Close
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Families exposed to community violence: The impact of maternal distress symptomatology on Latino and African -American young adolescents
(USC Thesis Other)
Families exposed to community violence: The impact of maternal distress symptomatology on Latino and African -American young adolescents
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. U M I films
the text directly from the original or copy submitted. Thus, some thesis and
dissertation copies are in typewriter face, while others may be from any type of
computer printer.
The quality of this reproduction is dependent upon the quality of the
copy submitted. Broken or indistinct print, colored or poor quality illustrations
and photographs, print bleedthrough, substandard margins, and improper
alignment can adversely affect reproduction.
In the unlikely event that the author did not send UM I a complete manuscript
and there are missing pages, these will be noted. Also, if unauthorized
copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and continuing
from left to right in equal sections with small overlaps.
ProQuest Information and Learning
300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA
800-521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
FAMILIES EXPOSED TO COMMUNITY VIOLENCE:
THE IMPACT OF MATERNAL DISTRESS SYMPTOMATOLOGY
ON
LATINO AND AFRICAN-AMERICAN YOUNG ADOLESCENTS
by
Eugene Francis Aisenberg
A Dissertation Presented to the
FACULTY OF THE SCHOOL OF SOCIAL WORK
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(SOCIAL WORK)
May 2002
Copyright 2002 Eugene Francis Aisenberg
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 3073736
Copyright 2002 by
Aisenberg, Eugene Francis
All rights reserved.
___ ®
UMI
UMI Microform 3073736
Copyright 2003 by ProQuest Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UNIVERSITY OF SOUTHERN CALIFORNIA
The Graduate School
University Park
LOS ANGELES, CALIFORNIA 90089-1695
This dissertation , w ritten b y
/P /s e v b e r j
Under th e direction o f h.J..L. D issertation
Com m ittee, an d approved b y a ll its m em bers,
has been p resen ted to an d a ccep ted b y The
Graduate School, in p a rtia l fu lfillm en t o f
requirem ents fo r th e degree o f
DOCTOR OF PHILOSOPHY
/ " D e a n i i o f G raduate S tu dies
D ate
DI SSER T A TION COMMITTEE
* ■ — — v a h
o Lsl - - (
Chairperson
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Eugene Francis Aisenberg Ferol Mennen
ABSTRACT
FAMILIES EXPOSED TO COMMUNITY VIOLENCE:
THE IMPACT OF MATERNAL DISTRESS SYMPTOMATOLOGY
ON LATINO AND AFRICAN-AMERICAN YOUNG ADOLESCENTS
Community violence is a serious risk factor that has potential effects on
children's health as well as school and social functioning (Jenkins & Bell,
1997). Typically, research has focused predominantly on African-American
youngsters while rarely obtaining information on parent’s exposure to
community violence (Aisenberg & Mennen, 2001). As a result, systematic
research has not yet been conducted on the effects of chronic community
violence and the caregiving environment (Osofsky & Scheeringa, 1997).
OBJECTIVE: This study aims to: 1) identify and measure the type, frequency,
and intensity of exposure to community violence experienced by 80 pairs of
Latino and African-American mothers and their children attending two public
middle schools in the same low income neighborhood; 2) measure the
behavioral and psychological effects of community violence exposure upon the
sample population; and 3) examine and specify the pathway of influence
(moderator and/or mediator) of maternal distress symptomatology, specifically,
PTSD, depression, and anxiety, upon child PTSD and behavior problems.
METHODS: Utilizing a random stratified sampling design in which children
were stratified according to gender and race/ethnicity, this study gathered
1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
comprehensive information from children, mothers, and teachers. Multiple
regression equations were conducted to ascertain if maternal distress
symptomatology acts as a moderator or a mediator of child symptomatology or
child behavior problems.
RESULTS: The mother-child pairs have substantial exposure to community
violence. One in three children were victims of violence and 76.2% were
witnesses to violence in their lifetime. Over 58% of the mothers had been a
victim of violence. Analysis reveals that maternal distress symptomatology
acts as a mediator of child symptomatology.
IMPLICATIONS: Findings highlight that maternal distress symptomatology is
more important than community violence exposure in contributing to the child’s
heightened behavioral problems. Schools and mental health practitioners
should assess for parent’s exposure to community violence when conducting
assessments of children and intervene with the parents.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEDICATION
This dissertation is dedicated to
my wife, Grace,
and my children, David, Samuel, and Rebecca,
whose support and patience made this work possible
and whose love is a treasured gift and blessing.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGEMENTS
This dissertation reflects years o f my work and the contribution o f many people through their
mentorship, support, and friendship. I am very grateful to them.
Ferol Mennen, my dissertation advisor, who has been a wonderful mentor and friend;
Penny Trickett, who has broadened my understanding o f neighborhood violence, and
Gretchen Guiton, who has provided valuable feedback and insight.
Kathy Ell and Bruce Jansson, exemplary scholars, who have been foundational in my
conceptualization o f community violence.
The members o f the UCLA Trauma Psychiatry Program, in particular, Bill Saltzman, who
has been a strength of encouragement, and Robert Pynoos, director of the Program.
Aracelis E. Francis, director o f the Council on Social Work Education Minority Fellowship
Program, who remains a beacon to the Fellows.
Rev. Lawrence Shelton, who always believes in me unconditionally, the parish community o f
St. Anselm and Sr. Mary Ann Connell, whose life and heart enabled me to celebrate my life.
Luis Hernandez and Rich Boccia, principals of the selected schools, and their teachers.
The interviewers: Therese Palacio, Cynthia Lemus, Bonnie Sanger, and Annette Saladana.
My mother, Frances, who gave me life and so much more, my sisters, Lolly and Malenie, my
family, Michael Davis, and the Allen, Anderson, and Barrera families.
And also to the National Institute o f Mental Health, which funded the work o f this
Dissertation as a Minority Dissertation Fellowship Grant, I R03 MH60502-01.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
DEDICATION PAGE ii
ACKNOWLEDGEMENT PAGE iii
LIST OF TABLES vi
LIST OF FIGURES vu
I. INTRODUCTION
STATEMENT OF THE PROBLEM 1 -5
PARENT-CHILD INTERDEPENDENT COPING MODEL 5-7
SPECIFIC AIMS 7-8
II. SUMMARY OF THE LITERATURE
INTEGRATED THEORETICAL FRAMEWORK 9-14
METHODOLOGICAL ISSUES 15-21
EPIDEMIOLOGICAL INFORMATION 22-27
PSYCHOLOGICAL AND EMOTIONAL EFFECTS 27-31
COMORBIDITY 31-32
PSYCHOBIOLOGICAL EFFECTS 32-33
PARENTAL DISTRESS 33-36
RESILIENCY/PROTECTIVE FACTORS 36-39
SUMMARY OF LITERATURE REVIEW 39
HYPOTHESES 40
HI. METHODS
SAMPLING PLAN 41 -46
DESIGN 46-47
INSTRUMENTATION 47-59
DATA COLLECTION 59-63
DATA ANALYSIS PLAN 63-65
IV. RESULTS
DESCRIPTION OF SAMPLE
CHILD SELF-REPORT EXPOSURE AND SYMPTOMS
MATERNAL REPORT ON CHILD EXPOSURE
MATERNAL REPORT ON CHILD SYMPTOMATOLOGY
MOTHER SELF-REPORT ON EXPOSURE AND SYMPTOMS
TEACHER REPORT ON CHILD BEHAVIOR
CORRELATIONS
MODERATOR EFFECTS
MEDIATOR EFFECTS
66-72
72-79
79-82
82-84
84-92
92-94
94-98
98-100
100-102
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS continued
V.
VI.
VII.
DISCUSSION
STRENGTHS AND LIMITATIONS
AREAS FOR FUTURE RESEARCH
IMPLICATIONS FOR SOCIAL WORK PRACTICE
REFERENCES
APPENDIX
CORRELATION MATRIX
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF TABLES
Table 1. Summary of Variables and Measures 49-50
Table 2. Descriptives o f Young Adolescent Students 66
Table 3. Descriptives o f Mothers of Middle School Students 68
Table 4. Child Lifetime Exposure to Discrete Violent Events 73
Table 5. PTSD Criteria Scores 77
Table 6. Maternal Report on Child Violence Exposure 81
Table 7. Standardized Regression Coefficients for Measures o f 83
Child Behavior Problems
Table 8. Maternal Lifetime Exposure to Discrete Violent Events 87
Table 9. Measures o f Moderator Effects 99
Table 10. Standardized Regression Coefficients for Measures o f 101
Mediator Effects on Child Behavior Problems based on
Mothers and their reports on Child at Neighborhood School
Table 11. Standardized Regression Coefficients for Measures o f 102
Mediator Effects on Child Behavior Problems based on
Latina Mothers and their Reports on Child
vi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES
Figure 1. Parent-Child Interdependent Coping with Community Violence Model
Figure 2. Exposure to Community Violence Model o f Effects
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 1
INTRODUCTION
Statement of the Problem
Community violence is a pervasive reality in the United States. During the
1990’s, focus escalated on children’s exposure to violence as crime surged and
community violence became a national public health concern (Gloderich, 1998; Koop &
Lundberg, 1992). Recent years have witnessed a national decline in the number of
homicides and serious violent crime such as assaults and robbery. Nevertheless,
violence and violent injuries among children have not decreased (American Academy
of Pediatrics, 1999; U.S. Department of Justice, 1998). Children are twice as likely as
adults to be victims o f serious violent crime and three times as likely to be victims of
simple assault (Sickmund, Snyder & Poe-Yamagata, 1997; Snyder, 1998). In addition,
increasing numbers o f children are repeatedly exposed to violence as observers or as
friends or relatives o f victims. The National Survey of Adolescents found that 23% o f
adolescents reported having been both a victim of assault and a witness to violence
(Kilpatrick, Saunders, Resnick, & Smith, 1993). Other studies have reported that
between 50% and 96% o f urban children have witnessed at least one act of community
violence in their lifetimes (Gorman-Smith & Tolan, 1998; Schwab-Stone et al., 1995;
Singer, Anglin, Song & Lunghofer, 1995).
These statistics underscore that exposure to community violence as a victim or
as a witness is a serious risk factor in the lives o f children. Exposure to community
violence is a chronic and cumulative stressor that has potential effects on children's
1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
physical and emotional health as well as school and social functioning (Jenkins & Bell,
1997; Pynoos, Steinberg & Wraith, 1995). Also, it shapes and influences children’s
sense o f safety, their worldview, and expectations for future happiness (Garbarino et al,
1991; Pynoos, Steinberg & Goenjian, 1996). This is especially true for many low-
income and immigrant children who live in high crime neighborhoods and encounter
violence on a repetitive basis as well as face the threat and fear o f violence in their daily
lives.
Studies o f community violence have revealed positive correlations between the
degree of exposure and levels o f psychological and behavioral effects as reported by
youngsters (Burton, Foy, Bwanausi, Johnson & Moore, 1994; Lynch & Cicchetti, 1998;
Osofsky, Wewers, Hann & Fick, 1993). Children exposed to community violence may
exhibit somatic symptoms and co-morbid depressive and anxiety disorders (Pynoos et
al., 1998; Singer et al., 1995). They also are at risk for posttraumatic stress disorder
(PTSD) or posttraumatic symptomatology (Boney-McCoy & Finkelhor, 1995; Fairbank,
Schlenger, Saigh & Davidson, 1995; Horowitz, Weine & Jekel, 1995; Kliewer, Lepore,
Oskin & Johnson, 1998; Saltzman, Pynoos, Layne, Steinberg & Aisenberg, 2001).
Many violence-exposed children experience difficulties concentrating in the classroom
and show impaired academic achievement. Some engage in aggressive, delinquent and
high risk sexual behaviors and others are at increased risk for substance abuse and
dependence (Farrell & Bruce, 1997; Kilpatrick, Aciemo, Saunders, Resnick & Best,
2000; Parker & Randall, 1996; Saigh, Mroueh, & Bremner, 1997). Also, evidence
suggests that the effects o f traumatic exposure to violence for children extend beyond
2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
current distress and impairment to include long-term disruption of normal development
(Goenjian et. al., 1999; Malinkosky-Rummell & Hansen, 1993; Pynoos, Steinberg, &
Piacentini, 1999; Pynoos et al., 1995).
Despite the known vulnerability of children, serious limitations in our
knowledge prevail. A major limitation is the lack o f consensus in the definition and
operationalization o f what constitutes community violence (Guterman, Cameron &
Staller, 2000). Typically, researchers develop screening measures for their particular
studies or modify existing measures. However, these measures define community
violence in divergent ways. For example, some define community violence to include
hearing gunshots while others foil to distinguish between acts o f violence occurring in
the family home and those in the neighborhood. Also, self-report instruments designed
to measure the frequency and severity o f children’s exposure to community violence do
not screen customarily for the same items of violence exposure. These idiosyncratic
approaches make it very difficult to generalize results across studies.
Most o f the violence measures are used as general screening tools and are not
sufficiently sensitive to capture the domain o f exposure to community violence (Horn &
Trickett, 1998). Only a few o f the measures o f community violence exposure have
psychometrically tested properties (Guterman et al., 2000).
While the risk for exposure to community violence is disportionately higher
among the poor, people o f color, and those who live in densely populated urban areas
(Foy & Goguen, 1998; Hill & Madhere, 1996), neighborhood violence is not restricted
to a specific ethnic group o f people or to inner city communities. Nevertheless,
3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
investigation of the frequency o f exposure to community violence and its impact has
focused primarily on African-American school-age children and adolescents (Bell &
Jenkins, 1993; Hill & Madhere, 1996). Scant research exists on the prevalence and
psychological consequences of exposure to community violence of Latino children
(Aisenberg, 2001; Attar, Guerra & Tolan, 1994; Freeman, Mokros & Poznanski, 1993;
Gorman-Smith & Tolan, 1998). In addition, the experience o f Latino immigrant
children with regards to community violence has not been well-studied (Kataoka et al.,
in review).
“Exposure to community violence is not merely a function of race but is more
likely a factor of socioeconomic status and setting” (Cooley et al., 1995, pg. 1362).
Community level analyses suggest that forms o f disadvantage such as the concentration
of single-parent families, female-headed households, substandard housing, family
concentration and residential stability may constitute potent sources of risk for children
and adolescents (Duncan & Aber, 1997; Gephart, 1997). However, research on
community violence often fails to assess and measure these risk factors when screening
children exposed to neighborhood violence.
To date, nearly all studies that measure exposure to community violence have
relied on self-reports by school-age children and adolescents or a combination of
children’s self-reports and a parent’s report on their child’s exposure. Information on
the parent’s/caregiver’s own exposure to neighborhood violence and their own distress
symptomatology is rarely obtained. Research on the interpersonal context o f children
coping with community violence is particularly sparse (Aisenberg & Ell, in review). As
4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
a result, systematic research has not been conducted on the effects of recurrent
community violence on the caregiving environment (Osofsky & Scheeringa, 1997).
The absence o f information on parental exposure greatly contributes to a significant gap
in our knowledge, namely, what are the associations between exposure to community
violence, parental distress, and adjustment in children. Scant attention has been given
to developing theoretical models that could account for such associations (Cicchetti &
Lynch, 1993; Kliewer et al., 1998).
The study o f the interaction o f trauma and parental psychopathology among
children exposed to community violence is therefore particularly crucial (Osofsky &
Scheeringa, 1997). The investigator has developed a conceptual model o f Parent-Child
Interdependent Coping with Community Violence (Aisenberg & EU, in review).
FIGURE 1. Parent — Child Interdependent Coping with Community Violence Model
Exposure to Community Violence
Coping and
Functioning
Child ^ Parent ■+ Social
Support
Internalizing Symptoms
/ I \
Anxiety Depression PTSD
Social
Support
Externalizing Internalizing Symptoms
Behavior Problems . . .
/ ▼ \
Anxiety Depression PTSD
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
While recognizing the importance of individual treatment of the child and/or
intervention with the child, this model focuses on the interaction between parent and
child in their coping responses to community violence. According to this model,
community violence directly and negatively impacts the parent as well as the child.
Experienced on a repetitive basis, community violence taxes a parent's resources,
creates stress, and may impair a parent's ability to be emotionally responsive and to
folfill parental tasks and responsibilities effectively (Pianos, Zayas & Busch-Rossnagel,
1997). Also, it affects the parent’s and child’s sources of social support such as peers,
families, churches, schools, and neighborhood institutions. The model maintains that
variations in neighborhood characteristics, such as the percent of the population living
in poverty, residential stability, and percent o f the population working, affect children’s
risk (Duncan & Aber, 1997; Sampson, Raudenbush, & Earls, 1997).
The model posits that the coping and functioning of the parent and child are
interwoven and exert a profound influence upon each other. It maintains that
community violence has indirect effects that contribute to negative child and parent
outcomes as a result o f a bi-directional relationship between parent and child coping
and functioning. The model purports that intervening to increase the social support of
the parent and/or intervening directly with the parent will result in improved
psychological, emotional, and behavioral outcomes for the child. Although this model
is not tested in this study, it serves as a useful conceptual lens and framework for the
research.
Informed by this model, the following model will be tested in the present study.
6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
FIGURE 2. EXPOSURE TO COMMUNITY VIOLENCE MODEL OF EFFECTS
Witnessing
Exposure to Community Violent
Mother Child
I
PTSD
Depression
Anxiety
Behavior Problems PTSD
Depression
Anxiety
Specific Aims
Informed by the model of Parent-Child Interdependent Coping, this present
study examines the behavioral, emotional, and psychological effects o f witnessing
community violence upon a community sample o f 80 Latino and African-American
mothers and their children attending two urban public middle schools in a large, diverse
city o f Los Angeles County. Utilizing a random stratified sampling design in which 6th ,
7th , and 8th grade students were stratified based on race/ethnicity, gender, and level of
violence exposure, the study has six specific aims. It seeks to:
I) identify and measure the type and frequency o f lifetime exposure to discrete
community violence events experienced by male and female African-American and
Latino students and compare differences based on students’ self-reports and
mothers’ reports on their child’s exposure;
2) assess and measure child distress symptomatology, namely PTSD, depression, and
anxiety, based on self-report;
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3) assess and measure child behavior problems based on reports from mothers and
teachers;
4) describe and identify lifetime maternal exposure to community violence and other
forms o f violence such as sexual abuse and domestic violence based on self-reports;
5) assess and measure maternal distress symptomatology, specifically PTSD,
depression, and anxiety, based on self-report; and
6) examine, identify, and compare the possible moderating and/or mediating effects of
mothers’ distress symptomatology upon children’s PTSD or behavior problems.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 2
SU M M A RY O F LITERATU RE
This chapter presents an overview of the theoretical basis for the present
investigation and highlights the conceptual model that serves as a framework for the
study. It summarizes the literature regarding salient methodological and measurement
issues operative in the study o f community violence that limit the advancement o f the
field and outlines how the study will address these issues. It then highlights the
prevalence o f community violence and posttraumatic stress disorder and depression
among children and adults in the general population and summarizes what is known of
the role and impact of gender on differential rates o f exposure and symptomatology.
The chapter summarizes the findings on the psychological and emotional effects of
exposure to community violence on children and their mothers and concludes with
identifying documented factors that contribute to the resiliency o f the child and family.
An Integrated Theoretical Framework
Four overarching conceptual frameworks illustrate the context and processes in
which children and their families cope with the daily stress of community violence:
stress theories, social ecology theory, social support and coping theory, and
interdependent coping theory. These theories serve as the theoretical basis for the
present investigation.
Stress Theories
Aneshensel (1992) defines stress as a "state of arousal resulting either from the
presence o f socio-environmental demands that tax the ordinary adaptive capacity o f the
individual or from the absence o f the means to attain sought-after ends" (pg. 19).
9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Ecological, sociological, psychological, and biological stress theories
underscore the risk of negative effects of stress on child and adult well-being (Lazarus,
1966; Pearlin, 1989; Selye, 1976; Weiner, 1994). In particular, exposure to prolonged
stress can overtax the system o f the body and debilitate the immune system, leading to
health problems and /or mental health disorders.
Congruent with family stress theory (Boss, 1987) and family systems theory
(Steinglass, 1987), research suggests that if one or more family members experience
distress symptoms, the family as a unit will also be affected. The family may experience
changes in expression o f affect, organization, daily interaction, and beliefs (Wilson &
Kurtz, 1997).
Researchers continue to examine interactional processes among biological,
psychological and social factors that characterize the experience of stress (Mazure,
1995). Recent inquiry has revealed the key roles o f the hypothalamic-pituitary-adrenal
axis (HPA) and the corticotropin-releasing factor (CRF) in stress response and
adaptation (Miller, 1995).
Social Ecology
Bronfenbrenner (1979) posits that a child is embedded in an interconnected
system o f family, peer, school, and community. This complex system exerts dynamic
and reciprocal influences on the behavior of family members and influences the coping
and outcome o f the child. Not only does the environment directly affect a person’s
behaviors but the individual’s perception o f his or her environment is seen as crucial
for predicting those behaviors.
It is in and through this context that the child and family experience community
violence and struggle to cope with it. Intrinsic child factors that affect the youngster's
10
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
coping with exposure to community violence include the child's temperament, intelligence,
developmental stage, competencies, preexisting psychopathology, and previous
experiences. External factors that influence a child's coping are the parent’s characteristics,
their responsiveness to the child, the neighborhood, and school (Hoagwood & Jensen,
1996). Factors that impact the parents’ responsiveness to their child as well as their child's
reaction to the traumatic stress o f witnessing community violence include the parents' own
grief/stress reactions to the event, their previous exposure to trauma/loss, their preexisting
psychopathology, such as depression, their warmth and attachment to the child, and
secondary stresses and stress reminders (Pynoos et al., 1995).
Cicchetti and Lynch (1993) assert that a transactional ecological framework is
important for understanding the behavioral, psychological, social, and developmental
sequelae for children and adolescents exposed to community violence. The transactional
model maintains that interrelations between biological, emotional, cognitive domains as
well as interpersonal and environmental factors, including familial, societal, community,
and cultural, influence children’s development (Cicchetti & Toth, 1997). This model
allows for an examination of the way that multiple systems of a child’s environment
(e.g., individual child, parent, family interaction, neighborhood) influence each other,
and in turn, influence developmental outcomes for children (Myers, 1997). This
framework suggests that exposure to community violence may lead directly to
maladaptive outcomes for children, or indirectly contribute to negative child outcomes
by impairing the parents’ ability to cope with community violence and safeguard their
children from the negative effects o f such exposure.
1 1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Cognitive Appraisal. Coping, and Social Support
The theory o f cognitive appraisal (Lazarus, 1986; Lazarus & Cohen, 1977) helps
inform the investigation o f neighborhood violence. Lazarus and his colleagues (1977,
1984) posit that the impact o f a stressor is determined in part by the individual’s
perception of that event. Perception and coping mechanisms play a significant role in
determining how a person experiences, responds to and reports acts as violent and
stressful (Guterman et al., 2000).
Social support and coping theories draw attention to the day-to-day interactional
processes engaged by children and their families. Social support serves to buffer a
person from the adverse effects of stressful events or experiences in two principal ways.
First, social support can intervene between a stressful event and an individual stress
reaction by attenuating or preventing a stress appraisal response. Second, social support
may intervene between the experience of stress and the onset o f a pathological response
by eliminating the stress reaction or by influencing physiological processes (Cohen &
Willis, 1985). In addition, social support can sustain and strengthen an individual and
family and thus may help one to recover more readily from the impact o f a stressful life
event (Stroebe et al., 1996).
Research findings substantiate that increasing social support helps protect the
child and promotes positive adaptation and coping (Bowen & Chapman, 1996; Werner,
1992). The buffering qualities o f support provided by a parent not only affect the
child’s cognitive appraisal and interpretation of the stressor o f community violence but
also increase the child’s knowledge o f coping strategies as well as the child’s self
esteem (Robinson & Garber, 1995).
1 2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Coping theory maintains that the meaning of a threat or traumatic loss can lead
to a major shift in an individual's perceptual sensitivities (van der Kolk, 1989). What
other individuals in the situation are doing and how they are coping profoundly
influence how an individual copes with a stressor or problem (Coyne & Smith, 1991).
Interdependent Coping— Parent-Child Interaction
Following a developmental perspective view of coping, interdependent coping
theory recognizes that a child is embedded in an interconnected system o f family, peers,
school, and community. It posits that coping and adaptation are multi-determined and
interactional processes in which parental behavior and coping response affect children
in an interdependent and reciprocal manner (Walsh, 1996). First, parental behavior
affects the child's immediate emotional reactivity in a given situation. Second, it may
affect the child's beliefs and cognitions about the world and the parent. In turn, this may
affect the way the child responds to the parent's behavior. For example, community
violence may undermine a youngster's sense of security and trust in their parent's ability
to protect them (James, 1994). Third, parental behavior may exert a direct effect on the
coping strategies that the child utilizes (Calkins, 1994). The buffering qualities o f
support provided by the parent not only affects the child's interpretation o f the stressor
but also enhances the child's knowledge of coping strategies as well as the child's self
esteem (Robinson & Garber, 1995).
Congruent with a family systems perspective o f coping, interdependent coping
takes into account the mediating influence of family processes in surmounting crisis or
chronic hardship (Walsh, 1996). How a family responds to and manages a disruptive
experience such as witnessing a drive-by shooting, how it attempts to buffer the
13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
resulting stress, how it seeks to effectively reorganize and endow the experience with
meaning influences immediate and long-term adaptation for the family members and for
the family unit (Antonovsky & Sourani, 1988; Rolland, 1994). Parents who perceive
themselves as being able to cope successfully with danger are likely to be better able to
convey a sense o f security and confidence to their children (Baker, 1990). Children
whose parents provide a good model of coping are themselves likely to enjoy higher
self-esteem and greater confidence in their own ability to cope (Garbarino et al., 1992).
Research further highlights the importance of interdependent coping. According
to studies of the mother-child relationship during war, "the level o f emotional upset
displayed by the child's parents, not the war situation itself, was the most important
factor in predicting the child's response to the war" (Garbarino et. al., 1992, pg. 54).
Researchers in Israel found that during air raids children reported greater anxiety when
their parents over-reacted, appeared unable to respond competently, or appeared to
disagree over the appropriate action to take (Bat-Zion & Levy-Shiff, 1995). The
parents' coping response directly influences the coping response o f the child and is a
significant and crucial mediator o f the child's distress (Pynoos et al., 1995).
Research in related areas supports the relevance of examining a mother’s
exposure and distress symptomatology as well as her child’s exposure and distress
reactions in the present study. Research in areas such as childhood cancer, abuse, and
war-like experiences suggests that if one or more family members experience distress
symptoms, the family members and the family as a unit will also be affected (Ell, 1996;
Garbarino et al., 1992).
14
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Methodological issues
Definition
Studies examining children’s exposure to community violence are not uniform
in their definition and operationalization of what constitutes community violence
(Guterman et al., 2000). For example, Osofsky (1995) defined exposure to com m unity
violence as “... frequent and continual exposure to the use o f guns, knives, and drugs,
and random violence” (pg. 782). Some researchers classify exposure to community
violence into the categories o f direct personal experience (vicitimization), exposure
through witnessing violence (saw it happen to someone else), and vicarious experience
(knowing someone who was harmed by violence) (Foy & Goguen, 1998). Other
researchers have classified exposure to community violence into the two broad
categories o f witnessing violence and being a victim (Horowitz et al., 1995; Kliewer et
al., 1998). Some studies include hearing of violence through stories told by family
members, peers, and neighbors as a category of community violence (Horn & Trickett,
1998) while another study distinguishes between assaultive violence, criminal violence,
and weapon related violence (Stein et al., 2001). The reliability and validity o f such
categories has not been conclusively demonstrated (Overstreet, 2000).
