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
/
Attachment style, interpersonal guilt, parental alcoholism, parental divorce and eating disordered symptomatology in college women
(USC Thesis Other)
Attachment style, interpersonal guilt, parental alcoholism, parental divorce and eating disordered symptomatology in college women
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI films
the text directly from the original or copy submitted. Thus, some thesis and
dissertation copies are in typewriter face, while others may be from any type of
computer printer.
The quality of this reproduction is dependent upon the quality of the
copy submitted. Broken or indistinct print, colored or poor quality illustrations
and photographs, print bleedthrough, substandard margins, and improper
alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript
and there are missing pages, these will be noted. Also, if unauthorized
copyright material had to be removed, a note w ill indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and continuing
from left to right in equal sections with small overlaps.
ProQuest Information and Learning
300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA
800-521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ATTACHMENT STYLE, INTERPERSONAL GUILT, PARENTAL
ALCOHOLISM, PARENTAL DIVORCE AND EATING DISORDERED
SYMPTOMATOLOGY IN COLLEGE WOMEN
by
Gretchen Roberta Reichardt
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
EDUCATION (COUNSELING PSYCHOLOGY)
December 2002
Copyright 2002 Gretchen Roberta Reichardt
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 3093974
_ _ ®
UMI
UMI Microform 3093974
Copyright 2003 by ProQuest Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UNIVERSITY OF SOUTHERN CALIFORNIA
THE GRADUATE SCHOOL
UNIVERSITY PARK
LOS ANGELES, CALIFORNIA 90089-1695
This dissertation, written by
Gretchen Roberta Reichardt
under the direction o f h e r dissertation committee, and
approved by all its members, has been presented to and
accepted by the Director o f Graduate and Professional
Programs, in partial fulfillment of the requirements for the
degree of
DOCTOR OF PHILOSOPHY
Director
Date Decem ber 18, 2002
^Dissertation Committee
Chair
^ -------------------
C --------------
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ii
DEDICATION
This manuscript is dedicated to my parents and my husband. Thank you for
your support and patience.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Ill
TABLE OF CONTENTS
Page
DEDICATION i i
LIST OF TABLES IV
LIST OF FIGURES v
ABSTRACT VI
Chapter
1. INTRODUCTION 1
Need for the Study
Purpose of the Study
Overview of the Study
Theoretical Framework
Definition of Terms
Review of the Literature
Research Hypotheses
2. METHODOLOGY AND PROCEDURES.................................... 31
Research Design
Procedures
Data Analysis
3. RESULTS...................................................................................... 48
Correlations Between the Variables
Measurement Models
Path Models
4. DISCUSSION................................................................................. 64
Summary of the Study
Limitations
Future Directions
BIBLIOGRAPHY 76
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
iv
LIST OF TABLES
Tables Page
1. Description of the Women in the Study....................................... 35
2. Means, Standard Deviations, and Intercorrelations
Among the Variables...........................................................
3. Summary of Fit Indexes for Models 1-4..................................... 51
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
V
LIST OF FIGURES
Figures Page
1. Original proposed model depicting the hypothesized
relationships among the variables....................................... 32
2. Model 1— Measurement model depicting first-order
latent constructs and path model........................................ 52
3. Model 2— Measurement model depicting first-order
latent constructs................................................................... 53
4. Model 2— Path model................................................................... 54
5. RMSEA model comparison........................................................... 55
6. Model 3— Measurement model depicting first-order
latent constructs................................................................... 59
7. Model 3— Path model................................................................... 60
8. Model A — Measurement model depicting first-order
latent constructs................................................................... 61
9. Model A — Path Model................................................................... 62
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
vi
ABSTRACT
This study used structural equation modeling (SEM) to test several
competing models describing relationships among family variables (parental
divorce and separation, parental alcohol abuse), interpersonal guilt,
attachment style, and eating disordered symptomatology in a female college
population. In the model with the best fit, Parental Divorce and Parental
Alcoholism were not directly associated with Eating Disorder
Symptomatology, but indirectly influenced Eating Disorder Symptomatology
through Attachment. Interpersonal Guilt made no contribution to this model
and appeared to confound the effects of Attachment. However, correlational
analysis indicated that Interpersonal Guilt had a positive relationship with
both Attachment and Eating Disorder Symptomatology. These data
emphasize the complexity of disordered eating and provide future directions
for the prevention and treatment of the continuum of disordered eating.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1
CHAPTER I
INTRODUCTION
Research reveals a disturbing rise in the incidence of eating disorder
symptomatology. These symptoms appear to be particularly prevalent
among young college women. As many as 20% of college women report
that they engage in these behaviors (e.g., Halmi, Falk, & Schwartz, 1981;
Striegel-Moore, Silberstein, Frensch, & Rodin, 1989). In fact, rates may be
even higher. For example, Mintz and Betz (1988) found that as many as
64% of college women display some form of eating disordered behavior.
