5| ‘
. ﬁnuuﬁﬁx
v. u :11... J . . . .
is .{r (2.5... . .
. . 2 3. .
. it . r» i;

vVIi) .0
51¢ 1:!
I}! 2|).

3:. i a

. . hm ukré. at. 5.

‘3: :w.) :.‘JE..J:I.!
52...»... iii
3:).

3.2.3..
3:15.:
..
ﬂuids}...- II.
.3215]: .1.
X

.41: . 1x.

a}! 1.. ,
5).. . ,
. . .
‘ . . :« :1. I ~ ‘

ax:

 

 

 

MICHIGAN STATE UNIVERSITY I RIES

l llllllll’ llll'lllllllllllllllllll
12

3 93 01417 1957

 

ll

 

 

 

 

 

 

 

 

 

THESIS "
’

This is to certify that the

thesis entitled

Understanding the Relationship Between
Child Abuse and Delinquency

presented by

Robert Keller

has been accepted towards fulfillment
of the requirements for

 

Master of Science degree in Criminal Justice

 

/

Ma' rpr fessor

DatefP"al4'-' gj

0-7639 MS U is an Afﬁrmative Action/Equal Opportunity Institution

 

_._.___~______‘ ~—-__. _..._~

-._—____ __.-__, r

 

 

LmﬁARY
Michigan State
Unlverslty

 

 

 

PLACE II RETURN BOX to remove this checkout from your record.

TO AVOID FINES roturn on or More data duo.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

utig';;{} g, 6:: :3
I
(5's? 4’ "I.
L 9 1+LZ I
AU 0 2 I

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

-L__|

 

 

 

T—l—T—l

 

MSU I. An Afﬁrmative Adlai/Equal Opponunlly Imnwon

 

UNDERSTANDING THE RELATIONSHIP BETWEEN CHILD ABUSE AND
DELINQUENCY

By

Robert Keller

A THESIS

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Criminal Justice

1995

ABSTRACT

UNDERSTANDING THE RELATIONSHIP BETWEEN CHILD ABUSE AND
DELINQUENCY

By

Robert Keller

The purposes of this research were to examine 1) the relationship between child
abuse and delinquency; 2) the relationship between the type of child abuse, either physical
or sexual, and the type of subsequent delinquency, either violent and aggressive or sexual;
and 3) the eﬁ‘ects of intervening variables such as race, age of onset, frequency of abuse,
and intensity of abuse. The survey design method was used in this study, which was
conducted at three juvenile facilities in Lower Michigan. Parental consent was obtained
for each respondent prior to obtaining verbal assent. The data collection instrument was a
questionnaire. The ﬁndings indicate that there is a relationship between child abuse and
delinquency, however, there is no statistical relationship between type of abuse and the
type of subsequent delinquency. Nevertheless, the relationship between type of abuse and
type of delinquency was aﬁ‘ected when controlling for respondents' race and the intensity
of abuse.

ACKNOWLEDGMENT

The author realizes that without the guidance, understanding, and inﬂuence ﬁ'om
many peOple, this study could not be ﬁnished. First, the author would like to recognize
Dr. Rosy Ekpenyong—Rowan for her superb leadership. Her patience and guidance are
greatly appreciated. Without her assistance, the author would not have endured the many
struggles that were faced to complete this study. She was a remarkable beneﬁt.

The author would also acknowledge his wife, Kristy. Her patience and support
provided a belief that this study could be accomplished.

Also, the author would like to thank the administrators at the three facilities.
Without their understanding of the complex nature of child abuse and the need for
additional research in this area, the study would not have been completed. Also, the
author strongly appreciates all the subjects who volunteered for participation.

Finally, the author would like to recognize the statsitical assitance that Dr. Randy
Fotiu and Tom Kueckel provided. Without the help of these two individuals, the author

could not have resolved the many statistical confusions.

iii

TABLE OF CONTENTS

LIST OF TABLES

LIST OF FIGURES

INTRODUCTION
Statement of Problem .
Purpose of Study
Signiﬁcance of Study .
Research Questions
Deﬁnitions

THEORETICAL FRAMEWORK
Social Learning Theory
Bandura's Four Learning Processes
Provoking Behavior
Maintaining Behavior .
Review of Literature .

RESEARCH METHODOLOGY
Purpose
Study Setting .

Research Questions

iv

11

12

I4

16

18

22

22

23

23

Population
Research Design
Data Collection Instrument
Variables
Independent Variables.
Dependent Variables .
Intervening Variables .
Operational Deﬁnitions
Data Collection
Statistical Procedures .
Limitations of Study .

LITERATURE REVIEW
Physical Abuse
Sexual Abuse .

DATA ANALYSIS AND PRESENTATIONS
Demographic Characteristics .
Physical Abuse

Level I Discipline
Level 11 Physical "Harm"
Sexual Abuse .

Delinquency Data

Relationship Between Child Abuse and Delinquency .

Research Question 1

24

24

24

25

25

25

25

26

27

28

28

29

29

48

54

55

61

62

69

74

77

78

79

Research Question 7 .
Research Question 7 .
Research Question 8 .
Research Question 8 .
/ Relationship Between Physical Abuse and Subsequent Delinquency .
Research Question 7 .
Research Question 8 .
Research Question 8 .
Research Question 6 .
/ Relationship Between Sexual Abuse and Delinquency
Research Question 3 .
Research Question 7 .
Research Question 7 .
Research Question 8 .
Research Question 8 .
Other Delinquent Responses .
Research Question 5 .
SUMMARY AND CONCLUSIONS .
Interpretation of Findings
Limitations of Research
Signiﬁcance of Research

BIBLIOGRAPHY

82

87

89

90

93

95

101
103
104
105
105
107
110
112
114
115
115
120
123
129
I31

132,"

vii

GLOSSARY . . . . . . . . . 135
APPENDIX A - QUESTIONNAIRE

APPENDIX B - CONSENT FORM

APPENDIX C - LETTERS FROM INSTITUTIONS AND THE UNIVERSITY

COMMITTEE ON RESEARCH INVOLVING HUMAN SUBJECTS (UCRIHS)

Table 1
Table 2
Table 3
Table 4
Table 5
Table 6

Table 7

Table 8
Table 9
Table 10
Table 11

Table 12

Table 13
Table 14
Table 15
Table 16

Table 17

LIST OF TABLES

Frequency Distribution of Respondents' Age .

Frequency Distribution of Respondents' Race.

Prior Placement History

Frequency Distribution of Respondents’ Parents' Marital Status
Respondents Who were Disciplined .

Frequency Distribution of Individual who Disciplined Respondent

Frequency Distribution of Reasons why Respondents were

Disciplined by their Parents

Frequency Distribution of Parental Methods of Discipline

Frequency Distribution of Frequency of Discipline

Received Injury as Result of Discipline

Frequency Distribution for Type of Injury Received .

Frequency Distribution of Respondents who w

”Harmed"

ere Physically

Frequency Distribution of Father as Perpetrator

Frequency Distribution of Mother as Perpetrator

Frequency Distribution of Other Perpetrators of Physical "Harm”

Frequency Distribution for Type of "Hann' for Other Perpetrators

Frequency Distribution of Respondents who were Sexually

Vi"I

viii

56

57

59

63

64

66

67

69

70

70

74

Table 18
Table 19
Table 20
Table 21
Table 22

Table 23

Table 24

Table 25

Table 26

Table 27

Table 28

Table 29

Table 30

Table 31

Table 32

Table 33

Table 34

Frequency Distribution for Perpetrator of Sexual Abuse
Frequency Distribution of Frequency of Sexual Abuse
Frequency Distribution of Crime for Current Adjudication
Frequency Distribution of Crime for Prior Adjudication
Frequency Distribution of Child Abuse Victims

Cross tabulation of Child Abuse and Type of Subsequent
Delinquency . . .

Cross tabulation of Child Abuse and Type of Subsequent
Delinquency for White Respondents .

Cross tabulation of Child Abuse and Type of Subsequent
Delinquency for Black Respondents .

Cross tabulation of Child Abuse and Type of Subsequent
Delinquency for ”Other" Race Respondents .

Cross tabulation of Age of Onset for Abused Respondents and
Type of Subsequent Delinquency . .

Cross tabulation otFrequency of Abuse and Type of Subsequent
Delinquency .

Frequency Distribution of Intensity of Child Abuse .

Cross tabulation of Subsequent Delinquency and Intensity of
Child Abuse . . . .

Cross tabulation of Physical Abuse and Type of Subsequent
Delinquency . .

Cross tabulation of Physical Abuse and Type of Subsequent
Delinquency for White Respondents . .

Cross tabulation of Physical Abuse and Type of Subsequent
Delinquency for Black Respondents . .

Cross tabulation of Physical Abuse and Type of Subsequent
Delinquency for ”Other” Race Respondents .

74

76

77

78

80

80

83

85

87

89

91

92

94

96

97

98

Table 35

Table 36

Table 37

Table 38

Table 39

Table 40

Table 41

Table 42

Table 43

Table 44

Table 45

Table 46
Table 47
Table 48

Table 49

Cross tabulation of Age of Onset of Physical Abuse and Type
of Subsequent Delinquency for Abused Respondents.

Cross tabulation of Frequency of Physical Abuse and Type
of Subsequent Delinquency for Abused Respondents.

Cross tabulation of Intensity of Physical Abuse and Type
of Subsequent Delinquency for Abused Respondents.

Frequency Distribution of Respondents who used Interventions
After Physical Abuse .

Cross tabulation of Sexual Abuse and Sexual Delinquency

Cross tabulation of Sexual Abuse and Sexual Delinquency for
White Respondents . . . .

Cross tabulation of Sexual Abuse and Sexual Delinquency for
Black Respondents . . .

Cross tabulation of Sexual Abuse and Sexual Delinquency for
”Other” Race Respondents

Cross tabulation onge of Onset for Sexual Abuse and Type
of Subsequent Delinquency for Sexually Abused Respondents

Cross tabulation of Sexual Crime and Frequency of Sexual Abuse
for Sexually Abused Respondents . . .

Cross tabulation of Sexual Delinquency and Intensity of Sexual
Abuse for Sexually Abused Respondents .

Frequency of Respondents’ Involvement with Property Crimes
Frequency of Respondents’ Involvement with Vandalism
Frequency of Respondents’ Involvement with Truancy

Frequency of Respondents’ Involvement with Substances, either
usage or sales .

100

102

103

105

106

107

108

109

111

112

114

116

117

118

119

Figure 1
Figure 2
Figure 3
Figure 4

Figure 5

LIST OF FIGURES

Bar Graph Display of Respondents' Age

Bar Graph Display of Respondents' Race

Bar Graph Display of Marital Status of Respondents' Parents
Bar Graph Display of Methods of Discipline .

Bar Graph Display of Injuries .

57

58

60

65

67

INTRODUCTION

During the last decade, juvenile crime has risen nationwide to epidemic
proportions. Crimes are being committed by an increasing number of children, and the
crimes are increasingly more serious and violent oﬁ‘enses than hitherto. According to the
FBI's Uniform Crime Report (UCR, 1992), the rate of violent juvenile crime has increased
by 27% over the last decade. Also, the arrest rate for murder among black and white
juveniles has risen by 145% and 48% respectively. During the past decade, there has been
a 79% increase in the number of juveniles who commit murders with guns (UCR, 1992).
In Michigan alone, during a three year period ﬁ'om 1988 to 1991, the juvenile arrest rate
for murder increased by 70% (Detroit Free Press, 1992). To halt this increase in crime
requires uncovering the reported determinants of this incredible increase in juvenile crime.
Is there a single contributor to this social dilemma or are the determinants a combination
of several factors?

Besides having an interest in abuse and delinquency as a research t0pic, the
researcher had past job experience with juvenile delinquents. It was during this job
experience that the researcher observed the number of residents who could recall abusive
experiences. One resident spoke about the time his mother was a prostitute and her
"pimp" beat him with a 2 by 4 to the point of his eye swelling shut. Another resident
related an incident when his father was drunk and ”whipped" him with a belt. During the
time frame that the researcher worked with juvenile delinquents, the researcher observed
that those residents who were physically abused were more abusive, either physically or

verbally. The researcher also noticed that those residents who were adjudicated for sexual

crimes had been sexually abused as a child. It was these incidents and observations that
peaked the researcher’s curiosity to study abuse and its relationship with delinquency.
Statement of the Problem

The National Council on Crime and Delinquency has identiﬁed four major
contributors to the phenomenon of “the serious juvenile oﬁ‘ender.” These include: 1) the
age of ﬁrst referral to the juvenile court, 2) the lack of commitment by youth to school and
corresponding existence of educational deﬁcits, 3) the youth's involvement in and level of
substance abuse, and 4) family disorganization including the presence of abuse and the
absence of love.

Recent statistics report a signiﬁcant increase in the incidence of reported child
abuse both within and outside the immediate family circles. Gelles (1980) used a Child
Abuse Index to study this phenomenon. He found that between 1.4 and 1.9 million
children are vulnerable to physical injury by their parents (Gelles, 1980). In 1980, The
National Center on Child Abuse and Neglect estimated that every year 250,000 children
fall victims to physical child abuse. A decade later, child abuse ﬁgures increased to the
millions. For example, the Michigan State Police (Michigan State Police, 1991) estimated
that in 1990, 2.5 million children were victims of child abuse in the United States. For the
State of Michigan, the State Department of Social Services reported that every year of
the last decade, over 5,000 children fell victim to physical injury (Michigan Department of
Social Services, 1992). With regard to sexual abuse, The National Center on Child Abuse
and Neglect (1980) estimates that the annual incidence of child sexual abuse is between
60,00 and 100,000. According to the Michigan Department of Social Services (1992),
2,553 children were victims of substantiated sexual abuse in the State of Michigan (1992).
These ﬁgures represent a signiﬁcant proportion of the adolescent population that is
exposed to either physical or sexual abuse.

There is evidence that most abuse occurs within the family environment, and that

children exposed to an abusive family environment have a higher probability of becoming

abusive towards others than those children who are not exposed to such an environment.
Theoretical and empirical evidence show that children learn behaviors vicariously through
observation or through direct experience. When abuse is present in the family
environment, children often learn that this abusive behavior is acceptable. Several theories
explain how a child learns acceptable and unacceptable behavior within the family or
quasi-family environments and other social contexts. Skinner's Operant Conditioning
theory (1953), examines how behavior is learned and shaped by consequences that
positively or negatively reinforce the learned behavior. Also, Sutherland's Differential
Association theory (Sutherland, 193 9) explains that criminal behavior is learned through
associations with individuals or groups who engage in criminal behaviors. Bandura's
Social Learning theory (Bandura, 1977) explains how behavior is acquired through
observation, direct experience, and modeling.
Purpose of Study

This study proposes a theoretical ﬁ'amework for examining the alleged relationship
between prior child abuse and subsequent delinquent activity, the nature of that
relationship, that is if the type of delinquent activity varies according to the form of child
abuse experienced by an adolescent, and if other reactions result ﬁom the prior abuse
experience. Generally, the objectives of this study are to 1) determine if a relationship
exists between prior child abuse experience and subsequent delinquency, 2) determine if a
relationship exists between the form of abuse experienced and the type of manifested
delinquent act, 3) determine if physical abuse propenses the abuse victim to violent,
aggressive crimes, and if sexual abuse inclines the victim to sexual crimes, 4) determine if
other responses to the abuse are apparent among abused children, 5) determine if there
are protective factors which may prevent an abused child from acting out through
delinquency, and 6) determine if there are factors which may affect the alleged relationship
between prior child abuse and subsequent delinquency.

Srgr_uﬁcance of Study
This study proposes to examine the relationship between prior child abuse and

subsequent delinquency. A ﬁnding that such a relationship exists can have important
policy implications especially in terms of prevention and treatment interventions for both
the abused child and the family.

A major premise of this study is that a child who experiences child abuse will learn
to become abusive. Ifthis premise is found to be true then interventions will be identiﬁed
to sever the learning of the abusive behavior and to re-direct the child's behavior into a
socially acceptable one.

The family is a major agent for socialization and if abuse is experienced within the
family there is signiﬁcant possibility that the child may not be socialized properly. The
abuse that the child experiences is a major inﬂuence on later behavior. If policy can be
developed to target the whole family for treatment this may minimize any possible future
victimization. The experienced child abuse has implications of being the cause of later
delinquency. Ifthis is substantiated then interventions will be identiﬁed to halt any ﬁxture
victimization by encompassing the whole family into treatment. By incorporating the
family into treatment the cycle of abusive behavior can be broken.

Remh Questions
The research questions for this study are as follows:

1) Is there a relationship between prior child abuse and subsequent delinquency?

2) Is there a relationship between the form of child abuse the adolescent

experienced and the type of subsequent delinquency?

3) Are those children who are physically abused more likely to commit violent,

aggressive behavior than those children who are not physically abused?

4) Are those children who are sexually abused more likely to commit sexual

crimes than those children who are not sexually abused?

5) Are there other delinquent or deviant responses to either prior physical or
sexual abuse?
6) Are there protective factors which may prevent an abused child ﬁ'om acting out
through delinquency?
7) Is the alleged relationship between prior child abuse and subsequent delinquency
aﬁ‘ected by demographic characteristics such as race and age of onset of abuse?
8) Is the alleged relationship between prior child abuse and subsequent delinquency
aﬂ‘ected by the ﬁequency and/or intensity of the abuse?
Deﬁnitions
prior physigg child abuse: the exposure to physical abuse by either observation or direct
experience including the intentional and non accidental inﬂiction of physical harm or injury
to a child through severe spanking, hitting, slapping, or beating with or without an object.
W: the exposure to sexual abuse by either observation or direct experience
which includes sexual touching, fondling, masturbation, molestation, or sexual
intercourse.
W: the adolescent's delinquent response to being a victim of

abuse including violent, aggressive crimes, sexual crimes, or other delinquent or deviant

responses.
violem, aggressive crimes: include crimes of murder, rape, assault, battery, robbery, arson,

vandalism, and malicious destruction of prOperty as deﬁned under Michigan Compiled
Laws (MCL - see Glossary).

We; include criminal sexual conduct (csc) I, H, III, IV as deﬁned under MCL.
(see Glossary).

other delingum or deviant reactions: include prostitution, running away ﬁ'om home,
truancy from school, incorrigibility, petty theft, and drug or alcohol abuse.

waging for child abuse: if the child received any professional counseling for the child

abuse.

frmbebg bf thg abuse: number of times the abuse occurred.

WM: extent of physical and psychological harm as a result of abuse.
W: how long the abuse occurred in months or years.
MM: whether the victim of the abuse was male or female.
WM: whether the perpetrator of the abuse was male or female.
14mm: biological father or mother, step mother or father, Sibling, foster or
adoptive mother or father, grandmother or father, aunt or uncle, or non-family member.
mtgmgfo—rabm: how old the child was when the abuse occurred.

gag: the race of the adolescent which includes Black, White, Hispanic, and other races.
prbtmb' mibg involvement or similar ageng: involvement of protective services or a
similar agency alter the child was victimized. V

br_iminal behavibr: any action which may be considered a crime including violent,
aggressive crimes, sexual crimes, and deviant reactions as deﬁned above.

fanﬂy; traditional or quasi-family environment, that is, any environment in which a child is
reared by an adult or parental ﬁgure, including biological, adoptive, foster, or step
families, or an institution and in which the abuse occurred.

Section Two of this research contains the theoretical framework for this study.
Section Three contains the literature review. Section Four discusses the methodology of
this research. Section Five contains the analysis of the data collected and the presentation
of the ﬁndings from the analysis of data. Section Six contains the conclusion of this study

and recommendations for further research.

THEORETICAL FRAMEWORK

Several theories have attempted to explain the determinants of delinquent or
criminal behavior. Few of these theories have focused on the alleged linkage between child
abuse and delinquency. The Social Learning theory is one of the theoretical perspectives
which appear to be relevant to an understanding of the alleged relationship. Under this
theory, most of the learning of behavior is related to dynamics within the family or similar
social groups. Within the family parents act as role models for their children. Children
learn by vicarious observation, by direct experience of behavior by parents or parental
ﬁgures, and or by direct experience of forces impacting on life such as abuse experience.
Once the learning occurs, children model the behavior observed and/or experienced.
Several theories have been developed to examine how behavior is learned vicariously and
experientially. This Study proposes to adopt the Social Learning theory as the ﬁamework
for understanding the learning and modeling of abusive behavior, and the relationship
between past child abuse experience and subsequent delinquent or deviant conduct. The
premise is that abuse begets abuse (Bandura). Accordingly, children who have
experienced abuse are very likely to internalize and re-enact abusive or other forms of
delinquent and deviant behavior. The Social Learning theory attempts to explain the chain
of events that links prior abuse experience with future delinquent or deviant conduct.

In his Operant conditioning theory Skinner (1953) examines how behavior is
shaped through reinforcement. Skinner explains that an initial act or stimulus elicits a
response or consequence. This act is reinforced by positive or negative consequences.
Negative reinforcement can occur through scolding, loss of privileges, and so forth. It

elicits a halt in that behavior. Positive reinforcement can occur through clapping, a smile,

and so forth, and will increase the likelihood that the initial action will be repeated. Also,
the strength of an action is accomplished or increased by positively reinforcing the action.
In addition, the action is strengthened by making the response very consistent and
ﬁ'equent. Therefore, where an action is strengthened through positive reinforcement, that
is, through. consistent and frequent positive consequences, it will reoccur. An action that
is negatively reinforced will cease to exist or will be eliminated. According to the operant
conditioning theory, when positive reinforcement occurs, the behavior is more likely to be
repeated than when the behavior is negatively reinforced.

In a family environment that engages in physical or sexual abuse, the child may
observe or directly experience the abuse. The child learns to associate the action, physical
or sexual abuse, with the reinforcement. If there is positive reinforcement for the parent
associated with the abuse then the child will become conditioned, learning that the abuse
is an acceptable behavior. Positive reinforcement of the abuse achieves a desired
response for the parent including power and control over the child, pleasurable feelings,
frustration reduction, and so forth. These forms of positive reinforcement cause the child
to learn to become abusive. Once the child learns that the abuse is an acceptable behavior
then the child will act out or re-enact the abuse. For example, a child who is part of a
physically abusive family learns to associate the physical abuse with the reinforcement of
the action. In this situation, physical abuse may cause the child to behave in a certain
manner. The child may perceive this power and control over behavior as rewarding for
the parent. Thus, the child learns that physical abuse is an acceptable means of controlling
behavior.

Similarly to physical abuse, a child who is part of a sexually abusive family
becomes conditioned to sexual abuse. The child learns to associate the action, sexual
abuse, with the type of reinforcement that is present. The reinforcement that is most

likely present is pleasure. The child perceives the pleasure associated with the action as

positive reinforcement. Once the sexual behavior is learned through positive
reinforcement the child will re-enact this behavior. The child becomes conditioned to
achieve sexual pleasure through this abusive behavior. Therefore, the child has learned
by the reinforcement of the sexual abuse, pleasure, that this sexual behavior is acceptable.

In his diﬁ‘erential association theory, Sutherland (193 9) explains how behavior is
acquired or learned through associations with intimate primary groups and individuals.
Predicated upon seven general principles, the theory holds that: 1) criminal behavior is a
learned behavior, 2) the process of learning criminal behavior is the same as the process of
learning lawful behavior, 3) criminal behavior is learned through a process of diﬁ‘erential
association, the primary association with people who engage in crime, 4) the learning of
criminal behavior is also determined by the ﬁequency, consistency, intensity, and duration
of contacts with criminal behavior and criminal patterns, 5) certain factors exist which
aﬂ‘ect diﬁ‘erential association or affect the consistency and frequency of criminal contact,
6) cultural conﬂict can cause differential association which causes criminal behavior, and
7) social disorganization is the basic cause of criminal behavior.

Diﬁ‘erential association theory explains that criminal or delinquent behavior is
learned in much the same fashion as the learning of confomring behavior. Whether
conforming or criminal behavior is acquired will depend on whether the primary
association occurs most ﬁequently with law-abiding or criminal groups and individuals.
When the primary association with criminal groups and individuals outweighs the
association with conforming groups and individuals then the person will acquire criminal
behavior. The process of acquiring criminal behavior through diﬁ‘erential association is
marked by four concepts: 1) priority, 2) ﬁequency, 3) intensity, and 4) duration. Priority
refers to the association of primary importance or the type of early learning exposure.
Where early learning exposure is with criminal signiﬁcant others, behavior will be

fashioned in a criminal manner. Frequency, intensity, and duration refer to the amount of

10

time and the strength of exposure to criminal groups and individuals. Thus, the earlier the
exposure to criminal groups and individuals, and the more frequent, intense, and durable
the association, the more likely the individual will learn to act out delinquent or criminal
behavior. An individual who associates on a consistent and ﬁequent basis with people
who engage in crime will learn criminal behavior and will begin to engage in similar
criminal behavior.

Sutherland's analysis of learning in intimate primary groups and the accompanying
attributes of priority, frequency, intensity, and duration, ﬁnds pertinent application to the
phenomenon of child abuse as it unfolds within the family or similar social contexts in
which the child grows. The family is one of the most signiﬁcant intimate primary groups.
What is learned within the family, and the negative quality and content of learning remain
one of the most decisive forces in the outcome of socialization, emotional, and behavioral
adjustment. An abusive family environment negatively socializes the child to learn and
accept abuse as the conventional mode of behavior. 1

From differential association theory it may be deduced that a child who is exposed
to physical abuse will learn physically abusive behavior. However, particularly signiﬁcant
in the child's ﬁrture behavior are the frequency, consistency, intensity, and duration of the
physical abuse. Frequent, persistent, severe, and long term exposure to physical abuse
within intimate primary groups is likely to be internalized as normal behavior. It is also
likely to translate itself into violent and aggressive behavior in such an abused child, who
will be more likely to learn and act out physically abusive behavior than the child who was
never physically abused. Therefore, the earlier the exposure to abuse, the more frequent,
severe, intense, and durable that exposure, the more likely the child will learn and act out
physically abusive behavior through violent and aggressive conduct.

Unlike the past two theories which explain how behavior is acquired through

reinforcement or association, Bandura's social learning theory explains how behavior is

I]

acquired through an interactional process of observation, direct experience, and modeling.
According to Bandura (1977), when an individual observes others' actions, that person
forms a concept of behavior performance. On a later occasion this individual models and

re-enacts the behavior which was learned and retained from past observation or

experience.

Soci Learnin theo

An in-depth explanation of the social learning theory must include Rotter's basic
developmental assumptions of the social learning theory. Rotter (1982) explains that the
social learning theory was originally developed "to predict and change behavior of
individuals more eﬁciently. " The theory is founded upon several assumptions including
the idea that:

1) The unit of investigation for the study of personality is the interaction of the

individual and his or her meaningful environment.

2) Not all behavior of an organism may be useﬁrlly described with personality

constructs. Behavior that may be usefully described by personality constructs

appears in organisms at a particular level or stage of complexity and a particular

level or stage of development.

3) A person's experiences (or a person's interactions with his or her meaningﬁrl

environment) inﬂuence each other.

4) Behavior as described by any personality constructs has a directional aspect. It

may be said to be goal directed. The directional aspect of behavior is inferred ﬁom

the eﬁ‘ect of reinforcing conditions.

Rotter’s basic assumptions of the social learning theory added to Bandura's
explanation of social learning theory, as applied to aggressive behavior, form the

theoretical framework for this thesis. A complex theoretical ﬁamework examining the

12

learning of behavior requires investigation of how behaviors are learned, what provokes or
instigates learned behaviors, and what maintains learned behaviors.

Behavior is deveIOped through a complex learning process which entails the
interaction with a signiﬁcant environment including signiﬁcant others. Rotter's ﬁrst
assumption of the social learning theory explains that an individual learns behavior through
the interaction with a meaningful environment. The meaningﬁrl environment may include
family, neighbors, peers, television, and so forth. Bandura reﬁned Rotter’s ﬁrst
assumption by examining how interaction with the environment occurs. According to
Bandura, the learning of behavior through the interaction with the environment occurs as
observation or direct experience. Direct experience may be deﬁned as an individual
having participation in a speciﬁc mode of behavior. Importantly, it is a learning
mechanism which is engaged during the interaction with the environment, whether the
interaction is through direct experience or observation. A learning mechanism maintains
that the behavior will be cognitively retained and used in the ﬁrture. Bandura's theory
attempts to delineate the processes or stages through which» behavior is learned, the
mechanism for maintaining or retaining it, and ﬁnally how it is subsequently provoked or
induced.

Bandura's Four Learning Prgxzesses

According to Bandura, learning of behavior occurs in four processes or stages:
input, retention, output, and reinforcement. The ﬁrst process is the attentional process or
input stage. During this stage an individual pays attention, recognizing the important
features of the observed or experienced behavior. It is those behaviors with important or
rewarding characteristics which receive more attention than unrewarding behaviors. In
addition to rewarding behaviors, an individual will pay closest attention to behavior

exhibited by individuals who possess prestige and power. After observing behavior an

13

individual will selectively code the information into symbols which then act as a guide for
subsequent behavior.

The selective input of symbolic information of observed or experienced behavior is
inﬂuenced by certain factors such as the observers' personal characteristics (i.e. age,
gender, and race), the modeled activity, and associations with other individuals. The
associations with other individuals determine what type of behaviors an individual
repeatedly observes or experiences and therefore learns most thoroughly. An individual
who is repeatedly exposed to certain behaviors and who intemalizes these behaviors as
rewarding will code this behavior as eﬁ‘ective, thus exhausting the selective coding and
input stage.