Self-report instruments are also divergent. For example, some instruments foil
to include questions regarding exposure to sexual abuse and domestic violence. These
important questions are often excluded due to the reticence o f school administrators and
parents to give permission to assess these sensitive areas (Cooley et al., 1995). Other
researchers operationalize community violence by including exposure to gang members,
bullying, drugs, or car chases (Guterman et al., 2000). As a result of the lack of
15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
consistency in defining and operationalizing community violence as well as the use of
divergent self-report instruments to assess children’s exposure to community violence,
it remains difficult to draw firm conclusions from the current literature and to generalize
results across studies.
Evidence o f a differential impact o f being a witness or a victim o f community
violence is inconclusive (Fitzpatrick & Boldizar, 1993; Horowitz et al., 1995; Kliewer
et al., 1998). Some researchers employing factor analysis found no support for the
distinction between witnessing and victimization (Hastings & Kelly, 1997; Singer et al.,
1995). Rather, they identified two categories of experiences: those that are life
threatening and those that are not. However, there exists little evidence that these
categories lead to unique developmental outcomes (Overstreet, 2000).
The lack o f consistent differences between witnesses and victims o f community
violence may be the result o f several factors (Mazza & Overstreet, 2000). First, the
exposure to a specific violent event may be perceived differently by two individuals and
thus impact them in divergent ways. Second, most measures o f exposure to community
violence foil to assess the child’s proximity to the event or the relationship of the child
to the victim or perpetrator (Guterman et al., 2000). These are two principal factors
associated with the severity o f response to a traumatic event such as community
violence (Zeanah & Scheeringa, 1996).
For purposes o f this study, community violence refers to violence that occurs in
the child's environment, i.e. neighborhood, school, or outside the neighborhood, but
outside the home. Such violence is distinct from domestic violence which involves
family members and usually occurs in the family home (Wallen & Rubin, 1997).
16
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Community violence involves direct exposure through victimization and also through
witnessing o f violence. It does not include hearing from others about violent acts. It
includes interpersonal conflicts and events where force, threats, or other means are used
with the intent o f causing physical harm, injury, or death to another person (Rosenberg
& Fenley, 1992).
For purposes o f this study, victimization is understood as harm that occurs to
individuals because other individuals behaved in ways that violated social norms
(Finkelhor & Kendall-Tackett, 1997). This understanding distinguishes exposure to
community violence from other stressors, such as natural disasters or chronic illness
(Margolin & Gordis, 2000).
Self-Report Instruments
Two principal limitations exist with regards the use o f self-report instruments to
assess and measure community violence. First, the lack of uniformity among self-report
instruments designed to assess the frequency and severity of children's exposure to
community violence makes it very difficult to generalize results and to advance the field
(Cooley et al., 1995). Second, most o f the instruments lack known psychometric
properties. For example, the Richters and Saltzman (1990) instrument is widely used
and often cited in the literature. However, no psychometric properties have been
published (Saltzman, personal communication, January 2000). Several studies have
reported recent attempts to address this limitation and make psychometric advances in
this area (Cooley, 1995; Hastings & Kelley, 1997; Selner-O’Hagan et al., 1998).
However, progress remains in its early stages.
17
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Four additional limitations are noteworthy: a) these instruments often do not
adequately distinguish between violence in the community and violence in the home;
b) most fail to take into account or specify the child's relationship to the victim or
perpetrator; c) the impact o f the event upon the child from the perspective o f the child is
not usually examined by the instruments currently used, and d) most measures are
additive in nature, that is, they tally the number o f distinct kinds of events and
frequency to which the child reported being exposed. This rudimentary method assigns
the same weight to witnessing a murder as to witnessing a fistfight in the schoolyard. It
is doubtful that these two heterogeneous events are comparable in their severity or
effects (Horn & Trickett, 1998).
In using the violence exposure screening instruments developed for this study
and patterned after the Richters and Saltzman instrument (1990), this study attempts to
address some o f these limitations by obtaining measures o f three forms o f exposure:
direct victimization, witnessing violence, and total exposure (victimization and
witnessing). It ascertains the child’s familiarity with the victim and/or perpetrator such
as parent, sibling, other relative, friend, schoolmate or stranger. Also, the study helps
clarify the severity o f events by assessing the personal impact o f the violent experience
from the subjects’ own perspective in asking what the child and mother deem to be the
most upsetting event in their lives. This qualitative information helps inform the
research regarding the individual’s perception o f the seriousness and severity of discrete
events.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Divergent Reports
The vast majority o f studies o f community violence have relied on retrospective
child-self report and/or parent report. Such an approach prohibits a comprehensive and
reliable assessment o f emotional and behavioral outcomes. Also, some researchers
argue that exclusive use of retrospective child-report raises the possibility o f shared
method variance (Kliewer et al., 1998; Lynch & Cichetti, 1998). Recall bias is another
inherent limitation. Retrospective instruments administered on a one-time basis are
subject to recall biases that reduce accuracy (Eckenrode & Bolger, 1995; Hilton, Harris,
& Rice, 1998).
Researchers document that parent and child reports o f child exposure to
neighborhood violence are not in agreement. Findings reveal that parents consistently
report lower levels o f child exposure than their child does (Cooley, Turner, & Beidel,
1995; Farver, Natera, & Frosch, 1999). In their study o f 9 to 12 year-old children, Hill
and Jones (1997) found that parents significantly underestimated their child’s exposure
to community violence. Poor concordance between child and parent reports of
children’s violence exposure may be due to several factors. 1) parents may not be aware
o f the extent o f their child’s exposure, since much of the violence that children
experience occurs near or at school; 2) parents may repress violence exposure
information as either a passive or active coping strategy; or 3) parents and children may
construe what constitutes an act o f violence in different ways or perceive the threat o f
the event differently (Goldstein & Conoley, 1997; Richters & Martinez, 1993; Stephens,
1997).
19
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rarely have the perceptions of children and parents regarding the severity of
violent events been investigated. Nevertheless, perception is a critical factor in
determining threat as well as the meaning they attach to the event (Hill & Madhere,
1996).
Significant divergence between parent and child reports o f child symptoms has
been reported (Kendall-Tackett et al., 1993). Most studies have found only low to
moderate correlations between informant reports (Achenbach et al., 1987; Angold et al.,
1987). Lower parent-child agreement is also found for internalizing symptoms than for
behavior problems or externalizing symptoms (Achenbach et al., 1987). While parents
are generally good evaluators o f their child’s externalizing behaviors they tend to be
less accurate in assessing their internalizing symptoms (Costello & Angold, 1988,
Applebaum & Bums, 1991). This is likely due to the fact that parents'judgments about
their child's symptoms tend to be highly related to their own level o f distress (Everson
et al., 1989; Newberger et al., 1993). Also, the absence o f strong correlations between
informants suggests that these different measures tap sources of variances that are not in
common (Fergusson & Horwood, 1987, 1989).
A major measurement issue is whether the information gathered from different
informants should be combined or treated separately (Kuo, Mohler, Raudenbush &
Earls, 2000). Typically, analysis employs separate regression analysis for each
informant or averages the scores o f exposure to community violence from parent and
child reports and then performs a single regression analysis. In the present study
separate regression analysis for each informant will be carried out.
20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Study o f Diverse Racial/Ethnic Groups
Scant research examines the prevalence and psychological consequences of
exposure to community violence among Latino children. Previous research that
included Latino children failed to report an analysis o f ethnic differences (Freeman et
al., 1993; Springer & Padgett, 2000) or found no differences associated with ethnicity
(Attar et al., 1994; Pynoos et al., 1987). However, these studies relied on children's
self-reports and were limited in other ways as well. For example, Attar's research did
not look at childhood PTSD while Pynoos' research focused on a single violent incident.
Recent studies have included Latino children in their samples. For example,
Singer and colleagues (1995) and the Project on Human Development in Chicago
Neighborhood Study (Sheidow, Gorman-Smith, Tolan & Henry, 2001) included Latino
youngsters in their large samples. However, no findings have yet been reported by
these studies regarding ethnic differences. Most recently, Kataoka and colleagues (in
review) found that 31% o f Latino immigrant children had been exposed to community
violence and manifested PTSD symptomatology or depressive symptoms. Aisenberg
(2001) reported that Latina mothers and their preschool children experienced high
levels of exposure to community violence and distress symptomatology (PTSD,
depression, and anxiety). Seventy-seven percent of the mothers and eighty one percent
of the preschool children had been a victim or a witness o f community violence in their
lifetime. Thirty percent o f the preschoolers scored at or above the clinical borderline
range for total behavior problems. Twenty-three percent o f the mothers scored in the
clinical range for depression and sixteen percent reported moderate levels of PTSD.
21
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Epidemiological Information
Poverty
In 1999 the nation’s poverty rate was 11.8% while California had a poverty rate
of 13.8% (US Census Bureau, 2000). In Los Angeles County, the percentage o f the
population estimated to be poor was even higher, 16.3%, for the same year. The
established federal poverty line was $16,700 for a family o f four (United Way o f
Greater Los Angeles, 1999). Overall, according to U.S. Census 2000 statistics, 22.8%
o f Hispanics and 23.6% of African-Americans in the United States live below the
nation’s poverty line.
The percentages are higher for children. One o f every three Latino immigrant
children residing in the United States live below the poverty line (35% o f children from
Mexico and 27-30% o f children from Central American countries) (Hernandez &
Darke, 1999). This statistic carries much significance in Los Angeles County, which is
home to increasing numbers o f Latino immigrants.
Exposure to community violence
Epidemiological data on the national prevalence and incidence of children's
exposure to community violence in the United States is lacking (Amaya-Jackson &
March, 1995; Phillips & Cabrera, 1996). Due to the divergent methods of data
collection and lack o f uniformity in defining community violence, statistics estimating
the rates and types o f violence exposure vary widely. However, information on certain
cities is available. In 1996, there were 1,140 firearm-related homicides in Los Angeles
County (Los Angeles County Coroner, 1996). Researchers estimate that children
22
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
witness 20% of the homicides committed in Los Angeles (Groves, Zuckerman, Marans
& Cohen, 1993). In studies o f youth in Chicago, 45% reported having witnessed more
than one violent event (Bell & Jenkins 1993) and 30% had seen three or more such
events (Gorman-Smith & Tolan 1998). Taylor, Zuckerman, Harik, and Groves (1994)
found 10% o f the children ages one to five attending a pediatric primary care clinic in
Boston had witnessed a shooting or stabbing and 47% had heard gunshots. Richters and
Martinez (1993) found that 61% o f the first and second grade students and 72% of the
fifth and sixth grade students o f an elementary school in Washington, DC reported
witnessing at least one act o f community violence in their lifetime. In their study of 6-
10-year old boys in New York city, Miller and colleagues (1999) found that 35%
reported witnessing a stabbing, 33% had seen someone get shot, and 25% had witnessed
someone get killed. In a sample o f 1000 African-American elementary and high school
students, Shakoor and Chalmers (1991) found that nearly three in four students reported
having witnessed at least one instance of someone being robbed, stabbed, shot, or
killed. Nearly half claimed to have been victimized by such events.
Exposure Gender Differences
Divergent findings exist pertaining to exposure to community violence and
gender. While definitive patterns are not yet evident, more studies report insignificant
results for the demographic variables o f age, gender, and ethnicity as potential risk
factors for exposure to community violence (Foy & Goguen, 1998). Cooley and her
colleagues (1995) reported no significant gender differences while other studies have
found that males reported more exposure to community violence once sexual assaults
were excluded (Fitzpatrick & Boldizar, 1993). Boney-McCoy & Finkelhor (1995)
23
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
reported significant gender differences in their national sample of adolescents with 35%
o f the males reporting to have been a victim of violence as compared to only 13% of
females.
Posttraumatic Stress Disorder
The National Survey o f Adolescents found that over 20% of the youth who
reported exposure to community violence met criteria for PTSD (Kessler et al., 1995).
Higher percentages have been consistently found in smaller samples. Saltzman and
colleagues (in press) reported that 25% of middle school students screened positively
for PTSD symptomatology. Breslau and colleagues (1991) found that life threat,
witnessing the injury or death o f another, as well as physical assault, produced lifetime
PTSD rates of approximately 25% among young adults living in the Detroit area.
Overall, they estimated the lifetime rate of PTSD in the general population o f young
adults to be 9%. Richters and Martinez (1993), in their study o f community violence,
estimated that 40% o f children and adolescents manifested PTSD symptomatology. In
their study of non-referred urban youths exposed to community violence, Berman and
colleagues (1996) found that 34.5% met full criteria for PTSD. Children under age 11
who experience a life-threatening event are three times more likely to develop PTSD
(Davidson & Smith, 1990). Saigh and colleagues (1996) reported that the prevalence of
PTSD following the occurrence o f sexual abuse is fairly constant among children, with
approximate rates o f one-third to one-half manifesting PTSD symptomatology.
The National Comorbidity Study estimated lifetime prevalence of PTSD in
adults to be 7.8% (Kessler et al., 1995). Prevalence is higher among women (10.5%).
24
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Kilpatrick and Resnick (1993) reported that 39% of women who had experienced
aggravated assault and 35% who were raped developed PTSD.
Child PTSD and Gender Differences
Findings with respect to gender differences in child PTSD severity are
inconclusive (Foy et al., 1996). Berman and colleagues (1996) did not find gender
differences. In some studies, girls reported higher distress scores only in cases of
natural disaster or accidents (Curie & Williams, 1996; Vemberg, LaGreca, Silverman &
Prinstein, 1996). However, in those studies that report significant differences, girls had
higher distress scores (Foy et al., 1996; Yule & Canterbury, 1994). Using the Impact of
Events Scale (Horowitz, 1984), Springer and Padgett (2000) found that 58.9% o f the
females reported severe PTSD symptomatology, compared to 44.2% o f males among a
school-based, nonclinical sample o f young adolescents.
Adult and Child Depression
Rates o f clinical depression in women between 25 and 44 years o f age are
generally estimated between 8 to 10% (Kessler et al., 1996). Epidemiological data
identifies variable risk for depression within the general population by gender, age,
racial/ethnic group, and socioeconomic status (Reynolds, 1994). A consistent finding in
the literature is that women have higher rates of depression (Klerman & Weissman,
1989; Kessler et al., 1994). Also, rates o f major depressive disorder (MDD) have been
found to be significantly higher among people meeting federal poverty criteria (Bruce,
Takeuchi & Leaf 1991).
While not definitive, current research indicates that Latinos may be at higher
risk for disorders such as depression and anxiety (Organista, 2000). In the National
25
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Comorbidity Survey (NCS) (Kessler et al., 1994), Latinos had significantly higher
prevalence o f current affective disorders compared with non-Latino Whites and
African-Americans. On the other hand, other studies reveal that 6-10% o f Mexican-
Americans in California suffer depression, a rate consistent with that of the general
population (Vega, Kolody, Aguilar-Gaxiola, Alderete, Catalano, & Caraveo-Anduaga,
1998). Adebimpe (1994) found that African-Americans and Caucasians did not differ
in prevalence o f affective disorder when rates were corrected for age, gender, SES, and
marital status.
The rates o f depression in youngsters increase significantly as children move
into adolescence. A number of epidemiological studies have reported that from 2.5 to
4% o f children and between 4.5 and 8.3% of adolescents in the United States suffer
from depression (Birmaher, Ryan, Williamson et al., 1996). A study sponsored by the
National Institute o f Mental Health estimates that the prevalence o f any depression
among children ranging from 9 to 17 years in age is more than 6 % in a six month
period, with 4.9 % having major depression (Schaffer, Fisher, Dulkan et al., 1996). It is
estimated that twenty percent o f adolescents will have experienced at least one episode
o f depression by the time they reach their 18 th birthday (Compass, Conner & Hinden,
1998). Earlier studies o f elementary school children found prevalence rates of
depression ranging from 5.4 to 6.3% when using self-report or interview instruments
(Poznanski et al., 1984; Strauss et al., 1984).
Gender Differences for Child Depression
In his sample o f African-American children, Fitzpatrick (1993) found that
gender moderated the impact o f exposure to violence on depressive symptoms. Girls
26
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
were more likely to report increased depressive symptoms. The literature consistently
reports that adolescent girls are more likely to report increased depressive symptoms
than adolescent boys by a 2 to 1 ratio (Compass et al., 1998). These findings are
consistent with data suggesting that the development and prevalence of depression is
greater in females and adolescents than in males and elementary school-age children
(Reynolds & Johnston, 1994).
Psychological and Emotional Effects
Children exposed to community violence may be particularly vulnerable to
emotional and developmental problems which affect their ability to succeed in school
(Gaensbauer, 1996). Kliewer and colleagues (1998) and Gorman-Smith and Tolan
(1998) gathered data from multiple informants on child’s exposure and symptomatology
and found that exposure to community violence contributed significantly to predicted
child symptomatology. In general, most studies report a positive relationship between
exposure to community violence and depression (Farrell & Bruce, 1997; Freeman et al.,
1993; Lynch & Cicchetti, 1998; Overstreet et al., 1999). In their study of inner-city
adolescent boys, Gorman-Smith and Tolan (1998) found that violence exposure was
significantly related to anxiety and depression. Schwab-Stone and colleagues (1995)
found that exposure to violence predicted depressed/anxious mood among 6th , 8^, and
lO1 * 1 graders. However, these studies did not distinguish between violence occurring in
the home versus in the community. In their sample o f 3,735 high school students,
Singer and colleagues (1995) reported that exposure to community violence accounted
for 7% o f the variance in depressive symptoms, even after controlling for demographic
variables such as age and family structure.
27
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Children exposed to violence may withdraw, or manifest aggressive or
disruptive behavior (Keane, 1996). Also, they may manifest eating disturbances,
speech and language delays, or suicidal ideation (Speier, Sherak, Hirsch & Cantwell,
1995). Younger adolescents in an urban school district who were exposed to
community violence were found to report more internalizing symptoms than older
adolescents (Schwab-Stone et al., 1999).
Experiencing one's environment as threatening is also associated with negative
mental health outcomes among youngsters (Aneshensel & Sucoff, 1996). Research has
shown daily stressors to play a more central role in the development and maintenance o f
psychological problems compared with major life events (Berton & Stabb, 1996).
Exposure to community violence on a repetitive basis may also desensitize
children and adolescents to the dangers and harms of violence. Also, their hearts may
become hardened and view violence as part of their “normal” landscape. The stress
associated with community violence also contributes to a sense o f hopelessness among
children and adolescents. Related to this sense o f hopelessness is a shortened sense of
the future. Many youngsters who have been reared in an environment of chronic
violence have difficulty envisioning themselves in meaningful future roles (Wallach,
1993).
The cumulative effects o f exposure to chronic community violence not only
have negative behavioral or psychological effects but also have detrimental effects on
children’s academic achievements. Children who perform poorly in school may be
suffering from intrusive thoughts and impairment in concentration that accompany
PTSD (Ell & Aisenberg, 1998).
28
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Posttraumatic Stress Disorder
A PTSD diagnosis for children was first specified in the DSM-III-R (APA,
1987). Early research initially examined PTSD in children exposed to single trauma
events, such as a school sniper attack (Pynoos et al., 1987), school shooting (Schwartz
& Kowalski, 1991), or a school bus kidnapping (Terr, 1981). PTSD manifested in
children exposed to large-scale natural disasters such as Hurricanes Andrew and Hugo
has also been examined (Goenjian et al., 1996; Greca et al., 1996; Lonigan et al., 1994;
Shannon et al., 1994; Shaw, Applegate, & Schorr, 1996; Vemberg et al., 1996). In
addition, children living in war zones who exhibit PTSD have been studied (Nader et
al., 1993; Sack, Clarke, & Seeley, 1995; Hubbard, Realmuto, Northwood, & Masten,
1995; Schwarzwald, Weisenberg, Solomon, & Waysman, 1994). Other research has
shown clear associations between exposure to violence in the form o f child abuse and
domestic violence and PTSD symptoms even in infants and toddlers (Drell et al., 1993;
Famularo, Fenton, Kinscherff, Ayoub, & Bamum, 1994; Saigh et al., 1996).
In recent years, researchers have documented a strong and consistent
relationship between exposure to community violence and PTSD (Boney-McCoy &
Finkelhor, 1995; Horowitz et al., 1995; Kliewer et al., 1998; Saltzman et al., in press).
However, rarely do studies control for concurrent life stressors in the lives of
youngsters. Most o f the samples are drawn from poor, high crime neighborhoods.
Families experiencing the stressor o f chronic community violence are likely to
experience many other life stressors as well such as family financial concerns and
problems, divorce, and overcrowding (Kliewer et al., 1998). These other stressors
29
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
could cause a spurious correlation between violence exposure and psychological
adjustment problems (Kliewer et al., 1998).
An important consideration is that many studies fail to distinguish between
PTSD symptomatology and clinical PTSD. Researchers often report that the sample
manifests heightened levels o f PTSD symptomatology but fail to report if the sample
meets the full diagnostic criteria for PTSD according to DSM-IV.
There is less consistency in methods used in research examining etiologic
factors in child PTSD (Foy et al., 1996). Whereas early work on PTSD in adults
developed around the study o f reactions of men to combat and women to rape, the lack
o f a trauma type for children contributes to the lack o f uniformity in the study o f PTSD
in children. What is known about PTSD etiology in children exposed to violence is that
they can develop PTSD either through direct exposure (e.g. being shot), through
observation (e.g. o f domestic violence), or through verbal transmission o f traumatic
information (e.g. being told about a friend shot in a drive-by incident) (Saigh, 1991).
Although PTSD symptomatology in children resembles PTSD in adults, children
differ in their reactions to traumatic stress (Pynoos & Nader, 1993). Child PTSD is
characterized by persistent re-experiencing of the trauma in various ways; recurring
unpleasant or frighting nightmares, sudden feelings and actions associated with a belief
that the stressfid event is still ongoing; intense psychological stress when exposed to
situations similar to those in which the trauma occurred; persistent symptoms of
increased psychological arousal, including difficulty felling asleep, incontinence, poor
concentration, irritability, exaggerated startled responses, and increase physiological
reactivity when exposed to stimuli reminiscent o f the traumatizing event, aggressive
30
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
behavior problems, moodiness, feelings of guilt; avoidance, and psychological numbing
o f responsiveness (Parker & Randall, 1996; Pynoos & Nader, 1993; Zeanah &
Scheeringa, 1996). Children with PTSD often show diminished interest in previously
enjoyed activities and may also show evidence o f restricted affect and accompanying
feelings of detachment or estrangement from others (Amaya-Jackson, 1995).
Regression in developmental achievements also is common, including regression to
thumb sucking or enuresis (Drell et al., 1993).
Debate has recently centered on whether the diagnostic criteria can be applied
adequately and appropriately to children and adolescents (Sauter & Franklin, 1998). As
March (1999) points out, the DSM-IV framework underemphasizes developmental
differences and social contextual factors. Thus, it may be less than adequate for
assessing childhood PTSD. No clear consensus has emerged. In addition, little is
known about cultural influences on the assessment and manifestation of PTSD (Allen,
1994).
Comorbidity
Children manifesting PTSD symptoms frequently present with other symptoms.
Moreover, many o f the symptoms of PTSD comprise the diagnostic criteria o f major
depressive or anxiety disorders (Aciemo, Kilpatrick & Resnick, 1999). Depressive
symptomatology is the most common co-occurrence (Amaya-Jackson & March, 1995)
with up to 80% of children with PTSD symptomatology also meeting criteria for
depressive symptoms (Pynoos, 1998). Also, PTSD can mimic a disruptive behavior
disorder (Osofsky, 1995). Comorbidity between PTSD and other anxiety disorders has
also been noted, including ADHD (Cuflfe et al., 1994; Glod & Teicher, 1996).
31
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Among adults, comorbidity o f PTSD with other psychiatric disorders is
common. Kessler and colleagues (1995) reported that nearly 48% o f the men and
women with PTSD had a lifetime history of major depression.
Epidemiological studies indicate that between 15 and 38 percent o f nonreferred
children with anxiety disorder meet criteria for at least one additional anxiety disorder
(Kashani & Orvaschel, 1988; McGee et al., 1990). Estimates o f prevalence of
depressive disorders in anxious children range considerably from 7 to 69 percent (Bird,
Gould, & Staghezza, 1993; McGee, Feehan, Williams et al., 1990; Kashani et al., 1987).
Besides true comorbidity, PTSD symptoms are often confounded by spurious
comorbidity resulting from overlap between criteria. As in traumatized adults,
depression and anxiety are also prevalent among traumatized youngsters (Yule, 1991).
The findings of Goenjian and colleagues (1995) indicate an interactive relationship
among child PTSD symptoms, depressive symptoms, and separation anxiety disorder.
Psvchobiological Effects
The day-to-day stress o f exposure to community violence may contribute to the
development o f headaches, asthma, allergies, and ulcers in youngsters. Stressful life
events, such as witnessing community violence, can also alter a wide range of
immunological functions (Miller, 1995). "Stress-induced alterations in immune
functioning have resulted in increased risk of infectious diseases and other illnesses in
children" (Miller, 1995, pg. 10).
Perry (1997) reported that children exposed to trauma, such as community
violence, often manifost alterations in their overall arousal, increased startle reposes and
sleep disturbance.
32
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Perry (1994) posits that "alarm reactions" induced by traumatic events are
associated with abnormal patterns of catecholamine activity. This may result in altered
development of the central nervous system, with consequent dysregulations of the
cardiovascular system as well as o f fear-enhanced startle responses and growth
hormone transmission and regulation.
Exposure to the chronic stress and fear o f community violence may also lead to
a dysregulation o f the hypothalamic-pituitary-adrenal (HP A) axis, a major stress-
regulating system. This system contains two feedback mechanisms, negative and
inhibitory, integral to the production of cortisol. Two different types of physiological
dysregulation may occur with exposure to chronic or traumatic stress such as exposure
to community violence. The first type involves an enhanced negative feedback
mechanism in the HP A axis. This mechanism leads to higher basal cortisol levels and is
linked with prolonged “fight or flight” responses, increased responsiveness to stress,
and PTSD symptomatology. The second type o f dysregulation involves a reduced
negative feedback mechanism that results in decreased cortisol production. This
reduction contributes to a decrease in responsiveness to stress and may lead to
depression (Golier & Yehuda, 1998; Nelson & Carver, 1998).
Parental Distress
Parents living in communities with high rates o f violence frequently express
helplessness and frustration with their inability to fulfill their responsibility to protect
their children (Garbarino et al., 1992; Richters & Martinez, 1993). This sense of
powerlessness and hopelessness counteracts parents’ deep-rooted desire to shield and
safeguard their children from harm. When their surroundings are imbued with danger,
33
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
parents may be afraid to allow their child to explore their environment and may adopt
an overly restrictive parenting style hi an attempt to protect their child (Wallen &
Rubin, 1997).
The stress precipitated by exposure to community violence negatively impacts
the ability o f many parents to parent well and to be responsive to the needs o f their
children. Research findings substantiate that parents with higher levels of emotional
distress tend to display lower levels o f parenting efficacy (Gondoli & Silverberg, 1997).
The stress associated with community violence may also contribute to parental
depression and parents’ diminished sense of self-esteem and self-efficacy.
Parents exposed to community violence may experience distress symptoms
similar to those experienced by their children. They may become numbed,
overwhelmed, fearful or anxious. These feelings are likely to be communicated to the
child in verbal and non-verbal ways and result in negative feelings. As a result, when
parents are overwhelmed or distressed by violence, the opportunity to establish and
enjoy a secure and positive attachment with their child may become weakened.
