Close
About
FAQ
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
/
Parenting practices among non-offending mothers of sexually abused girls and its impact on the abused girls' behavioral adjustment: perspectives from a multigenerational, longitudinal study
(USC Thesis Other)
Parenting practices among non-offending mothers of sexually abused girls and its impact on the abused girls' behavioral adjustment: perspectives from a multigenerational, longitudinal study
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
PARENTING PRACTICES AMONG NON-OFFENDING MOTHERS OF
SEXUALLY ABUSED GIRLS AND ITS IMPACT ON THE ABUSED GIRLS’
BEHAVIORAL ADJUSTMENT: PERSPECTIVES FROM A
MULTIGENERATIONAL, LONGITUDINAL STUDY
by
Kihyun Kim
______________________________________________________________
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(SOCIAL WORK)
August 2007
Copyright 2007 Kihyun Kim
ii
Acknowledgements
This dissertation project has been sustained by an old proverb – “we are all
dwarfs standing on the shoulders of giants.” Were it not for many giants that I have
had the fortune of encountering in my life, this dissertation would not have come to
see the light of the day.
I began my postgraduate career as a research assistant for Professor Penelope
Keith Trickett’s longitudinal studies that I joined in the first term of my doctoral
education. It must have been a rare fortune for a fledgling student researcher to
partake in her lab and to be exposed to a cutting-edge research. Ever since then, she
has stayed with me throughout the journey as a mentor, guide, and exemplar. I owe
her the most substantial debt during my years at the University of Southern
California (USC)’s School of Social Work.
It was my good fortune that I met scholars with unsurpassed intellectual
vibrancy, rigor, and commitment at USC. Professor John Horn represents those
scholarly virtues in person. Two of his graduate seminars in measurement and
developmental research introduced me to an open-ended horizon of quantitative
inquiries. He graciously agreed to oversee my dissertation project, but passed away
before the completion of my dissertation. I hope he would have been pleased to
recognize his voice and spirit in this final product. Professor Jack McArdle has
generously stepped in on behalf of Professor John Horn. This dissertation could not
have benefited more from his insights that come from his extensive practice and a
keen sense of modeling analyses. His inputs have proved to be critical in translating
iii
my abstract ideas into analyzable terms. Professor Ferol Mennen has produced
incisive comments on the entire dissertation manuscript. Her extensive clinical
experiences have deepened my understanding of the target population of this study.
Her warm support during my PhD education is truly appreciated. Although left USC
before I joined it, Professor Jennie Noll (University of Cincinnati) has left a deep
impact on this project. I had a chance to collaborate with her last fall. Her thoughtful
comments and balanced judgments were critical for the elaboration of the ideas that
culminated in this dissertation. The ideas of this dissertation were first germinated in
the weekly-tutorials that I have had with Professor Ron Astor. He has been a great
source of support and encouragement during my USC days.
My special thanks also extend to the three educational institutions which have
nurtured my intellectual curiosity, stimulated the commitment to societal problems,
and sharpened my analytic skills. The USC School of Social Work has provided me
with generous supports throughout my PhD education. Especially, this dissertation
project has benefited much from the USC Urban Initiative. The exemplary
community of the dedicated scholars in the School of Social Service Administration
at the University of Chicago convinced me of the value of social work as a catalyst
for societal reform and taught me what a social work research could and should be
like. I spent almost ten years at Ewha Woman’s University in Korea, both as an
undergraduate and graduate student. The liberal arts education that I received at
Ewha has always been at my heart throughout my subsequent career as a student and
social worker. Professor Young Choi was my principal mentor at Ewha over close to
iv
a decade and an integral part of my Shakesperean days. Now I understand how
difficult it must have been for her to see her protégé to embark on a completely
different and uncertain voyage as a social worker. May she be pleased with her
pupil’s work albeit in a different walk of life.
Back in the 70’s and 80’s in Seoul, Korea, I was one of very few kids who
had a working mother. My mother, Kyung-ja Lee, has successfully reconciled her
motherhood and professional career. It was on her shoulder that I had envisioned my
future and came to appreciate my motherhood. I thank her dearly, so dearly for her
unfailing support and as an exemplar both as a mother and a professional. I also
thank my daughter’s grandparents as well as my own. My own grandmother, Eul Tae
Shin, often played a mother role during my childhood in lieu of her daughter. Her
embrace was the warmest of all one can imagine and has always been remembered
as such throughout my life. She would have been most pleased to see her
granddaughter’s accomplishment she warmly encouraged long ago. My
parents-in-law, Dong Sung Lim and Ilsoon Kang, deserve my sincerest gratitude.
They are the most wonderful grandparents that anyone can hope for. Without their
unconditional love and support for my daughter, I could not have completed this
dissertation. My brother, Sung Ho Kim, and my sister, Kihye Kim, have been a great
source of support. They have reminded me of how fraternity can make our lives so
much richer and joyful.
I cannot thank enough my husband, Byungil Lim, for his love, patience, and
understanding. I look forward to the future that we will make together. My
v
seven-year-old daughter, Yeo-Eun, has stayed with me in this City of Angel
throughout my PhD years. I was not so much available to her as I wanted to be for
the time working on this dissertation – that may be why children’s resilience was
such a compelling framework to me as a mother and a researcher! She often made
my graduate student’s life complicated, but never failed to provide inspiration and
rejuvenation every day. Hopefully, she would find my shoulder firm and high
enough to step on and envision her future someday.
Last but not least, I thank my father, Chang Yul Kim. It was my greatest
fortune to be raised by this sharp, passionate, and committed intellectual. He has
demonstrated by example that integrity is one virtue that cannot be compromised
under any circumstances. He passed away shortly before the completion of this
dissertation, but has been with me throughout as the greatest source of my
self-confidence. To my father, this dissertation is dedicated with the deepest
gratitude.
vi
Table of Contents
Acknowledgements......................................................................................................ii
Table of Contents ........................................................................................................vi
List of Tables ............................................................................................................viii
List of Figures .............................................................................................................ix
Abstract ………………………………………………………………………………x
Chapter I. Introduction.................................................................................................1
1.1 Background and Significance .............................................................................1
1.2 Purpose of the Research......................................................................................5
Chapter II. Parenting Practices of Non-Offending Mothers of Sexually Abused Girls ..........6
2.1 Psychosocial Characteristics of Non-Offending Mothers of Sexually Abused
Girls: Empirical Findings ...........................................................................8
2.2 Descriptive Studies on Parenting Functioning of Non-Offending Mothers of
Sexually Abused Girls..............................................................................10
2.3 The Developmental-Ecological Perspective of Parenting ................................11
2.3.1 Mothers’ Childhood Sexual Abuse Experiences and Parenting...................12
2.3.2 The Notion of Intergenerational Transmission of Parenting Practices.........19
2.3.3 Dissociative Symptoms and Parenting Behavior .........................................22
2.3.4 Social Networks Surrounding Families........................................................25
2.4 An Integrative Model of Parenting Practice for Non-Offending Mothers of
Sexually Abused Girls..............................................................................29
Chapter III. Non-Offending Mothers’ Parenting and Abused Girls’ Behavioral
Adjustment...................................................................................................32
3.1 Resilience Framework ......................................................................................32
3.1.1 Theoretical Discussions of Resilience..........................................................32
3.1.2 Ordinary Human Adaptive Systems: Parenting as a Protective Process......36
3.2 Child Sexual Abuse and Resilience in Behavioral Adjustment........................37
3.2.1 Child Sexual Abuse as a Threat to Development of Behavioral Competence.........38
3.2.2 The Role of Non-Offending Mothers in the Post-Abuse Adjustment of
Sexually Abused Girls.............................................................................41
3.2.3 Methodological Issues in Past Studies..........................................................43
3.3 Integration: The Generic Conceptual Model ....................................................45
Chapter IV. Method ...................................................................................................48
4.1 Methods for Data Preparation...........................................................................48
vii
4.1.1 Participants ...................................................................................................48
4.1.2 Individual Subjects and Mother-Daughter Pairs ..........................................49
4.1.3 Measures of G
1
Mothers’ Childhood Developmental History .....................51
4.1.4 Measures of G
1
Mothers’ Parenting Practices..............................................53
4.1.5 Measures of G
1
Mothers’ Current Functioning ............................................53
4.1.6 Measures of G
2
Daughters’ Behavioral Adjustment ....................................54
4.1.7 Missing Data.................................................................................................56
4.1.8 Model Fitting Indices....................................................................................56
4.2 Methods for Model Organization......................................................................57
4.2.1 Structural Models of Relationships: G
1
Mothers’ Parenting........................57
4.2.2 Structural Models of Relationships: G
2
Daughters’ Behavioral Outcomes.............59
Chapter V. Results .....................................................................................................63
5.1 Data Description ...............................................................................................63
5.2 Testing G
1
Mothers’ Parenting Models ............................................................63
5.2.1 G
1
Baseline Models ......................................................................................63
5.2.2 Results of G
1
Mothers’ Parenting Models....................................................66
5.2.3 Results of Multi-Group Analysis of G
1
Mothers’ Parenting Models ...........72
5.2.4 Summary of Findings of G
1
Mothers’ Parenting Analyses ..........................72
5.3 Testing G
2
Daughters’ Behavior Outcome Models ..........................................73
5.3.1 G
2
Baseline Models ......................................................................................73
5.3.2 Results of G
2
Daughters’ Behavior Outcome Models..................................75
5.3.3 Results of Multi-Group Analysis of G
2
Daughters’ Behavior Outcome
Models.....................................................................................................78
5.3.4 Summary of Findings of G
2
Daughters’ Behavior Outcome Analyses........81
Chapter VI. Discussion ..............................................................................................85
6.1 Major Findings and Reevaluation Of Theories.................................................85
6.1.1 G
1
Mothers’ Parenting Practice Models .......................................................85
6.1.2 G
2
Daughters’ Behavioral Adjustment Models............................................91
6.2 Implications for Social Work Intervention .......................................................95
6.3 Limitations of the Study and Directions for Future Research ..........................96
6.4 Conclusion ........................................................................................................99
References................................................................................................................101
APPENDICES .........................................................................................................118
Appendix A: Distribution of the Major Dependent Variables..............................118
Appendix B: Mother’s Social Support Questionnaire (MSSQ) ............................124
viii
List of Tables
Table 1. Partial correlation coefficients among the study variables 64
Table 2. Model fit comparison for G
1
baseline models 65
Table 3. Model fit comparison for G
1
parenting models 68
Table 4. Parameter estimates for the G
1
parenting models 70
Table 5. Model fit comparison for G
2
baseline models. 75
Table 6. Model fit comparison for the models predicting G
2
behavior 77
adjustment
Table 7. Statistical comparison of G
2
integrative model 79
ix
List of Figures
Figure 1. Integrative model of parenting practice for non-offending 31
mothers of sexually abused girls
Figure 2. Integrative model of structural relationships among maternal 47
factors and the abused and comparison girls’ behavioral
adjustment
Figure 3. Summary of model comparison tests for G
1
parenting baseline 65
models
Figure 4. Summary of model comparison tests for G
1
parenting 67
Figure 5. Standardized parameter estimates from Model 7 71
Figure 6. Summary of model comparison tests for G
2
behavioral 74
adjustment baseline models
Figure 7. Summary of model comparison tests for G
2
behavioral adjustment 76
Figure 8. Unstandardized parameter estimates from multi-group analyses — 82
enjoyment of a child and parental role
Figure 9. Unstandardized parameter estimates from multi-group analyses — 83
authoritarian control
Figure 10. Distribution of G
1
mothers’ parenting – Enjoyment of a child and 118
parental role
Figure 11. Distribution of G
1
mothers’ parenting – Authoritarian control 119
Figure 12. Distribution of G
2
daughters’ behavioral adjustment – 120
Delinquent misbehaviors
Figure 13. Distribution of G
2
daughters’ behavioral adjustment – 121
Aggressive behaviors
Figure 14. Distribution of G
2
daughters’ behavioral adjustment – 122
Withdrawal/depressed behaviors
Figure 15. Distribution of G
2
daughters’ behavioral adjustment – 123
Immature/bizarre behaviors
x
Abstract
The primary goal of this study was to examine parenting practices among
non-offending mothers of sexually abused girls, and to investigate its long-term
impact on their daughters’ behavioral adjustment. While support from non-offending
mothers has been suggested as a potent resilience-enhancing factor in sexually
abused girls’ post-abuse adjustment, rigorous studies in the parenting practices of
these mothers has been lacking, as well as its impact on the abused girls’ adjustment.
In this study, guided by a developmental-ecological perspective of parenting,
the five determinants of parenting were included in the parenting model. Based on
the existing empirical studies, different patterns of relationships among the
determinants were hypothesized and a series of model comparison tests were
conducted. Further, children’s self-reported behavior problems were introduced to
the model and the relationships between maternal factors and their children’s
behavior outcomes were explored.
Methodologically, this study utilized a sample from a multigenerational,
longitudinal study in which both families of sexually abused girls and
demographically-similar comparison families participated. The present study is one
of a few studies to compare maternal impact on sexually abused girls’ post-abuse
adjustment to a demographically-similar comparison sample, as well as to investigate
the impact using multiple informants and over the long term. This multigenerational,
longitudinal approach provided unusual conditions for examining whether and how
xi
maternal factors impact on adaptation among sexually abused children, which
constitute substantial improvements over past research.
The results from a series of model comparison tests indicated that
non-offending mothers’ childhood experiences of punitive discipline, as well as
current dissociative symptoms, were important determinants of parenting practices
among the non-offending mothers. Further, the current findings reiterated that
maternal factors — especially their parenting — had beneficial impact on decreasing
behavior problems among sexually abused girls. The results also showed that, in the
relation between maternal factors and children’s behavioral adjustment, different
causal mechanisms may operate between the sexually abused and their comparison
groups.
The significance of these findings is discussed in the context of theories
concerning childhood trauma and later parenting, intergenerational transmission of
child maltreatment, as well as resilience among abused children. Practical
implications of the findings are also presented and further discussed in terms of
intervention programs for families where intrafamilial child sexual abuse occurs.
1
Chapter I. Introduction
1.1 Background and Significance
Child maltreatment is a serious and devastating societal problem. During
2005, an estimated 3.6 million children in the United States were investigated by
Child Protective Service (CPS) agencies, due to alleged maltreatment (NCANDS;
USDHHS, 2007). An estimated 899,454 children were determined by the agencies to
be victims of child abuse and neglect. Child sexual abuse constituted approximately
10% (n = 83,310) of the total child maltreatment cases substantiated by CPS
agencies. According to CPS, child sexual abuse was the third most common form of
child maltreatment, following neglect and physical abuse.
It is alarming to find that as many as 83,310 children experience sexual abuse
during one year. To make matters worse, these data are very stringent estimates,
because they include only the cases reported to and substantiated by State Child
Protective Services. Given that child sexual abuse is largely underreported to
authorities (Elliott & Briere, 1992), and has the lowest substantiation rate among all
CPS investigations (English, Marshall, Coghlan, Brummel, & Orme, 2002), it is
likely that the published statistics represent only the tip of the iceberg.
In fact, in the latest National Incidence Study (NIS-3), 300,200 children were
found to have been sexually abused during 1993, with an incidence rate of 6.8 per
1,000 children for females. A recent meta-analysis of 16 community sample surveys
estimates the prevalence of child sexual abuse as being 16.8% for women and 7.9%
for men in the United States (Gorey and Leslie, 1997), after adjusting for sample
2
related variation, response rates, and differences in definitions. It is clear that a
substantial proportion of children and adolescents growing up in the United States
experience sexual trauma during childhood.
In addition to the pervasiveness of the problem, many new studies in the last
30 years have documented that the effects of childhood sexual abuse are largely
deleterious and encompass a wide array of physical, psychological, behavioral, and
interpersonal domains of human functioning (see, for reviews, Beitchman et al.,
1992; Keldall-Tackett, Williams, & Finkelhor, 2001). The effects of childhood
sexual abuse are found to be long lasting, and its negative influences can persist
through various points of development (see, for review, Trickett & Putnam, 1998).
Further, the emerging evidence is beginning to show that these effects of childhood
sexual abuse are not merely probabilistic risk factors, but also causal mechanisms for
suicide attempts (Nelson et al., 2002), substance abuse (Dinwiddie, Heath, Dunne, &
Bucholz, 2000; Kendler et al., 2000), psychiatric challenges (Kendler et al., 2000)
and re-victimization (Noll, Bonanno, Trickett, & Putnam, 2003).
Given the magnitude and pervasiveness of the problem, it is very important to
understand the adjustment patterns of sexual abuse victims, and to apply such
knowledge strategically to facilitate positive adaptations among the abused children.
Child sexual abuse research suggests that the reactions and support non-offending
mothers give to their sexually abused children following their victimization are
particularly influential in predicting adjustment outcomes. In fact, Spaccarelli and
Kim (1995) suggest that support from primary caregivers is the single most
3
important determinant of sexually abused children’s resilience (i.e., positive
adaptation in spite of high risk) (Fraser, Richman, & Galinski, 1999; Masten, 2001).
Emerging evidence suggests that parental support may be a more potent determinant
of abused children’s successful adaptation than sexual abuse characteristics per se
(e.g., perpetrator, age of onset, duration) (Fromuth, 1986; Johnson & Kenekl, 1991;
Tremblay, Herbert, & Piche, 1999).
While it is possible that maternal support may protect sexually abused
children from maladjustment, an emerging body of literature on non-offending
mothers of sexually abused children also indicates that these mothers face
considerable challenges following the disclosure of their children’s abuse, and may
need support themselves. These challenges include psychological distress (e.g.,
Lewin & Bergin, 2001; Newberger, Gremy, Waternaux, & Newberger, 1993), the
mothers’ own childhood abuse histories (e.g., Kim, Noll, Putnam, & Trickett, in
press), and dysfunctional family environments (e.g., Avery, Hutchinson, & Whitaker,
2002; Kim et al., in press), which have been strongly suggested in parenting
literature as being important determinants of parenting behavior.
Moreover, studies of childhood sexual abuse consistently report an
intergenerational association between mothers’ childhood sexual abuse histories and
their daughters’ childhood sexual abuse experiences (Avery, Hutchinson, &
Whitaker, 2002; Briere & Runts, 1988; Browne & Finkelhor, 1986; Deblinger et al.,
1993; Egeland, Jacobotiz, & Papatola, 1987; Faller, 1989; Finkelhor et al., 1997;
Kaufman & Zigler, 1987; Leifer, Shapiro, & Kassem, 1993; McCloskey & Bailey,
4
2000; Oates et al., 1998; Spieker et al., 1996; Zuravin et al., 1996). Some studies
indicate that a mother’s childhood sexual abuse history may be one of the most
important factors increasing the likelihood that her child will experience sexual
abuse during childhood (Finkelhor et al., 1997; McCloskey & Bailey, 2000).
Emerging literature in this area suggests that parenting practices affected by a
mother’s childhood sexual abuse experience may be an important intergenerational
mechanism (Burkett, 1991).
However, it is largely unknown whether there exist risk factors associated
with the non-offending mothers which pose additional burdens on their abused
daughters’ development and, if so, what they may be (Noll, 2005). Except for a few
descriptive findings on the parenting characteristics of these mothers, information on
parenting practice among this population is extremely limited.
Therefore, in order to boost the likelihood of resilience in sexually abused
children, it is very important to understand parenting functioning among mothers of
sexually abused children, and how their parenting styles affect the ways in which
their abused children adjust post-abuse. Knowledge of how maternal factors affect
adjustment among the abused children is vital for informing intervention efforts that
seek to alleviate the adverse effects of child sexual abuse on development of the
children. Intervention elements addressing mothers’ needs may supplement a number
of existing treatment programs and family preservation programs for these children.
5
1.2 Purpose of the Research
The specific aims of the current study are as follows:
Specific Aim 1:
To understand parenting practices among non-offending mothers of sexually
abused children
a. To examine structural relationships among maternal risks, resources,
and parenting practices
b. To employ multi-group analysis to examine whether the pattern of
relationships in non-offending mothers of sexually abused girls differ
from those in mothers of non-abused comparison girls
Specific Aim 2:
To understand the relationship between non-offending mothers’ parenting
practices and their abused daughters’ behavior outcomes
a. To examine the impact of mothers’ parenting practices on their abused
children’s behavioral adjustment in the context of various maternal
risks and resources
b. To employ multi-group analysis to examine whether the pattern of
relationships between maternal factors and children’s behavioral
outcomes in the abuse group differs from that in the non-abuse
comparison group
6
Chapter II. Parenting Practices of Non-Offending
Mothers of Sexually Abused Girls
When child sexual abuse occurs, mothers are often blamed for failing to
protect their daughters from the perpetrators. In the case of intrafamilial sexual abuse,
even more negative attributions are given to these mothers. Historically, these
mothers are described in many clinical reports as being “collusive”, meaning that the
mothers play a central role in the onset, development, and maintenance of child
sexual abuse in the family. Upon disclosure, typically, these mothers are thought to
deny abuse allegations and fail to provide appropriate protection and support for
their abused daughters.
In family systems perspective, the “collusion” is explained in the context of
the mother’s defense against loss she experienced during her own childhood.
According to this perspective, mothers who had needs for love unmet by their own
mothers demand that their daughters play a mother-wife role in the family and,
consciously or unconsciously, allow their children to be molested by their male
partners (Lustig et al., 1966; Zuelzner & Reposa, 1983).
In the feminist perspective, the mothers of sexually abused girls are viewed
as victims of domineering male perpetrators’ violent tyranny at home (Herman,
1992; Wattenberg, 1985; Cammaert, 1988). The “collusion” in this perspective is
interpreted as a manifestation of “powerless and helpless” states of these victimized
mothers (Cammaert, 1988; Herman, 1992; Russell, 1986; Wattenberg, 1985).
7
The family systems perspective emphasizes individual histories and
role-reversed family dynamics, while the feminist perspective turns its attention to
societal factors like gender dynamics. However, these two perspectives convey a
very similar stereotypic image of the mothers of sexually abused girls: the mothers
are depicted as dependent, powerless and helpless women incapable of protecting
their daughters (Joyce, 1997).
Yet, when child sexual abuse is disclosed, the mothers are also suggested as
being “the subject of hope for rescuing the victim” (Nakhle Tamaraz, 1996, p.76),
and are supposed to take primary responsibility for helping their abused children’s
psychosocial adjustment. Moreover, child-welfare professionals use the mothers’
reactions to their children’s sexual abuse, such as the level of belief and
supportiveness they exhibit upon disclosure, as critical information for deciding
whether to place the abused children in out-of-home care (Cross, DeVos, &
Whitcomb, 1994; Hunter, Coulter, Runyan, & Everson, 1990). Research has found
that lower levels of maternal support are associated with mothers losing custody of
their children (Elliott & Briere, 1994) and placement in foster care (Leifer, Shapiro,
& Kassem, 1993). Family disruptions often exacerbate the guilty feelings of child
sexual abuse victims, which in turn compound their adjustment outcomes (Pintello &
Zuravin, 2001). In fact, some studies suggest that support from non-offending
mothers is the single most important determinant of how well sexually abused
children successfully adapt (Fromuth, 1986; Spaccarelli & Kim, 1995; Tremblay,
Herbert, & Piche, 1999).
8
2.1 Psychosocial Characteristics of Non-Offending Mothers of
Sexually Abused Girls: Empirical Findings
In contrast to the attention these mothers receive when their daughters’ sexual
abuse is disclosed, empirical research has been lacking in psychosocial functioning
of non-offending mothers of sexually abused children. Empirical studies which
examine these mothers in their own right are recent and, so far, scant (Elliott &
Carnes, 2001; Nakhle Tamaraz, 1996). We know little about the psychosocial
characteristics of these mothers, and for the most part their own challenges and needs
have been relatively overlooked.