The second process is the retention stage. During this stage, the information that
was coded into symbols will be retained for ﬁrture behavior. Behavior that is eﬁ‘ective is
more likely to be retained and used in the future than non-eﬁ‘ective behavior. In addition
to eﬁ‘ective behavior, behavior that is exhibited by individuals whom the observer
perceives as having power and prestige will be selectively coded and retained for ﬁxture
use. It is the retention of symbolic information that will be used as modeling cues to guide
future behavior. Also, the repeated exposure to certain behaviors will increase the
likelihood that retention of this behavior will occur. When an individual observes
behavior, possibility exists that the behavior will be mentally rehearsed. When an
individual mentally rehearses the observed behavior it is more likely than not that the
behavior will be retained.

A successful retention process means that there is signiﬁcant probability that the
retained behavior will be modeled in the ﬁrture and reproduced. The third process of
learning through modeling is the reproduction or output stage. This stage involves
converting retained symbolic information into action. The individual models the behavior
that was once observed and mentally retained. The modeled behavior is perceived by the

14

individual as most eﬁ‘ective. Since the output into action is rarely correct the ﬁrst time,
corrective adjustments must occur to reﬁne the behavior. Modeled behavior most often is
reﬁned by informative feedback and consequences. If the initial behavior is not reﬁned it
will continue until it is reﬁned.

The ﬁnal process is the reinforcement process. The reinforcement process may
take place at several times and in different forms. An individual may complete all the past
processes: attention, retention, and reproduction, but if the behavior is negatively
reinforced the learning of behavior may rarely be engaged, and the behavior will cease to
exist. However, behavior which is positively reinforced will be put into action. As
explained earlier, a behavior that is rewarding will have more attention than unrewarding
behavior. The reinforcement of a rewarding behavior affects the attention to that
particular behavior and thus the retention and reproduction of the behavior.

Provoking Behavior

After the learning of behavior occurs and an individual has completed all four
processes there is a signiﬁcant possibility that the observed or experienced behavior will be
modeled. What requires explanation is how learned behaviors are maintained and
provoked into action After observed or experienced behavior has been mentally retained,
a stirmrlus is required to instigate behavior into action. The social learning theory divides
these stimuli into two broad classes: 1) biological based stimuli and 2) cognitively based

Biologically based stimuli include external adversive stimuli such as threats,
physical assaults, and verbal insults. These stimuli provoke an emotional response from an
individual. The individuals behavioral response will depend upon the seriousness of the
threat and the learned mode of response for c0ping with the threat. The mode of
response, as well as the retention of symbolic information, is acquired through observed or
experienced behavior. The symbolic information is then used as a predictive cue to

15

regulate and guide behavior. Therefore, these adversive stimuli provoke an emotional
response which is dependent upon predictive cues gained from past experiences.

In reality, an individual‘s emotional response is distinct from another individuals
response. This diﬁ‘erence is due to an individual's predictive cues, which are situational.
predictive cues are a combination of past learning experiences retained for ﬁrture uses.
For example, one individual when provoked may withdraw from the situation rather than
another individual who may use aggressive techniques.

The second broad class of stimuli is cognitively based. Cognitively based stimuli
are considered when an individual has the mental capacity to recognize future
consequences of a stimulus. The anticipated consequences may take several fomrs, such
as positive or negative reactions, physical pain, pleasant stimulation. Anticipated
consequences may be formed from past experiences and observations. An individual
behaves in a certain manner due to the anticipated beneﬁts that are derived ﬁ'om the
stimuli. Therefore, behavior in this sense is based upon the intrigue of anticipated beneﬁts.

The anticipated consequences of a stimulus is inﬂuenced by past experience and
observing others' behavior, including the retention of symbolic information ﬁom observing
individuals who posses power and prestige. When an individual observes or experiences
behavior, there are modeling cues which are acquired. These modeling cues are similar to
symbolic information in that they guide and regulate behavior. The modeling cues provide
an individual the possibility to imitate the behavior that was once observed or experienced.
Therefore, the modeling cues are stimuli which an individual acquires from observing or
experiencing past behavior and encourages imitation of that behavior.

There are two important inﬂuences of a cognitively based stimulus. These
inﬂuences are the anticipated consequences and the acquired modeling cues. Both
inﬂuences are gained from past experiences and observing others' behavior. When an

individual observes or experiences behavior there is a mental retention of the outcome and

16

the behavior. The cognitive component allows an individual to imitate past behavior when
a similar stimulus instigates a response. For example, a child may behave aggressively

after observing an adult behaving aggressively.

Maintaining Behavior

The third component of behavior is what maintains learned behavior. This
maintenance of behavior ultimately refers to the reinforcement of the behavior. According
to Rotter, it is the directional aspect of behavior which is inferred ﬁ'om the eﬂ‘ect of the
reinforcing conditions. The reinforcement of the behavior may be positive or negative and
it can increase, eliminate or reinstate a particular mode of behavior. The social learning
theory examines three types of reinforcement: external, vicarious, and self-produced, their
inﬂuence upon behavior.

The ﬁrst type of reinforcement is external. External reinforcement refers to the
anticipated beneﬁts of a particular behavior. These beneﬁts may include tangible rewards,
power/control, status, prestige. It is the functional value of these rewards which regulates
an individual's behavior to obtain the beneﬁts. Conversely, it is the anticipated
punishments which inﬂuence an individual to eliminate that particular mode of behavior.
Therefore, a particular behavior will be increased if it is rewarded and decreased if it is
punished.

The second form of reinforcement is vicarious or observed reinforcement.
Vicarious reinforcement affects an individual in much the same way as external
reinforcement. An individual observes behavior as well as the reinforcing consequence or
outcome of the behavior. If the observed reinforcement is a reward then the individual
will be more likely than not to imitate that behavior. Likewise, if the observed outcome is
punished, the behavior will not be imitated. Seeing other people rewarded for their

behavior ﬁmctions as a motivatOr to produce similar behavior and thus be rewarded.

17

Thus, vicarious reinforcement, especially reward, inﬂuences an individual to model the
observed behavior.

The last form of reinforcement is a self-produced mechanism. The reinforcement
is cognitively developed by an individual. An individual will behave in a manner which
gives him or her a feeling of self-satisfaction and self-worth. Conversely, behavior which
results in self-criticism will be eliminated. Often times, an individual will also respond to
their own behavior on the basis of how others have reacted. Thus, an individual judges
how others have reacted to a particular behavior, whether it is self-worth or self-criticism,
and behaves in accordance to these self-produced mechanisms.

Bandura's social learning theory ( 1977) is a powerﬁrl theoretical tool for
understanding how behavior is acquired and modeled through direct experience and
observational learning. A child learns by experiencing and observing the behavior of
signiﬁcant others. Within a family, the child perceives the parents in powerful, prestigious
positions. It is those rewarding behaviors of a parent that a child pays closest attention to
and subsequently will retain. If the parent's behavior is rewarding, such as having power
and control over the child, winning an argument with a spouse, or obtaining some self-
satisfaction, the child selectively codes this information and retains the information for
future use. Once the symbolic information is retained a stimulus is needed to provoke the
child to imitate the behavior. If this occurs, the child has converted the retained symbolic
information into action, therefore modeling the behavior. If physical abuse is present
within the family, the child has a signiﬁcant probability of imitating the physically abusive
behavior.

As applied to child abuse, Bandura's model provides a compelling explanation for
the experiential learning of abuse, the mental retention of abuse symbols and the
subsequent reproduction or reenactment of those symbols through abuse, delinquent, or

other deviant conduct. The child who directly experiences child abuse codes the

18

information into symbols and retains the symbolic information for ﬁrture modeling or re-
enactment.

Within a physically abusive family, the child may observe or directly experience
this physically abusive behavior. The child may be the victim of abuse or may observe a
parent or parental ﬁgure abusing the other parent, or a parent abusing a sibling. Regardless
of the perceived or actual purpose of the physical abuse, it is the symbolic meaning of
power, control, domination, and self-satisfaction associated with the abuse that will be
symbolically coded and mentally retained for ﬁrture reenactment. The strength of the
child's retention of the abuse will be determined by the repetition of the physical abuse.
The more the physical abuse is repeated, the more likely the child will retain it as a model
of behavior. After the child retains the concept that physical abuse is powerful and
rewarding, particular stimuli will provoke the symbolic information into action, inducing
the child to imitate the aggressive and violent behavior to obtain his or her desires.

Similar to the child within a physically abusive family, a child within a sexually
abusive family either learns from direct experience of the sexual abuse or by observing a
parent sexually abuse a sibling. Unfortunately, the child retains a concept that such sexual
behavior is an eﬂ‘ective, acceptable, rewarding sexual conduct with the undertones of
power and control. Also, the acceptance of the inappropriate sexual behavior by the child
will increase with the repetition of the sexual abuse. Once the child has retained the
information that this form of sexual conduct is acceptable, then the child will re-enact the
behavior when provoked by the appropriate stimuli, whether or not the abusive implication
is appreciated by the child. Therefore, the child is modeling the sexual behavior that the
parents or parental ﬁgures exhibited within the family environment.

Rg/jgw bf Literature
Several past studies have examined the acquisition or learning of aggressive,

physical behavior by the child and the diﬂ‘erent factors within the family, such as abuse,

19

which relate to the learning of aggressive or physical behavior. Most importantly, the
studies have shown that the interaction between adult family members and children is a
contributing factor in the acquisition of aggressive behavior. These studies include those
of: Akers (1979), Bandura (1977), Brown (1982), Cemkovich and Giordano (1987),
Fagan and Wexler (1987), Gelles (1987), Kratcoski and Kratcoski (1983),
Kruschitt(1987), Papemey and Deisher (1983), Strauss (1991). Fagan & Wexler, 1987;
Gillespie, Seaburg, & Berlin (1977); Kratcoski (1982); Kratcoski & Kratcoski (1982);
Owens & Straus (1975); Ryan (1991). Bentovin (1988), Bentovin and Boston (1988),
Browne and Finkelhor (1984), Finkelhor (1980), McDonald (1981), Mrazek (1981),
Mrazek and Mrazek (1981), Topper and Aldridge (1981).

Several scholars (Fagan & Wexler 1987; Gillespie, Seaburg, & Berlin 1977;
Kratcoski 1982; Kratcoski & Kratcoski 1982; Owens & Straus 1975; and Ryan 1991)
have attempted to explain the relationship between child abuse and subsequent
delinquency through application of the social learning theory. The social learning theory is
a powerful theoretical explanation investigating how an individual acquires behaviors
through observation or direct experience and models those acquired behaviors. Applied
speciﬁcally to child abuse, the social learning theory may be a compelling explanation that
explains how an individual learns abusive behaviors and imitates those abusive behaviors.
In eﬁ‘ect, the individual transmits from victim to victirnizer.

Fagan & Wexler (1973) in their research study the family inﬂuences of violent
delinquents. The researchers determine that the child learns violence within the home
during early adolescence. Fagan & Wexler explain that it is the family process which
establishes the reinforcement of behavioral norms. From these behavioral norms, a child
will likely model the exhibited behavior. Therefore, it is the family as a socializing

institution which inﬂuences adolescent behavior and contributes to violent delinquency.

20

Fagan & Wexler apply the social learning theory to explain that families of violent
delinquents will be characterized by high incidence of family violence and child abuse.

Gillespie et al. (1977) examine the delinquent behavior of maltreated children. The
researches examine intra-familial behavior as a vehicle for the acquisition of aggressive
behavior. Gillespie et al. explain that the frequency of occurrence will affect the
acceptance of the behavior and the likelihood that the behavior will be imitated. The
researchers also explain that the status of the person performing the maladaptive behavior
aﬁ‘ects the chances of the behavior being modeled; the higher the status of the person
performing the behavior, as perceived by the learner, the more likely modeling will take
place.

In his research, Kratcoski (1982) uses a social learning framework to explore the
roots of violent juvenile behavior and the possible relationship between child abuse and
violent behavior directed towards members of the immediate family or caretakers.
Kratcoski explains that family violence emphasizes the parent-child relationship and early
childhood experiences as linked to behavior patterns later in life. These early experiences
shape the basic personality and the interaction with the individuals own child, therefore
being transmitted to the next generation. Kratcoski explores the potential that children
who are disliked or abused by their parents may turn out to be abusers. Also, those
children who witness family members resort to aggressive behavior as problem or conﬂict
resolution mechanisms are likely to incorporate aggressive behavior into their behavior
patterns.

Kratcoski & Kratcoski (1982) explore a possible relationship of being a child abuse
victim to aggressive delinquent behavior. The researchers examine the eﬁ‘ects of early
childhood experiences and the acquisition of aggressive behavior. A child who observes
parental uses of aggressive behavior, either to solve family problems or to release

frustrations, is likely to incorporate this behavior into personal behavior patterns. Thus, an

21

abused child may have a signiﬁcant potential to incorporate aggressive behavior as a
behavior mechanism.

Owens & Straus (1975) investigate the relationship between exposure to violence
as a child to the approval of violence as an adult. The researchers explain that violence is
a learned behavior and that much of the learning takes place in childhood through the
actual experience or observation of violence. Exposure to violence in childhood has
lasting and deep seated effects on later attitudes of violence. More importantly, the
observation and experience of violence are powerful learning experiences which provide a
road map of behavior, not just attitudes. Owens & Straus explain that, "the observation of
violent behavior leads to the imitation of that violent behavior."

In her study, Ryan (1991) applies the social learning theory to the learning of
deviant sexual behaviors. She explains that exposure to deviant models may result in the
imitation of deviant behaviors. A child who is exposed to a deviant sexual model, such as
through sexual victimization, may incorporate a reinforcement of the behavior which will
lead to a pattern of deviant sexual response. Her major premise is that deviant sexual
behaviors are learned behaviors and exposure through observation or experience is

required for a child to learn deviant sexual behaviors.

RESEARCH METHODOLOGY

Juvenile violent crime rate has reached an unprecedented high nationwide (U CR,
1992). The National Council of Crime and Delinquency (1987) has identiﬁed family
disorganization, including the presence of abuse and the absence of love, as a contributing
factor to the serious juvenile oﬁ‘ender. In 1990, over 2.5 million children were abused in
the United States (Michigan State Police, 1991). There is theoretical and empirical
evidence that abuse victims develop serious behavior problems in later life. Unless
attention is directed at severing the alleged link between child abuse and subsequent
delinquent activity, the incidence of violent juvenile crime will continue to rise.

hum

This study proposes a theoretical ﬁ‘amework for examining the reported link
between prior child abuse and subsequent delinquent activity. A survey design will be
used to test the theoretical framework. The study will examine: 1) if a relationship exists
between prior child abuse and subsequent delinquency, 2) if a relationship exists between
the form of prior child abuse the adolescent experienced and the type of subsequent
delinquency, 3) if physical abuse propenses the abuse victim to violent, aggressive crimes,
4) if sexual abuse inclines the victim of abuse to sexual crimes, 5) if other deviant or
delinquent reactions to the abuse are prevalent among abused children, 6) if there are
protective factors which may prevent an abused child from acting out through
delinquency, 7) if the relationship between prior child abuse and delinquency is affected by
demographic characteristics such as gender, race, and age of onset, and 8) if the
relationship between prior child abuse and subsequent delinquency is affected by the

ﬁ'equency, and/or intensity of the abuse.

22

23

Mien—ins
This study will be conducted at three juvenile institutions in Lower Michigan. Two

of the institutions are private treatment facilities. The institutions' resident population
consist of 48 and 100 delinquent adolescent males, respectively. The third facility is a
state run detention facility that houses 40 delinquent adolescents. One private facility and
the state facility hold serious, violent oﬁ‘enders, some of whom have been adjudicated of
murder, rape and criminal assault. The second private facility holds less serious oﬁ‘enders,
who have typically been adjudicated of theft, status oﬁ‘enses, and drug crimes. All three
institutions hold adolescents whose ages range from 13 to 18 years.

Research Questions
The research questions for this research are as follows:
1) Is there a relationship between prior child abuse and subsequent delinquency?
2) Is there a relationship between the ﬁ'om of prior child abuse the adolescent experienced
and the type of subsequent delinquency?
3) Are those children who are physically abused more likely to commit violent, aggressive
crimes than those children who are not physically abused?
4) Are those children who are sexually abused more likely to commit sexual crimes than
those children who are not sexually abused?
5) Are there other delinquent or deviant responses to either prior physical or sexual abuse?
6) Are there protective factors which may prevent an abused child from acting out through
delinquency?
7) Is the alleged relationship between prior child abuse and subsequent delinquency
aﬁ‘ected by demographic characteristics such as race and age of onset of the abuse?
8) Is the alleged relationship between prior child abuse and subsequent delinquency
affected by the ﬁ'equency and/or intensity of the abuse?

24

Pbbulatibn
The total population for this research is 148 delinquent adolescents housed in the

three Lower Michigan juvenile facilities. The sampling frame for this study will be the
residents' name list at the three institutions. The entire population of 148 residents will be
included in this study. Thus, the population will not be sampled. The researcher will use
the entire institutional population because of the unsuccessﬁrl attempt to obtain a non-
institutional, non-delinquent comparison group.
Research Design

The researcher proposes to investigate the alleged relationship between prior
experience of child abuse and subsequent delinquent and deviant behavior among
adolescents. The survey design will be used in this research. According to Hagan (1989),
the survey design is most appropriate for collecting data that deals with sensitive issues
such as child abuse and delinquent or deviant conduct. Also, a survey design is
appropriate for collecting data about involvement in activities and attitudes and
perceptions about events. Thus, the survey design will be used for this study. By using a
survey design and guaranteeing conﬁdentiality and anonymity, the respondent may be
more likely to expose past involvement in abusive and delinquent situations than if a
survey design had not been used.

Data Collectibn Instrumgnt

The data collection instrument will be a questionnaire including a self-report
component. There will be Likert type questions addressing the adolescents’ prior
experience of abuse and their subsequent delinquent conduct. The researcher proposes to
examine if there is a relationship between prior abuse and subsequent delinquency, the
nature and extent of that relationship, and the effect of other intervening variables on that

relationship.

25

ngables
Independent Variable

The independent variable for this research is prior child abuse. The attributes for prior
child abuse are as follow:
Kaliﬂlﬁ Attributes
prior child abuse prior physical abuse
prior sexual abuse
Dependent Variable
The dependent variable for this research is subsequent delinquent activity. The attributes

of the dependent variable include violent, aggressive crimes, sexual crimes, and other

delinquent reactions.
Mg Attributes
subsequent delinquent activity violent aggressive crimes
sexual crimes
other delinquent reactions
Intervening Variables

Intervening variables are factors which may have an effect upon the alleged relationship
between prior child abuse and subsequent delinquency. These intervening variables
include ﬁ'equency, intensity of the abuse, age of onset for abuse, race, and intervention by

protective services or similar agency.

Variable
age of onset for abuse ﬁ'equency of the abuse
race intensity of the abuse

intervention of protective services or similar agency

26

Qpemb'onal Deﬁnitibns
pribr pmcal bhild Quse: the exposure to physical abuse by either observation or direct
experience including the intentional and non accidental inﬂiction of physical harm or injury
to a child through severe spanking, hitting, slapping, or beating with or without an object.
W: the exposure to sexual abuse by either observation or direct experience
which includes sexual touching, fondling, masturbation, molestation, or sexual intercourse.
W: the adolescent's delinquent response to being a victim of
abuse including violent, aggressive crimes, sexual crimes, or other delinquent or deviant
reactions.
viblgrt, agggessive crimes: include crimes of murder, rape, assault, battery, robbery, arson,
vandalism, and malicious destruction of property as deﬁned under Michigan Compiled
Laws (MCL - see Glossary)
We; include criminal sexual conduct (csc) I, II, III, IV as deﬁned under MCL
(see Glossary). ’
bthg dglinguent br deviant reactions: include prostitution, running away ﬁ'om home,
truancy from school, incorrigibility, petty theft, and drug or alcohol abuse.
W: if the child received any professional counseling for the child
abuse.
Wm: number of times the abuse occurred.
W: extent of physical and psychological harm as a result of abuse.
mm: how long the abuse occurred in months.
W: whether the victim of the abuse was male or female.
mm: whether the perpetrator of the abuse was male or female.
W1: biological father or mother, step mother or father, sibling, foster or
adoptive mother or father, grandmother or father, aunt or uncle, or non-family member.
WM: how old the child was when the abuse occurred.

bag: the race of the adolescent which includes Black, White, Hispanic, and other races.

27

jammy; grvibss invblvsment 9r similar agengg: involvement of protective services or a
similar agency after the child was victimized.
W: any action which may be considered a crime including violent,
aggressive crimes, sexual crimes, and deviant reactions as deﬁned above.
My; traditional or quasi-family environment that is any environment in which a child is
reared by an adult or parental ﬁgure including biological, adoptive, foster, or step families
or an institution and in which the abuse occurred.

Data Collection

The survey design will require the use of a self report questionnaire. The
researcher will personally administer the questionnaire to participating adolescents at two
of the facilities. Since the researcher is an employee at the third facility, the researcher will
elicit the assistance of a graduate student, not aﬂiliated with the facility, to administer the
questionnaire. This student will be responsible for assigning ID code numbers to
consenting residents at the facility where the researcher is an employee. After ID code
numbers have been assigned, the resident name list will be destroyed. This graduate
student has no aﬁliation or association with any of the three facilities, with the residents,
or with the parents.

Because of the respondents’ age and social circumstances, parental consent will be
obtained prior to distributing the questionnaire to the adolescents. A copy of the parental
consent form is attached in Appendix B. Only those adolescents whose parents return a
signed consent form will participate in the study. On the introductory statement to the
questionnaire is a sentence which explains that voluntary participation is indicated by
completing and returning the questionnaire. Also, this writer guarantees that this study
will be conducted in strict observance of conﬁdentiality and anonymity. To assure
anonymity the respondents will be instructed to omit their name or any identifying marks
on the questionnaire. To assure conﬁdentiality, respondents at each institution will

complete the questionnaire at a designated time and place within the facility without the

28

presence or assistance of institutional employees. Respondents will also be instructed that
participation is voluntary and that withdrawal is permitted at any time during the
administration of the questionnaire. Also, the respondent will be allowed to leave blank
any question that he does not want to answer. The respondent will be instructed to place
the completed questionnaire in an unmarked, manila envelope distributed with the
questionnaire, close the envelope, and place it into the designated box in the room. The
session will last 40 - 45 minutes. The researcher will be the only person to see the data
and the data will only be used for the purpose of this research.
Statistical Procedures

Statistical procedures for this research will attempt to answer the research
questions and ultimately determine if there is a relationship between prior child abuse and
subsequent delinquency. Statistical procedures will determine the nature of the
relationship between variables and the extent of that relationship. Statistical procedures
for this study include Chi-square, correlation coeﬁicients, and measures of associations.

Limitatibns of Study

Limitations for this study include the lack of random sampling, the inexistence of a
comparison group, and the use of a self-report questionnaire. The lack of random
sampling does not allow for the use of inferential statistics. It limits the researcher to only
answering the research questions. Due to the unsuccessful attempt to obtain a non-
institutional or non-delinquent population there is an inexistence of a comparison group.
The use of a non-delinquent comparison group would have allowed for comparisons to be
statistically computed between the institutional and non-institutional groups. Also, the use
of a self-report questionnaire is a limitation in this study. The use of a self-report
questionnaire allows the respondent to be potentially inaccurate or deceitful when

answering the questionnaire.

LITERATURE REVIEW

Several researchers (Alfaro 1981, Benoit & Kennedy 1992, Brannon, Larson, &
Doggett 1989, Fagan & Wexler 1987, Gillespie, Seaburg, & Berlin 1977, Kratcoski 1982,
Kratcoski and Kratcoski 1982, Longo 1981, McCord 1983, Owen & Straus 1975, Pfouts,
Schopler, & Henley 1981, & Widom 1989) have investigated the alleged relationship
between child abuse and delinquency. Some researchers (Alfaro 1981, McCord 1983,
Widom 1989) have focused on physical abuse and neglect, other researchers (Fagan &
Wexler 1987, Gillespie et al. 1977, Kratcoski 1982, Owen & Straus 1975, & Pfouts et al.
1981) have focused on the physical abuse present within the family. Additionally, some
researchers (Brannon et al. 1989, Benoit & Kennedy 1992, & Longo 1981) have
investigated the eﬁ‘ects of sexual abuse on delinquency. These researchers have
consistently found this relationship to exist; however, they have cautioned that the
relationship is not direct, but is mediated by the inﬂuence of other factors. Each
researcher has oﬁ‘ered differing explanations for the existence of the relationship between
child abuse and delinquency. This review shall ﬁrst concentrate on those studies which
investigate the eﬁ‘ects of physical abuse on delinquency and those factors which inﬂuence
the relationship. Secondly, this review shall examine those studies which investigate the
eﬁ'ects of sexual abuse on delinquency and those factors which mediate the alleged
relationship.

Physical Abbse

Alfaro (1981) conducted a 20 year longitudinal study to examine the possible
linkage between child abuse and later subsequent delinquent behavior. Alfaro proposed to
show that abused children have a greater likelihood of becoming delinquent than non-

29

30

abused children. The researcher collected ofﬁcial records from child protective service
agencies and courts ﬁom the 1950's to determine the presence of abuse within a family.
These statistics were compared and analyzed with court records ﬁ'om the 1970's that
would indicate delinquent behavior or ungovemability exhibited by the adolescent. These
two samples were then analyzed to determine the likelihood of delinquent behavior by an
abused child.

Alfaro obtained data for this study by examining the case ﬁles ﬁ'om child protective
service agencies and courts ﬁ'om the years 1952-53. The researcher was able to obtain
information about 5,136 children and 1,423 families that were reported for suspected child
abuse or neglect. Information was collected about the suspected abused child, the
siblings, and the family history. These children that were indicated as possible abuse
victims were traced through the records of the Family Court or the Children's Court for
subsequent delinquency. From the 5,136 children that were suspected as abused, 3,705 or
72% had involvement with the court for delinquent behavior or ungovemable behavior.

A second set of information about adolescents was obtained ﬁom the court
records or Probation Intake Unit during 1971-72. These records provided information
about delinquent histories of 1,936 adolescents and 1,851 families. The histories of these
adolescents and families were traced backwards to determine any prior involvement in
child abuse or neglect cases. Therefore, Alfaro's research traced the direction of an
abused child after contact with the child protective service unit as well as retroactively
ﬁ'om contact with the juvenile court system.

The! ﬁndings from Alfaro's research indicated that an empirical relationship
between child abuse and delinquency does exist. Alfaro indicated that as many as 50% of
the families that were reported for child abuse during the 1950's had at least one child who
was involved in the court process for delinquency. However, it is not guaranteed that an
abused child will always become delinquent or that a delinquent child will always have an

abusive past.

 

31

From the 1970's data set collected from the juvenile intake unit, 21% of the males
and 29% of the females who were reported as delinquent had been reported as abused or
neglected children when they were younger. Additionally, Alfaro indicated that there is an
association between those children who were reported as abused in that they tend to be
more violent than non-abused children. From the 1970's sample, 28.6% of the adolescents
who were adjudicated for homicide, rape, and assault had been reported as abused or
neglected as children. Therefore, Alfaro determined that the relationship between child
abuse and subsequent delinquency does exist and that abused children may be violent
children.

However, Alfaro stated that the abuse cannot be used as a predictor of behavior.
Alfaro indicated that a relationship between child abuse and delinquency does exist,
however, not all child maltreatment, abuse or neglect always leads to delinquent behavior.
Some factors may inﬂuence the strength or weakness of this relationship. Alfaro indicated
that peer relationships or the neighborhood environments may help to strengthen the
relationship. Fortunately, community services such as counseling, parenting classes, day
care, and foster care may weaken the relationship between child abuse and delinquency.
These community services may inﬂuence the child and the subsequent behavior by
reducing stress and teaching appropriate parental behavior, thus weakening the
relationship between child abuse and delinquency.

Alfaro's research does determine the existence of a relationship between child
abuse and delinquency. However, the ﬁndings underestimated this relationship and
indicated it to be a conservative relationship. Alfaro indicated that collecting the data was
a problem due to the destruction of past records. The researcher indicated that using
oﬁcial records often causes an under reporting of actual child abuse cases. It is known
that not all examples of child abuse are reported, hence this affected the actual cases used
for the purpose of this study. Alfaro indicated that these reasons caused the conservative
ﬁndings of the relationship.

32

Alfaro determined that abused and neglected children have a greater likelihood of
becoming delinquent than non-abused or non-neglected children. Though the ﬁndings of
the research showed that this relationship is conservative, Alfaro oﬂ‘ered his opinion as to
why this is true. Alfaro explained the use of oﬁicial records and the loss of data are
signiﬁcant contributors to the conservativeness of the relationship. However, Alfaro
indicated that the relationship does exist and that interventions and social policy must be
created to reduce the effects of child abuse. He advised that community services such as
counseling and parenting classes be used as interventions to sever the existence of the
relationship between child abuse and delinquency.

Kratcoski and Kratcoski (1982) investigated the relationship between being the
object of physical abuse as a child and later manifestations of violence directed toward
members of the immediate family or caretakers. The purpose of this research was to
examine the possible relationship between being a victim of physical abuse and later eﬁ‘ects
of acting violent toward family members or caretakers. The researchers deﬁned violent
behavior as ”the illegal use or threat of force against persons. " The acts that were
included in this deﬁnition were: assault, aggravated assault, robbery, aggravated robbery,
rape, arson, threatening behavior, menacing, kidnapping, injury to persons, burglary,
breaking and entering, and murder. The researchers deﬁned abuse as: ”any non accidental
physical injury inﬂicted on a child by a parent or other caretaker deliberately or in anger.”
Kratcoski and Kratcoski examined the case ﬁles of 863 adjudicated, male, juvenile
oﬁ‘enders. Also, caseworkers or psychologists, who were familiar with a particular case,
were interviewed to obtain information that was not provided with the case ﬁles.
Occasionally, the researchers had to make a detennination about abuse based upon
situations that were described within the case ﬁles.