The traumatic stress derived from violence exposure likely has adverse effects
on the exchange o f social support among family members and the child’ s ability to cope
(Barrera & Li, 1996). When parents feel helpless or stressed in confronting the day-to-
day reality o f community violence they become less able to moderate its deleterious
effects in the lives o f their children. Research documents that the influence of parents is
the most important mediating factor in the lives o f children (Wallen & Rubin, 1997). In
an exploratory study with Latina women and their preschool children, the investigator
34
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
found that increased maternal distress symptomatology (PTSD, depression, and anxiety)
contributed to heightened symptomatology and behavioral problems in their young
children (Aisenberg, 2001). Findings highlight that mother’s distress symptomatology
has a mediating effect on child’s behavior problems. The more distressed the parents,
the less the child is able to successfully cope, and the more the child manifests
symptomatology (Foy et al., 1996).
Two important parental characteristics identified in the literature, which affect
children’s reactions to trauma and manifestation o f symptomatology, are: 1) parent’s
own symptoms o f PTSD, and 2) levels of general parental psychopathology (Famularo
et al., 1994; Shahinfar & Fox, 1997). Foy and colleagues (1996) found a positive
correlation between children's symptoms and trauma-related symptoms in their parents
with regards to child PTSD. In mother-child dyads in which children had been severely
maltreated, children were more than twice as likely to be diagnosed with PTSD when
their mother also had PTSD (Famularo et al., 1994). In other studies, parental distress
or the presence o f parental psychiatric disorders predicted higher levels of PTSD in the
child (Kolko, 1996; Stoddard, 1996).
Maternal Depression
A consistent finding in the literature is the negative relationship between
maternal depression and child outcome (Downey & Coyne, 1990). Studies comparing
depressed mothers with non-depressed mothers have found that depressed women's
behavior is less responsive and less competent with their children (Gelfand & Teti,
1990; Goodman, 1992). Research documents that school-age children o f depressed
mothers are at increased risk for the development o f psychopathology, including
35
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
internalizing and externalizing problems, greater social and academic impairments, and
poorer physical health (Beardslee & Wheelock, 1994; Downey & Coyne, 1990;
Hammen, 1991). Children of mothers suffering depression are also at risk to experience
disturbed attachment behaviors (Campbell et al., 1993; Gelfand, Messinger & Isabella,
1995). Living in an environment o f recurring community violence likely contributes to
these interdependent responses.
Resiliency/Protective Factors
When community violence characterizes the day-to-day environment of a child,
the child is challenged to develop coping styles and strategies to respond to threats to
life and to limitations on daily childhood activities. Although many children who are
exposed to the stress o f community violence develop behavioral or emotional problems,
not all do. In fact, most children exposed to violence do not develop PTSD symptoms
or become perpetrators o f aggression. Rather, they are resilient in the face of such
stressful life events and appear to develop healthy psychosocial functioning (Rutter,
1987).
Resilience, as applied to children in a stressful environment, refers to the
capacity to maintain healthy functioning in an unhealthy, high-risk setting, or the
maintenance o f mastery or competence under stress (Garmezy, 1981). Resilience is the
individual's predisposition to resist or overcome the potential negative consequences of
the risk and develop in a healthy manner (Engle, 1996). Resiliency involves the
ongoing interactions between the protective factors within the child, the vulnerability
factors within the child, protective factors and vulnerability stressors within the child's
36
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
environment of family, peers, neighborhood, school, and society as well as particular
risk factors (Mash & Dozois, 1996).
Recent studies of resilient children have identified important internal protective
factors such as intelligence, easy temperament, and locus o f control which enhance or
promote the child's resiliency in the face o f risk and severe stressors (Werner, 1992).
However, these protective factors are seen mostly as attributes of the individual child
and often emerging despite living in a dysfunctional family (Walsh, 1996). From this
perspective, the parents and family are viewed as impeding the growth o f the child,
limited in their psychological availability to the child, or limited in their ability to
provide their son or daughter with a sense of safety, stability, and support (Garbarino,
1995). Consistent with this view, the parent or family system is less likely to be
engaged in the treatment process of a child manifesting symptomatology resulting from
exposure to community violence. Rather, the child is likely to be engaged in individual
or group treatment.
Unfortunately, few studies on community violence have considered the family
as a potential source o f resilience and as a resource for the child. Typically, the focus of
assessment and treatment has centered on the child. This traditional emphasis on the
child neglects the fact that the most significant factor in buffering the negative effects of
stress is a positive relationship with a parent or significant adult figure who can provide
a supportive climate and attachment with the child (Engle, 1996; Werner, 1995).
Indeed, most children are able to cope with a violent environment and maintain
resilience as long as their parents are not stressed beyond their capacity to cope.
Research demonstrates that a resilient child usually had the opportunity to establish a
37
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
close bond and positive relationship with at least one competent and emotionally stable
adult person who is attuned to his or her needs (Werner, 1995).
The traditional emphasis on the child also fails to take into account what we
have learned from social support theory and the ecological perspective— namely, that
interactional processes are at work between the child and his/her immediate family
(Engle, 1996). While many factors affect a child's development and coping, the most
important processes exist in the interactions o f the child and parent. The nature o f the
individual attributes o f the child transforms the family just as the particular
characteristics o f the family influence the quality and direction of the child's
development (Lemer, 1978). Thus, the consequences o f exposure to community
violence are not the same for all children and adolescents and appear to be dependent on
family characteristics that serve to buffer or exacerbate the deleterious effects of
traumatic exposure (Gorman-Smith & Tolan, 1998).
Contrary to the perspective o f the individual child, studies of traumatized or
high-risk youth reveal that intervening with the parent or conjointly with parent and
child may be more effective and produce better child outcomes (Celano, Hazzard, Webb
& McCall, 1996; Henggeler, 1997). These studies indicate that when parents and
families are viewed and empowered as the primary change agents in the life of their
child favorable therapeutic outcomes can be attained (Henggeler, 1997). For example,
Multisystemic therapy (MST), which has demonstrated substantial effectiveness in
addressing serious behavior problems presented by young people, has found that the
probability o f favorable long-term outcome is determined primarily by the mother's
strength and competence (Henggeler & Boruin, 1990; Henggeler et al., 1995). In their
38
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
research with girls who had been sexually abused and their non-offending mothers,
Celano and colleagues (1996) found that treatment of the parent alone as well as
conjoint treatment resulted in greater reduction of child distress symptomatology than
the provision of individual treatment to the child victim.
Summary of Literature Review
The literature clearly supports the present investigation o f Latino and African-
American mother and child exposure to community violence and associated
symptomatology. Previous research has focused primarily on the experience o f
African-American children and adolescents and the direct effects o f exposure in their
lives. Few studies have examined the experience o f Latino youngsters. Research in
areas other than community violence has shown that parental symptomatology
negatively affects children's reactions to trauma and their symptomatology (Famularo
et al., 1994; Shahinfar & Fox, 1997). However, scant research has investigated the
direct effects o f community violence in the lives o f parents o f exposed to violence. As
a result, the effects of community violence among Latino families are largely unknown.
Also, little is known about the indirect effects o f maternal distress in predicting child
psychological and behavioral problems. An examination o f both direct and indirect
predictors o f child symptomatology is essential if we are to folly understand the impact
o f community violence. The conceptual and empirical literature reviewed support this
examination.
39
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Hypotheses
The study tests the following hypotheses:
1) Children exposed to elevated levels o f community violence will manifest
elevated distress symptomatology scores based on self-report.
2) Children exposed to higher levels o f community violence based on self-
report will manifest higher total behavioral problem scores.
3) Mothers exposed to elevated levels o f community violence will manifest
elevated levels of distress symptomatology based on self-report.
4) Higher scores o f maternal distress symptomatology predict higher scores o f
children’s PTSD or total behavior problems.
5) Maternal distress symptomatology mediates children’s PTSD and/or total
behavior problems scores.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 3
M ETH O D S
Sampling Plan
The present study built on the work conducted by the UCLA Child Trauma
Reduction Program o f the UCLA Trauma Psychiatry Service (Saltzman et al., 2001).
This school-based program provides treatment for students who manifest PTSD
symptoms and grief as a result of traumatic exposure to violence and loss. The
investigator served as a coordinator o f the program. Prior to initiating treatment, an
assessment was conducted to identify students who reported histories of significant
trauma and who reported clinically significant symptoms of posttraumatic stress or
depression. The UCLA program systematically screened 1200 middle school students
for violence exposure and PTSD from two middle schools in a major city in Los
Angeles County.
Both selected middle schools draw the vast majority o f their students from the
same six square mile neighborhood o f the selected city. This neighborhood contains
high-density housing with a population o f 6.17 dwelling units per acre. It has
disproportionate numbers o f low-income families with 26% of the households below
the poverty level and 18% receiving public assistance. Seventy-five percent o f the
students from the study’s schools meet the low-income criteria to participate in the
government’s free lunch program.
The neighborhood is also a high crime community as noted by police statistics.
Nearly 70% o f the city’s violent crime is annually committed in the six square mile
region. A prior large-scale survey o f violent victimization among ninth grade students
41
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
attending a public high school in this neighborhood revealed that twenty-two percent
reported having been shot at, 8% reported that they had been stabbed, 20% reported
having been badly beaten, and 9% reported having being sexually assaulted (Rohrbach,
Mansergh, Fishkin & Johnson, 1996). O f the students surveyed, 48% reported that they
had a friend who was injured or killed by gunshot, and 28% reported that they had a
relative who was injured or killed by gunshot.
The students reside in a diverse, urban city with a population o f approximately
135,000 according to the United States Census 2000. The ethnic diversity o f these
urban schools is reflected in the fact that 46% of the students are Latino, 39% African-
American, and 12% Caucasian. Approximately 40% have limited English proficiency
and primary language proficiency in Spanish.
According to 1999 California Crime Index statistics, a monthly average o f 137
confirmed cases of crimes such as armed robbery and assault were committed within
the city. However, only 2 homicides occurred. This low figure contrasts with figures
of preceding years. In 1998, 10 homicides took place and for the preceding five years
an average of 12 homicides occurred in the city.
According to 2000 statistics, the median income for a family of four in the
study’s city was $58,423 compared to $49,800 for residents of Los Angeles County
(California Department o f Finance, 2000). Despite this high median income, many
families in the selected municipality experience poverty. Over one of every five
families residing in this city survive on annual incomes o f less than $15,000 (City
Report, 2001).
42
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Los Angeles County provides an excellent setting for this research. It is a large
metropolitan area with an ethnically diverse population o f over 41% Latinos, 10.3%
African-Americans, along with 36.9 % White and 11.5% Asian (California Department
o f Finance, 1997). Significantly, Los Angeles County will have a Latino population
over 50% by the year 2002 (McCormack, 2000).
At the non-neighborhood school’s fall registration, 90% o f the parents signed an
active consent form giving permission for their son or daughter to participate in the
screening. At the neighborhood school site, the nature and purpose o f the screening, the
kinds of information that would be asked, and the length o f time needed to complete the
measures were explained to the students via a presentation to their English classes.
Letters were then sent to their mothers by mail requesting their permission to complete
the screening measures. Fifty-six percent of the mothers consented to have their child
complete the assessment measures. Overall, 73% o f the mothers gave consent for their
child to participate in the UCLA screening.
Students for whom permission was received completed the UCLA Community
Violence Exposure Survey (Saltzman, Layne, & Steinberg, 1998), which is a modified
version o f the Richters and Saltzman (1990) violence questionnaire. Also, they
completed the UCLA PTSD Reaction Index—Revised Adolescent Version (RI-R;
Rodrigruez, Steinberg, & Pynoos, 1999), and the Reynolds Adolescent Depression
Scale (RADS, Reynolds, 1987) in a large group format o f 20 -40 students. This
screening took place in the schools’ cafeterias over a period o f 3-4 months and
comprised the first stage of the present study.
43
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The present study extended the UCLA program in two ways: 1) by randomly
selecting students from the large pool o f students who completed the UCLA screening
measures and having them complete additional standardized measures; and 2) by
interviewing the mothers o f these students and having the teachers complete measures
on these students.
Prior to undertaking the 2n d phase o f the present study, the investigator presented
and discussed the research protocol with each school principal and received their
support and permission. The investigator then made a presentation to the Board of
Education o f the designated School District and received its permission to initiate and
implement the study. In addition, the investigator met with the English teachers at both
schools at a departmental meeting to inform them of the nature and purpose of the study
and invite their participation. Teacher participation was 100%.
In the second stage o f assessment young adolescents were randomly selected to
and stratified according to race/ethnicity (African-American, Latino), gender, and level
of violence exposure (high, low) to complete additional measures as part of the second
stage o f assessment.
The criteria o f high exposure was based on the categorization of the Project on
Human Development in Chicago Neighborhood Study (Gorman-Smith, 1998). In this
particular study, 3 or more experiences of distinct types of community violence were
determined to constitute a high level o f violence exposure. This criteria was followed
in the present study. Thus, low level o f exposure was defined as 2 or fewer experiences
o f community violence.
44
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The investigator then sought to enlist the participation of the mothers o f the
selected students. The mothers were recruited via letter that was sent by mail to the
family home. The letter explained the nature and purpose of the study in English and
Spanish and detailed the confidentiality of information that would be safeguarded. In
addition, it affirmed the voluntary nature o f the consent. The letter requested that an
attached consent form be returned within a two-week period to their child’s English
teacher indicating the mother’s preference.
The initial goal was to recruit 120 mother-child pairs. Criteria for inclusion in
the study were: participants must be the biological mother of a 6th , 7lh or 8th grade
student attending either o f the 2 selected middle schools or a student at such schools,
must be residents of the identified neighborhood, they must be African-American or
Latino and must be able to speak English or Spanish.
Initially, 200 letters were sent by mail to the mothers of the selected students.
Responses were received from 170 mothers, with 60 affirming their participation in the
study. Another 200 mothers were contacted via letters. One hundred forty responses
were received, however, the vast majority declined to participate in the study.
In the recruitment process, a number of procedures were implemented to
encourage parental participation. Compensation ($25 per mother) and choice o f
availability to interview the mothers in their homes were made. Nevertheless, parent
participation in the study proved challenging. Despite repeated attempts to increase the
sample size, including follow-up letters to the mothers and phone calls, only 22.5%
(N=90) agreed to participate. Five mothers were foster mothers and did not meet the
criteria o f being the biological mother. Another five mothers were lost to attrition.
45
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Thus, eighty mother-child pairs comprise the subjects o f the present study.
African-American Boys
N = 19
Latino Boys
N = 22
African-American Girls
N = 14
Latina Girls
N = 25
Study Design
The study incorporates a multi-method, multi-informant assessment o f multiple
domains through the use o f standardized and unstandardized measures. This
methodology provides a more reliable ascertainment o f symptoms across multiple
domains, identifies objective and subjective responses to divergent traumatic events,
and evaluates symptom severity.
The principal variables o f interest are: community violence exposure, PTSD,
depression, and anxiety among mothers and children, as well as child behavioral
problems. For purposes o f this study the main variables o f interest are treated as
continuous variables.
Utilizing a two-stage assessment procedure, eighty Latino and African-
American mothers and their young adolescent children participated in the present study.
The study
1) examines and compares the types, frequencies, and levels o f child lifetime
exposure to community violence as self-reported by a sample of sixth,
seventh, and eighth grade students and reported on the students by their
mothers;
2) examines and compares the students’ distress symptomatology, specifically,
PTSD, depression, and anxiety based on self-report;
46
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3) measures child behavioral problems based on mothers and teachers reports;
4) examines and compares the types, frequencies, and level o f mothers’ lifetime
exposure to community violence based on self-report;
5) measures and compares the mothers’ self-reported distress symptomatology
6) examines and identifies the possible moderating and mediating effects o f
mothers’ symptomatology upon child PTSD and total behavior problems.
Young adolescents were chosen as the subjects for the research because
previous research has shown that children under 9 years of age have been fairly
unreliable in their symptom reporting (Schwab-Stone et al., 1996). Furthermore,
younger children may not be able to reveal and express their internal experiences and
many o f the symptoms o f PTSD (Bingham & Harmon, 1996).
The decision to implement the study in a school setting as opposed to
community mental health centers or private offices is based on two key advantages:
1) schools are an appropriate and desirable setting for assessment since they lack the
stigma often associated with community mental health centers (Heflinger & Nixon,
1996); and 2) schools are settings in which symptoms associated with PTSD are likely
to emerge (Pynoos, 1998).
Measures
Mothers, their children, and the English teachers of the children completed
multiple paper and pen measures on a variety of domains. Research affirms the use of
multiple informants to provide more reliable data (Holmbeck, Li, Schurman, Friedman,
& Coakley, 2002, March, 1999). Studies consistently report elevated symptoms o f
PTSD or child behavior problems in children exposed to community violence. Thus, it
47
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 1. Summary o f Variables and Measures
Variable Measure Constructs
Assessed
Sclf-
Reuort
Mother
Self-
Reuort
Child
Mother
Reuort
on Child
Teacher
Reuort
on Child
Community
Violence
Exposure
Self-Assessment of Exposure
to Community Violence
(SAVE-P) (Aisenberg, 1998)
Victimization,
witnessing of
community and
domestic
violence
X
UCLA Community Violence
Survey
(Saltzman et al., 1998)
Victimization,
witnessing of
community
violence
X
Parent Assessment of Child
Exposure (PACE)
(Aisenberg, 1998)
Victimization,
witnessing of
community
violence
X
PTSD Posttraumatic Diagnostic
Scale (PDS)(Foa, 1995)
Intrusive
Thoughts,
Avoidance,
Hyperarousal,
Fear,
Attention/Sleep
Disturbance
X
SCID DSM-IV PTSD
symptoms
X
PTSD-Reaction Index
Adolescent Version
(Rodriguez et al., 1999)
Intrusive
Thoughts,
Avoidance, Fear
Recurrence,
Attention/Sleep
Disturbance
X
KSADS-E, PTSD subscale DSM-IV PTSD
criteria
symptoms
X
Depressed
Mood
BSI-depression subscale
(Derogatis, 1983)
Depression X
Reynolds Adolescent
Depression Scale (RADS)
(Reynolds, 1987)
Depression X
49
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 1. continued
Variable Measure Coastructs Self-
Reuort
Mother
Self-
Renort
Child
Mother Teacher
Assessed Reuort
on Child
Reuort
on Child
Anxiety BSI-anxiety subscale
(Derogatis, 1983)
Anxiety X
MASC
(March et al., 1997)
Anxiety,
Separation
Anxiety
X
Behavior
Problems
CBCL Total Score
(Achenbach, 1991)
Demanding/
Suspicious,
Sexual Acting
Out, Withdrawn/
Secretive,
Dissociative/
Hyperactive
Aggressive/
Delinquent
Bizzare/
Destructive
X
Parent Rating Scale
(Conners, 1998)
Oppositional,
ADHD,
Hyperactive,
Global Index
X
Teacher Rating Scale
(Conners, 1998)
Oppositional,
ADHD
Hyperactive,
Global Index
X
Teacher-Child Rating Scale
(Hightower, 1985)
Acting Out,
Behavior and
Learning
Problems
X
Sexual
Abuse
Sexual Abuse Trauma Index
of TSC-40
(Briere, 1996)
Sexual Abuse
Trauma
X
Domestic
Violence
Psychological Maltreatment
of
Women Inventory
(Tolman, 1989)
Dominance-
isolation;
Verbai-
emotional abuse
X
Achieve/
Ability
Teacher-Child Rating Scale
(Hightower, 1985)
Social Skills,
Assertiveness,
Task
Orientation,
Peer Sociability
X
Family
Resources
Demographic Form
Family Income,
Parents’
Education
X
50
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
To assess violence exposure, mothers completed the Self-Assessment of
Exposure to Community Violence, Parent Version, (SAVE-P) (Aisenberg, 1998),
developed by the investigator. This questionnaire measures the mother’s own lifetime
exposure to violence at various intervals of time, e.g. within 6 months, within the past
year, over one year ago. The mothers then completed the following self-report
measures: the Posttraumatic Diagnostic Scale, (PDS, Foa, 1997), the Trauma Symptom
Checklist-40 (Briere & Runtz, 1989; Elliot & Briere, 1992), the depression and anxiety
subscales o f the Brief Symptom Inventory (BSI) (Derogatis & Melisaratos, 1983), the
Rosenberg Self-Esteem Scale (Rosenberg, 1965), and the Structured Clinical Diagnostic
Interview (SCID).
The mothers also answered several measures reporting on their child’s
experiences, including: the Parent Assessment o f Child Exposure to Community
Violence (PACE) (Aisenberg, 1998), the Child Behavior Checklist (CBCL)
(Achenbach, 1991), and the Conners Parent Rating Scale (Conners, 1998). In addition,
the mothers completed a demographic information form specifically designed for this
study.
The young adolescents’ completed the following self-report measures: the
UCLA Community Violence Exposure Survey (Saltzman, Layne, & Steinberg, 1998),
the UCLA PTSD Reaction Index-Revised Adolescent Version (RI-R; Rodriguez et al.,
1999), the Reynolds Adolescent Depression Scale (RADS; Reynolds, 1987), the
Multidimensional Anxiety Scale for Children (MASC, March et al., 1997), and the
PTSD module from the K-SADS.
51
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The teachers reported on their observed behavior of the sample’s youngsters by
completing the Teacher-Child Rating Scale (TCR-S) (Hightower et al., 1986) and the
Conners Teacher Rating Scale (Conners, 1998).
M aternal Self-Report Measures
Self-Assessment of Exposure to Community Violence-P (SAVE-Pl (Aisenberg.
19981.
The investigator developed this 54-item violence survey questionnaire.
Patterned after the widely used Richters and Saltzman Survey o f Exposure to
Community Violence (1990), this self-report measure inquires about the mothers’
lifetime exposure to community violence as well as their exposure to other forms of
violence, including domestic violence and sexual abuse. It ascertains her experience o f
violence within various time intervals (e.g. past six months, past year, over a year), the
different settings of the occurrence, the frequency o f occurrences, and her relationship
to the perpetrator or victim. In addition, it ascertains if the child was a witness to the
violent event. Finally, the SAVE-P instrument obtains qualitative information on the
mother’s perception o f the most upsetting or disturbing experience in her life.
This instrument was cross-translated into Spanish to ensure clarity and enhance
comprehension and reliability. Pilot testing o f this instrument revealed face validity and
ease of comprehension of this measure. A version of this instrument was used in a prior
study focusing on Latino preschool children and their mothers (Aisenberg, 2001).
Posttraumatic Diagnostic Scale fPDSl fFoa. 19971
This instrument measures adult PTSD and is the only self-report measure to
assess all six criteria for PTSD according to the DSM-IV. It yields both a dichotomous
52
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
PTSD diagnostic score and a continuous symptom severity score (Weathers & Keane,
1999). Test-retest reliability for symptom severity is .83 over a two-week interval. A
Cronbach alpha o f .92 has been calculated on the items comprising the Symptom
Severity Score, indicating that the Symptom Severity Score is internally consistent. The
PDS’ sensitivity has been reported to be 82% and its specificity to be 76.7% (Foa,
1995).
Traum a Sym ptom Checldist-40 (TSC-401 (Briere & Runtz. 1989: Elliot &
Briere. 19921
This 40-item self-report instrument evaluates symptomatology in adults arising
from childhood or adult traumatic experiences. It measures posttraumatic stress along
with 6 subscales (e.g., anxiety, dissociation, depression, and sexual abuse). Reliability
values for the subscales range from .64 to .77. The TSC-40 is a relatively reliable
measure with alphas for the full scale averaging .89 to .91 (Briere, 1997).
Psychological Maltreatment o f Women Inventory fPMWD (Tolman. 19891
The PMWI is a 58-item scale that contains two subscales: dominance-isolation
and verbal-emotional abuse. Items are rated are on a Likert scale o f 1 = never to 5 =
very frequently. High reliability (alpha > .90) has been reported (Tolman, 1999).
Convergent validity has been shown with moderate to high correlation with the Conflict
Tactics Scale (Strauss, 1979) and the Brief Symptom Inventory General Symptom
Index (Derogatis & Melisaratos, 1983). The PMWI measure has been used to
successfully differentiate among women who were battered, relationship distressed but
nonbattered women, and women in nondistressed relationships (Tolman, 1999).
According to Tolman (1999), it is premature to establish a cutoff score. However, the
53
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
battered women in the validation sample had a mean score o f 61.7 (SD =25.5) for the
dominance/isolation subscale and a mean score o f 70.7 (SD =21.5) for the
emotional/verbal subscale.
Brief Symptom Inventory (BSD (Derogatis & Melisaratos. 1983)
The depression and anxiety subscales o f the Brief Symptom Inventory (BSI)
each contains 6 items with a severity scale of 0 to 4 indicating the degree to which the
participant was disturbed by each item during the preceding month. A higher score
indicates greater psychiatric symptomatology (Canetti, Bachar, Galili-Weisstub, Kaplan
De Nour & Shalev, 1997). A total T-score o f 63 is considered the clinical cut-off point
for each o f the subscales (Derogatis, 2000). The internal consistency reliability
coefficients for the two subscales using Cronbach’s alpha are as follows: Depression
subscale .85 and Anxiety subscale .81. Test-rest reliabilities are .84 and .79
respectively.
Stability o f Self-Esteem Measure fRSEl (Rosenberg. 19651
The Stability of Self-Esteem Measure is a ten-item measure o f global self
esteem. Reviewing seven studies, Wylie (1989) cited internal consistency reliabilities
ranging from .72 to .87 and test-retest reliabilities o f .85 after two weeks. Scoring
ranges from 1 = strongly agree to 4 = strongly disagree. The scoring is reversed on five
items. Thus, a high score reflects low self-esteem. A cut-off score of 20 has been
identified in the literature (Gottesman & Lewis, 1983; Neuling & Winfefield, 1988).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Measures of Maternal Reports on Child
Parent Assessment o f Child Exposure to Community Violence-PACE
(Aisenberg. 1998~)
The investigator also developed a parent report on child exposure to community
violence, the Parent Assessment o f Child Exposure to Community Violence (PACE)
(Aisenberg, 1998). This 60-item measure asks a parent to indicate to the best of their
knowledge their child's experiences o f different acts o f community violence within
various time intervals (e.g. past six months, past year, over a year ago) as well as the
frequency of such exposure. The questionnaire assesses for the child's experiences o f
being a victim of violence as well as the child's relationship to the victim and/or
perpetrator. Also, it surveys the frequency of exposure to other forms of trauma, such
as exposure to domestic violence. Lastly, the measure gathers qualitative information
regarding the child’s most upsetting experience related to community violence based on
the mother’s perspective.
Child Behavior Checklist (CBCL1 (Achenbach. 19911.
The CBCL is a widely used measure that examines behavioral and emotional
problems for youngsters of ages 4-17. Mothers rated their child on 113 items, using a
3-point scale o f not true at all (0), somewhat true (1) or very true (2) for behaviors
within the past six months. T-scores have been developed to allow comparison by
gender and age. On this scale a score o f 60 or above represents the clinical cut-off point
with scores between 60 and 63 representing the borderline range (Achenbach, 1991).
For purposes o f analysis, only the Total Score was used. It indicates overall
behavioral problems comprised from eight syndrome scales—withdrawn, somatic
55
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
complaints, social problems, anxious/depressed, thought problems, attention problems,
delinquent behavior, and aggressive behavior. Validity is supported by numerous
studies which have reported significant correlations between the CBCL and other child
problem measures (Achenbach, 1991). According to the manual, the one-week test-
retest reliability of the Total scores is r = .91 for non-referred children.
