.
Journal of Clinical Child Psychology, 12, 337-346.
Mastria , E. 0 ., Mastria, M. A., & Harkins, J. C. (1979). Treatment of child abuse by
behavioral intervention: A case report. Child Welfare, 58, 253-261.
Mayhall, P. D., & Norgard, K. E. (1983). Childabuse and neglect: Sharing responsibility. New
York: Wiley.
Minuchin, S. (1974). Families and family therapy. Cambridge: Harvard University Press.
Mrazek, P. B. (1983). Sexual abuse of children. In Lahey, B. B. & Kazdin, A. E. (Eds.),
Advances in clinical child psychology (Vo!. 6, pp . 199-215). New York: Plenum Press .
Parke, R., & Collmer, M. (1975). Child abuse : An interdisciplinary analysis. In M.
Hetherington (Ed.), Review of childdevelopment research (Vol, 5, pp. 509-590). Chicago:
University of Chicago Press .
Patterson, G. R. (1977). A performance theory for coercive family interaction. In R. Cairns
(Ed.), Social interactions:Methods, analysis, and illustrations. Society for Research in Child
Development Monographs, University of Chicago, Chicago, IL.
Patterson, G. R., Ray, R., Shaw, D., & Cobb, T. (1969). A manual forcoding familyinteractions.
New York: Microfiche Publications.
Patterson, G. R., Reid, J. B., Jones, R. R., & Conger, R. E. (1975). A social learningapproach to
family intervention (Vo!. 1). Eugene, OR: Castalia .
Reid, J. B., Taplin, P. S., & Lorber, R. (1981). A social interactional approach to the treat -
ment of abusive families. In R. B. Stuart (Ed.), Violent behavior: Social learningapproaches
to prediction, management, and treatment (pp. 83-101). New York: Brunner/Maze!.
Reynolds, G. S. (1975). A primer of operant conditioning (rev. ed .). Glenview, IL: Scott,
Foresman.
Rohrbeck, C. A., & Twentyman, C. T. (1986). Multimodal assessment of impulsiveness in
abus ing, neglecting, and nonmaItreating mothers and their preschool children. Jour-
nal of Consulting and Clinical Psychology, 54, 231-236.
Sandler, J., Van Dercar, C; & Milhoan , M. (1978). Training child abusers in the use of
positive reinforcement practices . Behaviour Research and Therapy, 16, 169-175.
Schaffer, H. R., & Crook, C. K. (1979). Maternal control techniques in a directed play
situation. Child Development, 50, 989-996 .
TREATING THE CHILD ABUSER 287

Scott, W. O. N ., Baer, G., Chri stoff, K A., & Kelly, J. A. (1984). The use of skills training
pro cedures in th e treatment of a child abu sive par ent: A sys tematic case study. Journal
of Behavior Therapy and Experimental Psychiatry, 15, 329-336.
Shure, M. B., & Spivak, G. (1972). Mean s-end th inking, adjus tment, and social class
amo ng elementary school-aged children . Journal of Consulting and Clinical Psychology,
38, 348-353.
Terting er, D. A., Gre ene, B. F., & Lutzker, J. R. (1984). Home safety: Development and
validation of one component of an ecobehavioral treatment program for abused and
negl ected children. Journal of Applied Behavior Analysis, 17, 159-174.
Wahler, R. G. (1980). The ins ular mother: Her problem in parent child treatment. Journalof
Applied Behavior Analysis, 13, 207-21 9.
Wahler, R. G., Leske, G. , & Rogers, E. S. (1979). The insular family: A deviance su ppo rt
sys tem for oppositional children . In L. A. Hamerlynch (Ed .), Behavioral systems for the
developmentally disabled: I. School and family environments. New York: Brunner /Maze!.
Wolfe, D. A. (1985). Child-abusive parents: An empirical review and anal ysis. Psychological
Bulletin, 97, 462-482.
Wolfe, D. A. , Kelly, J. A., & Drabman, R. S. (1981). "Beat the buz zer": A method for
tra ining an abusive mother to decrease recurrent child conflict s. Journal of Clinical Child
Psychology, 10, 114-116.
Wolfe, D. A., Fairbank, J. A., Kelly, J. A. , & Bradlyn, A. S. (1983). Child abu sive parents:
Physiological responses to stre ssful and nonstre ssful behavior in children. Behavioral
Assessment,S, 363-371.
Wolfe, D. A., & Sandler, J. (1981). Training abu sive parents in effective child management.
Behavior Modification, S, 320-335.
Wolfe, D. A., Sandler, J., & Kaufman, K. (1981). A competen cy-based parent training
pro gram for child abu sers. Journal of Consulting and Clinical Psychology, 49, 633-640.
Wolfe, D. A., St. Lawrence, J. S., Graves, K., Brehon y, K , Bradlyn , A. S., & Kelly, J. A.
(1982). Intensive behavioral parent train ing for a child abu sive mother. Behavior Thera-
py, 13, 438-451.
 PART V

CONCLUSIONS
 CHAPTER 12

FUTURE DIRECTIONS
JAMES GARBARINO

INTRODUCTION
In this chapter we will consider future directions in understanding
factors contributing to child abuse and neglect. What does the future
hold for this important field? We may examine five such directions:
expanding definitions, increasing polarization of family experiences,
proliferating linkage between child maltreatment and other dimensions
of developmental risk, intensifying debate over the nature of communi-
ty responsibility for children, and growing importance of psychological
maltreatment as an integrating concept in the study of child abuse and
neglect.
1. Expanding definitions . Child maltreatment is a social judgment.
Thus, we create rather than discover categories of abuse and
neglect.
2. Polarization of family experiences. Socioeconomic polarization, cou-
pled with geographic segregation, predicts increasing concentra-
tion of risk for child maltreatment among low resource, high-
stress families, who are increasingly estranged from affluent and
socially connected families.
3. Proliferating linkage. As the definition of child abuse and neglect

JAMES GARBARINO • Erikson Institute for Advanced Study in Child Development, Chi-
cago, Illinois 60610.

291
292 JAMES GARBARINO

broadens, the link between child maltreatment and other devel-

opmental problems will increase. This includes linking physical
abuse to subsequent criminal aggression, and sexual abuse to
subsequent sexual assault. It also includes the linkage between
prevention of child abuse and social reform.
4. Community responsibility for children. As definitions of child mal-
treatment expand and prevention initiatives increase, debate
over the proper normal role of the community in the lives of
children will intensify. Is child rearing presumed private until
proven otherwise, or does the community have a vested right of
access for purposes of preventing child maltreatment?
5. Psychological maltreatment. As research moves forward, there is
growing recognition that psychological maltreatment is the com-
mon thread that binds together all forms of maltreatment and
accounts predominantly for developmental outcomes.

EXPANDING DEFINITIONS
Child maltreatment is not a natural fact; it is a social judgment. In
the future we must attend more to this basic principle. We do not "dis-
cover" child abuse and neglect; rather, we "create" it. Instead of being
some set of objective categories of action (as is sometimes implied by
standard definitions of "acts that harm the child"), child maltreatment is
the product of child advocacy in raising the minimal standard of care for
children. It is a social judgment that particular patterns of behavior are
sufficiently inappropriate and dangerous as to warrant community ac-
tion. Some parental treatment of children is judged to be inappropriate
but not dangerous (e.g ., letting children watch violently explicit televi-
sion or permissively indulging children). Other potential treatment is
dangerous but not thought inappropriate (e.g., playing football, or cir-
cumcision). To call something child maltreatment means that it meets
both criteria (e.g., beating a child with board, taking sexually explicit
pictures of the child, tying the child to a bed).
The key to this process is its historical dimension. It moves forward
through a series of negotiated settlements between professional exper-
tise and citizen values. We have seen this process at work actively in the
last 25 years, as we see clearly in the case of vehicular neglect.
What proportion of injuries to children as occupants of automobiles
were the result of child neglect in 1959? Virtually none: We did not have
a minimal standard of care for children in automobiles then. We do now,
and most injuries are now neglect related (because most injuries-and
FUTURE DIRECTIONS 293

90+ % of deaths) are preventable if the minimal standard (car seats) is

met. Through a process of advocacy based upon professional expertise a
minimal standard was created. We now have a standard at a given time ,
in a specific political context.
Of course, one effect of this process is to increase the amount of
child maltreatment (at least temporarily), as we raise the minimal stan-
dard and to create minimal standards in domains in which none existed
before. This applies particularly to educational neglect, ps ychological
maltreatment, and minimal physical abuse particularly (in the past two
decades, at least).
The recent Straus and Gelles (1986) national survey of two-parent
families may be seen as evidence of this phenomenon. From 1975 to
1985, Straus and Gelles found a 50% decrease in the (self-reported)
physical abuse of children (at least 3-year-olds) among two-parent fami-
lies. One interpretation of this evidence is to see in it the taking hold of
the minimal standard of care with respect to physical assault. The field
created physical child abuse in the period from 1962 to 1975 and then
began to prevent it in the period from 1975 to 1985. We may envision this
process continuing among families receptive to prevention messages
and programs (i.e ., normal families with sufficient resources to act re-
sponsively, and with sufficient prosocial relationships to motivate such
responsible actions). The reciprocal of this hopeful prognostication is,
however, a recognition of growing polarization of experience (the haves
and the have nots) with concomitant bad news for the children of so-
cioeconomic impoverishment.

