, . 3 PA . II

5%.”.

4‘
.

y
n
W. T.»

. .f

. é?!

ﬂawxdlmum.
1.0.}. . .

4.3%.. .

X .3 n. .K
it. . I. .
z}!!- it!!!
4 1‘5.
.‘tilvfld. 01’
L 5.. . ...£.!l.i‘

 

. . H . v .ii'l'i
evvwm... Lilia-aw

' waft?!“ “

«gag-w

‘ 11
.3‘4! v ‘1.) )1! n [.‘n I!»

nvlfl-lln"lli«la‘fviillanvy I... II

III Pll?.|£ae.lolpll l1, IIIJJ‘I

I‘IIOIviuw \lo. EE'I-’ |, ‘.L

 

its will“millllllimlllmll

a
07 "5315 3 1293 01834 2059 r

 

 

 

 

 

LIBRARY

Michigan State
University

 

 

 

This is to certify that the

thesis entitled

SOCIAL SUPPORT IN THE LIVES OF CHILDREN
EXPOSE TO DOMESTIC VIOLENCE

presented by
Elaine Shpungin

has been accepted towards fulﬁllment
of the requirements for

Masters degree in Psychology

 

 

professor

/, , .
,1 1 “mm, 1/ 1/4 '
Date $1?“

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

 

PLACE IN RETURN Box to remove this checkout from your record.
TO AVOID FINE return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

23.13? 2 4 2010
2122 1D

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

we chlRC/DucDmpﬁS—pu

SOCIAL SUPPORT IN THE LIVES OF CHILDREN
EXPOSED TO DOMESTIC VIOLENCE ’

By

Elaine Shpungin

A THESIS
Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of
MASTER OF ARTS
Department of Psychology

1 999

ABSTRACT

SOCIAL SUPPORT IN THE LIVES OF CHILDREN
EXPOSED TO DOMESTIC VIOLENCE

By

Elaine Shpungin

The impact of social support in the lives of children exposed to their
mothers’ abuse was examined. The behavioral, emotional, social, and academic
adjustment of 80 children, ages 7 to 11, were measured using mother, child, and
school reports. The results revealed that social support in children’s lives had
both a moderating and main effect on their adjustment. Speciﬁcally, the number
of signiﬁcant peOpie in the children’s lives were found to buffer their behavioral
adjustment from the negative effects of witnessed violence. The size of
children’s social networks, as well as support from mothers and other signiﬁcant
adults, were also found to be correlated with positive outcomes. Implications for

social services and community-based interventions are discussed.

To my late grandmother, whose warmth and intellect will always be with me

Clo

ACKNOWLEDGMENTS

I would like to thank, ﬁrst and foremost, my advisor and mentor, Aiytia
Levendosky, for her support and guidance throughout this challenging process.
Her empathy, advice, and patience were invaluable at every step of the way. i
would also like to thank my committee members: Cris Sullivan, Anne Bogat, and
Deb Bybee, each of whom made vital contributions to the conceptualization,
analysis, and writing of my thesis. in addition, I would like to extend my
appreciation to Pennie Foster-Fishman for her encouragement and support
during a bend in the road when I most needed it. I feel fortunate to have had a
group of strong, competent, and successful women to help me on this
demanding journey.

i also want to thank my family, who has always believed in me and has
been there for me in many ways throughout this time. Finally, last but not least, I
would like to thank my husband who has been a beacon of light through the
most arduous parts of this journey, and a fount of laughter and support through

every day of graduate school.

TABLE OF CONTENTS

LIST OF TABLES ............................................... vii
INTRODUCTION ................................................ 1
CHAPTER 1: CHILDREN EXPOSED TO THEIR MOTHERS’ ABUSE ........ 3
Women with abusive partners ................................. 3
Effects on children of witnessing their mothers’ abuse .............. 4
Behavioral difﬁculties .................................. 4

Emotional difﬁculties ................................... 7

Social adjustment .................................... 10

Academic adjustment ................................. 12

CHAPTER 2: PROTECTIVE FACTORS .............................. 14
Risk and Resilience ........................................ 14

Social support and stress .................................... 16

Social support and children exposed to domestic violence .......... 18
CHAPTER 3: HYPOTHESES ...................................... 23
CHAPTER 4: METHOD .......................................... 25
Participants .............................................. 25
Procedure ............................................... 26
Measures . ............................................... 27
Witnessed Violence Scale - Child Report .................. 27

Witnessed Violence Scale - Mother Report ................ 28

Experience of Violence - Mother ......................... 28

Perceived Competence Scale ........................... 29

Childhood Depression Inventory ......................... 29

Behavioral Rating Scale ............................... 30

Number of Suspensions ............................... 30

Mean Semester Grade ................................ 30

Child Social Support Questionnaire ...................... 31

CHAPTER 5: RESULTS .......................................... 33
Demographic Variables ..................................... 34
Violence and Adjustment .................................... 34
Descriptives ......................................... 34

Child Report ........................................ 36

Mother Report ....................................... 36

Social Support and Adjustment ............................... 37
Descriptives ......................................... 37

Buffering ........................................... 39

Main Effect ......................................... 39

CHAPTER 6: DISCUSSION ....................................... 41
Violence and Adjustment .................................... 41
Descriptives ......................................... 41
Results ...... 42

Social Support and Adjustment ............................... 44
Buffering ........................................... 45

Main effects ......................................... 46
REFERENCES ................................................. 55
APPENDICES .................................................. 63
Appendix A: Witnessed Violence Scale - Child Report ............. 64
Appendix B: IMtnessed Violence Scale - Mother Report ............ 66
Appendix C: Experience of Violence - Mother .................... 67
Appendix D: Perceived Competence Scale ...................... 71
Appendix E: Childhood Depression Inventory .................... 77
Appendix F: Behavioral Rating Scale .......................... 80
Appendix G: Number of Suspensions .......................... 82
Appendix H: Child Social Support Questionnaire .................. 83
Appendix I: Table 1 ........................................ 85
Appendix J: Table 2 ........................................ 86
Appendix K: Table 3 ........................................ 87
Appendix L: Table 4 ........................................ 89
Appendix M: Table 5 ....................................... 90
Appendix N: Table 6 ....................................... 91
Appendix 0: Table 7 ....................................... 92
Appendix P: Table 8 ........................................ 93
Appendix Q: Table 9 ....................................... 94

vi

LIST OF TABLES

Table 1: Correlations Among Subscaies of the Behavior Rating Scale ...... 85

Table 2: Descriptives of Continuous Adjustment, Violence,
Social Support, and Demographic Variables ................... 86

Table 3: Correlations of Adjustment, Violence, Social Support,
and Demographic Variables ................................ 87

Table 4: Percent of Children Reporting Witnessing Verbal Conﬂict

and Abuse .............................................. 89
Table 5: Percent of Children Reporting Witnessing Threats

and Physical Abuse ...................................... 90
Table 6: Children’s Last Contact with Biological Fathers ................. 91

Table 7: Frequency of Children’s Contact with Biological Fathers
in the Last 4 Months ...................................... 92

Table 8: Percent of Children Listing Mothers, Fathers, and Assailants
as Social Supports ....................................... 93

Table 9: Number of Confidants as Buffer for Negative Effect of Witnessed
Violence on Children’s Total Behaviors ....................... 94

vii

INTRODUCTION

In the last decade, researchers have turned their attention to the children
of women experiencing partner violence. Although this research has
demonstrated the behavioral difﬁculties experienced by children exposed to
abuse against their mothers in comparison to children from non-violent homes, it
is currently lacking in several important ways. First, the research in this area has
not fully examined how the severity of violence to which the children are exposed
affects their adjustment. Second, factors that may contribute to the resilience of
children exposed to domestic violence have not been explored. in addition, this
literature has given limited attention to several speciﬁc areas of the children’s
development, such as their social and academic adjustment. The current
research also contains little data on children’s own reports of their adjustment.
Finally, this literature is limited by its sole reliance on descriptive measurements
of children’s adjustment in areas such as academic success and aggression,
which can be greatly enhanced by measurements of children’s actual
performance (e.g., grades, disciplinary actions in school).

The following study was designed to addresses these limitations and
broaden our understanding of children exposed to domestic violence in several
important ways. Eighty children, aged seven to eleven, of both genders, whose
mothers were experiencing partner abuse, were interviewed. First, the
relationship between the severity of violence to which the children were exposed

and their behavioral, emotional, social, and academic adjustment was examined

utilizing a combination of maternal and child reports, as well as school records.
Second, the protective nature of social support, as well as the relationship
between the severity of violence and the amount of social support in the
children’s lives, was examined. The following literature review summarizes our
present knowledge of the adjustment difficulties of children exposed to abuse
against their mothers and demonstrates the importance of examining these

currently unexplored or limited areas of the literature.

Chapter 1
CHILDREN EXPOSED TO THEIR MOTHERS’ ABUSE

Women with abusive partneg

Domestic violence, or more speciﬁcally, violence against women, is
gaining recognition as a serious and pervasive problem in this country. it is
estimated that between 21% and 34% of women in the US. will be slapped,
punched, kicked, choked, stabbed, shot, or raped by their intimate male partners
sometime in their lives (Browne, 1993; Straus, Gelles, & Steinmetz, 1980; Straus
& Gelles, 1986). Research on women with abusive partners has shown that the
violence has a variety of detrimental effects on their lives and well-being. For
instance, these women are more likely to experience depression than women
who have not experienced intimate partner violence (e.g., Sato & Heiby, 1992;
Vitanza, Vogel, & Marshall, 1995). They are also more likely to be isolated and
to have fewer friends, family, or community support who could help them leave
their assailant or ﬁnd safety (e.g., Nielsen, Endo, & Ellington, 1992; Sullivan,
Basta, Tan, & Davidson, 1992). Because a great deal of women experiencing
partner abuse reside with their children, researchers also became interested in
examining whether similar negative effects of the violence (e.g., depression), as
well as other adjustment difﬁculties (e.g., discipline problems), are experienced

by their children.

Effeete en ehilgren ef witneeeing their metheﬁ’ ebeee

Behevierel gifﬁegltiee. A number of important concerns prompted
researchers to investigate the possibility of elevated levels of aggressive or
acting out behaviors in children exposed to marital violence. Early clinical
reports and anecdotes from shelter workers suggested difﬁculties in this arena
(e.g., Hilbennan & Munson, 1977-78). Theories about the inter-generational
transmission of violence through the modeling of aggressive behaviors in the
home (e.g., Emery, 1982) also raised concerns. Finally, the relationship
between non-violent parental conﬂict and later conduct problems in children
(e.g., Amato & Keith, 1991) encouraged researchers to investigate children of
abused women.

Thus, most of the studies examining the effects of witnessing violence
against their mothers include measures of children’s behaviors such as the Child
Behavior Checklist (CBCL; Achenbach & Edelbrock, 1983; 1986;
Achenbach,1991) or the Peterson-Quay Behavior Problem Checklist (BPC;
Quay, 1977; Quay & Peterson 1979; 1983). The CBCL is a list of behavioral
items which parents (in this case, usually mothers) rate using a Likert-type scale
on how applicable each item is to their child. This measure contains a broad-
band grouping of extemelizing behaviors which include the aggression and
delinquency subscales. Children’s scores can be compared to that of a normed
sample of same age peers, which contains cutoffs for clinically problematic
scores. The BPC assesses deviant behaviors in children, and also contains

items which parents rate on a Likert type scale. This measure is comprised of

four subscales (mm mm W
and W) and also contains group norms.

The studies reviewed here are comprised mostly of school aged children,
as this is the targeted population for this study. In addition, a small number of
studies of younger children and slightly older children are included, to
demonstrate that these difﬁculties also apply to other age groups.

Overall, the research on behavioral adjustment utilizing the CBCL and
BPC, as well as studies using additional measures such as structured interviews
(e.g., Holden 8. Ritchie, 1991 ), demonstrate that children exposed to violence
against their mothers are at higher risk for behavioral problems than are children
from comparable community samples or children from normed samples.
Although these studies vary on the speciﬁcs of the behavioral problems,
important similarities exist among them. For instance, Davis & Carlson (1987)
found that children in their sample received higher parent ratings on aggressive
behaviors, while Holden & Ritchie (1991) discovered a trend towards aggressive
behaviors in their structured child interview. In their studies of boys, Jaffe,
Wolfe, Wilson, & Zak (1986) and Rosenbaum & O’Leary (1981 ) found elevated
extemalizing behaviors and elevated conduct disorder scores, respectively.
Elevated conduct disorder and externalizing scores were also evident in studies
of both boys and girls done by Jouriles, Murphy, & O’Leary (1989) and O’Keefe
(1994a; 1994b; 1995).

Behavioral problems were also seen for samples of preschool children.

For instance, Fantuzzo et al. (1991) showed young children (ages 3 1/z to 6 1/2)

residing at domestic violence shelters to be rated higher on externalizing
behaviors than children living in homes with domestic violence, who, in turn, had
higher externalizing scores than a non-violence community sample. In their
sample of preschoolers (ages 3-5), Graham-Bermann & Levendosky (1998)
found children from violent homes to be rated higher on externalizing, and to
exhibit more aggressive play, more violence, anger, and fnistration, and more
negative interactions with the caretakers in playroom observations.

Studies of children from outside the United States, including Australia
(Mathias, Mertin, & Murray, 1995), Israel (Stemberg et al., 1993), and Mexico
(McCloskey, Figueredo, & Koss, 1995) have also demonstrated higher
externalizing scores for children who have witnessed their mothers’ abuse.

Although these studies show that women experiencing partner abuse are
more likely to rate their children as having behavioral problems than mothers of
comparison community samples, the studies are limited in a number of ways.
First, the mothers’ reports are conﬁned to the time they spend observing their
children, and may reﬂect, at least in part, an interaction between the child and
the mother, rather than the child’s global behavior in all areas of functioning.
Second, as discussed by Stemberg et al. (1993), women who are experiencing
violence from their partners may not judge their children’s behaviors with
objectivity, due to their own feelings of guilt, depression, or frustration that may
be associated with the abuse. Finally, an increased externalizing, aggression, or
delinquency score on the CBCL or BPC does not provide a speciﬁc measure of

problematic behavior for children. Thus, in addition to a mother-reported rating

of extemalizing behaviors, modeled on the CBCL, the proposed study included a
measure of the actual number of times children were disciplined for disruptive
behavior in school (i.e., suspended). This information broadens understanding
of children’s adjustment in two important ways: ﬁrst, it provides an applied
measure of children’s adjustment in addition to mothers’ reported rating of their
aggressive behaviors. Second, in addition to the information obtained on their
behavior at home (which is more visible to the women), it provides a measure of
how the violence at home may inﬂuence their behavior at school.