Some studies challenge the earlier mother-blaming propositions either
theoretically (e.g., Herman, 1981; Wattenberg, 1985) or empirically (e.g., Deblinger
et al., 1993). In their recent literature review, Elliott and Carnes (2001) report that
the majority of mothers are supportive and protective, although a substantial number
are not. According to their review, the reported percentage of support/protection that
non-offending caregivers provide following the disclosure of child sexual abuse
ranges from 27% to 69%.
In terms of childhood histories, a number of studies have reported a
disproportionately higher rate of childhood sexual abuse experiences among
non-offending mothers of sexually abused girls. Two studies suggest that a mother’s
personal childhood sexual abuse history may be one of the most important factors
increasing the likelihood that her child will experience sexual abuse during
childhood (Finkelhor et al., 1997; McCloskey & Bailey, 2000). Mothers’ childhood
9
sexual abuse experience was also found to be an important factor in understanding
the mothers’ own psychosocial challenges (Kim et al., in press).
The psychological functioning of non-offending mothers has received more
empirical attention than other areas of maternal functioning. The findings of the
existing studies generally support the hypothesis that non-offending mothers
experience considerable psychological distress following the disclosure of their
daughters’ sexual abuse. Compared to a normative sample of mothers, non-offending
mothers exhibit a heightened level of general symptom distress (Deblinger et al.,
1993; Newberger et al., 1993). Empirical studies also found higher levels of
depression and anxiety (Kim et al., in press; Lewin & Bergin, 2001), as well as
elevated symptoms of PTSD (Green et al., 1995) among this population.
Families in which child sexual abuse occurs have been described as less
cohesive (Jackson et al., 1990), and as having high levels of family stress (Gale et al.,
1988). Other characteristics found to be elevated in this population include domestic
violence (Avery, Hutchinson, & Whitaker, 2002; Kim et al., in press; Herman &
Hirschman, 1981), and substance abuse (Kim et al., in press; Shah, Dali, & Heinrichs,
1995). Overall, the existing findings suggest that the non-offending mothers of
sexually abused children likely confront multiple challenges, while they have limited
familial psychological, emotional, and social resources to facilitate their coping with
stressful events and everyday adversities.
10
2.2 Descriptive Studies on Parenting Functioning of
Non-Offending Mothers of Sexually Abused Girls
Recently, Lewin and Bergin (2001) conducted an observational study and
found that mothers of sexually abused children showed diminished maternal
attachment behaviors compared with mothers of comparison girls on all four
dimensions that have been suggested as influencing the quality of attachment (i.e.,
acceptance, sensitivity, cooperation, and accessibility) (Ainsworth et al., 1978). In a
similar vein, Leifer and her colleagues (2001; 2004) suggest that these mothers may
have experienced fewer positive relationships with their own mothers during their
childhoods, as well as more severe childhood maltreatment than the mothers of
comparison girls. They found that mothers’ experiences of discontinuity of care
during childhood, as well as current negative or ambivalent attachment relationships
with their own mothers, were in turn significant predictors of their daughters’ sexual
abuse experiences (Leifer, Kilbane, Jacobson, & Grossman, 2004) and their own
supportive behavior toward their abused children (Leifer, Kilbane, & Grossman,
2001).
Other than that, however, rigorous empirical studies examining the parenting
characteristics of non-offending mothers are rare. The existing studies suggest that
non-offending mothers experience some psychosocial challenges following the
disclosure of the child sexual abuse. They experienced more childhood adversities
than mothers of non-abused children. However, the interrelationships among
maternal developmental histories, current functioning, and parenting, have not been
11
rigorously researched. Knowledge of determinants of parenting practices among this
population is extremely limited.
2.3 The Developmental-Ecological Perspective of Parenting
To examine parenting practice among non-offending mothers of sexually
abused girls, this study was guided by a developmental-ecological perspective of
parenting (Belsky, 1984; Luster & Okagaki, 2005). The root of this perspective is
traced back to Bronfenbrenner’s (1979) ecological approach to human development,
which highlights the embeddedness of individuals within a larger, multi-layered set
of systems, beginning with the surrounding macrosystem (e.g., culture, societal
values, policy), followed by exosystems (e.g., schools, community), and culminating
in the intimate microsystem of the immediate family.
Belsky (1984) discussed the determinants of parenting practice in a way
similar to Bronfenbrenner’s, with four nested, interactive systems of influence. These
include mothers’ ontogenetic origins, and the personal psychological resources they
bring to parenting situations; including, for example, mothers’ history of being
abused, how they were parented, and whether they have tenuous mental health. The
microsystem highlights the immediate family environment involving family stress,
spousal support, and the child’s characteristics. The exosystem includes elements
surrounding the immediate family, and involves social networks and support, as well
as neighborhoods and community. Finally, the largest layer indicates the
macrosystem. Examples of these levels include cultural sanctioning of parenting
12
behaviors, values, and attitudes, and social policy influencing family and
childrearing.
The developmental-ecological perspective of parenting provides a
predominant set of constructs for this study in examining the parenting practices of
non-offending mothers of sexually abused girls. Guided by the perspective, this
study focuses on four determinants of parenting behaviors among non-offending
mothers of sexually abused girls: 1) mothers’ childhood sexual abuse experience; 2)
mothers’ childhood experiences of punitive discipline by their own mothers; 3) the
level of mothers’ current dissociative symptoms; and 4) social support mothers
receive from their family-of-origin members.
2.3.1 Mothers’ Childhood Sexual Abuse Experiences and Parenting
Maternal history of childhood sexual abuse is an important factor for
understanding parenting among non-offending mothers of sexually abused girls. The
parenting practices of abused mothers are suggested as an important
intergenerational mechanism in child sexual abuse. The existing body of literature on
child sexual abuse provides evidence that the parenting styles of child sexual abuse
survivors may differ from those of mothers with no known childhood abuse histories.
As early as the 1980s, clinicians were touching upon how childhood sexual
abuse survivors expressed concerns about themselves as parents (Gelina, 1983;
Herman, 1981; Sroufe et al., 1985). Gelina (1983) noted that survivors had
difficulties maintaining a balance between discipline and affection with their
13
children, and expressed a desire to avoid motherhood because of their negative
experiences growing up. Herman (1981) also observed that survivors seemed unable
to obtain satisfaction from parenting and had unreasonable expectations for their
children. Similarly, Sroufe, Jacobvits, Mangelsdorf, DeAngelo, and Ward (1985)
observed 18 mothers reporting a history of familial child sexual abuse, and found
that the mothers were more likely to interact with their sons in a subtly seductive
manner, which could be indicative of generational boundary resolution.
Throughout the 1990s, some studies investigated the bivariate association
between mothers’ childhood sexual abuse histories and their parenting behaviors
(Burkett, 1991; Cohen, 1995; Lyons-Ruth & Block, 1996). These studies also
support a relationship between childhood sexual abuse trauma and the parenting
behaviors of abuse survivors. Burkett (1991) conducted semi-structured interviews
with 20 mothers who had been sexually abused as children, and observed their
interactions with their own children. The results showed that the abused mothers
focused more on themselves than on their children’s needs, talked about their
children as if they were their close friends, and relied more on their children for
emotional caretaking. Lyons-Ruth and Block (1996) studied 45 mothers who had
experienced sexual abuse trauma during childhood, in which they observed the
mothers with their 18-month-old infants. The researchers found that the maternal
history of childhood sexual abuse was significantly associated with restricted
maternal affect and decreased involvement with the infants. The mothers with
childhood sexual abuse trauma spent less time in the room with their infants, were
14
less engaged when with them, and tended to show flat affect in relation to their
infants. Finally, using a standardized parenting measure (Parenting Skills Inventory;
Nash & Morrison, 1984) Cohen (1995) compared 26 mothers with childhood sexual
abuse trauma and 28 mothers with no such history. On all seven subscales of the
parenting skill measure (i.e., role image, objectivity, expectations, rapport,
communication, limit-setting, and role support), they found a significant group
difference, characterized by a tendency of the abused mothers to be less skillful in
maternal functioning than the non-abused mothers. The most salient differences were
detected in role-support, communication, and role image.
Only very recently has systematic inquiry using multivariate analysis
methods emerged in this area. Cole, Woolger, Power, & Smith (1992) compared
group means of parenting dimensions among incest survivor mothers, mothers whose
fathers were alcoholic but not sexually abusive, and mothers with no known risk.
They found that the incest survivor mothers scored significantly lower in the
consistency and maturity demands dimensions of parenting than mothers whose
fathers were alcoholic.
In a similar study, Banyard (1997) showed that mothers with childhood
sexual abuse trauma had a more negative view of themselves as parents, and reported
more frequent use of physical strategies to resolve conflicts with their children.
Using a community sample of 516 low-income mothers, the study found that these
relationships held true even after controlling for other negative experiences in their
15
families-of-origin, such as neglect, physical abuse, and relationship quality with their
own parents.
Using a clinical sample, Ruscio (2001) found that mothers with a history of
childhood sexual abuse trauma tended to be more permissive in their parenting
practices than mothers in the normative sample (Parenting Practices Questionnaire;
Robinson, Mandelco, Olsen, & Hart, 1995). Lower rates of authoritative parenting
were also found among mothers whose personal childhood sexual abuse included
penetration, whereas no mean differences were detected between mothers with
childhood sexual abuse trauma and the normative sample. In subsequent analyses,
the researchers failed to find unique relationships between mothers’ childhood sexual
abuse trauma and permissive parenting after controlling for physical abuse and the
experience of growing up with an alcoholic in the home, whereas childhood sexual
abuse trauma retained its negative association with a lower rate of authoritative
parenting.
One study (Zuravin & Fontanella, 1999) showed that the parenting
differences between mothers with childhood sexual abuse trauma and mothers
without such trauma were completely accounted for by the mothers’ other childhood
experiences, such as physical abuse and neglect. Childhood sexual abuse experience
had significant bivariate relationships with a tendency to use violent tactics toward
children, and a sense of competency as parents, but they failed to maintain a
significant relationship once a set of growing-up experience variables was introduced
to the equation.
16
Overall, the existing studies suggest that mothers with a personal history of
childhood sexual abuse trauma may parent differently from mothers without such
trauma, in terms of generational boundary setting, warmth and caring, using more
physical strategies to resolve conflict with their children, and being less consistent,
or less organized. These characteristics could be interpreted as a possible indication
of the mothers’ distancing themselves from the demands of parenting. However, it is
questionable whether the observed parenting differences are attributable to childhood
sexual abuse in and of itself, or whether they are a function of other negative
growing-up experiences which co-occurred with the childhood sexual abuse, or both.
In addition, some recent studies suggest that a mother’s childhood sexual
abuse experience exerts its influence through trauma-related symptoms. So far, the
empirically-examined mediators or moderators include depression (Roberts et al.,
2004; Zuravin & Fontanella, 1999), anxiety (Roberts et al., 2004), overall
psychological functioning (Marcenko et al., 2000), anger (Dilillo, 2000), the quality
of relationship with one’s partner (Alexander et al., 2000; Cole et al., 1992) and
social support (Marcenko et al., 2000; Meyers et al., 2003; Ruscio, 2001).
With regard to mothers’ psychological functioning, studies using community
samples report a significant mediating effect of anxiety (Roberts et al., 2000) and
anger (Dilillo, Tremblay, & Peterson, 2000) on the relationship between mothers’
early sexual abuse trauma and later parenting practices. In contrast, Zuravin and
Fontanella (1999) examined 516 low-income, urban mothers, and failed to find a
significant mediation effect of depression on the relationship between childhood
17
sexual abuse experience and the frequency of use of verbal abuse, severe violence
and perceived level of parenting competence. Depression was independently
associated to the mothers’ frequency of verbal abuse and perceived level of parenting
competence.
Study findings are also inconsistent with regard to the function of mothers’
social support. Using a community sample of mothers with 5- to 8-year-old children,
Alexander, Teti, and Anderson (2000) found that the relationship between mothers’
childhood sexual abuse trauma and parenting was significant only when the mothers
were dissatisfied with their male partners. Abused mothers who enjoyed satisfying
relationships with their male partners did not report engaging in emotionally
dependent relationships with their children. Using a sample referred from outpatient
mental health services, Ruscio (2001) also found a significant indirect effect of social
support on the relationship between mothers’ childhood sexual abuse trauma and less
authoritative parenting.
In contrast, two studies failed to find a significant mediating function of
social support on the relationship between mothers’ childhood sexual abuse trauma
and parenting. In a study of adolescent mothers (Meyers & Battistoni, 2003), social
support had a direct effect on parenting, independent of childhood sexual abuse.
Marcenko’s study of TANF (Temporary Assistance for Needy Families) mothers
(2000) reports that the level of social support was not associated with either mothers’
childhood sexual abuse trauma or with their parenting attitudes.
18
Overall, most of the studies investigate how mothers’ childhood sexual abuse
experience exerts its negative influence through their psychological functioning or
social support. But the results of the various studies do not yield a coherent picture of
the functions of mothers’ psychological functioning and social support on the
relationship between their childhood sexual abuse experiences and parenting styles.
There may be several reasonable explanations for these inconsistencies. First,
many of the studies contain relatively high-risk samples (e.g., low-income
welfare-dependent mothers or adolescent mothers) and have no meaningful
comparison groups. Seemingly, under multiply-challenged situations such as teen
parenthood and severe poverty, neither psychological functioning nor social support
has mediating effects. They independently contribute to mothers’ parenting, or are
not a significant predictor for mothers’ parenting behaviors. But the number of
studies that examined these effects among mothers with childhood sexual abuse
histories is limited, and any definitive conclusions for the function of suggested
mediators are premature.
Moreover, there are inconsistencies in the operational definitions of parenting.
The parenting domains investigated in the existing literature vary from mothers’
parenting confidence and overall parenting skills, to specific strategies for discipline
and resolving conflicts. The existing studies have also investigated the determinants
of abused mothers’ parenting in isolation of important factors (e.g., childhood
parenting experiences as well as current risks and resources).
19
Based on the existing findings, this study examines three competing
hypotheses predicting the effects of the maternal childhood experience of sexual
abuse on their later parenting. The first hypothesis is that mothers’ parenting results
directly from their childhood sexual abuse experiences. This hypothesis has received
the most research attention over the years. The second hypothesis is that the
association between maternal childhood sexual abuse experience and parenting is
spurious. In other words, parenting practices are caused by other childhood
upbringing experiences, such as experience with one’s own parents or
family-of-origin, which co-occurred with childhood sexual abuse. Although not as
popular as the first hypothesis, support can be also found for this perspective. Finally,
this study considers a third hypothesis that rivals both of the previous hypotheses:
that the connection between mothers’ childhood sexual abuse experience and later
parenting is mediated by current risk and resources, such as the mothers’
psychological functioning and the level of social support available to them.
2.3.2 The Notion of Intergenerational Transmission of Parenting Practices
There is a strong commonsense notion of continuity in parenting behaviors
across generations. Intergenerational continuity in parenting is defined as the degree
to which past experiences with one’s parents during childhood is reproduced during
interactions with one’s children (Scaramella & Conger, 2003). As Maccoby and
Martin (1983) note, social learning theory has provided the predominant theoretical
explanations for understanding the effect of parenting experiences in the
20
family-of-origin context on later parenting behaviors. That is, children learn specific
parenting behaviors in the social context of their family-of-origin via modeling and
identification. Such behaviors become their parenting repertoire, and consciously or
unconsciously unfold later when they parent their own children.
Empirically, the existing studies have addressed this hypothesis in extreme
forms of parenting behavior, such as harsh and abusive parenting. A substantial
component of this research has supported the intergenerational hypothesis. For
example, earlier works provide evidence for intergenerational similarities with
regard to harsh parenting (Simons, Whitebeck, Conger, & Chyi-In, 1991), parental
rejection (Whitebeck, Hoyt, Simons, Conger, Elder, Lorenz, & Huck, 1992), and
generational boundary disturbances (Jacobvitz, Morgan, Kretchmar, & Morgan,
1991). Recent studies using multigenerational, prospective designs also yield
findings similar to previous research. Based on the Iowa Family Transition Project,
Scaramella and Conger (2003) report an intergenerational relationship in hostile
parenting. In the Rochester Youth Development Study, Thornberry and his
colleagues (2003) utilized a measurement model of parenting which consisted of the
parent-child affective tie and consistent discipline, and found a significant direct
effect of parenting from one generation to the next.
However, intergenerational continuity of parenting is not deterministic.
Social learning theory has undergone considerable restructuring, in which strong
emphasis has been placed on cognitive processes (Maccoby & Martin, 1983).
Likewise, the suggested effect size of the intergenerational continuity coefficients
21
among the existing empirical studies is moderate to modest (Dubow, Huesmann, &
Boxer, 2003) at best. Depending on the context, findings for continuities and
discontinuities are mixed in existing studies.
To illustrate, Scaramella and Conger (2003) found that a child’s negative
emotional reactivity conditions the intergenerational continuity of hostile parenting.
Specifically, when the child’s level of negative emotional reactivity was high, the
relationship between grandparents’ and parents’ hostility was significant. But under
the condition of a lower level of negative emotional reactivity, no such association
was detected. The researchers speculate that, in a situation of high parenting stress,
parents are more likely to react to their child’s behavior with automatic behaviors
learned in their families-of-origin.
The current study attempts to examine parenting practice among
non-offending mothers of sexually abused girls, using the mothers’ experiences
during their own childhoods as one of the determinants of their current parenting
practices. Based on the literature on intergenerational transmission of hostile
parenting, this study hypothesizes that the mothers’ childhood experiences of their
own mothers’ discipline strategies may be a significant predictor of their current
parenting styles. Under such highly stressful situations as the disclosure of their
daughters’ child sexual abuse, the association may be stronger. This study also
expects that the intergenerational association of parenting behaviors may differ
according to the daughters’ childhood sexual abuse status: the magnitude of
22
intergenerational continuity may be greater among the mothers of sexually abused
girls than among mothers of non-abused girls.
2.3.3 Dissociative Symptoms and Parenting Behavior
History alone does not determine later parenting functioning. Mothers’
contemporaneous factors are more important than, or at least as important as, their
childhood experiences. Parenting theorists suggest that maternal psychopathology is
a potent determinant on mothers’ parenting practices (e.g., Belsky, 1984).
This study focuses on maternal dissociative symptoms. According to the
DSM-IV-TR (APA, 2000), the essential feature of dissociative disorder is a disruption
in the usually integrated functions of consciousness, memory, identity, or perception.
The primary symptoms include amnesia, depersonalization, derealization, identity
confusion, and identity alteration. According to the trauma model, dissociation is a
strategy for dealing with severe, chronic childhood trauma and has been recognized
as being closely linked with early adverse childhood experiences (Ross, 1999).
Further, the trauma literature suggests that pathological dissociative
tendencies may be an important link to understanding higher rates of revictimization
among childhood abuse survivors, as well as subsequent experiences of violence of
their offspring (Chu, 1992; Noll et al., 2003). It suggests that damaged vigilance
systems, such as pathological dissociative tendencies of child abuse victims, may
serve to impair their judgment, increase the likelihood of misinterpreting dangerous
cues, and lead to revictimization. It is also possible that mothers who were abused as
23
children may be prone to making inaccurate or uninformed decisions about potential
dangers to their children, which may contribute to sexual abuse being developed and
maintained within the family. Given the high rates of childhood sexual abuse and
other abuse experiences among non-offending mothers of sexually abused children,
the degree to which dissociative symptoms contribute to parenting may have
important implications among these families in understanding intergenerational
phenomena of child sexual abuse.
While there exist rich descriptions of how depressed mothers parent
differently from non-depressed comparison mothers (Downey & Coyne, 1990), there
are extremely few studies on the parenting of mothers with dissociative symptoms.
Kluft (1987) provided an anecdotal preliminary report for possible parenting
differences among this population. Benjamin and her colleagues (1996) also
examined parenting experiences among dissociative patients in inpatient and partial
hospital facilities. They found that mothers with dissociative disorder scored lower
than control mothers and mothers with other psychiatric challenges in factors such as
supportive discipline, showing affection, attachment behaviors, and acting to
promote the child’s development. The qualitative study provides further description
of how the dissociative symptoms manifest in interactions with children (Benjamin,
Benjamin, & Rind, 1998). The mothers with dissociative disorders expressed fear of
parenting, and described themselves as hostile and less affective parents. In addition,
these mothers reported occasionally having cognitive distortions about their maternal
identity and being dependent on their children in daily functioning.
24
Given these reported parenting experiences of mothers with dissociative
disorder, one may speculate that the parenting behaviors of these mothers reflect the
affective and cognitive symptoms that characterize dissociation. However, it should
be noted that the researchers utilized a clinical sample with very severe forms of
dissociation, as a function of recruiting the participants in in-patient or partial
psychiatric hospital facilities. Thus, any conclusion concerning the parenting
characteristics of these mothers is tentative.
A couple of studies examined the function of dissociative symptoms on
parenting practices in a community sample. Vezina, Mireille, Robert, and Pierre
(2005) examined the role of depression and dissociation in the relationship between
child sexual abuse and later parenting practices. Their results revealed that
dissociation predicted all parenting dimensions examined in the study, including the
use of less positive reinforcement, lack of monitoring and supervision of the child,
inconsistency in applying discipline, and use of corporal punishment. Dissociation
also mediated the association between childhood maltreatment (physical abuse,
emotional abuse and neglect) and inconsistency in applying discipline.
Similarly, using a college sample, Narang and Conteras (2000) found a
significant relationship between childhood physical abuse experience and the current
level of dissociation. The results also revealed that dissociation mediated the
relationship between childhood physical abuse experience and the current level of
physical abuse potential, as measured by a self-report questionnaire. In summary,
these two pioneering studies provide some evidence that dissociation may be an
25
important mechanism by which parents’ trauma experiences are communicated to
the next generation.
Based on the extant preliminary studies on the relationship between
dissociation and parenting behaviors, this study expects that the level of dissociative
symptoms may be significantly and positively associated with the mothers’ current
parenting practices. The study also expects that maternal dissociation would mediate
the relationship between mothers’ earlier childhood experiences of sexual abuse and
punitive discipline, and their later parenting behaviors.
2.3.4 Social Networks Surrounding Families
Personal social networks consist of “those people outside the household who
engage in activities and exchanges of an affective and/or material nature with
members of the immediate family” (Cochran & Brassard, 1979, p.601). Social
support refers to those activities and exchanges that lead an individual to believe that
s/he is cared for, loved, and valued, and that s/he belongs to a network of
communication and mutual obligation (Cobb, 1976).
Social support has been suggested to be an important determinant of
parenting (e.g., Belsky, 1984; Cochran & Niegro, 1995; Crockenberg, 1988). The
parenting literature suggests that the connections mothers have to individuals in their
social environments can convey social expectations and serve as guidelines for
appropriate and inappropriate parenting behavior. Empirical studies along this line of
reasoning have found that a lack of social embedding (i.e., social isolation) is often
26
linked to abusive parenting practices (Corse, Schmid & Trickett, 1990; Polansky,
Ammons, & Gaudin, 1985). Support provided to mothers can directly reduce the
number of stressful events that they experience. Mothers’ perceptions of being cared
for and loved, and having available confidants to whom they can turn in stressful
situations, are suggested to have a direct and positive influence on their nurturing
parenting behavior, and on their ability to cope with everyday adversities (Crittendon,
1985).