POLARIZATION OF FAMILY EXPERIENCES

The mid-1960s were a time of socioeconomic optimism (and with
good reason), because the peculiar workings of the post-1950s economy
seemed to promise further extensions of affluence. Policy analysis sug-
gested that poverty could be neutralized by institutional initiative. And
indeed poverty was diminished dramatically for the elderly (who had
traditionally been disproportionately poor) . Poverty among families
with children likewise declined.
But, the 1980s have been notable for growing poverty among fami- .
lies with children-now affecting about 25 % of the young in America.
This reflects public policy-a retreat from some key subsidy programs.
But it also reflects an end to the peculiar economic conditions that gave
rise to the spreading affluence of the 1960s and 1970s (Garbarino, 1988).
The changed climate has exposed the underlying dynamics driving im-
294 JAMES GARBARINO

poverishment in the fourth quarter of the twentieth century in the

United States (including higher standards of minimal competence for
participation in the affluent economy, public policies geared to enhanc-
ing the opportunities for the already affluent, and an individualistic
ideology). All of the aforementioned facts portends growing polarization
of family experiences, with a significant proportion (up to 25%) of fami-
lies in even more deteriorated socioeconomic environments. Such an
environment provides one feature of the future for studying and dealing
with children at-risk for maltreatment.
We can see this insidious dynamic at work in many metropolitan
and rural communities, where impoverishment is already dominant for
a large segment of the population (a segment already often geograph-
ically concentrated in high-risk areas) . Growing public discussion of the
"underclass" serves to highlight this process of polarization. What is its
relevance to child maltreatment?
In principle, nearly everyone is a potential child abuser. We know
this from history and from Stanley Milgrim's laboratory studies of the
"Eichmann Effect" (in which normal people perform acts of brutality
under the impetus of situational stress and role pressure). Some few
"saints" might be incapable of attacking or abandoning a child. Never-
theless, we must acknowledge significant de facto variations in risk .
Child maltreatment becomes an ever more likely fact of life as we de-
scend the socioeconomic ladder. Among the underclass it is so wide-
spread that in most areas there is little or no attempt to employ the same
standards of care that apply elsewhere in the community. Even with a
triage system operating to screen in only the most immediately and
indubitably threatening cases, protective service agencies are over-
whelmed once they take the initiative in identifying child maltreatment
in such environments.
But this matter of incidence is not the whole story. The polarization
of family experience is occurring at a time when overall standards of
social competence for full participation are rising. Thus, anything that
diminishes academic success or employability becomes an ever more
serious liability. Child maltreatment does this, and thereby participates
in the various cycle of socioeconomic polarization. This leads to a third
issue: proliferating linkage.

PROLIFERATING LINKAGE
Over the last twenty years child maltreatment has emerged as a
central theme in efforts to understand the origins and consequences of
FUTURE DIRECTIONS 295

developmental risks. Few psychological or social pathologies are unre-

lated (at least statistically) to child maltreatment. The reported conse-
quences of child maltreatment include death, disability, delinquency,
deviant personality, learning problems, and communication disorders,
to list some of the more prominent.
Turning to the matter of causation, the origins of child maltreatment
are found in a wide range of factors: psychological deficiency, so-
ciological disability, social-psychological stress, and cultural impedi-
ments to caring for children. Thus, child maltreatment is linked to an
extremely wide range of problems. Such linkage has proliferated as
research has addressed a growing list of possible correlates and as the
process of definition has created more and more categories of child
maltreatment.
The future appears to hold more intense scrutiny of the links be-
tween child maltreatment and problems. For example, just how power-
ful is the role of child maltreatment in one generation in producing child
maltreatment in the next? As this question is refined, attention focuses
on the difficult task of specifying under what conditions intergenerational
transmission takes place. Is child maltreatment an independent contrib-
utor to school failure or is it but another consequence of the same en-
vironmental and personality factors that produce school failure?
To the degree that improved research reaffirms the commonly per-
ceived widespread linkage of child maltreatment to socioeconomic im-
poverishment then the public debate may come to focus on the issue of
whether or not we can separate child abuse prevention from social re-
form. Will programs do the job, or must we have structural reform of a
dramatic kind? Even to pose this question is to engage a fourth future
concern: the nature of community responsibility for children.

COMMUNITY RESPONSIBILITY FOR CHILDREN

At its core, child maltreatment exists as a community judgment
about the standards of care. Little wonder, then, that one of the key
issues for the future is the nature of community responsibility for chil-
dren. In the future, as in the past and present, the question is: What is
the appropriate role for the community in the lives of children? Alter-
native models exist.
Soviet ideology casts parents as subsidiary to the state; parents rear
citizens with delegated authority (delegated by the state to the parent). In
the Puritan society of New England, parents acted as agents of God,
with the Church as guarantor of God's interest in his children on earth.
296 JAMES GARBARINO

Secular, individualistic American society recognizes parents as having

natural authority over children by virtue of parenthood. The communi-
ty's role is to intervene only when parents have abused or neglected
their children.
Any other intrusion is generally unjustified (as the very word intru-
sion makes clear: the community is outside) unless it is voluntarily invit-
ed . All this has important implications for preventing child maltreat-
ment and neutralizing or ameliorating risk.
Several years ago, I visited Sweden and Germany, and then re-
turned to the United States. In Sweden, it was taken for granted that
parents would produce their children regularly for community-spon-
sored evaluations (as ,part of a general preventive orientation to chil-
dren). In Germany, such a model was thought to be an "intolerable
intrusion." However, when families move, they routinely register with
the local authorities (including children). Back home, this also was
thought to be an intolerable intrusion. What, then, is the point? Differ-
ent models (and perhaps levels) of community responsibility for chil-
dren are possible, and in fact exist. However, communities take for
granted as normal what they have, and consider weird, impossible, or
immoral what they do not.
Child abuse prevention efforts hinge upon the community's concep-
tion of its relationship to children and their parents. Various models
exist. One can define children as the property of parents and trusts that
this proprietary relationship will ensure a minimal standard of care; but
this model is largely discredited. Another defines the parent-child rela-
tionship as one of custody: parents have custody of children, with the
community's responsibility being that of guardian of last resort (in cases
of parental failure or voluntary request for services). This is the domi-
nant American model. A third approach asserts the community's vested
interest in children, and proposes some sort of shared custody of chil-
dren between parents and community.
I belie ve that the latter approach is the best context in which to
operate child abuse prevention efforts. It provides the basis for a suppor-
tive relationship that begins prenatally and that defines community con-
nections as one element of the minimal standard of care for children. As
such, it is the very foundation for child abuse prevention initiatives that
go beyond enhancement and offer realistic prospects for dealing with
the difficult dynamics posed by the concentrations of risks . It provides
the philosophic and political foundation for such measures as compre-
hensive home health visiting that begins during pregnancy and con-
tinues as appropriate over the first 2 years of a child's life (d. aids,
FUTURE DIRECTIONS 297

Henderson, Chamberlin, & Tatelbaum, 1986). It is also what is needed to

prevent psychological maltreatment.

PSYCHOLOGICAL MALTREATMENT
As the study of children at-risk matures, I believe it will turn in-
creasingly to the concept of psychological maltreatment as its unifying
theme. If we can set minimal standards of care that address directly
emotional and intellectual development, identity, and self-esteem, we
as a society will have arrived at a mature conception of the social dimen-
sion of normality. Armed with this conception, we will be able to formu-
late better policy and practice for preventing developmental risk.
A national survey revealed that roughly three quarters of American
adults believe that repeated yelling or swearing at a child leads to long-
term emotional problems for the child much of the time (National Com-
mittee for Prevention of Child Abuse, 1987). This is the cornerstone for
community action to prevent one form of psychological maltreatment.
The future could provide increasing specificity in research and program-
ming around the component concepts of psychological maltreatment:
rejection, terrorization, ignoring, isolation, and corruption (Brassard,
Germain, & Hart, 1987; Garbarino, Guttmann, & Seeley, 1986). Such a
development would advance our understanding of developmental risk
and child maltreatment.

SUMMARY
Child maltreatment has emerged as a core issue for those of us
concerned with the quality of life for children. Virtually nonexistent as a
topic of study by students of child development until the 1970s, it has
come to center stage. The future promises to continue this trend, with
growing attention to issues of social context (defin itions, polarizations of
family experience and concentration of risk , and community responsibil-
ity) and the psychological processes linking maltreatment to develop-
ment.