Emgjienel gifﬁeeltiee. The nature of the violence in the children’s homes
and the increased prevalence of depression and anxiety among their mothers
also spawned research interest into the emotional problems of children from
violent homes. The most common measure of children’s emotional adjustment
is the intemelizing broad band score from the CBCL, which includes
anxiety/depression, withdrawal, and somatic complaints. Most studies utilizing
this measure found higher internalizing scores along with the elevated
extemalizing scores described earlier (e.g., Graham—Bennann 8. Levendosky,
1998; Jaffe, Wolfe, Wilson, 8. Zak, 1986; Mathias, Mertin, & Munay, 1995).
Other studies found evidence for elevated internalizing scores only, in the
children from the domestic violence samples (Christopoulos et al., 1987; Holden
& Ritchie, 1991). However, at closer inspection, the results from these studies
are not easily distinguishable from the results of the others. For instance, in the
Christopoulos et al. (1987) study, it was shown that outlier girls in the community

sample affected the scores, while the rest of the children had both internalizing

and externalizing scores above the mean. In addition, although children in the
violence sample did not have internalizing scores that were signiﬁcantly higher
than children in the comparison community sample, their externalizing scores
were signiﬁcantly higher than the CBCL normed sample. Thus, similar to other
studies, the children in the Christopoulos et al. (1987) domestic violence sample
were exhibiting elevated externalizing, as well as internalizing scores. Finally,
several studies found higher total CBCL problem scores, which reﬂects an
overall elevated prevalence of both behavioral and emotional problems (e.g.,
Holden & Ritchie, 1991; Kolbo, 1996; Mathias, Mertin, & Murray, 1995; Wolfe,
Jaffe, Wilson, & Zak, 1985).

However, similar to the research on behavioral adjustment, the research
on the emotional adjustment of children exposed to their mothers’ abuse, which
utilizes the CBCL only, has several important limitations. First, although many of
these studies have found children to have elevated internalizing scores, most of
these studies do not report differences in the speciﬁc syndromes which comprise
this score (i.e., anxious/depressed, withdrawn, and somatic complaints). Of
these syndromes, childhood depression, which can have grave consequences
for children’s development, is particularly important to assess. Research has
shown that childhood depression is associated with difﬁculties in several areas,
including cognitive, social, and interpersonal (Hammon & Rodolph, 1996). In
addition, longitudinal studies suggest that childhood depression is a good
predictor of adult depression (e.g., Harrington, Fudge, Rutter, Pickles, & Hill,

1990). Finally, recent research has demonstrated that the prevalence of

depression among children is rising (e.g., Gershon, Hamovit, Guroff, &
Numberger, 1987; Ryan et al., 1992). Analyses of the possible sources of such
increases have suggested that the rise may be due partly to current social
variables such as family dysfunction and lack of resources and social supports
(Hammon & Rudolph, 1996). Thus, the relationship between childhood
depression, domestic violence, and social support is particulaity important to
examine at this time. However, even those studies that have examined the
depression syndrome of the CBCL internalizing scale (e.g., Davis & Carlson,
1987; McCloskey, Southwick, Femandez-Esquer, & Locke, 1995) are still limited
by another major concern: their sole reliance on the report of the mothers.

As discussed above (in the behavioral problems section), there are
several limitations to assessing children’s adjustment based only on their
mothers’ reports. These limitations are especially pronounced in the case of
emotional adjustment, since it is difﬁcult for mothers to accurately report their
children's thoughts and feelings (e.g., sadness, depression). Thus, it is
important to include a child self-report measure of depression when studying
children from violent homes. One such measure is the Childhood Depression
Inventory (CDl; Kovacs, 1981 ), a modiﬁed children’s version of the Beck
Depression Inventory (BDI; Beck, Ward, Mendelson, Mock, & Enbaugh, 1961).
This instrument asks children to rate their feelings of sadness, apathy, and
depression in the preceding two weeks. Two studies of children exposed to their
mothers’ abuse, which utilized the CDI, found that they rated themselves as

experiencing signiﬁcantly higher levels of depression than the normed sample

(Spaccarelli, Sandler, & Roosa, 1994; Stemberg et al., 1993). Tang (1997),
utilizing a similar child self report instrument, the Depression Self-Rating Scale
(Asamow & Carison, 1985), found elevated depression levels in the domestic
violence sample in comparison to both children from the maritally discordant and
the non-violence groups. it is important to note here that two of these studies
(Spaccarelli, Sandler, & Roosa, 1994; Tang, 1997) did not ﬁnd elevated
internalizing scores on the mothers’ rating of the CBCL, indicating support for the
utilization of multiple reports of depression. Thus, in addition to a mother-
reported rating of internalizing behaviors, modeled on the CBCL, the proposed
study includes the CDI, which assesses children’s reports of their depressive
feelings.

W According to Bandura’s (1986) social cognitive
theory, children’s behavior is inﬂuenced by their sense of self-efﬁcacy, which is
derived from a combination of personal standards and a comparison of
themselves to other children. According to this theory, a home environment of
violence could inﬂuence both of these areas, causing children to have lower
perceptions of their own social competency and to behave accordingly. A child
self-report measure created by Harter (Perceived Competence Scale (PCS),
1982; Self-Perception Proﬁle for Children,1985) assesses these areas of child
adjustment by examining the way children perceive their interactions with other
children and adults, as well as how they compare themselves to their peers in
sports, school, and physical appearance.

The small number of studies that have utilized the Harter with children of

10

abused women have yielded mixed results. Two studies with school aged
children from violent homes (Kolbo, 1996; Spaccarelli, Sandler, & Roosa, 1994)
have found their self-competency scores to be lower than the normed
comparison sample (although there were sex differences in both studies).
However, a study utilizing a pictorial version of the Haiter with preschoolers
(Fantuzzo et al, 1991), and two studies of school-aged children (Christopoulous
et al., 1987; O’Keefe, 1994b) found no differences between the children in the
violence sample and those in the comparative sample.

A number of studies utilizing other measures of social adjustment
(Fantuzzo et al., 1991; Graham-Bermann & Levendosky, 1998; Rosenberg,
1984; Wolfe, Jaffe, VWson, & Zak, 1985) also found children in the violence
samples to have lower social competency scores. It is possible to explain some
of the differences in the results by examining the differences in methodology of
the studies. For instance, the Christopoulos et al. ( 1987) study, which did not
ﬁnd differences between the groups, had at least one confounding factor: almost
half of the women in the non-violence comparison sample (n=19) reported
incidents of "less severe” forms of physical violence from their partners, such as
pushing and throwing things. Thus, the children in the two groups may have had
similarly low competency scores because both sets of children were exposed to
violence. There is not enough data at this point to speculate about the causes of
all the differences in the studies of this area. It seems clear, however, that the
lack of studies assessing children’s social competency, and the relative

disagreement among existing studies, makes this an important area of further

11

research. Thus, the proposed study will include Harter’s PCS among its
measures.

Aeegemie edjgetment. Another area of the literature that has been
relatively neglected is the academic adjustment of children exposed to domestic
violence. Children’s academic performance has obvious implications both for
how they perceive themselves and for their success in this society. A small
number of studies indicate that children exposed to violence against their
mothers perform worse on tests of academic achievement than children from
comparative samples (Moore et al., 1990; Wolfe, Zak, Wilson, & Jaffe, 1986;
Mathias, Mertin, & Murray, 1995). However, no studies have examined the way
children’s actual performance in school (i.e., their grades) may be affected by
domestic violence in the home.

There are several important differences between children’s grades and
their performance on achievement tests. The most important of these is the
dynamic way in which grades are assigned to children by their teachers, as
opposed to the standardized, non-interactive way in which children take
achievement tests. In his theory of child development, Vygotsky (1978) talked
about the major difference between a child’s developmental level as determined
by his or her independent problem-solving skills, and that child’s level of actual
achievement under some adult guidance or supervision. Success in elementary
school depends more on this latter type of achievement, since classroom
interactions do not take place in a standardized test-taking manner, with the child

working alone to solve problems. That is, unlike achievement test scores,

12

children’s grades are based in part on the interaction between the child and his
or her teacher, and include the teacher’s assessment of the child’s initiative,
efforts, and motivation. Thus, a measure of children’s grades provides us with a
more accurate understanding of their academic success and adjustment. The
drawbacks of utilizing grades as a measure of academic adjustment lie in the
variability between different teachers’ grading styles, and the lack of
standardization in grade assignment. However, when examined across different
school systems and teachers, grades should provide a unique and meaningful
measure of children’s academic achievement. The proposed study will,
therefore, include a measure of children’s grades for the segment of the school
year closest to the time of interviewing, and examine how this measure relates to

the other variables in the study.

13

Chapter 2

PROTECTIVE FACTORS

Riek eng Reeilienee
One important, though often unstated, goal of research with children of
abused women is to improve their well-being and enhance their development. In
order to best accomplish this, it is necessary to examine not only the risk factors
that negatively affect this development, but also the positive, or protective factors
that promote resilience in such high risk populations. According to Masten,
Best, and Garrnezy (1990),
Resilience is the process of, capacity for, or outcome of successful
adaptation despite challenging or threatening circumstances.
Psychological resilience is concerned with behavioral adaptation, usually
deﬁned in terms of internal states of well-being or effective functioning in
the environment or both [and] protective factors moderate the effects of
individual vulnerabilities or environmental hazards so that the adaptational

trajectory is more positive than would be the case if the protective factor
were not operational. (P. 426)

Discussing his research of childhood risk and resilience, Rutter (1990)
emphasized the interaction that operates between risk and protective factors.
According to this moderating model, the protective factors do not cause
resiliency, but act indirectly to buffer the negative effects of the risk factors.

Literature on risk and resilience has identiﬁed several protective factors
that can buffer children from various negative risk factors (e.g., maternal mental
health, child temperament, social support). However, thus far, the literature on

children of abused women has focused mostly on the children’s negative

14

adjustment and has done little to examine the relationships of risk and protective
factors in their lives. A small number of studies have examined the role of
certain variables that were predicted to contribute to the negative outcomes of
children from violent homes, or protect them from the negative effects of
witnessing violence. These risk and protective variables include child
temperament, parenting, gender, age (Holden & Ritchie, 1991), parental
aggression, family size, SES, parental drug and alcohol use (O’Keefe, 1994b),
maternal stress, negative life events, and family crises (Wolfe, Jaffe, Wilson, &
Zak, 1985). However, as seen from this list, even these studies tend to focus
more heavily on risk factors that can contribute to maladaptive adjustment,
rather than examining variables that can contribute to the resiliency of children.
Of the protective variables currently found in the literature, the buffering
effects of social support in the lives of children exposed to domestic violence are
particularly important for us to examine. First, research with women who have
experienced partner violence has shown them to be especially vulnerable in this
area, as they are often Isolated from friends, family, and social services, by the
perpetrator and by the abuse (e.g., Nielsen, Endo, & Ellington, 1992; Sullivan,
Tan, Basta, Rumptz, & Davidson, 1992). This research has also found that the
presence of social support is a positive moderator in the adjustment of these
women to the violence in their lives (Sullivan, Campbell, Angelique, Eby, &
Davidson, 1994). Second, as is described in the following section, there is some
research that suggests that social support can enhance the resiliency of children

exposed to their mothers’ abuse. Finally, as will be brieﬂy summarized, social

15

support has been found to be a signiﬁcant buffer in the lives of children exposed
to other stressors (e.g., poverty, divorce).
i I n r

The deﬁnition of social support used in this study is the perception by an
individual that he or she is cared for and loved, esteemed and valued, and a
member of a network of communication and mutual obligation (Cobb, 1976;
Cohen & Wills, 1985). Until recently, there has been little research on the
protective nature of social support in children. In contrast, the adult literature
has shown social support to act as a buffer against the effects of a wide range of
risk factors, including death of loved one, chronic illness, and major life changes
(e.g., Cobb, 1976; Cohen 8. Vlﬂlls, 1985). Social support is thought to beneﬁt
individuals in a multidimensional way, by providing them with “emotional
sustenance, informational guidance, or tangible assistance” (Hauser & Bowids,
1990, p.399). Although there is no agreement as to exactly how social support
works to protect and help peOpIe, two main models of the positive effects of
social support exist in the adult literature.

The ﬁrst one, known as the Mememedel, argues that social support is
helpful in ameliorating the effects in people under high stress only (i.e., an
interaction effect). The second model proposes that social support has a
positive mm regardless of the level of stress in people’s lives. The
research on social support has found signiﬁcant support for both the buffering
(e.g., Dalgard, Bjork, & Tambs, 1995; Licitra-Kleckler & Waas, 1993; Llabre &

Hadi, 1997; Wasserstein & La-Greca, 1996; Ystgaard, 1997) and the main effect

16

models (e.g., Collins, Dunkel-Schetter, Lobel, & Scrimshaw, 1993; Jonge,
Janssen & Van-Breukelen, 1996; Li, Seltzer 8. Greenberg, 1997), and the debate
continues to be a controversial one. In their review of the earlier literature of
buffering and main effects, Cohen and Wills (1985) discuss ways in which
measures of social support and stress that are used in studies, and the way the
data is analyzed, can often bias the results towards one of the two models. For
instance, measures that assess a person’s level ef integratien into a large
network or community (e.g., number of people in one’s social support network)
are more likely to show evidence for the main effect model. In contrast,
instruments that assess speciﬁc areas of social support relevant to the stressor
(e.g., social companionship, esteem) have a greater chance of showing a
buffering effect. Studies assessing the presence of a conﬁdant, or a person to
whom the respondent can talk to about private and personal issues, also show
consistent buffering effects. According to Cowen & Willis, a multiple regression
is the preferable analysis for showing these effects in studies using continuous
variables, while a chi-square or log—linear analysis may be used with
dichotomous variables. As concluded by the authors, both models have been
supported by the research, and both contribute to our understanding of the
different ways in which social support may operate to help people.

Based on these two models, questions assessing both speciﬁc types of
social support (e.g., companionship, esteem) and global social support (e.g.,
number of people in social network, overall satisfaction with support) were

examined in this study.

17

i I n hil n x m i vi I n
Although few studies exist at this time on the effects of social support on
children exposed to domestic violence, the mixed results from this limited
research suggest that certain kinds of social support protect children from the
negative effects of domestic violence on their adjustment.