Overall, a number of studies show that mothers with higher levels of social
support are generally more nurturing and consistent in their parenting, and less likely
to use punitive strategies such as scolding and ridiculing (for review, see McLoyd,
1990). They report that social support is related to more sensitive and accepting
parental behaviors, and enhances parenting behavior by providing additional
resources for childcare assistance, emotional guidance, and economic support (e.g.,
Jarrett, 1995; Mason, Cauce, Gonzales, Hiraga, & Grove, 1994; Taylor & Roberts,
1995; Unger & Watersman, 1985).
Increases in social contact, social interaction, and the provision of social
resources are not always associated with better parenting outcomes, however. The
effectiveness of social support is not equivocal, but depends on population
characteristics and the specific sources and types of social support. For example,
studies of adolescent mothers recognize that the social support these mothers receive
helps them to provide better parenting environments for their children. But the same
studies also suggest that support from their male partners is less beneficial for teen
27
mothers’ well-being and parenting practices than for their adult counterparts
(Shapiro & Mangelsdorf, 1993). Moreover, because of the developmental issues in
adolescence, studies have found that the family-of-origin is not only a source of
support for these mothers, but also a source of conflict (Davis, Rhodes, &
Hamilton-Leaks, 1997; Dukewich, Borkowski, & Whitman, 1996; Richardson,
Barbour, & Bybenzer, 1991; Voight, Hans, & Bernstein, 1996).
Lyons, Henly, and Scherman (2005) examined the influence of informal
social support among families receiving child welfare services. The results generally
showed that social support had a positive influence on depression, financial strain,
and negative life events, and that it enhanced positive parenting. However, mothers
with more informal support available tended to use more negative parenting practices
as well. The effect of depression on mothers’ negative parenting practices was
greater for mothers with more informal support. The results may suggest that the
effectiveness of relying on informal support might be compromised if the supporters
in the network share the same parenting knowledge, values and experiences as
families receiving child welfare services.
In particular, the present study focuses on social support non-offending
mothers receive from their family-of-origin members. This dimension of social
support is particularly important among this population. Leifer and her colleagues
(2001; 2004) suggest that quality relationships and actual support from the
non-offending mothers’ own mothers are significant predictors of the non-offending
mothers’ supportive behaviors, as well as an important mechanism explaining
28
intergenerational phenomena in child sexual abuse. On the other hand, Trickett and
Susman (1989) found that mothers who experienced physical abuse in childhood, but
who do not abuse their children, demonstrated a clear perception that they parent
differently from their own parents, and received less support from their parents.
However, this logical pattern was not found among the intergenerational group (i.e.,
abused parents who also abuse their children). The study results indicate that social
support from family-of-origin members in this population may function differently
from that of non-abused counterparts. Given the high rate of child maltreatment
experiences that have been reported among non-offending mothers, it is also
presumed that the beneficial effects of support from family-of-origin members may
be compromised among this population.
In summary, based on the existing literature, the current study hypothesizes
that the social support mothers receive from their family-of-origin may be
significantly, positively associated with the mothers’ parenting practices. With
regard to the moderating effects of daughters’ sexual abuse status, competing
hypotheses predict positive or non-significant effects of social support from the
family-of-origin on parenting among non-offending mothers of sexually abused girls.
The first hypothesis is based on Leifer and her colleagues’ studies, which emphasize
the positive function of support from family-of-origin members among this
population. In contrast, based on the high rates of adverse childhood experiences
among these mothers, it is also expected that the beneficial effects of social support
from family-of-origin members may be compromised.
29
2.4 An Integrative Model of Parenting Practice for Non-Offending
Mothers of Sexually Abused Girls
To examine the contribution of mothers’ childhood experiences (i.e., history
of being sexually abused during childhood, mothers’ own parents’ punitive
discipline) and current functioning (i.e., dissociative symptoms and social support
from family-of-origin sub-network) to their parenting practices, an integrative model
of the suggested relations among the variables was specified. The specification of the
model emanated from review of the existing empirical studies, as well as conceptual
hypotheses posited by Belsky (1984). A graphic representation of this model is
displayed in Figure 1.
This model makes explicit a number of hypotheses, including:
Hypothesis 1 (Path a, b, c, d & e):
Mothers’ parenting practices will be predicted by their childhood sexual
abuse experiences, their own mothers’ discipline strategies, their current
dissociative symptoms, and the amount of social support they receive from
their family-of-origin, as well as their daughters’ childhood sexual abuse.
Hypothesis 2 (Path f,g,h & i):
The relationship between mothers’ childhood sexual abuse and their current
parenting practices, as well as between punitive discipline experience in
childhood and current parenting, will be mediated by the level of dissociative
30
symptoms, as well as by the amount of social support they receive from their
family-of-origin.
Hypothesis 3 (Path c
(A)&(C)
, b
(A)&(C)
):
The relationships between their own mothers’ discipline strategies and the
non-offending mothers’ current parenting practices, as well as between social
support from their families-of-origin and their current parenting practices,
may differ according to their daughters’ sexual abuse status.
31
[Abuse Group]
[Comparison Group]
Figure 1. Integrative model of parenting practices for mothers in abuse and comparison
groups. Note. A (Abuse) and C (Comparison) in parentheses indicate a parameter varies
over groups.
Mothers’
Childhood
Sexual Abuse
History
Mothers’
Dissociative
Symptoms
Mothers’
Parenting
Practices
b
(A)
a
c
(A)
d
Daughters’
Childhood
Sexual Abuse
Mothers’
Own Mothers’
Discipline
Punitiveness
Social Support
Mothers Receive from
Family-of-Origin
Sub-Network
g
h
i
f
e
Mothers’
Childhood
Sexual Abuse
History
Mothers’
Dissociative
Symptoms
Mothers’
Parenting
Practices
b
(C)
a
c
(C)
d
Daughters’
Childhood
Sexual Abuse
Mothers’
Own Mothers’
Discipline
Punitiveness
Social Support
Mothers Receive from
Family-of-Origin
Sub-Network
g
h
i
f
e
32
Chapter III. Non-Offending Mothers’ Parenting and
Abused Girls’ Behavioral Adjustment
The resilience framework provided basic theoretical grounding for this study,
which explores the relationship between the parenting practices of non-offending
mothers and the post-abuse adjustment of sexually abused girls. This section
provides an overview of the current theoretical discussions on resilience, and then
discusses parenting as a resilience-enhancing factor in abused children’s
development.
3.1 Resilience Framework
3.1.1 Theoretical Discussions of Resilience
Resilience refers to “a dynamic process encompassing positive adaptation
within the context of significant adversity” (Luthar, Cicchetti, & Becker, 2000,
p.543). The definition suggests two fundamental conditions constituting the notion of
resilience: (1) experience of threats to normal development of competence; and (2)
the achievement of positive adaptation in spite of high risks.
By definition, resilience presupposes past or present exposure to risk or
adversity. The term risk itself stemmed from epidemiological research seeking the
causes of disease (Costello & Angold, 2000). In the context of developmental
psychopathology, risk represents probabilities demonstrating good scientific
evidence of being associated with undesirable developmental outcomes or digression
from normal developmental trajectories. While early studies have paid attention to a
33
single risk factor, increasing evidence shows that risk factors often co-occur with
other risk factors, and that cumulative risks are more predictive of children’s
maladjustments than a single risk event (e.g., Sameroff, Seifer, Barocas, Zax, &
Greenspan, 1987). In terms of the practicality of these two approaches, Masten
(Masten, 1999; Masten and Powell, 2003) suggests that the comprehensive
cumulative risk approach may result in a better overall prediction of outcomes, but
that the aggregation can obscure dynamic interrelations among risk and resources in
one’s life. A single risk approach has the vantage point of delineating the specific
processes of risks, resources, and adaptations.
Resilience also requires the achievement of positive adaptation in the context
of significant risk. In resilience research, the term, “positive adaptation” has been
operationalized in a variety of ways. For example, Rutter (1987, 1990) defines it as
the positive end of the distribution of developmental outcomes, while some
researchers conceptualize positive adaptation as the absence of disorder or mental
health problems. In the case of Masten and her colleagues, positive adaptation is
characterized as showing competence in developmental tasks, as well as
demonstrating an absence of symptoms (Masten, 1994; Masten, Best, & Garmezy,
1990). The current controversies over the conception of positive adaptation are in
some part attributable to two different approaches in understanding the resilience
process (i.e., person-oriented approach vs. variable-oriented approach). In the
absence of a universal and consistent definition of resilience, it is critical for
34
investigators to clearly state the operational definition of resilience that their studies
utilized (Luthar, Cicchetti, & Becker, 2000).
The diverse conceptions of positive adaptation also stem from
multidimensional characteristics of the resilience construct. It is unreasonable to
assume that a group of children showing positive adaptation in one specific domain
will demonstrate the same pattern in other domains. This implies that an individual’s
resilient state should be determined by considering multiple domains of outcomes. In
variable-based analysis, it is very important to stipulate the domain of outcomes
investigated, and to avoid overly generalizing the findings across domains (Cicchetti
& Garmezy, 1993; Luthar, 1993).
Moreover, positive adaptation is not a static concept. Increasing evidence
shows that individuals vary in how they sustain their level of adaptation over time.
Some at-risk children show fairly stable trajectories, while others excel at a particular
point in time and then subsequently deteriorate (Tarter & Vanyukov, 1999). This
indicates that both short- and long-term longitudinal investigation of positive
adaptation is important in understanding resilience.
A number of theorists on the topic of resilience emphasize that resilience is a
dynamic process (Egeland, Carlson & Sroufe, 1993; Luthar, Cicchetti, & Becker,
2000; Masten, 2001). This stance is obviously in contrast with earlier concepts of
resilience, which tended to understand positive adaptation under adverse life
conditions as personality-like traits inherent in individuals (e.g., Anthony, 1974).
Rather, Masten states that resilience is a “description of a general pattern” or
35
phenomenon, “whereas diagnosis [of being resilient] occurs when the individual is
matched to the pattern” (Masten, 2003, p.4).
Accordingly, areas of interest in the contemporary resilience research extend
beyond individual attributes and encompass various levels of environment
surrounding individuals. The roots of these areas of interest can be traced back to
ecology models (Bronfenbrenner, 1977; Cicchetti & Lynch, 1993; Sameroff &
Chandler, 1975), which highlight the embeddedness of children within a larger,
multi-layered set of systems. In this line of reasoning, an individual’s resilience is
determined independently and interactively by this nested set of social systems.
As a protective process for individuals at high risk, Masten (2001)
emphasizes the importance of ordinary human adaptive systems. She hypothesizes
that resilience arises from the operation of common human adaptational systems (e.g.,
a child’s cognitive abilities, parenting quality, good schools, neighborhood quality,
and policy), rather than from rare or extraordinary processes (Masten and Powell,
2003). Supporting her hypothesis, evidence has continued to accumulate for the
association of such resources with positive outcomes among children at high risk
(e.g., Luthar et al., 2000; Masten, 2001; Masten & Coatsworth, 1998).
All in all, resilience research aims to understand the process of significant
risk leading to positive adaptation in spite of the experience of such risk. Specifically,
it searches for protective processes restoring and facilitating human adaptation in the
context of risks, and strives to find similarities in, or differences between, the
protective processes vis a vis given risk and population characteristics. Such efforts
36
to promote competence and resilience in children at risk have potential to guide
policy and practice.
3.1.2 Ordinary Human Adaptive Systems: Parenting as a Protective Process
Adaptation or maladaptation depends upon environmental support.
Developmental scientists have hypothesized that a key process by which individual
children and adolescents successfully adapt to their environment and achieve
developmental milestones is through quality parenting. In a normal developmental
process, primary caretakers provide the means for the fulfillment of basic needs for a
child to thrive. Through myriads of reinforcement and caretaker-child interactions, a
child imitates parental behavior (Bandura, 1977), internalizes parental values
(Lepper, 1982), experiences early prototypic interpersonal relationships (Bowlby,
1969), and acquires basic skills to master his or her environment.
Consistent with this view, resilience investigators have found effective
parenting in high risk contexts to be associated with competence in children. In other
words, parenting provides protective influence. The “distal” environmental risks (e.g.,
dangerous neighborhoods or poverty), or the effects of a child’s earlier adverse
experiences, may be mitigated by “proximal” protective processes, such as a quality
home environment characterized by warm parents who carefully monitor their
children (Miliotis, Sesma, & Masten, 1999).
To illustrate, ethnographic studies of ethnic minority youths living in
low-income, inner-city communities have indicated that effective parenting in these
37
communities has powerful protective and mitigating effects (e.g., Jarret, 1997; 1999).
Empirical studies have suggested that maternal control and firm discipline predicted
better school achievement among these youths (Gonzales, Cauce, Friedman, &
Mason, 1996). Maternal warmth served to buffer the negative effects of a delinquent
peer group, and mitigated the combined effects of father absence and a delinquent
peer group (Mason, Cause, Gonzales, & Hiagara, 1994). Similarly, close parent-child
relationships and high parental involvement in the child’s education were associated
with school success among homeless families (Miliotis, Sesma, & Masten, 1999).
In summary, based on the previous studies of resilience, it is plausible to
assume that a sexually abused child who is in a more favorable home environment
would do better than a sexually abused child who is in a less favorable proximal
setting, and would do equally well as non-abused counterparts.
3.2 Child Sexual Abuse and Resilience in Behavioral Adjustment
Behavioral competence refers to positive adaptation in the conduct domain.
Investigating behavioral outcome among sexually abused females is especially
important for a number of reasons. Firstly, a salient developmental task for
school-aged children and adolescents is the ability to follow rules of conduct in the
home, school, and community. Following rules of conduct is a key criterion of
age-appropriate competence, along with academic and social competence. Moreover,
the descriptive literature on sexually abused children and adolescents suggests that
38
there exist considerable differences between abused children and adolescents and
their non-abused counterparts in behavioral characteristics.
3.2.1 Child Sexual Abuse as a Threat to Development of Behavioral
Competence
Internalizing behavioral symptoms has been one of the most frequently
examined areas in child sexual abuse research. Current child sexual abuse studies
have accumulated fairly consistent findings regarding the association between child
sexual abuse and internalizing symptoms. These studies have indicated a heightened
level of depressive symptoms (Bailey & McCloskey, 2005; Black, Dubowitz, &
Harrington, 1994; Feiring, Taska, & Lewis, 1999), trauma symptoms (Feiring, Taska,
& Lewis, 1999), and overall internalizing behaviors (Einbender & Friedrich, 1989;
Mannarino, Cohen, Smith, & Moore-Motily, 1991; Tremblay, Hebert, & Piche,
1999; Walrath, Ybarra, Holden, Liao, Santiago, & Leaf, 2003) among sexually
abused children. The levels of internalizing behavioral symptoms among this
population were found to be higher than the norm and the community comparison
group, and comparable to the psychiatric outpatient clinical control group (Friedrich,
Beilke, & Urquiza, 1988; Mannarino, Cohen, & Gregor, 1989). These associations
were supported in a six- and twelve-month follow-up study (Mannarino, Cohen,
Smith, & Moore-Motily, 1991).
Compared with internalizing symptoms, externalizing behavioral symptoms
have been relatively less researched among sexually abused children and adolescents.
39
Some child sexual abuse studies have focused more on “sex-typed” forms of
delinquency and acting out than on other forms of delinquency and physical
aggression (Einbender & Friedrich, 1989; Elze, Auslander, McMillen, Edmond, &
Thompson, 2001; Feiring, Taska, & Lewis, 1999; Runtz & Briere, 1986; Trickett &
Gordis, 2004). In addition, controversies exist over the relationship between child
sexual abuse and externalizing behavioral symptoms.
Using a prospective design, Widom and Ames (1994) compared the criminal
records of a group of adults abused and/or neglected as children, and a comparison
group not known to have been abused or neglected. They found that early childhood
sexual abuse did not increase the likelihood of later delinquent and criminal
behaviors, but was significantly related to the rate of arrest for becoming prostitutes.
Zingraff, Leiter, Myers, and Johnsen (1993) report similar findings. They found that
sexually abused children and adolescents did not differ from community comparison
groups with respect to property and violent offenses. Significant group difference
was found only in status offenses. Compared with impoverished children, sexually
abused children did not differ with respect to any type of offenses examined in the
study (i.e., property offense, violent offense, status offense).
In contrast to Widom and Ames (1994), Swanston and her colleagues suggest
that child sexual abuse may be a significant risk factor for aggression and delinquent
behavior (Swanston, Parkinson, O’Toole, Plukett, Shrimpton, & Oates, 2003). One
of the important caveats in Widom and Ames’s study (1994) is that information on
offending and aggression was obtained from official criminal records. These may be
40
vulnerable to system bias and court processing factors, and may exclude less serious
offenses, which can be an important limitation for capturing adolescents’ delinquent
and aggressive behaviors. Instead, Swanston and her colleagues (2003) examined the
relationship using three different types of information, including self-reports,
mother’s reports, and official records. During the nine years after the child sexual
abuse was substantiated, sexually abused children showed more self-reported
criminal behavior and mother-reported aggressiveness than the comparison group.
Similar to Widom and Ames (1994), group differences were not detected in juvenile
criminal records. In alignment with this study, an increasing number of studies have
reported an elevated level of externalizing behavioral symptoms among this
population. These studies include clinical samples (Black, Dubowitz, & Harrington,
1992; Friedrich, Beilke, & Urquiza, 1988; Mannarino, Cohen, & Gregor, 1989;
Walrath, Ybarra, Holden, Liao, Santiago, & Leaf, 2003) and college samples (Runtz
& Briere, 1986), as well as 12-month follow-up studies (Mannarino, Cohen, Smith,
& Moore-Motily, 1991).
Some studies have explicitly focused on differences in the reactions of boys
and girls to sexual abuse. Generally, they have failed to find any consistent gender
differences in the behavioral symptom profiles of sexually abused boys and girls.
Using a community sample of sexually abused adolescents, and a matched
comparison group, Garnefski and Diekstra (1997) failed to find a significant gender
by sexual abuse interaction on aggressive/criminal behaviors. Adolescents of both
genders with a history of childhood sexual abuse had higher internalizing (i.e.,
41
emotional problems) and externalizing (i.e., aggressive/criminal behavior,
addiction-risk behavior) behavioral symptoms than the comparison group
adolescents. In this particular study, the sexually abused girls exhibited
aggression/criminal and suicidal behaviors 13 times more often than their
non-abused female counterparts. Swanston and her colleagues (2003) also failed to
find gender differences in any of the measures of aggressive, delinquent or offending
behavior. The results of review studies (Jumper, 1995; Kendall-Tackett et al., 1993)
suggest that boys and girls who were sexually abused do not significantly differ in
subsequent psychological and behavioral adjustment.
The similarities between sexually abused girls and boys contrast with the
behavioral patterns observed among the general population, in which boys are more
likely to exhibit externalizing symptoms, while girls are more likely to exhibit
internalizing symptoms. Current findings suggest that gender may not be a
significant predictor of offending in the sexually abused group. Childhood sexual
abuse, however, may be a particularly significant factor for externalizing behaviors
among girls.
3.2.2 The Role of Non-Offending Mothers in the Post-Abuse Adjustment of
Sexually Abused Girls
In the area of childhood sexual abuse research, non-offending mothers’
reactions and support following their children’s victimization have been found to be
particularly influential in explaining the variability of the child victims’ adaptations.
42
Abused children with supportive caretakers show fewer symptoms of distress (Conte
& Schuerman, 1987; Morrison & Clavenna-Valleroy, 1998), and fewer
psychological symptoms (Mannarino & Cohen, 1996). Sexually abused children who
have better relationships with their mothers exhibit fewer internalizing and
externalizing symptoms (Hazzard, Celano, Lawry, & Webb, 1995), while sexually
abused children with mothers who have a rejecting parenting style tend to have more
emotional (Deblinger, Steer, & Lippmann, 1999) and behavioral (Lovett, 1995)
problems. This association is also found in the retrospective studies of adult
survivors of childhood sexual abuse. The level of support survivors received from
parents predicted later effects of childhood sexual abuse on outcomes such as
distress (Roesler, 1994) and psychological adjustment (Everson et al., 1989), and
moderated survivors’ attitudes toward men in adulthood (Wyatt & Mickey, 1987). In
fact, emerging evidence suggests that parental support may be a more potent
determinant of abused children’s later adaptations than sexual abuse characteristics
per se (Johnson & Kenkel, 1991; Tremblay, Herbert, & Piche, 1999).
The current study attempts to go beyond such general findings and examines
the way specific maternal factors exert influence on abused daughters’ post-abuse
adjustment. While the existing studies have accumulated fairly clear evidence
indicating that parental support predicts better prognosis among sexually abused
children, findings vary in terms of sources of support (e.g., mother, father, or friend)
and types of support (e.g., abuse-specific support vs. global, non-abuse specific
support). For example, contrary to most other studies, Lynskey and Fergusson’s
43
(1997) longitudinal birth-cohort study shows that paternal care was a significant
predictor of their abused children’s positive adjustment outcomes, but maternal care
was not. It is not yet clear whether abuse-specific or overall care is a better predictor
of abused children’s adjustment (Elliott & Carnes, 2001).
In addition, the existing studies suggest a direct relationship between
children’s self-reported distress and parents’ self-reported distress among this
population (Hanson, Saunders & Lyppovsky, 1992). In a similar vein, social support
literature has found that children’s social networks are closely related to their
mothers’ social networks. Some of these parent-network members are active in the
children’s networks. The social resources available to parents have direct influences
on various behavioral outcomes in their children (Cochran & Walker, 2005). These
findings imply that the relationship between mothers’ parenting practices and
children’s behavioral adjustment needs to be examined in a broader context of
mothers’ social and psychological resources. The current study aims to examine the
relationship between mothers’ parenting and their children’s behavioral adjustment
simultaneously with risks and resources provided by mothers. However, such
rigorous studies have been lacking in the current literature.
3.2.3 Methodological Issues in Past Studies
The methodological limitations of past research should also be noted. The
first issue is method variance in the relationship between maternal factors and
children’s adjustment. Many studies utilized mothers’ ratings of their children’s
44
behavioral symptoms (Hazzard, Celano, Lawry, & Webb, 1995), or children’s
reports of their parents’ parenting styles (Deblinger, Steer, & Lippmann, 1999). Such
research design is especially vulnerable to cognitive distortions of raters. It is also
unclear how much the variance of children’s adjustment is explained by maternal
factors (true relationship) or an operation of the same method (informant). Moreover,
many of these studies used a cross-sectional design, and thus provide limited
information on the causal direction of the relationship between maternal factors and
children’s adjustment. Issues such as common method variance may be more
problematic in cross-sectional design.
In addition, many of the existing studies focus on within-group analysis. Few
studies include meaningful comparison groups. This posits particular challenges for
testing whether similar or different causal mechanisms operate in the process of
behavioral adjustment among sexually abused girls, and among their non-abused
comparison counterparts.
Finally, many studies contain mixed samples of intra- and extra-familial
sexual abuse (Newberger, Gremy, Waternaux, & Newberger, 1993), in spite of the
empirical evidence suggesting that non-offending mothers’ reactions to their
children’s sexual abuse may differ according to their own relationships with the
perpetrators (Pintello & Zuravin, 2001). Some studies did not provide adequate
information on the sexual abuse characteristics of the samples (e.g., Shah, Dali, &
Heinrichs, 1995), while other studies utilized clinical samples of sexually abused
45
children, in which the reports of abuse solely depended on the participants’ reports
(e.g., Timmons-Mitchell, Chandler-Holtz, & Semple, 1997).