Unfortunately, the mental health professions have not been especially
successful at treating these disorders (Gremillion, 1992; Mahon, Bradley,
Harvey, Winston, & Palmer, 2001; Richards et al., 2000). This is particularly
unfortunate in light of the profound emotional and physical costs of these
disorders, which have the highest mortality of any psychiatric disorder
(Neumaerker, 2000; Vitiello & Lederhendler, 2000;). Many with these
disorders develop chronic health problems (American Psychiatric
Association, 1994), depression (J. F. Schumaker, Warren, Carr, Schreiber&
Jackson, 1995; Willcox & Sattler, 1996) and suicidality (Pook, Conti, &
Lester, 1996). Moreover, it appears that eating disordered behaviors and
symptoms occur on a continuum so that sub-clinical forms of the disorders
may not only cause considerable emotional distress, but also predict later,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
I
2
clinical eating disorders (e.g., Dancyger & Garfinkel, 1995; Mintz & Betz,
1988; Stice, Ziemba, Margolis, & Flick, 1996; Tylka & Subich, 1999).
Part of the difficulty treating these symptoms and behaviors may be
due to their varied and complex etiology and the consequent difficulty
identifying the most critical variables to address in treatment (Killen et al.,
1996). Eating disorders appear to be multi-determined, with interplay of
social, psychological, interpersonal, and biological factors (Richards et al.,
2000; Sands, Tricker, Sherman, Armatas, & Maschette, 1997). Early
theorizing about eating disorders focused on problematic family functioning
and relationships (Bruch, 1978; Minuchin, 1974; Minuchin, Rosman, & Baker,
1978). More recent researchers have emphasized socio-cultural factors (i.e.,
Murray, Touyz, & Beumont, 1996; Tiggemann & Pickering, 1996) and their
relationship to development of eating disorders, with eating disorders often
being viewed as “culturally bound” within western society (Banks, 1992; Lee,
1995). However, the most current research points to development of eating
disorders in nonwestem cultures, and identifies eating disorders as cutting
across cultural, social, economic, racial, and gendered lines (Cuijpers,
Langendoen, & Bijl, 1999; Mullholland & Mintz, 2001; Rieger, Touyz, Swain &
Beumont, 2001). Therefore, it is important to explore some of the other
variables that may be associated with development of sub-clinical and clinical
eating disorders. Such explorations ultimately may be useful to mental
health professionals in terms of prevention and treatment.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3
The relationship between insecure attachment style and eating
disorders seems well established (see O’Keamey, 1996; Ward, Ramsay, &
Treasure, 2000). Relatively little is known, however, about cognitive-affective
mechanisms and experiences that may be involved in the internalization and
expression of attachment style and that are potentially related to the
development of these symptoms. Research that identifies and examines
these relationships is crucial in furthering our understanding of these
problems and aiding in development of more effective treatment plans to
address internal experiences of afflicted individuals.
The purpose of this study was to present, describe, and empirically
test a model of the etiology of eating disorders among college students.
Specifically the model hypothesized a particular process (Attachment) and a
particular affective experience (Interpersonal Guilt) whereby eating
disordered mentality and pathology may be transmitted to the vulnerable
individual.
Need for the Study
To explore these hypothesized relationships between family-of-origin
variables, attachment style, interpersonal guilt, and eating disordered
symptomatology within a young adult population may advance the
understanding of the mechanisms through which childhood experiences and
perception of childhood familial relationships may affect later psychosocial
functioning. This may have implications for interventions by providing
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4
practitioners with a structure for understanding the eating disordered
individual.
The following sections of this document will advance a theoretical,
empirical, and clinical rationale for inclusion of specific variables in the
current model to be tested. It will also delineate family variables that are
theoretically or empirically related to attachment style, interpersonal guilt,
and/or eating disordered behaviors.
Purpose of the Study
Researchers have identified several correlates of disordered eating
behavior, including attachment, alcoholic family systems, parental divorce,
and interpersonal guilt. Most, however, have utilized univariate statistical
methods to explore these relationships. This approach does not allow for
potential between-variable associations. Indeed, in a current literature
review of risk factors for disordered eating, Mussell, Binford, and Fulkerson
(2000) advocated use of multivariate procedures to further our
comprehension of eating disordered symptomatology, antecedents,
correlates, and consequences.
Structural equation modeling (SEM) was used in this study to test a
model of variables that have shown independent connection with eating
disordered behaviors in prior research. This methodology is especially useful
for testing theoretical conceptualizations (Newcomb, 1990).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5
Overview of the Study
This study represents one of the first attempts to empirically evaluate
attachment and interpersonal guilt as cognitive-emotional processes that
translate family experiences, dynamics, and interpersonal interactions into
eating disordered behaviors. The study utilizes the perspectives of
psychodynamic, attachment, evolutionary psychological, and prosocial
instincts theories to explain the development of relational bonds and
interpersonal guilt and ways in which these may be related to the
development of behaviors and attitudes associated with eating disorders.
If attachment, interpersonal guilt or both are identified as intervening
mechanisms, they will be important to address in both prevention and
treatment of these problems.