A study by Kolbo (1996) examined the social support networks of 60
children from violent homes, aged 8 to 11, using a measure designed by the
researcher: the Supportive Relationship Questionnaire (SRQ). This
questionnaire is speciﬁcally aimed at assessing social support that is related to
the violence in the home. That is, three of the four categories of social support
questions included in the measure (emotional, instrumental, and informational)
focused on support received by the child around the issue of exposure to
violence (e.g., “When your mom and dad [or her boyfriend] ﬁght is there
someone you talk to who listens to you?“) The third category
(social/companionship) asked children who they like to have fun with. Children
were then asked to rate the level of support offered by each person they named
using a Likert-type scale of 0 (did not make them feel better at all) to 3 (made
them feel a lot better). Both the severity and frequency of the violence to which
children were exposed was measured using a modiﬁed version of the Conﬂict
Tactics Scale (CTS; Straus, 1979), which was completed by their mothers.
Measures of child behavior using the CBCL and child self worth (Self-Perception
Proﬁle for Children; Harter, 1985) were also used to determine the relationship

between exposure to violence, social support, and child adjustment.

18

The mean problem behavior score for the children in this study was
signiﬁcantly above that of the normed sample for a non-clinical population, and
not signiﬁcantly different from the normed clinical sample. However, the global
self-worth scores for the children in the study were also signiﬁcantly higher than
the normed scores presented by Harter. In other words, children in this study
had signiﬁcantly higher levels of behavioral adjustment difﬁculties, but displayed
higher levels of feelings of self worth than the children in the normed samples. A
moderating relationship between high levels of social support and ratings of self-
worth was also found, but only in boys. it is important to note here that,
regardless of the level of social support, low levels of exposure to violence were
associated with higher ratings of self-worth for all children, whereas exposure to
higher levels of violence produced more variability in children’s self-worth
ratings.

The authors hypothesized that the lack of buffering for giiis may be
explained by the caregiving role that women and girls often take in our society,
further increasing their own psychological stress in high risk situations.

However, it is also important to note that the study did not address social support
received by the children in other arenas, apart from exposure to violence.
Although children were asked who they like to have fun with, the support they
receive for problems with schoolwork or interpersonal difﬁculties was not
assessed. Similarly, they were not asked about positive support they may
receive from people that goes beyond simply “having fun" (e.g., sharing secrets,

learning skills, feeling cared for). Thus, it is possible that the negative effects of

19

exposure to violence on girls are buffered by social support in some of these
other areas and the ﬁndings may be reﬂecting a gender difference in the kind of
support that is sought or received by girls and boys, rather than a gender
difference in the buffering role of social support in general.

Another study also examined the relationship between social support
networks and the adjustment of children exposed to domestic violence
(Rawlins,1993). Seventy-one children, half of whom resided in shelters for
domestic abuse, participated in the study. The Self Perception Proﬁle and
CBCL were used to assess social competency and child behavioral problems,
respectively. No differences in size of social support networks were found
between the children exposed to violence and the non-exposed group, although
children in the non-exposed group were more likely to include their mother,
father, and other family members as part of their support network. Regardless of
group, children with larger support networks received higher social competency
scores, with the lowest social competency scores being received by children in
the violence-exposure group who had small social support networks. Similaiiy to
Kolbo’s study, social support was not found to affect behavioral scores for either
group of children. However, the researchers did not examine the children’s
satisfaction with the social support they received from the different people they
reported, or evaluate how different kinds of social support (e.g., from mothers vs.
fathers) may have moderated the children’s adjustment.

O’Keefe (1994b) has also shown that children’s behavioral and emotional

adjustment are moderated by social support. This study examined both the

20

effects of social support received by the child’s family from agencies and outside
individuals, and the quality of the parent-child relationship as rated by both
parents and children. It was found that both higher levels of family support and
higher quality levels of parental support protected children from the negative
effects of witnessing violence (i.e., higher internalizing and externalizing scores
on the CBCL).

in contrast, two other studies which included social support measures did
not ﬁnd that social support buffered children from the violence they witnessed
(McCloskey, Figuerdo, & Koss, 1995; McCloskey, Southwick, Femandez-
Esquer, & Locke, 1995). However, in the former study, only parental and sibling
support was assessed, limiting the conclusions to these types of child support. It
is also important to note that this study found children from violent homes to
report less overall social support than the comparative non-community sample.
The study also did not assess social support in a comprehensive manner, only
asking the number of relatives living in close proximity to the child. Thus, it is
impossible to determine whether these relatives served as a source of social
support, as a source of stress and aggravation, or played no role in the child’s
life at all.

Due to the limitations of the above studies, as well as to their small
number, it is too early to draw conclusions from this research. Rather, these
preliminary ﬁndings serve to encourage further exploration into different aspects
of social support (e.g., quantity, satisfaction) as It relates to the buffering of

children from the effects of exposure to domestic violence.

21

Finally, it is important to note that social support has been found to be a
signiﬁcant moderator for other high risk groups of children. Studies, including
ones that are longitudinal in nature (e.g., Werner, 1993), have found that
children that are homeless and economically distressed (Graham-Bermann,
Coupet, Egler, & Mattis, 1996), experiencing high amounts of stress (e.g.,
Wertlieb, Weigel, & Feldstein,1987), or at high risk for academic problems (e.g.,
Quamma 8: Greenberg,1994) are buffered from the negative effects of these
stressors by social support.

Thus, although the literature on social support in the lives of children from
violent homes is limited, there is evidence from this and other research to
suggest the important role that social support plays in protecting children from
exposure to violence. The purpose of this study was, therefore, to examine the
relationship between children’s exposure to violence and their adjustment for

both speciﬁc (i.e., buffering) and global (i.e., main) effects of social support.

Chapter 3

HYPOTHESES

First, the literature on the effects on children of witnessing their mothers’
abuse has demonstrated that these children experience more behavioral and
emotional difﬁculties than their peers from non-violent homes. Less attention
has been given to children’s social and academic adjustment, or their own
ratings regarding the effects of the violence in their homes. Although these
areas have not yet been examined in depth, there is some evidence to suggest
that children’s social and academic adjustment may also be detrimentally
affected by witnessing violence against their mothers. Thus, it was hypothesized
that higher levels of exposure to violence would relate higher levels of behavioral
and emotional difﬁculties and lower levels of social and academic adjustment.

Second, the literature on risk and resilience shows that social support is
an important protective factor, or moderator, in the lives of adults and children
experiencing a variety of negative life events. The literature has also shown
main effects of social support in people’s lives, which do not interact with stress.
Based on some preliminary studies of children exposed to domestic violence,
and on the literature on stress and social support in general, it was hypothesized
that both buffering and main effects for social support would be found in the
study. That is, social support was predicted to protect children from the negative

effects of witnessing violence on their behavioral, emotional, social, and

23

academic adjustment, while also having a positive main effect on their

functioning in these four areas.

24

Chapter 4

METHOD

P i i n

The participants were part of a larger longitudinal study of women with
abusive partners and their children. The eighty children and their mothers, who
participated in this study, were recruited from two domestic violence shelters and
one social service agency that provides services for lower socio-economic
families in a mid-sized Midwestern town. Women with at least one child aged 7
to 11, who had experienced abuse from their partners in the last 4 months, were
included in the study. If their mothers agreed to participate in the study, children
were approached separately to assess their interest and minimize possible
coercion from their mothers. In families where more than one eligible child was
interested in participating, one child was chosen randomly to receive the full
interview, while the other children were given a shortened "mock" interview to
minimize feelings of resentment among the children.

Almost half the women in the study identiﬁed themselves as Caucasian
(49%), with 39% identifying themselves as African American, 5% as
Hispanic/Latina, 5% as bi-racial, 1% as Asian, and 1% as Native American.
Forty-four percent of the children were identiﬁed as African American, with 40%
being identiﬁed as Caucasian, 10% as bi-racial, 5% as Hispanic, and 1% as

Asian. The women in the study averaged 31 years of age, with the majority of

25

the sample (77%) being under 35 years old. Their children’s ages ranged from 6
to 11 years, with an average of 8 years. Fifty ﬁve percent of the children in the
study were female. The mean monthly family income of the women in the study
was $1199, with a minimum of no income (unemployed) and a maximum $3472
a month. A little more than 22% of the women had not attended high school,
26% had high school diplomas or GED’s, 7.5% had graduated from a trade
school, 39% had attended some college, and 5% were college graduates.

Pr r

The interviewers were undergraduate females who participated in the
project for college credit. All interviewers committed to two semesters
(approximately 30 weeks) with the project in order to receive intensive training in
interviewing skills with both adults and children, sensitivity training in working
with a diverse community population, and knowledge about domestic violence,
following a manual developed for this project (Sullivan, Juris, Gauthier, Nguyen,
& Prewitt, 1997).

The interviewers were divided into ”interview teams" consisting of child
interviewers (undergraduates trained in general theory and child interviewing
skills), babysitters for younger children (as needed), and a team leader (an
experienced child interviewer who also had adult interviewing training and skills).
The team leader was in charge of coordinating the team and arranging the
interview. The women were ﬁrst contacted by telephone, following proper
precautions to assure their safety (i.e., inquiring whether this was "a good time to

talk"). If attempts at telephone contact were unsuccessful, the team leader

26

followed a tracking procedure designed for the project, which included contact
persons, school personnel, and other sources who were given by the woman in
case she could not be reached. An appointment was arranged with the woman
which was convenient to her and her children’s schedule, and the team drove to
the home (or other arranged location).

A majority of the interviews (about 90%) were conducted in the women’s
homes, with the remainder being conducted in other safe places that were
convenient for the participants (e.g., the project office on campus, a domestic
violence shelter). All interviews took place in separate rooms for each member
of the family to ensure privacy and conﬁdentiality. The children’s interviews
lasted an average of 110 minutes, with a range of 53 minutes to 445 minutes in
length. The women’s interviews were longer, with a mean of 145 minutes, and a
range of 80 minutes to 300 minutes in duration. All women were paid $15.00 for
the interviews, while all children were given their choice of $5.00 or a toy of that
value.

Macaw

The study utilized a number of preexisting standardized measures as well
as some that were designed or modiﬁed speciﬁcally for this research.

Meme Vielenee Sm - 93M Rem Children were asked three
questions using a Likert response scale (1 ="never" to 4="a lot") which assessed
their awareness of verbal abuse against their mothers (e.g., "There are many
things people do to annoy or hurt each other such as making fun of someone or

calling them names and saying things to make them feel bad. How often has

27

IF"

[assailant] said or done any of these type of things to your mom?"). Children
were also asked to respond to two questions which assessed their awareness of
physical abuse against their mothers (e.g., "How often have you seen or heard
[assailant] hurt your mom physically? By that I mean anything from grabbing or
slapping her to punching or kicking her, anything like that”) using a Likert
response scale (1 ="never" to 6="more than 4 times a week”). Scores on these
ﬁve items were standardized to create an overall Child Experience of Abuse
Scale (a=.81). This measure is found in Appendix A.

m \_/'I_O_I§ﬂ9§ Sea; - Meme: Rem Mothers were asked to
respond to three items assessing how often their children witnessed their
emotional abuse, threats against them, and their physical abuse on a Likert
scale (1 ="never" to 4="often") for emotional abuse and (1 ="never" to 5="3 or 4
times a week") for threats and physical abuse. The responses to these items
were then combined to form an overall mother report of Child Witnessing
(a=.81). This measure is found in Appendix B.

Exmegee e! \_ijl_e_r19_e Sm - Meme; Mothers were asked to rate each
of 22 emotional abuse items (e.g., "How often has [the assailant] ridiculed or
criticized you?") on a Likert scale (1 ="never" to 4=often')(a=.90). In addition,
mothers were given a 12—item scale assessing the types of injuries they may
have received (e.g., “Have you ever experienced internal injuries?"), using a
'yes" or 'no" format (a=.77). Finally, mothers responded whether they have ever
experienced the physical abuse described by 7 items modiﬁed from the Conﬂict

Tactics Scale (CTS; Straus, 1979)(e.g., "To the best of your recollection, has

28

[the assailant] ever pushed, grabbed, shoved, or slapped you?"). If the women
responded afﬁrmatively, they were asked whether or not this abuse occurred in
the last four months (a=.62). The responses from these three scales (Emotional
Abuse, Injury, and Physical Abuse) were standardized to form an overall abuse
scale (a=.77). This measure is found in Appendix C.

Eemeg meetegee Sale. The Perceived Competence Scale (PCS;
Harter, 1985) was used to assess the way children feel about their social
functioning in several areas (e.g., academic, athletic, behavioral). The PCS
presents children with 22 items (e.g., ”Some kids feel that they are very good at
their school work, but other kids worry about whether they can do the school
work assigned to them") and asks them to choose which type of kids they are
more similar to. The children are then asked to what degree they are like the
children they chose (i.e., ”really like them“ or ”sort of like them”). The items in
the PCS can be combined to form ﬁve subscales and one global self-worth
subscale. However, due to the large number of independent variables examined
in this study, only the Global self-worth scale was utilized (0:65). This measure
can be found in Appendix D.

Chm W Inventery. To assess whether children were
experiencing depressive symptoms, the Childhood Depression Inventory (CDI;
Kovacs, 1981) was used. The CDI contains 27 items which ask children to
indicate, using a Likert scale of 1 to 3, how they have felt in a particular area in
the past two weeks (e.g., 1="l feel like crying once in a while"; 2="l feel like

crying many days"; 3="l feel like crying every day“). Children are then given a

29

total score for depression. The CDI has Chronbach’s alphas ranging from .80 to
.94 (Saylor, Finch, Spirito, & Bennett, 1984). The alpha for the sample in this
study was .81. This measure is located in Appendix E.

Behevier Being m A measure with items similar to those found in
the Child Behavior Checklist (CBCL; Achenbach & Edelbrock, 1986) which asks
mothers to rate their children on items such as I‘argues a lo ," ”unhappy, sad, or
depressed", was developed for this study. Mothers use a Likert-type scale to
rate how much each item applies to their child (1 ="not at all tnie' to 4="very
true“). The 20 items in the instrument include an lntemalizing (a=.73) and an
Extemalizing subscale (a=.73), as well as an overall measure of children’s
behavioral and emotional adjustment (Total Behaviors scale; a=.83). This
measure can be found in Appendix F.

Mm 91 Mme. Children’s externalizing behavior was also
assessed with an item which asked mothers how often in the last four months
their child had been suspended. These measures can be found in Appendix G.

Mean Semester Stale.- Quarteiiy report cards collected from the
children’s schools were utilized to assess children’s academic adjustment.
Because children represented more than thirty different schools in the area and
ranged in grade level from kindergarten to 6th grade, their report cards
presented several different grading systems (e.g., check, check-plus, check-
minus vs. A, B, C, D, E). Thus, the different grading systems were trichotimized
to form a standardized system that could be compared across children. The ﬁrst

level included marks that indicated above average performance in the subject

30

D.