From the sample, 223 or 26% of the children had experienced some form of
physical abuse. The researchers divided the sample into abused and non-abused
delinquents and analyzed any possible differences between the groups. Both the abused

33

and non-abused groups had more than 75% of the youths who had committed at least one
violent act. However, 25% of the abused delinquents directed the violence toward family
members or signiﬁcant others, whereas, 14% of the non-abused delinquents directed the
violence toward family members or signiﬁcant others. Therefore, the ﬁndings from the
research indicated a slight difference between the frequency of violent acts committed by
abused and non-abused delinquents. However, there was a difference where the violent
act was directed. The abused delinquent was more likely than the non-abused to direct the
violence toward family members or signiﬁcant others.

For the purpose of this research Kratcoski and Kratcoski examined the case ﬁles of
incarcerated male, juvenile offenders. However, the researchers often had to make a
determination of abuse based upon the social workers' or psychological reports contained
within the case ﬁles. The researchers may have been inaccurate and biased when
abstracting information from the ﬁles. The possible inaccuracy by the researchers would
aﬁ‘ect the total number of abused adolescents and the ﬁndings of the study. The
researchers needed an independent rater to abstract information from the case ﬁles. This
would eliminate any possible miscalculations or biases by the researchers. Furthermore,
the researchers stated that the violence directed at signiﬁcant others did not always result
in an oﬁicial arrest. Therefore, this information had to be abstracted ﬁom the case ﬁles.
This is a problem due to the possible biases of the researchers when abstracting
information ﬁ'om the ﬁles.

Kratcoski (1982) conducted a study to determine if there is a relationship between
child abuse and an adolescent's subsequent violent behavior directed toward family
members or caretakers. Kratcoski incorporated the Social Learning theory which
maintains that violence is a learned behavior oﬂen learned during early life experiences.
To determine if there is a relationship, Kratcoski surveyed case ﬁles of delinquent
adolescents at four facilities in Ohio. All subjects are incarcerated as serious, male,

juvenile offenders. The researchers collected demographic information about the

34

adolescents as well as information about the nature of oﬁ‘ense and number of oﬁ‘enses
contained in the case ﬁles. Information about the family and possible child abuse was
obtained from psychological reports in the ﬁles. Often, the researcher made a
determination of abuse according to situations described in the records. For the purpose
of this research, violent behavior was deﬁned as "the illegal use or threat of force against
people" and abuse was deﬁned as ”any non accidental physical injury inﬂicted on a child by
a parent or other caretaker deliberately or in anger”. Additionally, signiﬁcant others
included parents, stepparents, foster parents, adult relatives, siblings, close peer groups,
and caretakers such as institutional staﬂ‘, teacher, or youth leaders. In total, the sample
consisted of 863 cases of which 223 offenders (26%) experienced some form of physical
abuse.

Kratcoski compared the characteristics of the abused and non-abused oﬂ‘enders
from the sample to determine the extent of violence directed toward family members and
signiﬁcant others. Analysis of violent behavior indicated that 26% of abused adolescents
acted out violently toward signiﬁcant others as compared to 14% of the non-abused
oﬁ‘enders. Additional analysis of those adolescents who committed violent crimes against
people indicated that 45% of the abused, adolescent offender directed the violence
towards family members or caretakers as compared to 18% of the non abused offenders.
Therefore, an abused adolescent is more likely than a non-abused adolescent to act out
violently towards an individual and more importantly towards family members or
signiﬁcant others. Kratcoski concluded that those adolescents who abused their parents
and acted violent towards signiﬁcant others tended to come ﬁom families in which
violence, disruption, and discord were everyday occurrences.

Kratcoski had to make determinations of possible child abuse based upon
psychological reports contained within the case ﬁles of the incarcerated offenders. The
ﬁndings of the research may be biased due to the researcher abstracting the information

from the ﬁles. The researcher may have been biased when trying to abstract information

35

that would ﬁt the purpose of this study. Also, the researcher indicated that the ﬁndings of
the study may be distorted since some psychologists when interviewing the adolescents
never brought up the subject of discipline. This would have possibly uncovered additional
instances of abuse. Furthermore, the information about abuse was obtained from the
psychological interviews with the adolescents. During these interviews, these adolescents
may have been deceitﬁrl or exaggerated their relationship with their parents and the type
of discipline used within the family enviomment. This also would distort the ﬁndings of
the research. Also, Kratcoski did not examine any possible factors which may prevent an
abused child ﬁom directing their violence toward family members. The researcher
seperated the abused offenders from the non abused offenders, however, the researcher
did not investigate those factors which may be different between samples.

Alfaro (1981), Kratcosld (1982),and Kratcoski and Kratcoski (1982) examined the
relationship between child abuse and subsequent delinquency. All three researchers
concluded ﬁ'om their ﬁndings that abused delinquents have a greater liklihood of being
violent than non abused delinquents. Kratcoski and Kratcoski et al. determined that those
delinquents who were abused directed their violent actions toward family members or
signiﬁcant others more often than non-abused deliquents. Alfaro indicated that other
Ectors such as peer relationships or community services may inhibit or strenghten the
relationship between child abuse and subsequent delinquency. Other researchers
investigated other factors which may mediate the relationship between child abuse and
delinquency. Some researchers explained the effects of abuse by examining family
violence and family criminality ( Fagan and Wexler 1987) or by controlling for gender and
race (Widom, 1989). Other researchers investigated the interaction between the child and
the parent (Pfouts et al. 1981 and McCord 1983) and the role of the child within the
family (Gillespie et al. 1977, Owen and Straus 1975, and Pfouts et al. 1981).

Fagan and Wexler ( 1987) examined the process of learning violence within the

family environment. The purpose of this study was to examine the family interactions and

36

environments of chronic violent juvenile oﬁ‘enders and their inﬂuence on subsequent
delinquency. Fagan and Wexler interviewed 98 violent, juvenile delinquents and their
maternal caretakers. The adolescents were identiﬁed by the juvenile courts in different
metropolitan cities: Memphis, Boston, Newark, and Phoenix. The adolescents were all
adjudicated violent felons for murder, rape, armed robbery, aggravated assault, arson, and
voluntary manslaughter. Data was collected ﬁom child protective service records, juvenile
court records, and ﬁom family interviews. The interviews provided information about the
family interaction of the violent juvenile offenders. The ofﬁcial data were used to validate
delinquency, family child abuse and disturbances histories. Also, a self report delinquency
scale was used to obtain information about the adolescents’ involvement in criminal
activities.

The researchers deﬁned four norms to leaming violence within a family
environment. The four norms were: violence are 1) legitimacy, 2) useﬁilness for obtaining
a goal, 3) authority, 4) dispute resolution through maintaining control. These four norms
establish a framework within a family as to how violence was learned. Fagan and Wexler
determined that violence was taught through the family interactions. They reported three
types of family interaction: 1) ”interactionist”, which were families that exhibited a high
degree of interaction and bonding, 2) "hierarchical”, which were families characterized by
parental dominance, and 3) ”antisocial", which were families marked by criminality and
family violence. Fagan and Wexler determined that these three family interactions were
the means of how violence was taught within the family. The researchers argued that
violence was reinforced within the family environment as well as outside the family by
"social constructs" including schools, peers, and neighborhoods.

The results from the family interviews indicated that 45% of the families reported
some form of violence occurring within the home environment and 38.4% of the violent
youths responded that they were hit by their parent. Additionally, 37% of the adolescents
and their siblings were hit with an object used by the parent. Oﬁicial reports ﬁ'om child

37

protective service agencies indicated that 14% of the adolescents had been reported as
physically abused and 11% of the adolescents were exposed to spousal abuse. Combined,
these data indicate an overwhelming majority of the violent adolescents had been exposed
to violence within their family environment.

Fagan and Wexler investigated several family variables that related to subsequent
delinquency by the child. They reported that family violence among parents and by
parents toward children contributed to the adolescents understanding of violence as an
acceptable behavior. These violent adolescents observed or experienced violent behavior
including physical abuse within the family. The pattern of family behavior that Fagan and
Wexler reported was that violent delinquents viewed their parents as having ﬁequent
contact with the law and being violent. The researchers determined that family violence
and Emily criminality exhibited strong associations with the juveniles’ subsequent
delinquency and violent behavior. The researchers indicated that these factors contributed
to the adolescents’ criminal involvement by providing negative role models.

Fagan and Wexler determined that the family interactions and environments have a
direct eﬁ‘ect on an adolescent's perception of violence. The researchers indicate that an
adolescent who was raised in a violent family environment was more likely to be violent
than an adolescent who was not raised in a violent family. However, the researchers failed
to explain how each type of farme interaction was related to subsequent delinquency. The
types of Emily interactions were used to classify families but statistical analysis was not
completed. Statistical analysis was completed only for those families that were considered
violent, therefore, eliminating any comparision between types of family interaction.

McCord (1983) conducted a 40 year longitudinal study to examine the effects of
child abuse and neglect. The researcher obtained information on the lives of 232 males
ﬁ'om case ﬁles during the years 1939 through 1945 and ﬁ'om interviews conducted with
counselors assigned to the various families. The researcher also developed a comparison

group by matching a similar boy of similar background with subjects in this study group.

38

The study group was divided into four categories: abused, neglected, rejected, and loved.
These categories were determined by the researcher according to how the males had been
treated as children McCord deﬁned abused as those children "consistently subjected to
punitive, physical punishments". Neglected were those children whose ”parents interacted
inﬁ'equently, showing neither affection nor rejection". Rejected were those children whose
”parents demonstrated repeated displeasure with the child". Finally, loved were those
children ”who had at least one parent who showed concern and was pleased with the
child's behavior". In total there were 49 abused, 48 neglected, 34 rejected, and 101 loved
adolescents. These groups were traced forward forty years to determine the effects of
child abuse and/or neglect.

Similarly, McCord classiﬁed the families according to the primary mode of
discipline. From the interviews conducted with the families' counselors, parents were
considered aggressive if ”they threw objects, or attempted to injure someone in response
to frustration or annoyance. " Also, the parents were considered punitive if ”corporal
punishment was the primary mode of discipline. " The researcher rated 92% of the mothers
and 84% of the fathers as aggressive and determined that 88% of all the adolescents were
raised in homes considered to be punitive.

McCord's analysis of the mode of discipline and parents’ interaction with the
children indicate that 35% of the abused and 44% of the rejected individuals had high rates
of parents who were classiﬁed as aggressive, whereas, those children classiﬁed as loved
were least likely of any group to have aggressive parents (15%). Those children who
were classiﬁed as abused, 76% of the fathers were clearly dominant and most often
punitive towards the child.

Between the years 1975 and 1979, the records of the study group were traced to
determine the effects of abuse and/or neglect on criminal involvement. Court records
were analyzed to determine an individuals criminal involvement. The effects of parental

abuse, neglect, and rejection have a deﬁnite impact on criminal involvement. Among the

39

97 neglected or abused children, 74% had committed some form of serious crime. Serious
crime was deﬁned as theft, auto theft, breaking and entering, burglary, and assault.
Furthermore, those children (29%) who had been rejected by their parents were found to
have a higher rate for juvenile criminal behavior than those who were abused or neglected,
10% and 15% respectively.

McCord also investigated what factors mitigate the possible eﬂ‘ects that abuse,
neglect, or rejection may have upon a child. She determined that having an affectionate
parent as well as a mother that was self conﬁdent made a child less vulnerable to
delinquency. Surprisingly, McCord found that being held back in school had the highest
mitigation rate (77%) overall. Additionally, she also determined that those children who
returned to school were less vulnerable (70%) than those children who did not return to
school.

McCord's longitudinal research is vital to the understanding of child abuse and
neglect and its relationship with subsequent delinquency. This study examined the long
term eﬁ‘ects that are present within a sample of men who were abused, neglected, rejected,
or loved by their parents. McCord detennined that children who have been abused,
neglected, or rejected have higher rates of delinquency than those children who were not
abused, neglected, or rejected. McCord's ﬁndings indicate that the interaction between
child and parent is a signiﬁcant factor that affects the relationship between child abuse and
subsequent delinquency. In fact, those children who had at least one loving parent were
the least likely of all the groups to engage in delinquent or criminal behavior. Those
children who comprised the abused, neglected, and rejected groups were subjected to
parents who were aggressive, punitive, conﬂictual, and uncaring. These parental
behaviors contributed to strengthening the relationship between child abuse and
delinquency. However, McCord indicated that maternal self conﬁdence and an individuals

education are factors which may hinder the possible eﬂ‘ects of child abuse.

40

Cathy Widom (1989) also examined the relationship between abuse and/or neglect
and the propensity to engage in juvenile criminal behavior and/or violent criminal behavior.
The purpose of the research was to determine if an abused child is at a higher risk to
commit delinquent oﬂ‘enses than a non abused child, if certain factors such as gender and
race increase the risk for delinquency, and if certain behaviors, sexual and/or violent, are a
result of the past victimization. Widom compiled a large sample of 908 substantiated child
abuse and neglect cases ﬁom the preceding 20 years and compared this sample with a
non-abused matched comparision group. A comparision group was matched as closely as
possible for similar sex, race, age, and appropriate socio-economic status. To obtain data
about child abuse for this study, Widom used ofﬁcial reports ﬁ'om the juvenile court and
probation department during the years 1967 through 1971 in a metropolitan area of the
Midwest. She also used ofﬁcial arrest records to determine the presence and severity of
delinquent or criminal behavior.

The results of this study indicated that abuse and neglect does put a child at an
increased risk for future delinquency. Widom determined that 26% of the subjects in the
abuse/neglect group committed delinquent act(s) compared to 17% of the non-abused
group. She also found a signiﬁcant difference in the number of offenses committed by the
two groups: 370 total offenses in the abuse/neglect group compared to 197 total oﬁ'enses
in the non-abused group. These ﬁndings indicate an increase of risk to engage in
delinquent oﬂ‘enses by an abused child.

Generally, Widom concluded the risk of delinquency involvement was 1.89 times
higher among abused or neglected children than among non-abused children. Widom also
investigated if there was an increase of risk for delinquency based on the child's gender.
Widom reported that when gender was controlled for, being a male victim of
abuse/neglect increased the risk from 22% to 33%, but for a female victim the risk
increased from 11% to 19%. She determined that being a victim of abuse or neglect has a

stronger delinquency effect upon black children than non black children and more so on

41

black males than black females. Speciﬁcally for black males, 50% of abuse/neglect group
had arrests compared to 25% of non-abused black males. For black females, 26% of the
abuse/neglect group had arrests versus 13% of non abused group. These differences for
both black males and females were signiﬁcantly more than the differences determined for
white adolescents. The difference of delinquency involvement for white males was 26% of
abuse/neglect group compared with 21% of the non abused white males. Also for
females, the difference was 17% of the abuse/ neglect group had arrests versus 11% of the
non abused white females.

Widom investigated if there are certain criminal behaviors that are a result of past
childhood victimization. She concluded that an increase of risk for delinquency by those
children who are abused and neglected was primarily associated with property crimes such
as stealing and vandalism and status oﬁ‘enses including truancy and ungovemability.
However, the effects of abuse and neglect are not restricted to only these certain
behaviors. Widom also investigated if there was a relationship between criminal sexual
conduct and past sexual victimization. She separated those children who had only been
sexually victimized from the rest of the abuse/neglect sample to determine if this
relationship does in fact exist. Widom determined for this sample that there was not a
direct link between past sexual victimization and criminal sexual behavior, though she
admits this is contrary to past research. Furthermore, Widom investigated if victims of
physical abuse are more violent than non-physically abused children. Overall, those
children who were physically abused did not differ in regards to violent offenses than non-
physically abused children. However, those children that were neglected had the highest
rates for violent crimes than any other group. This was especially true for neglected
males.

Widom concluded that there is an increase of risk for delinquency by an abused or
neglected child than a non-abused or neglected child. Unfortunately, Widom fails to

report a distinction between those children who have been abused and those children who

42

have been neglected. She reported that there were similar effects for the abused and
neglected groups. The only distinction that she reported was the difference for neglected
children to be more violent than abused children. However, she admitted this was
contrary to past research and explained the ﬁndings as an illustration of the complex
nature of child abuse. Additionally, she investigated the effects of abuse and neglect when
controlling for gender and race. Widom determined that black abused and neglected
males had the highest increase of risk for delinquency than any other group. However
using oﬁcial data often over represents minority groups including black males and
females. This over-representation may indicate a bias and discrimination of the ﬁndings of
this research.

Gillespie, Seaburg, and Berlin (1977) use a social learning theoretical ﬁamework
to investigate the learning of delinquent behavior through the exposure to abuse: physical,
sexual, and/or emotional. The researchers proposed to examine the internal dynamics of
Emily behavior and its relationship to the learning of aggressive behavior. Gillespie et. al.
hypothesized that a child who experienced abuse either through observation or direct
experience will be more likely to demonstrate delinquent behavior than a child who was
not exposed to an abusive family. Speciﬁcally, the researchers explained that those
children who were abused, having suffered from the aggression, will be more likely to not
be aggressive, but engage in escape crimes such as truancy and running away in order to
avoid further aggression. Gillespie et al. also explained that those children who are
siblings of abused children, having observed high levels of abuse but not been the
recipients of the aggression, will be more likely to exhibit aggressive and anti-social
behavior due to modeling the aggressive behavior.

Gillespie et al. obtained data ﬁ'om the Arizona State Department of Economic
Security for child abuse and from ofﬁcial juvenile court records. The total sample of
adolescents that were included in both records was 774. The sample was divided into

what type of abuse was present within the family environment. The categories were:

43

physical abuse, sexual abuse, emotional abuse, and physical abuse neglect. The siblings of
the abused adolescents who had not been directly abused but had observed the abuse,
were also used for the purpose of this study to determine if the sibling did model the
observed, abusive behavior. A comparison group of 900 reported juvenile oﬁ‘enders with
no abusive family history were used for this study to validate or invalidate a modeling
process of the familial abuse.

The researchers divided the crimes into 57 categories and analyzed two of these
categories: escape crimes (ninning away, truancy, and missing juvenile) and aggressive
crimes (disturbing the peace, ﬁghting, assault, assault with deadly weapon, aggravated
assault and battery, armed robbery, and strong arm robbery). Data analysis consisted of
doing ﬁ'equency tabulations and a three way chi-square test for escape crimes and
aggressive crimes.

The results of the data analysis indicated that children who are physically abused
are more likely to become involved in delinquent activities than those children who are not
physically abused. Gillespie et al. compared the ﬁ'equency rates of aggressive and escape
crimes committed by physically abused children with the ﬁ'equency rates for aggressive
and escape crimes committed by the comparison group, the non-abused oﬁ‘ender. The
researchers found that the physically abused children had a lower ﬁequency rate than the
comparison group for aggressive crimes: 17.2% versus 23.5%, and a higher ﬁequency for
escape crimes: 92.2% versus 76.5%. However, the siblings of the physically abused
juvenile offender had a frequency of 17.2% for aggressive crimes compared with a 7.8%
ﬁequency by the physically abused child. Furthermore, Gillespie et al conducted analysis
of escape crimes. The abused children had a 92.2% ﬁequency, whereas their siblings had
an 82.2% frequency.

The ﬁndings of the analysis supported the researchers hypothesis of observation
and direct experience of abuse. Gillespie et al. hypothesized that the child who observed

the parent abusing a sibling will be more likely to demonstrate aggressive behavior

through a modeling effect than a child who directly experiences the abuse. The researchers
hypothesized that the parent is a person of status within the family and rarely receives a
punishment for being abusive, therefore the child vicariously learns that the abusive
behavior is an acceptable and possible rewarding behavior. Conversely, the researchers
hypothesized that those children who were abused will experience the abuse as punishment
and will have a reduced tendency to manifest aggressive behavior. The ﬁndings from the
research indicated that there was a direct tendency for abused children, having directly
experienced the abuse, to avoid ﬁrrther punishment and aggression by committing escape
crimes such as truancy and running away. Also, those siblings of abused children, who
observed abuse but had not been the direct recipient of the abuse, had a higher ﬁ'equency
to commit aggressive crimes. This supported the researchers hypothesis that those
children who observe aggressive behavior will be more aggressive than those children who
are recipients of the abuse.

Gillespie et al. investigated the inter-familial patterns of behavior to demonstrate
that aggression is learned through vicarious observation. Gillespie et al. separated the
sample into four different categories but only conducted analysis on those children who
either directly experienced physical abuse or observed physical abuse. The researchers
Eiled to explain if sexual behavior as a result of sexual abuse was vicariously learned or
what eﬁ‘ects emotional abuse or neglect may have upon subsequent behavior. The
researchers did not expand upon either type of abuse and possible effects of either abuse.
Also, the ﬁndings ﬁ'om this research indicated that the observation of abuse is a more
powerful learning tool of aggression than the direct experience of abuse. These ﬁndings
are consistant with the ﬁndings of Owen and Straus (1975) and contrary to the ﬁndings of
Pfouts et al. (1981).

Owen and Straus (1975) investigated the relationship of exposure to violence as a
child with the approval or use of violence to achieve personal or political ends as an adult.

The researchers deﬁned exposure to violence as: l) observing violence, 2) being a victim

45

of violence, and 3) committing violence or acts violently. The researchers stated that
violence is a learned behavior and much of the learning takes place during childhood years
through observation, direct experience, or acting violently. Therefore, the research
hypothesis of this study is that the more a person is exposed to violence as a child through
observation, victimization, or committing violent acts, the greater the approval or use of
violence as an adult.

Data for this research is a reanalysis of data collected for The National
Commission on the Causes and Prevention of Violence. The National Commission was a
nationwide survey of a sample of 1,176 individuals. The surveys consisted of self-
reported behaviors concerning the respondents’ use and approval of violence. The
respondents were 18 years of age or older and they were selected by means of cluster
sampling. A cluster was Operationally deﬁned as a city block.

Owen and Straus conducted statistical analysis to examine the eﬁ‘ects of exposure
to violence as a child and its relationship to interpersonal violence. Correlation
coeﬁcients were computed for individuals who committed violence, who observed
violence, and who were victims of childhood violence. Correlation coefﬁcients were used
to determine the strength of the relationship between exposure to violence as a child and
involvement in interpersonal violence. The individuals who acted violently as children had
the highest correlation coefﬁcient (.31) for interpersonal violence. Additionally, those
individuals who were victims of violence as children had a correlation coeﬁcient of .21
and those individuals who observed violence as children had a correlation coeﬂicient of
.29. Therefore, the researcher found that all three aspects of exposure to violence, acting
violently, observation, or victimization, produce correlations that support the research
hypothesis, that is, that there is a relationship between exposure to violence as a child and
the use or approval of interpersonal violence as an adult.

Owen and Straus investigated the relationship between exposure to violence as a

child through acting violent, victimization, or observation and the use ”of violence as an

46

adult. The researchers concluded that those children who observed violence had a higher
correlation to act violent than those children who were victimized by physical violence as a
child. Consistent ﬁndings were reported by Gillespie et al. (1977). Gillespie et al.
concluded that those children who observed physical violence within the family were more
likely than physically abused children to be violent. Additionally, Gillespie et al. and Owen
and Straus explained that children who were victims of physical abuse were more likely
than non-victims of abuse to engage in escape crimes such as truancy and running away
rather than violent, aggressive crimes. Inconsistent ﬁndings were reported by Pfouts et al.
(1981). Pfouts et al. concluded that children who were victims of physical abuse were
more violent than those children who observed violence. Pfouts et al. indicate that 48.9%
of the adolescents who were victims of physical abuse engaged in severe delinquent
behavior, as opposed to 20.4% of the adolescents who were bystanders to family violence.

Pfouts, Schopler, and Henley (1981) explored the extent of family violence and its
association with childhood deviance. The researchers investigated if the variations in
deviance are caused by the emotional climate of the family, the relationship between the
child and the parent, and the role of the child in the structure of family violence. The role
of the child in the structure of family violence was operationally deﬁned as an observer or
victim of the abuse. The researchers deﬁned family violence as abuse, either paternal or
maternal, and inﬂicted upon any member of the family. The researchers hypothesized that
the relationship between the child and the parent and the role of the child in the structure
of family violence will determine the extent of deviance. Speciﬁcally, children who were
bystanders to family violence will exhibit minimal deviance if their relationship with their
parents has some positive features and mild deviance if their relationship is characterized
by parental rejection. Also, children who were victims of family violence will exhibit mild
deviance if their relationship with their parents has some positive features and severe

deviance if their relationship is characterized by parental rejection.

47

The researches examined the case ﬁles of 800 families investigated for abuse by the
North Carolina Protective Service Workers during the years 1971 to 1977. From the total
population of 800, a sample of 73 families were selected based upon past case histories of
substantiated abuse, either involving a child or a spouse. Information was collected about
these Emilies ﬁ'om the initial reports of the abuse investigation. Also, the researchers
conducted interviews with the social workers or supervisors responsible for each case.
Based upon the interviews with the social workers, Pfouts et al. determined the
relationship between the child and the parent and the extent of childhood deviance.

The researchers analyzed the characteristics of the abusive families to determine
the frequency of victimization and the type of perpetrator of the abuse. From the sample
of 73 families that have substantiated histories of abuse, 50% involved violence directed
toward the child, 20% toward the 'wife, and 30% involved abuse of both the wife and the
child. Analysis of the perpetrator showed that fathers and father ﬁgures were the sole
abusers in 47.4% of the families while the mother was the sole abuser in 35.1% of the
Emilies. Therefore, 80% of the physical abuse was directed toward both the child and/or
the mother and was committed (47.4%) by the father or father ﬁgure within the Emily
environment.

Pfouts, Schopler, and Henley discovered a widespread amount of deviant behavior
among those children who either observed the physical abuse or were victims of physical
violence within the family. Pfouts et al. determined that the deviant behaviors that were
exhibited most often were lying and cheating (45%), truancy (40.0%), assault (30.3%),
stealing (29.9%), and vandalism (24.1%). Additionally, the researchers discovered
physical eﬁ‘ects such as depression (64%), and hyperactivity (31.1%) as results from
physical violence that is present within the family environment.

The results of the research indicate that those children who are victims of Emily
violence have a higher degree of deviancy than those children who are witnesses to family

violence. The ﬁndings also indicate that 48.9% of the adolescents who were victims of

48

physical abuse engaged in severe delinquent behavior as opposed to 20.4% of the
adolescents who were bystanders to family violence. Of the 99 children who manifested
symptoms of abuse, 92 had engaged in some form of delinquent behavior. Additionally,
the researchers determined that abuse victims exhibited the most extreme forms of
delinquent behavior. These children were most often involved in delinquent behaviors
such as assault, stealing, vandalism, and truancy.

Pfouts et al. examined the ofﬁcial investigation reports of child abuse from the
Protective Service Unit in North Carolina to determine the emotional climate of the Emily
and the role of the child in the structure of family violence. By using oﬁicial reports, these
data for the research tend to be biased toward those families from the lower socio-
economic scale and minority groups. Therefore, the sample may not be representative of
the characteristics of the general population. Additionally, the researchers determined the
presence of abuse and the emotional climate of the family ﬁ'om the Protective Service
reports. These reports typically contain more information about those children who
exhibit behavioral problems as opposed to those children who are withdrawn or who do
not exhibit any behavioral problems. Therefore, this information provided contained a bias
in the direction of deviancy and affected the ﬁndings of the research.

Saga] Abuse

Brannon, Larson, and Doggett (1989) explored the extent and the origins of sexual
abuse perpetrated upon incarcerated male juvenile offenders. The purpose of this study is
to investigate and ﬁrrther clarify the prevalence of sexual exploitation of youth who are
entering the juvenile justice system. The authors investigated the extent and origins of
abuse through analysis of structured, conﬁdential interviews conducted by two staﬁ‘
psychologists within the setting of a residential program. These interviews were
conducted with a population of 63 incarcerated, male, juvenile offenders. These
adolescents have been adjudicated for a variety of offenses including crimes against

property and persons. Additionally, eleven of the 63 participants were adjudicated for sex

49

related oﬁ‘enses. The authors conducted analysis of the psychological interviews using
both chi- square and descriptive statistics. Initially, the authors differentiated between
sexual abuse and molestation by operationally deﬁning sexual abuse as occurring through
the use of force, threat, or violence, and molestation as being tricked, seduced, or
manipulated into sexual relations without being intimidated by verbal threats and/or
physical coercion.

Overall, results of the study report 70% of the participants as having been
victimized by either sexual abuse or molestation, 57% of the victimization having occurred
through molestation. The authors also investigated the types of molestation or abuse
experienced by these adolescents and the age at which the molestation or abuse occurred.
The ﬁndings indicated that the majority of the molestation involved intercourse (50%) or
fellatio (33%). Also, the authors discovered the mean age of the molested adolescents to
be 9.3 years of age. However, the authors indicated that the mean age for those who were
sexually abused was 8 years of age. The ﬁndings indicated that females consisted of 58%
of the victims and males 42% of the victims.

Brannon, Larson, and Doggett also investigated the perpetrators of the molestation
or abuse among those incarcerated, adolescent males. The mean age of the male
perpetrator was 27 years of age compared to 17 years of age for the female perpetrator.
An in depth investigation of the female perpetrator uncovered that those females may be
ﬁiends of older siblings or baby-sitters who actively victimize young children.