Theoretical Framework
This section addresses theoretical perspectives on the etiology of
eating disorders. Attachment theory provides a perspective that may help to
explain why individuals develop disordered eating from a family-based
context. Evolutionary psychology allows for the proposition that there are
certain biologically based internal mechanisms which promote survival for the
individual, and that these mechanisms have a psychological component and
impact.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6
Attachment Theory
Attachment theory speaks to bonding patterns between caregivers
and infants (Bowlby, 1969, 1973, 1980,1988). It refers to bonds formed
between caregivers and infants wherein ideally the caregiver acts as a
“secure base” (Bowlby, 1988) from which the child may wander and then
return. Children create internal maps or “ working models” which represent
experiences of the caregiver, self, and environment and which become
templates for relationships outside of the family (Bowlby, 1988). These
working models take different forms, depending upon the character of the
dyadic relationship between caregiver and child. These are classified as
“attachment styles."
Attachment styles
Attachment styles appear to be stable over time (Bretherton, 1985;
Main & Cassidy, 1988). That is, attachment styles in adulthood appear to
parallel those of childhood based on internalized working models (Pfaller,
Gerstein, & Kiselica, 1998).
Ainsworth and her colleagues (Ainsworth, Blehar, Waters, & Wall,
1978) found that infant attachment could be grouped into secure versus
insecure styles. Insecure styles were divided into anxious-ambivalent and
avoidant categories. More recently, Main and Solomon (1990) have
proposed a third insecure attachment style (disorganized or “D” type).
Researchers, though, have found much overlap between the several
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7
insecure attachment styles. Moreover, attachment occurs on a continuum
from “secure” to “insecure” and although it is possible to divide individuals
into attachment categories based on their interaction patterns, these are not
mutually exclusive. Therefore, for the purposes of this research, attachment
style was considered on a continuum from secure to insecure and was
measured as such.
Secure styles develop in response to a parenting style that allows the
child security and safety, yet encourages exploration. Insecure styles
develop in response to a care-taking style that is unpredictable or
nonexistent. To develop insecure attachment styles predicts not only difficulty
in adult relationships, but also a host of other pathological behaviors and
attitudes, including eating disorders. Consistent findings link eating disorders
to insecure attachment styles (O'Keamey, 1996; Ward et al., 2000).
One explanation for these findings highlights the link between the
developmental tasks of adolescence (separation from parents or caregivers
and beginning to establish independence) and the finding that most eating
disorders begin at this developmental stage. Attachment theorists contend
that adolescence is a time when attachment bonds are particularly sensitive
to disruption (Holmbeck & Hill, 1986), and that eating disordered behaviors
may indicate difficulties coping with the adolescent task of separation-
individuation (Striegel-Moore, 1995).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8
As previously noted, though, the manner in which disruptions in the
attachment processes might translate into eating disorders remains
unknown. One of the most recent reviews of attachment in eating disorder
research calls for “development of multifactorial and process-oriented models
of the role of attachment functions in the pathogenesis of eating disorders”
(O’Keamey, 1996, p. 120). The authors of a second recent review note that
“an association [between attachment style and eating disorder subgroups] is
likely to be complicated and it may be more fruitful to study specific aspects
of attachment rather than global attachment style” (Ward et al., 2000, p. 47).
The use of structural equation modeling with latent variables allows
exploration of specific antecedents of attachment as well as the effect of
attachment on eating disorders.
Family Systems Theories
The focus in this section will be on family systems theories. According
to family systems theorists, families may have delicate emotional balances
that children help to maintain. For example, Modell (1971) suggested that
people have an “unconscious bookkeeping system” which monitors the
distribution of available good within the family so that the current situation of
other family members determines how much “good” one is allowed to
possess. Similarly, Boszormenyi-Nagy and Spark (1973) introduced the
concept of the “ family ledger,” a familial accounting system of who,
psychologically speaking, owes what to whom. They introduced concepts
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9
such as “legacy” and “loyalty” to explain the set of expectations and
responsibilities toward each other which family members possess.
Protection of the family and loyalty to it have been implicated in the
development of eating disorders (Minuchin, 1974; Minuchin etal., 1978).
Loyalty to the family may prevent an individual from developing a sense of
autonomy and realizing his or her own personal potential, because to do so
would be seen as a transgression against the family. In this way, a child is a
“captive” of his or her environment/family. For that child to develop a
psychosomatic illness such as an eating disorder may represent a protective
defense against opposing anxieties (i.e., loss of family versus loss of self).
Weiss (1986, 1993) suggested that children may experience feelings
of guilt when they believe that their motives, traits or behavior threaten
attachment to parents or unbalance the family ledger. A child may give up or
change goals in order to maintain attachment to caregivers, thus leading to
intrapsychic conflict and potential psychological problems.