REFERENCES
Brassard , M., Germain, R., & Hart, S. (Eds.). (1986). Psychological maltreatment of children
and youth. New York: Pergamon Press.
298 JAMES GARBARINO

Garbarino, J. (1988). The future as if it really mattered. Longmont, CO: Bookmakers Guild .
Garbarino, J., Guttman, E., & Seeley, J. (1986). The psychologically battered child: Strategies for
identification, assessment, and intervention. San Francisco: Iosey-Bass.
National Committee for Prevention of Child Abuse (1987). Public attitudes andactions regard-
ing child abuse and its prevention. Chicago, IL: Louis Harris Public Opinion Poll.
Olds, D., Henderson, C.; Chamberlin, R., & Tatelbaum, R. (1986). Preventing child abuse
and neglect: A randomized trial of nurse home visitation. Pediatrics, 78, 65-78 .
Straus, M., & Gelles, R. (1986). Societal change and change in family violence from 1975to
1985 as revealed by two national surveys. Journal of Marriage and the Family, 48, 465-
479.
 AUTHOR INDEX

Abbott, D.A ., 235, 247 Anderson, c.. 180, 195 Barnett, PoAo, 180, 195
Aber, J.L. , 110, 112, 129, Anderson, E.S., 159, 165 Barone, N ., 29, 50
140, 141 Andreas, T., 25, 53 Barone, VoJ., 8, 17, 239,
Achenbach, T.M. , 95, 105 Aragona, J.A., 130, 131, 244, 279, 284
Acuff, D.5., 25, 53 140, 159, 164, 238, 248 Basham, RoB o, 157, 160, 166
Adams, S.L. , 235, 247 Arend, R., 121, 140 Baskiewicz, A., 211, 218
Adams, w., 29, 50 Arkowtiz, H. , 208, 220 Bassetti, M., 86, 105
Adel son, E., 152, 167 Asp , E., 57, 80, 161, 167 Bauman, MoK., 211, 217
Adler, R, 193, 195 Atkeson, B.M., 190, 196, Baumrind, D., 174, 195
Adleson, L., 230, 244 270,286 Beardslee, w.s.. 181, 194,
Adrian, c.. 182, 185, 196 Atteberry-Bennett, J., 64, 195
Ainsworth, M.D.5., 12, 78,79 Becerra, I.M., 64, 80
16, 113-115, 140, 151, Augoustinos, M., 93, 105 Becerra, R.M o, 25, 51, 111,
164, 205, 217 Aurand, j.c.. 212, 220 142
Alderson, L.R , 230, 244 Azar, S.T., 8, 19,233,234, Beckman, P., 212, 218
Allen, J.P., 129, 140 244, 245, 270, 284 Beezley, P., 214, 219
Allen, R., 211, 212, 221 Azrin, N.H ., 272, 284 Beidel ., o.c.. 179, 198
Altemeier, W.A ., 153, 154, Beilke, R, 251, 257, 266
158, 164, 170, 203, 220 Baer, D.M ., 113, 140 Bell, Go, 270, 284
Alter-Reid, K., 283, 284 Baer, G., 277, 282, 287 Bell, x.. 8, 18
Alvy, K.T., 64, 78, 100, 105 Baharal, R., 130, 133, 140 Bell, RQ o, 171, 195
American Association for Baker, L., 177, 178, 195, 196 Belsky, J., 4, 6, 13, 17, 91,
Protecting Children, 28- Bakwin, H ., 230, 244 104, 105, 109, 113, 115,
31,38,42,48-51,199,217 Balla, Do, 194, 196 118, 136, 137, 139, 140,
American Humane Asso- Ballen, Eo, 243, 246 143, 146, 151, 152, 154,
ciation, 7, 17, 28, 38, 39, Bandura, s ., 113, 140, 163,165,202,217,269,284
51 233,244 Bemporad, J. 181, 195
Amish, P.L., 89, 90, 108 Bank, L., 127, 144 Benedict, M ol., 211, 221
Ammerman, RT., 3, 5, 6, Barahal, RMo, 159, 165 Berk, RA., 60, 82
9, 11-14, 17, 97, 105, Barbieri , M.K., 71, 79 Berliner, L., 71, 73, 79,
199, 202, 207, 211-217, Barmayer, c .n.. 230, 244 260, 263, 265, 266
220, 283, 284 Barnes, K.To, 233, 234, 244 Besharov, o .j. 4, 17, 60,
Andereck, N.D., 41, 52 Barnett, D., 111, 118, 141 62, 64, 65, 79, 80

299
300 AUTHOR INDEX

Best, CL., 93, 94, 96, 108 Bromet, E.J., 157, 165 Caspi, A., 151, 155, 162,
Bickett, AD., 239, 245, Bronfenbrenner, U., 33, 166
247 51, 151, 161, 162, 165 Cassisi, J.E., 5, 9, 17, 199,
Biglan, A., 154, 168 Brooks, B., 43, 44, 46, 53 207, 216, 217
Bijou, s.w. 113, 140 Brown, C, 211, 218 Castle, R.L. , 251, 267
Billings, AG., 154, 165 Brown, R.H ., 63, 79 Catalano, R., 162, 170
Billingsley, A, 271, 285 Browne, A., 76, 77, 79, Caulfield, C, 128, 142,
Birrell, J., 213, 217 264, 266, 283, 284 270, 271, 285
Birrell, R., 213, 217 Browne, D.H., 212, 217, Ceresnie, 5., 6, 12, 19,
Bittner,S., 235, 244 234,245 159, 170, 214, 215, 220
Blackford, KA , 176, 195 Brunk, M., 275, 276, 285 Cerny, J.A., 179, 197
Bleck, L.C, 259, 262, 267 Brunnquell, D., 10, 12, 17, Chalmers, M.A., 86, 87,
Blehar, M.C, 114, 115, 154, 156, 158, 165, 166, 93, 94, 96, 107
140, 151, 164, 205, 217 204, 206, 216, 218 Chamberlin, R., 297, 298
Block, J., 155, 165 Bueno, G., 239, 247 Chandler, M., 151, 163,
Block, J.H., 155, 165 Bulkley, J., 58, 70, 79, 80 169
Blumberg, M.L., 135, 141 Burgdorff, K, 134, 141 Chapman, M., 188, 197
Blythe, B.J., 250, 251, 266 Burge, D., 182, 185, 196 Charnov, E.L., 115, 143
Boardman, H .E., 230, 244 Burger, A.M., 136, 143 Chesney, B.H., 211, 220
Bohmery S.; 156, 168 Burgess, A.W., 262-264, Child Protection and
Bonner, B.L., 15, 19, 250, 268 Treatment Act (Public
252, 255, 256, 268 Burgess, R.L., 95, 13, 17, Law 93-247), 66-68, 79
Boriskin, J.A., 4, 6, 12, 18, 105, 110, 113, 130, 131, Child Protection Report,
203,218 134, 137, 139, 141, 146, 231,245
Bourne, R., 56, 57, 79, 159, 165, 170, 202, 208, Child Sexual Abuse and
234, 239, 243, 247 213, 217, 219, 269-271 , Pornography Act, 69, 79
Bousha, D.M. , 8, 17, 128, 281, 285 Children's Defense Fund,
141, 208, 217, 272, 284 Burk, J.P., 182, 196 57,79
Bowlby, J.B., 113, 141 Burke, AE., 179, 197 Chisolm, B.A, 59, 62, 63, 79
Bradley, R.H ., 161, 165 Burkhart, B.R., 46, 51 Christensen, A., 189, 195
Bradlyn, AS., 270, 273, Bush, B.A., 23 Christoff, K.A., 277, 282,
274, 285, 287 Buttenweiser, E., 86, 87, 287
Brande~ A., 253, 267 93, 94, 96, 107 Christoffel, KK, 32, 51
Brassard, J., 160, 165 Christy, M., 255, 267
Brassard, M.R., 4, 17,87, Caffey, J., 230, 245 Cicchetti, D., 93, 105, 110-
94, 105, 106, 154, 168, Caldwell, B.M ., 161, 165 112, 116, 118, 124, 128-
297 Campbell, D.T., 99, 101, 130, 132, 137, 140-143,
Braswell, L., 255, 256, 267 103,105 145, 151, 153, 154, 156,
Braunling-McMorrow, D., Campbell, R.V., 15, 18, 163, 165, 168, 205, 211,
240, 246 239, 240, 243, 245, 246 217
Braunwald. KG., 112, Cantwell, D.P., 178, 195 Clarfield, 5., 227, 245
116, 118, 124, 141, 145 Caplan, G., 225-228, 245 Clark Foundation, 57, 79
Brehony, K , 273, 287 Carlsmith, J.M. , 57, 58, 61, Clark, D.C, 185, 196
Breitenbucher, M., 11, 17, 83,98,108 Clark, L.A ., 155, 170
160, 161, 163, 166, 201, Carlson, V., 112, 116, 118, Clark, M.e., 157, 160, 168
212,218 124, 141, 142, 145 Clasen, R.W., 29, 52
Bretherton, I., 115, 141, Carroll, J.C, 163, 165 Cobb, T., 273, 286
151, 165 Carson, B.A., 8, 19 Cochran, M., 160, 165
Breznitz, Z., 188, 195 Carstens, CC, 229-231, Cohen, J.A., 249, 251, 257,
Brody, G., 185, 186, 196 245 261,266
AUTHOR INDEX 301