Also, analysis of the identity of the perpetrator indicated that 60.8% of the
perpetrators of molestation were a known individual to the adolescent and 26% were
Emily members. Data analysis of the origins of sexual abuse indicate that the victims of
sexual abuse were victimized in equal numbers (37.5) by either members of the immediate
family or individuals who were known by the victims. Additionally, all the perpetrators of

molestation were reported to have been males and the mean age was 21.8 years of age.

50

Brannon et al. conducted research to determine the extent and frequency of sexual
abuse and molestation among incarcerated, juvenile offenders. The researchers obtained
data by conducting psychological interviews with the subjects. These interviews may be
problematic due to the subjects possibly being deceitﬁrl when answering the questions.
Also, the subjects may have exaggerated the extent and details of the sexual abuse that
was perpetrated upon them. By exaggerating the circumstances of the abuse, the
adolescent rationalizes why the abuse occurred to him. Furthermore, the researchers did
not make any comparisons between the type of abuse encountered and the form of crime
the adolescent committed. The authors neglected to establish any form of relationship
between abuse and crime. VVrthin the research, the authors establish that 11 of the
subjects had been adjudicated for a sexual related offense. However, the researchers
never conducted any analysis to determine the severity of abuse of these sex offenders, or
if there was a relationship between abuse and sexually offending. Though the research
provides substantial information about the extent of abuse among incarcerated, juvenile
oﬁ‘enders it does not provide any information about a linkage between sexual abuse and
subsequent delinquent behavior.

Benoit and Kennedy ( 1992) examine the ﬁ'equency and intensity of physical and
sexual abuse among incarcerated adolescent sexual offenders within a secure residential
treatment program in Florida. The researchers hypothesized that the incidence of sexual
abuse will be more frequent and intense for adolescent sexual offenders and that the
incidence of physical abuse will be more ﬁ'equent and intense for the violent adolescent
offenders. The researchers developed a control group of non-sexual oﬂ‘enders allowing
for a comparision between sex offenders and non-sex offenders.

The total population for this study was 100 males ages 12-18 years of age. All
subjects had been adjudicated for a felony crime and incarcerated within a secure
residential setting. Data for this study was obtained through reviewing the records of

those subjects who had been randomly selected. The records were reviewed by graduate

51

students, who had no investment in this study, and divided into four categories with 25
subjects in each category. The categories are: 1) non-aggressive offenders, and 2)
aggressive offenders, (in both categories 1 and 2 the offenders had not committed any
known sexual crimes), 3) female molesters and 4) male/female molesters. The researchers
statistically computed chi-square to make comparisons between the experimental and
control group in order to prove or disprove the research hypothesis.

The ﬁndings from the data indicate that there is a widespread amount of physical
and/or sexual abuse that is prevalent across all four categories of oﬂ‘enders. Benoit and
Kennedy determined that 34% of the population had been sexually victimized during their
development. Additionally, 26% of the sexual offenders had been sexually victimized as a
child. Eleven subjects who were sexually victimized as a child had been victimized by
more than one individual. Additionally, the researchers determined that the mean age of
victimization was 6.59 years of age. Furthermore, the researchers determined that 52% of
the population had been exposed to physical abuse during their developmental years.
Also, 10 of the subjects who had been physically abused received permanent physical
injury as a result of the physical abuse.

Benoit and Kennedy determine that sexual and physical victimization is present
across all categories of oﬁ‘enders and there is no difference in the ﬁ'equency or intensity of
past abuse. They also determine that an abusive past is not a sufﬁcient reason for
victirnizing others. Furthermore, the relationship between being a victim of abuse and the
subsequent offending is not a direct relationship and that other variables must be present.
However, the researchers make no indication as to what these other factors may be. The
researchers indicate that past victimization does not determine an individuals behavior and
that others factors must combine with the abusive past to compel an individual to victimize
others.

The researchers examine the frequency and intensity of sexual and physical abuse

present among adolescent, sex offenders. The researchers determined that both forms of

52

abuse, sexual and physical, are present among all categories of offenders. The categories
of offenders included sexual offenders and non-sexual oﬁ‘enders. However, the
researchers failed to make any distinction of intensity of abuse between the categories of
offenders. Also, the researchers did not report or analyze if any relationship existed
between the abuse that was suﬂ‘ered and the form of crime committed. The researchers
also stated that the relationship between being a victim of certain types of abuse and
subsequent oﬁ‘ending is not a direct relationship, that other factors must be present.
However, the researchers failed to determine these other factors.

Longo (1981) researched the effects of early exposure to sexual activity among
adolescent, sexual offenders. He employs the social learning theory for the purpose of this
research, which provides a framework that the adolescent began to learn about sexual
experiences at an early age and with individuals who were typically older than the
adolescent. Longo explains that this early exposure to adult sex can be a traumatic
experience whether or not coercion was involved. This traumatic experience may cause
concerns by the adolescent over his sexuality and Longo hypothesized that the adolescent
will act out sexually to overcome possible feelings of inadequacy.

Longo administered a conﬁdential questionnaire to residents of a sexual offenders
program in Florida. The total sample for the study is seventeen (17). All subjects are
nineteen years of age or younger and have been tried and sentenced as adults. Of the total
sample, eight (8) are black and nine (9) are Caucasians. Data based upon the adolescents’
past sexual history and involvement was collected and presented by Longo. Data analysis
determined that 47% of the adolescents were victims of sexual molestation as a child. Of
those victims of child molestation, 76% reported that their ﬁrst sexual experience was
before age twelve. Data analysis also determined that the mean age of ﬁrst sexual
experience was 9.1 and it occurred on the average with an individual 8 years older. The

data also indicates that the mean age when ﬁrst learned about sex as 9.5 years of age.

53

The adolescents involved in this study are all sexual offenders and all have had an
early exposure to sex and sexual experience. Also, almost 50% of the sample were
victims of molestation or sexual abuse. The results indicated that these sexual oﬂ‘enders
have had sexual experiences during their elementary school years and before the onset of
puberty. The results of this study indicate an atypical sexual development by these
adolescents. The researcher concluded that the inadequate development resulting ﬁ'om
early sexual exposure and victimization ﬁ'om molestation form a traumatic experience in
the lives of these oﬁ‘enders. Longo then concludes that this sexually traumatic experience
is a learning experience and may be the primary cause of the adolescent sexually acting
out.

The researcher obtained data for this study by administering a conﬁdential
questionnaire to the subjects. The use of a questionnaire allows the respondent to be
potentially deceitful when answering the questions. Also, the research population for this
study was 17; this is a small sample and limits the Opportunity to make generalizations to
the entire population. The researcher determined that those adolescents who were
adjudicated for a sexual offense had a higher occurrence of sexual trauma during their
childhood years. However, the experience of sexual trauma was not exclusive to only
sexual offenders. There was the occurrence of sexual trauma among non-oﬁ‘ending
adolescents as well. Furthermore, the researcher indicated that other Ectors play a part in
the relationship between abuse and subsequent offending, however, the researcher did not

make any indication as to what these factors are.

DATA ANALYSIS AND PRESENTATION OF FINDINGS

The purpose of this research is to provide a theoretical ﬁamework for

understanding the learning and modeling of abusive behavior as it relates to the link

between past child abuse experiences and subsequent delinquent activity. In this chapter,

the ﬁndings of the data analysis as they relate to the research questions are presented.

The research questions are as follows:

1)
2)

3)

4)

5)

6)

7)

8)

Is there a relationship between prior child abuse and subsequent delinquency?
Is there a relationship between the form of child abuse the adolescent
experienced and the type of subsequent delinquency?

Are those children who have been physically abused more likely to commit
violent, aggressive behavior than those children who have not been physically
abused?

Are those children who have been sexually abused more likely to commit
sexual crimes than those children who have not been sexually abused?

Are there other delinquent or deviant responses to either prior physical or
sexual abuse?

Are there protective factors which may prevent an abused child from acting out
through delinquency?

Is the alleged relationship between prior child abuse and subsequent
delinquency affected by demographic characteristics such as race and age of
onset of abuse?

Is the alleged relationship between prior child abuse and subsequent
delinquency affected by the ﬁ'equency and intensity of the abuse?

54

55

The writer examined the data generated ﬁom the study using univariate, and bivariate
descriptive statistics. No inferential statistical analyses were conducted due to the lack of
random sampling. The researcher had to use the entire institutional population due to an
unsuccessful attempt to obtain a non-institutional, non-delinquent comparison group.
According to University guidelines on research ethics in any research involving juveniles,
the researcher must obtain prior written parental consent and prior verbal assent by the
juvenile. One hundred and forty-eight delinquents were contacted at three Lower
Michigan juvenile facilities. Advance written notice was sent to parents together with a
request for written parental consent. To participate in this study, each potential
respondent had to submit prior verbal assent. The total number of actual respondents was
35, which is a 23.6% response rate. The researcher had diﬂiculty obtaining parental
consent partly because of the relationship between parent and prospective institutionalized
respondent, partly because of fear that negative information may be divulged, and partly
because of the lack of compensation for parental consent. The researcher acknowledges
that having to obtain parental consent contributed to the low response rate.

In the next pages, the writer will ﬁrst examine univariate demographic data
presented in frequency tables and graphs. This will be followed by the presentation of
bivariate data on the physical abuse and sexual abuse variables. Finally, ﬁndings on the
respondents' delinquent involvement will be presented and examined. These ﬁndings will
help determine if there is a relationship between prior child abuse and subsequent
delinquency. The researcher will also examine the effects of age of onset of abuse, race,
frequency, and intensity of the abuse on the alleged relationship between prior child abuse
and subsequent delinquency.

Dembmhic Characterisﬁss

In total, forty parents provided consent for their children to participate in the

study. During the data collection procedure ﬁve potential respondents refused to give

56
verbal assent for participation in the study. Therefore, the total research population was

35. All respondents were male and court mandated to a juvenile treatment Ecility in
Lower Michigan.

Table 1: Ersgbsngy Distribution of Respondsnts' Age

 

Late: f 2/2
12 1 2.9
13 1 2.9
14 4 11.4
15 s 14.3
16 18 51.4
17 6 17.1
TOTAL 35 100

Table 1 presents the age distribution of the respondents. The respondents' ages
ranged from 12 to 17 years old, with an average age of 15.6 years of age.
However, 68.5% of these respondents were in the older age bracket of 16 and 17
years. The majority of the respondents, that is 51.4% , were 16 years of age.
Figure 1 is a bar graph display that also illustrates the ages of respondents. Figure

1 shows the inequality of ages within the research population.

57

1”!

OOInBOﬂW.‘

 

 

X8 Respondents'

Figure 1: Bar Graph Display of Respondents’ Age

Table 2 presents the race of the respondents. Sixteen of the respondents (45.7%)
were White, 12 (34.2%) were Black, 2 (5.7%) were Hispanic, 1 (2.9%) was a Native
American, and 4 (11.4%) indicated "other" as a response.

Table 2: Ersqusng Digributibn bf Rsspondents' Rag

 

LMQE f °A
White 16 45.7
Black 12 34.3
Hispanic 2 5.7
Native American 1 2.9
Other 4 1 1.4
TOTAL 35 100

58

L I

l

at
3888883888
can an an

COIQSOGH.‘

  

 

 

°
.1

White Black Hispanic Native Other
American

Figure 2: Bar Graph of Respondents' Race

Table 2 and Figure 2 present the race of the respondents. In total, sixteen of the
respondents (45.7%) were White, 12 (34.2%) were Black, 2 (5.7%) were Hispanic, 1
(2.9%) was a Native American, and 4 (11.4%) indicated ”other" as a response. According
to the 1990 Michigan Census, the current Michigan population is about 81.4% White,
13.9% Black, 2.2% Hispanic, and .6% Native American. Comparison between the
research data and census data shows that compared to the respective racial representation
in the general population, there is a disproportionate representation of races in the three
Ecilities. For example, Whites are underrepresented by 35.7% while Black youths are
overrepresented by 20.3% and Hispanic youths are overrepresented by 3.5%. This leads
the researcher to conclude that there is a disproportionate representation of race according
to present day population trends. This ﬁnding of disproportionate representation is
consistent with ﬁndings by other researchers such as Sabil (1989), Seidman (1978), and

Velde 1977. These researchers concluded that there is a disproportionate number of

59

minority individuals, especially Black individuals, who are incarcerated or committed to

institutions for delinquent behavior.

Table 3: Pribr Plagment Histoty

 

X f f/_0

yes 23 65.7
no 12 34.3
TOTAL 35 100

I

Table 3 is a ﬁ'equency distribution of prior placement at a juvenile facility other
than the current placement. The above table shows that 65.7% of the respondents had
prior placement history while 34.3% of the respondents had no such history. In summary,
a majority of these respondents had multiple placement history, ranging ﬁ'om 1 to 9 prior
placements. Multiple placements usually occur as a result either of severe behavioral
problems that required transfer to more secure facilities or subsequent delinquent acts
committed by the juvenile which required further placement. This ﬁnding suggests that a
majority (65.7%) of the respondents suffer ﬁ'om serious behavioral or criminal problems,

which have resulted in multiple placements.

Table 4: Frequensy Distribution of Respondents’ Pgents’ Mmtal Stags

 

2; Marital Status f 319
Married 11 31.4
Separated 6 17.1
Divorced 13 37.1
Ngyer; Married 5 14.3
TOTAL 35 999*

'* due to rounding the result 99.9%

60

Table 4 presents the marital status of the respondents' parents. Thus, 31.4% of the
parents were married, while 17.1% and 37.1% respectively were separated or divorced,
and 14.3% of the parents were never married. This ﬁnding shows that over half(54.2%)
of the respondents were raised in homes whose parents did not have a stable marriage
which resulted in divorce or separation. The following bar graph, Figure 3, also displays

the marital status of the respondents’ parents.

CODnSOGHO'.

 

 

Figure 3: Bar Graph Display of Marital Status of Respondents' Parents

In summary, all respondents (35) for this research were male adolescents between
the ages of 12 to 17 with an average age of 15.6 years of age. From the ﬁndings 45.7% of
the respondents are White, 34.3% of the respondents were Black, 5.7% were Hispanic,
and 2.9% were Native American. These ﬁgures indicate a disproportionality among the

races. Very indicative is the ﬁnding that 65.7% of the respondents had multiple placement

61

history ranging from 1 to 9 prior placements. This ﬁnding suggests that over half of the
research population are serious offenders or have behavior problems requiring multiple
placements. While 31.4 % of the respondents came from two parent married Emilies,
over half (54.2%) of the respondents came from divorced or separated Emilies. In
summary, the ﬁndings indicate that over half of the respondents were raised in divorced or

separated families and have a history of multiple institutional placements.

Physical sbuse

Speciﬁcally, the research questions focus on the presence of prior physical or
sexual abuse experiences. The researcher will ﬁrst present the ﬁndings on prior physical
child abuse experiences followed by the presentation of the ﬁndings on prior child sexual
abuse experiences. Prior physical abuse experience is deﬁned for this research as: ”the
exposure to physical abuse by either observation or direct experience including the
intentional and non accidental inﬂiction of physical harm or injury to a child through
severe spanking, hitting, slapping, or beating with or without an object.”

For the purpose of this research, the researcher used two levels of physical abuse
operationalizations. The ﬁrst level is abuse that occurs under the guise of discipline and is
accidental, meaning that the original intention was discipline. The second level is non-
accidental and the original intention was to hurt the child. The term discipline was used
because it is a word that is easily recognized by adolescents. The researcher surmised that
because of antecedent conditions such as drinking, drugs, family ﬁ'ustrations, work
stressors, or cultural norms, parents may react to the point of becoming abusive to their
children even though their original intent was to discipline the child. The researcher was
able to distinguish between pure discipline and abusive discipline by asking the
respondents to indicate if they were ever injured during the course of being disciplined.
The researcher then used injury during a disciplinary action as an indicator of physical

abuse. The respondents were also asked speciﬁc questions about physical injury.

62

Level I Discipline

Table 5: Rssmndsnts wbb wers disciplined

 

X f 312
yes 35 100
do 0 0
TOTAL 35 100

Initially, the researcher investigated discipline among the respondents. Every
respondent felt that they had been disciplined during their lifetime. Typically, an individual
who cares about a younger child will respond with discipline to correct a misbehavior.
The researcher was less concerned with pure discipline cases than with cases of injury
during the course of discipline. However, before investigating the number of injuries that
occurred during the course of discipline, the research will present some characteristics of
the respondents' discipline. These characteristics include the person who disciplined, the

reasons for , and the frequency of the discipline.

Table 6: Frmusng Digribbtion bf Individuals who Disciplined Respondents

)_(, Individuals f 94
biological parents 25 71.4
adoptive or step parents 13 37.1
mother only 2 5.7
father only 0 0.0
step parent only 7 20.0
foster parent 2 5.7
other relative 17 48.5

63

Table 6 describes the individual who usually disciplined the respondents. Since,
the respondents were allowed to identify diﬂ‘erent individuals, the total number of
individuals is greater than .35. The ﬁndings in Table 6 show that 71.4 % of the
respondents were disciplined by their biological parents while 37.1% were disciplined by
adoptive or step parents. The fact that 71.4% of the respondents were disciplined by a
biological parent leads the researcher to surmise that discipline is an intrafamilial act.
However, only a small ﬁ'action of respondents (5.7%) indicated that they had been
disciplined only by their mother and none of the respondents indicated that they had been
disciplined only by their Ether. Furthermore, 20% of the respondents indicated that they
were disciplined only by their step parent. This higher percentage for step parents may be
due to a step parent's lack of attachment to a child as compared with a biological parent's
attachment with that child. Furthermore, almost half of the population, 48.5%, indicated
that they were disciplined by another relative such as aunt, uncle, or grandparent. This
suggests that either parent of the respondents allowed extended family members to
intervene with their child or that these respondents lived with family members other than

their parents.

Table 7: Fregbsnsy Distribution bf Reasons Rsspondents were Disciplined by their

Parents
Liaising f 912
late for curfew 21 60
didn't complete chores 23 65.7
wet the bed 2 5.7
talked back to adult ' 28 80
other 7 20

Table 7 displays some of the reasons respondents gave for being disciplined by

their parent. The respondents were allowed to select as many reasons that applied.

Therefore the total ﬁ'equency of responses is greater than 35. Eighty percent (80%) of the
respondents indicated that they were disciplined because they talked back to an adult,
65.7% indicated that they had been disciplined for failure to complete their chores. Sixty
percent (60%) of the respondents indicated that they were disciplined because of being
late for curfew, while 5.7% indicated that they were disciplined for wetting the bed.
Other factors that were listed by the respondents included getting into a ﬁght, lying to a
parent, stealing, and smoking cigarettes.

The above actions are situations that can cause the parent to react to the
respondent's behavior. This, creates an illusion to the parent and child that the parent is
truly disciplining the child. However, it is the parents' overreaction to this misbehavior

which inﬂicts an injury to the child that is being investigated.

Table 8: Ermpsngt Distribptipp bf Parentﬂ Methgﬁs prissiplins

ZLMQEQCI: f °_/9
spanking 28 80
hit by ﬁsts 13 37.1
hit by an object 14 40.0
loss of privileges 17 48.5
other 4 h 11.4

Table 8 and Figure 4 display the methods of discipline the respondents were
exposed to as children. Similar to the reasons for discipline, the respondents were allowed
to select as many methods of discipline that applied. Therefore, the total frequency of
methods is greater than 35. Spanking and the loss of privileges were the two most

common disciplinary methods reported by respondents.

65

CDIRSOC'O'I.‘

 

 

89mins l'ltby I-lltby Lost Other
ﬁsts object privlega

Figure 4: Bar Graph Display of Methods of Discipline

Table 8 and Figure 5 display the methods of discipline the respondents were
exposed to as children. Eighty percent (80%) of the respondents reported to have been
spanked while 48.5% of the respondents replied that they had privileges withheld. The
ﬁnal two listed methods of discipline, being hit with ﬁsts and being hit with an object, are
violent and aggressive actions by the parent. Almost half, or 40%, of the respondents
indicated that they had been hit with an object and 37.1% of the respondents indicated that
they had been hit with ﬁsts. Both of these types of disciplines are physical actions against
a child. Considering the situations in which these actions occurred, the child may learn to
associate violent physical response as an appropriate form of response to a situation of

confrontation.

66

Table 9: Frmueng Distribution of Fregueng of Discipline

W t E/a
once a week 3 8.6
twice a week 13 37.1
once a month 4 11.4
twice a month ' 8 22.9
other 5 17.1
no response 1 2.9

The above table presents the frequency of discipline that the respondents were
exposed to as children. The highest frequency was reported by reSpondents who were
disciplined twice a week (37.1%) followed by respondents who reported being disciplined
twice a month (22.9%). However, the weekly frequencies combined show that 45.7% of
the respondents were subjected to weekly discipline. Furthermore, 11.4% of the
respondents indicated being disciplined once a month. There were 5 respondents who

selected the ”other” category and explained this to occur ”all the time."

Table 10: Rspeivﬁ Injug as s Rssplt beissipline

 

mu 1‘ %
yes 1 1 31.4
no 24 68.6
TOTAL 35 100

Table 10 shows that 31.4% of the respondents indicated that they suﬁ‘ered various
types of injuries ﬁ'om the discipline experience. These injuries include receiving bruises,
stitches, a black eye, or being knocked unconscious as evidenced ﬁ'om Table 11. The type
of resulting injury is related to the methods of discipline. As previously reported, 37.1%
of the respondents were hit with ﬁsts while 40% of the respondents indicated that they

67

were hit with an object. Such disciplinary methods undoubtedly have potential for

inﬂicting different degrees of injuries on a young child.

Table 11: Fr u Di ri tion rT of In' R ceiv

lunar! t“ E/a
bruises 8 72.7
stitches 3 27.2
broken bones 0 0.0
black eye 5 45.4
knocked unconscious 2 18.2
other 2 18.2

Table 11 presents the types of injuries the respondents incurred in the course of being
disciplined. The respondents were allowed to select a combination or multiple responses
for injuries received. Figure 5 displays the percentages of respondents who were injured

during the course of discipline

”-4 72.7

OOIHSOﬂ‘.‘

 

 

Bruins Stitches Black eye Knocked Other
uneonaﬁous

Figure 5: Bar Graph Display of Injuries

68

Table 11 and Figure 5 display the injuries received by respondents ﬁom a
disciplinary action. From the eleven respondents who indicated injuries while being
disciplined, 72.7% (8) had received bruises and 45.4% (5) received a black eye.
Furthermore, 27.2% of the respondents received stitches and 18.2% (2) indicated that
they had been knocked unconscious. Both individuals who replied ”other” indicated that
they had teeth knocked out. These injuries are not only serious but they can leave severe
emotional scars, especially when administered by an adult whom the child looks up to.
Some of the discipline that the respondents were exposed to by an adult are violent and
aggressive. When a child is exposed to this type of violence and aggression, the parent or
adult teaches and reinforces that violence and aggression are acceptable forms of behavior.
Therefore, the child learns to behave aggressively.

In reviewing the ﬁndings on discipline, it may be noted that all the respondents
(35) indicated that they had been disciplined as a child. The most frequent disciplinary
methods included withholding privileges, grounding, spankings, hitting with ﬁsts, and
hitting with an object. In fact, 40% of the respondents were hit with an object and 37.1%
of the respondents were hit with ﬁsts by an adult. Furthermore, 31.4% of the sample
indicated that they had been injured as a result of being disciplined. These experiences are
traumatic for a child and they can shape what the child perceives as acceptable or

unacceptable behavior.

69

Level 11 Physical "Harm”

The ﬁrst operational level of physical abuse was discussed in the previous section.
The focus was an action by the parent that initially starts as discipline, but which resulted
in unintentional or accidental injury to the respondent. In the following section, the
researcher will examine the second operational level of physical abuse. The second level
of abuse focuses on physically abusive or harmﬁil actions by the parents independent of
the child's behavior. These situations may have occurred due to the perpetrator being a
violent person, wanting control over the child, being drunk or under the inﬂuence of
drugs, or reacting to a stressﬁil situation. Whatever the reason for the adult's action, the
respondents were asked to identify if they had ever been physically "banned” by an adult
and, if so, what happened and who committed this action. The researcher used the term
"harm" in place of ”abuse” because believing that had the term "abuse" been used then the

respondents would have been more reluctant to identify this as having occurred to them.

Table 12: Frequeng Distribution of Respondents who were Physically Harmed

 

f %
yes 21 60
no 14 40
TOTAL 35 100

An overwhelming 60% of the respondents (21) responded that they had been
physically ”banned by an adult”. The high percentage of respondents (60%) who were
physically ”harmed” indicates that a substantial number of respondents in this study were
exposed to situations that were outright abusive. More enlightening than the numbers
subjected to this abusive experience is the identity of the perpetrator as shown in Tables

13 through 15.

70

Table 13: Frmueng Distribution of Fathors as Pe_roetrators

X, injorios f 2/2

spanked 15 100
beaten 5 33.3
punched 5 33.3
hit with an object 7 46.6

Table 13 presents ﬁndings on the father as the perpetrator of physical abuse. Out
of 21 respondents who were physically "harmed", 15 respondents, or 71.4%, indicated
that they had been "banned" by their father. Surprisingly, this frequency of 15 was the
same for physical harm administered by a mother. This suggests that among these
respondents, physical abuse is an intra-familial act that is perpetrated with equal ﬁequency
by either parent. Table 13 also presents the abusive actions the father administered to the
respondents. All 15 respondents who indicated that they were physically ”harmed"
responded that they had been spanked by their father. While 33.3% indicated that they
had been beaten, 33.3% reported that they had been punched, and 46.6% indicated that
they had been hit with an object.

Table 14: Fraueng Distribution of Mothers a Perpetrators

2; in' . f 940
spanked 15 100
beaten 3 20

punched 2 13.3
hit with an object 4 26.6

Table 14 displays the ﬁndings on the mother as a perpetrator of physical ”harm".
Fifteen respondents (71.4%) indicated that they had been "harmed" by their mother. This

percentage is equal to that of fathers as perpetrator. Table 14 also presents what abusive

71

actions the mothers inﬂicted on the respondents. All 15 respondents indicated that they
had been spanked, while 20% indicated being beaten, 13.3% indicated being punched, and
26.6% as being hit with an object.

In summary, 71.4% indicated that they had been physically ”harmed” by either
their mother or father. The most common form of physical "harm" was spanking followed
by hitting with an object. Also, the respondents' fathers were more likely than mothers to
either hit with an object or punch their child. Comparing the two parents, the fathers were
more likely than mothers to be violent and physical by beating, punching, or hitting with
an object.

Unfortunately, a high frequency of respondents were exposed to physical abuse by
their parents. Parents are supposed to be role models for their children. The parents who
are physically abusive to their children teach them that physical behavior is appropriate.
Overall, 71.4% of those respondents who indicated that they had been physically "banned"
indicated that the perpetrator was either their mother or father. This is a large proportion
of respondents who were exposed to abuse within their immediate family environment.

The types of behavior that these respondents were exposed to as children were
aggressive and violent. The parents punched their children and hit them with objects such
as belts and boards. The theoretical ﬁamework is that a child's parent is expected to be a
role model for the child, teaching the child behavior that is acceptable or unacceptable by
societal standards. The theoretical framework also states that exposure of children to
violent and aggressive behavior may result in an internalization of violent and aggressive
behavior as acceptable and appropriate. It stands to reasons that if a child intemalizes
violent behavior as acceptable behavior then the child is likely to exhibit violent and

aggressive behavior, which may in turn manifest itselfin delinquent acts.

72

Table 15: Fr n itri uti n f her Pe trat rs of Ph sical Harm

 

)_(, individual f 2/3
foster parent 1 4.7
aunt 7 33.3
uncle 4 19
step parent 5 23.8
other 6 28.5
TOTAL 23 65.7

Respondents also indicated that other individuals, such as foster parents, aunts,
uncles, step parents, and other family members inﬂicted physical "harm" on them, as seen
in Table 15. Respondents were able to select as many individuals as applied to them for
this category.

The largest percentage of respondents (33.3%) reported that an aunt inﬂicted
physical "harm”. Uncles and step parents were about equal, 19% and 23.8%, respectively.
Furthermore, 28.5% indicated that other family members, such as grandparents, inﬂicted
physical "harm” while only a small percentage, 4.7, indicated that they were "banned” by

foster parents.

Table 16: Frmoenoy Distribution for Types of "Harm" by Other Pemetrators

z, injuries f ‘Zo
spanking 20 90.9
beaten 7 3 1.8
punched 8 36.3
hit with object 11 50

Table 16 presents the types of abuse inﬂicted by individuals such as aunts, uncles,
foster parents, and step parents. Similar to the injuries inﬂicted by parents, the highest

percentage (90.9%) was for spanking, followed by being beaten (31.8%) and being

73

punched (36.3%). Additionally, 50% had been hit with an object such as a belt or a board
of wood.

In summary, the above ﬁndings show that 60% of the respondents were physically
"harmed” as children. The ﬁndings also show that respondents were most likely exposed
to this abuse by either their father or mother. Seventy-one percent (71.4%) responded
that the ”harm" was inﬂicted by their father or their mother. This suggests the intrafamilial
relationship of the physical abuse and the poor role modeling the parents exhibited. Also
signiﬁcant is the number of respondents who indicate that they were physically hit by an
adult, including being punched or beaten or being hit with an object. Unfortunately, the
respondents may internalize and learn that aggressive behavior is acceptable and

consequently incorporate this behavior into their lifestyles.