Within households where the child perceives one or both parents to be
functioning poorly, in order to maintain attachment, he or she might attempt
to “balance” the family system by taking care of the parent(s) who he or she
views as not functioning well. Reasonably, certain family experiences will
make a child more prone to developing a sense of responsibility and care-
taking attitude toward a parent. Although many different types of family
systems could elicit this behavior or attitude, this particular study focused on
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10
the empirically, clinically, and theoretically relevant family variables: parental
alcoholism and parental divorce or separation.
Evolutionary Psychology
This branch of psychology identifies and explains the development of
psychological mechanisms based on evolutionary concepts. Interpersonal
guilt as a construct and its inclusion in the current model is explained by
application of evolutionary psychology, in particular prosocial instincts theory.
Prosocial instincts theory
The premise of this theory is that humans have a biological instinct to
be altruistic or push toward self-sacrifice for survival of the group. A central
tenet of prosocial instincts or “scientific altruism” (Trotter, 1919, as cited in
Friedman, 1985) is that individuals experience “altruistic loyalty”
accompanied by emotions such as concern, empathy, and guilt that are
assumed to be “bottom level emotions” rather than veiled needs for individual
self-preservation (Friedman, 1985). Attachment theory claims that
individuals are motivated to maintain ties (attach) to significant others.
Prosocial instincts theory expands this notion by postulating that the
particular experience of guilt feelings, related to significant others is a
biologically-based mechanism that serves to allow maintenance of these ties.
A further assumption is that regardless of the quality of the
relationship, children are motivated to maintain ties to caretakers. Thus, guilt
may serve to attach an individual more strongly to significant family
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
11
members. However, the individual may be impeded in reaching desired
goals or development if either (a) the child’s concern for a family member or
feelings of responsibility overshadow his or her developing autonomy or (b) if
secure attachment to a caregiver is unlikely based upon the caregiver’s
psychological or physical unavailability. In these cases, guilt may be or
become “maladaptive.” This construct of interpersonal guilt and its various
components will be further explicated later.
Interpersonal guilt may be an important cognitive-affective state for
psychological practitioners to be aware of and acknowledge. To address
painful affects which are related to concern for others or fear of harming
others acknowledges the human tendency toward altruistic behaviors. It also
positively reframes negatively experienced affects as understandable
concern for others while simultaneously allowing the client to explore
maladaptive qualities of the experience.
Summary of Theory
These several theories were the basis for the conceptual model that
informed this study. Specifically, that model posits that excessive
interpersonal guilt may be fostered in family environments wherein the child
perceives that he or she is unable to successfully aid a parent or caregiver
for whom he or she feels responsible. This failure results in the child turning
against the self in an attempt to “equalize” the available amount of family
good.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12
Empirical research shows a relationship between the experience of
excessive interpersonal guilt and depression (L. E. O’Connor, Berry, Weiss,
& Gilbert, 1999), obsessive-compulsive symptomology (Escherick, L. E.
O’Connor, Berry, & Weiss, 1999), and substance abuse (L. E. O’Connor,
Berry, Inaba, Weiss, & Morrison, 1994). Individuals may experience guilt
when separating psychologically, physically, and socially from caregivers;
they may be more likely to experience excessive guilt if they perceive that
they are responsible for a loved one. Adolescence is a time of separation-
individuation. Most eating disorders develop during this time (Gandour,
1984; Levine, 1987). Additionally, eating disorders consistently have been
linked to disruptions in the attachment processes. Although this connection
has been strongly documented, the nature of the connection is unclear. The
present study aims to explore one potential way in which attachment and
eating disorders might be linked.
Definition of Terms
Eating Disordered Symptomatology
Eating disordered behaviors and symptoms occur on a continuum
(Hart & Ollendick, 1985; Rodin, Silberstein, & Striegel-Moore, 1985; Stice,
Killen, Hayward 8 c Taylor, 1998). For the purposes of this study, the term
“eating disordered symptomatology” reflects symptoms that are related to
eating disordered mentality, attitudes, behaviors, and symptoms. The
behaviors that will be assessed include skipping meals to lose weight,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13
avoidance of certain foods because they will cause weight gain, episodic
binge eating, and using purging or restriction to control and/or lose weight.
Interpersonal Guilt
Guilt has been variously defined and conceptualized in the literature.
It was originally defined as an emotion resulting from a personal, intrapsychic
conflict (Ausubel, 1955). However, more recent conceptualizations highlight
interpersonal aspects of guilt (Baumeister, Stillwell, & Heatherton, 1994) and
describe guilt as a “relationally based affect” (Lopez et al., 1997). In this
study the “interpersonal” nature of guilt is emphasized and guilt is viewed as
an emotion which allows maintenance of attachments, first to parents and
siblings, and later to others in one’s social environment (Baumeister et al.,
1994). The specific definition of “guilt” which will be utilized in this study is
that proposed by Friedman (1985): “ the appraisal, conscious or unconscious,
of one’s plans, thoughts, actions, etc., as damaging, through commission or
omission, to someone for whom one feels responsible" (p. 530).