(e.g., grades representing "Excellent” and "Good," ”Exceptional Achievement”
and "Has Mastered Skill,“ or ”Excellent Progress" and 'Satisfactory'). The
second level included grades that indicated average performance in the subject
(e.g., grades representing ”Average,” “Needs Improvement,” or "Working on
Skill"). The third level included grades that indicated poor performance in the
subject (e.g., grades representing "Poor“ and ”Failing,“ “Does not perform skill,”
”Unsatisfactory," or "Area of Concern").

Children’s grades across seven subjects that were common across
schools (Reading, Writing, Spelling, Mathematics, Science, and Social Studies)
were averaged for a mean score for the semester. For the 17 children (40%)
that did not have grades for spelling, an average of the six remaining subjects
was taken. An additional child had an average that was comprised of English
and Mathematics grades only.

M M 5.09993 QuestmgaiLe. Children’s perceptions of social
support were measured utilizing the Child Social Support Questionnaire,
modiﬁed from the Adult’s Social Support Questionnaire (Bogat, Chin, Sabbath 8.
Schwartz, 1983). This measure includes ﬁve questions which assess the
quantity, role, and type of social support in different areas of children’s lives.
Each of the ﬁve questions addressed a speciﬁc area of social support that could
act as a buffer for the children (e.g., companionship, care, esteem). Examples
of questions are: "Who listens to you when you need to talk about something
personal, for example, about problems between your parents, problems with

friends, things like that?", and "Who can you really count on to always be there

31

for you?"

Global, or main effects of social support were measured in two ways: total
number and satisfaction with support. For total number, the total number of
people mentioned in the social support section was calculated, as well as the
total number of times that mothers, fathers, and other adults were mentioned in
the children’s social support sections. Children were also given a Satisfaction
scale consisting of 5 items that asked them to rate how they feel about £931
mum and the way in whieh they spend time with their friends and family, on a
Likert-type scale of 1="Very Happy" to 7="Very Unhappy“ (a=.22). Because the
items in this Satisfaction scale had low intemal reliability (a=.22), they were
deemed inappropriate for analyses in the study. Buffering effects were
measured by calculating the total amount of egn_ﬁd_agte (people who the child
feels cares about them, listens to them, and can be counted on) in each child’s

social support section. The social support measures can be found in Appendix

H.

32

Chapter 5

RESULTS

Some of the scales used in the analyses were incomplete due to
participant refusal or missing data. Five children did not respond to questions for
the Children’s Depression Inventory and three children did not respond to
questions about witnessing violence. Subsequently, when scaled, these
participants were not included. Due to high correlations between the
lntemalizing, Extemalizing, and Total Scales of the Behavior Scale (see Table
1), only the Total Scale was examined. School data were obtained for only 52%
(n=42) of children due to the unavailability of records for many children and
refusal by schools to disclose records for some children. However, analyses did
not reveal any signiﬁcant differences between the children for whom school
records were obtained and those for whom school data were missing on any
adjustment variables (i.e., CDI Total, Total Behaviors, Harter Global),
demographic variables (i.e., age, gender, minority status) or for self-reports of
level of violence witnessed. All variables exhibited a range of scores showing
adequate variance among the sample (see Table 2). In addition, a correlation
matrix which includes all variables used in analyses can be found in Table 3.

For this study, a signiﬁcance level of e<.1 rather than .05 was utilized in
order to equalize the Type 1 and Type II errors. With a sample of 80

participants, the utilization of a e<.05 would yield about 50% power to detect a

33

statistically signiﬁcant predictor that accounts for 5% of the variance in the
dependent variable. Using a signiﬁcance level of e<.10 increases the power to
70% while still holding Type I error probability at 10%. In addition, because the
hypotheses in this study were directional, one-tailed analyses were conducted
when testing the hypotheses. Cohen’s (1992) criteria were used in judging effect
sizes: I_'=.10 is considered a small effect, [=30 a medium effect, and 5.50 a
large effect.
in hi V ri l

Correlations were conducted to examine whether the age of the children
in the study was related to their adjustment or witnessing of violence. T-tests
were conducted to examine whether there were differences between boys and
gins on any of the outcome variables or their reported levels of witnessing
violence. T-tests were also conducted to examine whether minority versus white
children differed on adjustment or how much violence they witnessed. It was
found that age was signiﬁcantly related to the CDI Total scores (_i=-.30; p_=.009),
a medium effect accounting for 9% of the variance. It was also found that boys
had signiﬁcantly more suspensions than girls (i (2, 78)=-0.23; p_=.039; girts’ mean
=.15; boys’ mean=.50). Consequently, for all tests using the CDI Total and the
number of suspensions, age and sex were used as covariates, respectively. No
signiﬁcant differences were found between children of different ethnic/racial
backgrounds (i.e., minority vs. white) on any adjustment, social support, or
violence variables. Children’s age or sex was also not signiﬁcantly comelated

with any other outcome variables, reported levels of witnessed violence, or social

34

support variables.
\fielenee end Adjuetment

W Because items comprising the scale of children’s reports of
witnessed violence were standardized using z-scores, Tables 4 and 5 present
the frequencies of children’s responses to separate verbal and physical abuse
items. On average, 89% of the children reported witnessing at least some verbal
conﬂict in the home (yelling and arguing), with 46% reporting high levels ("a lot").
Seventy four percent of the children also reported witnessing at least some
verbal abuse towards their mothers (ridiculing and physical threats), with 37%
reporting high levels (”a lot” or once a week or more). In addition, 62% of
children reported witnessing at least some physical assaults against their
mothers, with 21% of the children reporting high levels (once a week or more).

Children’s Total Behavior scores ranged from 0 to 30, with a maximum of
38 points possible. Only one child, with a score of 30, fell in the fourth quartile
for Total Behaviors. The majority of children (58%) had Total Behaviors falling in
the second quartile. The scores of the remaining children were almost equally
divided between the ﬁrst and third quartiles.

Twenty percent of the children had been suspended from school at least
once in the last four months, with boys having signiﬁcantly more suspensions
than gins. Only eight percent of children had mean semester grades below
average (equivalent of D or F) and the majority of children (66%) had mean
grades of 2.5 or higher, on a scale of 2="average or C" and 3="above average

or A and B.“

35

The children’s average Harter Global scores 093.21) were comparable to
those of the normed sample (2(_=3.10), which was composed of 6‘“ and 7‘“ grade
children (Harter, 1985). The average CDI Total scores of the sample 09.03)
were also comparable to that of the normed sample of school aged children
()_(=9.28) (Kovacs, 1985).

Child Repert. It was hypothesized that children who reported witnessing
higher levels of violence would have higher depression scores as measured by
the CDI, more behavioral and emotional problems as measured by Total
Behaviors, a higher number of suspensions from school in the past semester,
lower social competency as measured by the Harter Global, and lower academic
achievement as measured by mean semester grades.

Pearson r correlations were used to test the relationships between
children’s reported levels of violence and the ﬁve adjustment variables. It was
found that children’s self reports of witnessed violence were signiﬁcantly
correlated to their Total Behaviors scores, at the p<.10 level (r_=.11; p_=.087),
which is considered a small effect. No signiﬁcant relationships were found
between how much violence children reported witnessing and their scores on the
CDI, the Harter Global, their mothers’ reports of how often they were suspended
in the past semester, or their mean semester grades.

MetlleLReegﬂ, Because there was only one signiﬁcant ﬁnding for
children’s reports of their mothers’ abuse, correlations were conducted between
their mother’s reports of violence and the children’s adjustment. A signiﬁcant

positive correlation was found between mothers’ reports of how much abuse

36

their children witnessed and their ratings of their children’s adjustment on the
Total Behaviors (p.32; p_=.057), a medium effect accounting for 10% of the
variance. Mothers’ reports of their children’s witnessing of violence were also
signiﬁcantly correlated to the number of times children were suspended in the
past semester (r=.14; 9:057), and the children’s ratings on the Harter Global
(_r=-.13; p_=.060), small effect sizes accounting for 2% of the variance each.
Similarly, a signiﬁcant positive relationship was revealed between mothers
reports of how much overall abuse they experienced and their children’s scores
on the Total Behaviors (_r=.37; e=.000) (a medium effect size accounting for 14%
of the variance), as well as their number of suspensions (i;=.26; p_=.006) (a small
effect size accounting for 7% of the variance). The children’s reports of how
much violence they witnessed were highly correlated to these maternal
perceptions of child witnessed abuse (r_‘=.39; p_<.005), as well as to the women’s
reports of how much abuse they actually experienced (i;=.39; p_<.005). These
were medium effect sizes accounting for 15% of the variance each.

' l. n A ' m n

W The number of individual (non-repeating) people mentioned

in children’s social support sections ranged from 3 to 27, with an average of 11
individual people across the ﬁve questions. The total number of people
mentioned in children’s social support sections (including individuals repeated for
different questions) ranged from 5 to 44, with an average of 17 total across the
ﬁve questions. Overall, 76% of people mentioned in the social support sections

were adults (other than mothers and fathers‘).

37

Seventy four children (93%) mentioned their mothers at least once in the
social support section, while forty ﬁve children (56%) listed their biological
fathers for at least one area of the support measure. it is important to note here
that, although all children in the study have regular contact with their mothers
(i.e., reside with her), only 3% of children (n=2) said they saw their biological
fathers sometime this week, and almost half of the children (n=37) had not seen
their fathers for at least one year. (See Table 7 and 8 for frequency of children’s
contact with biological fathers).

Of those biological fathers that were listed by children as sources of social
support, 47% (n=21) were also the women’s abusers in this study. In addition, 13
children (16%) listed assailants that were not their biological fathers in at least
one area of the social support section. Forty six percent of these men (n=6)
were the children’s stepfathers, 39% were their mother’s current boyfriends
(n=5), and 15% were described as their mothers’ former boyfriends (n=2). (See
Table 7 for distribution of mothers, fathers, and assailants as they were listed
among the ﬁve social support areas).

The majority of the children (96%) listed at least one friend of their age
group in their social support section, witha range of 1-10, and a mean of 4
friends being mentioned. Twenty nine percent of children also mentioned at
least one adult friend (with a range of 1-4 and a mean of 2) in their social support
section. In addition, 77% of children mentioned a relative (e.g., cousin, aunt) in
their social support section, with a range of 0-14 relatives and an average of 3

relatives per social support section being mentioned.

38

Biﬂering. It was hypothesized that social support would buffer, or protect,
children from the negative effects of witnessing violence on their adjustment. As
predicted, a multiple regression revealed that the number of conﬁdants present
in the children’s lives (persons who care, listen to, and can be counted on
according to the children), signiﬁcantly buffered children from the negative
effects of witnessed violence on their behavioral adjustment (p=.062) (See Table
9). That is, the interaction between high levels of witnessed violence and the
presence of many conﬁdants, moderated the relationship between violence and
children’s Total Behaviors scores, reﬂecting better lntemalizing and extemalizing
adjustment.

ELEM. It was also hypothesized that global types of social support
would have a main effect on children’s functioning. As expected, correlations
examining the relationship between children’s adjustment and the tetel nemeer
gieeeple in their social support section revealed a signiﬁcant relationship to
children’s Harler Global scores (i;=.13; p_=.064), a small effect size. Signiﬁcant
positive correlations were also found for the total number ef timee edulte (other
than mothers and fathers) were listed (out of 5 possible areas of support) and
the children’s mean grades (.26; e=.024), their Harter Global scores (i;=.16;
e=.041), and their CDI Totals (p.12; e=.082), all small effect sizes, accounting
for 7%, 4%, and 1% of the variance, respectively. No other signiﬁcant
relationships were found between the total number of people in children’s social
support networks and their adjustment scores, or for the total number of areas in

which children had supportive adults and the number of times they were

39

suspended.

A correlation examining the relationship between children’s adjustment
and the total number of areas (out of 5 possible) that werewere listed in the
social support section revealed a signiﬁcant relationship with children’s Harter
Global scores (i_=.27; (2:005) and the number of times children were suspended
(gr-=14; e=.059), both small effect sizes. Correlations did not reveal any
signiﬁcant differences between the other adjustment variables (Total Behaviors,
CDI, and mean grades) and how often mothers were listed in the social support
section.

Correlations examining the relationship between children’s adjustment
and the total number of times [alum (biological fathers and stepfatheis) were
listed in children’s social support section were not signiﬁcant for any of the
adjustment variables, regardless of whether fathers were also assailants or not.

The total number of tests performed was 54, indicating that about 3

signiﬁcant ﬁndings may have been obtained by chance.

40

Chapter 6

DISCUSSION

Overall, this study supports the hypotheses that domestic violence has
negative effects on children’s adjustment, especially in the behavioral arena,
while social support has both buffering and main effects on the lives of children
exposed to their mothers’ abuse.

Vi l n n A ' men

meringue; Overall, children in the study witnessed a wide range of
abusive behaviors towards their mothers, with more than a third (74%)
witnessing verbal abuse and more than half (62%) witnessing physical assaults.
On average, about one fourth of the children reported witnessing high levels of
verbal and physical abuse. This data highlights the importance of quantifying the
violence to which children are exposed to, rather than simply determining
whether or not children witnessed any abuse. In this way, samples of children in
different settings (e.g., shelter vs. non-shelter), or with different adjustment
proﬁles, can be examined in terms of levels and types of violence witnessed.

Children’s behaviors, as rated by their mothers, fell in the middle range of
the Behavior Rating scale, with a majority of the children (58%) having scores in
the second quartile, and only one child having a score high enough to reach the
fourth quartile. Although this scale has not been standardized on a community

sample of children, the distribution of the scores suggests that, overall, children

41

in the sample are not experiencing unusual levels of behavior or emotional
problems (i.e., the scores are skewed towards the lower end of the Behavior
Scale).

Overall, children’s mean semester grades also indicated a satisfactory
level of academic adjustment, with the majority of children (66%) having grades
of 2.5 or higher (the equivalent of C+ or higher), and only 8% of children having
a "D" or 'F" average for the semester. Thus, despite the violence in their homes,
the children in this sample are making good progress in the academic arena of
their lives. Future research can examine other factors, besides social support,
that may contribute to children’s academic resilience in elementary school.

In addition, mothers reported that about 20% of the children were
suspended from school in the last four months, with boys having signiﬁcantly
more suspensions than giiis. A lack of a control community sample (of children
without violence in the home) makes it difﬁcult to determine whether this
suspension frequency falls within normal limits for children in this age group.

Children’s Harler Global scores and CDI scores were both comparable to
the normed samples of school aged children. This suggests once more, that it
may be the level or type of violence witnessed by children, rather than the
presence or absence of violence in the home, that affects children’s adjustment,
at least according to their own ratings.