Conners Parent Rating Scale-Revised (CPRS-R1 (Conners. 19981
The CPRS-R is an 80-item measure that focuses on disruptive child behaviors
often related to ADHD. It has internal reliability ranging from .75 to .92 with an overall
correct classification rate o f 93.4% (Conners et al., 1998). Important subscales include
Oppositional, ADHD, PTSD, Hyperactive, and Global Index. T-scores o f at least 65
signify clinical caseness.
Student Self-Report Measures
UCLA Community Violence Exposure Survey (Saltzman. Lavne. & Steinberg.
19981
This measure is an adapted version o f the Survey o f Exposure to Community
Violence (Richters & Saltzman, 1990) that has been widely used in school settings to
screen for exposure to community violence. This self-report measure includes 38 items
covering different types o f traumatic exposure. These events include witnessing or
being a victim o f a serious accident, serious life-threat, beating, stabbing or knife attack,
shooting, kidnapping, attempted/committed suicide or homicide, and having a family or
close friend who was badly hurt in an accident or violent incident. For purposes o f
analysis, exposure to disaster and accidents were not included.
56
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The UCLA PTSD Reaction Index-Revised Adolescent Version fRI-R:
Rodriguez. Steinberg. & Pvnoos. 1999)
This self-report instrument measures the presence and frequency over the past
month o f all 17 DSM-IV PTSD symptoms. It is rated on a 5 point Likert scale ranging
from 0 (none of the time) to 4 (most all the time). The RI-R is a revised version o f the
DSM HI-R UCLA Reaction Index (Frederick, Pynoos, & Nader, 1992) and worded for
the adolescent age group. It is appropriate for youth aged twelve and older, and may be
administered either to individuals or in a group setting. Twenty o f the 22 items assess
core PTSD symptoms; two additional questions assess associated features including fear
of recurrence and trauma-related guilt. An evaluation of the psychometrics o f the
DSM HI-R version reported excellent test-retest reliability. Kutlac and colleagues
(2000) reported a Chronbach’s Alpha of .92 and evidence of good convergent validity.
Additionally, a cutoff score o f 30 has been identified as having good sensitivity and
specificity in detecting cases o f moderate PTSD and a score o f 40 to detect cases o f
severe to very severe PTSD (Pynoos et al., 1993). For purposes of analysis, the cutoff
score o f 30 was used.
Reynolds Adolescent Depression Scale (RADS) (Reynolds. 1987)
The RADS is a self-report measure o f depressive symptoms among adolescents
between the ages o f 12 and 18. It consists of 30 items with a four-point Likert-type
response format (1-4), from “almost never” to “most o f the time”. A cut-off score o f 77
represents clinical levels o f depression. The RADS has demonstrated excellent validity
and reliability. A range of studies have reported alpha coefficients ranging from .90
through .96, indicating high internal consistency o f the instrument (Reynolds, 1987). It
57
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
has been widely used in school settings and has extensive large-sample normative data.
With a sample o f over thousand young adolescents ages 12 to 14, Reynolds (1987)
found a strong correlation o f 0.70 between the RADS and the Children’s Depression
Inventory (Kovacs, 1992).
Multidimensional Anxiety Scale for Children (MASC) (March et al.. 19971
The MASC is a self-report instrument designed specifically to address a wide
spectrum o f anxiety symptoms in children and adolescents from 8 to 17 years of age. It
contains 39 items and has four subscales: physical symptoms, social anxiety, harm
avoidance, and separation anxiety. Also, it contains an Anxiety Disorder Index that is
used to identify children and adolescents who may be experiencing clinically significant
levels of anxiety symptoms (March, 1997). T-scores above 65 have been identified to
differentiate children and adolescents with a diagnosis o f an anxiety disorder from those
without such a diagnosis.
Two separate school-based population studies demonstrate that this instrument
possesses excellent internal reliability (March et al., 1997).
Teacher Completed Measures
Conners Teacher Rating Scale-Revised (CTRS-Rl (Conners. 19981
The Conners Teacher Rating Scale-Revised improves upon the CTRS by
updating item content to reflect current conceptualizations of childhood disorders, by a
more representative sampling, and by a stable factor structure confirmed with advanced
statistical techniques (Conners et al., 1998). A useful advantage o f the CTRS-R is that
its factors correspond with CPRS-R factors. The same subscales found in the Teacher
Rating Scale are contained in the Parent Rating Scale.
58
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Hightower Teacher-Child Ratine Scale fHightower et al.. 1986^
The Hightower measure is a 42-item checklist o f child behavior problems and
provides a series of ratings about school behavior. The first section contains eighteen
items that are ranked from 1 = not a problem to 5 = very serious problem. It contains
three 6-item subscales: Acting Out (e.g. disruptive in class, overly aggressive to peers),
Shy-Anxious (e.g. timid, worried), and Learning Problems (e.g. poor work habits,
underachieving).
The second section focuses on school competencies. It contains 24 items that
are ranked on a 5-point scale in term o f how well they describe the child, with
indicating 1 not at all and 5 indicating very well. For analysis, these items are reversed
coded. This section has four subscales: Frustration-Tolerance (e.g. copes well with
failure, accepts imposed limits), Assertive Social Skills (comfortable as a leader,
defends own views under group pressure), Task Orientation (well organized, completes
work), and Peer Sociability (makes friends easily, classmates wish to sit near this child).
Hightower and colleagues (1986) reported reliability coefficients ranging from
.85 to .95 and test-retest coefficients ranging from .61 to .91 for the T-CRS scales.
Data Collection
Information on sociodemographic variables collected in the present study
include: child's gender, grade, race/ethnicity, mother's race/ethnicity, her educational
and income levels, marital status, employment, number o f children in family, and
number o f adults residing in the family home. A form was developed for this purpose.
In addition, census data and crime rates o f the designated neighborhood and the entire
city were obtained.
59
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The study utilized a two-stage assessment procedure. At the initial stage, 1200
students completed the violence and distress symptoms’ questionnaires completed the
measures in a large group format in the school cafeteria during one class period o f the
regular school day. The students placed their name on each measure. Proper spacing in
the room was made to ensure the confidentiality o f responses. Students were instructed
to not share their responses with the other participants. This screening process took
place over a period of 3 months. All students completed English-version instruments.
Prior to the administration o f the instruments, the students were explicitly
instructed not to exaggerate or conceal instances o f violence to which they had been
exposed. The limits o f confidentiality were detailed, including the need to inform the
proper authorities if suspected or actual abuse was disclosed. While completing the
measures, very few students exhibited discomfort or distress signals. Those students
who manifested discomfort were invited to stop and allowed to process their experience
outside the presence o f the other students by members of the UCLA Trauma Psychiatry
Service with extensive clinical training and experience. No referrals were necessary.
The second stage involved the assessment o f the 80 students randomly selected
from the larger pool o f students according to the stratification criteria. Trained
interviewers met with these students and conducted face-to-face interviews with them
on an individual basis. Limits of confidentiality were explained prior to the
administration of this stage’s two measures: the Multidimensional Anxiety Scale for
Children (MASC, March et al., 1997) and the Schedule for Affective Disorder and
Schizophrenia for School-Age Children (KSADS-E). Four students manifested
difficulty answering the questions in English and thus completed these measures in
60
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Spanish. These interviews were conducted 3-6 months following the completion o f the
initial student screening measures that comprised stage one.
The mothers who affirmed their willingness to participate in the study were
contacted by telephone by an interviewer who scheduled an appointment to complete
the study’s measures. Interviewers met with the mothers on an individual basis
following the initial screening o f the students. The mothers were given the choice of
being interviewed in their home or at their child’s school. All mothers preferred being
interviewed in their home.
These interviews took place 3 to 9 months following the initial screening o f their
children. It took approximately 1 1/2 to 2 hours for the mothers to complete the study’s
measures. All but 3 Latina mothers completed the measures in Spanish.
Prior to the administration o f the study’s measures, the mother read and signed
the University Institutional Review Board’s (IRB) approved informed consent form that
denotes the nature and purpose o f the study as well as the limits o f confidentiality.
Only after answering all questions and receiving the signed informed consent did the
interviewer proceed. The mothers were explicitly told not to exaggerate or conceal
instances of violence to which they or their child in middle school had been exposed.
All instruments were available in English and Spanish and participants were asked
which version they preferred.
To facilitate better understanding o f the questions and lessen reporting errors
and missing data, the interviewer read the questions o f the various measures aloud in
the subjects’ preferred language if the subject indicated or demonstrated difficulty with
61
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
reading the measures. Following completion o f the measures, the interviewers
debriefed with the subjects to ensure their well being.
Upon request or if deemed appropriate, referrals to local community agencies
and programs were made to allow families to address issues or problems in a more
thorough manner. Ten percent o f the mothers requested such information. In general,
the mothers who participated in the study expressed that the interview process was both
helpful and informative to them.
The mothers received a small stipend to compensate them for their time. The
students were not compensated for their participation.
The English teachers completed the Hightower Teacher-Child Rating Scale
(Hightower et al., 1986) and the Conners Teacher Rating Scale (Conners, 1998) based
on their observations of the designated students classroom performance and behavior.
The amount of time needed to complete these measures was approximately IS minutes.
They filled out the measures within the same timeframe that the mothers completed
their measures. The teachers received ten dollars as compensation for their time and
effort.
Prior to conducting the student and mother interviews, the investigator trained
four interviewers in general interview techniques and the procedure of obtaining
informed consent and administering each o f the study’s measures to be completed by
the student or mother. The training included how to recognize distress signals and how
to sensitively respond to the subjects. In addition, two interviewers received a
specialized two-day training on the administration of the PTSD module o f the KSADS-
E by Dr. Ned Rodriguez of the UCLA Trauma Psychiatry Program.
62
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Two interviewers were second year graduate students in the Masters Program of
the School o f Social Work at the University of Southern California and had completed
coursework in research methodology, one had an MSW degree and had clinical
experience, and one was a resident medical student. Two assistants were fluent in
Spanish. Adherence to the instructions of each measure by the interviewers was
assessed throughout the data collection process. The investigator provided regular bi
weekly supervision and continuous feedback to the interviewers in order to maintain
standardization o f the interview protocol.
Data Analysis Plan
Descriptive statistics were used to examine the sociodemographic characteristics
of the African-American and Latino families. Univariate t-tests using Scheffe’s
procedure to adjust alpha for multiple comparisons and bivariate correlations were
conducted. T-Tests were performed to examine group differences based on gender,
ethnicity, school, level o f exposure, parents’ years of education, and family income.
Multiple regression analysis was conducted to examine moderating and mediating
effects.
Scores from the BSI subscales as well as the scores from the MASC and
Conners’ Scales (Parent and Teacher) use area T-scores. Like all T-scores, the area
T-score is characterized by a distribution with a mean of 50 and a standard deviation of
10. However, the area T-score has significant advantages over linear T-scores
(Derogatis, 2000). Area T-scores reflect meaningful and accurate percentile
equivalents. Thus, a T-score o f 60 always places the respondent in the 84th percentile of
the norm and a T-score o f 40 places the respondent in the 16th percentile. This
63
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
information is valuable for interpreting the clinical meaning and significance o f an
individual’s responses (Derogatis, 2000).
Following the recommendation o f Patterson and Bank (1986) the individual
maternal scores on PTSD, depression, and anxiety were evaluated individually but also
converted to z-scores. To provide a more reliable measure, the z-scores were then
combined to form one score, maternal distress symptomatology.
With information gathered from multiple informants, separate regression
analysis for each informant was conducted. To test for possible moderating effects of
maternal distress symptomatology, the regression methodologies proposed by Aiken
and West (1991) were used. Child total exposure to community violence (self-report),
maternal distress symptomatology (z-score), and their interaction term were entered
hierarchically as predictors o f scores for child PTSD. Each effect was tested while
partialling out the effects o f equal and lower order. If this interaction is significant, then
maternal distress symptomatology functions as a moderator.
Following the recommendations o f Baron and Kenny (1986), a series of
regression equations was then conducted to ascertain and verify the presence o f a
mediational relationship between mother’s distress symptomatology (the mediator) and
child’s PTSD (the outcome variable). First, the mediator, mother’s distress symptoms,
was regressed on the independent variable, child’s exposure to community violence;
second, the dependent variable, child’s behavior problems, was regressed on child’s
exposure to community violence; and third, the dependent variable was regressed on
both the independent variable and on the mediator. If a previously significant relation
64
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
between the independent and dependent variable is no longer significant or is
significantly attenuated than mother’s distress symptomatology functions as a mediator.
Additionally, similar analysis for moderating and mediating effects was
conducted using the mother’s report o f their child’s exposure to community violence,
maternal distress symptomatology (z-score) and the child’s total behavior problem
score.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 4
R ESU LTS
Description o f the sample
Eighty children and mothers completed the assessment battery. Table 2
summarizes descriptive data o f the children comprising the sample. The children
ranged from 11 to 14 years o f age (M = 12.41; SD = .82). The mean age o f Latino
children was 12.32 (SD = .84) and the mean age o f African American children was
12.55 (SD = .79). Eighty-three percent of the children were bom in the United States.
Gender was nearly equally divided with 41 boys and 39 girls. In terms o f ethnicity,
forty-seven youngsters were Latino and 33 were African-American. Among Latinos,
53.2% were girls while girls comprised 42.4% of the African-American cohort. A t-
test revealed a statistically significant gender difference in age among African-
American youngsters (t = 2.689; p = .011). On average, African-American girls (12.93;
SD = .62) were older than the boys (12.26; SD = .81).
Table 2. Descriptives of Young Adolescent Students
Name of Variable Number
fn=801 (n=4T> fn=331
%_________ Latino African American
Age o f Child
11
12
13
14
Gender o f Child
Male
Female
Nativity o f Child
United States
Mexico
El Salvador
10
31
33
6
41
39
66
1 3
1
12.5%
38.8%
41.3%
7.5%
51.2%
48.8%
82.5%
16.2%
1.3%
17.0%
40.4%
36.2%
6.4%
46.8%
53.2%
70.2%
27.7%
2. 1%
6. 1 %
36.4%
48.5%
9.1%
57.6%
42.4%
100%
66
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
From among the 1200 students who completed the UCLA violence screening
measures, seven hundred students were randomly selected to serve as a comparison to
the study’s sample students. This data was inputted at the final stages o f the
dissertation, after the study’s sample had completed their measures. The findings reveal
that the study’s sample is representative of the larger group in terms o f gender and age.
The mean age in years of the seven hundred students is 12.28 (SD = .85). Gender was
nearly evenly distributed with females comprising 50.9% o f the surveyed students.
Table 3 summarizes demographic data o f the mothers comprising the sample.
The mothers ranged in age from 29 to 54 years (M = 39.23; SD = 5.90). African-
American mothers were substantially older than Latina mothers. The mean age in years
for African-American mothers was 43.12 (SD = 5.73) compared to 37.13 (SD = 4.92)
forLatinas. This difference is statistically significant (t = 4.902; p < .01). The majority
o f the mothers were Spanish-speaking only (58.5%) with an additional 9.2% conversant
in English and Spanish. Fifty-nine percent o f the mothers had immigrated to the United
States.
Among the sample, 55.4% o f the families had monthly incomes o f one thousand
dollars or less. This figure is more alarming when one considers that it includes income
generated by extended family members living in the same household (20%). On the
other hand, 15.4% o f the families reported monthly income greater than three thousand
dollars. Fifty-seven percent o f the mothers reported working at least on a part-time
basis and 73.6% o f the fathers are employed.
67
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
than Latino families. Thirty-three percent of African-American families have monthly
incomes greater than three thousand dollars and only 6.1% have monthly incomes of
less than one thousand dollars. Overall, 36.4% o f African-American families have
monthly incomes o f less than fifteen hundred dollars. These figures contrast sharply
with the income o f Latino families among which 31.9% have monthly incomes less than
one thousand dollars. Overall, 68.1% o f Latino families earn less than fifteen hundred
dollars each month. A t-test reveals that these differences in income are statistically
significant (t = 2.856; p < .01).
Most parents are employed. Among Latinos, 51.2% o f the women and 86.1% of
the men are employed. Among African-American parents, more women held jobs than
men did. Sixty-nine percent (68.7%) o f the mothers reported being employed compared
to 47.1% o f the fathers. Sixty-nine percent (68.7%) o f mothers reported being
employed while 47.1% o f the fathers were held jobs. Mean scores reveal that more
Latina mothers (M = 1.49, SD = .51) were employed than African-American mothers
(M = 1.32, SD = .48). However, these scores yield no statistical significant difference
(t= 1.308; p = .20).
Parents’ Educational Differences based on Gender and Ethnicity
The parents’ educational backgrounds vary significantly and contrast with the
selected city’s norm in which 77.5% o f the residents are high school graduates. For the
total sample o f mothers, 53.8% o f the mothers had less than 12 years of schooling.
Similar findings are found with regard to the education level of the fathers
residing in the family home. Like the mothers, over half o f the fathers (58.9%) received
only some high school education and did not graduate from high school.
69
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A further examination o f ethnic and gender differences regarding educational
attainment reveals some clear contrasts. The mean years o f education for Latino fathers
is slightly higher, (7.87; SD = 3.30) than for Latina mothers (7.47; SD = 3.13).
However, the percentage o f Latino fathers and mothers who obtained only an
elementary level o f education is relatively the same, fifty-five percent. Overall, 76.9%
of Latino fathers possessed less than 12 years of formal classroom instruction with less
than one in five (17.9%) having successfully graduated from high school.
On the whole, African-American mothers attained substantially higher levels of
education than Latinas. The mean years of education o f African-American mothers
(12.52; SD = 2.27) is nearly five grade levels higher than the educational experience of
Latina women (7.47; SD = 3.13). A t-test indicates that this difference is significant
(t = 7.893; p < .01). Whereas only 12.1% of African-American mothers failed to
graduate from high school, 81% of Latina mothers did not complete high school. While
45.5% o f African-American mothers received their high school diploma and an
additional 42.4% gained college experience only 14.9% o f Latinas graduated from high
school and just one had some college experience (2.1%).
Findings reveal that African-American men received more education than
Latino men. The majority o f African-American men had graduated from high school
(52.9%) with 29.4% having received some college education. Only 11.8% o f the
African-American fathers failed to obtain their high school diploma. The mean level of
education o f African-American men is 11.59 (SD = 3.81), nearly four grades higher
than for Latino fathers (7.87; SD = 3.30). T-test analysis confirms that this difference in
years of education is statistically significant (t = 3.492; p < .01).
70
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Compared to African-American women, more African-American men graduated
from high school (52.9% to 40.9%) but more African-American women attended
college than African-American men (42.4% to 29.4%). The mean education level of
African-American women (12.52; SD = 2.27) was higher than African-American men
(11.59; SD = 3.81).
Regression analysis failed to find any statistically significant relationship
between years o f education and employment status across gender and ethnicity. Thus,
years of education are not a predictor o f employment status. Further analysis indicated
that race/ethnicity (R square = .174; p < .01) and father’s education (R square = .116;
p < .01) account for 29% o f the variance in family income. However, using years of
education as a predictor o f family income revealed statistical significance only for
Latino fathers’ education. It accounted for 10% o f the variance in family income
(R square = .104; p < .05).
Latino families were larger in size, on average, than African-American families.
While 55.8% of Latino families had 3 or fewer children (M = 3.30, SD = 1.28), 77.3%
o f African-American families had 3 or fewer children (M = 2.64; SD = 1.73). The vast
majority o f the study’s children were bom in the United States (78.4%) with 20% bom
in Mexico.
Latina mothers were younger, on average, than African-American mothers. The
mean age of Latinas was 37.16 years (SD = 4.85) while the mean age o f African-
American mothers was 43.73 years (SD = 6.92). T-test analysis confirms that this
difference in age is statistically significant (t = -4.451; p < .01).
71
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Child and Parent Differences based on Child’s School
The distribution o f students from both schools is nearly equivalent, 42 attended
the neighborhood school and 38 attended a different school to which they were bused.
However, there were marked differences in the composition o f the sample drawn from
these schools. At the neighborhood school, 19 were boys (45.2%) and 23 were girls
(54.8%). A near inverse percentage existed at the non-neighborhood school in which
22 were boys (57.9%) and 16 were girls (42.1%). The vast majority o f students from
the neighborhood school were Latino (N =29,69%) while Latino students (N = 18)
comprised 47.4% o f the students from the non-neighborhood school.
Substantial differences also exist between the parents o f the students from the
respective schools. The mean years of education of the mothers’ with students
attending the neighborhood school is 8.98 (SD = 3.87) and the fathers’ mean education
is 7.23 (SD = 3.50). The parents of students at the non-neighborhood school had
significantly more education. These mothers had a mean education o f 10.29 years
(SD = 3.40) while the fathers had a mean education o f 11.14 years (SD = 3.03).
Significant differentials in income were also noted. Sixty-two percent o f the families of
the neighborhood school had monthly incomes o f less than $1500 compared to only
40.7% o f the families o f the non-neighborhood school.
Child Violence Exposure Scores— Self-Report
Child Exposure to Community Violence
The lifetime exposure to discrete violent events reported by the sample
o f youngsters is substantial as revealed by Table 4.
72
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE 4. Child Lifetime Exposure to Discrete Violent Events
Prevalence Rates % (n=47) (n=33)
Type of Violence_____ Number Victim Witness Latino African American
Punched 21 27.5%
--------
23.4% 33.3%
Threatened physical harm 14 17.5%
---------
17.0% 18.2%
Beaten up 9 11.2%
--------
12.8% 9.1%
At home during burglary 8 10.0%
---------
5.0% 12.1%
Threatened with weapon 6 7.5%
---------
4.3% 12.1%
Chased by a car 5 7.5%
---------
4.3% 9.1%
Shot at 5 6.3%
--------
6.4% 6.1%
Attacked with knife 4 5.0%
--------
4.3% 6.1%
Seen someone beaten 45 56.3% 55.3% 57.5%
Seen family punched 21
---------
26.3% 27.7% 24.2%
Seen someone threatened 19
---------
23.8% 23.4% 24.2%
Seen someone shot at 17
---------
21.3% 27.7% 12.1%
Seen someone choked 13
---------
16.3% 14.9% 18.2%
Seen someone stabbed 12
---------
15.0% 17.0% 12.1%
Seen dead body 11
---------
13.8% 14.9% 12.1%
Lifetime Victimization 44 55.0% 51.1% 60.6%
Lifetime Witnessing 61
-------
76.3% 76.6% 77.8%
A total o f sixty-three young adolescents have been exposed to community
violence in their lifetime (78.8%). Most of the subjects’ exposure occurred more than
six months prior to the time o f completion of the violence questionnaire.
While the young adolescents’ lifetime exposure percentage is high, the number
of times that they have been exposed to community violence is relatively small (M =
3.48; SD = 3.45). O f those who have been victims of violence, 43.8% were victimized
on only one occasion and 23.8% were victims on two occasions. Overall, 55 youngsters
had never been a victim or been a victim only once (68.8%). In terms o f witnessing
violence, 68.8% reported witnessing 2 or fewer violent events in their lifetime.
Further examination reveals that 5% of the young adolescents had been a victim
of violence but not a witness, while 25% had been a witness to violence but never a
73
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
victim. Fifteen young adolescents reported having no exposure to violence of any kind
(18.8%).
The most common event o f victimization was being punched (27.5%). Nearly
eight percent (7.5%) have been threatened with a weapon and 6.3% revealed that they
have been shot at in their lifetime. In terms o f witnessing violence, seeing someone get
beaten was the most frequently reported exposure (56.3%). Over twenty-one percent
have seen someone get shot at and 13.8% have seen a dead body in the community.
The level of total exposure was somewhat unevenly distributed with 56.25% o f
the young adolescents reporting a high level of exposure to community violence (3 or
more total events). Twenty-five boys (60.9%) and 20 girls reported high levels of
exposure (51.3%). Among African-Americans, 52.6% o f the boys and 50% of the girls
had high levels of exposure. Similar percentages were found among Latino youngsters.
Sixty-eight percent of Latino boys and 52% of the girls reported high levels of
exposure.
There are no significant differences on exposure to violence by race or gender.
Although the mean total exposure for Latinos (4.09; SD = 3.48) is higher than that of
African-Americans (3.18; SD = 3.13) and 57.4% o f Latino youngsters reported a high
level of exposure to community violence compared to 51.5% o f African-American
youngsters, t-test analysis failed to reveal any statistical differences (t = 1.191; p = .24).
Similarly, although the boys’ mean total exposure (3.62; SD = 3.51) is higher
than the girls’ mean (3.33; SD = 3.42), and boys are more likely to be victims of
violence than girls (60% to 40%) while girls are more likely to witness acts of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
community violence (78.4% to 67.5%), these differences fail to attain statistical
significance (t = .370; p = .71).
The scores o f the 700 students are quite comparable to the study’s sample in
terms o f exposure to community violence. Among the 700 students, 51.5% had high
levels o f exposure, with 51.8% having been a victim of community violence at least
once in their lifetime, and 74.1% having witnessed an act of community violence.
Cronbach’s alpha reveals good reliability for the subscales o f victimization (.93) and of
witnessing (.88).
T-test analysis confirms that based on total exposure and gender no significant
differences exist between the study’s sample and the larger UCLA sample from which
the study’s sample was drawn (t = .080; p = .94).
In contrast with the study’s sample, significant gender differences were found,
however, among the sample o f 700 students. Whereas 59.7% o f the boys acknowledged
having been a victim o f community violence in their lifetime, only 44.3% o f the girls
had been a victim. Over seventy-seven percent of the boys had witnessed an act of
community violence (77.4%) compared to 70.9% o f the girls. Overall, a higher
percentage o f boys had high levels o f exposure (58.6%) compared to the girls (44.8%).
This difference was statistically significant (t = 3.669; p < .01).
Differences based on School
Frequency o f exposure to community violence was comparable among students
from both schools. Twenty-six o f the students from the neighborhood school (61.9%)
had a high level o f violence exposure compared to 20 students attending the other
school (52.6%). Twenty-three students (54.8%) from the neighborhood school had
75
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
been victimized by violence and 30 students had witnessed acts o f community violence
(71.4%). At the non-neighborhood school, 22 students had experienced victimization
(57.9%) while 30 had witnessed violence (78.9%).
Differences exist in students’ scores o f PTSD and depression. Students from the
non-neighborhood school manifested greater distress symptomatology than students
from the neighborhood school but these differences were not statistically significant.
For example, the mean PTSD score o f the students at the non-neighborhood school was
17.97 (SD = 17.14) compared to 13.27 (SD = 10.39) for the neighborhood school
students. Overall, twenty-two percent of the students from the non-neighborhood
school scored in the moderate to severe range o f PTSD and 14.3% scored in the clinical
range for depression. In contrast, 9.5% o f students from the neighborhood school met
PTSD criteria and 7% scored in the clinical range for depression.
Child Symptomatology Scores—Self-Report
PTSD-Reaction Index Revised Adolescent Version
The overall mean PTSD score was 16.09 (SD = 14.50). Although Latino
children scored higher on average (M = 16.37; SD = 15.99) than African-American
children (M = 14.67; SD = 14.24), no statistically significant difference was found (t =
.469; p = .634). However, significant gender differences exist (t = 2.180; p = .033). On
average, girls scored higher than boys. The mean score for girls was 18.4 (SD = 15.70)
compared to 11.12 (SD = 9.29) for boys.