Review of the Literature
The following review of the literature provides an empirical, theoretical
and/or clinical rationale for inclusion of each of the variables in the
hypothesized model.
Family Experiences and Eating Disorders
Research has established a link between various stressors and the
onset of eating disorder symptomatology. One general variable, which has
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14
received some recent attention in the literature, is the experience of “loss.” In
their recent literature review, Ward et al. (2000) noted that “unresolved
losses” appear to be prevalent within eating disordered populations. They
call for further exploration of the relationship of this variable to attachment
and eating disorders. Ward et al. (2000) do not, however, give examples of
what experiences might be considered “unresolved loss.”
Other researchers have noted that “losses,” such as separation from
family members, death of a parent or sibling, and divorce, are related to
eating disorders, and call for more exploration (Holden & Robinson, 1988;
Silber, 1986; Robinson & Andersen, 1985). Significant family losses
experienced by individuals in childhood and adolescence can include family
deaths, parental divorce, separation from parents or family members, and
emotional losses related to inadequate parenting or abuse. Two familial
“loss” variables that have shown a relationship in the literature with
development of eating disorders are parental alcohol abuse and parental
divorce.
Parental Substance Abuse
Children who grow up in families where one or both parents are
abusing alcohol may experience a lack of competent parenting that can be
conceptualized as a “loss.” Women who grow up in alcoholic families appear
to be at risk for eating disorders. A large body of clinical research reveals
high rates of alcoholism in parents and first-degree relatives of bulimic
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
15
women (Bulik, 1987; Hudson, Pope, Jonas, & Yrgelun-Todd, 1983; Kassett
et al., 1989; Mitchell, Hatsukami, Eckert, & Pyle, 1985). Binge eaters, too,
have reported significant rates of parental alcohol abuse (Kanter, Williams, &
Cummings, 1992).
In addition to clinical samples, undergraduate samples of women with
family history of parental alcohol abuse or experience with an alcoholic
significant other, evidence higher levels of general eating disordered
behaviors (Meyer, 1997) and bulimia (Claydon, 1987). A large sample of
adolescents from alcoholic families revealed significantly higher rates on
seven eating disorder symptoms (Chandy, Harris, Blum, & Resnick, 1994).
One recent study, utilizing 7,147 subjects also found a higher prevalence of
eating disorders in men who grew up within an alcoholic family (Cuijpers et
al., 1999).
Although there is a great deal of evidence linking eating disorder
symptomatology with experience of parental alcohol abuse, the strength of
that relationship varies across studies. For example, Mintz, Kashubeck and
Tracy (1995) found only a minimal relationship between parental alcoholism
and eating disorder symptomatology in a college sample. Orenstein, Davis,
and Wolfe (1993) found that high school students who were distressed by
their parents’ drinking did not differ from other youth on eating disorder
measures. However, in this study, participants whose parents had been
treated for alcohol and drug problems showed higher levels of eating
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16
disorder symptomatology as well as higher levels of substance use and
delinquent behavior.
The connection between parental alcohol abuse and eating disorder
symptomatology may not be direct, but rather mediated by one or other
factors. In this way, its effects are “indirect" One study examining parental
alcoholism and mediating effects of the construct “codependency” found that
the more characteristics of codependency displayed in female adult children
of alcoholics, the more eating disorder symptoms were displayed (Meyer,
1997). Given findings such as these, Mintz et al. (1995) asked that
researchers examine individual distress levels related to being an adult child
of an alcoholic, rather than to assume that adult children of alcoholic families
will necessarily display signs of pathology. Many variables might affect an
individual’s response to an alcoholic parent or parents. One is the specific
nature of the relationship that a child has with a parent, particularly the level
of responsibility or family loyalty that the adult child has felt. The current
study explored the possibility that the mechanisms of attachment and
interpersonal guilt mediate the relationship between parental alcoholism and
eating disordered symptomatology.
Parental Separation and Divorce
Rates of parental separation and divorce remain high. Therefore, this
is a common and significant loss event in the life of children. In a recent
guide to treatment and prevention of eating disorders, Clopton, Haas, Kent,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17
and Robert-McComb (2001) noted “ parental divorce” as a factor in the
development of eating disorders. This is supported by empirical studies
(Holden & Robinson, 1988; Robinson & Anderson; 1985 Silber, 1986).
However, these studies used very small samples, and were not concerned
with looking directly at parental separation and its relationship with eating
disorders.
Shisslak et al. (1998) examined risk factors for eating disordered
behaviors and symptoms in 523 preadolescent and adolescent (ages 9-15)
girls. They found that the experience of parental divorce or separation was
associated with unhealthy eating and weight control behaviors.