Result; As hypothesized, emdree’erepgﬂe of the levels of witnessed
violence against their mothers were positively correlated with their Total

Behavior scores. The amount of violence reported by children was not related to

42

their self reported levels of depression (CDI Total), their extemallzing behaviors
as evidenced by number of suspensions from school, or their self reported levels
of social adjustment (Harler Global). However, consistent with the current
literature, a signiﬁcant positive correlation was found between We
of their actual abuse and their children’s scores on the Total Behaviors scale, as
well as with the number of times children were suspended (e.g., Christopoulos et
al., 1987; Davis 8: Carlson, 1987; Jaffe, Wolfe, Wilson, & Zak, 1986; Holden &
Ritchie, 1991). Similarly, a signiﬁcant positive relationship was revealed
between W of how much violence was witnessed by their
children and children’s Total Behavior scores, number of suspensions, and
HarterGIobal scores.

There may be several explanations for why maternal reports of violence
were related to more children’s outcome variables than children’s reports. It is
possible that the children’s views of the violence in their homes did not relate to
their views of themselves, as reﬂected by the CDI and Harler scores. It is also
possible that children were inaccurate reporters of their own adjustment. Finally,
it is possible that children’s behaviors are more affected by the violence than
their emotional (i.e., depression) or social adjustment. Although the signiﬁcantly
elevated Total Behaviors ratings contained both lntemalizing and externalizing
items, only the extemallzing behaviors were able to be corroborated by another
source (the numbers of suspensions). Thus, it is possible that mothers were
exaggerating their children’s emotional distress and behavioral problems due to

their own feelings of concern or depression.

43

It is also important to note, that, although there was variance in the levels
of violence children reported witnessing, all the children in the sample had
mothers who had experienced some abuse in the past four months. Thus, it is
possible that it is the presence or absence of violence in the home that
signiﬁcantly affects children’s scores on the CDI and Harler. This could not be
tested due to the absence of a control group.

Children's social and emotional adjustment may also be inﬂuenced by
chronic, rather than recent experiences with violence. That is, the level of
violence witnessed over the past year, or the past several years, rather than in
the last four months, may be affecting children’s Harler and CDI scores.
Because these data represent Time 1 of a longitudinal study, the effects of
witnessing violence over time can be examined in the future.

In addition, although steps were taken to correct for this, the lack of
signiﬁcant ﬁndings may have been a result of low statistical power due to the
relatively small sample size. However, despite the lack of additional ﬁndings for
children’s self reports of witnessing, the overall combination of child and mother
reports suggests a signiﬁcant relationship between the violence in the home and
child adjustment difﬁculties, especially in the behavioral arena.

i l n A ' m n

Although previous literature has shown that high levels of speciﬁc types of
social support can buffer individuals from high levels of a variety of stressors
(e.g., Cohen & Hobennan, 1983; Flannery & Wieman, 1989; Power, 1988;

Koeske & Koeske, 1990; Wilcox, 1981), the buffering effects of social support in

44

the lives of children of abused'women have not been extensively studied. In
addition, the small number of studies examining the effects of social support on
the adjustment of children from violent homes have had mixed results. Whereas
some studies have found buffering effects for social support (Kolbo, 1996;
O’Keefe, 1994b), others have shown no effects (McCloskey, Figuerdo, & Koss,
1995; McCloskey, Southwick, Femandez—Esquer, & Locke, 1995) or main effects
only (Rawlins, 1993). Finally, as a result of an extensive review of earlier
literature, Cohen and Wills (1985) concluded that certain kinds of social support
were more likely to show buffering versus main effects for individuals.
Supporting this explanation of the diverse ﬁndings on social support in the
literature, the current study found both buffering and main effects for different
kinds of social support in different areas of the children’s development.

Buffering. As described in the literature (Cohen & Wills, 1985), buffering
effects were predicted to be found for social support that met speciﬁc functional
needs, such as esteem (caring) and informational support. As hypothesized, a
multiple regression revealed that the number of conﬁdants in the children’s social
support sections positively moderated the relationship between the levels of
witnessed violence reported by the children and their Total Behavior scores.
That is, children who reported witnessing very high levels of violence were
protected from the negative effects of this violence on their behavioral
adjustment by having many people in their lives who cared about them, listened
to them, and could be counted on.

This ﬁnding is similar to that of O’Keefe (1994b), who found both quality

45

and quantity of parental and outside support to buffer children from the negative
effects of violence on their behavioral adjustment (as measured by the CBCL).

In contrast to Kolbo’s (1996) results, this study found signiﬁcant buffering effects
for behavioral, rather than social adjustment. This suggests that a social support
measure that captures support in other areas of children’s lives (besides support
provided around the violence in their homes) is important to examine for this
population.

M It was hypothesized that main effects for global types of
social support (Le, a person’s embededness into a social network) would also
be found. Main effects were measured by calculating: (a) the total number of
peOple in children’s social support networks; (b) the total number of areas (out of
ﬁve) that mothers and fathers were listed; and (c) the total number of areas (out
of ﬁve) for which adults, other than their parents, were listed in children’s social
support networks.

As predicted, Pearson r correlations revealed that the total number of
people in children’s social support networks were signiﬁcantly positively related
to children’s Harter Global scores. That is, the more embedded in a social
support network that children perceived themselves to be, the higher they rated
themselves on a measure of social adjustment and self worth. This is consistent
with Rawlins’ (1993) ﬁnding that the size of children’s social support networks
was signiﬁcantly related to their social competency scores, for both children in
the "violence” and "non-violence” groups. These ﬁndings suggest that children’s

perceptions of how many people are available to support them in different areas

46

of their lives may be particularly relevant to their sense of self concept and their
adjustment in the social arena. Perhaps it is the feeling of having a large “safety
net" that allows children to develop a more positive image of themselves and
relate in a more effective way with others.

‘ The total number of times mothers were listed in children’s networks (out
of a possible ﬁve support areas) also showed a signiﬁcant positive correlation to
children’s Harter Global scores, and a signiﬁcant negative correlation to the
number of times they were suspended. Thus, children who perceived their
mothers as being sources of social support in more areas of their life, were more
likely to feel better about their social competency and self esteem, and were less
likely to engage in delinquent behaviors in school. Although support from
several different sources was shown to have a main effect on children’s Harler
Global scores, only support from mothers had a signiﬁcant effect on children’s
suspensions in school over the last semester. This suggests that for children of
abused women, a Impede: provision of social support by their mothers may be
critical in helping them control their behaviors in school, regardless of the level of
violence in the home.

In addition, the total number of times adults (other than their mother or
father) were listed as sources of social support in different areas of children’s
lives, was signiﬁcantly positively correlated to their Harter Global scores and
their mean school grades, and signiﬁcantly negatively correlated to their CDI
scores. Thus, children who have more adults who are able to provide support in

the different areas of their lives, are more likely to feel good about themselves, to

47

do better in school, and to rate themselves as having fewer feelings of sadness
and depression, than children who have adults in fewer (or no) areas of social
support. This ﬁnding is particularly important because no studies have examined
the impact of social support on children’s academic adjustment as measured by
their actual school grades. In addition, the presence of more adults in different
areas of social support was the only factor (of the ones tested) that signiﬁcantly
affected children’s scores on the CDI. Thus, this ﬁnding highlights the
importance of caring, supportive adults, other than the parents, in helping
children from violent homes adjust in several critical developmental arenas.

Although support provided from both mothers and other adults was
signiﬁcantly correlated to children’s adjustment in a variety of areas, the total
number of times fathers were listed in children’s social support sections were not
signiﬁcantly related to any of the outcome variables, regardless of whether or not
fathers were also assailants. These ﬁndings suggest that for school aged
children of abused women, a supportive, positive relationship with their mothers
or other caring adult, and their perception of how many people are supportive of
them, are more important to their well-being than their relationships with their
fathers.

There may be several reasons for these ﬁndings. First, whereas 93% of
children listed their mothers for at least one of the ﬁve social support areas, only
56% of children listed their fathers for at least one area of the support measure.
In fact, there was a higher percentage of children who listed an adult outside

their mother or father in at least one of the ﬁve areas (76%). Furthermore,

48

almost half (46%) of fathers listed as sources of support by their children, were
also their mothers’ abusers, potentially making these relationships confusing and
stressful for the children.

Finally, a vast majority of the children did not have regular contact with
their fathers, making their presence in the children’s lives less relevant. Only 3%
of children reported seeing their father sometime this week, and only one fourth
of the children (26%) saw their fathers within the last 5 months. In addition,
almost half of the children had not seen their fathers for a year or longer, or have
never seen their fathers. Considering this lack of contact, children mentioned
their fathers a surprising amount of times as sources of social support. Children
were most likely to say that their fathers "really care” about them (84%) and ”can
be counted on no matter what” (56%); percentages that were comparable to
those given for their mothers. However, children were about half as likely to say
that their fathers "listen“ to them (31%) or are "fun to talk to" (33%) and ”hang
out with (17%) as their mothers.

This data suggests that it is important for children to believe that
signiﬁcant peOple in their lives are emotionally invested in them, even if they do
not show it through their actions. Thus, fathers who are absent or are abusive to
their mothers, are still internalized as caring ﬁgures by these children. That is,
children may believe that in a hypothesized situation, their fathers would "be
there" for them, or that they care about the children "deep down inside." On the
other hand, children’s beliefs about their fathers roles as listeners, talkers, or

playmates, may be based more on their fathers’ actions and the lack of actual

49

time spent with the children.

In addition to fathers who were abusing their mothers, some children
(16%) also listed other male ﬁgures, who were their mothers’ assailants, as
sources of social support. The majority of these men were the children’s
stepfathers and their mothers’ current boyfriends. Similarly to the social support
proﬁle of biological fathers (assailant and non-assailant), non-father assailants
were also more likely to be listed as peOple who "really care" about the child and
”can be counted on no matter what" and much less likely to be listed for other
areas of social support (e.g., boyfriends were not listed at all for the other 3
areas). Contrary to the other assailants, two men who were categorized as
mothers’ former boyfriends, were listed only as people the children like to 'hang
out with."

Again, this data highlights the importance of asking children the ways in
which different people in their lives offer them support. Whereas fathers and
other male ﬁgures in children’s lives were mentioned as people who cared about
the children, this kind of theoretical support alone did not make a signiﬁcant
difference in children’s adjustment. This suggests that children need support
from a variety of areas in their life, and shows the importance of adults that listen
to, talk to, and spend quality time with school aged children. For these reasons,
the children’s mothers and other signiﬁcant adults in their lives were able to
make signiﬁcant contributions to their well-being, while their fathers (even the
non-abusing ones) could not.

While the current study enriches our understanding of the lives of children

50

of abused women, the results are qualiﬁed by several limitations which should be
addressed in future research. As mentioned previously, a larger sample of
children and their mothers would increase statistical power and provide more
variance on all measures. A larger sample would also allow for the examination
of gender differences, for which there was a lack of statistical power in this study.
A greater number of items on the social support measure would also provide
more variance for the children’s responses. In addition, a social support
satisfaction measure with better reliability would allow for examination of this
important aspect of the children’s support. Finally, the signiﬁcant ﬁndings need
to be interpreted cautiously due to the relatively large number of statistical tests
that were conducted (n=54).

Despite these limitations, several important ﬁndings and directions for
future research emerge from this data. First, this study indicates that not only
the presence of violence, as has been shown in the literature, but also higher
levels of violence witnessed by children, have negative effects on adjustment,
especially in the behavioral arena. This research also supports the current
ﬁnding in the literature that women’s reports of the levels of violence they
experience, (and their perceptions of how much violence their children witness),
are signiﬁcantly related to children’s outcomes.

The ﬁndings also show that speciﬁc types of social support can buffer
children from the negative effects of violence on their behavioral adjustment,
while global types of social support can have main effects on their development

in social, academic, and emotional arenas (including children’s self reported

51

outcome variables). Although it was hypothesized that conﬁdants buffer children
by providing esteem (caring) and informational support, it is not clear exactly
how the presence of conﬁdants in children’s lives protects them from the
negative effects of the violence. Future research that includes qualitative
measures (e.g., open-ended questions) of how children cope with witnessing
their mothers’ abuse and the role that conﬁdants play in their lives can clarify this
relationship.

The results also suggest that support from their mothers and other
signiﬁcant adults is more important to children than support from their fathers.
This appears to be due to the fathers’ lack of overall contact with the children,
and the lack of "quality time” spent with them (e.g., listening to them, talking to
them, and hanging out with them). In addition, the ﬁndings suggest, that,
although children’s social adjustment and self-esteem is affected by the total
number of people in their support network, their mothers’ support, and support
from other adults, there are certain areas of adjustment which are more affected
by their mothers’ support (e.g., extemallzing behaviors leading to suspensions),
and others which are more affected by signiﬁcant adults outside the family (e.g.,
academic adjustment and depression). Future research can explore the ways in
which mothers, versus other signiﬁcant adults, express their caring and concern
for the children, and the kinds of activities and topics that are shared between
the children and these adults.

In addition to introducing these important questions for further exploration,

the current ﬁndings have signiﬁcant implications for interventions with abused

52

women and their children. First, these ﬁndings suggest that separating children
from their abused mothers, as is often done in domestic violence situations, does
not maximize their adjustment. Rather, children’s mothers, who serve an
important supportive function in their lives, need to be provided with resources,
support, and respite care, that would allow them to be more emotionally and
physically available for their children. The mothers in the current study were
shown to be twice as likely as fathers to actively listen and talk to their children,
as well as spend time having fun with them. This increased attention and
“quality care” was shown to make a signiﬁcant difference in the children’s lives.
Yet, these women are given little outside support in order to continue, or expand,
this important role.

Because individual women have different needs, it is difﬁcult to prescribe
general services that would be helpful for all women in abusive relationships who
have school-aged children. However, services such as low-cost and ﬂexible
after school programs for children would allow women with unsupportive
partners to work or engage in other necessary activities while their children were
in a safe and stimulating environment. This would allow the women and their
children to spend more of their at-home time engaging in one-on-one activities.
At the same time, such programs would expand children’s social networks and
introduce them to signiﬁcant adults outside the home, both of which were shown
to have signiﬁcant positive effects on their adjustment.

A program that paired children from violent homes with community

volunteers, such as retired citizens or high school students, would also help

53

children by providing social support and building potential ”conﬁdant”
relationships. Again, all parents, especially those with unsupportive and abusive
partners, can provide better support to their children if their child-care duties are
regularly relieved by caring others.

It is also important for intervention programs to take a more individual
approach with abused women and their children, in order to provide support in
the areas most needed for that family. Thus, while some women are in need of
legal support, others may be struggling with housing issues. Similarly, while
some children have many friends and are in need of an adult to help them with
their homework, others need a group of peers with whom to share their
concerns. Thus, the ﬁeld needs to recognize the importance of offering support
to both abused women and their children, as well as conducting better
assessments as to which areas of support are most needed by each individual

family.