Sixteen percent o f the young adolescents met the criteria for PTSD. This
percentage reflects an elevated level o f PTSD compared to epidemiological findings of
76
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
less than 10% in the general adolescent population. Examining Table 5, the majority of
those who scored at or above the moderate range were girls and Latino (61.5%; N = 8).
Table 5. PTSD Criteria Scores
Moderate PTSD Severe PTSD
(score >.30) (score >40)
____________________N________ %____________ N_________ %__________________
Boys 3 7.0% 1 2.4%
Girls 5 12.8% 3 7.7%
Latino 5 10.6% 3 6.4%
African-American 3 9.1% 2 6.1%
Reynolds Adolescent Depression Scale
Young adolescents reported a mean total score o f64.32 (SD = 13.37). Fourteen
(17.5%) scored at or above the cut-off score of 77, thus meeting the criteria for
depression. O f the students with depression, 42.9% also met the criteria for PTSD (N =
6). This percentage signals the presence o f substantial comorbidity in symptomatology.
Gender and ethnic differences exist with regards to depression but failed to
reach statistical significance. For example, girls scored higher than boys on this measure
with a mean score o f 67.16 (SD = 14.14) compared to the boys’ mean score o f 61.65
(SD = 12.22). A t-test revealed that this difference is not statistically significant (t =
1.69; p = .09). Similarly, African-American students had higher depression scores than
Latino students (M = 64.59; SD = 12.31 compared to M = 64.11; SD = 14.32 for
Latinos), but this difference does not attain statistical significance (t = .146; p = .89).
MASC
The mean total score on this self-report instrument is 38.24 (SD = 15.19).
Although the mean total score for Latinos (39.95; SD = 13.75) is higher than for
77
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
African-Americans (35.61; SD = 17.10), no statistically significant differences were
noted (t = 1.121; p = .27).
Congruent with other research findings, girls are more anxious than boys. The
mean total score for girls was 42.49 (SD = 16.31) while the mean total score for boys
was 33.62 (SD = 12.54). T-test analysis confirms that this difference is statistically
significant (t = 2.580; p < .05).
Five students, one o f whom was a boy, scored in the clinical range (above 65).
This number represents 6.3% o f the children’s sample.
K-SADS-E rPTSD Module!
Using this structured diagnostic interview, nine students (11.3%) met the full
criteria for PTSD. Only two African-Americans, both girls, met the criteria compared
to seven Latinos, five o f whom were boys.
A lack o f concordance exists between the two measures o f child PTSD. Only
four students who met PTSD criteria according to the K-SADS scored in the moderate
or above range based on the PTSD-Reaction Index Revised. In addition, the gender
composition based on the K-SADS is almost the exact inverse of the gender
composition o f those who met PTSD criteria according to the Reaction Index-Revised
instrument.
Hypothesis O ne: Children exposed to elevated levels o f community violence based on
self-report will manifest elevated distress symptomatology scores.
Testing this hypothesis, a regression equation was conducted. It affirms that
higher child total exposure scores predict child PTSD based on self report, B = .432;
78
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
R square = .233; p < .01 and child total exposure scores predict child depression, B =
.415; R square = .118; p < .01. However, child total exposure scores failed to predict
child anxiety, B = .170; R square = .025; p = 6.01.
Controlling for child’s age and ethnicity, regression analysis reveals that 35% o f
the variance in the entire sample o f children’s PTSD scores is due to child total
exposure scores (R square = .264; p < .01) and gender (R square = .082; p < .05).
Regression analysis reveals noteworthy differences in child PTSD based on
ethnicity. Among African-Americans, 29% o f the variance is due to mothers’ education
and 18.8% is due to child-self-reported exposure. Among Latinos, child self-report
exposure explains 27.3% of the variance in child PTSD.
Child Violence Exposure Scores— Maternal Report
Child Exposure to Community Violence
At the beginning of the PACE questionnaire, the mothers responded to two
questions: Has your child ever been a victim o f violence? and has your child ever
witnessed an act of violence? Based on their responses, 12.5% o f their children have
been victims of violence and one in four children have been witnesses of violence in
their lifetime (25%). However, as the mothers completed the entire PACE
questionnaire, they reported substantially higher frequencies of child actual exposure
than previously indicated. In actuality, the mothers reported that 24 children have been
victims o f violence (30%) and 47 have been witnesses to community violence in their
lifetime (58.8%). O f those who were victims, 45.8% were victimized within the past
year. Although 23.8% of the youngsters had high levels o f lifetime exposure to
79
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
community violence, 36.5% o f the young adolescents had no exposure at all and 65.1%
had 2 or fewer total exposures. The mean total child exposure was 1.92 (SD = 2.14).
The most frequent occurrences o f victimization were being threatened with physical
harm (17.5%) and being beaten (15%). Of those threatened, all but one had been
threatened within the past year. Ten percent had been threatened on more than two
occasions. O f those beaten, ninety percent suffered the beating within the past year.
The most common perpetrator was a schoolmate (75%). Ten percent o f the children
had been robbed (N = 8). Two children were victims o f sexual abuse. Each child had
been abused more than one year ago, either by a father or by a friend of the family.
Only one child was ever shot at and only one child had ever been stabbed.
The most frequent type o f violence witnessed was seeing another person get
beaten (38.8%). The vast majority of this exposure occurred within the past year
(73.8%). While 17.5% of the students have witnessed a beating once or twice, 15%
have witnessed a beating on more than two occasions. The most common relationship
to the victim was that o f a friend (37.5%) or a schoolmate (37.5%). Overall, five
percent o f the children have seen a family member or relative get hurt by violence.
Numerous children have seen someone threatened with physical harm (18.8%)
with the vast majority o f these threats occurring within the past year (70%). Eleven
percent of the children (11.2%) have seen a dead body, not at a funeral, and 26.3% have
heard about a dead body.
Mothers reported that 47 children have heard gunshots near their home (58.8%)
and 12 have heard gunshots near their school (15.6%). One in five o f children have
80
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
seen a real gun. For purposes o f analysis, the category o f hearing gunshots was not
included. This more conservative approach was taken to ensure that the total frequency
o f exposure would not be overinflated.
According to the mothers, 33.8% o f the children experience fear of being hurt at
school. Twenty percent are afraid of being hurt at home and 20% fear being hurt in
their neighborhood.
In spite o f the high levels o f child exposure reported by the mothers and the
mothers’ perceptions o f their children’s fear, mothers often failed to discuss these
experiences with their children. Thirty-five percent o f the mothers (N = 28) reported
that they have never talked with their son or daughter about community violence and it
effects. Unfortunately, this is not uncommon. Parents are often hesitant or unsure o f
how to explain the reasons for violence or how to explain death or severe injury to their
child (Jackson, 1994).
TABLE 6. Maternal Report on Child Violence Exposure
Prevalence Rates % (0=47) (n=33)
Violence____________Number Victim Witness Latino African American
Threatened with harm 14 17.5%
---------
12.8% 24.2%
Beaten 12 15.0%
---------
11.9% 21.2%
Robbed 8 10.0%
---------
7.1% 15.2%
At home during burglary 7 8.8%
---------
14.3% 3.0%
Choked 5 6.3%
---------
0.0% 15.2%
Threatened with weapon 2 2.5%
---------
2.1% 3.0%
Sexually abused
2
2.5%
---------
2.1% 3.0%
Attacked with knife 1 1.3%
---------
2.1% 0.0%
Shot at I 1.3%
---------
0.0% 3.0%
Seen someone beaten 31 38.8% 29.8% 51.5%
Seen threat with harm 15
----------
18.8% 19.0% 18.2%
Seen dead body 9
----------
11.3% 14.9% 6.0%
Seen family get hurt 4 5.0% 7.1% 3.0%
Seen forced entry 2
----------
2.5% 2.4% 3.0%
Seen someone shot at 2 2.5% 4.3% 0.0%
81
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
In the study’s sample, the concordance between child and mother report o f the
child’s exposure to community violence varied substantially. Over half of the mothers
reported frequencies o f child’s total exposure to community violence lower than
reported by their children (55%). Thirty percent reported frequencies of total exposure
higher than indicated by their children while 15% reported the same total exposure.
Child Symptomatology Scores—Maternal Report
Child Behavioral Checklist
The mean CBCL Total T-score of the study’s young adolescents (50.77; SD =
12.92) is very consistent with the normed mean T-score of 50. Twenty-five percent of
the sample scored at or above the borderline score of 60 while twenty percent o f these
youngsters also scored in the clinical range (score above 63).
Based on gender, no statistically significant differences were found. The mean
Total score for girls was 51.07 (SD = 11.95) and 50.50 (SD = 13.90) for boys.
Further examination reveals gender differences among those who scored at or
above the borderline range. Compared to girls, a higher percentage of boys (29.4%)
scored at or above the borderline Total score compared to girls (20%). These
percentage differences were not statistically significant, however.
Investigation o f possible ethnic differences revealed that Latino children scored
slightly higher than African-American children in the CBCL Total Score. The mean
score for Latino students was 50.98 (SD = 13.39) while for African-American students
it was 49.05 (SD = 11.70). This difference was not statistically significant.
82
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Hypothesis Two: Children exposed to higher levels of community violence based on
self-report will manifest higher total behavioral problem scores.
Examining the relationship between child self-reported exposure scores and
child total behavior problem scores, regression analysis fails to support the
hypothesized relationship. Regressing child exposure on the Total behavior problem
score, B = -.014; R square = .000; p = .92.
Based on maternal reports of child’s exposure to community violence, however,
hypothesis two is affirmed. As noted in Table 7, controlling for mother’s education in
years and family income, mothers’ reports account for 13% of the variance in children’s
Total behavior problem score.
TABLE 7. Standardized Regression Coefficients for Measures of Child Behavior Problems
Factors Standardized Betas
Background characteristics
Mother’s Education -.102
Family Income -.185
R square change = .056
Mother’s Report on Child’s Exposure .383**
R square change = .131
F statistic for Model = 4.07*
Total R square change = .187
Adjusted R square change = .141
Conners Parent Rating Scale-Revised
This measure’s mean total score was 45.54 (SD = 43.00). The mean
T-scores of its subscales were: Oppositional (53.26; SD = 13.90), ADHD (55.14;
83
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Further univariate analysis reveals that one in five mothers had been victims of
violence without any other type o f exposure while 21.3% were solely witnesses to
violence. Overall, 23.8% o f the mothers had no exposure of any kind to community
violence while some had only one lifetime exposure (17.5%) or two exposures (20%) to
violence. Thirty-two mothers (40%) reported high levels o f exposure to community
violence (3 or more total exposures) with nearly all of these mothers (N = 28) reporting
exposure to different events of community violence on 4 or more occasions.
An examination o f the frequencies o f exposure to discrete events in Table 8
revealed that 28.8% of the women have been beatened. All such instances occurred
more than one year ago. Six children witnessed the beating (7.5%). Nearly one of
every five mothers (18.8%) has been a victim of robbery. Most of these robberies took
place more than a year ago (85%). A similar percentage of mothers (18.8%) have been
threatened with serious physical harm, with the vast majority of these threats occurring
over a year ago (85%).
Many o f the women in this study were victims of sexual abuse. Initially, 12.5%
of the mothers reported being a victim of sexual abuse as an adult and 21.3% reported
being sexually abuse as a child. Based on information provided in the instrument’s
qualitative question, however, the revised percentage of mothers who have been an
adult victim of sexual abuse is 21.3%. For those who were sexually abused as a child,
the revised percentage is 26.3%. Congruent with the literature, present findings indicate
that being a victim o f sexual abuse as a child heightens the risk o f being a victim of
abuse later in life. Of those women who were victims of sexual abuse as a child, two-
thirds were also victims o f sexual abuse as an adult.
85
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Six women have been victims of domestic violence (7.5%). However, with the
exception o f one mother, they have been exposed to other forms o f violence as well.
Over twelve percent (12.5%) were also victims o f sexual abuse.
Not only were a substantial number of mothers victimized in their lifetime but
over half have witnessed acts o f community violence. Nearly four o f every ten women
have seen someone beaten up (38.8%). Further indicative o f the violent milieu in which
the mothers reside, 20% have seen someone shot at, 70% have heard gunshots and
17.5% have seen a dead body.
For purposes of analysis, once again the more conservative approach was taken
in tabulating total frequency o f exposure. Thus, domestic violence exposure and
hearing gunshots were excluded. The mean total exposure was 2.83 (SD = 2.83). The
mean total for African-American mothers (3.61; SD = 3.37) is substantially higher than
Latina mothers (2.29; SD = 2.28). A t-test was conducted to examine possible ethnic
differences with regards to total lifetime exposure to community violence.
Homogeneity o f variance was tested and demonstrated. Analysis revealed that African-
American mothers have significantly more total exposure compared to Latina mothers
(t = 2.096; p < .05). However, in terms o f witnessing acts o f community violence,
Latinas saw more violence (M = 1.56; SD = .50) than African-American mothers (M =
1.27; SD = .46). This difference is statistically significant (t = 2.227; p < .05).
Fear is a reality in the lives of the mothers. Sixteen percent reported feeling
afraid at work, 28.8% are fearful at home, and 51.3% are afraid in their neighborhood.
In addition, a significant number o f mothers experience fear o f being hurt. Sixteen
86
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
percent indicated that they are afraid that someone might hurt them in their home and
32.5% are afraid o f being hurt in their neighborhood.
Less than one half o f the mothers (43.8%) reported having talked to someone
about their feelings related to their exposure to violence. Family members (30%) or
ministers (7.5%) were the principal sources o f contact. Twenty-seven mothers
indicated that they would like the opportunity to talk to someone about their
experiences related to violence (33.8%).
TABLE 8. Maternal Lifetime Exposure to Discrete Violent Events
Prevalence Rates % (n=47) (n=33)
Violence______________ Number Victim Witness Latina African American
Beaten 23 28.8%
--------
19.1% 42.4%
Sexually abused as child 21 26.3%
--------
23.4% 30.3%
Sexually abused as adult 17 21.3%
--------
17.0% 27.3%
Robbed 15 18.8%
--------
21.3% 15.2%
Threatened with weapon 14 17.5%
--------
14.9% 21.2%
At home during burglary 11 13.8%
--------
5.0% 21.2%
Choked 8 10.0%
--------
4.3% 18.2%
Domestic violence 6 7.5%
--------
4.3% 12.1%
Attacked with knife 5 6.3%
--------
4.3% 9.1%
Shot at 1 1.3%
--------
0.0% 3.0%
Seen someone get beaten 31
..........
38.8% 29.8% 51.5%
Seen someone shot at 16
--------
20.0% 14.9% 27.3%
Seen dead body 14
--------
17.5% 17.0% 18.2%
Seen someone choked 8
--------
10.0% 2.1% 21.2%
Seen forced entry in house 7
--------
8.8% 8.5% 9.1%
Seen family member hurt 4
--------
5.0% 4.3% 6.0%
Seen someone hurt at home 2
--------
2.5% 4.3% 0.0%
Heard gunshots near home 56
--------
70.0% 70.2% 69.7%
Maternal Symptomatology Scores—Self-Report
Brief Symptom Inventory—Depression and Anxiety subscales
The BSI has established community norms for females with T-scores at or above
63 determining clinical caseness. The mothers’ mean T-score for depression was o f
87
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49.95 (SD = 9.59) and 51.15 (SD = 11.30) for anxiety. Sixteen percent o f the mothers
reported elevated depressive symptoms above the cut-ofif score, a rate higher than the
rate of the general population. Twenty percent o f the mothers scored above the cut-ofif
score for anxiety.
African-American mothers scored higher on each subscale than did Latina
mothers. Their mean depression score was 51.82 (SD = 10.94) compared to 48.74
(SD = 8.50) among Latinas. For anxiety, African-American mothers reported a mean
score o f 52.03 (SD = 10.83) while the mean score for Latina mothers was 50.30 (SD =
11.62). T-tests revealed that these differences were non-significant (depression: t =
1.420; p = .16; anxiety: t = .323; p = .74).
Posttraumatic Diagnostic Scale
Research has determined a cut-ofif score of 21 to indicate moderate to severe
PTSD (Foa, 1995). In the present study, the mothers’ mean total symptom severity
score is 13.30 (SD = 10.84). This score falls in the mild range. While seventy percent
of the mothers scored in the mild to moderate range of PTSD, 24 met the criteria for
PTSD (30%). Although Latinas scored higher (14.13; SD = 10.43) on average than
African-American mothers (11.88; SD = 11.68), these differences in mean total score
were not significant (t = .891; p = .38).
For those mothers who scored a high level o f exposure, 51.3% met the PTSD
criteria. In contrast, among the mothers with low level of exposure only 16.3% met the
PTSD criteria. Among those mothers who experienced a high level of exposure to
community violence, ethnic differences are significant. Latinas had a much higher
88
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
mean total score (18.13; SD = 10.51) than African-American mothers (8.50; SD =
12.76) among this subsample. A t-test confirms that this difference is statistically
significant (t = 2.093; p < .05).
Stepwise regression analysis reveals that mother’s depression (R square = .286;
p < .01) contributes to 29% o f this variance in her PTSD score. This percentage points
to the high level o f comorbidity between depression and PTSD that is consistently
reported in the literature.
Thirty-nine percent o f the mothers with PTSD also scored in the clinical range
for depression (38.8%). Conversely, seventy-one percent o f the mothers meeting the
criteria for depression also met the criteria for PTSD (70.8%).
Structured Clinical Diagnostic Interview
Using this structured interview, which is considered the “gold standard” for
assessment o f clinical PTSD, 19.4% of the mothers met the diagnostic criteria. While
this percentage is slightly lower than the percentage found using the PDS instrument, it
is understandable due to the fact that the SCID is more stringent in its evaluation of
PTSD. In the present investigation, there exists a fairly strong correlation between
these two instruments (.558; p < .01) which suggests that the PDS has a good overall
level of diagnostic agreement with the SCID. Further analysis supports this finding as
41.4% (N =10) of those who met the PTSD criteria based on Foa’s instrument also were
identified as having PTSD based on the SCID.
Trauma Symptom Checklist
The mean total score for this self-report measure is 24.35 (SD = 18.79).
Mothers’ mean scores for the anxiety and depression subscales are as follows: 6.05
89
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(SD = 4.98) and 6.27 (SD = 4.95). Mothers reported a mean score o f 3.63 (SD = 3.48)
for the sexual abuse trauma index.
African-American women scored higher in each category with the exception of
the sexual abuse trauma index. The mean total score for African-American mothers
(27.14; SD = 21.64) was higher than the score of Latinas (22.93; SD = 17.25). For the
anxiety subscale, African-American women had a slightly higher mean score (6.62; SD
= 4.81) than Latina women (5.76; SD = 5.10). For the depression subscale, African-
American mothers reported a higher mean score of 7.33 (SD = 6.00) compared to Latina
mothers’ mean score of 5.73 (SD = 4.30). Only on the sexual abuse trauma index did
Latina women score higher (M = 3.80; SD = 3.38) compared to African-American
women, (M = 3.29; SD = 3.74). T-tests failed to reveal any significant differences for
total score (t = .792; p = .43) or for the various subscales.
Tolman’s Psychological Maltreatment of Women Inventory
While cutoff scores have yet to be established for Tolman’s instrument, 37% of
the study’s sample of mothers reported a total score o f 80 or above. The mean score
was 82.43 (SD = 27.31) with scores ranging from 58 to 173. This suggests that many
women may have experienced psychological maltreatment, which is often associated
with physical abuse. The mean scores o f African-American women (82.65; SD =
25.39) were slightly higher than the scores Latina women (82.2; SD = 28.75).
However, these scores fail to demonstrate statistically significant differences (t = .046;
p = .96).
90
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The Stability o f Self-Esteem Measure (Rosenberg. 1965’ )
A higher score indicates a person’s lower self-esteem on this measure. The
mothers’ mean score was 16.29 (SD = 5.43). Nearly twenty-two percent (21.3%) o f the
mothers scored at or above the cut-off score of 20, indicating that more than one in five
mothers experienced low self-esteem. T-test analysis revealed no significant
differences due to ethnicity (t = 1.483; p = .15).
Hypothesis Three: Mothers exposed to elevated levels o f community violence will
manifest elevated levels of distress symptomatology based on
self-report.
Regression analysis confirms the predicted relationship, namely, mothers
exposed to high levels o f community violence manifest higher levels o f distress
symptomatology. Mothers’ total exposure score accounts for eight percent of the
variance in her PTSD score (R square = .076; p < .05). Higher violence score predicts
twenty percent o f the variance in maternal depression score (R square = .196; p < .01)
and fifteen percent o f the variance in anxiety score (R square = .148; p < .01).
Overall, mothers’ exposure to violence and her depression and anxiety scores
explain 32% o f the variance in her PTSD scores.
Hypothesis Four: Higher scores of maternal distress symptomatology predict higher
scores of children’s PTSD or total behavior problems.
For the most part, this hypothesis was supported. For example, maternal distress
(Z-score) accounts for 11% of the variance on child PTSD (R square = .107; p < .01).
Maternal distress symptomatology predicted 27% o f child behavior problem score
(R square = .272; p < .01). However, when regressed on child depression score,
maternal distress score was not significant (R square = .000; p = .952).
91
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Regression analysis found that mothers’ total exposure score (R square = .230;
p < .01) and their anxiety T-score (R square = .147; p < .01) explained 38% of the
variance in externalizing behavior scores. Simultaneous entry revealed that overall
maternal distress symptomatology (R square = .240; p < .01) and her violence exposure
score (R square = .082; p < .05) explain 32% of the variance in children’s CBCL total
score.
Taking into account ethnicity, Latina mothers’ exposure to violence score
(R square = .319; p < .01) and their anxiety T-score (R square = .192; p < .01) explained
51% of the variance inchildren’s externalizing behavior scores. However, no such
statistically significant relationship exists for African-American mothers.
Child Behavior Scores—Teachers’ Reports
Hightower Teacher-Child Rating Scale
The English teachers o f the young adolescents reported a mean score of 30.47
(SD = 11.11) for Part One o f this instrument which focuses on the students’ acting out
behaviors and learning problems. They reported a mean score o f 62.35 (SD = 21.57)
for Part Two which focuses on tasks and sociability. The mean scores for the various
subscales are as follows: Part One: Acting Out (9.27; SD = 5.20), Shy/Anxious (8.99;
SD = 3.87), and Learning Problems (12.44; SD = 6.71); Part Two: Frustration/
Tolerance (14.92; SD = 5.71), Assertive Social Skills (17.40; SD = 5.73), Task
Orientation (16.36; SD = 7.71), and Peer Sociability (13.66; SD = 5.42).
Correlations among the subscales in Part One varied substantially. The subscale
of Acting Out was significantly correlated to the Learning Problems subscale (.591;
92
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
p < .01). However, the shy/anxious subscale was not correlated to either o f these two
subscales.
Among the four subscales in Part Two, each was highly correlated with the
others (.503 to .834; p < .01). In addition, all the four subscales were strongly
correlated to child learning problems, for example, from .560 for Assertive Social Skills
to .883 for Task Orientation.
Mean differences were noted for ethnicity and gender, however, none proved to
be statistically significant. African-American students scored higher on average than
Latino students on the subscales reflective o f problematic behavior: Acting Out,
Frustration /Tolerance, Task Orientation, and Learning Problems. Latino students
scored higher in the subscales that possess relational qualities: Shy/Anxious, Peer
Sociability and Assertive Social Skills.
This pattern was replicated when contrasting boys’ and girls’ scores on
Hightower’s instrument. Boys had higher mean scores in the areas of Acting Out,
Frustration/Tolerance, Task Orientation, and Learning Problems whereas girls scored
higher on average in the areas o f Shy/Anxious, Peer Sociability, and Assertive Social
Skills.
Conners Teacher Rating Scale
English teachers completed the Conners’ Scale on the students. They reported
the following mean T-scores for these subscales: Oppositional—51.87 (SD = 11.35);
Hyperactivity— 54.46 (SD = 13.60); ADHD—55.16 (SD = 13.64); PTSD—55.88 (SD =
13.62); and Global Index Total—53.99 (SD = 12.39).
93
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Overall, the teachers' mean total T-scores are lower than those reported by the
mothers in nearly all subscales. However, there are two exceptions. For the ADHD
subscale, the teachers reported a mean T-score of 55.16 (SD = 13.64), the same as the
mothers’ score o f 55.14 (SD = 14.50). Likewise, the teachers reported the same mean
T-score (55.88; SD = 13.62) as did the mothers (55.84; SD = 14.84) for the PTSD
subscale.
According to the teachers’ rating, 12.5% of the students scored above the cutoff
score o f 65 for Oppositional, 25% o f the students scored in the clinical range for
ADHD, 17.5% met the cutoff score for Hyperactivity, 23.8% met the criteria for PTSD,
and 18.8% scored in the clinical range for Global Total Score.
As in the case o f the mean T-scores, the percentage o f students reported by the
teachers to be in the clinical range is lower than reported by the mothers with the
exception of the ADHD and PTSD subscales. While the mothers reported that 20% of
the children scored in the clinical range for ADHD, 25% o f the students fell into the
category according to the teachers. In addition, according to the mothers, 22.5% of
their children met the clinical cutoff score for PTSD while the teachers reported that
23.8% of the children met the clinical criteria score.
Correlations
Pearson correlations between the broad scales o f the clinical outcome measures
and the exposure to community violence measures were computed in order to ascertain
how the different measures relate to each other. In many instances, only modest to
moderate correlations were found. Please see Appendix for all correlations.
94
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Correlations o f Child and Mother Violence Exposure Scores with
Symptomatology
The correlation between the student self-report of violence exposure, the
mothers’ self-report, and the mothers’ report on the child’s exposure reveals interesting
results. The mothers’ report on child’s exposure and the child self-report are not
significantly correlated (.118; p > .37). Nor are the mother and child self-reports
correlated (.029; p > .82). However, mothers’ report on child’s exposure is significantly
correlated with the mothers’ total exposure score (.633; p < .01).
Child’s self-report total exposure score is moderately correlated with the PTSD-
Reaction Index Revised score (.432; p < .01) and with the RADS (.415; p < .01). It is
not significantly correlated with the MASC or any parent-reported measure o f
symptomatology.
Mothers’ report on child exposure is significantly correlated with child’s total
behavior problem score on the CBCL (.269; p < .05) and with the subscales scores o f
Conners’ measure.
Mothers’ total exposure is significantly correlated with the CBCL Total score
(.342; p < .01) and with the Conners’ subscales scores.
Intercorrelations Between Measures
The MASC is not significantly correlated with any other measure o f child
symptomatology. For example, the correlation between the MASC score and the RADS
score failed to attain statistical significance (.232; p > .06). In addition, the MASC
scores are not significantly correlated with the PTSD Reaction Index-Revised score
(.143; p > .27).
95
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The PTSD Reaction Index-Revised score is moderately correlated with the
RADS score (.408; p < .01) as well as the maternal depression score (.352; p < .01) and
maternal anxiety score (.262; p < .05).
Among the mothers, the depression and anxiety subscales scores as measured by
the BSI and the self-esteem scores measured by the Rosenberg instrument are all
significantly correlated. The r ’s are as follows: depression to anxiety (.718; p < .01),
depression to self-esteem (.517; p < .01), and anxiety to self-esteem (.468; p < .01).
Also, the depression and anxiety subscales o f the BSI correlate significantly with Foa’s
PTSD score: depression (.520; p < .01) and anxiety (.493; p < .01).
All o f the Conners’ subscales based on maternal report are strongly correlated to
the CBCL Total Score with r ’s ranging from .724 to .794 (p < .01). In stark contrast,
none of the Conner’s subscales completed by the teachers are significantly correlated
with the CBCL Total score and on average the r ’s are quite low (.050).