Cohn, A.H., 61, 79, 99, Crook, c.x.. 276, 286 Dowdney, L., 136, 142
105, 232, 245, 253, 266 Cross, A.H ., 212, 218 Downey, G ., 151, 155, 166
Cohn, J.P., 187, 195 Cross, c.e.. 151, 162, 167 Drabman, R.5 ., 276, 281,
Colbus, D. , 8, 18 Crouter, A.e., 34, 51, 161, 287
Colletta, N .,D. , 154, 157, 165 Draper, P., 202, 208, 217
160, 166 Cumming, E., 226, 245 Drapier, P., 13, 17
Collins, D ., 127, 145 Cumming, J., 226, 245 Droegemueller, W., 4, 5,
Collmer, e. w., 4, 5, 19, Cummings, E.M ., 157, 7, 18, 55, 63, 81, 109,
89, 107, 109, 137, 144, 166, 188, 197 135, 143, 199, 219, 230,
192, 197, 200, 201, 219 Curtis, G ., 135, 142 246
Collmer, M ., 269, 270, 282, Cytryn, L., 183, 198 Drotar, D ., 211, 218
286 Dublin, c.c.. 135, 145
Commonwealth v. Adams, D'Zurilla, T.J., 278, 285 Dubow, E.P., 155, 166
73,79 Dachman, R.S ., 8, 18, 239, Dubowitz, H. , 23, 26, 27,
Conerly, S., 87, 105, 199, 245,246 51
200,219 Damon, L., 199, 200, 219, Dumas, J.E., 127, 142, 145,
Conger, R.D ., 95, 105, 260, 263, 266 194, 198
121, 130, 131, 141, 159, Daniels, }.H ., 153, 169 Duncan, D .P., 203, 213,
165, 170, 190, 196, 270, Dansky, L., 185, 196 219
271, 272, 281, 285, 286 Daro, D ., 32, 51, 99, 105 Durfee, M., 199, 200, 219
Conger, R.E., 272, 286 Davenport, Y.B., 183, 198 DUrlak , J.A ., 255, 267
Connecticut v. [arzbek, 73, Davidson, H .A. , 59, 64-
79 66,79,80 Easterbrooks, M.A. , 157,
Connell, D.B ., 116, 118, Davis, G .E., 102, 107 168
124, 144, 158, 169 Davis, H ., 194, 196 Eatman, R., 70, 80
Conte, J.R., 241, 245, 251, Dawson, B., 278, 279, 285 Eddelman, J., 240, 246
266 de Armas, A., 278, 279, Edelsohn, G ., 71, 82
Cook, T.D ., 99, 103, 105 285 Edwards, B., 238, 248
Cooper, S.P., 178, 195 DeConey, J.J., 62, 80 Egeland, B., 8-12, 17, 89,
Cornelius, D ., 251, 267 deLissovoy, Y., 6, 17, 89, 93, 105, 106, 110-113,
Comely, P.}., 157, 165 105, 200, 207, 218 115, 116, 124, 134, 136,
Corson, J., 65, 81 Denicola, J., 276, 282, 285 137, 142, 145, 154-161,
Costa, J.J., 60, 79 Diamond, L.J., 211, 214, 163, 165-167, 169,201,
Costello, A., 179, 198 216, 218, 219 204, 206, 212, 214, 216,
Coster, W., 112, 142 Dickens, B.M ., 61, 62, 66, 218
Cotterell, J.L. , 157, 166 80 Egolf , B., 88-90, 97, 106,
Coulter, M.L., 71, 82 Dietrich, KN ., 6, 12, 19, 107
Cowen, E.L. , 227, 245 158, 159, 166, 170, 214, Einbender, A.J., 5, 18, 199,
Cox, W.R., 230, 244 215,220 218, 257, 266
Coy v. Iowa 59, 75, 79 DiLalla , D .L. , 159, 166 Elder, G.H., [r., 151, 155,
Coyne, j.c.. 181, 195 Disbrow, M.A. , 128, 142, 162, 166, 169
Craig, M .E., 68, 81 270, 271, 285 Eldredge, R., 125, 126, 143
Crichton, L., 154, 165 Dishion, T.J., 127, 143, 144 Elmer, E., 4, 8, 17, 89, 91,
Crimmins, D .B., 274, 285 Doerr, H., 128, 142, 270, 93, 106, 110, 112, 114,
Crittenden, P.M ., 116, 118, 271,285 138, 142, 203, 218, 230,
124, 141, 153, 156, 158- Donovan, w., 204, 205, 245
160, 166, 206, 213, 217 218 Emde, R.N., 211, 218
Crnic, KA., 157, 160, 166 Dooley, D ., 162, 170 Emery, R., 177, 196
Crockenberg, S., 137, 142, Doran, L.D ., 128, 129, 144 Engfer, A., 151, 153, 157,
157, 163, 166 Dorr, D ., 227, 245 158, 161, 166, 167
302 AUTHOR INDEX

Eng lert, Y., 233, 245 Frazier, D., 253, 255, 266 George, c.. 125, 126, 142,
Epinosa, M.P., 156, 168 Freeman, M.L. , 25, 53 144, 156, 167
Erickson, M., 9, 10, 19, French, R. de S., 98, 108 Germain, RB., 4, 17, 87,
159, 166 Freud, A. , 58, 80, 98, 106 106, 297
Erickson, M.P. , 155, 167 Friedrich, WN., 4-6, 12, Ger sten , M., 112, 142
Ern st , E., 260, 263, 266 18, 199, 203, 218, 251, Giarretto, H., 261, 262,
Eron , L.D ., 155, 166 257,266 266
Estes, D., 115, 143 Fritz, G.S., 43-46, 51 Gibbens, T.E.N ., 135, 142
Estroff, T.W, 154, 159, 167 Frodi, A.M ., 12, 18, 118, Gibbs, M.S., 283, 284
Everson, M., 71, 82 129, 133, 142, 158, 167, Gil, D.G., 28, 39, 51, 86,
Eyberg, S.M. , 130, 131, 204, 205, 207, 218, 270, 106, 161, 162, 167, 213,
140, 159, 164, 275, 285 285 219
Fromuth, M.E ., 43-46, 51 Gilbreath, B., 154, 168
Factor, D.C. , 171 Furni ss-Tillman, T., 261, Gillespie, j.r, 194, 196
Fagot, B.l. , 132, 142 266 Gilliam , G., 24, 51, 134,
Fairbank, J.A., 270, 287 142, 156, 160, 161, 167
Falsey, S., 154, 164 Gabinet, L., 251, 266 Giovannoni, J.M., 25, 51,
Fantuzzo, J.W, 11, 15, 17, Gaboury, M.T., 57, 80 64, 80, 111, 142, 270, 285
234,245 Gaensbauer, T.J., 112, 116, Gladston, 135, 142
Fauber, R., 185, 186, 196 124, 142, 143, 156, 158, Glasgow, RE., 189, 195
Feiring, c.. 155, 169 167, 206, 218 Globe Newspaper Co . v.
Feldman, R., 156, 159, Gaines, R., 3, 18, 154, 159, Superior Court, 74
160, 169 167, 170, 201, 213, 218, Goins, c.. 15, 17
FeIner, R.D. , 194, 196 220, 251, 266 Goldberg, W.A ., 157, 168
Felton, D.K., 12, 18 Gallagh er, J.J., 212, 218 Goldfried, M.R, 278, 285
Ferrier, P.E., 236, 245 Gammon, G.D ., 184, 187, Gold stein, J., 58, SO, 98,
Feshbach, N .D ., 57, 80 198 106
Feshbach, S., 57, 80 Gan eles, D., 154, 168 Gold stein, M.J., 178, 196
Field, T.M., 154, 167, 170, Garbarin o, J., 5, 7, 12, 18, Gold ston , S. E., 100, 106
176, 197 24, 26, 33, 34, 51, 57, SO, Gold wyn, R., 153, 154, 169
Finkelhor, D., 27, 35, 43- 87, 91, 96, 106, 124, 135, Goodman, G.S., 72, 80
46, 51, 61, 77, 79, 80, 142, 154, 156, 160-163, Gordon , A.H., 116, 143
86, 88, 91, 106, 261, 264, 167, 170, 201, 218, 240, Gordon, D., 182, 185, 196
266, 271, 283-285 245, 270, 285, 291, 293, Gordon, M., 243, 246
Fischhoff, J., 6, 12, 19, 297,298 Gotlib, l. H ., 180, 181, 183,
159, 170, 214, 215, 220 Gardner, W.P., 115, 143 195, 196
Fisher, S.H. , 230, 245 Garmezy, N., 96, 106, 172, Gottfried , A.E. , 161, 168
Fleeson, J., 139, 145, 151, 196 Gottfried, A. W., 161, 168
170 Gath, A. , 211, 218 Gou ld, M., 154, 159, 167
Flicker, B.D ., 64, 80 Gavranidou, M., 157, 158, Graham, M.H ., 73, 80
Follingstad, D.R., 68, 81 161, 166 Grandjean, P., 57, 80
Fontana, V.J., 60, 62, 80, Gebhard, P.H ., 28, 43, 52 Grave s, K., 273, 287
203,218 Geca s, V., 161, 167 Gray, J., 91, 106
Forehand, R., 185, 186, Geffner, R , 88, 97, 106 Green, A.H., 3, 8, 18, 130,
196 Gelles , RJ., 4, 5, 18, 25, 143, 154, 159, 167, 170,
Forrest, D., 128, 129, 144 28, 29, 37, 39-42, 48-51, 191, 196, 201, 213, 218,
Fox, N.A. , 211, 218 53, 56, 80, 86, 88, 89, 91, 220, 251, 266
Fraiberg, S., 152, 158, 167 104, 106, 108, 134, 135, Green, B.V., 8, 17
Frame, R.E., 239, 246, 279, 156, 162, 170, 201, 212, Greenberg, M.T., 157, 160,
280, 282, 286 219, 234, 247, 283, 285 166
AUTHOR INDEX 303