REFERENCES

55

 

REFERENCES

Achenbach, T.M., & Edelbrock, C. (1983). Mengal fer the Chile Beheyjet
Cheekliet. Available from T.M. Achenbach, University of Vermont, Burlington, VT.

Achenbach,T.M.,&Edelbrock, C. (1986). M n If r h hil B h vi r

l' v' il h vi rPr ﬁl .Burlington, VT: Queen City
Printers.

Achenbach,T.M.(1991). M ii If h hil B h v'
31111991210111. Burlington, VT: University of Vermont Press.

van Aken, M.A.G. & Asendorpf, J.B. (1997). Support by parents,
classmates, friends and siblings in preadolescence: covariation and

compensation across relationships. I f i l P l
Reletiegehjge, 14, 79-93.

Amato, PR, 8: Keith, B. (1991). Parental divorce and the well-being of
children: A meta-analysis. Peyehelegigl Bulletin, 1 19, 26-46.

Asamow, J. R., & Carlson, G ..A (1985). Depression Self-Rating Scale:

Utility with child psychiatric inpatients. W
W 491 -499

Bandura, A. (1986). i l f n
mggjﬂyemeem, Englewood Cliffs, N.J.: Prentice Hall.

Beck, AT, Ward, CH. Mendelson, M.., Mock, I., &Erbaugh, J. (1961).

An inventory for measuring depression. MW 561 -
571.

Bogat, G. A., Chin, R. Sabbath, W., 8. Schwartz, C. (1983). M13
W (Technical Report #2). East Lansing: Michigan

State University.

Browne, A. (1993). Violence against women by male partners:

Prevalence, outcomes, and policy implications. W.
1077- 1087.

Christopoulous, C. Cohn, DA, Shaw, D.S., Joyce, 8., Sullivan-Hanson,
J., Kraft, S.P., Emery, RE. (1987). Children of abused women: l. Adjustment at

time of shelter residence. Jegmel ef Marriage and the Family, 49, 611-619.

Cobb, S. (1976). Social support as a moderator of life stress.
Peyeheeematic Medieine, 38, 300-314.

56

Cohen, S. 8. Wills, TA (1985). Stress, social support, and the buffering
hypothesis. Eeyehological Belletin, 98(2), 310 - 357.

Cohen, S. & chennan, H. (1983). Positive events and social supports as
buffers of life change stress. oumal f li i l P l 1 , 1983.

Collins, N.L., Dunkel-Schetter, C., Lobel, M. & Scrimshaw, SC. (1993).
Social support in pregnancy: Psychosocial correlates of birth outcomes and
postpartum depression. I f-P n Ii
1243-1258.

Davis, L.V., & Carlson, BE. (1987). Observations of spouse abuse: what
happens to the children? J f l I ' I 278-291.

Dalgard, O. S., Bjork, S. &Tambs, K (1995). Soaal support negative life
events and mental health. W 29-34.

Emery, RE. (1982). Interparental conﬂict and children of discord and
divorce. Eeyehelogigl Bulletin, 92, 310-330.

Fantuzzo, J.W., DePaola, L..,M Lambert, L, Martino, T., Anderson, 6., &
Sutton, S. (1991). Effects of interparental violence on the psychological

adjustment and competencies of young children. W
W 258-265

Flannery, R.B. &Wieman, D. (1989). Social support, life stress, and

psychological distress. an empirical assessment. ,[QumetgLQunjeaLEsyenem
4_5 867-871.

Garrnezy, N. (1985). Stress-resistant children. The search for protective

factors In J E Stevenson (Ed) WM
steheeethgtegy New York: Guilford.

Gershon, E., Hamovit, J., Guroff, J., & Numberger, J. (1987). Birth-cohort
changes in manic and depressive disorders in relatives of bipolar and

schizoaffective patients. Arehivee Qf Cﬁneml Peﬂhietg, 44, 314-319.

Graham-Bermann, S.A., Coupet, S, Egler, L., & Mattis, J. (1996).
Interpersonal relationships and adjustment of children in homeless and

economically distressed families. Jegmel ef Clinieel Chile Eeyehelegy, 25, 250-
261.

Graham-Bennann, S.A., & Levendosky, AA (1998). The social

functioning of preschool-age children whose mothers are emotionally and
physically abused-W 59-84.

57

Hammon, C., 8. Rodolph, K.D. (1996). Childhood depression. In E.J.
Mash & R.A. Barkley (Eds.) Chﬂefeyeneeethetm New York: Guilford Press,
pp. 153-195.

Harrington, R., Fudge, H., Rutter, M., Pickles, A., & Hill, J. (1990). Adult
outcomes of childhood and adolescent depression: Psychiatric status. ﬂehjyee
47 465-473.

Hauser, S. T. & Bowlds, MK. (1990). Stress, coping, and adaptation. In

Feldman, S.S.&Elliot,G.R.(Eds.),A hr h l ' h v I ' n.
Cambridge, MA: Harvard University Press.

Harler, S. (1982). The perceived competence scale for children. Child.
[2129mm 87-97.

Harler, S. (1985). the eel f;ee ereetien emﬁle fer ehjleren; Reyjejen rife the
Eer rig lyee Seel e fe: Children. Denver, CO: University of Denver.

Hilberrnan, E., & Munson, K. (1977-78). Sixty battered women.
Vietimelegy; Ag lnternetienel Jeumal, 2, 460-470.

Holden, G.W., & Ritchie, KL. (1991). Linking extreme marital discord,
child rearing, and child behavior problems: evidence from battered women. Qh_i1d_

W 311-327.

Jaffe, P.G., Wolfe, D.A., & \Mlson, SK. (1990). Ctitlegeeeieatteted
womem Newbury Park, CA: Sage.

Jaffe, P., Wolfe, D., Wilson, 8., 8. Zak, L. (1986). Similarities in behavioral
and social maladjustment among child victims and witnesses to family violence.

' ii If h hi 142-146.

Jonge, J., Janssen, P.P.M. & Van-Breukelen, G.J.P. (1996). Testing the
Demand-Control-Support Model among health-care professionals: A structural
equation model. Wemjne-Stgese, 19, 209-224.

Jouriles, E. N., Murphy, C. M, & O’ Leary, K. D. (1989). Interspousal

aggression, marital discord, &child problems. JeemaLeLQeneujtmemt‘et
W 453-455

Koeske, G.F. & Koeske, RD. (1990). The buffering effect of social
support on parental stress. Amerieen ,teumel ef Qrthoeeyehjetgy, 69, 1990.

Kolbo, JR. (1996). Risk and Resilience Among Children Exposed to
Family Violence. yieleeee aee Victime, 11, 113-128.

58

Kovacs, M. (1981). Rating scales to assess depression in school-age

children. Acte Paedeesychiatm, 46, 305-315.

Li, L.W., Seltzer, M.M. & Greenberg, J.S. (1997). Social support and
depressive symptoms: Differential patterns in wife and daughter caregivers.
0 mal f ‘ol'l'l 0|" I ‘rla BZP OllH.-|, in ‘ -no 0 il in e
828482006211.

Licitra-Kleckler, D.M. & Waas, GA. (1993). Perceived social support
among high-stress adolescents: The role of peers and family. Jeemel ef
Aeeleéeent Rﬁeareh, 8, 381-402.

Llabre, M.M. & Hadi, F. (1997). Social support and psychological distress
in Kuwaiti boys and girls exposed to the Gulf crisis. f ' i I hil

Peyehelegy, 28, 247-255.

Masten, A., Best, K., 8. Garmezy, N. (1990). Resilience and development:
Contributions from the study of children who overcome adversity. 291919201201

ene Peyeheeethology, 2, 425-444.

Mathias, J. L., Mertin, P., & Murray, A. (1995). The psychological
functioning of children from backgrounds of domestic violence, Aeettettae

W 47-56

McCloskey, L.A., Figueredo, A.J., & Koss, MP. (1995). The effects of

systemic family violence on children’s mental health. W
1239-1261.

McCloskey, L.A., Southwick, K., Femandez-Esquer, M.E., & Locke, K.
(1995). The psychological effects of political and domestic violence on Central
American and Mexican immigrant mothers and children. Jeemel ef mmmenjty

W. 1995.

Moore, T., Pepler, D., Weinberg, B., Hammond, L. Waddel, J., & Weiser,
L. (1990). Research on children from violent families. CaeadeteMeeteLtleattm
98, 19-23.

Nielsen, J. M. Endo, R..,K & Ellington, B. L. (1992). Social isolation and
wife abuse: a research report. In Viano, E. C. (Ed.) lntimete ielenm;

W. Washington D. C: Hemisphere Publishing
Corporation.

O’Keefe, M. (1994a). Linking marital violence, mother-child/father-child
aggression, and child behavior problems. Jeemel of Family Vieleeee, 9, 63-78.

59

O’Keefe, M. (1994b). Adjustment of children from maritally violent homes.
Families in Society: The Journal of Contemporary Human Services, 403-415.

O’Keefe, M. (1995). Predictors of child abuse in maritally violent families._
Jeemel ef Intemersenal Violence, 10, 3-25.

O'Sullivan, C. M.. The relationship between childhood mentors and

resiliency in adult children of alcoholics. EamﬂLQyLIamMAdiQIimﬂuaneﬂx.
1 46-59.

Power, M.J. (1988). Stress-buffering effects of social support: a
longitudinal study. Motivatien ene Emetion, 12, 1988.

Quamma, J.P. & Greenberg, MT. (1994). Children’s experience of life
stress: The role of family social support and social problem-solving skills as

protective factors. Jeumel ef Clinieel Chile Peyehelegy, 23, 295-305.

Quay, H. (1977). Measuring dimensions of deviant behavior: the behavior

problem checklist. Jeemal ef Aenonnal Chile Peyehelegy, 8, 277-289.

Quay, H.C., & Peterson, DR. (1979). Meneel fer the Behevier Pteelem
Cheektjet. Available from D.R. Peterson, 39 North Fifth, Highland Park, NJ
08904.

Quay, H.C., & Peterson, DR. (1983). Manual fer the Bevieﬁ Behevier
Preelem Checklist. Coral Gables, FL: University of Miami Press.

Rawlins, C. (1993). The social networks of children of battered women.
Poster presented at the 1993 Biennial Meeting of the of the Society for Research
in Child Development, New Orelans, Louisiana.

Rhodes, J. (1994). Natural mentor relationships among Latina adolescent
mothers: Psychological adjustment, moderating processes, and the role of early

parental acceptance. Amerign Jeumel ef Qmmenity Peyehelegy, 22, 211-227.

Rosenbaum, A. & O’Leary, K.D. (1981). Children: The unintended victims
of marital violence. A_reeriee_r_i Onheeeyehiatric Associetiee, 81, 692-699.

Rosenberg, MS. (1984). The impact of witnessing interparental violence
on children: implications for the concept of intergenerational violence. Paper
presented at the Second National Conference for Family Violence, August 7-10,
Durham, New Hampshire.

Rutter, M. (1987). Psychosocial resilience and protective mechanisms.
Amerign Jeumel ef Orthoeeychiatm, §7. 316-331.

60

Rutter, M. (1990). Psychosocial resilience and protective mechanisms. In
J.Rolf, A. Masten, D. Cicchetti, K. Nuechterlein, 8 S. Weintraub (Eds), 3151531151

iv f in h vl m f h h l (pp.181-911). New
York: Wiley.

Ryan, N.D., Williamson, D.E., Iyengar, S., Orvaschel, H., Reich, T., Dahl,
RE, 8 Puig-Antich, J. (1992). A secular increase in child and adolescent onset

affective disorder. ml Am ri nA m f hil
PM iatu, 31, 600-605.

Sato, RA, 8 Heiby, EM. (1992). Correlates of depressive symptoms
among battered women. Joemal of Family Woleng, 7, 229-245.

Saylor, C.F., Finch, A.J., Spirito, A., 8 Bennett, B. (1984). The children’s
depression inventory: A systematic evaluation of psychometric properties.

Jeumal ef @neelting and Clinical Peyeheiegy, 52, 955-967.

Spaccarelli, S., Sandler, IN, 8 Roosa, M. (1994). History of spouse
violence against mother: correlated risks and unique effects in child mental

health. Jeemal ef Eamily Vielenee, 9, 79-98.

Stemberg, K.J., Lamb, M.E., Greenbaum, C., Cicchetti, D., Dawud, S.,
Cortes, R.M., Krispin, O., and Lorey, F. (1993). Effects of domestic violence on

children’s behavior problems and depression. v I m n IP h l 2
44-52.

Straus, MA. (1979). Measuring intrafamily conﬂict and violence: The
Conﬂict Tactics (CT) Scales. Jeemal of Mam'age ane the Eamily, 5]. 75-88.

Straus, MA, 8 Gelles, R. (1986). Societal change and change in family
violence from 1975 to 1985 as revealed by two national surveys. JeumaLef

Marriage ane the Family, 48, 465-479.

Straus, M.S.Gelles, R.,8Steinmetz, S. (1980). getiiedelesedjm

W New York: Doubleday.
Sullivan, C.M., Basta, J., Tan, 0., 8 Davidson, WS (1992). Brief Report.

After the Crisis: A needs assessment of women leaving a domestic violence
shelter. Vielenee ane Vietime, 7, 267-275.

Sullivan, C.M., Juris, J., Gauthier, L., Nguyen, H., 8 Prewitt, A. (1997).
I : 'win hiI ‘ -H_ ,‘1 iv' -‘.'r'-ii.-I r :-I' -no

eensieeratiene, Technical report available through the ﬁrst author. Michigan
State University, East Lansing, MI.

Sullivan, C.M., Tan, C., Basta, J., Rumptz, M., 8 Davidson, WS (1992).

61

An advocacy program for women with abusive partners: initial evaluation.

Amerign Jeemal ef Cemmunity Esychology, 20, 309-332.

Sullivan, C.M., Cambell, R., Angelique, H., Eby, K.K., 8 Davidson, W.S.
(1994). An advocacy intervention program for women with abusive partners: six-

month follow-up. Amerieen Jeemal ef @mmenity Peyehelegy, 22, 101-122.

Tang, S.K.C. (1997). Psychological impact of wife abuse: Experiences of

Chinese women and their children. Jeemal ef Integeeteenal Vielenee, 12, 466-
478.

Vitanza, S., Vogel, L. C. M 8 Marshall, LL. (1995). Distress and
symptoms of posttraumatic stress disorder in abused women. \_/_iel_e_i3ee_8_
Vietime, 19, 23-34.

Vygotsky, LS. (1978). Cole, M., Steiner, V.J., Scribner, S., 8 Souberrnan,
E. (Eds.) Mine in 80ciety. Cambridge, Mass: Harvard University Press.

Wasserstein, S. B. 8 La-Greca, A. M. (1996). Can peer support buffer
against behavioral consequences of parental discord? MW

Eeyehelegy, 25, 177- 182.