Three studies that utilized relatively small samples, also have found
support for this association. In one study, the frequency of broken homes in
bulimic patients was significantly higher than for 40 matched obese patients
(19 of 40, versus 6 of 40). The researchers concluded that parental
separation is a template for a situation of intolerable choice that then is
reflected in the dichotomous thinking associated with bulimic behaviors
(Igoin-Apfelbaum & Apfelbaum, 1990). In another study that examined
effects of separation and divorce on 80 participants, ages 4-18, researchers
found that children of divorced parents were more likely to display eating
disorders, along with a host of other psychopathological symptoms such as
sleep disorders, depression, and language disorders (Soler, Bargada,
Molina, Bassas, & Vilaltella, 1996). A third study (n = 25 women) revealed
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
18
that parental psychiatric illness may be a risk factor for bulimia. It may
contribute to environmental effects through increased rates of divorce and
impaired parental relationships (Boumann & Yates, 1994).
Thus, there is some evidence to indicate that parental separation or
divorce may be a risk factor for eating disorders. This study examined a
model in which parental separation and divorce were hypothesized to
indirectly affect levels of eating disorder sympotmatology.
Family Experiences and Attachment Style
Family experiences affect the development of attachment style in
children and adolescents (Crittenden, 1988; Egeland & Stroufe, 1981;
McCarthy & Taylor, 1999; Schneider-Rosen, Braunwald, Carlson, &
Cicchetti, 1985). Once formed, these styles of attachment persist into
adulthood (Diehl, Elnick, Bourbeau, Labouvie-Vief, 1998; Mikulincer&
Florian, 1999). Theorists suggest that family-of-origin experiences are a
mechanism for sustaining attachment styles throughout adulthood
(Bartholomew & Horowitz, 1991; Bowlby, 1988). Several empirical studies
have examined the relationship between attachment styles of young adults
and specific family-of-origin variables, particularly those of parental
substance abuse and parental divorce and separation.
Parental Substance Abuse
Adult children of substance abusers may be more likely to show
insecure attachment patterns in childhood and adulthood. M. J. O’Connor,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
19
Sigman, and Brill (1987) found a relationship between moderate-to-heavy
alcohol use in mothers and insecure infant attachment Two unpublished
research studies reveal that adult children of problem drinkers develop
insecure attachment patterns (Latty-Mann, 1989, 1991, cited in McCarthy &
Taylor, 1999). Another study noted that adult children of alcoholics are more
likely to report insecure attachment styles (Brennan, Shaver, & Tobey, 1991).
However, at least one study has failed to demonstrate parental
alcoholism as a predictor of parental attachment (Mothersead, Kivlighan &
Wynkoop, 1998). And one study found differential effects by gender El-
Guebaly, West, Maticka-Tyndale, and Cohn (1993) found that female adult
children from alcoholic families had a dysfunctional attachment profile,
whereas males did not.
Parental Divorce and Separation
Research findings regarding the relationship between attachment and
parental divorce have been somewhat inconsistent, although the most recent
research increasingly has established a relationship between the two. One
recent study, conducted with a large, nonclinical population of adolescents
and young adults, found that both parental divorce and parental death
predicted dismissive or emotionally defensive styles of interaction (Brennan
& Shaver, 1998).
Other studies have shown that college women from divorced families
appear to have less secure attachment styles than individuals without the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20
experience of parental divorce (J. J. Evans & Bloom, 1996; Sprecher, Cate,
& Levin, 1998). Additionally, in studies using either the four-group divisions
(e.g., secure, preoccupied, dismissive, and fearful) or a continuous measure
of attachment, individuals from divorced families were less likely to show
secure attachment styles (Brennan & Shaver, 1998; Summers, Forehand,
Armistead, & Tannenbaum, 1998). Elsewhere, in a large, nationwide adult
sample, Mickelson, Kessler, and Shaver (1997) found that parental divorce
or separation was positively related to anxious attachment and negatively
related to secure attachment.
In earlier studies of college students where the three-group categorical
division has been utilized (e.g., secure, anxious-ambivalent, and avoidant),
no significant differences were found between individuals from intact families
versus those from divorced families (Brennan & Shaver, 1993; Feeney &
Noller, 1990; Hazen & Shaver, 1987). The most recent research utilizing the
most current four-category division of attachment (Bartholomew & Horowitz,
1991; Griffin & Bartholomew, 1994) indicates that there is an association
between parental separation or divorce and insecure attachment style in
offspring. As noted previously, this study conceptualized attachment as a
continuous variable (secure to insecure).
Interpersonal Guilt: Perceived Responsibility
An underlying assumption of this study was that the experience of
interpersonal guilt is related to prosocial instincts and attachment. Family
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
21
systems theorists maintain that individuals are powerfully motivated to foster
bonds with others that serve to protect not just that individual, but the
other(s), and the dynamic relationship between them. Thus, a child desires
to be protected by and to protect his or her parent.
The notion that the child may desire to protect his or her parent is very
important. It allows that a child of any given age, maturity level, or role may
have this desire, yet most likely will not be able to protect a parent in any
significant way. Guilt may then be the cognitive and/or affective response to
this perceived failure. The child may symbolically protect the parent from
whatever harm he or she perceives by turning against the self as a leveling
device and to balance the family ledger, or so as not to be “better off than" a
parent or sibling. Hence the development of emotional distress, and
potentially a psychological disorder such as an eating disorder, which
essentially allows an individual not to progress or be successful. This
conceptualization has roots in evolutionary psychology that identifies the
“ prosocial instincts theory” described in an earlier section as a key area of
human development.