Greene, B.F., 239, 241, HerrenkohI, E.C , 86, 88- Hyman, LA., 94, 107
244, 247, 279, 284, 287 90, 93, 95, 97, 106, 107,
Greenspan, 5.I., 152, 161, 128, 130-132, 143, 213, Iannotti, R.I., 190, 197
168 219 In re Freiberger, 74, 81
Gregg , G.5 ., 4, 8, 17, 203, Herrenkohl , R.C , 85, 86, Institute of Judicial Ad-
218 88-90, 93, 95, 97, 106, ministration /American
Griffin , n .c.. 207, 220 107, 128, 130-132, 143, Bar Association , 57, 61,
Gro ve, F., 121, 140 213,219 53,81
Gruber, K.J., 42, 51 Herrera, C , 154, 159, 167 Iowa Department of Social
Guttman , E., 26, 51, 87, Hersen , M., 3, 5, 6, 9, 12, Service s, 213, 219
96, 106, 297, 298 13, 17, 199, 207, 211- Irvin, N ., 211, 218
Gwartney-Gibbs , P.A., 217, 220, 283, 284 Isaacs, CD., 250, 251, 266
156, 168 Hershorn, M., 97, 107 Isabella, R., 115, 143
Hetherington, E.M ., 177,
Ha gan, R., 132, 142 194, 196 Jacobson, R.S., 128, 131-
Hagenhoff, C, 233, 245 Higgin s, J.E., 41, 53 133, 145
Hake, D.F., 272, 284 Hightower, A.D., 15, 17 [acobvitz, D., 89, 106, 110,
Hala sz, M.M ., 239, 245, Hiroto, D., 182, 185, 196 134, 136, 137, 142
247 Hodson, D. , 235, 246 Jaenicke , C , 182, 185, 196
Haley, J., 276, 285 Hoffman-Plotkin, D., 8, Jaffe, P.J., 186, 194, 198
Hall, O .K., 128, 130, 131, 18, 130, 132, 143, 156, Jameson, J.C , 116, 143
144, 209, 219 168 Janes, CL. , 184, 198
Hall , J.C, 207, 220 Hollenback, B.E., 190, 197 Jason, J., 41, 52
Hammen, C , 182, 185, Holz, W.C, 272, 284 [audes, P.K., 211, 214, 216,
196 Hop s, H ., 154, 168 218,219
Hammer, M. , 156, 159, Horowitz, R.M., 59, 64- Jen sen, P.5 ., 194, 196
160, 169 66, 80 John , K., 184, 187, 198
Hammond, M., 189, 198 Horwitz, B., 182, 196 John son, B., 138, 143, 213,
Hampton, R.L., 29, 37, 51 Hotaling, G.T., 35, 51, 86, 219
Harkin s, J.C , 272, 286 88, 91, 97, 106, 107, 271, Johnson, C , 252, 254, 267
Harpe r, L., 171, 195 285 Johnson , CF. , 29, 52
Harrold , M., 240, 243, 246 Hou se of Representatives Johnson , S.M ., 189, 195
Hart , S.N ., 4, 17, 87, 106, Report No . 685, 67, 81 Johnston , C, 131, 144,
154, 168, 297 Hov ell, M.F., 233, 245 159, 169, 270, 286
Hartman, C , 262-264, 268 Howes, C , 125, 126, 143, Joint Commission on Men-
Hartup, W.w., 177, 197 156, 168 tal Health of Children,
Haugaard, J.J., 27, 43-46, Hubinont, P.O ., 233, 245 226,246
52, 71, 72, 76, 81 Hu ebner, E.5 ., 253, 266 Jones, D., 97, 107
Hawkins, W.E., 203, 213, Huesmann, L.R., 155, 166 Jones, R.J., 42, 51
219 Hughes, H ., 88, 97, 106 Jones, R.R., 272, 286
Hazan, C , 152, 168 Hughs, J., 240, 246 [urecic , L., 15, 17
Heffron, W.M., 214, 219 Hunt, R.D ., 61, 82 Justice, B., 97, 107
Heide, J., 253, 266 Hunter, R., 134, 136, 137, Justice, R., 97, 107
Hekinans, E., 270, 284 143
Helfer, R.E., 282, 285 Hunter, W.M ., 71, 81, 82 Kadushin, A., 89, 90, 107,
Helgeson, V.S., 72, 80 Hurley, J., 154, 169 207,219
Henderson , C, 297, 298 Hutchings, B., 176, 196 Kahn, J., 181, 195
Henggeler, S.W., 275, 276, Hyatt, A., 157, 169 Kain , E.L., 162, 169
285 Hyde, J.N ., 153, 169 Kalichrnan, S.E., 68, 81
Hermalin, J., 243, 246 Hyd e, T.5 ., 179, 197 Kalmu ss, D., 156, 168
304 AUTHOR INDEX

Kamarck, T., 243, 246 Klein, D.N., 185, 196 Lewis , D.O., 194, 196
Kandel, E., 176, 196 Klein , M. , 203, 219 Lewis, M., 112, 124, 143,
Kaplan, M.G ., 158, 166 Klerman, G.L., 181, 187, 155, 169
Kaplan, 5.J., 154, 168, 251, 195, 198 Liddle, C , 136, 137, 144,
267 Knapp, R.E., 230, 247 153, 169
Kashani, J.H ., 182, 196 Knop, J., 176, 196 Lieberman, A., 121, 143,
Katz, M.H ., 161, 169 Knudsen, D.O., 29, 32, 52 152, 161, 168
Kaufman, J., 5, 18, 90, Knutsen, J.F., 136, 143 Light , R., 39, 52
107, 110, 128-130, 132, Kobak, R.R, 152, 168 Lightcap, J.L., 213, 219
136, 137, 139, 143, 153, Kohn, M.L. , 161, 168 Liker, J.K., 151, 162, 167
156, 168 Kolko, D.J., 240, 246, 271, Limber, 5., 72, 76- 78, 82
Kaufman, K.L., 15, 19, 283,285 Lipton, R, 158, 169
238, 248, 250, 252, 255, Korbin , J.E., 24, 52 Litz, J., 240, 246
256, 268, 274, 275, 282, Koss, M.P ., 43, 44, 46, 53 Liu, K., 32, 51
287 Kotelchuck, M., 153, 156, Loeber, R., 12, 18, 113,
Kazdin, A.E ., 8, 18, 255, 161, 168, 169 127, 128, 130, 144, 145,
267 Koverola, C, 238, 248 210, 216, 220, 252, 254,
Keller, M.B ., 181, 195 Kovitz, K., 131, 144, 159, 267, 270, 286
Kelley, M. , 211, 221 169, 252, 254, 267, 270, Long , F., 57, 80
Kelly, D.P., 71, 81 286 Long, N. , 185, 186, 196
Kelly, J.A. , 10, 14, 18, 269- Kramer, 5., 24, 52 Long, 5., 199, 200, 219
271, 274, 276-279, 281, Kreling, B., 227, 245 Lourie, R.5 ., 135, 145, 152,
282, 285, 287 Krieger, R, 251, 267 161, 168
Kempe, CH., 4, 5, 7, 18, Kuczynski, L., 128, 130, Lowe, A., 233, 245
55, 63, 81, 109, 135, 143, 131, 145, 159, 170, 188, Lubell, D., 260, 267
199, 219, 230, 246, 270, 197, 209, 220 Lubetsky, M.J., 13, 17,
285 Kumka, 161, 169 212, 214-217
Kemp e, R., 93, 106, 270, Kurland, J.A., 213, 219 Lucht, C, 231, 246
285 Lucht, CL. , 66, 81
Kendall, P., 255, 256, 267 Lachnmeyer, J.R., 283, 284 Lusk, R , 94, 107
Kennell, J., 203, 211, 218, LaGreca, 209, 220 Lutzk er, J.R , 8, 15, 17-19,
219 Lahey, B.B., 190, 194, 196, 200, 219, 225, 231, 239-
Kentucky v. Stincer, 75, 81 197, 270, 286 243, 244-247, 251, 266,
Kercher, G.A. , 43, 44, 46, Lamb, M.E., 112, 115, 116, 279, 280, 282, 284, 286,
52 124, 143, 158, 167, 204, 287
Ketcham, O.W., 59, 80 205, 218, 270, 285 Lutzker, 5.Z., 240, 246
Kidd , K.K., 180, 197 Lambert, R., 211, 219 Lynch, M.A. , 93, 107, 203,
Kievans, F., 243, 246 Lane, T.W., 102, 107 212, 219, 229, 247
Kilpatrick, D.G ., 71, 81 Last , J., 24, 52 Lyons-Ruth, K.I., 116, 118,
Kilstrom, J., 134, 136, 137, Lauderdale, M. , 34, 53 124, 144, 158, 169
143 Leach, C, 178, 195
Kinard, E.M. , 10, 18, 128, Leavitt, L., 204, 205, 218 Maccoby, E.E., 173, 174,
129, 143 Lee, CM., 183, 196 187, 198
King, M.P., 71, 81 Leiberman, P.H., 57, 61, MacFarlane, K., 199, 200,
Kinsey, A.C, 28, 43, 52 83 219
Kirkegaard-Sorensen, L., Leiderman, P.H ., 98, 108 MacMurray, B.K., 8, 19
176, 196 Lennon, M.C, 211, 221 Magnussen, M.G., 255, 267
Kirkham, M.A ., 283, 285 Leske, G., 282, 287 Main, M., 125, 126, 142,
Klaus, M. , 203, 211, 218, Levine, E., 253, 255, 266 144, 153, 154, 156, 167,
219 Levitt , M.J., 157, 160, 168 169
AUTHOR INDEX 305