The present study reports on findings from a multigenerational, longitudinal
study, in which both non-offending mothers and their sexually abused daughters
participated. The participants were assessed, starting within 6 months of the
disclosure of the daughters’ sexual abuse, and continuing throughout their childhood
and adolescence. A demographically-similar comparison group of mothers and
daughters was also recruited from the same communities where the abused girls’
families resided. The daughters’ childhood sexual abuse was substantiated and
stringently defined. Their behavioral outcomes were rated by the children themselves,
which enabled this study to overcome some limitations of the existing studies with
respect to common method variance.
3.3 Integration: The Generic Conceptual Model
To explore the relationship of non-offending mothers’ parenting as well as
other maternal factors to their daughters’ later behavioral adjustment outcomes, an
integrative model of the suggested relations among the variables was specified. The
specification of the model emanated from the existing empirical studies, as well as
conceptual hypotheses posited by Belsky (1984). A graphic representation of this
model is displayed in Figure 2.
46
This model makes explicit a number of hypotheses, including:
Hypothesis 4 (Path j & k):
Daughters’ behavioral adjustment outcomes will be predicted by their own
childhood sexual abuse status and mothers’ parenting practices
Hypothesis 5 (Path l & m):
Daughters’ behavioral adjustment outcomes will be predicted by the amount
of social support their mothers receive from their families-of-origin and the
mothers’ dissociative symptoms.
Hypothesis 6 (Path j
(A)&(C)
, l
(A)&(C)
):
The relationshipss of mothers’ parenting styles, and the levels of support
mothers receive from family-of-origin sub-networks, to the children’s
behavioral outcomes, will differ by the children’s sexual abuse status. More
specifically, based on the resilience perspective suggesting the protective
function of environmental resources in abused children’s lives, this study
predicts greater effects of maternal parenting and social support from the
maternal family-of-origin network among abused girls than among their
non-abused comparison group.
47
[Abuse Group]
[Comparison Group]
Figure 2. Integrative model of structural relationships among maternal factors and the
abused and comparison girls’ behavioral adjustment. Note. A (Abuse) and C (Comparison)
in parentheses indicate a parameter varies over groups.
Mothers’
Childhood
Sexual Abuse
History
Mothers’
Parenting
Practices
Daughters’
Behavioral
Adjustment
l
(A)
j
(A)
m
k
e
Mothers’
Dissociative
Symptoms
Social Support
Mothers Receive from
Family-of-Origin
Sub-Network
Daughters’
Childhood
Sexual Abuse
Mothers’
Own Mothers’
Discipline
Punitiveness
Mothers’
Childhood
Sexual Abuse
History
Mothers’
Parenting
Practices
Daughters’
Behavioral
Adjustment
l
(C)
j
(C)
m
k
e
Mothers’
Dissociative
Symptoms
Social Support
Mothers Receive from
Family-of-Origin
Sub-Network
Daughters’
Childhood
Sexual Abuse
Mothers’
Own Mothers’
Discipline
Punitiveness
48
Chapter IV. Method
4.1 Methods for Data Preparation
The current study utilized data from a prospective, longitudinal study of the
long-term impact of childhood sexual abuse on female development (Trickett &
Putnam, 1987). The study is the longest-running child sexual abuse study in the U.S.
It began in 1987, with support from the NIMH, and the sample was reassessed
approximately every 2 years after the initial assessment. The latest data collection
wave (Time 6) has just finished, with an excellent retention rate of 96%.
4.1.1 Participants
Participants of the current study include the mothers (Generation 1; G
1
) and
girls (Generation 2; G
2
) enrolled in the original study. The abused girls were referred
to the study by protective agencies in the greater Washington, D.C. metropolitan area.
Eligibility criteria for inclusion in the study were that: 1) the victim was female, aged
6 to 16 years; 2) disclosure of the substantiated child sexual abuse occurred within 6
months of participation in the study; 3) sexual abuse involved genital contact and/or
penetration; 4) the perpetrator was a family member such as parent, stepparent,
sibling, uncle, or the mother’s live-in boyfriend; and 5) a non-abusing parent or
guardian (usually the mother) was also willing to participate in the study.
Participants for the comparison group were recruited through ads posted in
community newspapers, welfare clinics, daycare agencies, and community facilities
located in the same neighborhoods in which the abused girls lived. The two groups
49
of girls were similar in terms of age, ethnicity, socioeconomic status (SES), and
family constellation (one- or two-parent families) (Noll, Trickett, & Putnam, 2003).
Possible abuse experiences among comparison girls were ascertained through
maternal interviews conducted at initial assessment, and further validated by
interviewing girls at later time points. It was learned that a few comparison girls
experienced child sexual abuse before the age of 18. For the comparison girls who
had experienced childhood sexual abuse, the nature of the abuse was examined. If
their experiences were similar to the sexual abuse criteria of the study, they were
included in the abuse group for the current analyses.
The current study focuses on G
1
mothers’ data from the initial assessment
(Time 1) and G
2
daughters’ data from the fourth assessment (Time 4). Time 4
corresponds to approximately seven years after the disclosure of childhood sexual
abuse. The mean age of G
2
girls was 18.6 years at Time 4 assessment. Girls began to
self-report their behavior problems from Time 4.
4.1.2 Individual Subjects and Mother-Daughter Pairs
The Time 1 sample consisted of 84 abused and 82 comparison participants,
comprising a total sample of 166 participants, and their non-offending caretakers or
primary caregivers. Because of 26 sibling cases, the total number of G
1
mothers who
participated in Time 1 was less than the number of G
2
daughters, totaling 134
mothers.
50
Next, “caretaker” or “primary caregiver” status was defined as having lived
with the G
2
child for most of the child’s life, including the two years prior to the
study, and all biological mothers were assumed to be the primary caregivers in this
sample. Of the 134 G
1
mothers who participated in Time 1, 128 mothers met these
criteria. 96% of the mothers (n=123) were biological mothers. Others included 4
adoptive mothers (3%) and 1 grandmother (1%). Because of possible confounding of
age and cohort effects, the grandmother’s data was removed from the final analysis
(total N
G1
= 127).
G
2
data for the current analyses were provided by a subset of the original
sample. Based on the 127 G
1
mothers’ data, G
2
information was linked. For sibling
cases, only one child was randomly selected. No significant group differences were
detected in any of the major demographic information or the study variables between
the G
2
original sample and this subset of girls.
The mean age of G
1
mothers was 35.4 years old (sd=5.5), with a range of 24
to 49 years old at study entry. All mothers ranged from low to middle SES, with a
mean Hollingshead (1975) score of 36.2 (sd=13.3), which is defined as blue-collared
working class. Almost half of the mothers (52.8%) were Caucasian, and the rest were
members of ethnic minorities (55 were African American, 4 Hispanic, and 1 Asian
American). The mean age of G
2
daughters was 18.6 years old (sd=3.3), with a range
of 10.6 to 25.9 years old at Time 4 assessment.
51
4.1.3 Measures of G
1
Mothers’ Childhood Developmental History
Information about G
1
mothers’ childhood developmental history was
obtained with the Mothers’ Developmental History Questionnaire (MDHQ) (Trickett
& Everett, 1988), which is a structured interview protocol developed for this study.
The MDHQ consists of four parts, including the G
2
child’s family background and
developmental history, and the G
1
mother’s family background and developmental
history. Most of questions in the MDHQ had response categories, but some were
open-ended questions (e.g., “Can you tell me a little about your child’s friends?
What is her best friend like?”). G
1
mothers’ perceptions of their own parents
(referred to as Generation 0; G
0
) were also obtained through G
1
mothers’
self-reports.
G
1
mothers’ childhood sexual abuse. At Time 1 assessment, trained
interviewers asked about the G
1
mothers’ developmental history, and then probed for
sensitive details about the G
1
mothers’ own childhood sexual abuse history. These
details include the age of onset, the type of abuse, and the perpetrator. The authors
conducted individual case reviews in order to determine the presence of the mothers’
sexual abuse. Subsequent changes were made in two mothers’ childhood sexual
abuse status. In one case, the mother was over 18 at the first time of experiencing
sexual abuse; and in the other case, the mother reported “just looking” as her
childhood sexual abuse experience. Sexual abuse history was given a binary code (0
= no abuse; 1 = abuse).
52
Using this measure, the previous study (Kim et al., in press) found that
thirty-nine mothers (32.5%) reported having histories of childhood sexual abuse,
while 81 mothers (67.5%) did not report such histories in this sample. The same
study also showed that a disproportionately larger number of G
1
mothers in the G
2
abuse sample reported childhood sexual abuse than G
1
mothers in the G
2
comparison
sample (45% vs. 16%;
2
= 11.4, p < .001). Among the 39 mothers reporting
childhood sexual abuse histories, the mean age at the onset of abuse was 8.5 years
old (sd = 3.9). Forty-one percent of the mothers reported that they had been abused
by their fathers or father figures (i.e., stepfathers, adoptive fathers, or foster fathers),
followed by other relatives (28%; cousins, uncles, or grandfathers), siblings (15%),
and non-family members (15%) during childhood.
G
0
mothers’ discipline punitiveness. At Time 1 assessment, the G
1
mothers
rated, on a 5-point scale (“1”= never, “5”= often), the frequency of specific
discipline strategies that their own G
0
mothers utilized in a series of nine questions
(e.g., explanations, physical punishment, time out, guilt induction). To obtain a basis
for measurement, exploratory common factors analyses of these nine items were
conducted. Parallel analysis (Horn, 1965) results indicated that two factors should be
extracted. Promax rotation to an Equamax target was used to maximize the simple
structure. For the current analysis, the factor that could be interpreted as frequency of
using punitive discipline strategies (e.g., physical or verbal punishment, and guilt
induction) was utilized ( = .72).
53
4.1.4 Measures of G
1
Mothers’ Parenting Practices
G
1
mother’s parenting. The Child Rearing Practices Q Sort (CRPQ) (Block,
1981) was utilized to obtain the G
1
mothers’ child-rearing practices. This measure
consists of 91 items that the mothers sort into seven piles, depending on how
descriptive each statement is of the mothers’ parenting attitudes. There is
considerable evidence for the test-retest reliability and validity of the CRPQ (e.g.,
Block, 1981). From this measure, two subscales of parenting were formed:
enjoyment of the child and parental role (13 items, = .82), and authoritarian control
(11 items, = .66). Specific procedures used to make the subscales are presented
elsewhere (Trickett & Susman, 1988). The subscales have a possible range of 1 to 91
for enjoyment and 1 to 77 for authoritarian control.
4.1.5 Measures of G
1
Mothers’ Current Functioning
G
1
mothers’ dissociatve symptoms. At Time 1 assessment, information on the
mothers’ current level of dissociative symptoms was obtained from The Dissociative
Experience Scale (DES; Bernstein & Putnam, 1986). The DES is a 28-item
self-report measure, which assesses the frequency of dissociative experiences, such
as amnesia, depersonalization, derealization, identity disturbance, and absorption and
enthrallment. A test-retest coefficient of .84 has been reported for the DES, and the
scale’s validity has been supported by its ability to distinguish subjects diagnosed
with a dissociative disorder from other subjects (Bernstein & Putnam, 1986). Alpha
for this sample was .77.
54
Social support G
1
mothers receive from family-of-origin sub-network. At
Time 1 assessment, information on mothers’ social support and network
characteristics was obtained using the Mother’s Social Support Questionnaire
(MSSQ) (Corse, 1983) (Appendix B). The MSSQ is a network-based inventory.
Mothers first identify network members who they perceive as providing social
support and indicate: (1) their relationship with the person; (2) how long they have
known the person; (3) the frequency of contact; and (4) the gender of the person.
Using a binary scale (“1”=yes; “0”=no), mothers then rate the availability of support
for each suggested support item for each supportive individual. The questionnaire
consists of 23 items. Based on 11 receiving-support items and 1 mutual-interaction
item, the previous study (Kim, Chung, & Trickett, in preparation) conducted
confirmatory factor analysis, and found that the two-factor model of social support
(emotional/information support vs. instrumental support) was the most accurate and
parsimonious representation of the social support structure in this sample. The
structure was invariant across support sources (family-of-origin sub-network vs.
non-familial sub-network) as well as sexual abuse status. For the current analysis,
simple linear composite score of instrumental support mothers receive from their
family-of-origin sub-network was utilized ( = .75).
4.1.6 Measures of G
2
Daughters’ Behavioral Adjustment
G
2
behavioral adjustment. Five behavior problem measures were obtained
based on Youth Self Report (YSR; Achenbach, 1991). The YSR consists of 120
55
items, which parallel those of the Child Behavior Checklist (CBCL; Achenbach,
1991, 1992), but the child herself checks one of three possible responses to indicate
the extent of a particular problem they may have (e.g., fears going to school). The
composite measures proposed in the manuals for these items are overlapping and
have inadequately low alpha reliabilities (Trickett, Noll, Reiffman, & Putnam, 2001).
To obtain a better basis of measurement, the previous study (Noll, Trickett, Horn,
Long, & Putnam, in preparation) factored polychoric intercorrelations among the
question items within a sample of 944 maltreated children. According to the study, in
a random one half of the sample, the scree test suggested five factors, which were
metrically invariant in the second random one half of the sample, and across gender
and occasion of assessment (Time 1 and Time 2 over a two-year follow-up of 496
subjects) (Trickett et al., 2001). The factors were interpreted as indicating: (a)
delinquent misbehavior (13 items), (b) immature or bizarre behavior (38 items), (c)
aggressive or bullying behavior (21 items), (d) expressions of withdrawal and
depression (33 items), and (e) expression of physical problems (13 items). Simple
linear composites of the items indicating each factor were obtained, and the
composites indicating the first four factors were utilized for the current analyses. The
intercorrelations among these measures and their alpha reliabilities (in italics along
the main diagonal) are presented in Table 1 (p.64). Intercorrelations among the YSR
measures ranged from .31 to .66.
56
4.1.7 Missing Data
Data for the scales were imputed using multiple imputation procedures
embodied in Schafer’s (1999) NORM program. This simulation-based analysis of
incomplete data replaces the missing information with a “best guess” via the
Estimation-Maximization (EM) algorithm (Dempster, Laird, & Rubin, 1977) and
iterative data augmentation procedures. First-order imputation was conducted for all
of the scales where mothers did not respond to a particular item. There were few
missing item responses. No items of any scales exceeded 4% in missing item
responses. Second-order imputation was also conducted for four composites of the
G
2
YSR, using longitudinal information (i.e., Time 4, 5 and 6 data).
4.1.8 Model Fitting Indices
The degree of discrepancy between the observations and the expectations in
each model was assessed using multiple goodness-of-fit indices. This study used a
chi-square distribution (
2
) indicating the overall absolute fit of the expectations
compared to the observations. Differences in
2
values were referred to assess the
relative adequacy of the models. Further, given the objectivity of identifying the
model that minimized the discrepancies, the Root Means Square Error of
Approximation (RMSEA) was an important index for the model decision (Browne &
Cudeck, 1993). RMSEA represents the misfit ratio per subject and per degree of
freedom. A value of less than 0.05 was considered to be an excellent fit, and a value
of less than 0.08 reasonable (Browne & Cudeck, 1993).
57
4.2 Methods for Model Organization
The suggested study hypotheses were formally tested by comparing the
integrative models (Figure 1 & 2) with theoretically-based alternative ones that
specify different patterns of relations among the variables. Following this logic, a
series of models positing specific hypotheses were examined. The first point was a
model for predicting mothers’ parenting practices. Next, G
2
daughters’ behavioral
outcome measures were introduced to the G
1
model. A series of models specified a
priori were compared and the hypothesized relations among maternal factors and G
2
behavioral outcomes were tested.
4.2.1 Structural Models of Relationships: G
1
Mothers’ Parenting
A prerequisite to evaluating the structural component was to determine
whether the two dimensions of G
1
mothers’ parenting practices (i.e., enjoyment of
child and parental role, authoritarian control) should be considered simultaneously,
or separately by testing a single-factor model of parenting practice. These were
examined by comparing between baseline and single-factor models:
Model 1: Baseline model for G
1
parenting
Included both dimensions of mothers’ parenting practices (i.e., enjoyment of
a child and parental role, authoritarian control) correlated with each other.
58
Model 2: One-factor model for G
1
parenting
Hypothesized that correlations between the two observed G
1
parenting
practice variables were explained by regression from an unobserved variable
representing “G
1
parenting practice” to the two subscales.
Next, the hypotheses specifying structural relationships among the variables
were tested in a series of model comparison tests. These include:
Model 3: G
1
Baseline model
Hypothesized no relations among the variables, and represents a baseline
against which the G
1
parenting models can be compared. From G
1
Integrative
model (Model 4), all the hypothesized paths were removed.
Model 4: G
1
Integrative model
Described in Section 2.4 (“Integrative Model of Parenting Practices for
Non-Offending Mothers of Sexually Abused Girls”). Represented the full
model, including both direct and indirect effects among the variables.
Model 5: Direct effect model
From Model 4, paths between G
1
mothers’ child sexual abuse and social
support, and between G
1
mothers’ child sexual abuse and dissociative
symptoms, were removed. This model postulates that G
1
mothers’ childhood
sexual abuse history, current social support from their family-of-origin, and
dissociative symptoms, predict their parenting independently.
59
Model 6: Spurious effect model
From Model 5, the path between G
1
mothers’ childhood sexual abuse and G
1
mothers’ parenting practice was removed. This postulates that the relation
between G
1
mothers’ childhood sexual abuse experience and later parenting
practices is spurious, after the G
0
mothers’ parenting variable is taken into account.
Model 7: Indirect effect model
To Model 6, four paths between G
1
mothers’ childhood variables and social
support, as well as between G
1
mothers’ childhood variables and dissociative
symptoms, were added. These paths allowed indirect effects of mothers’
childhood history variables on their current parenting practices.
Finally, the current study employed multi-group analysis and examined
whether the hypothesized pattern of relationships among the variables differs by the
child’s sexual abuse status. First, the sample was split into two groups, (sexually
abused vs. comparison) and the G
1
integrative model was simultaneously fit into two
groups. Next, all the effects on the mothers’ parenting practices were held equal
across samples, and then let free to vary one by one. One-degree-of-freedom, nested
chi-square tests of cross-group equality constraints (Raykov, 1997) were performed.
4.2.2 Structural Models of Relationships: G
2
Daughters’ Behavioral Outcomes
Based on the most accurate and parsimonious representation of the G
1
parenting data identified from the series of model comparison tests described above,
60
the relations between maternal factors and four different measures of G
2
behavioral
outcomes (i.e., withdrawal/depressed behaviors, immature/bizarre behaviors,
delinquent misbehaviors, aggressive behaviors) were examined. A prerequisite to
evaluating the structural component was to determine whether the four dimensions
should be considered simultaneously or separately, by testing one- and two-factor
models of behavioral outcomes. These were examined by comparing baseline,
one-factor, and two-factor models. These models include:
Model 8: Baseline model for G
2
behavioral adjustment
Included G
2
daughters’ four measures of behavioral symptoms correlated
with one another.
Model 9: One-factor model for G
2
behavioral adjustment
Hypothesized that correlations among the four observed behavioral
symptom variables were explained by regression from an unobserved
variable representing “G
2
global behavioral adjustment” to the four
subscales.
Model 10: Two-factor model for G
2
behavioral adjustment
Hypothesized that correlations among the observed behavioral outcome
variables were accounted for by regressions from two unobserved variables
representing “internalizing symptoms” and “externalizing symptoms”.
61
Next, similar to the procedures utilized for testing hypotheses of G
1
mothers’ parenting
practices, the relations between maternal factors and the daughters’ behavioral symptoms were
examined by comparing a series of models. These models include:
Model 11: G
2
Baseline model
Hypothesized no relations among maternal factors and G
2
behavioral
outcome variables. From G
2
Integrative model (Model 12), all the
hypothesized paths were removed.
Model 12: G
2
Integrative model
Described in Section 3.3 (“Integration: Generic Conceptual Model”). This
model postulates that mothers’ social support and dissociative symptoms,
as well as their parenting practices, affect daughters’ behavioral adjustment
outcomes.
Model 13: Mediation model
Removed two paths from Model 12: the direct path from mothers’ social
support from their family-of-origin sub-network to G
2
behavioral outcomes,
as well as the path from mothers’ dissociative symptoms to G
2
behavioral
outcomes. Represented a mediation model of the effects of G
2
child sexual
abuse on their behavioral adjustment via mothers’ parenting practices.
Finally, using multi-group analysis, the hypotheses specifying that different
causal mechanisms may operate in two groups of girls were tested (Model 14 – Model
62
17). As with the G
1
parenting analyses, all the effects on the daughters’ behavioral
adjustment were held equal across samples, and then were free to vary one by one.
63
Chapter V. Results
5.1 Data Description
Table 1 presents the correlations, means, and standard deviations of all the
study variables. In each analysis, G
1
mothers’ ethnic minority status, SES ratings,
and G
2
daughters’ ages were controlled for. Visual inspection of this table shows that
G
1
mothers’ enjoyment of a child and parental role is negatively correlated with their
level of dissociative symptoms. Mothers’ authoritarian control was positively
correlated with their own G
0
mothers’ discipline punitiveness, but both dimensions
of parenting were not correlated with any measures from the YSR. Group differences
were detected in the correlations between mothers’ social support and the G
2
measures from the YSR as well as between two G
1
parenting practice variables and
G
2
behavior outcome measures.
5.2 Testing G
1
Mothers’ Parenting Models
5.2.1 G
1
Baseline Models
Two different representations of G
1
mothers’ parenting practices were
examined in order to determine the baseline model in this sample. Two parenting
measures from Child Rearing Practice Q-Sort (CRPQ) (Block, 1981) were included
in the same model at this step. The first model (Model 1) represents a baseline model
hypothesizing correlation between two parenting variables. The second model
64
Table 1. Partial Correlation Coefficients among the Study Variables
1 2 3 4 5 6 7 8 9 10 11 M SD
[1.1] Total sample (N = 127)
1. G
1
Child sexual abuse - .3 .5
2. G
2
Child sexual abuse .27 - .6 .5
3. G
0
Discipline punitiveness .12 .12 .72 15.4 4.3
4. G
1
Social support (FOO) -.14 -.07 -.17 .75 1.5 1.1
5. G
1
Dissociative symptoms .12 .03 .22 -.17 .77 10.8 8.5
6. G
1
Enjoyment -.09 -.19 -.18 .09 -.28 .82 74.9 8.8
7. G
1
Authoritarian control -.03 .04 -.22 .00 .08 -.45 .66 33.4 8.0
8. G
2
Delinquent misbehavior -.03 .17 .13 .07 -.16 -.14 .19 .63 5.4 2.7
9. G
2
Aggressive behavior .09 .21 .03 .00 -.09 -.13 .09 .51 .79 10.4 6.3
10. G
2
Withdrawal/depressed .04 .13 -.01 - 07 -.14 -.10 .03 .42 .63 .88 18.6 9.0
11. G
2
Immature/bizarre -.03 .15 .04 .07 -.09 -.14 .05 .31 .56 .66 .81 28.2 8.0
[1.2] G
2
Abuse sample (N = 72)
1. G
1
Child sexual abuse .5 .5
2. G
2
Child sexual abuse - 1.0 .0
3. G
0
Discipline punitiveness .15 - 15.8 4.5
4. G
1
Social support (FOO) -.20 - -.10 1.3 1.1
5. G
1
Dissociative symptoms .16 - .19 -.10 11.1 8.1
6. G
1
Enjoyment -.02 - -.28 .08 -.38 73.4 9.0
7. G
1
Authoritarian control -.12 - -.28 .02 .03 -.40 33.2 8.8
8. G
2
Delinquent misbehavior -.15 - .11 .14 -.19 -.09 .24 5.8 2.7
9. G
2
Aggressive behavior .03 - .03 .17 -.10 -.23 .21 .47 12.0 6.1
10. G
2
Withdrawal/depressed -.01 - -.12 .14 -.17 -.19 .23 .43 .65 20.5 9.0
11. G
2
Immature/bizarre -.14 - -.01 .27 -.12 -.25 .23 .31 .49 .65 29.2 8.0
[1.3] G
2
Comparison sample (N = 55)
1. G
1
Child sexual abuse .2 .4
2. G
2
Child sexual abuse - .0 .0
3. G
0
Discipline punitiveness -.05 - 14.7 4.1
4. G
1
Social support (FOO) .11 - -.20 1.6 1.1
5. G
1
Dissociative symptoms .07 - .26 -.24 15.9 9.1
6. G
1
Enjoyment -.03 - .08 .05 -.13 76.9 7.0
7. G
1
Authoritarian control .14 - -.15 -.03 .17 -.54 33.6 7.0
8. G
2
Delinquent misbehavior .06 - .14 .06 -.18 -.12 .08 4.8 2.6
9. G
2
Aggressive behavior .08 - -.02 -.19 -.15 .18 -.16 .50 9.4 6.3
10. G
2
Withdrawal/depressed .07 - .15 -.34 -.16 .14 -.36 .36 .57 17.3 9.0
11. G
2
Immature/bizarre .06 - .11 -.24 -.06 .21 -.33 .23 .63 .67 26.8 7.5
Note.G
1
Mothers’ ethnic minority status, SES ratings, and G
2
daughters’ age were controlled
for in each analysis. G
1
, Generation 1; G
2
Generation 2; G
0
Generation 0; CSA, Child sexual
abuse; FOO, Family of origin. Reliability alphas in Italics in main diagonal for total sample
65
(Model 2) hypothesizes the single factor model of parenting practice (Figure 3).