Family Experiences and Guilt
Family-of-origin variables, which have shown association with guilt in
the literature, are described in this section. Although guilt has more global
features than just interpersonal, this study focused only on its interpersonal
features. That focus is reflected in this review.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
22
Empirical and clinical evidence and theoretical arguments indicate that
the emotional and/or physical loss variables of divorce and family alcoholism
previously noted to be related to eating disorders and attachment are also
associated with maladaptive guilt patterns.
Parental Substance Abuse and Dependence.
Guilt is frequently associated with parental alcoholism. One study
examining a combination of 277 adult children of alcoholics and those who
were not adult children of alcoholics found that adult children of alcoholics
were more likely to experience guilt, resentment, and anxiety than were
adults from nonalcoholic families. This effect was heightened for those who
were alcoholics themseives (Carpenter, 1995). Another study comparing 63
adolescents with at least one alcoholic parent to 161 adolescents with no
alcoholic parents found that children of alcoholics reported significantly
higher levels of guilt (Tomori, 1994).
Clinical evidence reveals that in alcoholic families, children’s anger at
their parents may be expressed as guilt (Breen, 1985), that maladaptive roles
adopted by family members include experiences of shame and guilt (Murphy,
1984) and that guilt is an important motivating feeling behind behaviors such
as overachievement, over control, and perfectionism. These behaviors are
often evidenced in “ hero” children of alcoholic families (Glover, 1994, p. 185).
Marcus and Tisne (1987) found that children with alcoholic mothers more
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23
often perceived their mothers to be using guilt as a means of psychological
control than their counterparts with nonalcoholic mothers.
Parental Divorce and Separation
Clinicians report that parental divorce is associated with guilt in
children (Canziani, 1996) and young adults (Lopez, 1991). Young adults in
particular may struggle with feelings of worry over parents’ well-being and
loyalty conflicts (Lopez, 1991). In a review of behavioral patterns and roles
that children of divorced parents may adopt, Klosinski (1993) found guilt-
related patterns of children’s inclination to sacrifice themselves for the sake
of their parents, “ parentization” where the child takes on the role of a
substitute partner, and feelings of guilt which arise with respect to the conflict
between the combined aspects of both power and helplessness inherent in
their situations.
These studies have measured and assessed guilt as a global
construct and have not examined the interpersonal nature of guilt relative to
the situation that is assumed to produce it. Much of the work is clinical in
nature and findings would be strengthened if corroborated by empirical
research. This study assessed interpersonal guilt and its relationship to
specific qualities or characteristics of family relationships, attachment style,
and eating disordered behaviors through empirical methods.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24
Attachment Style and Interpersonal Guilt
Because many of the family variables that are linked to insecure
attachment also have been linked to feelings of guilt and responsibility in
children and adults, it is possible to infer that attachment may be related to
interpersonal guilt. Theorists have suggested a connection between
attachment and the experience of guilt related to a caretaker or significant
other.
Some theorists believe that guilt develops within the very early infant-
caregiver bond. For example, Baumeister et al. (1994) noted that guilt
commences within the infant’s attachment to the caregiver and in the fear
and anxiety over breaking that attachment. Guilt thus operates as a force
that perpetuates the relationship. Evolutionary theorists such as Trivers
(cited in Friedman, 1985) have suggested that guilt materialized from natural
selection because it prevented humans from carrying out exploitative actions
that could be destructive to their relationships with others—relationships that
are critical to survival and reproduction. It is apparent that guilt has relational
roots. Guilt may be an expression of our ability to feel or experience others’
suffering and distress. Guilt may also be an experience that involves a fear
of alienating actual or potential attachments.
Some recent empirical evidence indicates an association between
attachment style and guilt. In a study of 153 university students, Leondari
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
and Kiosseoglou (2000) found a positive association between security of
attachment and freedom from guilt, anxiety, and resentment towards one’s
parents. In another study utilizing 41 college students, guilt scales correlated
inversely with secure attachment (Harder & Greenwald, 1999). One model
tested in this study hypothesized a relationship between the two variables.
The proposed attachment-interpersonal guilt link in this current study was its
most exploratory aspect.
Adult Attachment Style and Eating Disordered Behaviors
A recent review of eating disorders and attachment research reported
no significant overall relationship between insecure attachment and eating
disorder symptomatology (O’Keamy, 1996). However, in the few years since
the publication of this review, more research has been published in this area
than in all the time prior to 1996. The most recent review found consistent
evidence that attachment disturbance is linked to eating disordered
behaviors and development of eating disorders (Ward et al., 2000).