None o f the scores of the subscales between the Parent and Teacher versions of
the Conners instrument are significantly correlated.
Most o f the Hightower scores reported by the teachers were significantly
correlated with Conners’ Teacher scores. For example, the acting out subscale o f the
Hightower instrument strongly correlated with all subscales o f Conners’ Teacher
instrument with r ’s ranging from .777 to .838. Likewise, the learning problems
subscale strongly correlated with the various subscales o f the Conners’ instrument with
r’s ranging from .625 to .744. However, no scores on the Hightower measure were
significantly correlated with any Conners’ Parent-rated scores.
96
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The opposite occurred with the CBCL Total Score and the Conners’ instrument.
Whereas the Total Score was highly correlated with the various subscales of the
Conners’ Parent instrument no significant correlation was found among the Total Score
and the subscales scores o f the Teacher version of Conners’ instrument.
The total score o f Tolman’s instrument correlates significantly with the BSI
depression score (.314; p < .05) and with the self-esteem score from Rosenberg’s
measure (.335; p < .05).
The TSC-40 total score as well as its Sexual Abuse Trauma Index subscale are
significantly correlated with various other measures. The total score enjoys a strong
correlation with the PTSD score from Foa’s measure (.688; p < .01) and a modest
correlation with Tolman’s total score (.284; p < .05). The Sexual Abuse Trauma Index
is moderately correlated with Foa’s PTSD score (.393; p < .01).
Maternal Symptomatology Correlations
O f particular interest is the association of mothers’ exposure to violence and her
symptomatology scores. Mothers’ self-reported exposure to community violence is
modestly correlated with Foa’s PTSD score (.276; p < .05) and with the TSC total score
(.395; p < .01) and the Sexual Abuse Trauma Index (.275; p < .05). It is moderately
correlated with mother’s depression score (.443; p < .01) and her anxiety score (.385;
p < .01). Also, it is significantly correlated with maternal distress symptomatology
z-score (.431; p <.0l). Mothers’ total violence exposure score is modestly correlated
with her level of education (.280; p < .05) and her ethnicity (.278; p < .05).
97
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Significant correlations are found among various demographic items as well.
Mothers’ race/ethnicity is significantly correlated with her level o f education (.626;
p <.01); fathers’ employment (.413; p < .01) and fathers’ education (.449; p < .01) and
family income (.349; p <.01). Mothers’ level o f education correlates significantly with
family income (.316; p < .05) and with maternal self-esteem scores (.-331; p < .01).
Fathers’ education level correlates significantly with family income (.498; p < .01) as
well as with mother’s exposure to violence (.281; p < .05). Fathers’ employment is
moderately correlated with maternal depression (.325; p < .05). Marital status
correlates modestly with maternal self-esteem (.311; p < .05) and with sexual abuse
trauma scores (.313; p < .05).
Moderating Effects
Procedures outlined by Aiken and West (1991) were used to test if maternal
distress symptomatology is a moderator of child PTSD. Standardized scores were used
for analysis.
Child total exposure to community violence, maternal distress symptomatology,
and their interaction term were entered hierarchically as predictors o f scores for child
PTSD (See Table 9). Each effect was tested while partialling out the effects o f equal
and lower order. Scores on total child exposure to community violence (R2 change =
.231; p < .01) and maternal distress symptomatology (R2 change = .064; p < .01) were
significant as main effects, accounting for 30% o f the variance in scores on child PTSD.
However, no significant interaction between scores on total child violence exposure and
maternal distress symptomatology was found (R2 change = .002; p = .63). Thus,
98
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
maternal distress symptomatology is not a moderator and does not amplify or dampen
the impact o f exposure to community violence upon the child’s PTSD.
TABLE 9. Measures o f Moderator Effects
Measure Child Total Behavior Problem
Change in R Square F change Semipartial
Correlation
Child Exposure to Violence .231 21.628** .481**
Maternal Distress Symptomatology .064 5.538* .258*
A X B .002 -.070
R Square Change .297 .236
* p<.05; ** p<.01
The same procedures were followed to ascertain a moderating relationship based
on a variety of other variables. No moderating relationship for any variable was found.
For example, analysis was conducted to determine if African-American mothers’ years
of education served as a moderator of child PTSD. However, non-significant findings
resulted. Similarly, maternal distress symptomatology did not act as a moderator of
child’s self-reported exposure to community violence upon child’s PTSD. Also,
maternal distress symptomatology failed to act as a moderator o f child behavior
problems when using mother’s report of child violence exposure and child CBCL total
score. Scores on child’s violence exposure based on maternal report (R2 change =
.198; p < .01) and maternal distress symptomatology (R2 change = .225; p < .01) were
significant as main effects, accounting for 42% o f the variance in child’s CBCL total
score. However, after entering the interaction term, significance was not found between
99
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the scores on CBCL total and maternal distress symptomatology (R2 change = .005; p =
.62). The lack o f interaction signifies that child’s exposure to violence and maternal
distress symptomatology remain direct predictors of child behavior problems.
Mediation Effects
Hypothesis Five: Maternal distress symptomatology mediates children’s PTSD and/or
total behavior problem scores.
A series o f three multiple regression equations was computed to test for the
hypothesized mediator effects. The recommendations o f Baron and Kenny (1986) were
followed. In equation one, maternal distress symptomatology was regressed on child’s
violence exposure total scores (R2 change = .023; p < .19). This failed to attain
significance and thus maternal distress symptomatology does not act as a mediator of
child’s PTSD.
Other variables were examined to determine if they are mediators o f child PTSD
or behavior problems. With the exceptions of the neighborhood school and Latina
mothers reports on child, all other variables foiled to act as a mediator.
Focusing exclusively on the sample of children from the neighborhood school, a
mediating relationship was found as shown in Table 10. In equation one, maternal
distress symptomatology was regressed on child’s total exposure score based on
maternal report. In equation two, child’s total behavior score was regressed on child’s
exposure to community violence total score. In equation three, child’s total behavior
scores were regressed on both child total exposure score and maternal distress
symptomatology. The previously significant Beta coefficient (B = .47) for child
exposure was attenuated and no longer significant (B = .19; p = .10) while maternal
100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
distress was a significant predictor o f child total behavior problem scores (B = .49; p <
.01). Thus, the criteria were met and maternal distress was found to act as a mediator of
child total behavior problems.
TABLE 10. Standardized Regression Coefficients for Measures of Mediator Effects on
Child Behavior Problems based on Mothers and their Reports on Child at
Neighborhood School
Predictor Standardized Betas
Equation One:
Maternal Distress Symptomatology .55*
R squared change = .31
F change = 14.042*
Equation Two:
Child Total Exposure to Community Violence .47*
R squared change = .22
F change 8.258*
Equation Three:
Maternal Distress Symptomatology .49*
Child Total Exposure to Community Violence .19
*p<.01
Using Latina mothers’ report of their child’s total exposure and their CBCL total
behavior problem score, a significant mediating relationship was likewise found after
conducting another series o f regression equations. As Table 11 shows, in equation one,
maternal distress symptomatology was regressed on child’s total exposure score based
on mother’s report (R2 change = .272; p = < .01). In equation two, child’s total
behavior score was regressed on child's exposure to violence total scores (mother’s
report) (R2 change = .148; p < .02). In equation three, child’s total behavior problem
scores were regressed on both child total exposure and maternal distress
symptomatology scores. The outcomes reveal that maternal distress symptomatology
1 0 1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
was a significant predictor o f child total problem scores (B = .49; p < .01). The
previously significant relation between child’s exposure and behavior problem scores
was no longer significant (p = .39). In addition, the standardized coefficient for child
exposure in equation three was substantially less than in equation two (.385).
TABLE 11. Standardized Regression Coefficients for Measures of Mediator Effects on Child
Behavior Problems based on Latina Mothers and their Reports on Child
Predictor Standardized Betas
Equation One:
Maternal Distress Symptomatology
R squared change = .27
F change = 14.209**
Equation Two:
Child Total Exposure to Community Violence
R squared change = .15
F change 6.083*
Equation Three:
Maternal Distress Symptomatology
Child Total Exposure to Community Violence
*p < .05 **p<.01
Thus, each o f the criteria for a mediating relationship as outlined by Baron and
Kenney (1986) was met for this particular case. Whereas child exposure to community
violence continues to have an effect on child behavior problems, it has less a direct
effect than previously established (B = .141; p = .39). Therefore, in the case o f Latino
children as with the case o f all the students o f the neighborhood school, maternal
distress symptomatology is more important than exposure to community violence in
contributing to the heightened distress and behavior problems of the child.
102
.52**
.39*
.49**
.14
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter S
D ISCU SSIO N
The study provides valuable insight into the experience o f Latina and African
American mothers and their young adolescent children who collectively experience the
negative impact and on-going stress o f community violence. It highlights that both
mother and child have substantial exposure to community violence and experience
negative consequences. The study also renders important data regarding variability of
exposure and symptomatology when living in similar low-income neighborhoods with
high levels o f violence.
This study has several strengths, including gathering of comprehensive
information from multiple informants on mother and child exposure to community
violence and related distress symptomatology as well as the inclusion of a substantial
percentage o f Latino subjects. The findings extend our knowledge o f exposure to
community violence and PTSD symptomatology among Latinos whose adolescent
population is increasing more rapidly than that o f any other racial or ethnic group
(Ozzer, Brindis, Millstein, Knopf, & Irwin, 1997). The findings highlight the variability
o f exposure and symptomatology even among families residing in the same
geographical area.
Unique to this study is the examination o f the traumatic effects o f violence
exposure upon mothers and how it may affect their child. A major strength o f this study
is therefore the inclusion of mothers. By collecting data on Latina and African-
American mothers’ own levels of exposure to community violence, their traumatic life
events history, and their distress symptomatology, this study provides information
103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
crucial for the appropriate assessment and treatment o f an undeserved population. This
information contributes to a better understanding o f the mechanisms through which
neighborhood violence may lead to child PTSD and other distress symptomatology or
child behavior problems. Such data are currently absent from the literature. Typically,
family characteristics, such as discipline patterns, parenting practices, and family
constellations, are studied to determine if they act as moderators (Sheidow et al., 2001).
This study also contributes to the knowledge base by exploring the type o f
effects (moderator/mediator) of maternal distress symptomatology upon young
adolescent’s PTSD and behavior problems and specifying the relationship or pathway
of influence. No significant interaction between scores on child’s total exposure to
community violence and maternal distress symptomatology were obtained. Therefore,
maternal distress symptomatology is not a moderator in the relationship between scores
on child total violence exposure and child PTSD. In other words, maternal distress
symptomatology is not a mechanism that either amplifies or dampens the relationship
between child violence exposure and child PTSD.
Regression analyses revealed that maternal distress symptomatology does not
act as a significant mediator between scores of child total exposure to community
violence and child PTSD or child behavior problem total scores. However, for children
o f the neighborhood school as well as for Latino children at both schools, maternal
distress symptomatology acts as a mediator on child’s total behavior problem score
when using scores taken from mother’s report on child’s violence exposure. Thus, in
these two cases, the greater a mother’s distress symptomatology the more likely a child
exhibits increased behavioral and emotional problems.
104
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A possible reason why the findings of a mediating relationship occurs only
among the Latino children at both schools may be due to the substantial differences in
family income and education among Latino and African-American parents. The lack of
a mediating relationship for African-American children at the non-neighborhood school
suggests that their family income and parents’ education may serve as a protective
factor for the students attending that school. The opportunity for these students to build
relationships with those who reside outside their neighborhood also may contribute to
the study’s finding of a non-mediating relationship for the sample.
The findings o f significant mediating relationships o f maternal distress
symptomatology upon child total behavior problems based on mothers’ reports and not
based on children’s self-report measures points to a salient issue regarding divergence
between informants. In a study o f a clinical population of depressed youths, parent and
child reports differed in their reported levels of child depressive symptoms with a
consensus diagnosis of major depression made by clinicians (King et al., 1997). The
findings by King and colleagues revealed that only those diagnoses made according to
parental report were predictive o f a clinical diagnosis. However, it is not clearly known
if maternal bias due to her depressive symptomatology results in spurious rates of child
symptomatology (Mick, Santangelo, Wypij & Biederman, 2000). Reporter bias is a
possible contributing factor to the variance in reporting violence exposure. Young
adolescents may be prone to overreport their experiences in a large group format and/or
underreport their experiences to their parents. The present study’s findings o f a
mediating relationship highlight the clear need for further research to identify factors
that influence the reliability o f data collected, including reporter effects.
105
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Additional factors likely contribute to the divergence in reporting o f violence
exposure between child and mother. On the one hand, the lack o f concordance between
child self-report o f exposure to community violence and mother report of child’s
exposure is consistent with the literature that mothers generally underestimate their
child’s frequency of exposure to community violence (Gaensbauer, 1996). The lack o f
communication between parent and child regarding violence likely contributes to the
lack o f concordance. Many parents remain unaware or deny the multiple ways their
children can be exposed to and therefore traumatized by violence (Osofsky et al., 1993).
This is a serious issue. Children may fail to reveal their experiences of violence to their
parents.
While children may remember a specific trauma, they frequently experience
difficulty in discussing it and often refuse to acknowledge that which they have
previously described (Parker and Randall, 1996). Thus, recall bias or selective memory
may contribute to the underreporting o f exposure o f both children and their mothers.
Children and mothers may underreport violence exposure due to their life
experiences and perceptions o f what is threatening or harmful. Repeated exposure to
acts o f violence can dull one’s sense of its impact. Children and mothers whose day-to-
day reality includes exposure to violence are likely aware o f areas or circumstances that
are dangerous but minimize the threat or distance themselves from it in order to survive
both physically and psychologically. Thus, discordance found between mother and
child reports may be a function of differences in perception o f violent events. What a
parent may perceive to be violent or threatening may not be construed as such by a
child.
106
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
This is also an important consideration since a person’s perception of a given
event determines, in part, the level of stress experienced by that person (Lazarus, 1986).
One’s perception o f violence, rather than the frequency o f exposure, may be of
paramount importance for coping and functioning (Kuther, 1999).
Another contributing factor to the divergence is that the reporting o f violence
exposure may be a suppressor effect resulting from the administration o f the violence
measure to the students in a large group format. Students may have felt inhibited to
disclose despite the assurance o f confidentiality. Research findings indicate that
children often underreport exposure to community violence when they perceive that
privacy linked with the collection of data is less trustworthy (Turner et al., 1998).
Another factor that likely contributed to the low level o f mother-child agreement
in reporting exposure to violent events is the ordering of the items on the UCLA
Violence Survey Questionnaire. The items began with the most severe type o f exposure
and later inquire about witnessing of events. To bolster accuracy o f reporting, the less
stressful types o f exposure should be placed at the beginning o f the measure. Also,
instead o f beginning with the violence questionnaire it may be more appropriate to
begin with a potentially less stressful measure such as a depression measure and
introduce the violence measures later on in the interview process. More accurate
reporting may occur when rapport with the subject is more fully established and the
subject feels more comfortable in participating. This re-ordering o f items and measures
should also be applied to the mothers.
The study’s findings, while not clearly supporting the hypothesized mediating
relationship o f maternal distress symptomatology for the entire sample, do underscore
107
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the importance to assess the caregiver’s exposure to community violence not only in
future research but also in the treatment o f children for mental health and behavioral
issues (Aisenberg & Mennen, 2000). By assessing the mother’s own exposure as well
as their distress reactions, mental health practitioners can better determine any increased
risk for middle school aged children for distress symptomatology.
The study’s results alert us to the need to include parents in the treatment
process. Whereas violence prevention and treatment efforts predominantly focus on the
child, the finding o f a mediating function of maternal distress symptomatology reveals
the importance to intervene with the parent. This indirect effect of maternal distress
symptomatology suggests that successful intervention with the parent or caregiver may
diminish the distress the child experiences as well as the behavior problems or
symptomatology o f the child.
The study also points to the importance o f obtaining qualitative information
from the respondents as to what they deem to be the most upsetting experience of
violence during their lifetime. Rarely is such information gathered from the perspective
of the respondent. In some cases, respondents may report that witnessing the beating of
a friend is the most upsetting or distressful event even if they reported witnessing the
objectively more serious act o f a stabbing or shooting. It is crucial to gather
information on the respondent’s perception of the event and the severity o f its impact in
the person’s life.
The study highlights the appropriateness o f use of multiple measures to gain
insight into adherence to diagnostic criteria for symptomatology. For example, the
108
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
identification o f fewer students as meeting PTSD by the K-SADS is reasonable due to
its more stringent assessment o f the PTSD compared to the Reaction-Index Revised
measure. The lack of agreement in specifying students with PTSD between the
measures is noteworthy. Several factors may contribute to the divergence in
identification o f students with PTSD as measured by the K-SADS and the Reaction
Index-Revised instruments. One, adherence to the strict diagnostic criteria may be
problematic due to the fact that it may lack developmental sensitivity to appropriately
assess young adolescents. Two, the Reaction-Index may lack necessary specificity to
reliably distinguish between young adolescents with subthreshold symptomatology and
those who meet clinical criteria.
It should be noted that while the sociodemographic variables, parental education
and family income, fail to attain statistical significance as moderators o f child distress
and behavior problems, they likely contribute to enhanced coping resources among
African-American families. Further research with a larger sample size is warranted to
ascertain if years of education and family income serve as moderators o f child’s distress
and behavior problems and may mitigate some of the harmful effects o f traumatic
exposure to community violence.
There are six principal limitations that should be considered when interpreting
the data. First, the findings are not generalizable to other populations. The sample
consisted of low-income Latino and African-American families living in a somewhat
violent neighborhood. Families from divergent SES backgrounds as well as from less
violent neighborhoods should be included in future research to explore similarities and
differentials in exposure and symptomatology.
109
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
contributed to the lack o f expected correlations. Students were more likely to be
victims o f violence compared to their mothers and they were more likely to be recently
exposed to violence. Findings reveal that a much higher percentage o f students had
exposure within the past year than did their mothers. The vast majority o f the mothers’
exposure to community violence occurred over one year from the time of participating
in the study. As a result, both children and mothers have engaged in coping patterns
over extended periods o f time. Thus, the children may have already experienced
moderating effects o f their mothers’ symptomatology or education and engaged in
coping responses which make would diminish their own level o f distress
symptomatology.
The low correlations between the same constructs from the perspective of
different sources in the use o f the Conners’ instruments are noteworthy. Although the
factors o f each version of the instrument correspond to each other, the Parent and
Teacher reported scores are not significantly correlated with each other. This is not an
unusual finding in the literature, however. In their study, Chilcoat and Breslau (1997)
found a small but significant association between maternal psychopathology and rating
differences between mother and teachers. Thus, one possible reason for the non
significant correlation in the present study is that elevated rates o f maternal distress
symptomatology may contribute to the lack of concordance discovered between
mothers and teachers scores on Conners’ instrument.
Another explanation is that child adjustment is accurately assessed by the
mothers and teachers but the divergence in scores reflects the different contexts (home
and school) in which the child behaves and interacts. It may be that each informant
til
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
symptomatology. To ask which trauma caused PTSD may be reductionistic and
inappropriate given the multivariate methodology used in this study (Weathers &
Keane, 1999). Nevertheless, this study contributes to the knowledge base by gathering
detailed information through the use o f symptom checklists assessing for child and
maternal exposure to sexual abuse and domestic violence as well as assessing for
distress symptomatology. It is important for further research efforts to try and
disentangle and specify the relationship of exposure to community violence and distress
symptomatology as it is crucial to try and disentangle the effects of community violence
exposure from the effects of domestic violence.
Areas for Future Research
The study signals that further research on the dynamics o f the interdependent
processes engaged in by parents and their children coping with chronic community violence
is warranted. The findings and recent literature suggest that to adequately address the needs
of the child exposed to the cumulative effect of chronic community violence requires that
researchers and practitioners address the needs o f the parent who also experiences the
detrimental effects o f community violence. This requires a fundamental shift from
primarily focusing on the resiliency of the individual child to the resiliency o f the family
(Walsh, 1996).
Whereas interventions often target the child alone, this study supports a
fundamental shift from focusing primarily on the resiliency of the individual child to
focusing on the resiliency o f the family in order to address the collective needs o f
families suffering the traumatic effects of exposure to community violence.
Intervention and treatment should include the family.
113
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
There exists a paucity o f longitudinal research on the developmental effects of
community violence in the lives o f children (White et al., 1998). Child behavior
problems and clinical distress symptoms associated with the trauma o f community
violence exposure have immediate consequences that may influence a young
adolescent’s school performance. In addition, they may have a negative impact on the
youngster’s cognitive development, school behavior and performance as well as social
conduct in later years.
The long-term effects o f child trauma on personality development and psychological
fonctioning are not yet clearly understood. Therefore, it is crucial that longitudinal research
be conducted in the area o f community violence in order to identify links between disrupted
and normal development (Cichhetti, 1993; Pynoos, 1993). At different stages of
development, children face different development tasks that can be disrupted or hindered by
exposure to community violence. Developmentally, the impact of a parent’s exposure to
community violence may be more acute when the child is younger in age. At a young age,
a child is likely to have similar levels o f exposure as their mother. However, as the child
grows and matures, the child has more and more experiences outside the presence o f their
parent and utilizes more systems o f social support than just parents or family members. An
understanding o f the sequelae o f exposure to violence must be informed by an
understanding o f normal adaptation across developmental stages.
Longitudinal research is also important because it can address current limitations
in analysis. Typically, using only single informants or averaging scores o f multiple
informants is employed in regression analysis. However, this might lose important
114
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
information. In addition, these analytic techniques and structural equation modeling
(SEM) techniques are limited to a continuous outcome and do not allow for missing
data. On the other hand, hierarchical linear modeling techniques employed in
longitudinal research can address these limitations.
An important issue that warrants further investigation is the interaction o f risk
factors with parenting. Risk factors may accentuate or amplify the impact of depressive
symptoms resulting in more dysfunctional patterns o f caregiving and poorer child
outcomes than might be observed in the presence of depressive symptoms alone (Rutter,
1990). For example, depression is associated with cognitive distortions that heighten
perceptions o f negative events and stress. It may be that this contributes to the
comorbidity or to the discrepancies in maternal reports.
Clearly, more work is warranted to determine whether different outcomes are
associated with different types o f exposure of community violence (Mazza &
Overstreet, 2000). Also, more work is needed to investigate and specify the role of
perception with regarding to exposure to community violence.
A critical area o f future research needs to be an investigation o f the specific
processes and circumstances which serve as protective factors for families that allow for
or enable effective parenting under conditions o f high threat, stress, and neighborhood
violence (Jenkins and Bell, 1997). In addition, there is a particular need to better
understand how the coping o f parents relates to the coping o f children. Further
clarification and specification o f the interdependent processes through which family
factors relate and interact with exposure to community violence to affect potential
115
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
outcomes for children is clearly needed (Gorman-Smith & Tolan, 1998). Towards this
end, it would be beneficial to include fathers in future studies and gather information on
their traumatic exposure to violence and distress symptomatology. Unfortunately, this
information was not gathered in the present study.
Further study is warranted to understand what is the nature of the cultural and
contextual factors that shape the varied perceptions o f community violence and
manifestations of symptomatology cross-culturally (Guterman et al., 2000). Few
studies examine racial/ethnic or cultural differences in child PTSD, yet these factors
likely affect both exposure and response to trauma. Many studies o f PTSD in children
do not examine their previous exposure to trauma, prior conditions, or psychiatric
history, despite the fact the preexisting conditions and prior exposure increase the
vulnerability of children at times o f stress (Pfeflferbaum, 1997).
The study’s findings o f differences in symptomatology among students from the
two schools despite residing in the same neighborhood suggest that contextual factors
such as the school culture and environment may be important factors along with
familiar factors that serve to buffer the detrimental effects of violence exposure. In
contrast to the neighborhood school, the administration at the non-neighborhood school
took a very proactive approach to the students and families. As a result, the culture at
this school was notably different than the neighborhood school. Further research o f
such variables is needed.
One contextual factor that may contribute to differences in coping with exposure
to community violence and with negative outcomes is the frequent presence of
116
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
additional family members residing in the home of Latino families. The presence of
other adults and/or multiple caregivers in some families may be a protective factor
because it provides additional support for the parent (Randolph, Koblinsky & Roberts,
1996). Further research in this area is necessary.
Towards this end, the development and validation o f assessment instruments
that are clinically relevant and weigh the frequency and intensity o f exposure to
community violence is needed to more precisely measure the traumatic effects of
exposure to community violence. Given the wide range o f verbal skills and variability
in children’s ability to express their internal experiences, assessment tools should be
sensitive and specific to the diverse representations of traumatic responses in children
(Keane, 1996). Also, these tools should be culturally sensitive and appropriate.
The absence o f a consensus regarding how various events o f community
violence should be classified or weighed in terms of their severity hinders scientific
investigation (Selner-O’Hagan et al., 1998). Most measures o f exposure to community
violence do not assess the child’s relationship to the victim or perpetrator nor the child’s
proximity to the event. Unless researchers agree to define community violence in the
same way and assess community violence in similar ways, the ability to advance the
field will be hindered. A uniform and standardized method of assessing children’s and
parents exposure to community violence is critically needed to facilitate integration of
findings across studies and generate a more reliable data base for understanding the
differential impact o f various violent experiences across development (Overstreet,
2000).
117
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Implications for Social Work Practice
Community violence is a very complex problem that intersects with many
sectors o f systems o f care for children and adolescents, including schools, mental health
settings, hospitals, and the juvenile justice system. Social workers are likely to
encounter children in these settings who are experiencing the effects o f exposure to
community violence. However, in spite of the consistent evidence o f the frequency
with which children are exposed to community violence with its serious psychological
impact, assessment o f exposure to community violence in children is uncommon. Thus,
the trauma often remains unrecognized. The lack of assessment heightens the risk that
symptoms displayed by a child may be construed as related to circumstances other than
community violence (Ell & Aisenberg, 1998). For example, failure to obtain a history
o f the exposure to community violence could result in the treatment o f depression or
ADHD or other problems without attending to undiagnosed symptoms related to
exposure to community violence. The failure to regularly assess for exposure to
community violence also contributes to the likelihood that children exposed to
community violence do not receive appropriate or timely treatment intervention
(Aisenberg & Mennen, 2000).
Social workers need to routinely assess for exposure to community violence in
schools and mental health settings just as immunization screening is routinely
conducted as a preventive measure. As part o f regular practice, the initial assessment
should elicit information regarding type and extent o f the child’ s exposure to community
violence. Information should be obtained from multiple sources, such as caregivers and
118
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
teachers, as the child’s behavior in the context of home and school may differ. The
initial assessment should also include the primary caregiver’s reactions to the shared
trauma to determine if those reactions indicate increased risk for the child (Foy et al.,
1996). This assessment would include the extent to which the caregiver may transfer
her/his own distress feelings onto the traumatized child (Gaensbauer, 1996).
Community violence requires a comprehensive, multi-disciplinary response towards
its prevention as well as the treatment o f symptomatology resulting from traumatic
exposure to violence. Single focused interventions such as conflict resolution curriculum,
anger management, cognitive-behavioral treatment, or building more prisons have
demonstrated only limited effectiveness.
A new and more comprehensive intervention to enhance the child’ s coping and
resiliency is required. This intervention would address the impact o f exposure to
community violence upon and within the family, especially the parents. Parents are
potentially significant sources o f resilience in the face of the threat and harm o f
community violence. However, parents who are distressed by exposure to community
violence may be less able to be a resource in promoting the resiliency o f their daughter
or son. It is crucial, therefore, to evaluate the parents’ reactions to traumatic exposure
to community violence in determining any increased risk for children (Foy et al., 1996).