74

Ssxual abuse

This research also investigated the relationship between child sexual abuse and
subsequent delinquency. Before presenting the ﬁndings of this relationship, univariate
presentations will be shown on the frequency of sexual abuse. Respondents were asked to
indicate if they had been victims of sexual abuse. Respondents were asked if they had ever
been sexually victimized by an adult or if they had ever been made to perform a sexual act
on an adult. An afﬁrmative response to either situation was coded as prior sexual abuse.
Respondents were also asked to identify the perpetrators, the sexual act that took place, at

what age, and with what duration.

Table 17: Frmuenoy Distribution of Respondents who were Soxugly Victimizoo

 

Autism f 24»
yes 10 28.5
no 23 65.7
W 2 5.7
TOTAL 35 100

Table 17 shows that 28.5% or 10 respondents reported prior sexual abuse

victimization while 65.7% of the respondents reported no such experience.

Table 18: Froguonoy Distribution for Porpetrators of Sexual Abuse

P t t r f ?/_o
uncles 3 27.2
step parent 3 27.2
mother 1 9.0
father 0 0
other 8 72.7

75

From the ten respondents who indicated prior sexual abuse, several combinations
of individuals were indicated as the perpetrator. Therefore, the total number of
perpetrators was higher than the number of victims. The "other” category provided the
highest frequency (8) although there are several individuals who comprise this category,
such as grandfather, ﬁ'iend, neighbor. The next highest ﬁ'equency was for uncles and step
parents with a frequency of 3 each. Surprisingly, none of the respondents indicated their
fathers as perpetrator and only one respondent reported his mother as the aggressor.

An analysis of the respondents' age at the time of abuse revealed an age range of 4
to 13 years with a mean of 8.1 years of age. This ﬁnding suggests that although there are
victims who were abused in their older years (12 and 13), the majority of the victims had
been abused in the earlier years. In fact, well over half or 72.7% of the respondents had
been victimized by the age of 8. Unfortunately, the child is still developing during these
younger years and what results is a misconception of appropriate sexual behavior. These
respondents may learn that sexual behavior is appropriate with little children or that it is an
acceptable way of exerting control or expressing anger.

The respondents were also asked to indicate the duration of the sexual activities in
months and the ﬁ'equency of occurrence. “With regards to duration three respondents or
30%, answered that the abuse occurred only once. However, two respondents answered
that the abuse occurred over 5 and 7 years, respectively. The remainder of respondents
(5) indicated that the abuse occurred for 2 to 6 months. Due to extreme values reported
by respondents the median was calculated. Thus, the median value for duration was 2 to 6

months, suggesting that the sexual abuse did not occur for long periods of time.

76

Table 19: Ermueng Distribution of Frequenoy of Sexual Actions

L ro_u_rrence f °_/o
more than 1 a day 0 0
daily 1 9
weekly 5 45.5
other 4 36.4

With regards to the frequency of occurrence of sexual abuse, two respondents
(18.2%) indicated that the abuse took place either more than once a day or daily, ﬁve
(45.5%) responded that their abuse took place weekly. Those individuals who were
victimized once in their lives (4) indicated ”other” as an answer of frequency. The data
suggest that the respondent either had a one time inappropriate sexual contact or a weekly
recurrence of sexual abuse. Similar to the duration of abuse, reporting the mean ﬁequency
of occurrence would produce a skewed result based upon the four respondents who only
had the action occur once. Therefore, the median frequency of sexual abuse was
computed. The median frequency was weekly.

In summary, the ﬁndings show that 10 respondents responded that they were
sexually victimized. Sexual victimization included either performing a sexual act to an
adult or having an adult perform a sexual act to them. The ﬁndings show that there is an
equal percentage, 27.2%, of uncles and step fathers as perpetrators. The mean age of
onset is 8.1 years with a median value of 2 to 6 months for duration and a median value of
weekly for the frequency occurrence. When a child is exposed to sexual behavior at such
an early age and does not have a clear understanding of it, they are likely to intemalize a
poor conception of sexual behavior, causing the child to be overly sexual, sexually deviant,

or predatory and manipulative towards sex.

77

Delinguenoy Data
All the previously reported ﬁndings relate to the respondents' child abuse

experiences. Prior to investigating the alleged relationship between child abuse and
subsequent delinquency, the researcher will present ﬁndings on the respondent's
delinquency as measured by reported adjudications. An adjudication in the juvenile justice
system means that a juvenile has been found guilty of breaking the law. The respondents
were asked to answer questions to indicate the crimes for which they had been adjudicated
in the past and also crimes for which they were adjudicated at the time of the study. The
respondents were also asked to identify the total number of crimes for which they were

ever adjudicated.

Table 20: Frmuong Distribution of Qrimes for Cprrent Adjudioation

 

mm: f °_/9
murder 2 5.7
criminal assault 16 45.7
robbery 12 34.2
theft 16 45.7
prostitution 5 14.2
tans/ca; 6 17.1
Total 57 162

The above table is a ﬁequency distribution of adjudications for which each
respondents was in placement. The respondents were allowed to report as many
adjudications as applied to them. Therefore, the total number of adjudications is over 35.
The highest frequency of occurrence is 16, which was reported for both criminal assault
and for theft. The second highest frequency, 12, was for robbery. There were also 6
respondents who were adjudicated for rape/csc, 5 respondents for prostitution, and 2

respondents for murder. The following table displays frequencies of prior adjudications.

78

Table 21: Fregpong Distribution of Crimes for Prior Adjudications

 

lama Grim f 340
murder 1 2.8
criminal assault 7 20
theft 18 51.4
prostitution 0 0
rape/csc 3 8.5
Quanta! 21 60
Total 50 142.7

Again, respondents were able to report as many adjduciations as applied to them.
Therefore, the total number of adjduciations is higher than 35. The highest frequency, 21,
is for truancy, while the second highest, 18, is for theft, followed by criminal assault with a
ﬁ'equency of 7. There are also 3 respondents adjudicated for rape/csc and 1 for murder.

In this study, the crimes of murder, rape, assault, battery, robbery, and arson were
grouped as violent, aggressive class of crimes while truancy, prostitutiOn, and theft were
classiﬁed as non-violent crimes. Subsequent analysis of the ﬁndings will explore the
relationship between child abuse and subsequent delinquency. The highest frequency ﬁ'om
Table 20 and 21 combined is 60%. This frequency was for truancy, which is a non-violent
crime. However, the above tables also show that 47% of the adjudicated crimes were
violent in nature. The next highest frequency of occurrence, 34%, is for theft. Generally,
these ﬁndings are consistent with the delinquency literature that states that juveniles
commit more crimes of theﬁ than any other crime to attract attention towards themselves

(Gillespie et a1, 1977).

The Relationship Betweon Child Abuse and Delinqueng

The purpose of this research was to investigate the relationship between prior child

abuse and subsequent delinquency. For the purpose of this research, those respondents

79

who were injured during the course of being disciplined and those respondents who
responded that they had been physically ”banned” by an adult were grouped as physically
abused respondents. Furthermore, those respondents who indicated that they had
performed inappropriate sexual acts on adults as well as children or adults had performed
inappropriate sexual acts to the respondent were classiﬁed as sexually victimized
respondents. The groups of respondents who were physically abused and those
respondents who were sexually victimized were combined to form the group of abuse
respondents. In the following pages, the researcher will present the results of the analysis

of the alleged relationship between prior child abuse and subsequent delinquency.

Research Question 1: Is there a relationship between child abuse and subsequent
delinquency?

Using univariate and bivariate analysis, the researcher examined the relationship
between being abuse victims and having committed violent or non-violent crimes. Violent
crimes were previously deﬁned as murder, rape, assault, battery, robbery, and arson. Non-
violent crimes were deﬁned as truancy, prostitution, theft, incorrigibility, and drug abuse.
It is also important to recall that this research population is composed of adolescents who
are institutionalized and have been adjudicated for at least one criminal act.
Unfortunately, the lack of a non-delinquent sample or comparison group limits the
investigation of the effects of child abuse upon subsequent behavior. The researcher will
ﬁrst present the ﬁndings on respondents who are victims of either physical or sexual

abuse, and then their relation to delinquent activities.

80

Table 22: Frmbongg Distribution of Child Abug Vigims

 

hi1 11 V‘ im 1‘ %
yes 23 65.7
no 12 34.3
TOTAL 35 100

Table 22 shows that 23 respondents, 65.7%, were victims of either physical or
sexual abuse, whereas 12 respondents, or 34.3%, had no prior abuse experience. When
comparing the percentages of abuse and non-abuse victims, there is a diﬁ‘erence of 31.4%
higher for child abuse victims. This factor coupled with the understanding that all the
respondents were adjudicated for at least one crime indicates that there is a higher
percentage of child abuse victims who are delinquent than non abused and delinquent
respondents. Therefore, the difference of percentages between child abuse victims and
non victims shows that there is a higher percentage of 31.4% for child abuse victims as

compared to non-abuse victims.

Table 23: Cross tabulation of Child Abuse and Type of Subsmuent Delinguonoy

type of crime
child abuse violent non - violent totals
yes 16 7 (23) 65.7%
no 10 2 (12) 34.3%
totals (26) 74.3% (9) 25.7% (3 5) 100%

Table 23 is a cross tabulation of physical or sexual abuse with violent or non-

violent delinquency. The above table shows that 65.7% were abused while 34.3% (12) of

81

the respondents were not abused. Also, it shows that 26, or 74.3%, of the respondents
were adjudicated for violent crimes while, 9 or 25.7%, of the respondents were
adjudicated for non-violent crimes. These percentages show that 65.7% were victims of
child abuse and 74.3% of the population were involved in violent crimes.

In the previous section, it was explained that 65.7% of the research population was
abused and 34.3% was not abused. Combined with the factor that all the respondents
were adjudicated for a crime, the differences of percentages was calculated to determine
the extent of the relationship. The differences of percentages between child abuse victims
and non-victims indicate that abuse victims have a 31.4% higher rate to engage in
delinquent behavior.

Statistical analysis investigates the relationship between child abuse and subsequent
delinquency to determine if these variables are dependent upon each other. The Chi
square statistic was computed for the above cross tabulation of child abuse and type of
subsequent delinquency. Fisher’s Exact Test was employed due to one of the cells being
less than the required value of ﬁve. The probability level for a two tail test is .685. Given
this probability level, the relationship between child abuse and subsequent delinquency is
not statistically signiﬁcantly, thus suggesting a lack of dependence between child abuse
and type of subsequent delinquency. However, Chi square is affected by the sample size,
therefore, had the sample size been larger, the level of probability may have decreased
causing the relationship to be statistically signiﬁcant. Thus a Phi value was computed to
determine the strength of the relationship between child abuse and type of subsequent
delinquency. The Phi value is .116, thus establishing that this relationship is a weak
relationship within this sample. Further analysis of the relationship between child abuse
and subsequent delinquency will examine control variables such as race, age of onset for
abuse, frequency of abuse, intensity of abuse, and protective factors to determine if these

control variables have any effect upon the relationship.

82

The researcher determined that there is a relationship between child abuse and
delinquency when comparing the difference of percentages between child abuse victims’
and non-victims’ propensities to engage in delinquent behavior. Abuse victims have a
31.4% higher propensity for delinquency than non-abuse victims. This ﬁnding is
consistent with the ﬁndings of Alfaro (1981), Fagan and Wexler ( 1987), Gillespie et al.
(1977), Kratcoski and Kratcoski (1982), McCord (1983), Owen and Straus (1975), and
Widom (1989). These researchers concluded that there is a relationship between child
abuse and subsequent delinquency, that is, these two variables are dependent upon one
another. However, the ﬁndings of this study indicate an independence between child
abuse and type of subsequent delinquency, either violent or non-violent. This ﬁnding is
inconsistent with the ﬁndings of Alfaro (1981) and Owen and Straus (1975). These

researchers concluded that child abuse victims were more violent than non-abuse victims.

Research Question 7: Is the alleged relationship between child abuse and subsequent
delinquency affected by demographic characteristics such as race and age of onset of
abuse?

Research question 7 focuses on the effect of race on delinquency. Although
previous analysis indicated that child abuse and type of subsequent delinquency are
independent, the researcher surmised that certain control variables such as race and age of
onset may affect this relationship. The following section will examine the effects of race
on this relationship.

As previously stated, there is a disproportionate representation of races within the
three facilities. Whites constitute 45.7% of the research pOpulation, while Blacks are
34.3%, Hispanics are 5.7%, Native Americans are 2.9%, and "others” are 11.4%. These
percentages of the population compared with the current population ﬁom the 1990
Michigan Census indicates an under representation of Whites by 35.7% and an over

representation of Blacks and Hispanics by 20.3% and 3.5%, respectively. Analysis of the

83

"respondents' race may somehow aﬁ‘ect the relationship between child abuse and the type

of delinquency.
Table 24: gross tsbplation of Child Abuse and Type of Subsguent Delinguongg for
Whito Rosmndonts
‘ type of crime
child abuse violent non - violent totals
yes 8 5 (13) 81.3%
no 3 o (3) 18.8%
totals (11) 68.8% (5) 31.3% (16) 100%

Out of the 16 White respondents, 13 (81 .3%) were physically or sexually abused
while 18.8% responded that they had not been abused. The above cross tabulation also
shows that 68.8% of the White respondents were adjudicated for a violent crime and only
5 White respondents (31.3%) were adjudicated for a non-violent oﬁ‘ense. To examine the
dependence between child abuse and delinquency while controlling for White race,
Fischer‘s Exact Test was used due to one cell of the cross tabulation equaling zero. The
level of probability using Fischer’s two tailed test is .508 which determines that the
relationship is not statistically signiﬁcant. A Phi value was computed to determine the
strength of the relationship although the relationship is not statistically signiﬁcant. The Phi
value is .323 which indicates that there is some degree of strength between the variables of
child abuse and delinquency when controlling for the White race. Although when
controlling for White race the relationship between child abuse and delinquency is not
statistically signiﬁcant, there does appear to be some strength between them based upon
the phi value of .323. The fact that the relationship is not statistically signiﬁcant may be

affected by the relatively small sample size.

84

 

Table 25:

type of crime
child abuse violent non - violent totals
yes 5 0 (5) 41.7%
no 5 2 (7) 58.3%
totals (10) 83.3% (2) 16.7% (12) 100%

The above cross tabulation examines the relationship between child abuse and the
type of subsequent delinquency for Black respondents. In total, there were 12 Black
respondents (34.3%). The above table shows that 5 Black respondents (41.7%)
responded that they were abused while 7 (58.3%) were not abused. The table also shows
that 10 Black respondents were adjudicated for violent crimes and 5 of these violent
respondents had been abused and 5 had not been abused. Again Fischer's Exact Test was
employed due to cells of the cross tabulation being less than the required number of ﬁve.
The probability level calculated by using Fischer's two tailed exact test is .469 which
determines that the relationship is not statistically signiﬁcant. The fact that the relationship
is not statistically signiﬁcant may be due to the sample size as it is proven that Chi square
is aﬂ‘ected by the sample size. Therefore, the probability level may have decreased had the
sample size been larger. A Phi value was calculated to determine the strength of the
relationship although it is not statistically signiﬁcant. The Phi value is .377 indicating that
there is some relative strength between child abuse and delinquency when controlling for
Black race. This Phi value of .377 is slightly higher than the .323 Phi value calculated for
the White race. Therefore, there is a slightly stronger relationship based upon Phi values

85

between child abuse and subsequent delinquency for Black respondents than for White

respondents.
Table26: . .~ 2.» _l:_ i of 'lo ~. .- z. . T10' of 'I'n! en D-«lin - for
" " n n
type of crime
child abuse violent non - violent totals
yes 4 2 (6) 85.7%
no 1 0 (1) 14.3%
totals (5) 71.4% (2) 28.6% (7) 100%

Table 26 is a cross tabulation between child abuse and type of subsequent
delinquency for the ”other” category of race. As previously stated, Hispanics and Native
Americans were combined into this category for the purpose of analysis. There is a total
of 7 respondents who were either Hispanic, Native American, or "other”. Six of the 7
(85.7%) indicated that they were abused. Investigating the type of subsequent
delinquency, 5 respondents (71.4%) were adjudicated for a violent oﬂ‘ense while 2
(28.6%) were adjudicated for a non-violent oﬂ‘ense. Again, Fischer’s two tail exact test
was used to calculate dependence between child abuse and delinquency when controlling
for ”other" race. The level of probability from Fisher test is 1.00 which indicates that the
relationship is not statistically signiﬁcant, in fact, they are independent. A Phi value was
also calculated to determine the strength of the relationship between the two variables.
The Phi value is .258 which determines a very weak relationship between the variables of
child abuse and delinquency when controlling for ”other" race. The Phi value of .258 for
”other” race is the smallest Phi value in comparison to White and Black.

86

Therefore, using race as a control variable has not shown dependence between
child abuse and the type of subsequent delinquency at any level. However, the Phi value
of .323 and .377 for the White and Black races, respectively, are comparable. These Phi
values indicate that there is some degree of relationship between child abuse and
subsequent delinquency when controlling for race for this sample, although the
relationship is not statistically signiﬁcant. The lack of statistical signiﬁcance may be due to
the small sample size. It is well known that Chi square is affected by the sample size.
Thus had the sample been larger the ﬁndings may have shown a statistically signiﬁcant
relationship between child abuse and delinquency.

The above ﬁndings suggest that race does have a slight effect upon the relationship
between child abuse and delinquency. This ﬁnding is inconsistent with the ﬁndings of
Widom (1989). She concluded that there is a higher propensity among the Black race to
be abused and subsequently violent. However, this research shows comparable values for
both White and Black respondents when investigating the relationship between child abuse
and subsequent delinquency.

Research Question 7 focused on the effect of age on subsequent delinquency.
Although the present data shows a diﬂ‘erence of percentages, indicating a relationship
between child abuse and subsequent delinquency, the researcher introduced other control
variables to examine any possible changes in the original relationship. The next section
will investigate the eﬂ‘ects that the age of onset for the abuse, either physical or sexual,
may have upon the relationship between prior child abuse and subsequent delinquency.

It was assumed that the age of onset for the abuse could affect the relationship.
This assumption is derived from theoretical ﬁ’arnework which states that the earlier the
exposure of a child to certain behavior, for example abusive behavior, then the higher the
likelihood of internalizing and emulating that behavior. Although the researcher found no
statistical relationship between child abuse and type of subsequent delinquency, the

87

researcher will still explore whether introducing variables such as age of onset will alter

this picture.

Research Question 7: Is the alleged relationship between child abuse and subsequent
delinquency affected by demographic characteristics such as age of onset of abuse?

Table 27:

 

age of onset
crime 1 thru 6 7 and older totals
violent 8 7 (15) 68.2%
non - violent 4 3 (7) 31.8%
totals (12) 54.5% (10) 45.5% (22) 100%

Table 27 presents the ﬁndings of the age of onset for abuse and the type of
subsequent delinquency. In total, there were 23 respondents who reported being abused.
There are a total of 22 respondents who responded to this question. One respondent who
was abused did not indicate the age at which this began. Table 27 shows that 12
respondents (54.4%) were abused between the age of 1 and 6 while 10 (45.5%) were
abused alter the age of 6. The table also shows that 15 respondents (68.2%) were
adjudicated for violent crimes while 7 (31.8%) were adjudicated for non-violent crimes.
Of those 15 respondents who were violent delinquents, 8 (36.4%) were abused between 1
and 6 while 7 (31.8%) were abused after the age of 6. Of the 7 non-violent delinquents, 4
(18.2%) were abused between 1 and 6 years of age, while 3 (13.6%) were abused after 6

years of age.

88

Due to two of the cells being less than the required ﬁve respondents, Fischer's
Exact test was used to determine if there is dependence between the age of onset for
abuse and the type of delinquency. The probability level of F ischer's Exact Test using the
two tail method is 1.00 which indicates that age of onset for abuse and type of delinquency
are not dependent upon each other, in fact, they are independent of each other. Chi square
results are affected by the sample size thus it is likely that had the sample size been larger
than age of onset may have affected the original relationship. The Phi statistic was
calculated to examine the strength of the relationship between child abuse and
delinquency. The calculated Phi value is .035 indicating that the relationship is almost
non-existent.

Although the researcher determined that there is no relationship between child
abuse and type of subsequent delinquency, certain control variables have been examined to
determine if these variables may have an effect upon the original relationship. These
control variables have not changed the original lack of relationship between child abuse
and type of subsequent delinquency. In the following pages, the researcher will examine
the likely impact of frequency of prior abuse on the original relationship between child

abuse and type of subsequent delinquency.

89

Research Question 8: Is the alleged relationship between prior child abuse and subsequent
delinquency aﬂ‘ected by the frequency of the abuse?

Table28: 05.2.0.2. . ash-rm . 1- -._,- 3-1! h'Tfu'O 43.9-11
12:-Jigsaws:
Frequency of abuse
crime more times than weekly weekly or less totals
violent 11 5 (16) 69.5%
non - violent 5 2 (7) 30.5%
totals ( 16) 69.6% (7) 30.4% (23) 100%

The above table displays the ﬁndings on relationship of frequency of abuse and the
type of subsequent delinquency. In total, there are 23 respondents who were abused and
who indicated the ﬁ'equency of occurrence for the abuse. The respondents were combined
into two categories. These categories included the respondents who were abused ”more
times than weekly" and respondents who were abused "weekly or less frequently”.
Sixteen respondents (69.6%) were abused ”more than weekly", whereas 7 (30.4%) were
abused ”weekly or less”. From the 16 respondents who had the abuse occur "more times
than weekly”, 11 (68.8%) were adjudicated for a violent crime while 5 (31.3%) were
adjudicated for a non-violent offense. Of the 7 respondents who were abused ”weekly or
less ﬁ’equent", 5 (71.4%) were adjudicated for a violent crime and 2 (28.6%) were
adjudicated for a non-violent crime.

Chi square statistic was computed to determine dependence between the ﬁ'equency
of occurrence of the abuse and the type of subsequent delinquency. Due to one of the
cells being less than the required ﬁve, Fischer‘s Exact Test was calculated to determine

9O

dependence between ﬁequency of abuse and type of delinquency. Using the two tail result
of Fischer's Exact Test the probability level is 1.00, which indicates that the relationship
between frequency of abuse and delinquency is not statistically signiﬁcant and the two
variables are not dependent upon each other. This result may in part be due to the small
sample size. Therefore, a Phi value was calculated to measure the degree of strength
between these two variables. The Phi value is .026 which indicates that there is not much
strength between the frequency of abuse and delinquency. This Phi value is lower than the
original Phi value indicating that ﬁequency of abuse does not affect the relationship of
child abuse and subsequent delinquency.

The theoretical framework indicates that the learning of criminal or abusive
behaviors is affected by the frequency of occurrence. However, the above ﬁnding that
ﬁ'equency of occurrence of abuse does not affect the relationship between child abuse and
delinquency for this sample is contrary to the theoretical framework. This ﬁnding is also
inconsistent with past research conducted by Fagan and Wexler (1987) and Gillespie et al.
(1977). These researchers concluded that the ﬁequency of occurrence increases the

acceptance of the abusive behavior and its manifestation through delinquency.

Research Question 8: Is the alleged relationship between prior child abuse and subsequent

delinquency affected by the intensity of the abuse?

Research question 8 also asks if the intensity of the child abuse affects the
relationship between child abuse and type of subsequent delinquency. For the purpose of
this research, the intensity of the abuse was determined as the number of perpetrators of
abuse the adolescent was exposed to. Although the relationship between child abuse and
subsequent delinquency for this sample is not statistically signiﬁcant, the researcher will

present analysis of the intensity of the abuse to determine the effects that this variable may

91

have upon the relationship. In the following section, the results of analysis for intensity of
child abuse are presented.

 

Table 29: WW
Lﬂurnher t” 29

1 4 17.3

2 9 39.1

3 5 21.7

4 3 13.0

5 2 as

TOTAL 23 100

The above frequency distribution displays the number of abusers for each abused
respondent. The range of abusers is 1 to 5 abusers with a modal value of 2 abusers.
Similar to the age of onset for abuse and frequency of abuse, the intensity of abuse
predisposes that the respondent was abused. Therefore, the non-abused cases are dropped
ﬁom analysis. In order to examine if intensity has any effect, the researcher prepared a
cross tabulation of intensity of the child abuse and delinquency, while computing Chi

square and Phi for the cross tabulation.

92

Table30: 0' altar f 0m! n Dlin “M a 0 n “11 f 11 hi] ‘2 1'

intensity of abuse
crime lor 2 abusers 3 or more abusers totals
violent 8 8 (16) 69.5%
non - violent 5 2 (7) 30.5%
totals (13) 56.5% (10) 43.4% (23) 100%

Table 30 is a cross tabulation of type of subsequent delinquency and the intensity
of the child abuse for the 23 respondents in the abuse sub-sample. For the purpose of this
analysis, the respondents were divided into two categories, 1 and 2 abusers or 3 and more
abusers. There are 13 respondents (56.5%) who were abused by 1 or 2 individuals while
10 (43.4%) were abused by 3 or more abusers. Also from the above table, there are 16
respondents (69.5%) who were adjudicated for a violent crime while 7 (30.5%) were
adjudicated for a non-violent crime. Of the 16 violent delinquents, 8 (34.7%) had 1 or 2
abusers while 8 (34.7%) were abused by 3 or more individuals. Of the 7 non-violent
delinquents, 5 (71.4%) had 1 or 2 abusers while 2 (11.5%) were abused by 3 or more.
Due to one cell being less than ﬁve, Fischer’s Exact test was used to calculate the level of
probability. The level of probability is .652 which indicates that this relationship is not
statistically signiﬁcant and the variables are not dependent. A Phi value was calculated to
determine the strength of the two variables. The Phi value is .170 which is about equal to
the Phi value of .116 for child abuse and delinquency. Thus, intensity of abuse has little
eﬁ‘ect upon the relationship between child abuse and subsequent delinquency for this

research sample.

93

The ﬁndings indicate that the intensity of the abuse and delinquency are not
dependent upon each other and that they are not statistically signiﬁcant. However, this
may have been different had the sample size been larger. The Phi value for intensity of
abuse indicates that this variable has the strongest effect of the control variables upon the
relationship than either the age of onset or the frequency of abuse.

As stated in the preceding pages both bivariate and multivariate analysis show no
statistically signiﬁcant effect of prior child abuse experience or subsequent delinquency.
Nonetheless, this researcher surmised that failure to distinguish between types of abuse
may have masked the possible effect of abuse on delinquency. Accordingly, the researcher
proceeded to examine if children are likely to exhibit delinquent behavior depending on
whether they had experienced physical or sexual abuse. Secondly, the researcher sought
to determine if there is a variation in types of delinquent conduct depending on the

variables on type of abuse. The results of these analyses will be presented in the next

pages.
h i n D lin

The ﬁrst pertinent research question (#3) asked: " Are those children who are
physically abused more likely to commit violent, aggressive behavior than those children
who are not physically abused?” As previously stated, there are two operationalizations of
physical abuse. The ﬁrst level is accidental abuse which occurs under the guise of
discipline, when a parent overreacts to the child's behavior, thus causing an injury to the
child. The second level involves the classic case of abuse, that is, intentional or non-
accidental inﬂiction of physical injury. Delinquency included violent crimes and non-
violent crimes. In this study, the crimes of murder, rape, assault, battery, robbery, and
arson are considered violent crimes, whereas the crimes of prostitution, theft,

incorrigibility, and drug abuse are considered non-violent crimes.

94

As previously stated, all the respondents have been adjudicated for a crime which
eliminates the use of Chi square to determine dependence between physical abuse and
delinquency. Therefore, when examining if physical abuse is related to delinquency, the
diﬁ‘erence of percentage between physical abuse victims and physical abuse non-victims is
examined. There was 62.9% of the respondents who were physiwa abused compared to
37. 1% of the respondents who were not physically abused. Thus, there is a difference of
25.8% higher for physical abuse respondents than non-physical abuse respondents.
Therefore, there is a relationship between physical abuse and delinquency based on the
percentages of respondents who were physically abused or not physically abused.

Table3l: r l i Ph i f b uen D lin n
physical abuse

crime yes no totals

violent 17 9 (26) 74.3%

non - violent 5 4 (9) 25.7%

totals (22) 62.9% (13) 37. 1% (35) 100%

Table 31 shows that 62.9% of the respondents were physically abused while
37. 1% were not abused. Also from the above table, 74.3% of the respondents were
adjudicated for a violent crime, whereas 25.7% were adjudicated for a non-violent crime.
In sum, 77.2% (17) physically abused respondents had a violent crime adjudication while
22.7% (5) of the physically abused respondents had a non-violent crime adjudication. In
summary, 48% of the research population was physically abused and was later involved in

violent crime. However, 34.6% of the research population that was not physically abused

95

had subsequent violent crime adjudication. There is 74.3% of the respondents who were
adjudicated for violent crime, however 34% of these respondents were not physically
abused. With to 34% of the respondents having not been physically abused but they have a
violent crime adjudication, it is suggested that physical abuse does not lead an individual
to be violent.

When Fischer's Exact test was run, it gave a probability level of .598, thus
suggesting that the relationship is not statistically signiﬁcant. In other words, there is no
dependence between physical abuse and violent, aggressive crimes. A Phi value was
calculated to determine the strength of this relationship between physical abuse and
violent, aggressive crimes. The Phi value is .088 which indicates that the strength of the
relationship is very weak. Although a statistically signiﬁcant relationship was not
established for physical abuse and violent, aggressive crimes, the researcher still controlled
for the eﬁ‘ects of race, age of onset for the physical abuse, frequency of the abuse,
intensity of the abuse, and protective factors to determine if these variables have an eﬁ‘ect
on the relationship between physical abuse and violent, aggressive crimes. In the

following pages, the results of analysis of the control variables will be presented.