Malkin , CM., 112, 116, Megson, D.A ., 8, 18, 239, Myers, P.A., 253, 267
124, 143 246
Maney, A. , 66, 68, 81 Mehrn, J.G., 136, 143 Nagi, S., 86, 107
Manion, C, 238, 248 Meichenbaum, D., 255, Nalepka, C , 29, 38, 52
Mannarino, A.P., 249, 251, 267 National Center for State
255, 257, 261, 266, 267 Melendez, L., 243, 246 Courts, 66, 82
Margolis, E.T., 185, 196 Melnick, B., 154, 169 National Center on Child
Marks, S., 41, 52 Melton, G.B., 55-59, 61, Abuse and Neglect, 24,
Marneffe, C, 233, 245 64-66, 68, 71, 72, 76-78, 28, 29, 32, 34-37, 39,
Marrs, S.R., 241, 248 81,82 48-50,52
Martin, B., 177, 194, 196 Meltzer, N .J., 283, 285 National Committee for
Martin CA., 214, 219 Meredith, W.H. , 235, 247 Prevention of Child
Martin, CE., 28, 43, 52 Merikangas, K.R., 184, Abuse, 69, 70, 82, 297,
Martin, H .P., 93, 95, 107, 187, 198 298
130, 133, 140, 159, 165, Messmer, M.C, 239, 247 Navarre, E.L., 94, 107
214, 219 Michelson, L., 255, 267 Neff , C, 204, 218
Martin, J., 89, 90, 107 Midonick, M.L. , 59, 61, Nelles, W.B., 179, 197
Martin, J.A., 173, 174, 187, 62,82 Nelson, B.J., 66, 82, 236,
198, 207, 219 Milhoan, M., 272, 273, 286 247
Marvell, T.B., 59, 80 Miller, S.H ., 41, 52 New York Social Services
Mash, E.J., 131, 144, 159, Miller-Perrin, CL., 241, Law Sec. 413, McKin-
169, 270, 286 248 ney, 64, 82
Mask, E., 252, 254, 267 Milner, J.S., 91, 107, 209, Newberger, E.H ., 29, 37,
Mason, R , 41, 53 219 51, 52, 56, 57, 79, 153,
Massoth, N.A., 283, 284 Minuchin, S., 177, 196, 169, 234, 235, 239, 243,
Masten, A.5 ., 172, 196 276,286 244,247
Mastria, E.O., 272, 286 Missing Children Assis - Newman, M.R ., 225, 231,
Mastria, M.A., 272, 286 tance Act of 1984, 68, 82 239, 240-243, 245, 246
Maurer, A., 201, 219 Mitchel, L., 32, 51, 52 Nezworski, T., 115, 140,
Mausner, J.5., 24, 52 Mitchum, N .T., 259, 267 152, 165
Maximus, Inc ., 32, 41, 52 Mnookin, RH., 57, 82 Nguyen, T.V., 162, 166
Mayfield, A., 183, 198 Moen, P., 33, 51, 162, 169 Norelius, K.L., 283, 285
Mayhall, P.O ., 271, 286 Moore, CK. , 214, 219 Norgard, K.E., 271, 286
McCleer, S.V., 89,107 Moos, RH., 154, 165 Nover, R , 152, 161, 168
McCombs, A., 185, 186, Morgan, S.R., 8, 12, 19,
196 213, 215, 216, 219 O'Brien, S., 239, 245
McConaughy, S.H., 95, Morris, J.L., 29, 52 O'Connor, S., 153, 154,
105 Morse, H.A., 138, 143, 158, 164, 170, 203, 220
McCord, J., 8, 19 213,219 O'Leary, D., 156, 169
McGonigle, J.J., 13, 17, Mortimer, J.T., 161, 169 O'Toole , R, 29, 38, 52
215-217 Moser, J., 8, 18 Oates, R.K., 93-95, 107,
McGrath, M.L., 278, 279, Moses, J.T., 240, 246 108, 128, 129, 144, 199,
285 Mosk , M.D., 130, 146, 219
McGuffog, C , 155, 169 252, 253, 268 Ohio v. Roberts, 73, 82
McKnew, D.H ., 183, 198 Mrazek, D., 96, 107, 136, Oldershaw, L., 128, 130,
McNeil , L.E., 94, 105 142 131, 144, 209, 219
McShane, M., 43, 44, 46, Mrazek, P. B., 96, 107, ous, D., 296, 298
52 271,286 Olsen, K., 153, 165
McWilliams, S.A., 227, 245 Myer, M.H., 261, 267 Olson, D.H ., 177, 196
Mednick, S.A., 176, 196 Myers, J.E.B., 64, 82 Olweus, D., 127, 144
306 AUTHOR INDEX

Orvaschel, H., 180, 181, Porter, F.S., 259, 262, 267 Risin , L.I., 43, 44, 46, 53
184, 197 Potter, L., 132, 142 Rizley, R , 93, 105, 1l1 ,
Otto, R.K., 55, 71, 81 Power, E., 3, 18, 201, 218, 129, 137, 141, 145, 151,
251,266 163, 165
Paisley, P.O. , 235, 247 Poythress, N.G., 56, 57, Roberts, J., 93, 107, 203, 219
Papatola, K., 89, 106, 1l0, 61, 66, 72, 78, 82 Robertson , E., 176, 197
134, 136, 137, 142 Pratt, M., 227, 245 Robins on, D.R, 270, 284
Pardek, J.A., 234, 247 Prince v. Massachusetts, Robinson, M.G. , 152, 161,
Pardek, J.T., 234, 247 56,82 168
Parental Kidnapping Pre- Prinz, R.J., 193, 198 Robinson, N.M ., 157, 160,
vention Act of 1980, 68, Provence, S., 158, 169 166
82 Prusoff, B.A., 184, 187, Roe v. Wade, 56, 82
Parke, RD., 4, 5, 19, 89, 198 Rogers, B., 127, 145
107, 109, 137, 144, 192, Rogers, c.u.. 71, 82
197, 200, 201, 219, 269, Quinton, D., 136, 137, Rogers, E.S., 282, 287
270, 282, 286 142, 144, 145, 153, 169 Rohner, E.e., 155, 169,
Patterson, G.R, 1l3, 127, 212,220
144, 209, 210, 216, 220, Rachrnan , S.J., 250, 251, Rohner, RP., 155, 169,
272, 273, 286 267 212,220
Paulsen, M., 231, 247 Radda Barnen, 24, 53 Rohrbeck, c:«, 89, 90,
Paulsen, M.G., 64, 82 Radke-Yarrow, M., 157, 108, 270, 286
Paykel, E.S., 187, 198 166, 183, 188, 197, 198 Rosario , M., 156, 159, 160,
Peacock, A., 128, 129, 144 Ragozin, A.S. , 157, 160, 169
Pelcovitz, D., 154, 168, 166 Rosen, c.. 241, 245
251,267 Raphael, B., 193, 195 Rosenbaum, A., 88, 97,
Pelton , L.H ., 29, 32, 34, Rappaport, J., 226-228, 106, 107, 156, 169
53, 91, 107, 111, 144 247 Rosenberg , A.H ., 60, 82
Pensk y, E., 136, 137, 139, Ray, R., 273, 286 Rosenberg, D., 11, 17, 160,
140 Reed, RB. , 153, 169 161, 163, 166, 201, 212,
Perri, M.G. , 8, 19 Reichler, RJ., 179, 197 218
Perry, M.A. , 128, 129, 144 Reid, J.B., 12, 18, 113, 127, Rosenberg, M.s., 14, 19,
Perry, P., 214, 219 128, 130, 144, 145, 210, 61, 82, 101, 107, 214,
Peters , S.D ., 43, 46, 47, 53 216, 220, 252, 254, 267, 220, 237, 241, 242, 247
Petrila, J., 56, 57, 61, 66, 270, 272, 286 Rosenberg, R, 176, 196
72,78,82 Reid, j.c.. 182, 196 Rosenfield-Schlichter,
Phillips, S., 189, 195 Reidy, T.J., 130, 131, 145 M.D ., 239, 247
Phipps-Yonas, S., 175, 197 Reppucci , N.D., 14, 19, Rosenthal, J.A., 31, 33, 53
Pianta, s.c.. 157, 169 27, 43-46, 52, 64, 71, 72, Rosman, B.L., 177, 196
Pierce v. Society of Sisters, 76, 79, 61, 101, 107, 214, Ross, A.W., 275, 285
56,82 220, 237, 241, 242, 247 Rovine , M., 1l5, 1l8, 140
Pines, A.M., 42, 53 Reynolds, G.S., 272, 286 Rugg , D., 233, 245
Piotrkowski, c.s.. 161, 169 Rice, J.M., 15, 19, 200, Runyan, D.K., 71, 81, 82
Platt, J.J., 255, 267 219, 239, 246, 251, 266, Rusch , RG ., 207, 220
Plotkin, s.c.. 8, 19 279, 280, 282, 286 Russell, A., 28, 30, 32, 38,
Podorefsky, D., 194, 195 Richardson, M.T., 253, 266 39,53
Poertner, J., 238, 247 Richardson, RA., 134, Russell, c.s.. 177, 196
Pokracki, D., 227, 245 141, 270, 285 Russell, D.E.H ., 28, 43-
Polansky, N., 26, 53, 86, Ricks, M.H ., 152, 169 45, 47-49, 53, 86, 107
87, 93, 94, 96, 107 Rigler, D., 4, 5, 19, 89, 90, Rutter, M., 1l0, 136, 137,
Pollock, c.. 135, 145 108, 191, 192, 197, 201, 142, 144, 145, 153, 169,
Pomeroy, W.B., 28, 43, 52 220 172, 176, 178, 197
AUTHOR INDEX 307