Model fit indices indicated poor fit for both models, and improvement of fit from
Model 1 to Model 2 was not significant (
2
= 4.25,df = 3) (Table 2). In
subsequent analyses testing structural relationships, two parenting measures were
examined separately.
[Model 1. G
1
Baseline Model] [Model 2. One Factor for Parenting]
Figure 3. Summary of model comparison tests for G
1
parenting baseline model
Note. Mothers’ ethnic minority status, SES ratings and G
2
daughters’ age were partialled
out in each analysis. G
1
, Generation 1; G
2
Generation; G
0
, Generation 0; CSA, child sexual
buse; DP, Discipline punitiveness; Social, Social support from family of origin; DES,
Dissociative symptoms, Enjoy (Enj), Enjoyment of child and parental role; Control (Con),
Authoritarian control; Par, Parenting.
Table 2. Model Fit Comparison for G
1
Baseline Models (N = 127)
Model
2
df p RMSEA
Δ from base
2
df
Parenting model
Model 1 (baseline for G
1
parenting) 63 26 .00 .11
Model 2 (one factor for G
1
parenting) 67 23 .00 .12 4.25 3
G
1
Enj
G
2
CSA
G
1
DES
G
0
DP
G
1
CSA
G
1
Con
G
1
Par
G
1
Social
G
1
Enjoy
G
1
Social
G
1
DES
G
0
DP
G
1
Control
G
1
CSA
G
2
CSA
66
5.2.2 Results of G
1
Mothers’ Parenting Models
A graphic representation of each model for G
1
parenting model comparison
tests is provided in Figure 4. For each set of model comparison tests, the enjoyment
of child and parental role dimension was examined first, and the authoritarian control
dimension was then tested following the same procedures.
Table 3 presents the results of the model comparison tests. The two
dimensions of parenting yielded a similar pattern. The fit indices for the baseline
model (Model 3) indicated poor fit (
2
= 30.80, df = 15 for enjoyment,
2
= 26.58, df
= 15 for authoritarian control), but the fit indices for the G
1
integrative model (Model
4) provided an excellent fit (
2
= 3.36, df = 6 for enjoyment;
2
= 3.37, df = 6 for
authoritarian control), suggesting that this may be a reasonable representation of the
structure of the G
1
parenting data.
The direct effect model (Model 5) also yielded an acceptable fit (
2
= 15.69,
df = 10 for enjoyment;
2
= 15.43, df = 10 for authoritarian control), but compared
with the integrative model (Model 4) the fit significantly deteriorated (
2
= 12.33,
df = 4 for enjoyment,
2
= 12.06,df = 4 for authoritarian control).
The next model (Model 6) predicted that the relation between G
1
child sexual
abuse and parenting is not significant when examined in the context of other
maternal factors. As indicated by the fit indices, the fit for this model was also
acceptable (
2
= 15.76, df = 11 for enjoyment;
2
= 16.89, df = 11 for authoritarian
control) and deterioration in fit from Model 5 was negligible for enjoyment (
2
= .07,df = 1) as well as for authoritarian control (
2
= 1.46,df = 1). The final
67
[Model 3. G
1
Baseline model]
[Model 5. Direct effect model]
[Model 7. Indirect effect model]
[Model 4. G
1
Integrative model]
[Model 6. Spurious model]
Figure 4. Summary of model
comparison tests for G
1
parenting
Note. Mothers’ ethnic minority status,
SES ratings, and G
2
age were
controlled for in each model.
Covariance was given among three
exogenous variables. Social, Social
support from family-of-origin
sub-network; CSA, Child sexual abuse;
Par, Parenting; DP, Discipline
punitiveness; DES, Dissociative
symptoms
G
1
Par
G
2
CSA
G
1
Social
G
1
DES
G
0
DP
G
1
CSA
G
1
Par
G
2
CSA
G
1
Social
G
1
DES
G
0
DP
G
1
CSA
G
1
Par
G
2
CSA
G
1
Social
G
1
DES
G
0
DP
G
1
Par
G
2
CSA
G
1
Social
G
1
DES
G
0
DP
G
1
Par
G
2
CSA
G
1
Social
G
1
DES
G
0
DP
G
1
CSA
G
1
CSA
G
1
CSA
68
model (Model 7) reproduced Model 3, but the path between G
1
child sexual abuse
and parenting practice was removed, hence predicting indirect effects of G
1
childhood history variables. The fit indices indicated an excellent fit (
2
= 3.39, df =
7 for enjoyment;
2
= 4.95, df = 7 for authoritarian control). The deterioration in fit
from the integrative model was negligible (
2
= .03,df = 1 for enjoyment;
2
=
1.58,df = 1 for authoritarian control), and improvement in fit from Model 6 was
significant (
2
= 12.37,df = 5 for enjoyment;
2
= 11.94,df = 5 for
authoritarian control). Overall, alternative models allowing indirect effects showed
better fit. Model 7 was the most accurate and parsimonious representation among the
models for both dimensions of parenting in this sample.
Table 3. Model Fit Comparison among G
1
Parenting Models (N = 127)
Model
2
df
p
RM-
SEA
from
integrative
model
2
df
from
previous
model
2
df
[3.1] Enjoyment of child and parental role
Model 3 (G
1
baseline – no relations) 31 15 .01 .09
Model 4 (G
1
integrative model) 3 6 .76 .00 27 9
Model 5 (direct effect model) 16 10 .11 .07 12 4 12 4
Model 6 (spurious effect model) 16 11 .15 .06 12 5 0 1
Model 7 (indirect effect model) 3 7 .85 .00 0 1 12 5
[3.2] Authoritarian control
Model 3 (G
1
baseline – no relations) 27 15 .03 .08
Model 4 (G
1
integrative model) 3 6 .77 .00 23 9
Model 5 (direct effect model) 15 10 .12 .07 12 4 12 4
Model 6 (spurious effect model) 17 11 .11 .07 14 5 2 1
Model 7 (indirect effect model) 5 7 .67 .00 2 1 12 5
69
Table 4 presents the parameter estimates from each model. The estimates
from Model 7 for the enjoyment dimension of parenting indicated that the G
0
mothers’ discipline punitiveness was positively related to the G
1
mothers’ current
level of dissociative symptoms ( = .22), which was negatively associated with the
enjoyment dimension ( = -.25). This finding is in line with the indirect effect
hypothesis through dissociative symptoms. However, small path coefficients
between the mothers’ child sexual abuse experiences and other mediators do not
support the indirect hypothesis in mothers’ child sexual abuse. A graphic display of
these results is presented in Figure 5.
The estimates from Model 7 for the authoritarian control dimension of
parenting showed that G
0
mothers’ discipline punitiveness was positively related to
the G
1
mothers’ current level of dissociative symptoms ( = .22), but the dissociative
symptoms were not related to the authoritarian control dimension of parenting. The
G
0
mothers’ discipline punitiveness independently predicted the G
1
mothers’
authoritarian control style of parenting (s = - .29). These findings support the
intergenerational hypothesis of parenting style, but the direction was opposite to the
expectation, and negative. However, direct and indirect path coefficients of child
sexual abuse were small, and not accurate in predicting authoritarian control.
70
Table 4. Parameter Estimates for the G
1
Parenting Models (N = 127)
Variable M3 M4 M5 M6 M7
[4.1]Enjoymentofchildandparentalrole
(t) (t) (t) (t) (t)
G
1
CSA →G
1
Parenting .00=
-.01
(-.16)
- .02
(-.17)
.00 = .00 =
G
0
Parenting → G
1
Parenting .00 =
- .12
(-1.25)
- .12
(-1.32)
- .12
(-1.35)
- .12
(-1.27)
G
2
CSA → G
1
Parenting .00 =
- .15
(-1.71)
- .15
(-1.71)
- .16
(-1.86)
- .16
(-1.86)
G
1
Social support → G
1
Parenting
.00 =
. 02
(.23)
.02
(.24)
.02
(.26)
.02
(.25)
G
1
Dissociation → G
1
Parenting .00 =
- .25
(-2.74)
- .24
(-2.78)
- .25
(-2.81)
- .25
(-2.76)
G
1
CSA → G
1
Social support .00 =
- .15
(-1.63)
.00 = .00 =
- .15
(-1.63)
G
0
Parenting → G
1
Social
support
.00 =
- .11
(-1.17)
.00 = .00 =
- .11
(-1.17)
G
1
CSA → G
1
Dissociation .00 =
.10
(1.07)
.00 = .00 =
.12
(1.07)
G
0
Parenting → G
1
Dissociation .00 =
.22
(2.43)
.00 = .00 =
.22
(2.43)
[4.2] Authoritarian control
(t) (t) (t) (t) (t)
G
1
CSA → G
1
Parenting .00 =
- .12
(-1.32)
- .11
(-1.22)
.00 = .00 =
G
0
Parenting → G
1
Parenting .00 =
- .29
(-3.14)
- .28
(-3.18)
- .29
(-3.27)
- .30
(-3.21)
G
2
CSA → G
1
Parenting .00 =
.05
(.66)
.06
(.64)
.02
(.27)
.02
(.26)
G
1
Social support → G
1
Parenting
.00 =
- .05
(-.55)
- .04
(-.40)
- .03
(-.28)
- .04
(-.42)
G
1
Dissociation → G
1
Parenting .00 =
.12
(1.23)
.10
(1.20)
.10
(1.13)
.11
(1.19)
G
1
CSA → G
1
Social support .00 =
- .15
(-1.64)
.00 = .00 =
- .15
(-1.62)
G
0
Parenting → G
1
Social
support
.00 =
- .11
(-1.20)
.00 = .00 =
- .11
(-1.19)
G
1
CSA → G
1
Dissociation .00 =
.10
(1.06)
.00 = .00 =
.10
(1.05)
G
0
Parenting → G
1
Dissociation .00 =
.22
(2.37)
.00 = .00 =
.22
(2.37)
Note. The “ = ” indicates a fixed parameter (at “0”). CSA, Child Sexual Abuse; G
0
,
Generation 0; G
1
Generation 1; G
2
Generation 2.
71
[Enjoyment of child and parental role]
[Authoritarian control]
Figure 5. Standardized parameter estimates from Model 7.
Note. Continuous lines indicate accurately estimated coefficients (t value < ±1.96). Discontinuous
lines indicate estimable coefficients (t value < ±1.96). 0= indicates fixed parameter (at “0”). Age, SES,
and race were controlled for. Covariance was given among three exogenous variables in each model.
G
1
, Generation 1; G
2
Generation 2; G
0
Generation 0; CSA, Child sexual abuse; FOO,
Family-of-origin; DP, Discipline punitiveness.
G
1
CSA
G
1
Enjoyment
of Child &
Parental Role
.22
0 =
- .25
G
1
Social
Support
(FOO)
G
0
DP
G
2
CSA
G
1
Dissociative
Symptoms
0=
- .29
G
1
Social
Support
(FOO)
G
1
Authoritarian
Control
G
0
DP
G
1
CSA
G
2
CSA
.22
G
1
Dissociative
Symptoms
72
5.2.3 Results of Multi-Group Analysis of G
1
Mothers’ Parenting Models
The next set of analyses examined the group differences in structural
relationships hypothesized in the G
1
integrative model (Model 4). All the parameters
of paths hypothesizing the relationships between maternal factors and G
2
behavior
adjustment were first constrained across groups, and then allowed to vary one by one.
Improvements in fit from the no-difference model were examined using chi-square
difference tests. Results from a series of comparison tests indicated no significant
chi-square differences in any of the path coefficients in the model. It was reasonable
to conclude that structural relationships across the two independent samples were
similar.
5.2.4 Summary of Findings of G
1
Mothers’ Parenting Analyses
In summary, a series of model comparison tests showed that the indirect
effect model (Model 7) is the most accurate and parsimonious model in representing
the structural relationships among the variables. The pattern of relationships was
similar between the enjoyment and authoritarian control dimensions of parenting.
Specifically, the current data supported the indirect effect hypothesis of dissociative
symptoms between punitive parenting experiences in childhood and current
enjoyment of child and parental roles. This study also supported the direct effect
hypothesis of childhood parenting experience on authoritarian control. However, the
effects of current social support from family-of-origin were not accurate in both
dimensions of parenting. Both direct and indirect effects of mothers’ child sexual
73
abuse history on their parenting practice were not significantly different than zero.
Finally, no group differences by G
2
child sexual abuse status were detected in the
structural relationships hypothesized in the integrative model.
5.3 Testing G
2
Daughters’ Behavior Outcome Models
5.3.1 G
2
Baseline Models
Based on the G
1
parenting model (Model 7), four measures of G
2
daughters’
behavior symptoms self-reported at Time 4 assessment were introduced. In order to
determine the baseline model, four measures from the YSR were simultaneously
included in each model. The first model (Model 8) represents the baseline model,
hypothesizing that the four indicators are correlated with one another. Models 9 and
10 represent the single-factor and two-factor model, respectively (Figure 6).
Table 5 summarizes the results from the G
2
baseline model comparison tests.
Model fit indices indicated excellent fits for the G
2
baseline models examining the
function of enjoyment (
2
= 29.56, df = 31) as well as authoritarian control (
2
=
45.43, df = 35). Deterioration in fit from the baseline (Model 8) to the one-factor
model (Model 9) was significant for both dimensions of parenting (
2
= 22.62,df
= 11 for enjoyment;
2
= 25.50,df = 11 for authoritarian control). Deterioration in
fit from the baseline to the two-factor model (Model 10) was also significant (
2
=
14.75,df = 7 for enjoyment;
2
= 17.04,df = 7 for authoritarian control). In
subsequent analyses testing structural relationships, four measures from YSR were
examined in separate models.
74
Figure 6. Summary of model comparison tests for G
2
behavior adjustment baseline model
Note. Mothers’ ethnic minority status, SES ratings and G
2
daughters’ ages were partialled out in each
analysis. Covariance was given among the exogenous variables. G
1
, Generation 1; G
2
Generation 2;
CSA, Child sexual abuse; Social, Social support from family-of-origin; DES, Dissociation; Par,
Parenting; BHV, Behavior adjustment; E, externalizing behavior; I, Internalizing behavior.
G
1
CSA
G
2
DM
G
2
Ag
G
2
WD
G
2
IB
G
0
Par
G
1
DES
G
2
CSA
G
1
Par
E
I
G
1
Social
G
1
CSA
G
2
DM
G
2
Ag
G
2
WD
G
2
IB
G
0
Par
G
1
DES
G
2
CSA
BHV
G
1
Par
G
1
Social
G
1
CSA
G
2
DM
G
2
Ag
G
2
WD
G
2
IB
G
0
Par
G
1
DES
G
2
CSA
G
1
Par
G
1
Social
[Model 9. Single-factor
Model for G
2
Behavior
Adjustment]
[Model 10. Two-factor
Model for G
2
Behavior
Adjustment]
[Model 8. Baseline for G
2
Behavior Adjustment]
75
5.3.2 Results of G
2
Daughters’ Behavior Outcome Models
A graphical representation of each model for model comparison tests is
presented in Figure 7. In each set of model comparison tests, two measures of
internalizing symptoms (i.e., withdrawal/depressed and immature/bizarre behavior)
were examined first, and then measures of externalizing symptoms (i.e., delinquent
misbehavior and aggressive behavior) were tested.
Table 6 presents the results of the models examining the relationships among
G
1
mothers’ parenting and G
2
daughters’ behavior measures. Across four measures
of the YSR, the baseline model representing no relations showed an excellent fit.
Improvements in fit from the baseline model to Model 12 and Model 13 were not
significant. The same pattern was detected across two dimensions of parenting.
Table 5. Model Fit Comparison for G
2
Baseline Models (N = 127)
Model
2
df RMSEA
Δ from base
2
df
Δ fromprevious
2
df
[6.1] Enjoyment of a Child and Parental Role
Model 8 (G
2
baseline) 30 31 .00
Model 9 (one factor) 52 42 .04 23 11
Model 10 (two factor) 44 38 .04 15 7 8 4
[6.2] Authoritarian Control
Model 8 (G
2
baseline) 34 31 .03
Model 9 (one factor) 59 42 .06 25 11
Model 10 (two factor) 51 38 .05 17 7 8 4
76
Figure 7. Summary of model comparison tests for G
2
behavioral adjustment.
Note. Mothers’ ethnic minority status, SES ratings and G
2
daughters’ ages were partialled out in each
analysis. Covariance was given to the exogenous variables. G
1
, Generation 1; G
2
, Generation 2; CSA,
Child sexual abuse; Social, Social support from family-of-origin; DES, Dissociation; Par, Parenting;
BHV, Behavior adjustment.
G
1
CSA
G
0
Par
G
1
DES
G
2
CSA
G
1
Par
G
2
BHV
G
1
CSA
G
2
BHV
G
0
Par
G
1
DES
G
2
CSA
G
1
Par
G
1
Social
G
1
Social
G
1
CSA
G
0
Par
G
1
DES
G
2
CSA
G
1
Par
G
2
BHV
G
1
Social
[Model 11. G
2
Baseline Model]
[Model 12. G
2
Integrative Model]
[Model 13. G
2
Mediating Model]
77
Table 6. Model fit comparison for the models predicting G
2
behavior adjustment (N=127)
from
base
from
previous
Model
2
df
RM-
SEA
2
df
2
df
[6.1] Enjoyment of Child and Parental Role
Withdrawal/Depressed
Model 11 (baseline) 9.10 13 .05
Model 12 (integrative) 3.56 9 .00 5.54 4
Model 13 (mediation) 6.47 11 .03 2.63 2 2.91 2
Immature/Bizarre
Model 11 (baseline) 11.72 13 .00
Model 12 (integrative) 4.54 9 .00 7.18 4
Model 13 (mediation) 7.73 11 .00 3.99 2 3.19 2
Delinquent Misbehavior
Model 11 (baseline) 16.90 13 .05
Model 12 (integrative) 7.33 9 .00 9.57 4
Model 13 (mediation) 12.03 11 .03 4.87 2 4.70 2
Aggressive behavior
Model 11 (baseline) 12.48 13 .00
Model 12 (integrative) 5.16 9 .00 7.32 4
Model 13 (mediation) 6.64 11 .00 5.84 2 1.48 2
[6.2] Authoritarian Control
Withdrawal/Depressed
Model 11 (baseline) 21.44 17 .05
Model 12 (integrative) 17.44 13 .05 4.00 4
Model 13 (mediation) 19.48 15 .05 1.96 2 2.04 2
Immature/Bizarre
Model 11 (baseline) 23.13 17 .05
Model 12 (integrative) 18.35 13 .06 4.78 4
Model 13 (mediation) 20.59 15 .05 2.54 2 2.24 2
Delinquent Misbehavior
Model 11 (baseline) 33.93 17 .09
Model 12 (integrative) 24.78 13 .09 9.15 4
Model 13 (mediation) 28.27 15 .08 5.66 2 3.49 2
Aggressive behavior
Model 11 (baseline) 24.95 17 .10
Model 12 (integrative) 18.91 13 .06 6.04 4
Model 13 (mediation) 20.02 15 .05 4.93 2 4.93 2
78
5.3.3 Results of Multi-Group Analysis of G
2
Daughters’ Behavior Outcome
Models
The next set of analyses examined the group differences in structural
relationships hypothesized by the G
2
integrative model. This examination was
conducted using one-degree-of-freedom, nested chi-square tests of cross-group
equality constraints. Model 14 represents the baseline model, in which all the path
coefficients are constrained to be equal across groups. Model 15 allowed the path
from maternal social support to G
2
behavioral adjustment to vary across the two
groups. Model 16 allowed the effect of the mothers’ parenting to differ across the
two groups. Finally, Model 17 hypothesized that the effect of the mothers’
dissociative symptoms on their daughters’ behavior symptoms may differ between
the G
2
abuse and comparison samples.
Table 7 summarizes the results from a series of multi-group comparison tests
of the G
2
integrative model. First, results from a series of comparison tests
examining the function of social support provided evidence that the relationship
between social support from family-of-origin and G
2
internalizing behavior problems
(Model 15) may differ. When the path was allowed to vary, the improvement in fit
from the no-difference model (Model 14) was significant (
2
= 6.37,df = 1 for
withdrawal/depressed;
2
= 6.64,df = 1 for immature/bizarre) for the enjoyment
model as well as (
2
= 6.16,df = 1 for withdrawal/depressed;
2
= 6.61,df = 1
for immature/bizarre) for the authoritarian control model.