Parents and family members are rarely included in the delivery o f mental health
services for this population. Involving parents/caregivers in the treatment process is
crucial as many parents who are distressed or depressed foil to seek treatment. For
example, among Mexican Americans with mood or anxiety disorders, less than a third
119
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
receive any treatment in the general medical or mental health sector (Vega et al., 1998).
Involving parents in treatment enables the practitioner to address parental distress
resulting from their exposure to community violence and to alert the parents to the
distress o f their child. Also, it enables the practitioner to discuss ways parents can assist
their child such as providing appropriate avenues for their son or daughter to feel safe
and to give expression to his or her feelings and experiences regarding the trauma,
including grief In doing so, social workers can promote the resiliency of the individual
child and family members.
Towards a comprehensive assessment and intervention, social workers need to
initiate further research into the assessment and treatment o f children and parents who
are at risk of PTSD and other distress symptomatology as a result of exposure to
community violence (Tyano et al., 1996). Social workers must engage in research
regarding the effectiveness o f intervention programs and to promote the development of
appropriate, empirically based interventions that target the needs o f children and
families impacted by traumatic exposure to community violence.
Congruent with the profession’s person-in-environment perspective, social
workers must assess and access the various systems comprising the child’s
environment, in particular, the child’s parents. Also, social workers need to advocate
not merely for the treatment of symptomatology related to community violence
exposure but also to address the fundamental causes and risks factors that promote and
sustain community violence.
Social workers provide the overwhelming majority of mental health services in
the United States. Thus, social workers, as researchers and practitioners, are
120
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
prominently positioned and equipped to provide valuable leadership in promoting a
fundamental shift in the delivery o f services to include the assessment and treatment of
the individual child and the child's parents and family members.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
REFERENCE LIST
Achenbach, T.M. (1991). Manual for the Child Behavior Checklist/4~18 and
1991 profile. Burlington, VT: University o f Vermont Department o f Psychiatry.
Aciemo, R., Kilpatrick, D.G., & Resnick, H.S. (1999). Posttraumatic stress
disorder in adults relative to criminal victimization: Prevalence, risk factors, and
comorbidity. In P. A. Saigh and J.D. Bremner (Eds.), Posttraumatic Stress Disorder: A
Comprehensive Text (pp. 44-68). Boston: Allyn and Bacon.
Aiken, L.S., & West, S.G. (1991). Multiple regression: testing and interpreting
interactions. Newbury Park, CA: Sage.
Aisenberg, E. (2001). The Effects o f Community Violence Exposure Upon
Latina mothers and their preschool children. Hispanic Journal o f Behavioral Sciences,
23, 378-398.
Aisenberg, E. (1998). Parent Assessment o f Child’ s Exposure to Community
Violence (PACE). Unpublished manuscript. University of Southern California.
Aisenberg, E. (1998). Self-Assessment o f Exposure to Community Violence-
Parent Version (SAVE-P). Unpublished manuscript. University o f Southern California.
Aisenberg, E., & Ell, K. (Submitted for publication). Traumatic Effects of
Exposure to Community Violence: A Parent-Child Model of Interdependent Coping.
Aisenberg, E., & Mennen, F. (October, 2000). Children Exposed to
Community Violence: Issues for Assessment and Treatment. Child and Adolescent
Social Work Journal, 17, 341-360.
Amaya-Jackson, L., & March, J.S. (1995). Posttraumatic stress disorder. In J.S.
March (Ed.), Anxiety disorders in children (pp. 276-300). New York: Guilford Press.
American Academy o f Pediatrics Task Force on Violence (1999). The Role of
the Pediatrician in Youth Violence Prevention in Clinical Practice and at the
Community Level. Pediatrics, 103, 173-181.
American Psychiatric Association (1994). Diagnostic and statistical manual o f
mental disorders (4 ed.). Washington, DC: Author.
Antonovsky, A., & Sourani, T. (1988). Family sense of coherence and family
adaptation. Journal o f Marriage and Family, 50, 79-82.
122
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Attar, B.K., Guerra, N.G., & Tolan, P.H. (1994). Neighborhood disadvantage,
stressful life events, and adjustment in urban elementary school children. Journal o f
Clinical and Child Psychology, 23, 391-400.
Barrera, M., & Li, S. A. (1996). The relation of family support to adolescents’
psychological distress and behavior problems. In G.R. Pierce, B.FL Sarason, & LG.
Sarason (Eds.), Handbook o f Social Support and the Family (pp. 313-343). New York:
Plenum Press.
Baron, R.M., & Kenny, D.A. (1986). The moderator-mediator variable
distinction in social psychological research: Conceptual, strategic, and statistical
considerations. Journal o f Personality and Social Psychology, 51, 1173-1182.
Bat-Zion, N., & Levy-ShifF, R. (1993). Children in war: Stress and coping
reactions under the threat o f the Scud missile attacks and the effect o f proximity. In L.
Lewis & N. Fox (Eds.), Effects o f war and violence in children (pp. 143-161).
Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.
Beardslee, W.R. & Wheelock, I. (1994). Children of parents with Affective
Disorders. In W.M. Reynolds & H.F. Johnston (Eds.), Handbook o f Depression in
Children and Adolescents (pp. 463-479). New York: Plenum Press.
Berman, S.L., Kurtines, W.M., Silverman, W.K., & Serafini, L.T. (1996). The
impact o f exposure to crime and violence on urban youth. American Journal o f
Orthopsychiatry, 66, 329-330.
Berton, M.M., & Stabb, S.D. (1996). Exposure to violence and post-traumatic
stress disorder in urban adolescents. Adolescence, 3 1 ,489-498.
Bird, H.R., Gould, M.S., & Staghezza, B.M. (1993). Patterns o f diagnostic
comorbidity in a community sample o f children aged 9 through 16 years. Journal o f the
American Academy o f Child and Adolescent Psychiatry, 32, 361-368.
Boney-McCoy, S., & Finkelhor, D. (1995). Psychosocial sequelae o f violent
victimization in a national youth sample. Journal o f Consulting and Clinical
Psychology, 63, 726-736.
Boss, P.G. (1987). Family Stress. In M.B. Sussman & S. Steinmetz (Eds.),
Handbook o f marriage and the family (pp. 695-721). New York: Plenum Press.
Briere, J. (1997). Psychological Assessment o f Child Abuse Effects in Adults.
In J.P. Wilson & T.M. Keane (Eds.), Assessing Psychological Trauma and PTSD (pp.
43-68). New York: Guilford Press.
123
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Briere, J., & Runtz, M. (1989). The Trauma Symptom Checklist (TSC-33).
Early data on a new scale. Journal o f Interpersonal Violence, 4, 151-163.
Breslau, N., Davis, G.C., Andreski, P., & Peterson, E. (1991). Traumatic events
and posttraumatic stress disorders in an urban population of young adults. Archives o f
General Psychiatry, 48, 216-222.
Bruce, M.L., Takeuchi, D.T., & Leaf, P.J. (1991). Poverty and psychiatric
status. Archives o f General Psychiatry, 48,470-474.
Burton, D., Foy, D., Bwanausi, C., Johnson, J., & Moore, L. (1994). The
relationship between traumatic exposure, family dysfunction, and post-traumatic stress
symptoms in male juvenile offenders. Journal o f Traumatic Stress, 7, 83-93.
California Department of Finance Economic Research. (2000). Los Angeles
County. Sacramento, CA: Author.
Canetti, L., Bachar, E., Galili-Weisstub, E., Kaplan De Nour, A., & Shalev,
A.Y. (1997). Parental bonding and mental health in adolescence. Adolescence, 32,
381-394.
Celano, M., Hazzard, A., Webb, C., & McCall, C. (1996). Treatment of
traumagenic beliefs among sexually abused girls and their mothers: An evaluation
study. Journal o f Abnormal Child Psychology, 24, 1-17.
Chilcoat, H.D., & Breslau, N. (1997). Does psychiatric history bias mothers’
reports? An application o f a new analytic approach. Journal o f the American Academy
o f Child and Adolescent Psychiatry, 36, 971-979.
Cicchetti, D., & Lynch, M. (1993). Toward an ecological/transactional model
o f community violence and child maltreatment: Consequences for children’s
development. Psychiatry, 56, 96-118.
Cicchetti, D., & Toth, S.L. (1997). Transactional ecological systems in
developmental psychopathology. In S.S. Luthar, J.A. Burack, D. Cicchetti, & J.R.
Weisz (Eds.), Developmental Psychopathology: Perspectives on Adjustment, Risk, and
Disorder (pp. 317-349). New York: Cambridge University Press.
City of Pasadena (2001). www.ci.pasadena.ca.us
Compass, B.E., Conner, J.K., & Hinden, B.R. (1998). New perspectives on
depression during adolescence. In R. Jesson (Ed.), New Perspectives on Adolescent
Risk Behavior (pp. 319-362). Cambridge, England: Cambridge University Press.
1 2 4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Conners, C.K., Sitarenios, Parker, J.D.A., & Epstein, J.N. (1998). The Revised
Conners’ Parent Rating Scale (CPRS-R): Factor Structure, Reliability, and Criterion
Validity. Journal o f Abnormal Child Psychology, 26, 257-268.
Conners, C.K., Sitarenios, Parker, J.D.A., & Epstein, J.N. (1998). Revision and
Restandardization o f the Conners Teacher Rating Scale (CTRS-R): Factor Structure,
Reliability, and Criterion Validity. Journal o f Abnormal Child Psychology, 26, 279-
291.
Costello, E., & Angold, A. (1988). Scales to assess child and adolescent
depression: Checklists, screens, and nets. Journal o f the American Academy o f Child
and Adolescent Psychiatry, 27, 357-363.
Cooley, M.R., Turner, S.M., & Beidel, D.C. (1995). Assessing community
violence: The children's report of exposure to violence. Journal o f the American
Academy o f Child and Adolescent Psychiatry, 34, 201-208.
Craighead, W.E., Curry, J.F., & Ilardi, S.S. (1995). Relationship o f Children’s
Depression Inventory factors to Major Depression among adolescents. Psychological
Assessment, 7, 171-176.
Cufie, S.P., McCullough, E.L., & Pumariega, A.J. (1994). Comorbidity of
attention deficit hyperactivity disorder and posttraumatic stress disorder. Journal o f
Child and Family Studies, 3, 327-336.
Davidson, J.R.T., & Smith, R. (1990). Traumatic experiences in psychiatric
outpatients. Journal o f Traumatic Stress, 3,459-475.
Derogatis, L.R. (2000). Brief Symptom Inventory (BSI) 18: Administration,
Scoring, and Procedures Manual. Minneapolis, MN: National Computer Systems, Inc.
Derogatis, L.R., & Melisaratos, N. (1983). The brief symptom inventory: An
Introductory report. Psychological Medicine, 13, 595-605.
Downey, G., & Coyne, J.C. (1990). Children of depressed parents: An
integrative review. Psychological Bulletin, 108, 50-76.
Drell, M.J., Siegel, C.H., & Gaensbauer, T J. (1993). Post-Traumatic Stress
Disorder. In C.H. Zeanah (Ed.), Handbook o f Infant Mental Health (pp. 291-304).
New York: Guilford Press.
Duncan, G.J. & Aber, J.L. (1997). Neighborhood Models and Measures. In J.
Brooks-Gunn, G.J. Duncan, & J.L. Aber (Eds.), Neighborhood Poverty, vol. I (pp. 62-
78). New York: Russell Sage Foundation.
125
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Eckenrode, J., & Bolger, N. (1995). Daily and within-day event measurement.
In S. Cohen, R.C. Kessler, & U. Gordon (Eds.), Measuring stress: A guide for health
and social scientists (pp. 80-101). New York: Oxford University Press.
Ell, K. (1996). Social networks, social support and coping with serious illness:
The family connection. Social Science Medicine, 42, 173-183.
Ell, K., & Aisenberg, E. (1998). Stress-related disorders. In J.B. Williams & K.
Ell (Eds.), Advances in Mental Health Research for Social Work Practice (pp. 217-
256). Washington, DC: NASW.
Elliot, D.M., & Briere, J. (1992). Sexual abuse trauma among professional
women: Validating the Trauma Symptom Checklist-40 (TSC-40). Child Abuse and
Neglect, 16, 391-398.
Engle, P.L., Castle, S., & Menon, P. (1996). Child development— Vulnerability
and resilience. Social Science and Medicine, 43, 621-635.
Everson, M.D., Hunter, W.M., Runyon, D.K., Edelsohn, G.A., & Coulter, M.L.
(1989). Maternal support following disclosure of incest. American Journal o f
Orthopsychiatry, 59, 197-207.
Fairbank, J.A., Schlenger, W.E., Saigh, P.A., & Davidson, J.R. (1995). An
epidemiologic profile o f post-traumatic stress disorder: Prevalence, comorbidity, and
risk factors. In M.J. Friedman, D.S. Chamey, & A.Y. Deutch (Eds.), Neurobiological
and Clinical Consequences o f Stress (pp. 415-427). Philadelphia: Lippincott-Raven.
Famularo, R., Fenton, T., Kinscherff, R., Ayoub, C., & Bamum, R. (1994).
Maternal and child posttraumatic stress disorder in cases o f child maltreatment. Child
Abuse and Neglect, 18, 27-36.
Farrell, A., & Bruce, S. (1997). Impact o f exposure to community violence on
violent behavior and emotional distress among urban adolescents. Journal o f Clinical
Child Psychology, 2 6 ,2-14.
Farver, J.A., Natera, L.X., & Frosch, D.L. (1999). Effects of community
violence on inner-city preschoolers and their families. Journal o f Applied
Developmental Psychology, 20, 143-158.
Finkelhor, D., & Kendall-Tackett, K. (1997). A developmental perspective on
the childhood impact o f crime, abuse, and violent victimization. In D. Cicchetti &
S.L.Toth (Eds.), Rochester Symposium on Developmental Psychopathology, vol. 8.
Developmental Perspectives on Trauma; Theory, Research, and Intervention (pp. 115-
140). Rochester, NY: University o f Rochester Press.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fitzpatrick, K.M., & Boldizar, J.P. (1993). The prevalence and consequences
of exposure to violence among African-American youth. Journal o f the American
Academy o f Child and Adolescent Psychiatry, 3 2 ,424-430.
Foa, E.B. (1995). PDS (Posttraumatic Stress Diagnostic Scale) Manual.
Minneapolis, MN: National Computer Systems
Foy, D.W., & Goguen, C.A. (1998). Community violence-related PTSD in
children and adolescents. PSTD Research Quarterly, 9, 1-6.
Foy, D., Madvig, B.T., Pynoos, R.S., & Camilleri, A.J. (1996). Etiologic factors
in the development o f posttraumatic stress disorder in children and adolescents.
Journal o f School Psychology, 34, 133-145.
Frederick, C.J., Pynoos, R.S., & Nader, K. (1992). Child Post-traumatic Stress
Reaction Index. Unpublished instrument.
Freeman, L.N., Mokros, H., & Poznanski, E. (1993). Violent events reported
by normal urban school-aged children: characteristics and depression correlates.
Journal o f the American Academy o f Child and Adolescent Psychiatry, 32, 419-423.
Gaensbauer, T. (1996). Developmental and therapeutic aspects o f treating
infants and toddlers who have witnessed violence. In J.D. Osofsky & E. Fenichel
(Eds.), Islands o f Safety (pp. 15-20). Washington, DC: Zero to Three/National Center
for Clinical Infant Programs.
Garbarino, J., Kostelny, K., & Dubrow, N. (1991). No place to be a child:
Growing up in a war zone. Lexington, MA: Lexington Books.
Garbarino, J., Dubrow, N., Kostelny, K., & Pardo, C. (1992). Children in
Danger. San Francisco: Jossey-Bass Publishers.
Gephart, M.A. (1997). Neighborhoods and Communities as Contexts for
Development. In J. Brooks-Gunn, G.J. Duncan, & J.L. Aber (Eds.), Neighborhood
Poverty, vol. 1 (pp. 1-43). New York: Russell Sage Foundation.
Glod, C.A., & Teicher, M.H. (1996). Relationship between early abuse, PTSD,
and activity levels in prepubertal children. Journal o f the American Academy o f Child
and Adolescent Psychiatry, 35, 1384-1393.
Glodrich, A. (1998). Traumatic exposure to violence: a comprehensive review
of the child and adolescent literature. Smith College Studies in Social Work, 68, 321-
345.
127
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Goenjian, A.K., Stilwell, B.M., Steinberg, A M ., Fairbanks, L.A., Galvin, M.,
Karayan, I., & Pynoos, R.S. (1999). Moral development and psychopathological
interference with conscience functioning among adolescents after trauma. Journal o f
the American Academy o f Child and Adolescent Psychiatry, 38, 376-384.
Goldstein, A., & Conoley, J. (1997). Student aggression: Current status. In
A.P. Goldstein & J.C. Conoley (Eds.), School violence intervention: A practical
handbook (pp. 3-19). New York: Guilford Press.
Golier, J., & Yehuda, R. (1998). Neuroendocrine activity and memory-related
impairments in posttraumatic stress disorder. Developmental Psychopathology, 10,
857-869.
Gondoli, D.M., & Silverberg, S.B. (1997). Maternal emotional distress and
diminished responsiveness: The mediating role of parenting efficacy and parental
perspective taking. Developmental Psychology, 33, 861-868.
Gorman-Smith, D., & Tolan, P. (1998). The role of exposure to community
violence and developmental problems among inner-city youth. Development and
Psychopathology, 10, 101-116.
Gottesman, D., & Lewis, M.S. (1982). Differences in crisis reactions among
cancer and surgery patients. Journal o f Consulting and Clinical Psychology, 50, 381-
388.
Groves, B.M., Zuckerman, B., Marans, S., & Cohen, D.J. (1993). Silent
victims: Children who witness violence. Journal o f the American Medical Association,
269, 262-264.
Hammen, C. (1991). Depression runs in families: The social context o f risk
and resilience in children o f depressed mothers. New York: Springer-Verlag.
Hart, E.L., Lahey, B.B., Loeber, R., & Hanson, K.S. (1994). Criterion validity
o f informants in the diagnosis o f disruptive behavior disorders in children: A
preliminary study. Journal o f Consulting and Clinical Psychology, 62, 410-414.
Hastings, H.M., & Kelly, M.L. (1997). Development and validation o f the
Screen for Adolescent Violence Exposure (SAVE). Journal o f Abnormal Child
Psychology, 25, 511-520.
Henggeler, S.W. (1997). The development of effective drug-abuse services for
youth. In J.A. Egertson, D.M. Fox, & A l. Leshner (Eds.), Treating Drug Abusers
Effectively (pp. 253-279). Cambridge, MA: Blackwell Publishers.
128
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Hernandez,, D.J. & Darke, K. (1999). Socioeconomic and Demographic Risk
Factors and Resources Among Children in Immigrant and Native-Born Families: 1910,
I960, and 1990. In D.J. Hernandez (Ed.), Children o f Immigrants: Health, Adjustment,
and Public Assistance (pp. 19-125). Washington, DC: National Academy Press.
Hightower, A.D., Work, W.C., Cowen, E.L., Lotyczewski, B.S., Spinel], A.P.,
Guare, J.C., & Rohrbeck, C.A. (1986). The Teacher-Child Rating Scale: A Brief
Objective Measure o f Elementary Children’s School Problem Behaviors and
Competencies. School Psychology Review, 15, 393-409.
Hill, H.M., & Jones, L.P. (1997). Children’s and parent’s perceptions o f
children’s exposure to violence in urban neighborhoods. Journal o f the National
Medical Association, 89, 270-276.
Hill, H.M., & Madhere, S. (1996). Exposure to community violence and
African-American children: A multidimensional model of risks and resources. Journal
o f Community Psychology, 24, 26-43.
Hilton, N.Z., Harris, G.T., & Rice, M.E. (1998). On the validity o f self-
reported rates of interpersonal violence. Journal o f Interpersonal Violence, 13, 58-72.
Holmbeck, G.N. (1997). Toward terminological, conceptual, and statistical
clarity in the study o f mediators and moderators: Examples from the child-clinical and
pediatric psychology literatures. Journal o f Consulting and Clinical Psychology, 65,
599-610.
Holmbeck, G.N., Li, S.T., Schurman, J.V., Friedman, D., & Coakley, R.M.
(2002). Collecting and Managing Multisource and Multimethod Data in Studies of
Pediatric Populations. Journal o f Pediatric Psychology, 27, 5-18.
Horn, J.L., & Trickett, P.K. (1998). Community violence and child
development: A review o f research. In P.K. Trickett & C. Schellenbach (Eds.),
Violence against children in the family and community (pp. 103-138). Washington,
DC: APA Books.
Horowitz, K., Weine, S., & JekeL, J. (1995). PTSD symptoms in urban
adolescent girls: Compounded community trauma. Journal o f the American Academy
o f Child and Adolescent Psychiatry, 34, 1353-1361.
Horowitz, M. (1984). A cross validation o f the Impact o f Events Scale.
Journal o f Personality and Social Psychology, 51, 188-194.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Hubbard, J., Realmuto, G.M., Northwood, A.K., & Masten, A.S. (1995).
Comorbidity o f psychiatric diagnoses with posttraumatic stress disorder in survivors of
childhood trauma. Journal o f the American Academy o f Child and Adolescent
Psychiatry, 34, 1167-1173.
James, B. (1994). Handbook for treatment o f attachment-trauma problems in
children. New York: Lexington Books.
Jenkins, E.J., & Bell, C.C. (1997). Exposure and response to community
violence among children and adolescents. In J.D. Osofsky (Ed.), Children in a Violent
Society (pp. 9-31). New York: Guilford Press.
Kashani, J.H., Beck, N., Hoeper, E., Fallahi, C., Corcoran, C.M., McAllistar,
J.A., Rosenberg, T.K., & Reid, J.C. (1987). Psychiatric disorders in a community
sample o f adolescents. American Journal o f Psychiatry, 144: 584-589.
Kashani, J.H., & OrvascheL, H. (1988). Anxiety disorders in mid-adolescence:
a community sample. American Journal o f Psychiatry, 145: 960-964.
Kataoka, S., Jaycox, L.H., Stein, B.D., Wong, M., Escudero, P., Zaragoza, C., &
Fink, A. (submitted for publication). Meeting the Mental Health Needs o f Traumatized
Latino Children: Development o f a School-Based Program.
Keane, T.M. (1996). Clinical perspectives on stress, traumatic stress, and
PTSD in children and adolescents. Journal o f School Psychology, 34, 193-197.
Kessler, R.C., McGonagle, K.A., Zhao, S., Nelson, C.B., Hughes, M.,
Eshleman, S., Wittchen, H., & Kendler, K.S. (1994). Lifetime and 12-month
prevalence o f DSM-IH-R psychiatric disorders in the United States. Archives o f
General Psychiatry, 51, 8-19.
Kessler, R.C., Sonnega A., Bromet, E., Hughes, M., & Nelson, C.B. (1995).
Posttraumatic stress disorder in the National Comorbidity Survey. Archives o f General
Psychiatry, 52, 1048-1060.
Kessler, R.C., Nelson, C.B., McGonagle, K.A., Liu, J., Swartz, M., & Blazer,
D.G. (1996). Comorbidity o f DSM-IH-R major depressive disorder in the general
population: results from the US National Comorbidity Study. British Journal o f
Psychiatry, 168, 17-30.
Kilpatrick, D., & Resnick, H. (1993). PTSD associated with exposure to
criminal victimization in clinical and community populations. In J.R.T. Davidson 8c
E.B. Foa (Eds.), Post-traumatic stress disorder: DSM-IVand beyond (pp. 113-143).
Washington, DC: American Psychiatric Press.
130
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Kilpatrick, D., Aciemo, R., Saunders, B., Resnick, H., & Best, C. (2000). Risk
Factors for Adolescent Substance Abuse and Dependence: Data From a National
Sample. Journal o f Consulting and Clinical Psychology, 65, 1-12.
Kilpatrick, D.G., Saunders, B.E., Resnick, H.S., & Smith, D.W. (1995). The
National Survey ofAdolescents: Preliminary Findings o f Lifetime Prevalence o f
Traumatic Events and Mental Health Correlates. Charleston: Medical University of
South Carolina, National Crime Victims Research and Treatment Center.
King, C.A., Katz, S.H., Ghaziuddin, N., Brand, E., Hill, E., & McGovern, L.
(1997). Diagnosis and assessment of depression and suicidality using the NIMH
Diagnostic Interview Schedule for Children (DISC-2.3). Journal o f Abnormal Child
Psychology, 25, 173-181.
Klerman, G., & Weisman, M.M. (1989). Increasing rates o f depression.
Journal o f the American Medical Association, 261, 2229-2235.
Kliewer, W., Lepore, S.J., Oskin, D., & Johnson, P.D. (1998). The role of
social and cognitive processes in children's adjustment to community violence. Journal
o f Consulting and Clinical Psychology, 66, 199-209.
Koop, C.E., & Lundberg, G.D. (1992). Violence in America: A public health
emergency: time to bite the bullet back. Journal o f the American Medical Association,
267, 3075-3076.
Kovacs, M. (1992). Children’ s Depression Inventory manual. North
Tonawanda, NY: Multi-Health Systems.
Kuo, M., Mohler, B., Raudenbush, S., & Earls, F. (2000). Assessing exposure
to violence using multiple informants: Application of hierarchical linear model.
Journal o f Child Psychology and Psychiatry and Allied Disciplines, 41, 1049-1056.
Kuther, T.L. (1999). A developmental-contextual perspective on youth
covitimization by community violence. Adolescence, 34, 699-714.
Kutlac, M., Layne, C. M., Wood, J., Saltzman, W. S., Stuvland, R., & Pynoos,
R. S. (2000, July). Psychological adjustment in war-exposed secondary school students
two years alter the war: Results of a large-scale risk screening survey. In R. Stuvland
and M. Black (Chairs), UNICEF Psychosocial Projects in Bosnia & Hercegovina 1993-
1999. Paper symposium conducted at the conference, Psychosocial Consequences of
War: Results o f Empirical Research from the Territory o f Former Yugoslavia.
University o f Sarajevo, Bosnia.
1 3 1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LaGreca, A.M., Silverman, W.K., Vemberg, E.M., & Prinstein, M.J. (1996).
Symptoms of posttraumatic stress in children after Hurricane Andrew: A prospective
study. Journal o f Consulting and Clinical Psychology, 64, 712-723.
Lazarus, R.S. (1966). Psychological stress and the coping process. New York:
McGraw Hill.
Lazarus, R.S., & Folkman, S. (1984). Stress, Appraisal, and Coping. New
York: Springer.
Lazarus, R.S., & Cohen, J.B. (1977). Environmental stress. In I. Altman & J.F.
Wohwill (Eds.), Human behavior and the environment: Current theory and research
(pp. 89-127). New York: Plenum Press.
Lazarus, R.S. (1986). In S. McHugh & M.T. Vallis (Eds.), Illness behavior:
A multidisciplinary model (pp. 303-308). New York: Plenum Press.
Los Angeles County Coroner (1996).
Lynch, M., & Cicchetti, D. (1998). An ecological-transactional analysis of
children and correlates: The longitudinal interplay among child maltreatment,
community violence, and children’s symptomatology. Development and
Psychopathology, 10, 235-257.
Malinkosky-Rummell, R., & Hansen, D. (1993). Long-term consequences of
childhood physical abuse. Psychological Bulletin, 114, 68-79.