Research Question 7: Is the alleged relationship between prior physical abuse and violent,
aggressive crimes affected by demographic characteristics such as race and age of onset of

abuse?

Research question 7 focused on the effect of race on the relationship between child
abuse and subsequent delinquency. It has already been established in this research
population that there is a disproportionality between the races. Although previous analysis
has concluded that there is no relationship between physical abuse and violent, aggressive

crimes, the researcher examined the effects of race upon the relationship between physical

96

abuse and violent, aggressive crimes to determine if there is a masking eﬂ‘ect within the
relationship.

Table32: 1'0‘.‘ .2). : u . 'h ‘13; a- ._ Tru‘ of _. A! -- D-clin no for
W
physical abuse
crime yes no totals
violent 8 3 (11) 68.8%
non - violent 3 2 (5) 31.3%
totals (11) 68.8% (5) 31.3% (16) 100%

The above table is a cross tabulation of physical abuse and violent, aggressive
crimes when controlling for race. Table 32 is a cross tabulation for White respondents. In
total, there are 16 White respondents. Eleven (68.8%) White respondents were physically
abused and 5 (31.3%) were not physically abused. Eight physically abused White
respondents (50%) were adjudicated for a violent crime whereas 3 respondents who were
not physically abused were adjudicated for a violent crime. However, 3 physically abused
White respondents (18.7%) were adjudicated for a non-violent crime and 2 respondents
who were not physically abused were adjudicated for a non-violent crime.

The Fischer's Exact Test gave a probability level of 1.00 thus showing that the
relationship between physical abuse and violent, aggressive crimes for White respondents
is not statistically signiﬁcant. The variables physical abuse and violent, aggressive crimes
for White respondents are independent of each other. A Phi value was calculated to
determine the strength of the relationship between physical abuse and violent, aggressive
crimes for White respondents. The Phi value is .127 which indicates that the relationship

97

is a weak relationship. The fact that the relationship is not statistically signiﬁcant may be
due to the relatively small sample size which limits the use of Chi square. Therefore, when
controlling for race, speciﬁcally White respondents, there was an effect to the relationship
between physical abuse and violent, aggressive crimes. When comparing the original Phi
value of .088 for physical abuse and violent, aggressive crimes with the Phi value of .127

for White respondents the relationship becomes somewhat stronger.

Table33: r. guano ._' .=_ -.,,- arm f ’I’a'. n D-clin ‘ a or
1 k n
physical abuse
crime yes no totals
violent 5 5 (10) 83.3%
non - violent 0 2 (2) 16.7%
totals (5) 41.7% (7) 58.3% (12) 100%

Table 33 is a cross tabulation of physical abuse and type of subsequent delinquent
involvement when controlling for race. The above table is for Black respondents. There
is a total of 12 Black respondents. From the above table, 5 (41.7%) were physically
abused while 7 (58.3%) were not physically abused. Five physically abused Black
respondents (41.6%) were adjudicated for a violent crime and ﬁve Black respondents who
were not physically abused were adjudicated for a violent crime. Finally, two Black
respondents (16%) who were not physical abuse victims were adjudicated for a non-
violent crime. Fischer's exact test gave a probability level of .469, thus suggesting that the
relationship is not statistically signiﬁcant. A Phi value was calculated to determine the
degree of strength for this relationship. The Phi value is .377 which indicates that there is

98

some strength to the relationship for Black respondents. Furthermore, the Phi value of
.377 is much higher than the original Phi value of .088 and the Phi value of . 127 for White
respondents. Therefore, the relationship between physical abuse and violent, aggressive

crimes for Black respondents is stronger than for White respondents.

Table34: 0 .2..- 2.011 01"! 1.2 ‘v _ .' a o Tm‘ f - 1'19 11 lin 1121 fr
WM

physical abuse
crime yes no totals
violent 4 1 (5) 71 .4%
non - violent 2 0 (2) 28.6%
totals (6) 85.7% (1) 14.3% (7) 100%

Table 34 is a cross tabulation of physical abuse and violent, aggressive crimes for
”other" race respondents. Hispanic and Native American respondents were combined with
respondents who responded ”other" for the purpose of analysis. In total, there are 7
respondents who fall within this category. Six of the 7 respondents (85.7%) were
physically abused. In regards to criminal involvement, 5 respondents (71.4%) were
adjudicated for a violent crime while 2 (28.6%) were adjudicated for a non-violent crime.
Four of the physically abused respondents (57.1%) were adjudicated for violent offense
while 2 physically abused (28.5%) were adjudicated for a non-violent crime. There was 1
respondent who was not physically abused but yet adjudicated for violent crime. Fischer‘s
Exact test gave a probability level of 1.00 which suggests that there is no statistical
signiﬁcance for the “other” race. This determines that there is no dependence between

physical abuse and violent, aggressive crimes when controlling for "other” race. A Phi

99

value was calculated to determine the strength of the relationship, although it is not
statistically signiﬁcant. The Phi value is .258 which indicates some strength for the
relationship between physical abuse and violent, aggressive crimes for the ”other" race.
This .258 Phi value is not as high as the .377 Phi value for Black respondents although it is
higher than the original Phi value of .088 and the .127 Phi value for White respondents.

The results of analysis indicate that controlling for race does have an effect upon
the relationship between physical abuse and violent, aggressive crimes. The effect is most
noticeable for the Black respondents for which 'the relationship becomes stronger based
upon the original Phi value of .088 and the Phi value of .377 for Black respondents. The
"other" race category also experienced a stronger relationship, although the increase in
strength is not as strong; a Phi of .258 compared to a Phi of .088. However, none of the
relationships were statistically signiﬁcant. This may have resulted from the relatively small
sample size. This ﬁnding of a stronger relationship for physical abuse and violent,
aggressive delinquency for Black respondents is consistent with prior research conducted
by Widom (1989). In her research, Widom concluded that there is a higher propensity
among Blacks to be abused and subsequently violent.

Research Question 7 also indicates that the alleged relationship between prior
physical abuse and violent, aggressive crimes may be affected by the age of onset for
abuse. Although there is no statistical dependence between physical abuse and violent,
aggressive crimes, the researcher still examined the effects of age of onset. The age of
onset is when the child was ﬁrst exposed to abusive behaviors.

This analysis was limited only to respondents who claimed to have been victimized
and reported an age of onset. The following cross tabulation displays age of onset for
physical abuse and subsequent delinquent involvement to help determine whether the age

of onset of abuse is likely to incline the victim to delinquency.

100

Table35: 1‘0‘. ;O-a.oonA1‘01‘110; . h ic A -:. qTin- . 01—13011
WW

age of onset physical abuse

for
crime 1 thru 6 7 and older totals
violent 9 6 (15) 75%
non - violent 2 3 (5) 25%
totals (11) 55% (9) 45% (20) 100%

Twenty of twenty-two (22) physically abused respondents answered this question.
Eleven respondents (55%) identiﬁed ages of 1 thru 6 as when the abuse began, while 9
(45%) were physically abused after the age of 7. The table also displays the type of
subsequent delinquency for physically abused respondents. Fifteen respondents (75%)
were adjudicated for a violent offense while 5 (25%) were adjudicated for a non-violent
offense. Nine of the 15 violent offenders (60%) were physically abused between the ages
of 1 thru 6 while 6 of the physically abused respondents (40%) were abused after the age
of 7. Of the 5 nonoviolent delinquents, 2 (40%) had been abused between the ages of 1
and 6 while 3 (60%) had been abused after the age of 7.

The result of Fischer's Exact test gave a level of probability of .616 suggesting no
statistical signiﬁcance between physical abuse and violent, aggressive crimes when
controlling for the age of onset. The lack of statistical signiﬁcance may be reﬂective of the
small sample size as the sample size affects the results of Chi square. A Phi value was
calculated to determine the strength of the relationship even though there is not statistical
signiﬁcance. The Phi value is .174 which indicates a weak relationship. This Phi value of

101

.174 is slightly higher than the original Phi value of .088 for physical abuse and violent,
aggressive crimes, thus indicating a small increase of strength of the relationship.
Therefore for this sample, age of onset does have a slight effect upon the relationship
between physical abuse and violent, aggressive crimes, although this relationship is not

determined to be statistically signiﬁcant.

Research Question 8: Is the alleged relationship between prior physical abuse and

subsequent delinquency aﬂ‘ected by the ﬁequency of the abuse?

Research question 8 suggests that the frequency of the physical abuse may have an
effect upon the relationship between physical abuse and subsequent delinquency. The
response categories included: "everyday", ”more than once a day", ”more than once a
week”, or ”other”. These answers were combined to form "every day”, "more than
weekly”, and ”less than weekly” for the purpose of analysis. Therefore, in order to
examine the eﬁ'ects that frequency has upon the relationship of physical abuse and violent,
aggressive crimes a cross tabulation of ﬁ'equency and subsequent delinquency was

computed.

102

Table 36: 0 .2.» at- 0 Fr - ‘ 1'1 0 _' i z ‘0 a; a 9 Tu; f -0 gut '
In 11
Frequency of physical abuse
crime every day more than weekly less than weekly totals
violent 2 9 5 (16) 76.2%
non - violent 0 3 2 (5) 23.8%
totals (2) 9.5% (12) 57.1% (7) 33.3% (21) 100%

Twenty-one (21) out of 22 physically abused respondents answered this question
concerning frequency of abuse. There were 2 physically abused respondent who
responded that they were physically abused "every day”. Both these respondents had a
violent crime adjudication. Twelve physically abused respondents (57.1%) replied they
were abused ”more than weekly”. Of these 12, 9 (75%) had a violent crime adjudication.
Finally, 7 physically abused respondents (33 .3%) responded that they were abused ”less
than weekly”. Of these 7 respondents, 5 (71.4%) had a violent crime adjudication.

The Chi square value was .721 with a probability level of .697, thus suggesting
that there is no statistical signiﬁcance among the variables. This indicates that the
variables are independent of each other. Lambda was calculated to determine the level of
association between ﬁ'equency of physical abuse and subsequent delinquency. The lambda
value is .000 which indicates that when knowing the frequency of physical abuse there is
likelihood to make the same number of errors when trying to guess if the respondents
were violent or non-violent, whether knowing or not knowing the attributes. This lambda
value and the Chi square result indicate that there is independence between the frequency
of abuse and violent, aggressive crimes.

This ﬁnding that frequency of abuse does not affect the relationship between
physical abuse and delinquency is inconsistent with prior research conducted by Fagan and

Wexler (1987). These researchers concluded that the frequency of occurrence of a

103

behavior causes an increased acceptance of the behavior by an individual. Thus, the more
ﬁequent the occurrence of abuse, the higher the acceptance of being abusive. Fagan and
Wexler' 3 research sample only interviewed violent juvenile delinquents as opposed to this
research, which surveyed violent and non-violent. The researcher believes that the
different subjects lends to the discrepancy between ﬁndings concerning frequency of

abuse.

Research Question 8: Is the alleged relationship between prior physical abuse and

subsequent delinquency affected by the intensity of the abuse?

Research question 8 indicates that the intensity of the physical abuse may have an
impact upon the relationship between physical abuse and violent, aggressive crimes. For
the purpose of this research, the intensity of the abuse was calculated as the total number

of abusers for each physically abused respondent.

 

Intensity of physical abuse

crime 1 or 2 abusers 3 or more abusers totals
violent 3 14 (17) 77.3%
non - violent 3 2 (5) 22.7%
totals (6) 27.3% (16) 72.7% (22) 100%

Table 37 is a cross tabulation of the intensity of the physical abuse and type of
delinquent involvement for physically abused respondents. For the purpose of this

104

research, the intensity of the abuse was determined as the total number of abusers for each
abused respondents. The range of abusers is 1 to 5 abusers with a modal value of 2
abusers. For the purpose of analysis, abusers were combined into two categories, 1 or 2
abusers and 3 or more abusers. Six physically abused respondents (27.3%) were abused
by one or two individuals. Three of these six respondents were adjudicated for a violent
crime while three were adjudicated for a non-violent crime. Sixteen physically abused
respondents (72.7%) who were abused by 3 or more perpetrators. Of these 16, 14
(87.5%) were violent and 2 (12.5%) were non-violent offenders. The ﬁndings show that
the majority of respondents (87.5%) with 3 or more abusers had a violent adjudication
which lends support to the relationship between age of onset and violent, aggressive
crimes.

The Chi square Fischer's Exact test gave a probability level of .100 which suggests
a statistically signiﬁcant relationship at the .10 alpha level. The Phi value was .398 which
suggested a relatively strong effect of intensity on the relationship. The strength of the
relationship is also evident by the difference of proportions among 1 or 2 abusers who
were violent (50%) and 3 or more abusers who were violent (87 .5%). These data suggest
that abused children are likely to manifest violent subsequent delinquency if they were
exposed to multiple abusers. In other words, the more abusers, meaning the more intense

the abuse, the more likely the respondent was to become violent.

Research Question 6: Are there protective factors which may prevent an abused child

ﬁom acting out through delinquency?

The intent here was to determine if certain protective factors are likely to prevent
an abused child from acting out through delinquency. To determine the extent of these
factors the researcher asked respondents to indicate if they ever saw a doctor, a therapist,

or a protective service worker after the abuse experience. The researcher surmised that

105

such treatment interventions may mitigate the effect of abuse and minimize the chances of
violent delinquency among abuse victims. Therefore, the researcher investigated the
response to physical abuse by the respondents and compared these responses with the type

of crime, that is, either violent or non-violent, the respondent was adjudicated for.

Table38: Ff'n',_"11'1__.110_100n 1501010611 ‘ 10 .r'vn -rv- 1'1. 211' ’1
13.12.18
W911 f 22
yes 7 31.8
no 15 68.2
TOTAL 22 100

Table 38 presents the number of respondents who beneﬁted from some form of
intervention through the courts or their personal physician. Only 7 respondents (31.8%)
reported having had such interventions while 68.2% did not. The researcher also
examined the type of crime that these 7 respondents were adjudicated for. Surprisingly,
all 7 respondents were adjudicated for a violent offense. Therefore, obtaining help or

intervention does not keep an adolescent from acting out through delinquency.

A lin

Previous analysis suggest no statistically signiﬁcant relationship between child
abuse and subsequent delinquency or physical abuse and violent, aggressive crimes for this
research population. In the following section, ﬁndings of analysis are presented as they
relate to the alleged relationship between child sexual abuse and subsequent sexual crimes.

Research Question 3: Are those children who are sexually abused more likely to commit
sexual crimes than those children who are not sexually abused?

106

For the purpose of this research those respondents who indicated that they were
sexually ”hanned" by an adult were considered to be sexually abused. Additionally, those
respondents who were adjudicated for criminal sexual conduct, csc I, II, III, or IV, were
considered to be sexually delinquent. Analysis was then conducted for these two variables
to determine if there is a relationship between child sexual abuse and sexual crimes based

on recent adjudications.

Table 39: l i f D lin n

sexual adjudication

sexual crime yes no totals

yes 4 6 (10) 28.6%
no 3 22 (25) 71 .4%
totals (7) 20% (28) 80% (3 5) 100%

The above table shows that ten respondents (28.6%) were sexually victimized as
children whereas 25 respondents were not victims of child sexual abuse. Also ﬁ'om the
above table, 7 respondents (20%) were adjudicated for a sexual crime while 28 (80%)
were not adjudicated for a sexual crime. Four of the ten sexual abuse victims (40%) were
adjudicated for a sexual crime whereas 6 sexually victimized respondents (60%) were not
adjudicated for any sexual crime. Three respondents (8.5%) were adjudicated for a sexual
offense, however, they had not been sexually abused. There were 22 respondents who

indicated they were neither sexually abused nor adjudicated for a sexual crime.

107

Fischer‘s Exact test gave a probability level of .171, thus showing that the
relationship between sexual abuse and sexual delinquency is not statistically signiﬁcant. A
Phi value was calculated to determine the strength of the relationship although there is not
any dependence between variables. The Phi value is .276 which indicates that there is a
weak relationship between variables, although it is not very strong for this sample.

The researcher investigated the effects of certain control variables to determine if
they have any effect upon this relationship which may have been masked in the original
relationship. Control variables such as race, age of onset for sexual abuse, frequency of
abuse, intensity of abuse, and protective factors were examined. In the following pages,
the results of analysis of these control variables are presented to examine their effects

upon the relationship between child sexual abuse and sexual delinquency.

Research Question 7: Is the relationship between prior child abuse and subsequent

delinquency affected by demographic characteristics such as race?

Table40: r l in f A xualA'ui infrWhi
82312911119111:
sexual adjudication
semal crime yes no totals
yes 3 1 (4) 25%
no 4 8 ( 12) 75%
totals (7) 43.8% (9) 56.3% (16) 100%

The above table is a cross tabulation of sexual abuse and sexual adjudication when

controlling for race, speciﬁcally White respondents. There are 16 White respondents, 7

108

(43.8%) who were sexually abused and 9 (56.3) who were not sexually abused. Also, of
those 16 White respondents, 4 (25%) were adjudicated for a sexual crime and 12 (75%)
were not adjudicated for a sexual offense. Three sexually abused White respondents were
adjudicated for a sexual crime, while 4 sexually abused respondents were not adjudicated
for any sexual crime. Also, one White respondent was adjudicated for a sexual crime but
yet was not sexually abused. There were 8 White respondents who were not sexually
abused nor were they adjudicated for a sexual crime.

F ischer‘s Exact test gave a probability level of .261, thus suggesting that the
relationship between sexual abuse and sexual adjudication for White respondents is not
statistically signiﬁcant. A Phi value was calculated to measure the degree of association
between these two variables for White respondents. The Phi value is .363 which suggests
a small amount of strength between these sexual abuse and sexual delinquency for White
respondents. This Phi value of .363 is higher than the original Phi of .276 without
controlling for race. Therefore, controlling for race does have a slight effect upon the

relationship between sexual abuse and sexual delinquency.

Table41: r l in f xu A Ad' i in frBlak
893187114511:
sexualadjudication
sexualcrime yes no totals
yes 1 (1) 8.3%
no 23 9 (11) 91.2%

totals (2)16.7% (10) 83.3% (12) 100%

109

Table 41 is a cross tabulation of sexual abuse and sexual adjudication when
controlling for the race. Table 40 shows the result of the Black respondents. There are
12 Black respondents, 2 (16.7%) who were sexually abused and 10 (83.3%) who were not
sexually abused. There was 1 respondent (8.3%) who was adjudicated for a sexual crime
while 11 (91.7%) were not adjudicated. Zero Black respondents were sexually abused
and adjudicated for a sexual crime however there was one respondent who was
adjudicated for a sexual crime but was not sexually abused. Two sexually abused Black
respondents (16.6%) were not adjudicated for a sexual crime. The remaining 9 Black
respondents (75%) indicated not being sexually abused and no adjudication for a sexual
crime.

Fischer's Exact test gave a probability level of 1.00 thus suggesting that the
relationship is not statistically signiﬁcant. In fact, there is independence between the two
variables. A Phi value was calculated to determine the degree of strength between sexual
abuse and sexual crimes for Black respondents. The Phi value is .134 which indicates a
weak relationship for Black respondents.

Table42: 0‘; .a.t_121110f ‘ a- A 1‘ a l ' a A. 01.3111 f "0 hr" ' = 1'
Wants

sexual adjudication

sexual crime yes no totals
yes 1 1 (2) 28.6%
no 1 4 (5) 71.4%

totals (2) 28.2% (5) 71.4% (35) 100%

110

As explained earlier, Hispanics and Native Americans were combined into one
category of race, ”other", for the purpose of analysis. The above cross tabulation displays
the results of those "other" respondents who were sexually victimized and adjudicated for
a sexual crime. There are 7 respondents considered as ”other”, 2 (28.2%) were sexually
abused and 501.4%) were not sexually abused. Of the 7 respondents, one was both
sexually abused and adjudicated for a sexual crime and one was adjudicated for a sexual
crime, however, was not sexually abused. Also, one respondent was sexually abused but
not adjudicated for a sexual crime. The remaining 4 respondents had neither been sexually
abused nor been adjudicated for a sexual crime.

Again, Fischer's Exact test gave a probability level of 1.00 thus suggesting that the
relationship is not statistically signiﬁcant. This indicates that the variables sexual abuse
and sexual delinquency for the ”other” race are independent of each other. The Phi value
is .300 with a probability level of 1.00 suggest a weak relationship.

Therefore, race does have an effect on the relationship between prior sexual abuse
and sexual delinquency. The relationship between sexual abuse and sexual delinquency is
not statistically signiﬁcant. However, the relationship does become a stronger relationship
for speciﬁc races. This strength is based upon comparing Phi values with the original Phi
value of .276. The Phi value for the White race is .363 and the Phi value for the ”other"
race is .300, both are higher than the original Phi value. Therefore, controlling for race

does have an impact upon the relationship between sexual abuse and sexual delinquency.

Research Question 7: Is the relationship between prior child abuse and subsequent
delinquency affected by demographic characteristics such as age of onset?

Past research by Longo (1981) concluded that the earlier the exposure to the
sexual behavior the increased likelihood to be sexually deviant. Also, the theoretical

111

Work explains that the earlier the exposure the child has to a behavior the more likely
the child is to acquire and internalize the behavior. The age of onset is the child's ﬁrst

exposure to the sexual abuse.
Table43: 0 .2... -1; or 0 a: 0 01-; 0 ' -2 30.1" 2.4! m 0 . .- gut .‘~ll
' l R n 11
age of onset for sexual crime
sexual crime 1 thru 6 7 and older totals
yes 2 2 (4) 40%
no 3 3 (o) 50%
totals (5) 50% (5) 50% (10) 100%

Table 43 presents the ﬁndings on the effect of the age of onset on the relationship
between sexual abuse and sexual delinquency. Ten respondents (28.6%) were sexually
abused. The age range for respondents was 4-13 years. The mean value for the age of
onset was 8.3 years of age. Of the 10 who were sexually abused, 5 (50%) were abused
between the ages of 1 thru 6, while 5 (50%) were abused ﬁ'om 7 and older.

Fischer's Exact test gave a probability level of 1.00 thus suggesting that age of
onset of abuse and sexual delinquency is not statistically signiﬁcant. The Phi value is .069
which indicates that the relationship between age of onset and sexual delinquency is very
weak. This ﬁnding is inconsistent with the ﬁndings of Longo (1981). Longo concluded
that the earlier the exposure to sexual abuse and greater the likelihood to be sexually
deviant. One reason for the discrepancy is Longo interviewed only sexually delinquent

individuals who were residents of a sexual offender treatment program.

112

Research Question 8: Is the alleged relationship between prior physical abuse and
subsequent delinquency aﬂ‘ected by the frequency and intensity of the abuse?

In the following section the researcher will present analysis of the ﬁ'equency and
intensity of the sexual abuse to determine if these variables affect the relationship between
prior child sexual abuse and subsequent sexual delinquency. As previously deﬁned, the
frequency of the abuse is the number of times the abuse occurred, that is, "every day”,
”more than once a day”, ”more than once a week”, or "other". For the purposes of
analysis, frequency was combined into three categories, every day, more than weekly, and
less than weekly. The intensity of the sexual abuse was determined as the number of
abusers to which the respondent was exposed. Therefore, the more the number of abusers

the higher the intensity of the abuse.

Table44: 0 54-13111 f ' 2.. rim-s a r Frvc -‘ a 1" 7(1):. ‘t .1 fr ‘ 21
WW
Frequency of sexualabuse
crime every day more than weekly less than weekly totals
violent 1 2 1 (4) 40%
non - violent 0 3 3 (6) 60%
totals ( 1) 10% (5) 50% (4) 40% (10) 100%

Table 44 is a cross tabulation of sexual crimes and the ﬁequency of sexual abuse
for those respondents who were sexually abused. In total, 10 respondents were sexually
abused. F our of these respondents (40%) were adjudicated for a sexual crime while 6
(60%) were not. One respondent (10%) replied that the sexual abuse occurred ”every
day”, while 5 respondents (50%) indicated they were abused ”more than weekly”. Finally,

113

4 respondents (40%) indicated ”less than weekly" as a response. The one respondent
who indicated ”every day" as a frequency was adjudicated for a sexual crime. From the
ﬁve respondents who indicated "more than weekly”, 2 respondents were adjudicated for a
sexual crime while 3 were not. Finally, only one of the four respondents who replied ”less
than weekly” was adjudicated for a sexual crime.

The Chi square value was 1.87 with a probability level of .391. Given the
probability level of .391, the relationship is not statistically signiﬁcant. Lambda was
calculated to measure the degree of association between sexual delinquency and ﬁequency
of abuse. The Lambda value is .250 which suggests that knowing the frequency of sexual
abuse one could predict the likelihood of a sexual crime only 25% of the time. Therefore,
frequency of sexual abuse has a weak effect on the relationship between sexual abuse and

sexual delinquency for this sample.

114

Research Question 8: Is the alleged relationship between prior physical abuse and

subsequent delinquency affected by the intensity of the abuse?

Table45: ro; 2.0101 0 - .2. Dlin- - a : oInnil o A 4;; --fr
1 R n n

Intensity of sexual abuse

sexual crime 1 or 2 abusers 3 or more abusers totals

yes 2 2 (4) 40%
no 0 6 (6) 60%
totals (2) 20% (8) 80% (10) 100%

Table 45 is a cross tabulation of sexual crimes and the intensity of the sexual
abuse. The intensity of the sexual abuse is measured by the number of perpetrators, which
are in categories of 1 or 2 abusers or 3 or more abusers. Ten respondents were sexually
abused. The range of perpetrators was 2 to 5 with a modal value of 2. From the above
table, 4 respondents (40%) committed sexual crimes while 6 (60%) did not. Of the four
sexually delinquent respondents, 2 (50%) had 2 perpetrators while 2 (50%) had 3 or more
perpetrators.

Fischer's Exact test gave a level of probability of .13 which indicates that there is
slight statistical signiﬁcance between sexual delinquency and intensity of abuse. In fact,
had the sample size been larger than 10 it is quite possible that the level of probability
would have decreased. The Phi value is .612 which indicates a moderate strength between
sexual delinquency and intensity of abuse. This Phi value is also much higher than the
original Phi value of .276 which was calculated for sexual abuse and sexual delinquency.

115

Therefore, intensity of abuse does have an effect upon the relationship of sexual abuse and
sexual delinquency within this research sample. This ﬁnding of intensity is similar to the
ﬁndings of intensity for physical abuse. Therefore, the higher the number of abusers the
greater the likelihood to become delinquent.

h lin nR n A R 11

Research Question 5: Are there other delinquent or deviant responses to either prior

physical or sexual abuse?

In this section, the researcher examines the self report delinquency questions with a
response scale of 1-5. The self report questions are Likert scored questions. The
response categories are: "never", "1 or 2 times", "several times”, ”often", and ”very oﬁen".
The self report questions required the respondents to reveal crime information about
themselves. The crimes included real crimes such as property crimes and drug
involvement and status oﬁ‘enses such as vandalism and truancy. Property crimes included
stealing or theﬁ, stealing money from parents, stealing large amounts (850+), stealing cars,
and joy riding. Vandalism involves defacing public or private property. The drug crimes
consisted of drug use and sales by respondents. Analysis is conducted on the above

mentioned crimes for both physically and sexually abused respondents.

116

Table46: R n ’ nv v n wi r ri

f %
1 or 2 times 5 14.3
several times 8 22.9
often 11 31.4
very often 10 28.6 '1 missing case
totals 35 100

Table 46 examines the frequency of theft including stealing property or money.
One respondent did not choose to answer the question. Twenty-one respondents
indicated an involvement in a high number of these crimes as indicated by their answers of
either ”often" or "very often". The researcher examined those respondents who were
physically abused and their involvement in property crimes. Of the 22 respondents who
were physically abused, all 22 indicated that they committed at least one crime against
property. Nineteen of the 22 respondents (86.3%) indicated that they were involved in
property crimes at least "several times". When the responses for sexually abused victims
were examined, the researcher found that most respondents indicated a low involvement
of property crimes. These ﬁndings suggest that property crimes were more ﬁequent
among physically abused respondents than respondents who were sexually abused.

The researcher also examined the data for those respondents who admitted
committing the status offense of vandalism. These crimes involve destroying or defacing
public or private property such as breaking windows or street lights or throwing objects at

others' property.

117

Table 47: Etoqpopoy of Rospondsnts’ Idvolvomont with ngslism

f %
1 or 2 times 10 28.6
several times 11 31.4
often 6 17.1
very oﬂen 7 20 ‘1 missing case
totals 35 100

Table 47 shows that one respondent did 'not answer this question. Of 34 who
answered, twenty-one respondents (60%) indicated that they committed some type of
vandalism, however, the ﬁndings indicate a frequency of occurrence of either "1 or 2
times” or ”several times”. The remaining 37% indicate a high frequency of occurrence for
vandalism crimes.

The researcher also examined data for respondents who were physically and
sexually abused and the frequency of their involvement in vandalism. Of the 22
respondents who were physically abused, all indicated that they were involved in
vandalism at least once. Furthermore, 18 physically abused respondents (81.8%) indicated
that they were involved in vandalism ”several times” or more. ”Several times” was deﬁned
as more than 5. This is a high percentage of respondents who committed the offense of
vandalism. Of the sexually abused respondents (10), all reported some involvement with
vandalism. However, the majority of sexually abused respondents (70%) indicated a low
ﬁequency of occurrence, by answering either "1 or 2 times”. In summary, there was a
high number of respondents who were physically abused and had frequent involvement
with vandalism as compared to sexually abused respondents who had less frequent
involvement with crimes of vandalism. Therefore, physically abused respondents (81.8%)
were more likely to have involvement with crimes of vandalism than sexually abused

respondents (70%).