Werner, BE. (1993). Risk, resilience, and recovery: Perspectives from
the Kauai longitudinal study. Special Issue: Milestones in the development of
resilience. VI in n an P h hl 503-515.

Wertlieb, D., Weigel, C., 8 Feldstein, M. (1987). Stress, social support,
and behavior symptoms in middle childhood. JeemajeLQtjnijCme

Peyehelegy, 18, 204-211 .
Wilcox, BL. (1981). Social support, life stress, and psychological

adjustment: a test of the buffering hypothesis. WW
Esxgnoiogxza. 371-386.

Wolfe, D.A., Jaffe, P, Wilson, S. 8 Zak, L. (1985). Children of battered
women: The relation of child behavior to family violence and maternal stress.

Jeumal ef Qeneelting 8 Clinical Peyeholegy, 58, 657-665.

Wolfe, D.A., Zak, L., Wilson, S.K., 8 Jaffe, P. (1986). Child witnesses to
violence between parents: critical issues in behavioral and social adjustment.

Jeemal ef Aenerrnal Child Peyehelegy, 14, 95-104.

Ystgaard, M. (1997). Life stress, social support and psychological
distress in late adolescence. i I hi n hi ri E i mi l
82,277-283.

62

APPENDICES

63

APPENDIX A

WITNESSED VIOLENCE SCALE - CHILD REPORT
Now I’d like to ask you some questions about your mom and Assailaet’e name
1. Think about how things have been going in the last 4 months, like since about

hetieaz er eeeeifie time mark. Since then, how often would you say you’ve heard
your mom and Assailant’e name arguing?

(SHOW YELLOW CARD# 2)
NEVER ..................... 1
A LITTLE .................... 2
SOMETIMES ................. 3
A LOT ...................... 4
Not applicable ................ 8

2. How often would you say you’ve heard your mom and Aeeailaet’a name
yelling and screaming at each other, or one screaming or yelling at the other?
(SHOW YELLOW CARD# 2)

NEVER ..................... 1
A LITTLE .................... 2
SOMETIMES ................. 3
A LOT ...................... 4
Not applicable ................ 8

3. There are many things people do to annoy or hurt each other, such as
making fun of someone or calling them names, and saying things to make them
feel bad. How often has Aesailant’ a name said or done any of these type of

things to your mom, since about WM
(SHOW YELLOW CARD# 2)

NEVER ..................... 1
A LITTLE .................... 2
SOMETIMES ................. 3
A LOT ...................... 4
Not applicable ................ 8

4. About how often have you seen or heard Aaeai/aet’a name threaten to hurt
your mom, since about belieay er eeeeiﬁe time mark?

(SHOW YELLOW CARD# 2)
NEVER ............................ 1
ONCE A MONTH OR LESS ............ 2
2 OR 3 TIMES A MONTH .............. 3
ONCE OR TWICE A WEEK ............ 4
3 OR 4 TIMES A WEEK ............... 5
MORE THAN 4 TIMES A WEEK ........ 6
Not Applicable ...................... 8

5. About how often have you seen or heard Aeeailagt’a name hurt your mom

physically, since about helieay er eeeeiﬁe time mam? By that I mean anything
from grabbing or slapping her to punching or kicking her, anything like that.

(SHOW YELLOW CARD# 2)
NEVER ............................ 1
ONCE A MONTH OR LESS ............ 2
2 OR 3 TIMES A MONTH .............. 3
ONCE OR TWICE A WEEK ............ 4
3 OR 4 TIMES A WEEK ............... 5
MORE THAN 4 TIMES A WEEK ........ 6
Not Applicable ...................... 8

65

APPENDIX B

WITNESSED VIOLENCE SCALE - MOTHER REPORT

1. There are some things that men might do to annoy or hurt their partners such
as ridiculing, criticizing, controlling, or humiliating them. I want you to recall how

many times Chi/ea name has ever seen or heard Aeaailant’a name do any of
these things to you.

(SHOW YELLOW CARD# 2)
NEVER .................. 1
RARELY .................. 2
SOMETIMES .............. 3
OFTEN .................. 4

2. Now I would like to ask you questions about whether thle’e name might have
seen or heard any of the abuse you’ve experienced. How many times has

thle’e name ever seen or heard Aesailant’e name threaten you? Would you
say:

(SHOW PINK CARD# 3)
NEVER ............................ 1
ONCE A MONTH OR LESS ............ 2
2 OR 3 TIMES A MONTH .............. 3
ONCE OR TWICE A WEEK ............ 4
3 OR 4 TIMES A WEEK ............... 5
MORE THAN 4 TIMES A WEEK ........ 6
Not Applicable ...................... 8

3. How many times has thld’a name ever seen or heard Aaaailant’e name
physically harm you, or attempt to harm you?

(SHOW PINK CARD# 3)
NEVER ............................ 1
ONCE A MONTH OR LESS ............ 2
2 OR 3 TIMES A MONTH .............. 3
ONCE OR TWICE A WEEK ............ 4
3 OR 4 TIMES A WEEK ............... 5
MORE THAN 4 TIMES A WEEK ........ 6

Not Applicable ...................... 8

66

APPENDIX C

EXPERIENCE OF VIOLENCE - MOTHER

1. Now I have a list of some of the emotional things men might do to annoy or
hurt their partners and ex-partners. Using this card (SHOW YELLOW CARD #2)
could you tell me, to the best of your recollection, how many times in the last 4

months Assailant’s name did any of these things to annoy or hurt you?

 

NEVER ...............

RARELY ..............
SOMETIMES ..........

OFTEN ...............

Not applicable ..........
a. How often has he refused to talk to you? .........................

b. How often has he accused you of having or wanting other sexual
relationships? .............................................. .

...1

....2
....3

...4

....8

c. Told you about other sexual relationships he wanted or was having in order to

hurt you .....................................................
d. Tried to control your money ...................................
e. Tried to control your activities ................................

f. Lied to you or deliberately misled you . . . .......................
g. Called you names ...........................................
h. Tried to humiliate you ........................................
l. Ignored or made light of your feelings. . . . .......................
j. Ridiculed or criticized you ....................................
k. Criticized your family or friends to you . . ........................
l. Harassed your family or friends in some way . ....................
m. Discouraged your contact with family or friends ....................

n. Threatened to hurt your family or friends .........................

67

NEVER .................. 1

RARELY .................. 2
SOMETIMES .............. 3
OFTEN .................. 4
Not applicable .............. 8

0. Broken or destroyed something important to you ..................

p. Do you have or have you had pets in the last four months? (IF YES):
Abused or threatened to abuse your pets to hurt you ...............

q. Punished or deprived the children when he was angry at you ......... _
r. Threatened to take the children away from you ...................
s. Left you somewhere with no way to get home ....................

t. Threatened to end the relationship if you didn’t do what he wanted ......
u. Tried to force you to leave your home ..........................
v. Threatened to commit suicide when he was angry at you ............
2. How many times in the last 4 months did Aeaailant’e name threaten you in
any way? By that I mean said or did things that made you feel scared or in

danger, whether in person, over the phone, through the mail, or through other
people. Would you say:

(SHOW PINK CARD#3)
NEVER ............................ 1
ONCE A MONTH OR LESS ............ 2
2 OR 3 TIMES A MONTH .............. 3
ONCE OR TWICE A WEEK ............ 4
3 OR 4 TIMES A WEEK ............... 5
MORE THAN 4 TIMES A WEEK ........ 6
Not Applicable ...................... 8

Now I’d like to talk about the violence you’ve experienced from Aesai‘lant’e name.
First I have some speciﬁc questions to ask, but then we can talk a little more
informally about what happened if you want, Okay?

3. First I have a list of different types of physical violence that women have
experienced from their partners and ex-partners. I wonder if you could tell me,

to the best of your recollection, whether Aaeailant’a name has eﬂ done any of
the following things to you:

68

Has he ever: EVER IN THE LAST 4 MmtlTHS?

YES NO YES NO N/A

a. Broken your glasses or tom 1 . . . 2 1 ...... 2 ...... 8
your clothing?

b. Pushed, grabbed, shoved or 1 . . .2 1 ...... 2 ...... 8
or slapped you?

c. Punched, kicked, choked 1 . . .2 1 ...... 2 ...... 8
or burned you?

d. Thrown something at you, 1 . . . 2 1 ...... 2 ...... 8
tried to hit you with an object,
or hit you with an object?

.3
N
.—l
N
m

e. Forced sexual activity?

f. Tied you up or physically 1 . . . 2 1 ...... 2 ...... 8
restrained you in some way?

9. Threatened you with a gun or 1 . . . 2 1 ...... 2 ...... 8
knife or used a gun or knife
against you?

4. (IF SHE HAS BEEN HARMED AT ALL) Now I’m going to go through a list of
physical injuries and ask you yes or no if you ever suffered these injuries from
his abuse.

 

Did YOU ever suffer ___ER W
YES NO YES NO N/A
a. Soreness without bruises? 1 . . .2 1 ...... 2 ...... 8

.3
N
_s
N
m

b. Cuts, scrapes, bruises, red
marks, or welts?

c. Burns, including rug burns? 1 . . . 2 1 ...... 2 ...... 8
d. Loose or broken teeth? 1 . . . 2 1 ...... 2 ...... 8
e. Broken bones or fractures? 1 . . . 2 1 ...... 2 ...... 8

69

f. lntemal injuries?
9. Strains or sprains?

h. Pregnancy complications or
miscarriage?

I. Knife or gunshot wound?
j. Permanent scarring?
k. Any other physical injuries

I haven’t mentioned?
(Specify)

 

70

1 ...... 2 ...... 8
1 ...... 2 ...... 8
1 ...... 2 ...... 8
1 ...... 2 ...... 8
1 ...... 2 ...... 8
1 ...... 2 ...... 8

 

APPENDIX D

PERCEIVED COMPETENCE SCALE

What I Am Like

 

 

 

 

_ ﬂ- SAMPLE SENTENCE _
Really Sort Sort Really
True of of True
for me True True for me
for me for me

Some kids BUT Other kids
a. D D would rather would rather D D

play outdoors watch TV

in their spare

time
1. D D fSome kids BUT Other kids D D

eel that they worry about

are very whether they

good at their can do the

school work school work

assigned to
them

Some kids BUT Other kids ﬁnd
2' D 0 ﬁnd it hard to It’s pretty easy C] D

make friends to make friends.
3. D D Some kids do BUT Other kids don’t E] El

very well at feel that they

all kinds of are very good

sports when it comes

to sports.

71

 

Really Sort

True of

for me True
for me

 

4.0 CI

Some kids
are happy
with the way
they look

Some kids

Often do not
like the way
they behave

Some kids
are often
unhappy with
themselves

Some kids
feel like they
are just as
smart as
other kids
their age

Some kids
have a lot of
friends

Some kids
wish they
could be a lot
beﬂerat
spons

 

BUT

BUT

BUT

BUT

BUT

BUT

72

Other kids are
not happy with
the way they
look.

Other kids
usually like the
way they
behave.

Other kids are
pretty pleased
with
themselves.

Other kids
aren’t so sure
and wonder if
they are as
smart.

Other kids don’t
have very many
friends.

Other kids feel
they are good
enough at
spons.

 

Sort Really
of True
True for me
for me

Cl
El III
Cl El
Cl E]
E] El
Cl Cl

 

Really Sort Sort Really

 

 

True of of True
for me True True for me
for me for me
10. D D Some kids BUT Other kids wish D D

are happy their height or
with their weight were
height and different.
weight

11. D D Some kids BUT Other kids often 0 D
usually do don I do the
the right thing right thing

12 D D Some kids BUT Other kids do D D

don't like the like the way
way they are they are leading
leading their their life.

life

13 Ci 0 Some kids BUT Other kids can D D

are pretty do their
slow in schoolwork
ﬁnishing their quickly.
school work

14. D Ci Some kids BUT Other kids have D Ci

would like to as many friends
have a lot as they want.
more friends

15. D D Some kids BUT Other kids are D a

think they afraid they
could do well might not do
at just about well at sports
any new they haven’t
sports activity tried before.
they haven’t

tried before

73

 

Really Sort Sort Really

 

 

 

True of of True
for me True True for me
for me for me
Some kids BUT Other kids like
16' D D wish their their body the D D

body was way it is.
diﬂeient

17 D D Some kids BUT Other kids often D D

usually act don’t act the
the way they way they are
know they supposed to.
are supposed

to

18 Ci 0 Some kids BUT Other kids are D D

are happy often not happy
with with
themselves themselves.

as a person

19. D Ci Some kids BUT Other kids can 0 D

often forget remember
what they things easily.
learn
20. Cl C] Some kids BUT Other kids Ci D
are always usually do
doing things things by
with a lot of themselves
kids

21 D D Some kids BUT Other kids don’t D D

feel that they feel they can
are better play as well.
than others

their age at

sports

74

 

Really Sort

 

True of
for me True
for me

22. Cl CI
23. CI CI
24. D CI
25. CI CI
26. E] El

Some kids
wish their
physical
appearance
(how they
look) was
different

Some kids
usually get in
trouble
because of
things they
do

Some kids
like the kind
of person
they are

Some kids do
very well at
their
classwork

Some kids
wish that
more people
their age
liked them

Sort Really
of True
True for me
for me

 

BUT

BUT

BUT

BUT

BUT

75

Other kids like
their physical
appearance the
way it is.

Other kids
usually don't do
things that get
them in trouble.

Other kids Often
wish they were
someone else.

Other kids don't
do very well at
their classwork.

Other kids feel
that most
people their age
do like them.

CI E]
El CI
Cl E]
El Cl
E] El

 

Really Sort

True of

for me True
for me

 

27.0 0

29.0 0

30.0 0

32.0 0

In games and
sports some
kids usually
watch
instead of

play

Some kids do
things they
know they
shouldn't do

Some kids
are very
happy being
the way they
are

Some kids
have trouble
ﬁguring out
the answers
in school

Some kids
are popular
with others
their age

BUT

BUT

BUT

BUT

BUT

 

Other kids
usually play
rather than just
watch.

Other kids
hardly ever do
things they
know they are
not supposed to

Other kids wish
they were
different

Other kids
almost always
can ﬁgure out
the answers in
school

Other kids are
not very
popular.