Sack, WH., 41, 53 Seiffer, R , 180, 197 Snell, L., 194, 196
Salzinger, S., 154, 156, Seloes, J., 135, 142 Snyder, J., 234, 239, 243,
159, 160, 168, 169 Sen ate Report No . 123 247
Sam ero ff, A.J., 151, 163, Sera fica, E C , 211, 217 Solnit, AJ., 58, 80, 98, 106
169, 180, 197 Sgro i, S.M ., 259, 261, 262, Solomo n, C R, 212, 221
Sarnit, C, 251, 267 267 Soong, W., 260, 267
Sandgrund, A., 3, 18, 159, Shaffe~ D., 154, 159, 167 Soumenkoff, G., 233, 245
170, 201, 213, 218, 220 Shaffer, H.R .,211, 220 Spearly, J.L., 34, 53
Sandler, H.M ., 153, 154, Shanok, S., 194, 196 Spinetta, J.J., 4, 5, 19, 89,
158, 164 Shapiro, V., 152, 167 90, 108, 154, 170, 191,
Sandler, J., 238, 248, 272- Shaver, P., 152, 168 192, 197, 201, 220
276, 282, 285-287 Shaw, D., 273, 286 Spivak, G., 255, 267, 278,
Sands, K., 206, 218 Shelton, P.R , 251, 267 287
Sands, S.K , 156, 158, 167 Shermack, R, 241, 245 Sprenkle, D.H ., 177, 196
Santosky v. Kramer, 56, 82 Sherman, D., 160, 162, Sroufe, L.A. , , 8- 10, 17,
Sap erstein, L., 241, 245 167, 270, 285 89, 93, 105, 110-113,
Sarber, RE., 239, 247 Sherman, L.W , 60, 82, 115, 116, 121, 124, 137,
Sceery, A., 152, 168 154, 168 139, 140-142, 145, 151,
Schachtel, D., 15, 17 Sherman, T., 188, 195 155, 159, 166, 167, 170,
Schaeffer, S., 112, 124, 143 Sherrod, K.B., 153, 154, 205,220
Schaffer, H.R , 276, 286 158, 164, 170, 203, 220 St. Lawrence, J.S., 273,
Sch akel, J.A. , 94, 95, 107 Sherry, D., 204, 218 274, 285, 287
Schau ghency, E.A., 194, Shilan sky, M., 211, 221 St. Pierre, J., 208, 221
197 Sholomska s, D., 184, 187, Stahl, J., 116, 118, 124,
Schell enbach, CJ., 134, 198 144, 158, 169
135, 142, 159, 165 Shure, M.B., 255, 267, Stamler, J., 32, 51
Schillin g II , R.E , 283, 285 278, 287 Starr, RH., Jr., 4, 6, 7, 10,
Schindler, E , 208, 220 Siegal, M., 161, 162, 170 12-14, 19, 23, 53, 87, 91,
Schinke, S.P., 283, 285 Sigal, J., 283, 284 96, 108, 158-160, 166,
Schneewind, KA , 151, Silbert, M.H., 42, 53 170, 200, 202, 213- 215,
153, 167 Silver, H .K., 4, 5, 7, 18, 220
Schn eid er, C, 91, 108 55, 63, 81, 199, 219, 230, Steele, B.E , 4, 5, 7, 18, 55,
Schneider-Rosen, K, 112, 246 63, 81, 89, 108, 109, 134,
116, 124, 142, 145 Silver, L.B., 135, 145 143, 145, 156, 170, 199,
Schore, E.L., 178, 197 Silverman , EN., 4, 5, 7, 214, 219, 220, 230, 246
Schuerman, J.R , 251, 266 18, 55, 63, 81, 109, 135, Steinberg, L., 162, 170
Schulsinger, E , 176, 196 143, 199, 219, 230, 246, Steinmetz, S.K, 37, 39,
Schultz, L., 112, 116, 124, 247 41, 42, 53, 86, 88, 89, 91,
143 Silverman, W.K , 179, 197 108, 134, 145, 156, 162,
Schultz, WJ ., 229, 231, 247 Sisson, L.A ., 212, 220 170, 234, 247
Schwartz, B.A , 175, 197 Skeen, P., 235, 246 Stem, L., 203, 219
Scott , KG., 176, 197 Sku se, D., 136, 142 Stevens, J.H ., 160, 170
Scott, R.A., 212, 220 Slobogin, C , 56, 57, 61, Stillwell, S.L., 241, 247
Scott , W.O .N ., 277, 282, 66, 72, 78, 82 Stockard, J., 156, 168
287 Slotkin, J., 185, 186, 196 Stocking, S.H. , & Associ-
Seb es, J., 134, 142 Smetan a, J., 129, 133, 142 ates, 162, 167
Sedney, M.A. , 43, 44, 46, Smets, A.C , 177, 197 Stoll, K , 43-46, 51
53 Smith, C , 98, 108 Ston e, N .W., 211, 220
Seech, M., 88, 106 Smith , J.E., 250, 251, 267 Stonebr ook, B., 251, 266
Seeley, J.W., 26, 51, 87, 96, Smith , R, 194, 196 Storer, D., 178, 195
106, 297, 298 Snedecor, S.T., 230, 247 Stovall, A. , 15, 17
308 AUTHOR INDEX

Straker, G., 128, 131-133, Tufts New England Medi- VVaterman, J., 94, 107, 130,
145 cal Center, 257, 267 133, 140, 159, 165, 199,
Straus, M.A, 28, 37, 39- Turbett, P., 29, 38, 52 200, 219, 260, 263, 266
42, 48, 49, 51, 53, 86, Turner, S.M., 179, 198 VVaters, E., 112, 114, 115,
88, 89, 91, 108, 134, 145, Twentyman, C.T., 8, 11, 140, 141, 146, 151, 164,
156, 162, 170, 212, 220, 17-19, 89, 90, 108, 128, 205, 217, 220
234, 247, 293, 298 130, 132, 141, 143, 156, VVaters, G., 128, 130, 131,
Stringer, S.A ., 209, 220 168, 208, 217, 233, 234, 144
Strock, B.D. , 184, 198 244, 245, 270, 272, 284, VVatson, D., 155, 170
Sturkie, K., 260, 263, 267 286 VVatson-Perczel, M., 15,
Sudia, C.E ., 96, 108 Tyler, c.w., [r., 41, 52 18,239,246
Sugal, D .P., 255, 267 VVebb, M.E., 8, 18, 239,
Sugarman, D., 97, 107 246
United States v. Iron Shell ,
Sussman, E.J., 128, 145, VVebe~ R.A., 157, 160, 168
74,83
159, 170, 190, 197 VVebster-Stratton, c.. 189,
Urquiza, A.J., 251, 257,
Sylvester, C.E ., 179, 197 198
266
VVeeks, D.G ., 184, 198
Urzi , T., 59, 60, 82
Talbot, N .E., 162, 170 VVeinraub, M., 157, 160,
Taplin, P.S., 113, 128, 130, 170
Valentine, D.P., 25, 53
145, 210, 22, 252, 254, VVeissman, M.M ., 180,
Van Dercar, c., 272, 273,
267, 270, 286 184, 186, 187, 197, 198
286
Tatelbaum, R, 297, 298 VVells, E.A ., 128, 129, 144
Van Hasselt, V.B., 5, 6, 9,
Taylor, D., 115, 140 wells, S., 66, 68, 81
12, 13, 17, 199, 207,
Templin, K., 253, 267 211-217, 220, 283, 284 VVelsh, R.J., 214, 219
Terrell, D., 227, 245 VVerner, E.E., 172, 198
Vaughn, B., 204, 214, 218
Tertinger, D.A., 239, 247, VVesch, D., 239, 246, 279,
Vietze, P.M ., 153, 154,
279,287 158, 164, 170, 203, 220 286
Thibaut, J., 71, 82 Von Eye, A., 115, 143
VVest, M.O ., 193, 198, 211,
Thompson, R,118, 142 Vondra, J., 7, 18, 149, 154, 219
Thompson, R.A ., 56, 58, Whatley, M.H ., 236, 237,
162, 163, 165, 167, 170
64, 65, 82, 115, 118, 142, 243,247
143 Wheeler, J.R, 265, 266
Toedter, L., 89, 95, 97, 106 VVagne~ N .N ., 43-46, 51 VVhelan, J.P., 275, 276, 285
Tong, L., 94, 108 VVahler, R.G ., 127, 142, VVhitcomb, D., 70, 72, 74,
Tonge, w.t.. 178, 195 145, 194, 198, 282, 287 78,83
Tooman, P., 29, 50 VVaisbren, S.E., 211, 221 VVhite, K.M. , 234, 239,
Toro, P.A, 93, 108 VVald, M.S ., 57, 58, 61, 83, 243,247
Trainor, C.M ., 28, 30, 32, 98,108 White, R, 211, 221
38,39,53 VValker, A, 135, 142 VVhitnig, L., 27, 53
Traylor, J., 194, 196 VValker, C.E ., 15, 19, 250, VVhittaker, J.K., 160, 170
Treiber, EA., 190, 196, 252, 255, 256, 268 Wieder, S., 152, 161, 168
270,286 VValker, L., 71, 82 Williams. D.P., 26, 53, 86,
Trickett, P.K., 128, 130, wsn. S., 114, 115, 140, 87, 93, 94, 96, 107
131, 145, 159, 170, 190, 151, 164, 205, 217 VViIIiams, G.J., 252, 268
197, 209, 220 VValsh-AIIis, G., 184, 197 Wilson, A ., 227, 245
Tronick, E.Z ., 154, 170, VValters, c .c.. 209, 219 VViIson, H .B., 230, 247
187, 195 VVarner, v.. 184, 187, 198 Wilson, S.K., 186, 194, 198
Trowell, J., 251, 267 VVarren, D.A , 212, 221 Wimberley, RC., 91, 107
Trudell, B., 236, 237, 243, VVasserman, G.A ., 211, VVisconsin v. Yoder, 56, 83
247 212,221 VVittig, B., 113, 114, 140
AUTHOR INDEX 309