79
Table 7. Statistical comparison of G
2
integrative model (N=127)
from baseline
2
df
2
df
[7.1] Enjoyment of Child and Parental Role
Withdrawal/Depressed
Model 14 (baseline – no difference) 30.60 24
Model 15 (G
1
social effect differ) 24.23 23 6.37 1
Model 16 (G
1
parenting effect differ) 27.43 23 3.17 1
Model 17 (G
1
dissociation effect differ) 30.12 23 .48 1
Immature/Bizarre
Model 14 (baseline – no difference) 33.74 24
Model 15 (G
1
social effect differ) 27.10 23 6.64 1
Model 16 (G
1
parenting effect differ) 27.25 23 6.49 1
Model 17 (G
1
dissociation effect differ) 32.36 23 1.38 1
Delinquent Misbehavior
Model 14 (baseline – no difference) 22.21 24
Model 15 (G
1
social effect differ) 21.04 23 1.74 1
Model 16 (G
1
parenting effect differ) 22.21 23 .00 1
Model 17 (G
1
dissociation effect differ) 21.58 23 .63 1
Aggressive behavior
Model 14 (baseline – no difference) 27.56 24
Model 15 (G
1
social effect differ) 24.17 23 3.39 1
Model 16 (G
1
parenting effect differ) 22.25 23 5.31 1
Model 17 (G
1
dissociation effect differ) 27.43 23 .13 1
[7.2] Authoritarian Control
Withdrawal/Depressed
Model 14 (baseline – no difference) 31.91 24 1
Model 15 (G
1
social effect differ) 25.75 23 6.16 1
Model 16 (G
1
parenting effect differ) 19.68 23 12.23 1
Model 17 (G
1
dissociation effect differ) 31.66 23 .25 1
Immature/Bizarre
Model 14 (baseline – no difference) 30.20 24 1
Model 15 (G
1
social effect differ) 23.59 23 6.61 1
Model 16 (G
1
parenting effect differ) 20.86 23 9.34 1
Model 17 (G
1
dissociation effect differ) 29.14 23 1.06 1
Delinquent Misbehavior
Model 14 (baseline – no difference) 24.14 24 1
Model 15 (G
1
social effect differ) 22.59 23 1.55 1
Model 16 (G
1
parenting effect differ) 23.23 23 .91 1
Model 17 (G
1
dissociation effect differ) 23.82 23 .32 1
Aggressive behavior
Model 14 (baseline – no difference) 23.05 24 1
Model 15 (G
1
social effect differ) 19.48 23 3.57 1
Model 16 (G
1
parenting effect differ) 18.25 23 4.80 1
Model 17 (G
1
dissociation effect differ) 22.96 23 .09 1
80
The results also indicated that the function of parenting on G
2
behavior
adjustment may differ by G
2
child sexual abuse status. With regard to the enjoyment
dimension of parenting, when the path was allowed to vary (Model 16),
improvement in fit from the no-difference model (Model 14) was significant in the
models examining both immature/bizarre behavior (
2
= 6.49,df = 1) and
aggressive behavior (
2
= 5.31,df = 1). With regard to the authoritarian control
dimension, improvement in fit from the no-difference model (Model 14) was
significant in the models examining withdrawal/depressed behavior (
2
= 12.23,
df = 1), immature/bizarre behavior (
2
= 9.34,df = 1) and aggressive behavior
(
2
= 4.80,df = 1). No group differences were detected in any of the models: the
effects of the mothers’ dissociative symptoms on all four measures were similar
across the four measures in both parenting models.
Parameter estimates from the multi-group analyses are presented here
graphically. Figure 8 summarizes the results of the enjoyment dimension model, and
Figure 9 summarizes the authoritarian control dimension. In terms of the effect of
social support from the mother’s family-of-origin, the patterns were similar across
the two parenting models. The G
1
mothers’ social support from their family-of-origin
sub-network had positive impact in decreasing the G
2
s’ withdrawal/depressed
behaviors (unstandardized = -2.55 in enjoyment and -2.53 in authoritarian control)
in the comparison sample. However, the beneficial effects of social support from the
mothers’ family-of-origin were not detected in the abuse sample: the effect was not
significantly different from zero on withdrawal/depressed behaviors across two
81
parenting models, and significantly increased immature/bizarre behaviors
(unstandardized = 2.46 in both enjoyment and authoritarian control models).
In terms of the effect of maternal parenting practice on the daughters’
behavior adjustment, the enjoyment dimension of parenting did not predict any of the
YSR measures in the comparison sample. However, the impact of G
1
mothers’
enjoyment of child and parental roles was positive in decreasing immature/bizarre
(unstandardized = -.28) and aggressive (unstandardized = -.19) behaviors.
Maternal authoritarian control showed opposite patterns. Maternal
authoritarian control had a positive impact on decreasing withdrawal/depressed
(unstandardized = -.49), immature/bizarre (unstandardized = -.42), and
aggressive (unstandardized = -.24) behaviors. However, the impact of maternal
authoritarian control on daughters’ behavior measures was inaccurate in the abuse
sample. Moreover, maternal authoritarian control increased the level of
withdrawal/depressed behavior.
5.3.4 Summary of Findings of G
2
Daughters’ Behavior Outcome Analyses
In summary, a series of model comparison tests supported the no-relation model
across the four measures from YSR in the total sample. This pattern was similar for
the enjoyment and authoritarian control dimensions. However, in subsequent
multi-group analyses, the results suggested group differences in structural
relationships. First, the effect of social support mothers received from their
family-of-origin on G
2
internalizing behaviors significantly differed. Specifically, for
82
[Internalizing Behaviors]
[Externalizing behaviors]
Figure 8. Unstandardized parameter estimates from multi-group analyses – Enjoyment of
child and parental role. Note. Continuous lines indicate accurately estimated coefficients (t value <
±1.96). Discontinuous lines indicate estimable coefficients (t value < ±1.96). Bold arrows indicate
significant group moderators at p < .05. Parameter estimates for the abused group are depicted above
estimates for the comparison group. 0= indicates fixed parameter (at “0”). Age, SES, and race were
controlled for and covariance was given among three exogenous variables in each model.
*
p < .05,
n/s
p > .05. DP, Discipline punitiveness, Social, Social support from family-of-origin; Enj, Enjoyment of
child and parental role.
G
1
Enj
G
1
Social
G
0
DP
G
1
DES .44
*
1.49
n/s
-2.55
*
G
1
CSA
-.24
*
G
2
Withdrawal/
Depressed
G
1
Enj
G
1
Social
G
0
DP
G
1
DES .44
*
2.46
*
-1.16
n/s
-.28
*
.20
n/s
G
1
CSA
-.24
*
G
2
Immature/
Bizarre
G
1
Enj
G
1
Social
G
0
DP
G
1
DES .44
*
G
1
CSA
-.24
*
G
2
Delinquent
Misbehavior
G
1
Enj
G
1
Social
G
0
DP
G
1
DES .44
*
-.19
*
.16
n/s
G
1
CSA
-.24
*
G
2
Aggressive
Behavior
83
[Internalizing Behaviors]
[Externalizing behaviors]
Figure 9. Unstandardized parameter estimates from multi-group analyses – Authoritarian
control. Note. Continuous lines indicate accurately estimated coefficients (t value < ±1.96).
Discontinuous lines indicate estimable coefficients (t value < ±1.96). Bold arrows indicate significant
group moderators at p < .05. Parameter estimates for the abused group are depicted above estimates
for the comparison group. 0= indicates fixed parameter (at “0”). Age, SES, and race were controlled
for and covariance was given among three exogenous variables in each model.
*
p < .05,
n/s
p > .05.
DP, Discipline punitiveness, Social, Social support from family-of-origin; Con, Authoritarian control.
G
1
Enj
G
1
Social
G
0
DP
G
1
DES
1.47
n/s
-2.53
*
G
1
CSA
-.29
* G
2
Withdrawal/
Depressed
G
1
Con
G
1
Social
G
0
DP
G
1
DES
2.46
*
-1.13
n/s
-.02
n/s
-.42
*
G
1
CSA
-.29
* G
2
Immature/
Bizarre
G
1
Con
G
1
Social
G
0
DP
G
1
DES
G
1
CSA
-.29
*
G
2
Delinquent
Misbehavior
G
1
Con
G
1
Social
G
0
DP
G
1
DES
.11
n
/s
-.24
n/s
G
1
CSA
-.29
*
G
2
Aggressive
Behavior
.26
*
-.49
*
84
the comparison sample, the social support the mothers received from their family of
origin decreased their daughters’ problem behaviors in internalizing symptoms,
while the effect was not significantly different from zero for the abuse sample.
Further, the impact of maternal parenting differed by G
2
sexual abuse status.
Generally, for the comparison sample, the authoritarian control dimension showed a
protective impact, while, for the abuse sample, the mothers’ enjoyment of parenting
exerted a beneficial impact.
Depending on the measures of YSR, the relationships between mothers’
parenting and G
2
behaviors showed different patterns. Although the chi-square
difference tests for G
2
baseline models indicated that the baseline model
hypothesizing correlations among the four measures from the YSR is the most
accurate and parsimonious representation, the model fit for two-factor model also
showed an excellent fit. Considering relatively small sample size (n=127), it was
worth while to replicate the results based on the two-factor model of G
2
behavioral
adjustment (internalizing vs. externalizing behaviors). Parameter estimates based on
the two-factor model produced same patterns of relationships with the results based
on correlation model. For internalizing behaviors, social support from maternal
family of origin decreased such behaviors in comparison sample, but it increased in
abuse sample. For externalizing behaviors, mothers’ parenting practices had direct
impact, but specific dimensions of parenting exerted beneficial impact differed
across groups: enjoyment had beneficial effect in abuse sample, but authoritarian
control did in comparison sample.
85
Chapter VI. Discussion
The primary goal of this study was to examine parenting practices among
non-offending mothers of sexually abused girls, and to investigate its long-term
impact on their daughters’ behavioral adjustment. Guided by the
developmental-ecological perspective of parenting (Belsky, 1984; Bronfenfrenner,
1979), the five determinants of parenting were included in the parenting model.
Based on the existing empirical studies, different patterns of relationships among the
determinants were hypothesized, and a series of model comparison tests were
conducted. Further, children’s behavior problems, which were self-reported
approximately seven years after the disclosure of child sexual abuse, were introduced
to the model, and the relationships between maternal factors and their children’s
behavior outcomes were explored.
6.1 Major Findings and Reevaluation Of Theories
6.1.1 G
1
Mothers’ Parenting Practice Models
This study found that mothers’ childhood experiences of punitive disciplines,
as well as their current mental health, are important determinants of parenting
practices among non-offending mothers of sexually abused girls. The first
relationship studied related to mothers’ childhood experiences of their own mothers’
(G
0
mothers) discipline punitiveness. The results indicated that, depending on the G
1
parenting dimensions examined here (i.e., enjoyment of a child and parental role,
authoritarian control), the effects of G
0
mothers’ discipline punitiveness differed. G
0
86
mothers’ discipline punitiveness had a direct impact on the G
1
mothers’ level of
providing authoritarian control. This provides supportive evidence for the
intergenerational hypothesis of parenting. In predicting the G
1
mothers’ parenting
dimension of enjoyment of a child and parental role, the effect of the G
0
mothers’
discipline punitiveness was fully mediated through their current dissociative
symptoms, which is in line with the theory that maternal dissociative symptoms may
be a mechanism by which parents’ childhood traumas are communicated to the next
generation.
However, in terms of the direct relationship between the G
0
mothers’
discipline punitiveness and the G
1
mothers’ authoritarian control, it should be noted
that the direction of the direct effect is negative in this sample, meaning that more
punitive discipline experiences from G
0
mothers predicted less authoritarian control
among G
1
mothers. This finding is in contrast with the result detected in the
physically abusive families (Corse, Schnid, & Tirckett, 1990). This is also opposite
to the explanation for intergenerational phenomena in parenting. According to
Scaramella and Conger (2003), how one is parented during childhood becomes one’s
learned parenting repertoire, and automatically unfolds when one becomes a parent
in adulthood. In this line of reasoning, it has been hypothesized that the likelihood of
this automatic response may be greater under more stressful conditions. However,
the moderating effect of G
2
daughters’ child sexual abuse on this relationship was
not detected in this sample.
87
Interpretation of a direct, inverse relationship between the G
0
mothers’
discipline punitiveness and the G
1
mothers’ authoritarian control is challenging. It is
possible that the participating mothers who experienced punitive discipline during
childhood made conscious efforts to parent differently from their mothers. Also, it
should be noted that the distribution of the G
1
authoritarian control variable in the
current sample ranged from 18 to 51, out of possible range of 1 to 77. This suggests
that authoritarian control in this data may have some positive connotation. That is, it
may be close to moderate control, which has been suggested as one of the important
constituents of optimal parenting (Mason, Cauce, Gonzlales, & Hiraga, 1996), rather
than authoritarian control in its excessive form. The results from the G
2
behavioral
adjustment models also showed a similar tendency. For example, the mothers’
degree of providing authoritarian control had a positive impact in decreasing their
daughters’ behavior problems in the comparison sample.
It is reasonable to expect that mothers who experienced punitive discipline
during childhood may have difficulty in providing and maintaining an optimal level
of parental control with their children. This is reflected in the direct, inverse
relationship between childhood experiences of punitive discipline and current
authoritarian control in this data. However, this explanation still leaves unanswered
the question for the process of how punitive discipline experiences led to failure to
provide an optimal level of parental control, which is not mediated by the mothers’
current psychological functioning. Clearly, more studies are needed to accurately
explain the process by which the G
0
mothers’ discipline punitiveness directly leads
88
to the less authoritarian control dimension of parenting in this data, before any
definitive conclusion for this relationship is put forth.
In this sample, the mothers’ childhood sexual abuse histories were not a
significant predictor of their current functioning, nor their parenting practices.
Initially, three competing hypotheses of the function of the mothers’ childhood
sexual abuse in their later parenting practices (i.e., direct, indirect, and spurious
effect) were made. The current findings did not support any of these hypotheses.
No significant relationship between the mothers’ childhood sexual abuse and
their subsequent parenting may reflect the sample characteristics. In some ways, this
finding is opposite to the current body of literature on parenting practices among
childhood sexual abuse survivors, from which the hypotheses for this study were
formed. However, the same body of literature also provides evidence that mothers’
childhood sexual abuse histories are not predictive of their parenting practices,
among mothers with additional multiple challenges. For example, similar findings of
no significant relationship between mothers’ childhood sexual abuse and their
subsequent parenting have been reported in such high-risk samples as low-income,
welfare-dependent mothers, and adolescent mothers.
Similar findings have also been reported in the literature on non-offending
mothers. According to Elliot and Carnes’s review (2001), mothers’ childhood sexual
abuse was one of the most frequently researched predictors of their supportive
behavior following disclosure of their daughters’ child sexual abuse. The existing
literature shows fairly consistent findings, which indicate that mothers’ childhood
89
sexual abuse is not a significant predictor of their protective or supportive behavior
following the disclosure. Although in their early stages, studies on the non-offending
mothers of sexually abused girls have shown that these mothers confront multiple
psychosocial challenges, while they have limited familial psychological, emotional,
and social resources to facilitate coping with stressful events and everyday
adversities. In this multiply-challenging context, it is possible that maternal
childhood sexual abuse has limited predictive power of their parenting behavior.
In addition, it should be noted that, in contrast to mothers’ childhood
experiences of punitive discipline, a mediating hypothesis of dissociative symptoms
was not supported in childhood sexual abuse experience. This finding is in line with
previous findings (Narang & Conteras, 2000; Vezina, Mireille, Robert, & Pierre,
2005), which report that dissociative symptoms mediated the relationship between
childhood physical and emotional abuse, and later parenting, but the same pattern
was not detected for childhood sexual abuse trauma. Although the paucity of
research on the association between dissociative symptoms and parenting makes this
reasoning purely speculative, this may imply that, depending on types of childhood
trauma, different causal mechanisms operate in predicting parenting practices.
Overall, parenting has been suggested as one of the possible pathways
through which mothers’ experiences of childhood sexual abuse are transmitted to the
next generation, affecting their offspring’s development. However, the relationship
between childhood sexual abuse and parenting was not supported in this sample.
90
In addition to these consistent findings, there is a result that deserves further
interrogation. Social support from mothers’ family-of-origin was not predictive of
their enjoyment of a child and parenting role, nor the authoritarian control dimension
of parenting. Initially, two competing hypotheses were made, predicting either
beneficial or non-significant functions for the abuse sample. However, the effect was
not significant in the total sample, nor in multi-group analyses. This might be due to
the fact that for the current study the dimension of social support was limited only to
instrumental support from the family-of-origin sub-network. Based on these findings,
overgeneralization of the effect of social support should not be made across different
sources and types of social support. It is possible that other types of support, from
different sources, may exert positive impact, as the social support literature has
suggested. In order to more accurately capture the function of social support in this
population, more studies are needed to examine it from multidimensional,
multiple-sub-network perspectives.
Overall, this study showed that mothers’ childhood experiences of punitive
discipline, and their current dissociative symptoms, are important determinants of
parenting practices among non-offending mothers of sexually abused girls. However,
the question still remains regarding the different pattern of relationships detected in
the G
1
mothers’ childhood experiences of sexual abuse and punitive discipline. This
sample supported the intergenerational hypothesis of parenting, as well as the
mediation hypothesis of dissociative symptoms in the case of childhood experiences
of punitive discipline. By contrast, the mothers’ childhood sexual abuse did not show
91
such a pattern. Clearly, more theoretical discussions and empirical investigations of
explanatory mechanisms are needed, in order to accurately capture the relationships
between various types of childhood trauma and subsequent parenting.
6.1.2 G
2
Daughters’ Behavioral Adjustment Models
The G
2
daughters’ behavior problems that were self-reported approximately
seven years after the disclosure of sexual abuse, were introduced into the indirect
effect model of G
1
parenting. Initially, it was expected that the mothers’ social
support from their family-of-origin sub-networks, their dissociative symptoms, their
parenting styles, as well as their daughters’ own sexual abuse experiences, would
independently contribute to the daughters’ behavior problems. Guided by the
resilience perspective, this study also predicted that the beneficial effects of maternal
social support and parenting would be greater for the abuse sample.
In terms of the mothers’ parenting styles, the two dimensions of parenting
showed different patterns depending on the daughters’ child sexual abuse status. For
the comparison sample, the mothers’ authoritarian control had the beneficial impact
of decreasing their children’s behavior problems. However, for the abuse sample, the
mothers’ degree of enjoying a child and parental role had the protective impact of
decreasing the abused children’s behavior problems. The impact of authoritarian
control was imperceptible in the abuse sample.
These findings are consistent with the previous studies suggesting that their
mothers’ support and care has a protective impact on the behavioral adjustment of
92
sexually abused girls. Further, this study shows the ameliorating effect not only of
abuse-specific support, but also of overall care, such as parenting. Moreover, this
study suggests that mothers’ parental attitudes, and their subjective experiences of
parenting, such as enjoyment of child and parental roles, may exert a protective
impact on their sexually abused daughters, rather than the authoritarian control
dimension of parenting. It is speculated that mothers’ enjoyment of a child and
parental role related to the level of support for the abused children.
The second pattern in the findings of this study is the relationship between
social support from the maternal family-of-origin sub-network and children’s
behavior problems. Initially, based on children’s social network literature, as well as
the resilience literature on maltreated children, this study predicted a positive impact
of support from the mothers’ informal networks. This hypothesis was supported in
the comparison sample, but no beneficial effect was perceptible in the abuse sample.
In the case of immature/bizarre behavior, the presence of greater support from the
family-of-origin predicted even more behavior problems. Contrary to our
expectations, the current study showed that the beneficial effect of the social support
mothers received from their family-of-origin could be compromised in the abuse
sample.
In some ways, this finding is opposite to that of Leifer and her colleagues’
studies (2001; 2004), which suggested that mothers’ current quality relationships
with their own mothers were predictive of their daughters’ sexual abuse experiences,
as well as the mothers’ supportive behaviors following the disclosure of their
93
daughters’ child sexual abuse. However, when the direct effect of support from the
mothers’ family-of-origin sub-network on their children’s behavioral adjustment was
examined, the actual support mothers received from their maternal family-of-origin
in and of itself had either no significant impact, or a negative impact, on the abused
girls’ behavioral adjustment.
Interpretation of the direct, negative impact of the social support mothers
receive from their maternal family-of-origin sub-network on their children’s
internalizing symptoms is challenging. Given the higher rates of childhood trauma
experiences by G
0
parents, as well as more discontinuity of care from G
0
mothers
detected among the mothers of sexually abused girls in this sample (Kim et al., in
press), it is plausible that the quality of support from the family-of-origin
sub-network may be compromised in the abuse sample. Social support from family
of origin can be interference and perpetuation of negative functions in the abuse
sample.
It is also presumed that whether mothers clearly perceive their upbringing
and childhood trauma experiences, as well as whether such trauma is resolved, may
condition the effectiveness of support from the family-of-origin in abuse families. In
other words, among survivors of childhood trauma, the beneficial impact of support
from the family-of-origin sub-network can be compromised without adequate
appraisal of their own upbringing and abuse experiences. Future research should
address and scrutinize the relationships between the mothers’ perceptions of their
own parents’ parenting, whether and how they parent differently from their own
94
parents, and their current interactions with their family-of-origin. This may provide
additional insights for intergenerational continuity or discontinuity in child
maltreatment.
While maternal dissociative symptoms were an important determinant of
their parenting practices, according to this study, the direct effect hypothesis of the
effect of maternal dissociative symptoms on children’s behavioral adjustment was
not supported in this sample. Dissociation literature suggests that, compared with
other types of psychopathology, such as depression, the genetic mechanism is less
important in the etiology of dissociative symptoms (Ross, 1999; Macfie, Cicchetti, &
Toth, 2001). It is speculated that this may lead to a non-significant direct relationship
between maternal dissociation and daughters’ internalizing symptoms, such as
depressed/withdrawal and immature/bizarre behaviors. The effects of maternal
dissociative symptoms were fully mediated via the enjoyment dimension of
parenting in this data.
This study overcomes several methodological limitations in past studies, by
using multigenerational, longitudinal data, and provides a strong empirical basis for
the protective impact of non-offending mothers’ parenting, according to the existing
body of literature. Further, this study showed that both abuse-specific support, and
overall care — such as maternal parenting practices — had a positive impact on
children’s adaptations. The source of support was also important. It was found that
maternal support had a positive impact, but support from the mothers’
95
family-of-origin sub-network had limited impact for enhancing behavioral
adjustment among sexually abused girls.
6.2 Implications for Social Work Intervention
This data provides compelling evidence that the extent to which mothers
enjoy a child and parental role has a direct and positive impact in decreasing their
abused daughters’ behavior problems. In addition, maternal dissociative symptoms
are an important determinant of this dimension of parenting. This implies that
families where child sexual abuse occurs may benefit from intervention programs
which include some elements for their non-offending mothers.
In line with these findings, some intervention programs for sexually abused
children have incorporated their mothers as an important intervention element in
promoting the children’s psychosocial adaptations (e.g., Celano, Hazzard, Webb &
McCall, 1996; Cohen & Mannarino, 1993; Deblinger & Heflin, 1996). For example,
Deblinger and Heflin’s program (Deblinger & Heflin, 1996) allocates separate
sessions for mothers of sexually abused children, assists them in coping with their
own emotional distress, and engenders communication and management skills with
which to handle their children’s behavioral difficulties. Their outcome research
indicated that the mothers who received cognitive-behavioral intervention reported
greater improvement than the community control group or the child intervention only
group, on indicators such as parenting skills and children’s behavioral problems
(Deblinger, Lippman & Steer, 1996).
96
In addition to these skills training components, the current study suggests that
such elements as mothers’ mental health, and the experiential or attitudinal
dimensions of parenting, may improve outcomes for victims. Professionals working
with this population of mothers may also need to inquire about the mothers’ own
childhood traumas, and to explore how these past experiences may be associated
with current functioning as caregivers. With adequate child advocacy support, such
inquiries may improve outcomes for victims and their families.
Another practical implication for intervention programs suggested by this
study concerns social support from mothers’ family-of-origin sub-networks. Some
intervention programs found that parents’ social support systems were an important
predictor of treatment outcome (Cohen & Mannarino, 1998; Celano, Hazzard, Web
& McCall, 1996). However, the findings of the current study suggest that some
caution should be used when mobilizing informal support from clients’ natural
networks. The availability of support is important, but the quality of relationships
that recipients have with the members of their networks may be an important
consideration for professionals to make when working with these families. With
adequate assessments and, if needed, education of supportive individuals, families
may benefit from such support existing in their natural networks.
6.3 Limitations of the Study and Directions for Future Research
This study utilized a sample from a multigenerational, longitudinal study in
which both families of abuse and demographically-similar comparison families
97
participated. The present study is one of a few studies which compare the maternal
impact on sexually abused girls’ post-abuse adjustment to a demographically-similar
comparison sample, as well as to investigate the impact using multiple informants
and over the long term. This multigenerational, longitudinal approach provided
unusual conditions for examining whether and how maternal factors impact on
adaptation of sexually abused children, which constitute substantial improvements
over past research.
However, several limitations of the findings of this study are noteworthy.