Sort Really
of True
True for me
for me

 

ti Cl

0 0
0 0
0 0
0 0

 

76

APPENDIX E

CHILDHOOD DEPRESSION INVENTORY

Instructions: Kids sometimes have different feelings and ideas. This
questionnaire lists feelings and ideas in groups. From each group, I would like
you to pick one sentence that describes you best for the PAST TWO WEEKS.
Remember, there are no right or wrong answers. Just pick the sentence that
best describes the way you have been recently. (MARK AN "X" IN THE
APPROPRIATE BOX, OR ALLOW THE CHILD TO MARK THE BOX IF Sll-IE
WISHES TO DO 80).
Here is a sample question:

E] I read books all the time
Cl I read books once in a while
III I never read book

 

DOC]

DUE]

Item 1
I am sad once in a while
lam sad many times
I am sad all the time

Item 2
Nothing will ever work out for
me
I am not sure if things will work
out for me
Things will work out for me

Item 3
I do most things OK.
I do many things OK.
I do everything wrong

77

DDU

DUO

Item 4
l have fun in many things
I have fun in some things
Nothing is fun at all

ltem5
I am bad all the time
lam bad many times
lam bad once in a while

I Item 6
I think about bad things happening
to me
I worry that bad things will happen
to me
I am sure that terrible things will
happen to me

Item 7
I hate myself
I do not like myself
I like myself

ODD

Item 8
All bad things are my fault
Many bad things are my fault
Bad things are not usually my
fault

DUE]

Item 9
D I do not think about killing
myself
D I think about killing myself BUT
I would not do it
D I want to kill myself

Item 10
I feel like crying everyday
I feel like crying many days
I feel like crying once in a while

EJDCI

Item 1 1
Things bother me all the time
Things bother me many times
Things bother me once in a
while

DOD

Item 12
I like being with people
I do not like being with people
many times
D i do not want to be with people
at all

DD

78

Item 13
D I cannot make up my mind about
things
D It is hard to make up my mind
about things
D I make up my mind about things

Item 14
I look OK.
There are some bad things about
my looks
D I look ugly

DU

Item 15
D I have to push myself all the time
to do my schoolwork
D l have to push myself many times
to do my schoolwork
D Doing schoolwork is not a big
problem

Item 16
D l have trouble sleeping every night
D I have trouble sleeping many
nights
D I sleep pretty well

Item 17
I am tired once in a while
I am tired many days
I am tired all the time

DUO

Item 18
Most days I do not feel like eating
Many days I do not feel like eating
I eat pretty well

DUE]

Item 19

Item 24

D I do not worry about aches and D I can never be as good as other
pains kids
D I worry about aches and pains D I can be as good as other kids if I
many times want to
D I worry about aches and pains D I am just as good as other kids
all the time
Item 20 Item 25
D I do not feel alone D Nobody really loves me
D I feel alone many times D I am not sure if anybody loves me
D I feel alone all the time D I am sure that somebody loves me
Item 21 Item 26
D I never have fun at school D I usually do what I am told
D I have fun at school only once D ldo not do what I am told most
in a while times
D l have fun at school many times D I never do what lam told
Item 22 Item 27
D l have plenty of friends D I get along with people
D I have some friends BUT I wish D I get into ﬁghts many times
I had more D i get into ﬁghts all the time
D I do not have any friends
Item 23
D My schoolwork is alright
D My schoolwork is not as good

as before
I do very badly in subjects I
used to be good in

Copyright @1992 Maria Kovacs, Ph.D.
Multi-Health Systems, Inc. All rights reserved.

79

APPENDIX F

BEHAVIOR RATING SCALE
Below is a list of items that describe children and youth. For each item that
describes your child now or within the past 4 months, please circle the 2 if the
item is very true or often true of your child. Circle the 1 if the item is somewhat
or sometime true of your child. If the item is not true of your child, circle the 0.

Please answer all items as well as you can, even if some do not seem to apply

to your child.
Very True Somewhat Not True
or Often True or
True Sometime
True
1. Argues a lot ..................... 0 .......... 1 ........... 2
2. Can’t concentrate, can’t pay ......... 0 .......... 1 ........... 2
attention for long
3. Can’t sit still, restless, or hyperactive . 0 .......... 1 ........... 2
4. Clings to adults or too dependent . . . . 0 .......... 1 ........... 2
5. Complains of loneliness ............ 0 .......... 1 ........... 2
6. Day-dreams or gets lost in his/her. . . . 0 .......... 1 ........... 2
thoughts. 0 . . . . 1 ...... 2
7. Disobedient at home .............. 0 .......... 1 ........... 2
8. Fears he/she might think or do ...... 0 .......... 1 ........... 2

something bad

80

9. Feels or complains that no one ..... 0 .......... 1 ........... 2
loves him/her

 

10. Feels worthless or inferior ........ 0 .......... 1 ........... 2
11 . Gets in many ﬁghts .............. 0 .......... 1 ........... 2
12. Would rather be alone than with . . . . 0 .......... 1 ........... 2
others

13. Poor school work ............... 0 .......... 1 ........... 2
14. Poorly coordinated or clumsy ..... 0 .......... 1 ........... 2
15. Stubborn, sullen, or irritable ....... 0 .......... 1 ........... 2
16. Temper tantrums or hot temper . . . . 0 .......... 1 ........... 2
17. Underactive, slow moving or lacks . . . 0 .......... 1 ........... 2
energy
18. Unhappy, sad, or depressed ....... 0 .......... 1 ........... 2
19. Worries ....................... 0 .......... 1 ........... 2
20. In the last 4 months, has your child been on any prescription medication
(for example, Ritalin) for behavior problems? (Circle '1' or '2 below)

Yes. . . . 1

(If yes, what medication? )

NO . . . . 2

81

APPENDIX G

NUMBER OF SUSPENSIONS

1. Has (thle’e name) ever been suspended from school?

YES ............ . 1

NO .............. 2

1a. How many times?

INDICATE EXACT NUMBER (IF "0" WRITE ”0”)

82

APPENDIX H

CHILD SOCIAL SUPPORT QUESTIONNAIRE

lnetgectie a te Interviewer: The purpose of this section is to get as many people
in the child’s life that provide him/her with support. Make sure that you give the
child enough time to think about who these peOple might be. Prompt or probe
for responses, without being pushy. Put the ﬁrst name or each person, their last
initial, and their relationship to the child on each line. Do not exceed the ten
spaces. For example: child says she hangs out with Sally. Ask what Sally’s last
initial is. Ask child who Sally is (a friend, cousin, sister, etc.). If child can’t think
of anyone, put "no one." Indicate matemal/patemal if necessary to clarify who
you are talking about (e.g., if there are 2 Aunt Marys). Probe for approximate
age (e.g., 'ls Jose a grown-up?" ”ls Nina your age?").

Inetaigaiene te thle: I’d like to start out by asking you about peOple in your life
that you may do certain things with, or peOple that may do certain things for you.
I want you to list as many people as you can for each question. I will ask you for
their name, and their relationship to you (like mom, dad, friend, teacher). I will
also ask you for the ﬁrst letter of their last name.

1. Who do you hang out with, for example, at their house, your house, around
the neighborhood, or at school, etc.?

 

 

 

 

6) f)
b) 9)
c ) h)

 

 

d) I)
e) i)

 

 

2. Who do you think are fun people to talk with, for instance, about things you
like to do or TV shows, etc?

 

 

 

 

 

 

a) f)
b) 9)
C) h)

d) I)
e) i)

 

 

83

3. Who listens to you when you need to talk about something personal, for
example, about problems between you and your parents, problems with friends,

things like that?

a)
b)
c )
d)

8)

 

 

 

 

 

4. Who do you feel really cares about you?

a)

 

b)

 

0)
d)

 

 

9)

 

 

 

 

 

 

 

 

 

 

 

5. Who can you really count on to always be there for you?

a)

 

b)

 

C)
d)

 

 

e)

 

1’)
9)

h)

 

 

 

 

 

APPENDIX I

 

 

Table1
rr l i it Am I f h B h vi in I
lntemalizing Extemalizing Total
lntemalizing 1 .00 —
Extemalizing .90“ 1 .00 —
Total .85" .52” 1 .00

 

** signiﬁcant at the 0.01 level

85

APPENDIX J

 

 

Table 2

D; ri-iv- of onin . Ao‘ mn Vic-In - . il H. .n-

Demeggaehic Variables
Variable N Min. Max. Mean Std. Dev.
Total Behaviors 80 0.00 30.00 14.21 6.86
CDI Total 75 0.00 26.00 9.03 6.32
Harler Global 80 1.50 4.00 3.21 0.68
Mean Semester Grade 42 1.00 3.00 2.47 0.46
# of Suspensions 80 0.00 4.00 0.31 0.73
Child Witness of Abuse 79 082 1.91 0.57 0.72
Mom Report Own Abuse 80 -1.79 2.14 -4.56 E-16 0.83
Morn Report Child Witness 80 -0.83 2.56 0.68 0.84
Mom Social Support 80 0.00 5.00 2.90 1.38
Dad Social Support 80 0.00 5.00 1.30 1.37
Adult Social Support 80 0.00 5.00 1.66 1.36
Conﬁdant Social Support 80 0.00 5.00 1.25 1.24
Total Social Support 80 3.00 27.00 10.52 4.30
Age 80 6.00 1 1 .00 8.55 1 .50

 

Nete. Because the Likert responses for these scales had different ranges (e.g.,
1-4 vs. 1-6), the responses were standardized into z-scores. For non-
standardized report of children’s exposure to violence, see Tables 4 and 5.
Mom SS = number of times (out of 5) mother mentioned in social support.
Dad SS = number of times (out of 5) father mentioned in social support.

Adult SS = number of times (out of 5) other adults mentioned in social support.
Conﬁdant = person who cares, listens, and can be counted on.

Conﬁdent SS = number of conﬁdants mentioned in social support.

Total SS = total number of non-repeating persons in social support.

86

APPENDIX K

 

 

Table 3
err-l. ion oofA' inn Vi-.In .: ule .000 :no I-mmruhi

Variablee

Total CDI Harter Number of Mean
Behaviors Total Global Suspensions Grade

Total Behaviors 1.00
CDI Tototal .11 1 .00
Harter Global -.12 -.43*** 1.00
Number of 30*” -.06 .01 1.00
Suspensions
Mean Grade -.43*** -.40*** .15 -.12 1.00
Viﬁtness Scale .08 -.07 .00 -.04 .30”
(Child)
Witness Scale 32*“ -.02 -.13 .11 -.09
(Mom)
Experience of .37” -.03 -.05 .23“ -.05
Violence (Mom)
Adult SS .14 -.10 .04 -.07 .16
Total SS -.05 -.04 .13 .07 .02
Mom SS -.02 -.02 .27“ -.11 .11
Dad SS -.02 .18 -.03 -.08 -.02
Conﬁdant SS -.14 -.09 -.18 -.13 .26
Age .00 -.30*** -.04 .14 .01
Sex .16 -.17 .14 .23" .10
Race -.1 1 .02 .00 -.14 -.23

87

Table 3 (cont’d)

 

 

 

 

 

 

Witness Witness Experience Adult Total
_ Scale (Child) Scale (Mom) of Violence SS SS
Witness Scale 1.00
(Child)
Witness Scale 39*“ 1.00
(Mom)
Experience of 39*“ .67*** 1.00
Violence (Mom)
Adult SS -.02 -.05 -.029 1.00
Total SS -.05 -.18 -.12 .24“ 1.00
Mom SS -.24** -.10 .14 .28“ .07
Dad SS .04 -.04 .14 .09 .38”
Conﬁdent SS -.04 -.12 .07 .15
Age -.12 -.17 -.10 .05 -.06
Sex -.14 -.05 .01 .12 -.14
Race -.15 -.18 -.22** -.07 .16
Mom SS Dad SS Age Sex Race
Mom SS 1.00
Dad SS 36*“ 1.00
Conﬁdant SS .59” .52” -.014 .00 -.19*
Age -.05 -.25** 1 .00
Sex .08 -.13 .19 1.00
Race -.15 -.03 .06 -.13 1 .00

 

Nete: * p<.1; “p<.05, ***p<.01; Sex:1=female; 2=male; Race:1=white; 2=minority
Mom SS = number of times (out of 5) mother mentioned in social support.

Dad SS = number of times (out Of 5) father mentioned in social support.

Adult SS = number of times (out of 5) other adults mentioned in social support.

Conﬁdant = person who cares, listens, and can be counted on.

Conﬁdant SS = number of conﬁdante mentioned in social support.
Total SS = total number of non-repeating persons in social support.

88

APPENDIX L

 

 

Table4

P r n f hiI nR in Vlﬁn in Ve | nﬂi an A

Frequency of Witnessing Yelling Arguing Ridiculing of Mom
Never 13% 9% 19 %

Alittle 15% 9% 13 %
Sometimes 33% 30% 29.5%

A lot 39% 52% 38.5%

 

89

APPENDIX M

Table 5

Pr n hildrnR in Win in Thr An Phi IAb

 

 

Frequency of Witnessing Threats Physical Harm
Never 32% 38%
Once a month or less 23% 31%
2 or 3 times a month 10% 10%
Once or twice a week 8% 3%
3 or 4 times a week 4% 5%
More than 4 times a week 24% 13%

 

90

APPENDIX N

Table 6

 

 

hildrn’ L n wihBil i IF hr

Last seen Percentage of children
Sometime this week 3%

Less than 2 weeks ago BUT not this week 4%

2-3 weeks ago 8%

1-2 months ago 6%

3-5 months ago 5%

6-11 months ago 29%

12 or more months, or never seen father 46%

 

91

APPENDIX 0

Table 7

Fr n fhil n’s n wihBiI i IF hrinh L 4M nh

 

Frequency of Contact Percentage of Fathers Percentage of Fathers

 

not listed as Social listed as Social Support
Support (n=35) (n=45)

A lot 3% 4%

Sometimes 0% 24%

A little 23% 29%

Not seen in last 4 74% 42%

months

 

92

APPENDIX P

 

 

Table 8
Pern of hilor-n Li in- Moh- F h- .nAo- iln oli. ...
Support Source N Cares Count Listens Talks Hang
On Out
Mothers 74 97% 74% 76% 46% 22%
Biological Fathers
All 45 84% 56% 31 % 24% 1 1%
Non-Assailant 24 71% 58% 33% 33% 17%
Assailant 21 100% 52% 29% 14% 5%
Other Assailants 13
Step Father 6 100% 33% 17% 33% 17%
Mom’s Boyfriend 5 80% 60% 0% 0% 0%
Ex-Boyfriend 2 0% 0% 0% 0% 100%

 

93

APPENDIX Q

Table 9

Numeer ef Qenﬁeante ae Beﬂer fer Ngative Effeet ef Vlﬂtneseee Vielenm en
Children’s Total BehavioLs

 

 

Standard a R2 e-value

Step 1
Child Witness of Abuse 0.077 0.249
Number of Conﬁdants -0.132 0.125

0.024 0.199
Step 2
Witness X Conﬁdants -0.276 0.062
Change 0.030

Total 0.054 0.062

 

94

"Ilillllilllllllillllllii