Wolbert , W.A., 262-264, WurtelIe, S.K., 241, 248 Zak, L., 186, 194, 198
268 Wyatt, G.E., 43-48, 53 Zax, M., 180, 197
Wolf, B.M ., 157, 160, 170 Zigler, E., 5, 18, 90, 107,
Wolfe, D.A. , 4, 5, 10, 11, Xenak is, S.N., 194, 196 110, 136, 137, 139, 143,
13, 19, 89, 91, 93, 94, 96, 153, 162, 168, 170
100, 104, 108, 130, 146, Yanushefski, A.M ., 95, 106 Zimmerman, M.L., 262-
171, 173, 175, 183, 186, Ye, W., 184, 197 264,268
189, 192-194, 198, 199, Yllo, K., 86, 88, 91, 106 Zirpoli, T.]., 211, 221
201, 202, 207, 208, 221, Yoder, N .M., 211, 217 zen. D., 116, 118, 124,
238, 248, 252, 254, 268, Youngblade, L.M ., 109, 144, 158, 169
270, 273-276, 281, 282, 110, 113, 132, 134, 137, Zuravin, S.J., 33, 53, 243,
287 139, 141, 142, 146 248
Wolfe, V.V., 93, 94, 96, 108 Yule, B., 250, 251, 267 Zweier, D., 6, 12, 19, 159,
Wood, R.P., 255, 267 170, 214, 215, 220
Worland, J., 184, 198 Zahn-Waxler, C; 157, 166,
Wulff, L., 211, 221 183, 190, 197, 198
 SUBJECT INDEX

Abusive parents Child abuse and neglect (cont.)

intervention programs, 238-240 conceptual models (cont .)
Attribution theory, used with sex- social-cultural, 201
ually abused children , 264 social-interactional, 201, 202
American Academy of Pediatrics, 230 transitional model, 202
American Humane Association, 231 future directions
American Medical Association, 231 community responsibility for
Am erican Society for the Prevention children, 292, 295-297
of Cruelty to An imals, 229 expanding definitions, 291-293
polarization of family experi-
Battered Child Syndrome, 199, 230, ence s, 291, 293, 294
231 proliferating linkage, 291, 292,
Behavior therapy, with sexually 294,295
abused children, 265 ps ychological maltreatment, 292,
Behind Closed Doors, 234 297
Better Safe Than Sorry II film, 240 prevention, 231-283
role of stress, 234
Child Assessment Schedule, 179 Child Abuse Potential Inventory
Child abuse (CAP), 209, 210
coercive parent-child interactions, Child abuse prevention
208-211 hotlines
historical overview, 229-231 Conneticut's Care-Line, 242
multifactorial models, 270, 271, Michigan Warm Line, 242
283,284 Parents Anonymous, 242
research critique, 282, 283 Child abusive parents
Child Abuse/Mandatory Reporting research critique, 280, 281
Law, 231 treatment, 270-284
Child abuse and neglect Child Behavior Checklist, 189
con ceptual models, 200-203 Child maltreatment
psy chopathology, 200, 201 parental psychopathology, 191-193

311
312 SUBJECT INDEX

Child management training, with Goals for research, 85-108

abusive parents consequences of, 92-96
interventions, 271, 272 future policy considerations, 101-
parenting skills intervention, 272, 104
273 identification of causes, 89-92
Child physical abuse incidence/prevalence, 86-89
treatment, 251-256 prevention, 100, 101
Child Protection Report, 231 treatment, 96-100
Child sexual abuse Group therapy, with sexually abused
prevention, 236, 237, 240, 241 children, 259, 260
treatment, 256-265
educational interventions, 263, Individual psychotherapy, with sex-
264 ually abused children, 258, 259
family therapy, 261-263 International Society for Prevention
group therapy, 259, 260 of Child Abuse and Neglect,
individual psychotherapy, 258, 236
259
Children of depressed parents Joint Commission on Mental Health
behavioral problems, 182-184 and Illness, 226
cognitive and affective distur- Joint Commission on Mental Health
bances, 184-186 of Children, 226
parent/child interactions, 186-191
Maltreatment
Developmental risk factors risk factors
parental criminality, 176 early childhood, 203-208
parental immaturity, 175, 176 handicapped children, 211-215
DSM-III, 179 Massachusetts Society for the Preven-
tion of Cruelty to Children,
Educational interventions, with sex- 229,230
ually abused children, 263, 264 Minnesota Multiphasic Personality
"Eichman Effect," 294 Inventory (MMPI), 227
English Society for the Prevention of
Cruelty to Children, 229 National Committee for Prevention of
Epidemiology of maltreatment, 23-53 Child Abuse, 297
definition of, 24-28 New York Society for the Prevention
incidence and prevalence of, 28-47 of Cruelty to Children, 229,
general public surveys, 38-47 230
national surveys, 34-38
reported cases, 28-34 Pan Asian Parent Education Project,
Eyberg Child Behavior Inventory 242
(ECBI), 274, 275 Parent coping interventions, with ne-
glectful and abusive parents
Family therapy, with sexually abused treatment component, 276-280
children, 261-263 Parent training interventions, with
Family violence abusive parents
clinical features, 234, 235 group comparison studies, 274
SUBJECf INDEX 313

Parental depression, influence on Secondary prevention

child development, 180-182 child abuse/neglect, 232-234, 238,
Parental disorders 243, 244
child adjustment, 174-180 defined, 227, 228
immaturity, 175, 176 Sexual abuse
criminality, 176 prevention programs, 236, 237,
chronic illness, 176, 177 240,241
parent psychopathology, 177-180 Sexual Abuse Prevention Program for
Parental psychopathology Preschoolers (SAPP), 241
child-rearing patterns Social and emotional consequences,
influence on child development, 109-149
172-174 adolescence and adulthood, 134-
Parenting style 138
authoritative, 174 childhood, 126-133
demandingness, 173, 174 behavioral functioning, 128-133
responsiveness, 173, 174 family interaction, 127, 128
Physical abuse, seealso Child Physical definition problems, 111, 112
Abuse infancy and toddlerhood, 114,126
Prevention, see Child Abuse/Neglect, methodological issues, 112
Child Sexual Abuse theoretical issues, 113, 114
Prevention levels defined Social-interactional model, 201, 202
primary, 228, 229 Social-cultural model, 201
secondary, 227, 228 Sociological and ecological factors ,
tertiary, 226, 227 149-170
Prevention programs factors outside the family, 160-163
recommendations, 242-244 social network support, 160, 161
Primary prevention socialcultural milieu, 162, 163
child abuse/neglect, 235-236, 240- socioeconomic considerations,
244 161, 162
telephone hotlines, 242 factors within the family, 151-160
defined, 228, 229 adult psychological resources,
Primary Prevention Partnership, 242 154, 155
Project C.A.N. Prevent, 242 child characteristics, 158, 160
Project Ecosystems, 240 child-rearing history of parents,
Project Network, 242 151-154
Project 12-Ways, 239, 242, 243 marital relationship, 155, 158
Psychopathology model, 200, 201 Special abuser populations
treatment, 283
Red Flag/Green Flag People: Preven-
tion Book, 240
Research, current and future direc- Tertiary prevention
tions, 3-19 child abuse/neglect, 232, 236, 242-
definitional issues, 6-8 244
effects of maltreatment, 8-10 defined, 226, 227
identification of risk factors, 10-13 The Children's Hospital Program on
treatment and prevention, 14-16 Family Violence, 239
314 SUBJECT INDEX

The Kansas Child Abuse Prevention Trends in legislation, 55-83

Trust Fund Program, 238 legislative responses, 63-70
Transitional model, 202 Child Abuse Prevention and
Traumagenic dynamics, with sexually Treatment Act of 1974, 66-48
abused children, 264, 265 Children's Trust Funds, 69, 79
Treatment approaches federal legislation, 68, 69
child abuse issues, 249, 250 reporting statutes, 63-66
child physical abuse, 251-256 reforms in judicial procedure, 70-
review of studies, 252 78
new directions, 254-256 child testimony, 71, 72
child sexual abuse, 256-265 children's competency, 72
review of studies, 258-264 courtroom procedures, 74-76
definitional problems, 250, 251 hearsay exceptions, 73, 74
methodological problems, 251 state intervention, 56-63
new directions, 264, 265
physically abusive parents, 256
"provocative child behaviors," 255, United States (U.S.) Children's Bu-
256 reau, 231, 236