First, there exists a considerable time gap between mothers’ parenting and children’s
self-reported behavioral outcomes. Parenting information was obtained at Time 1
assessment, while children’s behavior outcomes were reported at Time 4, which
corresponds to approximately seven years after the initial Time 1 assessment. On one
hand, empirical support for the long-term impact of maternal parenting is
strengthened by using prospective, longitudinal data. On the other hand, it is also
reasonable to assume that some intervening mechanisms play, as maternal parenting
exerts its impact on children’s behavior outcomes over seven years. Future research
needs to direct its attention to partitioning the direct path between mothers’ parenting
and G
2
behavior outcomes detected in this data.
A similar point could be made for the G
2
children’s contribution. While the
G
1
mothers’ parenting model is relatively well-elaborated in the current conceptual
model, the G
2
children’s contribution to their behavior outcomes is sketchy. Feiring,
Taska and Lewis (1998; 2002) suggest that self-conscious emotions, including shame
98
and attribution, may be an important mechanism linking non-offending mothers’
parenting and their children’s adjustment. In addition, longitudinal studies show that
outcomes in one developmental stage cannot be presumed to generalize to other
periods in children’s lives (Fergusson & Horwood, 2003). Longitudinal investigation
of dynamic processes between maternal factors and children’s behavior adjustment
may have important implications for elucidating how maternal factors impact on
sexually abused girls’ behavior adjustment over the long term.
A caveat should be made when interpreting the function of G
1
mothers’
childhood sexual abuse experiences detected in the current analyses. This study
failed to find significant relationships between the mothers’ own childhood sexual
abuse experiences and other study variables included in the models. However, the
models proposed in the current study were not intended to be exhaustive. Moreover,
the previous study (Kim et al., in press) using the same sample, found that the
mothers’ childhood sexual abuse experiences were related to other psychological
functioning, such as depression and anxiety. The mothers’ childhood sexual abuse
experiences were important in understanding their overall psychosocial functioning.
Continuing efforts need to be made to elucidate the role of mothers’ child sexual
abuse in their own functioning, as well as their children’s functioning among this
population, before any definitive conclusions regarding the role of mothers’
childhood sexual abuse experiences are put forth in this population.
Finally, this study relied on the mothers’ retrospective reports of their own
childhood developmental histories. Although one review suggests that the reliability
99
and validity of autobiographical memory are better than commonly believed (Brewin,
Andrews & Gotlib, 1993), there exists the possibility of bias associated with the
mothers’ current states. In addition, this study stringently defined the daughters’
child sexual abuse status by limiting the abuse characteristics by family perpetrator,
genital contact, and substantiated cases. Therefore, the current findings may not
generalize to the entire abused population.
6.4 Conclusion
An important conclusion generated by this study is that mothers’ parenting
has a positive impact on the long-term behavioral adjustment of sexually abused girls.
The study also found that different causal mechanisms may operate in these
relationships, for abused and comparison girls. Moreover, the results reported here
indicate that important factors in understanding mothers’ parenting practices might
be their own mothers’ punitive discipline strategies during their childhoods, as well
as their current dissociative symptoms. Therefore, child sexual abuse victims may
benefit from intervention programs focused on enhancing mothers’ functioning.
Effective interventions for the mothers might include an element addressing their
childhood upbringing and current mental health challenges, as well as their parenting
practices. Finally, it is increasingly clear that there is a pressing need for more
studies that investigate the precise mechanisms by which maternal factors and
children’s characteristics dynamically relate over time, and lead to long-term
outcomes, if we are to better understand the adjustment patterns of sexually abused
100
girls and strategically utilize the knowledge to boost the likelihood of resilience
among these children.
101
References
Achenbach, T. M. (1991). Manual for the Child Behavior Checklist/ 4-18 and 1991
Profile. Burlington, VT: University of Vermont, Department of Psychiatry.
Achenbach, T. M. (1992). Manual for the Child Behavior Checklist/ 2-3 and 1992
Profile. Burlington, VT: University of Vermont, Department of Psychiatry.
Ainsworth, M. (1973). The development of infant-mother attachment. In B. C. H.
Ricciuti (Ed.), Review of child development research (Vol. III, pp. 1-94).
Chicago: University of Chicago.
Alexander, P. C., Teti, L., & Anderson, C.L. (2000). Childhood sexual abuse history
and role reversal in parenting. Child Abuse & Neglect, 24(6), 829-838.
Anthony, E. J. (1974). Introduction: The syndrome of the psychologically vulnerable
child. In J. A. C. Koupernik (Ed.), The child in his family: Children at
psychiatric risk (Vol. 3, pp. 3-10). New York: Wiley.
Arbuckle, J. L. (2003). Amos 5.0 user's guide. Chicago: Small Waters.
Avery, L., Hutchinson, D., & Whitaker, K. (2002). Domestic violence and
intergenerational rates of child sexual abuse: A case record analysis. Child &
Adolescent Work Journal, 19(1), 77-90.
Bailey, J. A., & McCloskey, L.A. (2005). Pathways to adolescent substance use
among sexually abused girls. Journal of Abnormal Child Psychology, 33(1).
Bandura, A. (1977). Social learning theory. Englewood Cliffs, N.J.: Prentice-Hall.
Banyard, V. L. (1997). The impact of childhood sexual abuse and family functioning
on four dimensions of women's later parenting. Child Abuse & Neglect, 21(11),
1095-1107.
Beitchman, J. H., Zucker, K.J., Hood, J.E., daCosta, G.A., & Akman, D. (1991). A
review of the short-term effects of child sexual abuse. Child Abuse & Neglect,
16, 101-118.
Belsky, J. (1984). The determinants of parenting: A process model. Child
Development, 55, 83-96.
Benjamin, L. R., Benjamin, R., & Rind, B. (1996). Dissociative mothers' subjective
experience of parenting. Child Abuse & Neglect, 20(10), 933-942.
102
Benjamin, L. R., Benjamin, R., & Rind, B. (1998). The parenting experience of
mothers with dissociative disorders. Journal of Marital and Family Therapy,
24(3), 337-354.
Bernstein, E. M., & Putnam, F.W. (1986). Development, reliability, and validity of a
dissociation scale. Journal of Nervous and Mental Disease, 174(12), 727-735.
Black, M., Dubowitz, J., & Harrington, D. (1994). Sexual abuse: developmental
differences in children's behavior and self-perception. Child Abuse & Neglect,
18, 85-95.
Block, J. H. (1981). The child-rearing practices report (CRPR): A set of Q items for
the description for parental socialization attitudes and values. Unpublished
manuscript, University of California, Berkeley.
Bowlby, J. (1969). Attachment and loss (Vol. 1). New York: Basic Books.
Brewin, C. R., Andrews, B., & Gotlib, I.H. (1993). Psychopathology and early
experience: A reappraisal of retrospective reports. Psychological Bulletin,
113(1), 82-98.
Briere, J., & Runtz, M. (1988). Symptomatology associated with childhood
victimization in a nonclinical adult sample. Child Abuse & Neglect, 12, 51-59.
Bronfenbrenner, U. (1977). Toward an experimental ecology of human development.
American Psychologist, 32, 513-531.
Bronfenbrenner, U. (1979). Contexts of child rearing: Problems and prospects.
American Psychologist, 34(10), 844-850.
Browne, A., & Finkelhor, D. (1986). Impact of child sexual abuse: Biological and
psychological perspectives. New York: Plenum Press.
Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K.
A. B. J. S. Long (Ed.), Testing Structural Equation Models (pp. 136-162).
Newbury Park: Sage.
Burkett, L. P. (1991). Parenting behaviors of women who were sexually abused as
children in their families of origin. Family Process, 30, 421-434.
Cammaert, L. P. (1988). Nonoffending mothers: A new conceptualization. In L. E.
Walker (Ed.), Handbook on sexual abuse of children. New York: Springer.
103
Cauce, A., Stewart, A., Rodriguez, M.D., Cochran, B., & Ginzler, J. (2003).
Overcoming the odds? adolescent development in the context of urban poverty.
In S. S. Luthar (Ed.), Resilience and vulnerability: Adaptation in the context of
childhood adversities. New York: Cambridge University Press.
Celano, M., Hazzard, A., Webb, C., & McCall, C. (1996). Treatment of traumagenic
beliefs among sexually abused girls and their mothers: An evaluative study.
Journal of Abnormal Child Psychology, 24(1), 1-17.
Chu, J. A. (1992). The revictimization of adult women with histories of childhood
abuse. Journal of Psychotherapy and Research, 1, 259-269.
Cicchetti, D., & Garmezy, N. (1993). Prospects and promises in the study of
resilience. Development & Psychopathology, 5, 497-502.
Cicchetti, D., & Lynch, M. (1993). Toward an ecological/transactional model of
community violence and child maltreatment: Consequences for children's
development. Psychiatry, 56, 96-118.
Cobb, S. (1976). Social support as a moderator of life stress. Psychodynamic
Medicine, 38, 300-314.
Cochran, M., & Brassard, J. (1979). Child development and personal social networks.
Child Development, 50(3), 601-616.
Cochran, M., & Brassard, J. (1995). Parenting and social networks. In M. H.
Bornstein (Ed.), Handbook of parenting: Status and social conditions of
parenting (Vol. 3). Hillsdale, N.J.: Lawrence Erlbaum Associates.
Cochran, M., & Walker, S.K. (2005). Parenting and personal social networks. In L.
O. T. Luster (Ed.), Parenting: An ecological perspective (2 ed., pp. 235-273).
Mahwah, NJ: Lawrence Erlbaum Associates.
Cohen, T. (1995). Motherhood among incest survivors. Child Abuse & Neglect,
19(12), 1423-1429.
Cole, P., & Putnam, P. (1992). Effects of incest on self and social functioning: A
developmental psychopathology perspective. Journal of Consulting & Clinical
Psychology, 60(2), 174-184.
Collin-Vezina, D. C., Cyr, M., Pauze, R., & McDuff, P. (2005). The role of
depression and dissociation in the link between childhood sexual abuse and later
parental practices. Journal of Trauma & Dissociation, 6(1), 71-97.
104
Conte, J. R., & Schuerman, J.R. (1987). Factors associated with an increased impact
of child sexual abuse. Child Abuse & Neglect, 11(201-211).
Corse, S. J. (1983). Social support networks of abusing and nonabusing parents.
University of Maryland, College Park, MD.
Corse, S. J., Schmid, K., & Trickett, P.K. (1990). Social network characteristics of
mothers in abusing and nonabusing families and their relationships to parenting
beliefs. Journal of Community Psychology, 18, 44-59.
Costello, E. J., & Angold, A. (2000). developmental psychopathology and public
health: Past, present, and future. Development & Psychopathology, 12(4),
599-618.
Crittenden, P. M. (1985). Social networks quality of child rearing and child
development. Child Development, 56(1299-1313).
Crockenberg, S. (1988). Social support and parenting. In B. L. W. T. M. Young (Ed.),
Research on support for parents and infants in the postnatal period (pp. 67-92).
New York: Ablex.
Cross, T. P., De Vos, E., & Whitcomb, D. (1994). Prosecution of child sexual abuse:
Which cases are accepted? Child Abuse & Neglect, 18(8), 661-677.
Davies, A. A., Rhodes, J.E., & Hamilton-Leaks, J. (1997). When both parents may
be a source of support and problems: An analysis of pregnant and parenting
female African American Adolescents' relationships with their mothers and
fathers. Journal of Research on Adolescence, 7(3), 331-348.
Deblinger, E., & Heflin, A.H. (1996). Treating sexually abused children and their
nonoffending parents: A cognitive behavioral approach. Thousand Oaks, CA:
Sage.
Deblinger, E., Hathaway, D.R., & Lippman, J., & Steer, R. (1993). Psychosocial
characteristics and correlates of symptom distress in nonoffending mothers of
sexually abused children. Journal of Interpersonal Violence, 8(2), 155-168.
Deblinger, E., Lippmann, J., & Steer, R. (1996). Sexually abused children suffering
from posttraumatic stress symptoms: Initial treatment outcome findings. Child
Maltreatment, 1(310-321).
Deblinger, E., Steer, R., & Lippmann, J. (1999). Maternal factors associated with
sexually-abused children's psychosocial adjustment. Child Maltreatment, 4,
13-20.
105
Dempster, A. P., Laird, N.M., & Rubin, D.B. (1977). Maximum likelihood
estimation from incomplete data via the EM algorithm (with discussion).
Journal of the Royal Statistical Society, B39, 1-38.
Dilillo, D., Tremblay, G.C., & Peterson, L. (2000). Linking childhood sexual abuse
and abusive parenting: The mediating role of maternal anger. Child Abuse &
Neglect, 24(6), 767-779.
Dinwiddie, S., Heath, C., Dunne, M.P., Bucholz, K.K., Madden, P.A.F., Sluttske,
W.S., Bierut, L.J., Statham, D.B., & Martin, N.G. (2000). Early sexual abuse
and lifetime psychopathology: A co-twin-control study. Psychological Medicine,
30, 41-52.
Downey, G., & Coyne, J.C. (1990). Children of depressed parents: An integrative
review. Psychological Bulletin, 108(1), 50-76.
Dukewich, T. L., Borkowski, J.G., & Whitman, T.L. (1996). Adolescent mothers and
child abuse potential: An evaluation of risk factors. Child Abuse & Neglect,
20(11), 1031-1047.
Egeland, B., Carlson, E., & Sroufe, L.A. (1993). Resilience as process. Development
& Psychopathology, 5, 517-528.
Egeland, B., Jacovitz, D., & Papatola, K. (1987). Intergenerational continuity of
abuse. In J. L. R. Gelles (Ed.), Child abuse and neglect: Biosocial dimension (pp.
255-276). New York: Aldine de Gruytem.
Einbender, A. J., & Friedrich, W.N. (1989). Psychological functioning and behavior
of sexually abused girls. Journal of Consulting & Clinical Psychology, 57(1),
155-157.
Elliott, A. N., & Carnes, C.N. (2001). Reactions of nonoffending parents to the
sexual abuse of their child: A review of the literature. Child Maltreatment, 6(4),
314-331.
Elliott, D. M., & Briere, J. (1994). Forensic sexual abuse evaluations of older
children: Disclosure and symptomatology. Behavioral Sciences and the Law, 12,
261-277.
Elze, D. E., Auslander, W., McMillen, C., Edmond, T., & Thompson, R. (2001).
Untangling the impact of sexual abuse on HIV risk behaviors among youths in
foster care. AIDS Education and Prevention, 13(4), 377-389.
106
English, D. J., Marshall, D.B., Coghlan, L., Sherry, B., & Orme, M. (2002). Causes
and consequences of the substantiation decision in Washington State Child
Protective Services. Children and Youth Services Review, 24(11), 817-851.
Everson, M. D., Hunter, W.M., Runyon, D.K., Edelsohn, G.A., & Coulter, M.L.
(1989). Maternal support following disclosure of incest. American Journal of
Orthopsychiatry59, 59, 197-207.
Faller, K. C. (1989). Sexual abuse: An exploration of the intergenerational
hypothesis. Child Abuse & Neglect, 8, 544-548.
Feiring, C., Taska, L.S., & Lewis, M. (1998). Social support and children's and
adolescents' adaptation to sexual abuse. Journal of Interpersonal Violence, 13,
240-260.
Feiring, C., Taska, L.S., & Lewis, M. (1999). Age and gender differences in
children's and adolescents' adaptation to sexual abuse. Child Abuse & Neglect,
23(2), 115-128.
Feiring, C., Taska, L.S., & Lewis, M. (2002). Adjustment following sexual abuse
discovery: The role of shame and attributional style. Developmental Psychology,
38(1), 79-92.
Fergusson, D. M., & Horwood, L.J. (2003). Resilience to childhood adversity: results
of a 21-year study. In S. S. Luthar (Ed.), Resilience and vulnerability:
Adaptation in the context of childhood adversities (pp. 130155). New York:
Cambridge University Press.
Finkelhor , D., Moore, D., Hamby, S.L., & Straus, M.A. (1997). Sexually abused
children in a national survey of parents: Methodological Issues. Child Abuse &
Neglect, 21(1), 1-9.
Fraser, M. W., Richman, J.M., & Galinsky, M.J. (1999). Risk, protection, and
resilience: Toward a conceptual framework for social work practice. Social
Work Research, 23(3), 131-143.
Friedrich, W. N., Beilke, R.L., & Urquiza, A.J. (1987). Children from sexually
abusive families" A behavioral comparison. Journal of Interpersonal Violence,
2(4), 391-402.
Friedrich, W. N., Urquiza, A., & Beilke, R.L. (1986). Behavior problems in sexually
abused young children. Journal of Pediatric Psychology, 11, 47-57.
107
Fromuth, M. E. (1986). The relationship of childhood sexual abuse with later
psychological and sexual adjustment in a sample of college women. Child Abuse
& Neglect, 10(5-15).
Gale, J., Thompson, R.J., Moran, R., & Sack, W.H. (1988). Sexual abuse in young
children: Its clinical presentation and characteristic patterns. Child Abuse &
Neglect, 12, 163-170.
Garnefski, N., & Diekstra, R.F.W. (1997). Child sexual abuse and emotional and
behavioral problems in adolescence Gender differences. Journal of the
American Academy of Child and Adolescent Psychiatry, 36, 323-329.
Gelinas, D. J. (1983). The persisting negative effects of incest. Psychiatry, 46,
312-332.
Gonzales, N. A., Cauce, A., Friedman, R.J., & Mason, C.A. (1996). Family, peer,
and neighborhood influences on academic achievement among
African-American adolescents: One-year prospective effects. American Journal
of Community Psychology, 24(3), 365-387.
Gorey, K. M., & Leslie, D.R. (1997). The prevalence of child sexual abuse:
Integrative review adjustment for potential response and measurement biases.
Child Abuse & Neglect, 21(4), 391-398.
Gorman-Smith, D., Tolan, P.H., & Henry, D. (1999). The relation of community and
family risk among urban-poor adolescents. In C. S. P. Cohen, & L.N. Robins
(Ed.), Historical and geographical influences on psychopathology (pp. 349-367).
Mahwah, NJ: Lawrence Erlbaum Associates.
Green, A. H., Coupe, P., Fernandez, R., & Stevens, B. (1995). Incest revisited:
Delayed post-traumatic stress disorder in mothers following the sexual abuse of
their children. Child Abuse & Neglect, 19, 1275-1282.
Hanson, R. F., Saunders, B.E., & Lipovsky, J.A. (1992). The relationship between
self-reported levels of distress of parents and victims in incest families. Journal
of Child Sexual Abuse, 1, 49-60.
Hazzard, A., Celano, M., Gould, J., Lawry, S., & Webb, C. (1995). Predicting
symptomatology and self-blame among child sexual abuse victims. Child Abuse
& Neglect, 19, 707-714.
Herman, J. (1992). Trauma and recovery: The aftermath of violence from domestic
violence to political terror. New York: Basic Books.
108
Herman, J., & Hirschman, L. (1981). Families at risk for father-daughter incest.
American Journal of Psychiatry, 138, 967-970.
Hollingshead, A. F. (1975). Four Factor Index of Social Status. Unpublished
manuscript, Yale University.
Horn, J. L. (1965). A rationale and test for the number of factors in factor analysis.
Psychometrika, 30(2), 179-185.
Hunter, W. M., Coultern, M.L., Runyan, D.K., & Everson, M.D. (1990).
Determinants of placement for sexually abused children. Child Abuse & Neglect,
14(3), 407-417.
Jackson, J. L., Calhoun, K.S., Amick, A.E., Maddever, H.M., & Habif, V.L. (1990).
Young adult women who report childhood intrafamilial sexual abuse:
Subsequent adjustment. Archives of Sexual Behavior, 19, 211-221.
Jarrett, R. L. (1995). Growing up poor: The family experiences of socially mobile
youth in low-income African-American neighborhoods. Journal of Adolescence
Research, 10(1), 111-135.
Jarrett, R. L. (1997). Resilience among low-income African American youth: An
ethnographic perspective. Ethos, 25(2), 218-229.
Jarrett, R. L. (1999). Successful parenting in high-risk neighborhoods. The Future of
Children, 9(2), 45-50.
Johnson, B. K., & Kenkel, M.B. (1991). Stress, coping, & adjustment in female
adolescent incest victims. Child Abuse & Neglect, 15, 293-305.
Joyce, P. A. (1997). Mothers of sexually abused children and the concept of
collusion: A literature review. Journal of Child Sexual Abuse, 6, 75-92.
Jumper, S. A. (1995). A meta-analysis of the relationship of child sexual abuse to
adult psychological adjustment. Child Abuse & Neglect, 19(6), 715-728.
Kauman, J., & Zigler, E. (1987). Do abused children become abusive parents?
American Journal of Community Psychology, 57, 186-192.
Kendall-Tackett, K. A., Williams, L.M., & Finkelhor, D. (1993). Impact of sexual
abuse on children: A review and synthesis of recent empirical studies.
Psychologicla Bulletin, 113, 164-180.
109
Kendler, K., Bulik, C., Silberg, J., Hettema, J., Myers, J., & Prescott, C. (2000).
Childhood sexual abuse and adult psychiatric and substance abuse disorders in
women. Archives of General Psychiatry, 57, 953-959.
Kim, K., Chung, I.C., & Trickett, P.K. (in preparation). Social network and social
support characteristics among mothers with childhood sexual abuse trauma:
Findings based on factor analytic approach.
Kim, K., Noll, J.G., Putnam, F.W., & Trickett, P.K. (in press). Psychosocial
characteristics of non-offending mothers of sexually abused girls: Findings from
a prospective, multigenerational study. Child Maltreatment.
Kluft, R. P. (1987). The parental fitness of mothers with multiple personality
disorder: A preliminary study. Child Abuse & Neglect, 2, 273-280.
Leifer, M., Kilbane, T., & Grossman, G. (2001). A three-generational study of
comparing the families of supportive and unsupportive mothers of sexually
abused children. Child Maltreatment, 6(4), 353-364.
Leifer, M., Kilbane, T., & Kalick, S. (2004). Vulnerability or resilience to
intergenerational sexual abuse: The role of maternal factors. Child Maltreatment,
9(1), 78-91.
Leifer, M., Kilbane, T., Jacobsen, T., & Grossman, G. (2004). A three generational
study of transmission of risk for sexual abuse. Journal of Clinical Child and
Adolescent Psychology, 33(4), 662-672.
Leifer, M., Shapiro, J.P., & Kassem, L. (1993). The impact of maternal history and
behavior upon foster placement and adjustment in sexually abused girls. Child
Abuse & Neglect, 17(6), 755-766.
Lepper, M. R. (1982). Social control processes, attributions of motivation, and the
internalization of social values. In D. N. R. E. T. Higgins, & W. W. Hartup
(Ed.), Social cognition and social behavior: Developmental perspectives.
Cambridge, Eng.: Cambridge University Press.
Lewin, L. C., & Bergin, C. (2001). Attachment behaviors, depression, and anxiety in
nonoffending mothers of child sexual abuse victims. 6, 4, 365-375.
Lovett, B. B. (1995). Child sexual abuse: The female victim's relationship with her
nonoffending mother. Child Abuse & Neglect, 19, 729-738.
Lustig, N., Dresser, J., Spellman, S., & Murray, T. (1966). Incest: A family group
survival pattern. Archives of General Psychiatry, 14, 31-40.
110
Luthar, S. S. (1993). Annotation: Methodological and conceptual issues in research
on childhood resilience. Journal of Child Psychology and Psychiatry, 34(4),
441-453.
Luthar, S. S., Cicchetti, D., & Becker, B. (2000). The construct of resilience: A
critical evaluation and guidelines for future work. Child Development, 71(3),
543-562.