118

The other status offense that is investigated is truancy. Current day adolescents
believe that it is acceptable to skip school and ”hang out”. However, skipping school is

considered a status offense for adolescents and they can be adjudicated for this action.

Table 48: WWW

f %
1 or 2 times 5 14.3
several times 13 37.1
often 8 22.9
very often 8 22.9 ‘1 missing case
totals 35 100

Table 48 presents the respondents’ answers to how often they either ran away
from home or they skipped school. Again, one respondent chose not to answer either
question pertaining to truancy. Almost half of the respondents, that is, 45.8% indicated a
frequency of either ”often” or ”very often” for truancy. There is also 37 . 1% who indicated
that they were truant ”several times".

When comparing the responses for physically or sexually abused respondents, 10
physically abused respondents replied "oﬁen" or "very often”. Four sexual abuse
respondents replied ”often" or "very often", whereas 6 replied "several times”. Therefore,
the physically abused respondents were involved in being truant from home or school
more oﬁen than those respondents who were sexually abused.

The researcher also investigated the ﬁndings of respondents' answers to drug use
and sales. The drugs in question were marijuana, cocaine, LSD, and crack both for use
and sales.

119

Table49: , "30.9115 0 'snun!‘ ’Ivlv ‘1 will we is 111 11‘ 1' ‘.:_ n
f %

lor2times 5 14.3

several times 8 22.9

often 11 31.4

very oﬁen 10 28.6 *1 missingcase

totals 35 100

The above table examines the respondents’ answers to their involvement with
either drug use or drug sales. One respondent did not answer this question. There are 34
respondents who indicated some type of involvement with substances. Twenty-one
respondents (60%) indicated that they were involved either ”oﬁen" or "very often”. When
examining the results of respondents who were physically or sexually abused, there was a
relatively equal distribution of respondents for each category, meaning that there is no
difference between physical abuse or sexual abuse respondents to engage in substance
abuse crimes. Therefore, the type of abuse did not predispose a respondent to engage in
drug use or sales.

The ﬁfth research question asked if there were other delinquent responses for
either prior physical or sexual abuse victims. The respondents answered amrmatively and
indicated these responses include property crimes, vandalism, truancy, and substance
abuse crimes. The physically abused respondents had a greater likelihood and frequency
of occurrence for property crimes, vandalism, and truancy. However, there was an equal
chance for physically or sexually abused respondents to be involved in substance abuse

crimes, including either usage or sales.

SUMMARY AND CONCLUSIONS

The purpose of this research was to examine the relationship between prior child
abuse and subsequent delinquency. The study incorporated a theoretical ﬁamework for
examining the learning and modeling of abusive behaviors and for understanding the
alleged relationship between prior child abuse and subsequent delinquency. Several past
researchers ( Fagan & Wexler (1987), Gillespie, Seaburg, and Berlin (1977), Kratcoski
(1982), Kratcoski and Kratcoski (1982), Owen and Straus (1975), and Ryan (1991) have
attempted to explain the relationship between child abuse _andsubseqnent_d£1}_'1‘193“3y

through the application of the social learning theory. The social learning theory is a

“us...— ..._.‘ w

 

 

powerful theoretical tool for explaining how an individual acquires behaviors through
direct experience or observation and then models those acquired behaviors. In eﬂ‘ect, the
individual turns from victim to victimizer. These past researchers determined that a
relationship between child abuse and subsequent delinquency exists. However, they also
cautioned that the relationship is not direct, but is affected by other factors.

Thus, the researcher developed a theoretical explanation based on the social
learning theory to examine the learning and modeling of abusive behaviors. The
theoretical explanation examines how behaviors are learned, what provokes or instigates
learned behaviors, and what maintains these learned behaviors. This theoretical
framework provides the background for understanding why abused children have a
propensity to become involved in delinquent activity.

In theory, behaviors are learned through direct experience or observation within an
individual's environment. An individual pays attention to behaviors that appear to be

rewarding or important. These rewarding behaviors are retained or internalized for future

120

121

use. These behaviors are invoked into action based on an outside stimulus which is either
biological or cognitive. Biological stimuli include external threats, physical assaults, and
verbal insults. These stimuli induce an emotional response by the individual to cope with
the stimuli. Cognitive stimuli are when the individual has the mental capacity to recognize
ﬁrture consequences of a stimuli. These future consequences are inﬂuenced by observing
others and observing past experience. Finally, the behaviors are maintained due to the
reinforcement of the behavior. If the behavior is perceived as having positive results then
ultimately it will be maintained. However, if the behavior is perceived as being negatively
reinforced then it will cease to exist or be extinguished.

Therefore, the researcher proposed to use the social learning theory to examine the
learning and modeling of abusive behaviors. The purpose was to determine if the type of
delinquent activity varies according to the form of child abuse experienced by an
adolescent. The researcher studied 35 delinquent adolescents at three Lower Michigan
juvenile treatment facilities to examine the relationship between prior child abuse and
subsequent delinquency. Speciﬁcally, the research questions included:

1) Is there a relationship between prior child abuse and subsequent delinquency?

2) Is there a relationship between prior child abuse the adolescent experienced and
the type of subsequent delinquency?

3) Are those children who have been physically abused more likely to commit
violent, aggressive crimes than those children who have not been physically
abused?

4) Are those children who have been sexually abused more likely to commit sexual
crimes than those children who have not been sexually abused?

5) Are there other delinquent or deviant responses to either prior physical or
sexual abuse?

6) Are there protective factors which may prevent an abused child from acting out
through delinquency?

122

7) Is the alleged relationship between prior child abuse and subsequent
delinquency affected by demographic characteristics such as race and age of
onset?

8) Is the alleged relationship between prior child abuse and subsequent

delinquency aﬂ‘ected by the ﬁ'equency and intensity of the abuse?

The researcher used a survey design and a questionnaire instrument to gather data
which address the respondents' prior abuse experience and their involvement in
delinquency. The total population was 148 delinquent adolescents at three Lower
Michigan juvenile facilities. Due to the University's ethical rules regarding juveniles,
parental consent had to be obtained prior to obtaining a respondent's verbal assent to
participate. Advance written notice was sent to prospective respondents' parents together
with a request for consent to participate. Many parents failed to submit their consent. It
can only be speculated that these parents were uninterested in the study, were
apprehensive of what might be disclosed, or just did not care. Thus only 35 respondents
for whom there was parental assent and respondent assent could be included in the study.
The researcher did not randomly sample due to an unsuccessful attempt to obtain a non-
delinquent population. Therefore, the entire population was used for statistical
procedures.

After data were collected, ﬁ'equency distributions, chi square , and Phi values were
used to analyze the data. The data show an age range of 12 to 17 with an average age of
15.6 years of age. In regards to race, 45.7% of the respondents were White, 34.3% were
Black 5.7% were Hispanic, and 2.9% were Native American. Over half, 54.2%, of the
respondents came from divorced or separated families and 31.4% were from married
families. Furthermore, 65.7% of the respondents had multiple placement histories ranging
ﬁ'om 1 to 9 facilities. The results of the data also indicate that 23 (65.7%) of the
respondents were abused as children while 12 (34.3%) were not abused. Similarly, the

123

data show that 26 (74.3%) of the respondents were adjudicated for violent offenses and 9
(25.7%) were adjudicated for non-violent offenses.

The results of the research show some ﬁndings that are signiﬁcantly different from
ﬁndings ﬁ'om past studies. Yet, there are also some consistent ﬁndings. Therefore, the
interpretation of the ﬁndings will be presented followed by a discussion of the limitations
and signiﬁcance of this 'study.

i i in

In this study, the researcher examined the relationship between child abuse and
subsequent delinquency and the relationship between the type of child abuse suffered and
type of subsequent delinquency. The researcher sought to determine if there is
relationship between physical abuse and violent, aggressive crimes and if there is a
relationship between sexual abuse and sexual crimes. The ﬁndings show that there rs a

-M .

 

higher percentageof child abuse victims who are delinquent than non—victims .andahigher
percentage of physical abuse victims who are delinquent than non-physical abuse victims.
These higher percentages suggest that there is a relationship between child abuse and

delinquency and a relationship between physical abuse and dellnquency This ﬁnding rs
consistent with prior “research by“ 51583581) Fagan & Wexler (1987), Gillespie,
Seaburg, & Berlin (1977), Kratcoski & Kratcoski (1982), McCord (1983), and Widom
(1989). These researchers determined that there is dependence between either child abuse
and delinquency or physical abuse and violent delinquency. However, the results show
that there is no statistical relationship between child abuse and type of delinquency, nor is
there a relationship between physical abuse and violent, aggressive crimes or sexual abuse
and sexual delinquency for this sample. The ﬁndings are inconsistent with past research
conducted by Alfaro (1981), Benoit & Kennedy (1992), Fagan & Wexler ( 1987),
Gillespie, Seaburg, & Berlin (1977), Kratcoski and Kratcoski (1982), Longo (1981),
McCord (1983), and Widom (1989). These researchers found that a relationship exists

between child abuse and type of subsequent delinquency. Widom (1989) found a

124

relationship to exist and also explained that the relationship is affected by control variables
such as race and gender. Fagan & Wexler (1987), Gillespie, Seaburg, & Berlin (1977),
and Kratcoski & Kratcoski (1982) explained that the process of learning violence occurs
speciﬁcally within the family environment. Longo (1981) and Benoit & Kennedy (1992)
determined that a relationship between sexual abuse and sexual delinquency does exist.

Alfaro (1981) examined child protective service case ﬁles and court records from
1952-53 and 1971-72, respectively, to obtain data concerning a child's exposure to
violence and later involvement with delinquent behavior. Alfaro determined that there is a
relation between child abuse and delinquency. Speciﬁcally, Alfaro concluded that those
children who were abused are more likely to be violent than non-abused children. This
ﬁnding is consistent with the ﬁndings of Fagan & Wexler (1987). Fagan & Wexler also
determined that a relationship between child abuse and delinquency exists.

The results of the present research are inconsistent with research conducted by
Alfaro and Fagan & Wexler. The results of this study show no statistical diﬂ‘erence
between abused and non-abused respondents' propensity to engage in violent oﬂ‘enses.
Although there was a high number of abused respondents who were violent (17), there
were 9 non-abused respondents who were violent. Fischer's Exact Test gave a probability
level of .598 suggesting that the relationship is not signiﬁcant. Therefore, there is no
statistical relationship or difference for abused, violent respondents and non-abused violent
respondents for this research. .

The lack of statistical signiﬁcance between child abuse and delinquency is also
inconsistent with the ﬁndings of McCord (1983). McCord investigated the case ﬁles of
232 males and interviewed the families to determine interactional patterns. McCord
determined that those children who were abused or neglected had a high rate of serious
crime involvement. McCord's research also indicated that the interaction between a child
and parent is a signiﬁcant factor that affects the relationship between child abuse and

delinquency. A statistical relationship between child abuse and delinquency was found for

125

the present research; there was a signiﬁcant number of respondents who reported being
severely disciplined or abused by their parents. Seventy-one percent of the respondents
were disciplined or "physically harmed” by their parent. There is indication that these
respondents were signiﬁcantly affected by their parents’ behaviors based upon the number
of respondents who were ”physically harmed" by their parents and who were later
delinquent.

This research also examined the relationship between physical abuse and violent,
aggressive crimes and found no statistically signiﬁcant relationship among variables.
Contrary to this research, Fagan & Wexler (1987), Kratcoski & Kratcoski (1982), and
Pfouts et al. (1981) found that there is a relationship between physical abuse and violent
actions. Fagan & Wexler (1987) and Kratcoski & Kratcoski (1982) found a relationship
to exist and that the violence is directed by the abused individual towards family members.
Pfouts et a1. concluded that children who are victims of family violence have a higher
degree of deviancy than family members who are witness to family violence. What is
signiﬁcant is the understanding that violence is learned within the family environment and
that a relationship exists between physical abuse and violent actions. The ﬁndings of this
study are inconsistent with those of past researchers. Past researchers conclude that there
is a relationship between physical abuse and type of delinquency.

The results of this research did not investigate the recipient of violent actions,
however, the research did examine the relationship between physical abuse and violent,
aggressive crimes. The ﬁndings of the study are inconsistent with those of past
researchers, Fagan & Wexler (1987), Kratcoski & Kratcoski (1982), and Pfouts et al.
(1981). However, the lack of relationship between physical abuse and violent, aggressive
behavior is consistent with Widom (1989). Widom concluded that those physically abused
children did not differ in regards to violent offenses than non-physically abused children.
Widom concluded that there is increased risk of delinquency for abused children. This risk

is for engaging in property crimes and status offenses such as truancy and incorrigibility.

126

Although this research did not investigate the difference between being the
recipient of family violence or observing family violence, the ﬁndings of this research show
that there is some difference in frequencies between respondents who were abused and
committed violent crimes and respondents who were abused and committed non-violent
crimes. Gillespie et al. (1977) and Owens & Straus (1975) found that children who were
victims of family violence engaged in more non-violent or esCape crimes such as truancy
and theft than children who witnessed family violence. The children who observed family
violence were more likely to engage in violent behavior. The ﬁndings of this research are
consistent with the results found by Gillespie et al. (1977), Owens & Straus (1975), and
Widom (1989), who concluded that abused children will engage in escape crimes in order
to avoid ﬁrrther aggression.

Another ﬁnding of this research that is consistent with past research is the
determination that Black respondents have a stronger relationship between physical abuse
and violent crime than White respondents. This ﬁnding is consistent with the results of
Widom (1989). She concluded that being a victim of child abuse has a stronger aﬂ‘ect
upon Black children, especially Black males, than non Black children. The ﬁnding of this
research determined that there is no statistical relationship between physical abuse and
type of delinquency. However, there is a stronger relationship between physical abuse and
violent crime for Black respondents as opposed to White respondents. The strength of the
relationship was determined by comparing the Phi values of .377 for Black respondents
and the Phi value of .127 for White respondents. In regards to sexual abuse and sexual
delinquency, White sexually abused respondents had a Phi value of .363 compared to a Phi
value of .134 for Black sexual abuse respondents. Therefore, White respondents had a
higher likelihood than Black respondents to be a sexual abuse victim and engage in
sexually delinquent behavior.

This research also investigated the relationship between child sexual abuse and

subsequent sexual delinquency. This research concluded that there is no statistical

127

relationship between child sexual abuse and sexual delinquency. This ﬁnding is consistent
with the results of Widom's research. She concluded that there was no direct linkage
between past child sexual abuse and sexual delinquency. However, Widom admitted that
this result is contrary to past research. Past research conducted by Benoit & Kennedy
(1992) and Longo (1981) concluded that there is a relationship between sexual abuse and
sexual delinquency. These researchers reasoned that there was an atypical sexual
development by sexual offenders which is a primary cause for an individual to sexually act
out.

The research questions also consider if the frequency, the intensity, or the age of
onset of abuse has any affect upon the relationship between child abuse and delinquency or
affect the relationship between physical abuse and violent crime or sexual abuse and sexual
delinquency. The rationale for investigating these factors is based upon the theoretical
ﬁarnework that holds that the more exposure to the behavior, the more likely the
individual will internalize this behavior and model its use. The theoretical ﬁ'amework also
explains that the more repeated the frequency of exposure a child has to a particular
behavior the more likely the behavior will be retained. Furthermore, the younger the child
is exposed to abuse, the more likely the child will be unable to determine that this behavior
is unacceptable.

The research examined the effects that age of onset has upon the relationship
between abuse and delinquency. According to the theoretical framework, the earlier a
child was abused the more likely the child is to become violent or sexually delinquent.
When examining the age of onset for abuse, either physical or sexual, this variable did not
aﬂ‘ect the relationship between abuse and delinquency. The relationship was determined
to have no statistical signiﬁcance and therefore no dependence between variables.

The researcher also did not ﬁnd any statistical relationship between frequency of
the abuse and subsequent delinquency. This ﬁnding is inconsistent with the ﬁndings of
Gillespie et al. (1977). Gillespie et al. determined that the ﬁ'equency of occurrence for

128

abusive behaviors leads to a higher acceptance and likelihood of modeling these behaviors.
However, the ﬁequency of occurrence for this sample did not affect the relationship
between abuse and subsequent delinquency.

However, the results of this research are consistent with the ﬁndings by Benoit &
Kennedy (1992). Benoit & Kennedy (1992) examined the frequency and intensity of
sexual abuse. These researchers determined that there is no difference in the frequency or
intensity of past sexual abuse when investigating the relationship. In this study, there was
no affect upon the relationship between sexual abuse and sexual delinquency when
controlling ﬁ'equency or intensity of the sexual abuse which is consistent with the ﬁndings
of Benoit & Kennedy (1992).

This research also examined the effects that protective factors may have on the
relationship between abuse and delinquency. The results of this research are inconsistent
with the ﬁndings of Alfaro in regards to protective factors. The results of this research do
not indicate a diﬂ‘erence in relationship when controlling protective factors such as
counseling or therapy. There were 31.8% of the respondents who utilized some type of
protective factor. However using a protective factor did not prevent a respondent ﬁ'om
becoming delinquent, nor did it show that it prevented a respondent from becoming
violent. Alfaro (1981) states that the relationship between abuse and delinquency is not
direct and may be affected by certain factors. Alfaro continued to explain that factors
such as community services including parenting classes, counseling, and foster care may
help to weaken the relationship between child abuse and delinquency. The ﬁndings of this
study do not support Alfaro's argument.

In conclusion, the ﬁndings of this research have been determined to be mostly
inconsistent with the ﬁndings of previous research. Past research has determined that
there is a relationship between physical abuse and violent crime and a relationship between
sexual abuse and sexual delinquency. However, the results of this study show that there is
no relationship between child abuse and type of delinquency. Speciﬁcally there is no

129

relationship between physical abuse and violent crime or sexual abuse and sexual
delinquency. However, there has been some consistency with past research. This research
is similar to research by Widom (1989). She concludes as does this research that there is
an eﬂ‘ect to the relationship between abuse and delinquency when examining race,
especially for the Black race. This research is also consistent with past research which
shows that there is an intrafamilial relationship of abuse. The majority of abused
respondents within this sample suffered the abuse at the hands of their parents. Although
the ﬁndings of this research are not completely censistent with ﬁndings of past research,
they show some results which call for further inquiry in future research.
am

As previously mentioned, a signiﬁcant limitation of this study is the small sample
size. This relatively small sample size has limited the use of statistical procedures which in
turn has limited the results of the study. Although statistical signiﬁcance was not found
for any relationship, this result may have changed had the sample size been larger. Chi
square, which determined dependence, is affected by the sample size, thus suggesting that
had the sample been larger then the results of Chi square would have been different. The
small sample size also limited the statistical use of control variables. It was difﬁcult to
conduct statistical procedures for control variables because the sample would be
subdivided into very small cells which limits the useﬁrlness of statistical procedures.
Therefore, the sample size had a direct impact upon the results of this study.

The researcher strongly suggests for ﬁlture researchers to expand the sample size.
This could be accomplished by adding additional facilities to broaden the number of
potential respondents. It is recommended that future researchers not undertake a study of
this magnitude without a sample of at least 100 respondents. Any sample size smaller
would be an attempt of futility. The sample size affected what statistics could be used and
aﬂ‘ected the results of the statistics that were used. Had the sample size been larger the

researcher could have used an analysis of variance (ANOVA) statistics. The researcher

130

relationship between physical abuse and violent crime or sexual abuse and sexual
delinquency. However, there has been some consistency with past research. This research
is similar to research by Widom (1989). She concludes as does this research that there is
an eﬁ'ect to the relationship between abuse and delinquency when examining race,
especially for the Black race. This research is also consistent with past research which
shows that there is an intrafamilial relationship of abuse. The majority of abused
respondents within this sample suffered the abuse at the hands of their parents. Although
the ﬁndings of this research are not completely consistent with ﬁndings of past research,
they show some results which call for ﬁrrther inquiry in future research.
Limitations of RCW

As previously mentioned, a signiﬁcant limitation of this study is the small sample
size. This relatively small sample size has limited the use of statistical procedures which in
turn has limited the results of the study. Although statistical signiﬁcance was not found
for any relationship, this result may have changed had the sample size been larger. Chi
square, which determined dependence, is affected by the sample size, thus suggesting that
had the sample been larger then the results of Chi square would have been different. The
small sample size also limited the statistical use of control variables. It was difficult to
conduct statistical procedures for control variables because the sample would be
subdivided into very small cells which limits the useﬁrlness of statistical procedures.
Therefore, the sample size had a direct impact upon the results of this study.

The researcher strongly suggests for ﬁlture researchers to expand the sample size.
This could be accomplished by adding additional facilities to broaden the number of
potential respondents. It is recommended that future researchers not undertake a study of
this magnitude without a sample of at least 100 respondents. Any sample size smaller
would be an attempt of futility. The sample size affected what statistics could be used and
aﬂ‘ected the results of the statistics that were used. Had the sample size been larger the

researcher could have used an analysis of variance (ANOVA) statistics. The researcher

131

could have also used more comprehensive combinations of control variables.
Unfortunately, with a small sample size any combination of control variables proved too
small to conduct statistics. Finally, a larger sample size would have aﬂ'ected the results of
Chi square which was used to determine dependence.

One explanation of the small sample size is the requirement to obtain parental
consent. The parents of prospective respondents were reluctant to provide consent partly
because of the relationship between parent and prospective respondent, partly because of
fear that negative information may be divulged, and partly because of lack of
compensation for the parent. These feelings caused a small number of parents to provide
consent.

Another limitation of this study is the lack of any comparison group. The
researcher tried unsuccessﬁllly to obtain a non-delinquent sample. However, the attempts
were futile. The researcher believes that when administrators of potential non-delinquent
populations were made aware of the content of the study, speciﬁcally child abuse, they
decided to not be involved. The research is also limited by not randomly selecting cases
for statistical procedures. The lack of random selection limited the researcher from being
able to conduct any inferential statistical analysis. Therefore, any result of the study
applies only to the research sample and not to the general public.

Future researchers should also be conscious of the time and monetary commitment
that a study such as this requires. The number of parental consent forms that were mailed
on three separate occasions is a large ﬁnancial undertaking due to the postage of the
consent forms and the self-addressed stamped envelope, and the copying fees required to
distribute the number of consent forms needed to obtain a sample of over 100. Future
researchers should also be conscious that there is a large gap of time between mailing
consent forms and receiving them back from parents. It is recommended that future
researchers try to obtain funding in order to provide some ﬁnancial assistance to the

demands.

BIBLIOGRAPHY

BIBLIOGRAPHY

Agnew, Robert.(1983). Physical Punishment and Delinquency._Lolsth_stl_d_S_o_ei_ety, 15,.
225-236.

Akers, Ronald. & Krohn, Marvin. (1979) Social Learning and Deviant Behavior. A
Speciﬁc Test ofA General Theory WM_ 44 636-655

Alfaro, Jose" report on Relationship Between Child Abuse and Neglect and Later Socially
Deviant Behavior" ed. Robert & Walker, Yvonne, MW
WWW Allanheld Pub 1981

Bandura, A (1977) Social Learning Theory. Englewood Cliffs, N .J. : Prentice Hall.

Becker, Judith, 'Eﬂ‘ects of Child Sexual Abuse on Adolescent Offenders" ed. Wyatt, Gail

E., & Powell, Gloria in Eff f hil Sage Pub., Newbury Park
Cal, 1988.

Blaske, David. & Bourdin, Charles.(l989). Individual, Family, and Peer Characteristics of
Adolescent Sex Offenders and Assaultive Offenders. ergttmmtsligobology, _2__5_, 846-
855.

Cernkovich, S. & Giordano, P. (1987). Family Relations and Delinquency. Criminology,
25: 295-321.

Doemer, William. (1987). Child Maltreatment seriousness and Juvenile Delinquency.
1W. 2. 197-224.

Fagan, Jeffery. & Wexler, Sandra. (1987). Family Origins of Violent Delinquents.
Criminology, L5. 643-669.

Farber, E. & Joesoph, J. (1978). The Maltreated Adolescent: Patterns of Physical Abuse.
WM, 2: 201-206.

Federal Bureau of Investigation, 1992 Uniform Crime Reports.

132

133

Garbarino, James "Child Abuse and Juvenile Delinquency : The Development Impact of
Social Isolation“ ed. Hunner, Robert & Walker, Yvonne, Exploring Rolstionship Em

Wm Allanheld Pub- 1981-

Gelles, R (1980). A Proﬁle of violence towards children in the United States. in Straus,

M. & Geller, R (eds) hi ' 10 i h ' F '1 (pp 82-
105). New York : Anchor/ Doubleday.

Hetherington, E.M., Stouwie, RJ., & Ridberg EH. (1971). Patterns of Family and Child
Rearing Attitudes Related to Three Dimensions of Juvenile Delinquency. 1ou_rr_ral of

Wm 28.. 160-176.

Hill, G. & Atkinson, M. (1982). Gender, Familial Control, and Delinquency. Criminology.
2o: 127-49.

Howes, C. & Espinosa, M. (1978). The Consequnce of Child Abuse for the Formation of
Relationships with Peers. CbinAbusooodﬂsng, _9__: 397-404.

Kratcoski, Peter. & Kratcoski, Lucille. (1983). The Relationship of Victimization Through
Child Abuse to Aggressive Delinquent Behavior. Moat, 1, 199-203.

Kruttschitt, Candance. & Ward, David(l987). Abuse Resistant Youth: Some
Factors which may Inhibit Violent Criminalm Behavior. Sooiabﬁobcos, do, 501-
517.

Mrazek, Patricia & Kempe, Henry 1 A hil r Th 'r F ili .
Pergarnon Press, Oxford 1981.

Michigan State Police. (1991). Child Abuse in Michigan.

Michigan Department of Social Services. (1992). Major Trends in Michigan's
Children Protective Services Fiscal Years 80-81 through 89-90. .

Pfouts, Jane, Schopler, Janice, & Henley, Carl ”Deviant Behavior of Child Victims
and Bystanders in Violent Families" ed. Hunner, Robert & Walker, Yvonne,

l 'n ' nshi B n hil A lin en Allanheld Pub.

1981.

Ryan, Gail & Lane, Sandy WM Lexington Books, Lexington
1991.

Skinner, B. F. ( 1953) Science and Human Behavior. New York: MacMillan.

Sutherland, E. (1939) Principles in Criminology. Chicago : J. B. Lippencott Co.

134

Welsh, Ralph. (1978). Delinquency, Corporal Punishment, and the Schools. Crims
W19. 336-355.

White, Jennifer. & Mofﬁtt, Terrie. (1980).Predictors of Childhood Conduct
Disorder and Adolescent Delinquency. Criminology, 3. 507-523.

GLOSSARY

136

750.5201). First degree criminal sexual conduct

Sec. 520b. (l) A person is guilty of criminal sexual conduct in the ﬁrst
degree if he or she engages in sexual penetration with another person and if
any of the following circumstances exists:

(a) That other person is under 13 years of age.

(b) That other person is at least 13 but less than 16 years of age and any of
the following:

(i) The actor is a member of the same household as the victim.

(ii) The actor is related to the victim by blood or affinity to the fourth
degree.

(iii) The actor is in a position of authority over the victim and used this
authority to coerce the victim to submit.

(c) Sexual penetration occurs under circumstances involving the commis-
sion of any other felony.

(d) The actor is aided or abetted by 1 or more other persons and either of
the following circumstances exists:

(i) The actor knows or has reason to know that the victim is mentally
incapable, mentally incapacitated, or physically helpless.

(ii) The actor uses force or coercion to accomplish the sexual penetration.
Force or coercion includes but is not limited to any of the circumstances
listed in subdivision (f)(i) to (v).

(e) The actor is armed with a weapon or any article used or fashioned in a
manner to lead the victim to reasonably believe it to be a weapon.

(f) The actor causes personal injury to the victim and force or coercion is
used to accomplish sexual penetration. Force or coercion includes but is not
limited to any of the following circumstances:

(i) When the actor overcomes the victim through the actual application of
physical force or physical violence.

(ii) When the actor coerces the victim to submit by threatening to use force
or violence on the victim, and the victim believes that the actor has the
present ability to execute these threats.

(iii) When the actor coerces the victim to submit by threatening to retaliate
in the future against the victim, or any other person, and the victim believes
that the actor has the ability to execute this threat. As used in this subdivi-
sion. “to retaliate" includes threats of physical punishment, kidnapping, or
extortion.

(iv) When the actor engages in the medical treatment or examination of the
victim in a manner or for purposes which are medically recognized as
unethical or unacceptable.

(v) When the actor, through concealment or by the element of surprise. is
able to overcome the victim.

'Assault" is any intentional. unlawful offer
ofccsponlinjrn'ytoanotherbyforceorforce
directed to person of another under
such circumstances as create a well-founded
fear of imminent peril. coupled with apparent
present ability to effectuate attempt if not pre-
vented. Peeple v. Bryant (1978) 264 N.W.2d
13. 80 MichApp. 428; People v. Carlson (1910)
125 N.W. 361. 160 Mich. 426, 136 AmSLRep.
447.

137

5. Battery—Natureandelemenssofoﬁune

'Battery'isthewillfultouchingofpersonof
anotherby awesororbysome substanceput
inmotionbyhim; a‘battery'istbecomum-
mation of the 391111. People v. Bryant (1978)
264 N.W.2d 13. 80 MichApp. 428.