Marans, S., & Adelman, A. (1997). Experiencing Violence in a Developmental
Context. In J.D. Osofsky (Ed.), Children in a Violent Society (pp. 202-222). New
York: Guilford Press.
March, J.S. (1997). Multidimensional Anxiety Scale for Children: Technical
Manual. New York: Multi-Health Systems Inc.
March, J.S. (1999). Assessment o f Pediatric Posttraumatic Stress Disorder.
In P. A. Saigh & J.D. Bremner (Eds.), Posttraumatic Stress Disorder: A Comprehensive
Text (pp. 199-218). Needham Heights, MA: Allyn & Bacon.
March, J.S., Parker, J.D., Sullivan, K., Stallings, P., & Conners, C.K. (1997).
The Multidimensional Anxiety Scale for Children (MASC): Factor Structure,
Reliability, and Validity. Journal o f the American Academy o f Child and Adolescent
Psychiatry, 36, 554-565.
Margolin, G., & Gordis, E.B. (2000). The Effects o f Family and Community
Violence on Children. Annual Review o f Psychology, 5 1 ,445-479.
132
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Mazza, J.J., & Overstreet, S. (2000). Children and adolescents exposed to
community violence: a mental health perspective for school psychologists. The School
Psychology Review, 29, 86-101.
McCormack, D. (Ed.). (2000). McCormack’s Guides: Los Angeles County
2000. Martinez, CA.
McGee, R., Feehan, M., Williams, S., Patridge, F., Silva, P.A., & Kelly, J.
(1990). DSM-IH disorders in a large sample o f adolescents. Journal o f the American
Academy o f Child and Adolescent Psychiatry, 29, 611-619.
Miller, L.S., Wasserman, G.A., Neugebauer, R., Gorman-Smith, D., &
Kamboukos, D. (1999). Witnessed community violence and antisocial behavior in
high-risk, urban boys. Journal o f Clinical Child Psychology, 28, 2-11.
Mink, E., Santangelo, S.L., Wypij, D., & Biederman, J. (2000). Impact of
Maternal Depression on Ratings o f Comorbid Depression in Adolescents with Attention
Deficit/ Hyperactivity Disorder. Journal o f the American Academy o f Child and
Adolescent Psychiatry, 39, 314-319.
Myers, T.G. (1997). Parenting and Community Violence: An Exploration of
Maternal and Child Adjustment. Unpublished dissertation. University of Maryland.
Nader, K., Pynoos, R.S., Fairbanks, L., Frederick, C., Al-Ajeel, M., &
Al-Asfour, A. (1993). A preliminary study of PTSD and grief among the children o f
Kuwait following the Gulf Crisis. British Journal o f Clinical Psychology, 32, 407-416.
Nelson, C.A., & Carver, L J . (1998). The effects of stress and trauma on brain
and memory: a view from developmental cognitive neuroscience. Developmental
Psychopathology, 10, 793-809.
Newberger, C.M., Geremy, I.M., Watemaux, C.M., & Newberger, E.H. (1993).
Mothers o f sexually abused children: Trauma and repair in longitudinal perspective.
American Journal o f Orthopsychiatry, 63x 92-102.
Neuling, S.J., & Winefield, H.R. (1988). Social support and recovery after
surgery for breast cancer: Frequency and correlates o f supportive behaviours by family,
friends, and surgeon. Social Science and Medicine, 27, 385-392.
Organista, K. (2000). Latinos. In J.R. White & A.S. Freeman (Eds.), Cognitive-
behavioral Group Therapy (pp. 281-303). American Psychological Association,
Washington, DC.
Osofsky, J.D. (1995). The effects of exposure to violence on young children.
American Psychologist, 50, 782-785.
133
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Osofeky, J.D., & Fenichel, E. (1994). Caring for infants and toddlers in violent
environments: Hurt, healing and hope. Arlington, VA; Zero to Three/National Center
for Clinical Infont Programs.
Osofeky, J.D., Wewers, S., Hann, D.M., & Fick, A.C. (1993). Chronic
community violence: What is happening to our children. Psychiatry, 56, 36-45.
Overstreet, S. (2000). Exposure to Community Violence: Defining the Problem
and Understanding the Consequences. Journal o f Child and Family Studies, 9, 7-25.
Overstreet, S., Dempsey, M., Graham, D., & Moely, B. (1999). Availability o f
fomily support as a moderator o f exposure to community violence. Journal o f Clinical
Child Psychology, 28, 151-159.
Ozzer, E.M., Brindis, C.D., Millstein, S.G., Knopf, D.K., & Irwin, C.E. (1997).
America’s adolescents: Are they healthy? San Francisco: University of California, San
Francisco, National Adolescent Health Information Center.
Parker, J., & Randall, P. (1996). Post-traumatic stress disorder in children: The
social work challenge. Journal o f Social Work Practice, 10, 71-81.
Pastore, D.R., Fisher, M., & Friedman, S.B. (1996). Violence and mental
health problems among urban high school students. Journal o f Adolescent Health, 18,
320-324.
Patterson, G.R., & Bank, L. (1986). Bootstrapping your way in the
nomological thicket. Behavioral Assessment, 8,49-73.
Pearlin, L.I. (1989). The Sociological Study of Stress. Journal o f Health and
Social Behavior, 30, 241-256.
Perry, B. (1997). Incubated in terror: neuro-developmental factors in the
“Cycle of Violence.” In J.D. Osofeky (Ed.), Children in a Violent Society (pp. 124-
149). New York: Guilford Press.
Perry, B. (1994). Neurobiological sequelae o f childhood trauma: PTSD in
children. In M.M. Murberg (Ed.), Catecholamine function in posttraumatic stress
disorder: Emerging concepts (pp. 233-256). Washington, DC: American Psychiatric
Press.
Phillips, D., & Cabrera, N. (1996). Beyond the Blueprint: Directions for
Research on Head Start’ s Families. Washington, DC: National Academy Press.
1 3 4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Pianos, R., Zayas, L.H., & Busch-Rossnagel, N.A. (1997). Mental health
factors and teaching behaviors among Iow-income Hispanic mothers. Families in
Society: The Journal o f Contemporary Human Services, 78, 4-12.
Poznanski, E.O., Grossman, J.A., Buchsbaum, Y., Banegas, M., Freeman, L., &
Gibbons, R. (1984). Preliminary studies o f the reliability and validity o f the Children’s
Depression Rating Scale. Journal o f the American Academy o f Child Psychiatry, 23,
191-197.
Pynoos, R.S., Frederick, C.J., Nader, K., Arroyo, W., Steinberg, A., Eth, S.,
Nunez, F., & Fairbanks, L. (1987). Life threat and posttraumatic stress in school-age
children. Archives o f General Psychiatry, 44, 1057-1063.
Pynoos, R.S., & Nader, K. (1988). Psychological first aid: For children who
witness community violence. Journal o f Traumatic Stress, 7,445-473.
Pynoos, R.S., & Nader, K. (1993). Issues in the treatment o f post-traumatic
stress in children and adolescents. In J. Wilson & B. Raphael (Eds.), The international
handbook o f traumatic stress syndromes (pp. 535-549). New York: Plenum Press.
Pynoos, R., Goenjian, A., Tashjian, M., Karakashian, M., Manjikian, R.,
Manoukian, G., Steinberg, A., & Fairbanks, L. (1993). Posttraumatic stress reactions
in children after the 1988 Armenian earthquake. British Journal o f Psychiatry, 163,
239-245.
Pynoos, R.S., Steinberg, A.M., & Goenjian, A. (1996). Traumatic stress in
childhood and adolescence: Recent developments and current controversies. In B.A.
van der Kolk, A.C. McFarlane, & L. Weisaeth (Eds.), Traumatic stress: The effects o f
overwhelming experience on mind, body, and society (pp. 331-368). New York:
Guilford Press.
Pynoos, R.S., Steinberg, A.M., & Wraith, R. (1995). A developmental model
of childhood traumatic stress. In D. Cicchetti & D.J. Cohen (Eds.), Developmental
Psychopathology, vol. 2. (pp. 72-95). New York: Wiley.
Pynoos, R.S., Goenjian, A.K., & Steinberg, A.M. (1998). A public mental
health approach to the postdisaster treatment o f children and adolescents. Child and
Adolescent Psychiatric Clinics o f North America, 7, 195-210.
Pynoos, R.S., Steinberg, A.M, & Piacentini, J.C. (1999). Developmental
psychopathology o f childhood traumatic stress and implications for associated anxiety
disorders. Biological Psychiatry, 46, 1542-1554..
135
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Randolph, S.M., Koblinsky, S.A., & Roberts, D.D. (1996). Studying the role of
family and school in the development o f African-American preschoolers in violent
neighborhoods. Journal o f Negro Education, 65, 282-295.
Reynolds, W.M. (1987). Reynolds Adolescent Depression Scale. Professional
Manual. Odessa, FL: Psychological Assessment Resources, Inc.
Richters, J.E., & Martinez, P. (1993). The NIMH Community Violence
Project: I. Children as victims and witnesses to violence. Psychiatry, 56, 7-21.
Richters, J.E., & Saltzman, W. (1990). Survey o f Children's Exposure to
Community Violence: Parent Report. National Institute o f Mental Health: Rockville,
MD.
Rodriguez, N., Steinberg, A.M., & Pynoos, R.S. (1999). UCLA PTSD Reaction
Index -Revised Adolescent Version. University o f California at Los Angeles.
Rohrbach, L., Mansergh, G., Fishkin, S., & Johnson, C.A. (1996). Survey of
Alcohol, Tobacco, and Other Drug Use, and Violence Among Pasadena/Altadena
Youth. University o f Southern California, Institute for Health Promotion & Disease
Prevention Research. Los Angeles, CA.
Rolland, J.S. (1994). Families, illness, and disability: An integrative treatment
model. New York: Basic Books.
Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ:
Princeton University Press.
Rosenberg, M.L., & Fenley, M.A. (1992). The Federal Role in Injury Control.
American Psychologist, 4 7 ,1031-1035.
Rutter, M. (1990). Commentary: some focus and process considerations
regarding effects o f parental depression on children. Developmental Psychopathology,
2 6 ,60-67.
Sack, W.H., Clarke, G.N., & Seeley, J. (1995). Posttraumatic Stress Disorder
across Two Generations o f Cambodian Refugees. Journal o f the American Academy o f
Child and Adolescent Psychiatry, 34, 1160-1166.
Saigh, P.A., Green, B.L., & Korol, M. (1996). The history and prevalence of
posttraumatic stress disorder with special reference to children and adolescents.
Journal o f School Psychology, 34, 107-131.
1 3 6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Saigh, P. A., Mroueh, M. & Bremner, J. D. (1997). Scholastic impairments
among traumatized adolescents. Behavior Research and Therapy, 3 5 ,429-436.
Saltzman, W.R., Pynoos, R.S., Layne, C.M., Steinberg, A.M., & Aisenberg, E.
(2001). Trauma- and grief-focused intervention for adolescents exposed to community
violence: Results o f a school-based screening and group treatment protocol. Group
Dynamics Special Issue: Group-based interventions for trauma survivors, 5, 291-303.
Sampson, R.J., Raudenbush, S.W., & Earls, F. (1997). Neighborhoods and
violent crime: a multilevel study o f collective efficacy. Science, 277, 918-924.
Sauter, J., & Franklin, C. (1998). Assessing post-traumatic stress disorder in
children: Diagnostic and measurement strategies. Research on Social Work Practice, 8,
251-270.
Scheeringa, M., Zeanah, C.H., Drell, M.J., & Larrieu, J.A. (1995). Two
approaches to the diagnosis o f posttraumatic stress disorder in infancy and early
childhood. Journal o f the American Academy o f Child and Adolescent Psychiatry, 34,
191-200.
Schwab-Stone, M.E., Ayers, T.S., Kasprow, W., Voyce, C., Barone, C., Shriver,
T., & Weissberg, R.P. (1995). No safe haven: a study of violence exposure in an urban
community. Journal o f the American Academy o f Child and Adolescent Psychiatry, 34,
1343-1352.
Schwab-Stone, M., Chen, C., Greenberger, E., Silver, D., Lichtman, J., &
Voyce, C. (1999). No safe haven. II: The effects of violence exposure on urban youth.
Journal o f the American Academy o f Child and Adolescent Psychiatry, 38, 359-367.
Schwartz, E.D., & Kowalski, J.M. (1991). Posttraumatic stress disorder after a
school shooting: Effects o f symptom threshold selection and diagnosis by DSM-IH-R,
or proposed DSM-IV. American Journal o f Psychiatry, 148, 592-597.
Schwarzwald, J., Weisenberg, M., Solomon, Z., & Waysman, M. (1994). Stress
reaction of school-age children to the bombardment by SCUD missiles: A I-year
follow-up. Journal o f Traumatic Stress, 7,657-667.
Selner-O’Hagan, M.B., Kindlon, D.J., & Buka, S.L. (1998). Assessing
exposure to violence in urban youth. Journal o f Child Psychology and Psychiatry and
Allied Disciplines, 39, 215-225.
Selye, H. (1976). The stress o f life. New York: McGraw-Hill.
Shaffer, D., Lucas, C.P., & Richters, J.E. (1999). Diagnostic assessment in
child and adolescent psychopathology. New York: Guilford Press.
137
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Shahinfar, A., & Fox, N.A. (1997). The Effects of Trauma on Children:
Conceptual and Methodological Issues. In D. Cicchetti & S.L. Toth (Eds.), Rochester
Symposium on Developmental Psychopathology, vol. 8. Developmental Perspectives on
Trauma; Theory, Research, and Intervention (pp. 115-140). Rochester, NY: University
o f Rochester Press.
Shaw, J.A., Applegate, B., & Schorr, C. (1996). Twenty-one-month follow-up
study of school-age children exposed to Hurricane Andrew. Journal o f the American
Academy o f Child and Adolescent Psychiatry, 34, 1185-1192.
Shakoor, B.H., & Chalmers, D. (1991), Co-victimization of African-American
children who witness violence: Effects on cognitive, emotional and behavioral
development. Journal o f the American Medical Association, 8 3 ,233-238.
Sheidow, A.J., Gorman-Smith, D., Tolan, P.H., & Henry, D.B. (2001). Family
and Community Characteristics: Risk Factors for Violence Exposure in Inner-City
Youth. Journal o f Community Psychology, 29, 345-360.
Sickmund, M., Snyder, H., & Poe-Yamagata, E. (1997). Juvenile offenders and
victims: 1997 Update on violence. Washington, DC: Office o f Juvenile Justice and
Delinquency Prevention.
Singer, M.I., Anglin, T.M., Song L.Y., & Lunghofer, L. (1995). Adolescents’
exposure to violence and associated symptoms of psychological trauma. Journal o f the
American Medical Association, 273, 477-482.
Smucker, M.R., Craighead, W.E., Craighead, L.W., & Green, B.J. (1986).
Normative and reliability data for the Children’s Depression Inventory. Journal o f
Abnormal Psychology, 9 7 ,492-495.
Snyder, H. (1998). Juvenile arrests 1997. Juvenile Justice Bulletin.
Washington, DC: Office o f Juvenile Justice and Delinquency Prevention.
Speier, P., Sherak, D., Hirsch, S., & Cantwell, D. (1995). Depression in
children and adolescents. In E.E. Beckham & W.R. Leber (Eds.), Handbook o f
depression (2n d ed.) (pp. 467-493). New York: Guilford Press.
Springer, C., 8c Padgett, D.K. (2000). Gender Differences in Young
Adolescents’ Exposure to Violence and Rates o f PTSD Symptomatology. American
Journal o f Orthopsychiatry, 70, 370-379.
Steinglass, P. (1987). A systems view o f family interaction and psychotherapy.
In T. Jacob (Ed.), Family interaction and psychopathology (pp. 25-63). New York:
Plenum Press.
1 3 8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Stephens, R. (1997). National trends in school violence: Statistics and
prevention strategies. In A.P. Goldstein & J.C. Conoley (Eds.), School violence
intervention: A practical handbook (pp. 72-90). New York: Guilford Press.
Taylor, L., Zuckerman, B., Harik, V., & Groves, B. (1994). Witnessing
violence by young children and their mothers. Journal o f Developmental and
Behavioral Pediatrics, 15, 120-123.
Terr, L.C. (1981). Psychic trauma in children: Observations following the
Chowchilla school bus kidnapping. American Journal o f Psychiatry, 138, 14-19.
Tolman, R.M. (1989). The initial development o f a measure o f psychological
maltreatment o f women by their male partners. Violence and Victims, 4, 159-178.
Tolman, R.M. (1999). The validation o f the psychological maltreatment o f
women inventory. Preliminary report. Violence and Victims, 14, 25-35.
Turner, C.F., Ku, L., Rogers, S.M., Lindberg, L.D., Pleck, J.H., & Sonenstien,
F.L. (1998). Adolescent sexual behavior, drug use, and violence: Increased reporting
with computer survey technology. Science, 280, 867-873.
Tyano, S., Iancu, J., Solomon, Z., Sever, J., Goldstein, I., Touviana, Y., &
Bleich,A. (1996). Seven-Year follow-up of child survivors of a bus-train collision.
Journal o f the American Academy o f Child and Adolescent Psychiatry, 35, 365-373.
U.S. Census Bureau (March Supplement 2000). Current Population Survey,
Ferret Report.
U.S. Department o f Justice (1998). National Crime Victimization Survey.
Criminal Victimization 1997: Changes 1996-1997 with Trends 1993-1997.
Washington, DC: Bureau of Justice Statistics.
United Way of Greater Los Angeles (1999). State o f the County Report: Los
Angeles 1998-99. Table 128, Los Angeles, CA.
Vega, W.A., Kolody, B., Aguilar-Gaxiola, S., Alderete, E., Catalano, R., &
Caraveo-Anduaga, J. (1998). Lifetime prevalence o f DSM-IH-R psychiatric disorders
among urban and rural Mexican Americans in California. Archives o f General
Psychiatry, 55, 771-778.
Vemberg, E.M., LaGreca, A.M., Silverman, W.K., & Prinstein, M.J. (1996).
Prediction o f posttraumatic stress symptoms in children after Hurricane Andrew.
Journal o f Abnormal Psychology, 105, 237-248.
1 3 9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Wallen, J., & Rubin, R.H. (1997). The role of the family in mediating the
effects o f community violence on children. Aggression and Violent Behavior, 2, 33-41.
Walsh, F. (1996). The concept o f family resilience: crisis and challenge.
Family Process, 35, 261-281.
Weathers, F.W., & Keane, T.M. (1999). Psychological assessment o f
traumatized adults. In P. A. Saigh & J.D. Bremner (Eds.), Posttraumatic Stress
Disorder: A Comprehensive Text (pp. 219-247). Boston: Allyn and Bacon.
Weiner, H. (1994). The revolution in stress theory and research. InR.P.
Liberman & J. Yager (Eds.), Stress in psychiatric disorders (pp. 1-36). New York:
Springer.
Weiss, B., Weisz, J.R., Politano, M., Carey, M., Nelson, W.M., & Finch, A.J.
(1991). Developmental differences in the factor structure o f the Children’s Depression
Inventory. Psychological Assessment, 3, 1-8.
Wemer, E.E. (1992). The children o f Kauai: Resiliency and recovery in
adolescence and adulthood. Journal o f Adolescent Health, 13, 262-268.
Werner, E.E. (1995). Resilience in development. Current Directions in
Psychological Science, 4, 81-85.
White, K.S., Bruce, S.E., Farrell, A.D., & Kliewer, W. (1998). Impact of
exposure on anxiety: a longitudinal study of family social support as a protective factor
for urban children. Journal o f Child and Family Studies, 7, 187-203.
Wylie, R.C. (1989). Measures o f self-concept. Lincoln: University of
Nebraska Press.
Yule, W., & Canterbury, R. (1994). The treatment o f post traumatic stress
disorder in children and adolescents. International Review o f Psychiatry, 6, 141-151.
Zeanah, C.H., & Scheeringa, M. (1996). Evaluation o f posttraumatic
symptomatology. In J.D. Osofeky & E. Fenichel (Eds.), Islands o f Safety: Assessing
and treating young victims o f violence (pp. 9-14). Washington, DC: Zero to
Three/National Center for Clinical Infant Programs.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX
CORRELATION MATRIX
Mom Age
M Age
1.000
Race School
Race/ethnicity .500 •• 1.000
School attends .099 .239 • 1.000
Mom education 303 •• .653 •• .174
Dad education .192 .449 •• .511
Family income .231 3 4 9 •• .321
Child gender .055 -.133 -.092
BSI depression .048 .172 -.007
BSI anxiety • 038 .073 .069
TSC total -.107 .119 -.057
TSC sex abuse -.214 -.071 -.104
Self Esteem -.199 -.176 -.143
Mother total exp. .182 .278 • -.008
Child total exp. -.007 -.095 .037
MR child exp. .221 .218 .011
PTSDRI .058 -.019 .118
RADS .046 .086 .087
Foa PTSD -.024 -.100 -.083
Psy. Maltreat .068 -.004 .061
MASC Total .006 -.141 .023
CBCL Total -.062 -.010 -.012
M Oppositional .052 .069 .098
M ADHD .165 .107 .167
M Hyperactive .072 .082 .076
M Global Index .119 .032 .105
T Oppositional -.079 -.049 -.038
T ADHD -.082 -.132 .036
T Hyperactivity -.069 -.053 -.073
TGlobal Index -.079 -.137 -.041
M 9 Mom Report
T s Teacher Report
•
»•
p < .05
p< .0 1
M edu D edu Income Gender Depress Anxiety TSC lot Sex Ab
1.000
.648 •• 1.000
3 1 6 •• .498 •• 1.000
-.057 -.024 .107 1.000
.010 -.103 -.283 •• .097
-.049 .028 -.153 .073
-.066 .036 -.056 .276 •
-.177 -.024 -.095 .303 *
-331 • -.153 -.378 •• .078
.280 • .281 • .048 .127
-.131 -.045 -.254 -.045
.235 .085 .220 .093
-340 • -.214 -.064 .225
-.042 .116 .124 .222
-.115 -.257 -.268 • .021
-.210 -.197 -.171 .164
-.090 -001 -.097 .224
-.203 -.085 -.236 .030
-.072 -.038 -.119 -.059
-.052 .025 -.042 -.036
.040 .012 -.053 -.057
-.097 -.052 -.070 -.009
.006 .072 .098 -.027
-.199 .086 .170 -.009
-088 .031 .060 -.048
-.179 .046 .142 -.015
1.000
.718 • • 1.000
.702 • • .792 •• 1.000
.594 •• .751 •• .903 •• 1.000
.517 •• .468 •• .385 •• .372
.443 • • .385 •• 395 •• .275
.132 .097 .110 .085
.201 .201 .307 • .148
.352 • • .262 • .258 • .260
-.050 .000 .129 .196
.520 •• .493 •• .688 •• .393
314 • .250 .284 • .184
-.057 -.130 -.153 -.022
.454 •• .496 •• .429 •• 352
.450 •• .434 •• 3 7 2 •• .230
.365 • • .409 •• 3 2 2 •• .177
.383 •• .401 •• .404 •• 391
.388 •• .445 *• 374 •• .259
-.027 -.052 .103 .137
.069 .150 303 .265
.060 -.033 .093 .109
.044 .052 .183 .240
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX continued
Esteem Mexp Ch exp M report React In RADS F PTSD Maltreat MASC CBCLT
(.000
.030 1.000
.309 .029 1.000
-.085 .633 *• .119 1.000
.144 -.010 .432 •• .093 1.000
.183 .250 .415 *• .148 .408 **
.457 •• .276 • .184 .058 .230 •
.335 • .234 .359 • .139 .393 ••
.028 -.227 • .160 -.122 .143
.295 • .342 • -.013 .269 • .116
.204 .305 • -094 .272 • .025
.220 .298 • .066 .444 •* .105
.214 JJ5I •• .163 .431 •• .213
.189 .330 •• .052 .398 •• .140
.044 -.007 .210 -.031 -.010
.226 -.021 .228 -.101 .037 ••
.059 -.010 .155 -.089 -.028
.158 -.002 .174 -.059 -.017
1.000
.037 1.000
.232 .295 * 1.000
.185 -.110 .045 1.000
.130 .265 • .396 •• .072 1.000
.015 .190 .189 -.021 .794 •*
.078 .147 .287 .086 .749 ••
.202 .285 • .251 .009 .724 ••
.147 .289 • J 0 7 • .038 .792 ••
.111 .086 .053 -.161 -.137
.084 .145 .035 -.144 -.039
.076 .038 .063 -.141 -.016
.071 .150 .045 -.130 -.073
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX continued
M Oppose M ADHD M Hyper M Global T Oppose T ADHD T Hyper
1.000
.843 •• 1.000
.839 •• .835 •• 1.000
.899 •• .911 •• .869 •• 1.000
-.088 -.097 -.035 -.009 1.000
.054 .084 .095 .119 .722 •• 1.000
.056 -.019 .062 .090 .875 •• .793 •• 1.000
.008 .010 .045 .085 .873 •• .938 •• 502
T Global
1.000
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Abstract (if available)
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
The flow of children through the foster care system
PDF
Communicating about health in Chinese, Korean, and Latino families in Los Angeles
PDF
Emergent literacy: A sense of becoming and being literate in Latino families
PDF
The role of affect in the marital satisfaction of monoethnic Latino, biethnic Latino, and European -American newlywed couples
PDF
Intra-familial child sexual abuse: The experience and effects on non-offending mothers
PDF
A cultural -ecological analysis of educational aspirations and satisfaction of educational opportunities of Asian -Americans and Hispanics: Understanding the interaction of race and class, paren...
PDF
Latino and African American mothers' involvement in their children's early education
PDF
The effects of community violence on the well-being of Latino mothers and their preschool children
PDF
Predictors of dysphoria and substance abuse among Latino and African American youth
PDF
Residential satisfaction of Chinese seniors living in senior housing
PDF
The effects of race, gender, and family background on children's educational attainment: Contemporary patterns and historical change
PDF
Parents as harbingers: A new model for differentially predicting life outcomes for African-Americans and Caucasian Hispanics
PDF
Romanticizing misogynistic images: Adolescent girls and rap music
PDF
Re-racing reality: Families formed across the color line reassert race and class
PDF
Psychological well-being, religious affiliation, and cultural identification in adult children of Jewish-Gentile interfaith marriages
PDF
The relationship between acculturation and the household structure of older Mexican Americans in Los Angeles
PDF
Stress, appraisal and coping among Japanese-American, Anglo-American, African-American and Mexican-American spousal caregivers of persons with dementia
PDF
Organizational constructs as predictors of client outcomes: A study of a family preservation intervention model
PDF
Comparative analysis of the stress process in AIDS caregivers: Gay male partners, Anglo women, and Latinas
PDF
The performance of race in Asian American drama
Asset Metadata
Creator
Aisenberg, Eugene Francis
(author)
Core Title
Families exposed to community violence: The impact of maternal distress symptomatology on Latino and African -American young adolescents
Degree
Doctor of Philosophy
Degree Program
Social Work
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
OAI-PMH Harvest,Social Work,sociology, ethnic and racial studies,sociology, individual and family studies,sociology, public and social welfare
Language
English
Contributor
Digitized by ProQuest
(provenance)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c16-210197
Unique identifier
UC11339073
Identifier
3073736.pdf (filename),usctheses-c16-210197 (legacy record id)
Legacy Identifier
3073736.pdf
Dmrecord
210197
Document Type
Dissertation
Rights
Aisenberg, Eugene Francis
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA
Tags
sociology, ethnic and racial studies
sociology, individual and family studies
sociology, public and social welfare