Lynskey, M., & Fergusson, D. (1997). Factors protecting against the development of
adjustment difficulties in young adults exposed to childhood sexual abuse. Child
Abuse & Neglect, 12, 1177-1190.
Lyons, S. J., Henly, J.R., & Schuerman, J.R. (2005). Informal support in matreating
families: Its effect on parenting practices. Children and Youth Services Review,
27(1), 21-38.
Lyons-Ruth, K., & Block, D. (1996). The disturbed caregiving system: Relations
among childhood trauma, maternal caregiving, and infant affect and attachment.
Infant Mental Health Journal, 17(3), 257-275.
Maccoby, E. E., & Martin, J.A. (1983). Socialization in the context of the family:
Parent-child interaction. In E. M. Hetherington (Ed.), Handbook of child
psychology Vol.4: socialization, personality, and social development. New
York: Wiley.
Macfie, J., Cicchetti, D., & Toth, S.L. (2001). The development of dissociation in
maltreated preschool-aged children. Development and Psychopathology, 13(2),
233-254.
Mannarino, A. P., & Cohen, J.A. (1996). Family-related variables and psychological
symptom formation in sexually abused girls. Journal of Child Sexual Abuse, 5,
105-120.
Mannarino, A. P., & Cohen, J.A. (1996). A follow-up study of factors that mediate
the development of psychological symptomatology in sexually abused girls.
Child Maltreatment, 1(246-260).
Mannarino, A. P., Cohen, J.A., & Gregor, M. (1989). Emotional and behavioral
difficulties in sexually abused girls. Journal of Interpersonal Violence, 4(4),
437-451.
Mannarino, A. P., Cohen, J.A., Smith, J.A., & Moore-Motily, S.M. (1991). Six- and
twelve-month follow-up of sexually abused girls. Journal of Interpersonal
Violence, 6(4), 494-511.
111
Marcenko, M. O., Kemp. S.P., & Larson, N.C. (2000). Childhood experiences of
abuse, later substance use, and parenting outcomes among low-income mothers.
American Journal of Orthopsychiatry, 70(3), 316-325.
Mason, C. A., Cauce, A., Gonzales, N., & Hiraga, Y. (1994). Adolescent problem
behavior: The effect of peers and the moderating role of father absence and the
mother-child relationship. American Journal of Community Psychology, 22(6),
723-743.
Masten, A., & Coatsworth, J.D. (1998). The development of competence in
favorable and unfavorable environments: Lessons from research on successful
children. American Psychologist, 53, 205-220.
Masten, A., Best, K., & Garmezy, N. (1990). Resilience and development:
Contributions from the study of children who overcome adversity. Development
and Psychopathology, 2, 425-444.
Masten, A. S. (1994). Resilience in individual development: Successful adaptation
despite risk and adversity. In E. W. G. M.C. Wang (Ed.), Educational resilience
in inner-city America: Challenges and prospects (pp. 3-25). Hillsdale, NJ,
England: Lawrence Erlbaum Associates.
Masten, A. S. (1999). Resilience comes of age: Reflections on the past and outlook
for the next generation of research. In J. J. M.D. Glantz (Ed.), Resilience and
development: Positive life adaptations (pp. 281-296). New York: Plenum.
Masten, A. S. (2001). Ordinary magic: Resilience processes in development.
American Psychologist, 56, 227-238.
Masten, A. S., & Powell, J.L. (2003). A resilience framework for research, policy,
and practice. In S. S. Luthar (Ed.), Resilience and vulnerability: Adaptation in
the context of childhood adversities (pp. 1-25). New York: Cambridge
University Press.
McCloskey, L. A., & Bailey, J.A. (2000). The intergenerational transmission of risk
for child sexual abuse. Journal of Interpersonal Violence, 15(10), 1019-1035.
Meyers, S. A., & Battistoni, J. (2003). Proximal and distal correlates of adolescent
mothers' parenting attitudes. Applied Developmental Psychology, 24, 33-49.
Miliotis, D., Sesma, A., & Masten, A.S. (1999). Parenting as a protective process for
school success in children from homeless families. Early Education &
Development, 10(2), 112-133.
112
Morrison, N. C., & Clavenna-Valleroy, J. (1998). Perceptions of maternal support as
related to self-concept and self-report of depression in sexually abused female
adolescents. Journal of Child Sexual Abuse, 7, 23-40.
Nakhle Tamraz, D. (1996). Nonoffending mothers of sexually abused children:
Comparison of opinions and Research. Journal of Child Sexual Abuse, 5(4),
75-104.
Narang, D. S., & Conteras, J.M. (2000). Dissociation as a mediator between child
abuse history and adult abuse potential. Child Abuse & Neglect, 24(5), 653-665.
Nash, L., & Morrison, W.L. (1984). Manual for the Parenting Skills Inventory.
Seattle, WA: Parenting Press.
Nelson, E., Heath, A., Madden, P., Cooper, M.L., Dinwiddie, S.H., Bucholz, K.K.,
Glowinski, A., McLaughlin, T., Dunne, M.P., Statham, D.J., & Martin, N.G.
(2002). Association between self-reported childhood sexual abuse and adverse
psychosocial outcomes: Results from a twin study. Archives of General
Psychiatry, 59, 139-146.
Newberger, C. M., Gremy, I.M., Waternaux, C.M., & Newberger, E.H. (1993).
Mothers of sexually abused children: Trauma and repair in longitudinal
perspective. American Journal of Orthopsychiatry, 63(1), 92-102.
Noll, J. G. (2005). Does childhood sexual abuse set in motion a cycle of violence
against women?: What we know and what we need to learn. Journal of
Interpersonal Violence, 20(4), 455-462.
Noll, J. G., Horowitz, L. A., Bonanno, G. A. Trickett, P. K. & Putnam, F. W. (2003).
Revictimization and self-harm in females who experienced childhood sexual
abuse. Journal of Interpersonal Violence, 18(12), 1452-1471.
Noll, J. G., Trickett, P.K., Horn, J.L., Long, J., & Putnam, F.W. (in preparation). A
mega-factoring of CBCL items for maltreated children.
Oates, R. K., Tennutt, J., Swanston, H., Lynch, D.L., & O'Toole, B.I. (1998). Prior
childhood sexual abuse in mothers of sexually abused children. Child Abuse &
Neglect, 22(11), 1113-1123.
Pintello, D., & Zuravin, S. (2001). Intrafamilial child sexual abuse: Predictors of
postdisclosure maternal belief and protective action. Child Maltreatment, 6(4),
344-352.
113
Polansky, N. A., Ammons, P.W., & Gaudin, J.M., Jr. (1985). Loneliness and
isolation in child neglect. Social Casework, 66, 38-47.
Raykov, T. (1997). Equivalent structural equation models and group equality
constraints. Multivariate Behavioral Research, 32(2), 95-104.
Richardson, R. A., Barbour, N.B., & Bubenzer, D.L. (1991). Bittersweet
connections: informal social networks as sources of support and interference for
adolescent mothers. Family Relations, 40, 430-434.
Roberts, R., O'Connor, T., Dunn, J., & Golding, J. (2004). The effects of child sexual
abuse in later family life: Mental health, parenting and adjustment of offsprign.
Child Abuse & Neglect, 28, 525-545.
Robinson, C. C., Mandleco, B., Olsen, S.F., & Hart, C.H. (1995). Authoritative,
authoritarian, and permissive parenting practices: Development of a new
measure. Psychological Reports, 77, 819-830.
Roesler, T. A. (1994). Reactions to disclosure of childhood sexual abuse: The effect
on adult symptoms. Journal of Nervous and Mental Disease, 182, 618-624.
Ross, C. A. (1999). Dissociative disorders. In T. Millon, Blaney, P.H., & Roger, D.
(Ed.), Oxford textbook of psychopathology (Vol. 4). New York, NY: Oxford
University Press.
Runtz, M., & Briere, J. (1986). Adolescent "acting-out" and childhood history of
sexual abuse. Journal of Interpersonal Violence, 1(3), 326-334.
Ruscio, A. M. (2001). Predicting the childrearing practices of mothers sexually
abused in childhood. Child Abuse & Neglect, 25, 369-387.
Russell, D. E. H. (1986). The secret trauma: Incest in the lives of girls and women.
New York: Basic Books.
Rutter, M. (1987). Psychosocial resilience and protective mechanisms. American
Journal of Orthopsychiatry, 57(3), 316-331.
Rutter, M. (1990). Psychosocial resilience and protective mechanisms. In A. S. M. J.
Rolf, D. Cicchetti, K.H. Neuchterlein, & S. Weintraub (Ed.), Risk and protective
factors in the development of psychopathology (pp. 181-214). New York:
Cambridge.
114
Sameroff, A., & Chandler, M.J. (1975). Reproductive risk and the continuum of
caretaking casualty. In M. H. F.D. Horowitz, S. Scarr-Salapatek, & G. Siegel
(Ed.), Review of child development research. Chicago: University of Chicago
Press.
Sameroff, A., Seifer, R., Barocas, R., Zax, M., & Greenspan, A. (1987). Intelligence
quotient scores of 4-year-old children: Social environmental risk factors.
Pediatrics, 79, 343-350.
Scaramella, L. V., & Conger, R.D. (2003). Intergenerational continuity of hostile
parenting and its consequences: The moderating influence of children's negative
emotional reactivity. Social Development, 12(3), 420-439.
Schafer, J. L. (1999). NORM: Multiple imputation of incomplete multivariate data
under a normal model (Version 2.0), Computer software available from
http://www.stat.psu.edu/~jls/misoftwa.html.
Shah, R. Z., Dali, P.W., & Heinrichs, T. (1995). Familial influences upon the
occurrence of childhood sexual abuse. Journal of Child Sexual Abuse, 4(4),
45-61.
Shapiro, J. R., & Mangelsdorf, S.C. (1994). The determinants of parenting
competence in adolescent mothers. Journal of Youth and Adolescence, 23(6),
621-641.
Simons, R. L., Whitebeck, L.B., Conger, R.D., & Conger, K.J. (1991). Parenting
factors, social skills, and value commitments as precursors to school failure,
involvement with deviant peers, and delinquent behavior. Journal of Youth and
Adolescence, 20(6), 645-664.
Spaccarelli, S., & Kim, S. (1995). Resilience criteria and factors associated with
resilience in sexually abused girls. Child Abuse & Neglect, 19, 1171-1182.
Spieker, S. J., Bensley, L., McMahon, R.J., & Fung, H. (1996). Sexual abuse as a
factor in child maltreatment by adolescent mothers of preschool aged children.
Developmental and Psychopathology, 8(3), 497-509.
Sroufe, L. A., Jacobvitz, D., Mangelandorf, S., DeAngelo, E., & Ward, M.J. (1995).
Generational boundary dissolution between mothers and their preschool
children: A relationship systems approach. Child Development, 56, 317-325.
Swanston, H. Y., Parkinson, P.N., O'Toole, B.I., Plunkett, A.M., Shrimpton, S., &
Oates, R.K. (2003). Juvenile crime, aggression and delinquency after sexual
abuse. British Journal of Criminology, 43(4), 729-749.
115
Tarter, R. E., & Vanyukov, M. (1999). Re-visiting the validity of the construct of
resilience. In M. D. G. J. L. Johnson (Ed.), Resilience and development: Positive
life adaptations (pp. 85-100). Dordrecht, Netherlands: Kluwer Academic
Publishers.
Taylor, R. D., & Roberts, D. (1995). Kinship support and maternal and adolescent
well-being in economically disadvantaged African-American families. Child
Development, 66(6), 1585-1597.
Thornberry, T. P., Freeman-Gallant, A., Lizotte, A.J., Krohn, M.D., & Smith, C.A.
(2003). Linked lives: The intergenerational transmission of antisocial behavior.
Journal of Abnormal Child Psychology, 31(2), 171-184.
Timmons-Mitchell, J., Chandler-Holtz, D., & Semple, W.E. (1997). Post-traumatic
stress disorder symptoms in child sexual abuse victims and their mothers.
Journal of Child Sexual Abuse, 6(4), 1-14.
Tremblay, C., Herbert, M., & Piche, C. (1999). Coping strategies and social support
as mediators of consequences in child sexual abuse victims. Child Abuse &
Neglect, 23, 929-945.
Trickett, P. K., & Everett, B. (1988). Mothers' Developmental History Questionnaire.
Unpublished manuscript, Bethesda, MD.
Trickett, P. K., & Gordis, E.B. (2004). Aggression and antisocial behavior in
sexually abused females. In B. M. Putallaz, K.L. (Ed.), Aggression, antisocial
behavior, and violence among girls: A development perspective (pp. 162-185).
New York: Guilford Publications.
Trickett, P. K., & Putnam, F.W. (1987). The Psychobiological effects of sexual abuse.
Bethesda, MD: DHHS, National Institute of Mental Health, Divison of
Intramural Research Programs.
Trickett, P. K., & Putnam, F.W. (1998). The developmental impact of sexual abuse.
In P. K. T. C. Schellenbach (Ed.), Violence against children in the family and
the community. Washington, D.C.: APA Books.
Trickett, P. K., & Susman, E.J. (1988). Parental perceptions of child-rearing
practices in physically abusive and nonabusive families. Developmental
Psychology, 24(2), 270-276.
116
Trickett, P. K., & Susman, E.J. (1989). Perceived similarities and disagreements
about childrearing practices in abusive and nonabusive families:
Intergenerational and concurrent family process. In V. G. D. Cichetti (Ed.),
Child Maltreatment (pp. 280-301). New York: Cambridge University Press.
Trickett, P. K., Noll, J., Reiffman, A., & Putnam, F. (2001). Variants of intrafamilial
sexual abuse experiences: Implications for short- and long-term development.
Development & Psychopathology, 13, 1001-1019.
U.S. Department of Health and Human Services, A. o. C., Youth, and Families.
(2007). Child Maltreatment 2005. Washington DC: U.S. Government Printing
Office.
Unger, D. G., & Wandersman, L.P. (1988). The relation of family and partner
support to the adjustment of adolescent mothers. Child Development, 59(3),
1056-1060.
Voight, J. D., Hans, S.L., & Bernstein, V.J. (1996). Support networks of adolescent
mothers: Effects on parenting experience and behavior. Infant Mental Health
Journal, 17(1), 58-73.
Walrath, C., Ybarra, M., Hilden, E.W., Liao, Q., Santiago, R., & Leaf, P. (2003).
Children with reported histories of sexual abuse: Utilizing multiple perspective
to understand clinical and psychosocial profiles. Child Abuse & Neglect, 27,
509-524.
Wattenberg, E. (1985). Ina different light: A feminist perspective on the role of
mothers in father-daughter incest. Child Welfare, 64, 203-211.
Whiteback, L. B., Hoyt, D.R., Simons, R.L., Conger, R.D., Elder Jr., G.H., Lorenz,
F.O., & Huck, S. (1992). Intergenerational continuity of parental rejection and
depressed affect. Journal of Personality and Psychology, 63(6), 1036-1045.
Widom, C. S., & Ames, M.A. (1994). Criminal consequences of childhood sexual
victimization. Child Abuse & Neglect, 18(4), 303-318.
Wyatt, G. E., & Mickey, M.R. (1987). Ameliorating the effects of child sexual abuse.
Journal of Interpersonal Violence, 2, 403-414.
Zingraff, M. T., Leiter, J., Myers, K.A., & Johnsen, M.C. (1993). Child maltreatment
and youthful problem behavior. Criminology, 31(2), 173-102.
Zuelzer, M. B., & Reposa, R.E. (1983). Mothers in incestuous families. International
Journal of Family Therapy, 5(2), 98-110.
117
Zuravin, S., & Fontanella, C. (1999). Parenting behavior and perceived parenting
competence of child sexual abuse survivors. Child Abuse & Neglect, 23,
623-632.
Zuravin, S., McMillen, C., & DePanfils, D., & Risley-Curtis, C. (1996). The
intergenerational cycle of child maltreatment: Continuity versus discontinuity.
Journal of Interpersonal Violence, 11(3), 315-334.
118
APPENDICES
Appendix A: Distribution of the Major Dependent Variables
1. G
1
Mothers’ Parenting - Enjoyment of a Child and Parental Role
Enjo ymen t of a Ch ild and Parental R ole
87.5
85. 0
82.5
80.0
77.5
75.0
72. 5
70.0
67.5
65.0
62.5
60. 0
57.5
55.0
52.5
50.0
T otal Sample
Frequency
30
20
10
0
Enjoyment
87.5
85.0
82.5
80.0
77.5
75.0
72.5
70.0
67.5
65.0
62.5
60.0
57.5
55.0
52.5
50.0
Ab use Sample
12
10
8
6
4
2
0
Enjoyment
90.0 85.0 80.0 75.0 70.0 65.0 60.0 55.0 50.0
Compar ison S ample
30
20
10
0
Figure 10. Distribution of G
1
mothers’ parenting – Enjoyment of a child and
parental role
119
2. G
1
Mothers’ Parenting - Authoritarian Control
Authoritarian Control
50.0
47.5
45.0
42.5
40.0
37.5
35.0
32.5
30.0
27. 5
25. 0
22.5
20.0
17.5
Total Sample
Frequency
30
20
10
0
Authritarian Control
50.0
47.5
45.0
42.5
40.0
37.5
35.0
32.5
30.0
27.5
25.0
22.5
20.0
17.5
Abuse Sample
12
10
8
6
4
2
0
Authoritarian Control
50. 0
47.5
45.0
42.5
40.0
37.5
35.0
32.5
30.0
27.5
25.0
22.5
20.0
Comparison Sample
10
8
6
4
2
0
Figure 11. Distribution of G
1
mothers’ parenting – Authoritarian control
120
3. G
2
Daughters’ Behavioral Adjustment – Delinquent Misbehaviors
Delinquent Misbehaviors
14.0 12.0 10.0 8.0 6.0 4.0 2.0 0.0
Total Sample
Frequency
50
40
30
20
10
0
Delinquent M.
14.0 12.0 10.0 8.0 6. 0 4. 0 2.0 0.0
Abuse Sample
30
20
10
0
Delinquent M.
14.0 12.0 10.0 8.0 6.0 4.0 2.0
Comparison Sample
30
20
10
0
Figure 12. Distribution of G
2
daughters’ behavioral adjustment – Delinquent
misbehaviors
121
4. G
2
Daughters’ Behavioral Adjustment – Aggressive Behaviors
Aggressive Behaviors
30.0
27.5
25. 0
22. 5
20. 0
17. 5
15.0
12.5
10.0
7.5
5.0
2.5
0.0
Total Sample
Frequency
30
20
10
0
Aggression
30.0
27.5
25.0
22.5
20.0
17.5
15.0
12. 5
10.0
7.5
5.0
2.5
0.0
Abuse Sample
20
10
0
Aggression
30.0 25.0 20.0 15.0 10.0 5.0 0.0
Comparison Sample
30
20
10
0
Figure 13. Distribution of G
2
daughters’ behavioral adjustment – Aggressive
behaviors
122
5. G
2
Daughters’ Behavioral Adjustment – Withdrawal/Depressed Behaviors
Withdrawal/Depressed Behaviors
45.0 40.0 35.0 30.0 25. 0 20. 0 15. 0 10. 0 5.0 0.0
Total Sample
Frequency
30
20
10
0
W/D
45.0 40.0 35.0 30.0 25.0 20.0 15.0 10.0 5.0
Abuse Sample
20
10
0
W/D
45.0 40.0 35.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0
Comparison Sample
14
12
10
8
6
4
2
0
Figure 14. Distribution of G
2
daughters’ behavioral adjustment –
Withdrawal/depressed behaviors
123
6. G
2
Daughters’ Behavioral Adjustment – Immature/Bizarre Behaviors
Immature/Bizarre Behaviors
50.0 45.0 40.0 35.0 30.0 25. 0 20.0 15. 0 10.0 5.0
Total Sample
Frequency
40
30
20
10
0
I/B
50.0 45.0 40.0 35.0 30.0 25.0 20.0 15.0 10.0 5.0
Abuse Sample
30
20
10
0
I/B
45. 0 40. 0 35.0 30.0 25.0 20.0 15.0 10.0
Comparison Sample
16
14
12
10
8
6
4
2
0
Figure 15. Distribution of G
2
daughters’ behavioral adjustment – Immature/Bizarre
behaviors
124
Appendix B: Mother’s Social Support Questionnaire (MSSQ)
ID #: _______________
I can go to this person for
advice
This person can come to
me for advice
I can talk to this person
about personal concerns
This person talks to me
about personal concerns
I borrow money or
material goods from this
person
This person borrows
money or goods from me
I can count on help from
this person in an
emergency
This person can count on
help from me in an
emergency
This person and I get
together to have fun or go
out
I ask this person for
advice about childrearing
This person asks me for
advice about childrearing
I talk to this person about
problems with children
Name Relationship Length Frequency 1
2
3
4
5
6
7
8
9
10
11
12
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
125
This person talks to me about
problems with their children
I tell this person happy news
about my children
This person tells me happy news
about their children
I get help from this person such
as babysitting, picking up the
kids, etc.
This person gets help from me as
described above
I talk to this person when I am
upset and just need to talk
This person comes to me when
they are upset and just need to
talk
I talk to this person when I am
upset with my children and need
to let off steam
This person talks to me when they
are upset with their children and
need to let off steam
When I go to this person for help
or support I usually find it
When this person comes to me for
help or support this person
usually finds it
Name Relationship Length Frequency 13
14
15
16
17
18
19
20
21
22
23
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Abstract (if available)
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Observed and self-reported childrearing in mothers with a history of sexual abuse
PDF
The impact of child maltreatment on the mental health of adolescents: a longitudinal study of social anxiety and self-perception
PDF
Untangling the developmental relations between depression and externalizing behavior among maltreated adolescents
PDF
Examining the longitudinal relationships between community violence exposure and aggressive behavior among a sample of maltreated and non-maltreated adolescents
PDF
A longitudinal study of job resources, burnout, and turnover among social workers: implications for evidence-based organizational practices
PDF
Dynamic analyses of the interrelationship between mothers and daughters on a measure of depressive symptoms
PDF
The link between maternal depression and adolescent daughters' risk behavior: the mediating and moderating role of family
PDF
The influence of a humorous intervention on parent-child sexual communication
PDF
Transformational leadership theory aligned-practices and social workers' well-being: an exploratory study of leadership practices in the context of stress and job burnout
Asset Metadata
Creator
Kim, Kihyun
(author)
Core Title
Parenting practices among non-offending mothers of sexually abused girls and its impact on the abused girls' behavioral adjustment: perspectives from a multigenerational, longitudinal study
School
School of Social Work
Degree
Doctor of Philosophy
Degree Program
Social Work
Publication Date
08/07/2009
Defense Date
06/21/2007
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
child sexual abuse,non-offending mothers,OAI-PMH Harvest,post-abuse adjustment of sexually abused children,resilience,trauma and parenting practices
Language
English
Advisor
Trickett, Penelope K. (
committee chair
), McArdle, John J. (
committee member
), Mennen, Ferol E. (
committee member
)
Creator Email
kihyunki@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-m730
Unique identifier
UC1211354
Identifier
etd-Kim-20070807 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-535174 (legacy record id),usctheses-m730 (legacy record id)
Legacy Identifier
etd-Kim-20070807.pdf
Dmrecord
535174
Document Type
Dissertation
Rights
Kim, Kihyun
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Repository Name
Libraries, University of Southern California
Repository Location
Los Angeles, California
Repository Email
cisadmin@lib.usc.edu
Tags
child sexual abuse
non-offending mothers
post-abuse adjustment of sexually abused children
resilience
trauma and parenting practices