750.316. First degree mtu'der

Sec. 316. Murder which is perpetrated by means of poison, lying in wait,
or other wilful, deliberate, and premeditated killing, or which is committed in
the perpetration, or attempt to perpetrate arson, criminal sexual conduct in
the ﬁrst or third degree, robbery, breaking and entering of a dwelling. larceny
of any kind, extortion, or kidnapping, is murder of the ﬁrst degree, and shall
be punished by imprisonment for life.

750.317. Second degree murder

Sec. 317. Saccnn pecans MURDER—All other kinds of murder shall be mur-
der of the second degree, and shall be punished by imprisonment in the state
prisonforlife, oranytermofyears, inthe discretionofthecourttryingthe
same.

750.529. Armed robbery; aggravated assault

Sec. 529. Any person who shall assault another, and shall feloniously rob,
steal and take from his person, or in his presence, any money or other

property, which may be the subject of larceny, such robber being armed with
a dangerous weapon, or any article used or fashioned in a manner to lead the
person so assaulted to reasonably believe it to be a dangerous weapon, shall
be guilty of a felony, punishable by imprisonment in the state prison for life
or for any term of years. If an aggravated assault or serious injury is
inflicted by any person while committing an armed robbery as defined in this
section, the sentence shall be not less than 2 years' imprisonment in the state
pnson. '

750.448. Soliciting and accosting

Sec. 448. Any person, male or female, 17 years of age or older, who shall
accost, solicit or invite another in any public place, or in or from any building
or vehicle, by word, gesture or any other means, to commit prostitution or to
do any other lewd or immoral act, shall be guilty of a misdemeanor.

750.71. Deﬁnitions

Sec. 71. DEFINITION or “BURN"—The term "burn" as used in this chapter
shall mean setting fire to, or doing any act which results in the starting of a
ﬁre, or aiding, counseling, inducing, persuading or procuring another to do
such act or acts. '

750.3773. Malicious destruction of personalty

Sec. 377a. Any person who shall wilquy and maliciously destroy or injure
the personal property of another, by any means not particularly mentioned or
described in the preceding section, if the damage resulting from such injury
shall excwd $100.00, shall be guilty of a felony. If the damage done shall be
$100.00 or less, such person shall be guilty of a misdemeanor.

138

750.5%. Second degree criminal sexual conduct

Sec. 520c. (l) A person is guilty of criminal sexual conduct in the second
degree if the person engages in sexual contact with another person and if any
of the following circumstances exists:

(a) That other person is under 13 years of age.

(b) Thatotherpersonisatleast 13 but lessthan 16 yearsofageand anyof
the following:

(i) The actor is a member of the same household as the victim.

(ii) The actor is related by blood or affinity to the fourth degree to the
victim. ‘

(iii) The actor is in a position of authority over the victim and the actor
used this authority to coerce the victim to submit.

(c) Sexual contact occurs under circumstances involving the commission of
any other felony. '

(d) The actor is aided or abetted by 1 or more other persons and either of
the following circumstances exists: .

(i) The actor knows or has reason to know that the victim 'is mentally
incapable, mentally incapacitated, or physically helpless.

(ii) The actor uses force or coercion to accomplish the sexual contact.
Force or coercion includes but is not limited to any of the circumstances
listed in sections 520b(1)(f)(i) to (v).'

(e) The actor is armed with a weapon, or any article used or fashioned in a
manner to lead a person to reasonably believe it to be a weapon.

(1’) The actor causes personal injury to the victim and force or coercion is
used to accomplish the sexual contact. Force or coercion includes but is not
limited to any of the circumstances listed in section 520b(1)(f)(i) to (v).

(z) The actor causes personal injury to the victim and the actor knows or
has reason to know that the victim is mentally incapable, mentally incapaci-
tated, or physically helpless.

(h) That other person is mentally incapable, mentally disabled, mentally
incapacitated, or physically helpless, and any of the following: .

(i) The actor is related to the victim by blood or affinity to the fa

750.520d. Third degree criminal sexual conduct

Sec. 520d. (1) A person is guilty of criminal sexual conduct in the third
degree if the person engages in sexual penetration with another person and if
any of the following circumstances exists: -

(a) That other person is at least 13 years of age and under 16 years of age.

(b) Force or coercion is used to acc0mplish the sexual penetration. Force
or coercion includes but is not limited to any of the circumstances listed in
section 520b(1)(f)(i) to (v).'

V (c) The actor knows or has reason to know that the victim is mentally

incapable, mentally incapacitated, or physically helpless.

(2) Criminal sexual conduct in the third degree is a felony punishable by
imprisonment for not more than 15 years.

Amended by P.AJ983, No. 158, § 1. Eff. March 29. 1984.
' Section 750.520u1)(f)(i) to (v).

139

750.520e. Pom-th degree criminal sexual conduct

Sec. 520e. (1)A person is guilty of criminal sexual conduct in the fourth
degree if he or she engages in sexual contact with another person and if any
of the following circumstances exists:

(a) Force or coercion is used to accomplish the sexual contact. Force or
coercion includes but is not limited to any of the circumstances listed in
section 520b(1)(f)(i) to (iv).‘

(b) The actor knows or has reason to know that the victim is mentally
incapable, mentally incapacitated, or physically helpless.

(c) That other person is under the jurisdiction of the department of correc-
tions, and the actor is an employee or a contractual employee of, or a
volunteer with, the department of corrections who has knowledge that the
other person is under the jurisdiction of the department of corrections.

(2) Criminal sexual conduct in the fourth degree is a misdemeanor punisha-
ble by imprisonment for not more than 2 years, or by a ﬁne of not more than
$511100, or both.

Amended by P.A.l983. No. 158. § 1, Eff. March 29, 1984; P.A.l988, No. 86, § 1, Eff.
June 1.
'Scction 750.520b(1)(f)(i) to (iv).

APPENDIX A

QUESTIONNAIRE

141

Hello,

My name is Robert Keller. 1 am a graduate student at Michigan State University. In most
cases, the university requires that students who want to get a graduate degree must go out
to the real world and study real life problems, and propose solutions to them. As part of
the requirements for my Master’s degree, I am conducting a study at some youth facilities
in this area. The purpose of this study is to determine what makes young people get in
trouble with the law. It is expected that the ﬁndings of the study will not only give us some
understanding of this problem. I hope that they will also help us devel0p effective
intervention program for youth in trouble, and proper programs to prevent others from
getting in trouble. Your participation in this study and your honest answers are extremely
important. First, they will help us understand things that happen to young people that may
be unpleasant. Second, they will help us understand youths’ point of view on why they get
in trouble with the law, they types of trouble youths get into and how best to prevent such
occurrences. '

This study is not conducted by your youth facility or for your youth facility. It is not
conducted by or for the court or any agency or person connected with your adjudication.
It will not affect your placement, level, status or privileges at this facility or with the court
or other agencies or persons involved in your adjudication or treatment. It will not affect
the length of time you will stay at this facility. It is not part of any treatment

Your participation in this study is voluntary. Your answers will be held strictly conﬁdential
and nobody, including me the investigator, will be able to trace the questionnaire to you.
The numbers on the questionnaire are Code numbers. They will be used only to keep count
of the questionnaires. Nobody, including me will be able to use them to identify you.
Therefore, DO NOT put your name or other identifying marks on your questionnaire. I
assure you that your answers will be kept strictly confidential and anonymom.

You may be wondering if your completed questionnaires will be shown to the staff at your
facility. No, they will not be shown to anybody. I will be the only person handling the
questionnaires once you have completed them. I will be glad to share the ﬁnal reports with
you, the staff at your facility or any other interested person. The reports will not contain
your name or anything that will identify you.

Please answer all questions honestly and to the best of your knowledge. However, you may
omit or leave blank any questions that you do not wish to answer. Also you are free to
withdraw from participation at any time during this session. You indicate your voluntary
agreement to participate by completing and returning your questionnaire. Once you
complete the questionnaire, please place it in the unmarked brown envelope in which the
questionnaire will be distributed. Close the envelope and place it in the designated box in
the room. Do not put your name or any other identifying mark on the envelope.

Thank you very much for your participation. If you wish to obtain a copy of the ﬁnal
reports, you may indicate so to your treatment director who will inform about how many
copies are requested at your facility. I will send the copies to the director for distribution
to those who requested them.

142

Please check the correct box with the answer you feel is correct
1. Sex: El 1. male E] 2. female

2. Your race: [:1 1. Black [3 2. Hispanic D 3. Native American [3 4. Oriental
[:1 5. White [3 6. other (please specify)

 

3. Age __ years.
4. How long have you lived in this facility? months.

5. Have you lived in other facilities?
D 1. yes D 2.. no (if no go to question 6)

if yes, how many facilities?

6. Before coming to this facility, what city were you living in?

 

7. Who were you living with?

 

 

El 1. mother D 2. mother and stepfather or boyfriend
Cl 3. mother and father CI 4. father and stepmother or girlfriend
D 5. grand mother / grand father
U 3. step parent(s) CI 6. aunt or uncle
U 4. foster parent(s) El 7. other
El 8. adoptive parents (please tell who)
8. Parents are: E] 1. Married D 2. Separated Cl 3. Divorced El 4. Never married
D 5. Other
9. Have you ever been disciplined?
U 1. yes Cl 2. no (if no go to question 16)
10. If yes, who would discipline you? (check all that apply)
[1 1. mother [I 5. step mother
D 2. father E] 6. grand mother
[I 3. step father D 7. grandfather

D 4. foster parent C] 8. other
[I 9. adoptive parent

 

143

ll 1. For what reasons were you disciplined? (if more than one check all that apply)
D 1. late for curfew
U 2. did not complete my chores.
D 3. wet the bed
['34. talked backtoan adult

D 5. other
(please tell what)

 

12. How were you disciplined? (if more than one check all that apply).
1. spanking

2. hitting with ﬁsts

3. hitting with an object (belt, board)

4. loss of privileges

5. other

DECIDE]

 

13. How often were you disciplined?
D 1. once a week
El 2. twice a week
El 3. once a month
C] 4. twice a month.
D 5. other

 

14. Did you ever receive any injuries when you were disciplined?
El 1. yes El 2. no (if no go to question 16)

15. If yes, what type of injuries? (check all that apply)
E] l. bruises
D 2. stitches
El 3. broken bone(s)
D 4. black eye
[:1 5. knocked unconscious
C] 6. other

 

16. Do you have any sisters?
E] 1. yes D 2. no (if no go to question 17)

if yes, how many?

17. Do you have any brothers?

I: 1. yes D 2. no (if no go to question 18)
if yes, how many? .

 

144

18. Were your brother(s) or sister(s) ever hit by the people they lived with?
El 1. yes

I] 2. no

When people live together they often have arguments for several reasons. Arguments are part
of living closely with one another. The next set of questions asks about those arguments between
people you were living with before being put in placement.

l9.Did the people you live with ever argue with each other?
U 1. yes
[:1 2. no (if no go to question 21)

20. If yes, how often?
[:1 1. always
[3 2. sometimes
E] 3. never

21. Did the people you were living with ever hit each other when they were arguing?
E] 1. yes, always
[:1 2. yes, sometimes
[3 3. no, never

This group of questions asks about why you are in placement and what kind of trouble with the
police you have been involved in. Remember that these questions are only for the purpose of
this study and in no way will answering them affect your placement, level or status at this
facility.

22. Type of offense leading to your present adjudication?
E] l. murder

. criminal assault

. vandalism

. robbery

. theft

. prostitution

. rape (csc 1, II, III, or IV)

. truancy

. other (please specify)

DDDDDDDD
\OOOQOMrhUJh-l

145

23. This adjudication was your being adjudicated.
E] 1. ﬁrst time (if ﬁrst time, go to question 25)
D 2. second time
[:1 3. third time
[I] 4. fourth time
D 5. more than four times (please tell how many times__)

24. Apart from the present adjudication, what other types of crimes have you been adjudicated
for? (if more than one check all that apply).

 

El 1. theft
[:1 2. vandalism
El 3. criminal assault
El 4. murder 1:] 7. truancy
D 5. criminal sexual conduct (csc) D 8. other
Cl 6. prostitution (please tell what)
25. How old were you when adjudicated for your ﬁrst time? years.

 

Many children have adults do wrong stuff to them. These questions may get very personal.
Please try to answer them the best way possible. Again, these questions will only be used for
this study and they will be kept only by the researcher and not given to anyone else. There will
be no way of placing your name with. any of your answers. .

26. As a child did an adult ever physically hurt you?
1. 13 yes 2. El no (if no go to question 31)

27. If yes, please indicate who physically hurt you and in what way by putting a check ( ) in
the boxes below. You may ﬁll as many boxes that applies

spanked beaten punched hit with hot water other

object poured on (please
body write)
1. father El 1:] D D
2. mother U El D E] D
3. foster parent '3 D D D El
4. aunt D D U E] El
5. uncle D E] El Cl El
6. step parent D E] E] D D
7. other people D E] D D E]

(Please SPOCify)

146

28. How old were you when you were ﬁrst physically hurt

by an adult? __ years.

29. How often did an adult physically hurt you?
1. [:1 every day [I 3. more than once a week
2. C] more than once a day D 4. other

 

(please indicate)
30. How long did an adult physically hurt you? months

 

31. Concerning being physically hurt by an adult when you were a child,
which of these following statements is true for you?
(if more than one is one check all that is me)

1. [I] I told somebody about it.
2.131 did not tell anybody about it.

3. El Somebody found out about it.
(state who and how ).

4.131 saw a doctor who treated the injures 1
received when I was spanked.

5. El I saw a therapist. .

6. D I saw a social worker.

7. D 1 saw a protective service worker
8. Cl Nobody ever found out. 10. D police came and took me from my home
9. D l was placed in a foster home. 11. D other

 

33. Did anyone ever make you do inappropriate sexual stuff T_O THEM?
1. El 1. yes 2. [3 no (if no go to question 38)

34. If yes, please indicate who and what happened by putting a check ( ) in the correct box
below. You may ﬁll as many boxes as applies?

touched masturbated fondled had sex other (please
private private indicate
parts parts what)

1. father 1:] Cl 1:] El El

2. mother I] U D D D

3. foster parent I] Cl D D D

4. aunt U D D D 1:]

5. uncle Cl 1:] D [I D

6. step parent El [:1 D C] D

7. other people 1:] D D D 1:]

(please Specify)

147

35. How old were you when this ﬁrst happened? years old.
36. How often did this occur?
D 1. every day 1:] 3. more than once a week
[:1 2. more than once a day 1:] 4. other

 

(please indicate)
37. How long did this occur? months.

38. Did anybody ever do inappropriate sexual stuff TO YOU?
1. [:1 yes 2. [3 no (if no go to question 43)

39. If yes, please indicate who and what happened by putting a check ( ) in the correct box.
You may check as many boxes as applies.

touched masturbated fondled had sex other (please
private private indicate
parts parts what)
1. father [:1 ‘ E] El [:1 El
2. mother Cl [:1 U U D
3. foster parent D D D D D
4. aunt [3 Cl C] C] [:1
5. uncle D D D E] El
6. step parent El [3 U E] D
7. other people D D E] El 1:]
(please M)
39. How old were you when this ﬁrst happened? years old.

40. How often did this occur?
[:1 1. every day Cl 3. more than once a week
U 2. more than once a day 1:] 4. other

 

(please indicate)

41. How long did this occur? months.

148

42.Concerning the inappropriate sexual stuff by or with an adult when you were a child,

which of these following statements are true for you?

(check all that apply)
D l. I told somebody about it.

D 2. I did not tell anybody about it.
E] 3. Somebody found out about it.

state who

 

D 4. 1 saw a doctor who treated the injm'es 1

received.
[:1 5. I saw a therapist.
I: 6. I saw a social worker.

[I 7. 1 saw a protective service worker.

E] 8. Nobody ever found out.
1:] 9. Placed in a foster home.
[:1 10. Police came.

43. Many chidren do things that the law prohibits. The following are some of those things.
Please check the correct box for the number of times that you may have been involved in these
activities. It is important to make sure that the correct answer box matches with the question that

you are answering. Also, if more than one box applies check all the appropriate boxes.

_1. drunk so much that I could not remember
afterward some of the things you had done.

2. used alcohol excessively

never 1 or 2

E]

El

3. gone for a ride in a car someone had stolen Cl

4. taken part in a "gang ﬁght"

5. carried a switchblade or other weapon.

6. bought or drank beer, wine, or other liquor
7. used narcotics other than marijuana.

8. smoked marijuana

Cl
E]
Cl
[:1
CI

El

DDDDDDD

several

times

[I]

DDDDDDD

often

Cl

DDDDDDD

ESE!
often -

El

DDDDUDD

149

mm

9. sold hard drugs such as heroin, coke, LSD U 1:]

10. sold marijuana to someone D D

11. taken part in a robbery involving the use El D
of physical force.

12. taken part in a robbery involving the use [:1 D
of a weapon. *

l3. taken part in a robbery. [I Cl

14. resisted arrest or fought with an ofﬁcer E] El
trying to arrest me.

15. hit a teacher or another adult [I] D

16. attacked or fought an ofﬁcer dying to D D
arrest someone.

17. thrown eggs, tomatoes, garbage, or anything-:1 El
else like this at a person, house, or building.

18. broken out any windows. E] E]

19. put paint on anything I wasn’t supposed E] El

' to be painting.

20. broken out any light bulbs on the street CI 1:]
or anywhere else.

21. taken things of value($2 to $50) that I C] C]
was not supposed to take.

22. taken a car for a ride without the D [:1

owner’s permission.

23. taken things of large value (over $50). D [:1

several
times

 

 

Ell]

CID

m

Molten
DC]
DE]
DE]
CID
DE]
DE!
DE]
DE]
EIEI
DE]
CID
DE!
DE]
DE!
DE!

150

never 1 or 2

 

24. skipped school without a legitimate excuse. [:1

25. taken money from home without the D
intention of returning it.

26. stolen or tried to steal a motor vehicle D
such as a car or a motorcycle.

27. ran away from home. El

28. attacked someone with the idea of seriously 1:]
hurting/killing them.

29. been paid for having sexual intercourse E]
with someone.

30. Had or uied to have sex with someone D
against their will.

 

CID

C]

DE]

 

acme xsrx

times 2mm
[:1 C] E]
[:1 Cl
Cl E] E]
E] E] E]
[1 Cl Cl
C] D E]
El [1 El

APPENDIX B

CONSENT FORM

152

MICHIGAN STATE UNIVERSITY

 

COMOFSOGAlmCEOSCﬂOOLOFWJUm EASTWG'IICHIGANOWIIIS
mum

Dear PW‘E‘JT",

My name is Robert Keller. I am a graduate student at Michigan State University. In most
cases. the university requires that the students who want a graduate degree must go out to
the real world and study real life problems. and propose solutions to them As part of the
requirements for my Masters degree, I am conducting a study at some youth facilities in this
area including the facility where your child is a resident. The purpose of the study is to
determine what makes young people-get in trouble with the law. It is expeded that the
findings of the study will not only give us some understanding of this problem I hepe that
they will also help us develop effective intervention programs for youth in trouble, and proper
programs to prevent others from getting in trouble.

This study is not conducted by your child‘s youth facility or for your child's youth facility. It is
not conducted by or for the court or any agency or person connected with your adjudication.
lt will not affect your child's placement. level, status. or privileges at their facility or with the
court or other agencies Or pemons involved in your child's adjudication or treatment. It will
not affect your child's length of stay at this facility. lt is not part of any treatment. Your child's
participation in this study is voluntary. Your child‘s answers will be held strictly confidential
and nobody, including me , the investigator, will be able to trace the questionnaire to your
child.

You may be wondering if your child‘s completed questionnaire will be shown to the staff at
your facility. NO. they will not be shown to anybody. I will be the only person handling the
questionnaires once your child has completed them. I will be glad to share the final reports
with you, the staff at your child’s facility or any other interested person. The reports will not
contain your child's name or anything that will identify him or her. .

The reason I am writing is to ask your consent so that your child may participate in this study
by answering a questionnaire. Your child must also agree to participate in the study.
However. he has a choice to participate or not to participate. Should your child decide to
participate in the study. he or she will not be asked his or her name or to identify themselves
in any way. Also, your child will be able to withdraw at any time without any penalty. and he
or she can decline to answer any specific question(s). It may require 45 minutes total time
for your child to con'plefe the questionnaire. As indicated above, this study is part of my
degree requirements. It is not part of any treatment. Whether your child participates or not.
there will be no negative consequences to your child' privileges. level, status. or program
placement at the child‘s facility. Your child will remain anonymous in any report of the
findings from the study.

Thank you for reading this letter and for your support of this study. Please sign the attached
consent form and place it into the enclosed self-addreesed. stamped envelope and send it to
me in the mail. Please contact me. Robert Keller. or Dr. Rosy Ekpenyoung at (517) 355-
0765 if you have any questions or concerns.

Si , rely,
. t C‘f
Robert Keller
MSU is an Woe Action/Equal Opportunity Institution

 

 

153

PARENT CONSENT FORM

My son. (name and Date Of
Birth) who is in placement at has my authorization to participate in a
study conducted by Robert Keller. a graduate student of Michigan State University. I
understand that my child has the option to participate but first my consent and signature are
needed. I also understand that my child can withdraw at any time and that all answers will
be kept confidential. i understand that there will be no harm or penalty to my child. I
understand that this study will not. affect in any way my son's level, status. placement.

treatment. or length of stay.

'I give my consent that my child may agree to participate in this study. I know that the

results of the study will be kept in strict confidentiality and not harm my child in any way.”

 

(your signature and date)

APPENDIX C

LETTERS FROM INSTITUTIONS AND UNIVERSITY COMMITTEE ON
RESEARCH INVOLVING HUMAN SUBJECTS

155

STIHTECM=RMC34K§AN

aces

5“:

     
 

MMLEEGov-mor

DEPARTMENT OF SOCIAL SERVICES

ZSSSommcmendAwomumFUD.Box30GELlamﬂthMbhbanIHEMB
GauupHJluanmumu

March 11, 1993

To whom it May Concern:

This letter confirms that Robert Keller has received the necessary
approval from the HDSS Division of Planning and Evaluation to proceed
with his proposed study.

Further he has discussed both his methodology and intended use of the
collected data with the local GVRC administrator. We are supportive
of his efforts and will be interested in reviewing the results.

ﬂaw

,Judy Arthur,, Director
Genesee Valley Regional Center
4287 H. Pasadena Avenue

Flint, MI 48504

1 cerely,

 
    

sgol

156

STATE OF MICHIGAN
m

JOHN ENGLER, Governor

DEPARTMENT OF SOCIAL SERVICES

235 South Grand Avenue. PO. Box 30037, Lansing. Michigan 48909
GERALD H. MILLER. Director

March4.1993
RobertKeller
c/oDr.RosyEI<penyoung
SGOBakerI-Idl
ELans’ng,Ml48864
Deaer.KelIer.

ThisletteristoinformyoutlutyulrequwtodoasmdyusingceneaeeVaneyRegiordCeuerlrmes
asresearchsubjectshasbeenapproved. Thisapp'ovaliscontingentonthefollowingcondltions:

1.Asrequiredmdersecﬁon7(1)(i)ofﬂieChidProteaionLaw,meDepamnaidSocidSavices
(DSS)cannotreIaseanyidentﬂyinginfomiationonanyperson namedinthecasereoord
wwwmmmenconsemammpemwmmybemudedwﬂw
awrittenconsent
ThematicmforﬂupasomlﬂewiewmtstspecﬂywtmmedMiszdo.

2. lnformationon hdlvidual clients must be kept conﬁdential.

3.AnyproblemsorcomplaiisiwdvingDSSclientsorstaﬂmstberepatedpromptlytothelocal
ofﬁoedirectorandtothisoffice.

4.Anychangoshueprofectdesimmbesummwtomisdﬂcetismwesmnsbnto
othercomties, badmlondotherpmgramareasand/ordieMgmmanimhmesanue
sfze,etc

5.AoopyoftheshrdyresdtsMbesubrnlttedtothisofﬁce.

ﬂyouammuembeghﬂusnnymmenensbtmrﬂBuneedtocaﬁmﬂnproieabeyau
eighemmutspleaseletmelmow.

Ifyouneedanyfutheraasfaanceorhaveanyquestions. pleasecontactAlanLabavltzat (517)335~3943.

  

s. ,! jl ,

Plam’mgandEvalmtionDivision

CC:NancyDuncan AlanLabavltz JudiArthur

1 5 7 Camp Highﬁelds
Special Families
Breakthrough
In Home Family. Carr.
First Contact

 

, .
Ll. 11ers people mama

22 March 1993
To whom it may concern,

Mr. Robert Keller has requested my permission to survey students
currently in residence at Camp Highfields. He has reviewed the
survey with me and explained how it will be administered and the
information used.

He has my permission to conduct this survey with the students at
Highfields under the following conditions:

1. that he gain written authorization from the parent or
guardian of each student to be surveyed,

2. .that he inform all students in Camp Highfields of the
nature and purpose of this survey, and that participation
in it is completely voluntary,

3. that he survey only those students who agree to
participate and whose parent/guardian has given written
authorization for this participation, and

4. that the identity of the students participating in this
survey be kept strictly confidetial.

Mr. Keller is aware of these conditions, and I know he will insure
that they are met. He is also aware that I am interested in seeing
a copy of his thesis once it has been completed.

   

es H. Hines, ACSW, BCD
irector of Treatment

Highﬁelds. Inc.
PI). Box 9"
Onondaga, Ml 49264
(317) 528-237
Fax 15!?) 628-343!

158

\VJN III II @COD (0)? nnTii©Irliii9©ml

I ORPORA ED

bow

3744mm“
WWW
[51714237451

ﬂl" I'OPﬁ

WWW

WWW

Munch 11, 1993

De. Rosy Elepwoung
Michigan WWW
560 mm
mm, MI. 48864

Dean DIL. Hepenyoung,

Plensebeddvuedthntm. Robentkelm, dgmduatesmdeniinym
mm, hesbeenineoniaotmihmnegmdinghisgnedualeﬂtws

m.KeuuiAemidaumetheuAeo‘smBoysvil€eWcuentsns

panto‘hiiaeseneehpopula/uon‘oenésﬂtesis. Wehdveducussed
«thepaniimldnso‘suehanmdeavoesevualmmdmmmking
mummmmww, 1993,Many
WW.

W,mm6imﬁzingthenechmim6msewningm€
We‘lhosedientsmommmheinthesmdy. Itisnfy
undenstandmgaﬂutm. Kellenhnsshanedﬂtesemocedunesuwym.

Atﬂvcstime, Boysvillebspeeasedatonssiltm. KelEeninﬂtis
memhstudyandmmopmtonummelientstompwte,
giumﬂteinvolvenmuvollmmy, andmenlnlpwnossmuneeeiyed.
:5 youhcweenyquestiansdbcwtthedboveinsmimonmd
Wdeiails, ptensecontncimeatﬁuWWPnogwn,
(517) 423-7451.

Thanhyou.

Sineenely,

comm“;

81. Chest”. Faeel, csc, Acsw
Regional We».
Boysville 04 W

cc: Robe/pt Kellen.

 

OFFICE or
RESEARCH
AND
GRADUATE
STUDIES

UniveraIty Commas on
Russell Involving
Rama Subjects
(UCRIHS)

Michigan State University
225 Administration Building
East Lansing, Michigan
488244046

517/355-2180

FAX: 517/336-1171

MSU is an a/ﬁmiarive-adion.
equal-opportunity insfimfron.

159

MICHIGAN STATE

 

  

UNIVERSITY
Iune9,l993

TO: Robert Keller
2375 Club Meridian #311
Okemos, MI 48864

RE: IRB #: 93-186
TITLE: UNDERSTANDINGTHERELATIONSHIP BETWEEN CHILDABUSE
AND DELINQUENCY
CATEGORY: Full Review
REVISION REQUESTED: NIA
APPROVAL DATE: May 3, 1993

The University Committee on Research Involving Human Subjects’ (U CRIHS) review Of this project
is complete. I am pleased to advise that the rights and welfare of the human subjects appear to be
adequately protected and methods to obtain informed consent are appropriate. Therefore, the
UCRIHS approved this project including any revision listed above.

UCRIHS approval is valid for one calendar year, beginning with the approval date shown above.
Investigators planning to continue a project beyond one year must seek updated certiﬁcation. Request
for renewed approval must be accompanied by all four of the following mandatory aswrances.

l. The human subjects protocol is the same as in previous studies.

2. There have been no ill effects suffered by the subjects due to their participation in the study.

3 There have been no complaints by the subjects or their representatives related to their
participation in the study.

4. There has not been a change in the research environment nor new information which would
indicate greater risk to human subjects than that assumed when the protocol was initially
reviewed and approved.

There is a maximum of four such expedited renewals possible. Investigators wishing to continue a
project beyond that time need to submit it again for complete review.

UCRIHS must review any changes in procedures involving human subjects, prior to initiation of the
change. Investigators must notify UCRIHS promptly of any pmhlems (unexpected side effects,
complaints, etc.) involving human subjects during the course of the work.

If we can be of any future help, please do not hesitate to contact us at (517) 355-2180 or FAX (517)

336-1171.

vid E. Wright, Ph.D.
UCRIHS Chair

Sincerely,

DEW :pjm

cc: Dr. Rosy Ekpenyong-Rowan

l. I ’ itllll.1lav.li‘il

 

 

 

"VIIIIIIIIIIIIIIIIIIT