Research comparing female clinical and non-clinical samples found
that secure attachment styles predict membership in non-clinical groups,
whereas insecure attachment styles predicted membership in eating
disordered groups (Chassler, 1997; Friedberg & Lyddon, 1996;). Candelori
and Ciocca (1998) found 83% of 36 inpatients were rated “insecure” on the
adolescent version of the Adult Attachment Inventory. In non-clinical groups,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
26
insecurely attached participants reported greater weight concerns than
securely attached participants, and thus were at greater risk of developing
eating disorders (Sharpe et al., 1998). Elsewhere, in a sample of 360 college
students and student nurses, those with eating/body concerns reported an
insecure attachment style (L. Evans & Wertheim, 1998).
Although there is a clear connection between insecure attachment and
eating disorder symptomatology, the origin and accompanying individual
experiences are less evident Ward et al. (2000) suggest that it may be
helpful to “ study specific aspects of attachment rather than global attachment
style in relation to eating disorder behavior” (p. 49). This study hypothesized
parental alcoholism and divorce as life experiences that might affect
attachment bonds and in turn affect the development of eating disordered
behaviors and symptomatology.
Interpersonal Guilt and Eating Disordered Symptomatology
Theoretically and clinically, guilt has been connected with eating
disorders. In a recent treatment handbook for eating disorders, Goodsitt
(1997) noted that patients with eating disorders seem to “ suffer profound guilt
for the wish to separate and for acts of separation and individuation” (p. 212).
He described this “ self-guilt” as a “pervasive sense of discomfort for simply
being or existing” (p. 212). Goodsitt asserted that the patient feels guilty for
occupying both physical and psychological space to the point that “ any act of
occupying psychological space is experienced as an immoral, hostile, and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
27
destructive act that deprives others of their psychological soace” (p. 212).
For those with eating disorders, eating means giving to oneself, but
necessarily depriving someone else of sustenance. This guilt that Goodsitt
describes is thus interpersonal in nature.
Friedman (1985) employed a qualitative method to examine eight
clients with anorexia nervosa. He found high levels of both survivor and
separation guilt as they have been defined in the current study. The author
called for more research on this topic based on his findings and on the
additional finding that when he addressed interpersonal guilt with these
clients, they found it to be a very helpful conceptualization.
At least three empirical studies have directly examined guilt and eating
disordered behaviors. One study testing the relationship between eating
disordered behaviors and the specific construct “interpersonal guilt” was a
study looking at 95 college females (Vilas, 1997). This research found an
association between self-hate guilt and eating disordered behaviors and no
relationship between survivor guilt and separation guilt and eating disorder
symptomatology. The study was correlational in nature, and did not look at
potential antecedents or associations between other variables.
Another study found a relationship between guilt over separation from
parents and eating disordered behaviors (Smolak & Levine, 1993). In this
study of college aged women, “ bulimic-like” women were found to experience
more guilt and conflict concerning separation from parents than did women
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
28
showing no eating disordered behaviors or women showing limited eating
disordered behaviors. Additionally, “ anorexic-like” individuals reported
significant levels of guilt and conflict about separation from parents. A third
study found that individuals with pathological feelings of guilt are more likely
to have eating disturbances (Bybee, Zigler, Berliner, & Merisca, 1996). In
this study, ineffective alleviation of guilt feelings was correlated with greater
eating disturbances. Additionally, women who reacted to guilt-producing
events with fewer intropunitive responses such as rumination and self-hatred,
and by confiding in others, distancing, and rationalizing their actions through
justifications and excuses were less likely to have eating disturbances.
A handful of studies identify a relationship between eating disordered
behaviors and concepts related to interpersonal guilt For example, Chassler
(1997) found that female bulimic and anorectic inpatients were more likely
than non-eating disordered females to report “ feeling responsible for parent
happiness” (p. 47). Other research has shown a link between
“codependency” in women and higher levels of eating disordered
symptomatology. According to Engel (1990) the definition of “ codependency”
includes the notion that one feels responsible for and takes care of others.
Meyer and Russell (1998) found that female college students who displayed
more codependent characteristics evidenced higher levels of eating
disordered behavior and conflictual separation from parents.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
29
Taken together, the above findings suggest that guilt related to the
fear of harming others may play a role in development of eating disordered
behaviors and that further confirmation is warranted.
Research Questions
1. Do family-of-origin experiences (parental divorce, parental
alcoholism) predict eating disorder symptomatology?
2. Do family-of-origin experiences (parental divorce, parental
alcoholism) predict adult children’s attachment style?
3. Do family-of-origin experiences (parental divorce, parental
alcoholism) predict the experience of interpersonal guilt?
4. Does attachment style predict eating disorder symptomatology?
5. Does interpersonal guilt predict eating disorder symptomatology?
6. Does attachment style predict interpersonal guilt?
Research Hypotheses
Based on the previous research questions, the following hypotheses
were formulated:
1. Familial experiences of parental alcoholism and/or divorce will
affect the development of attachment style, which in turn will affect the
development of eating disordered symptomatology.
2. Familial experiences of parental alcoholism and/or divorce will
affect the development of interpersonal guilt, which in turn will affect the
development of eating disordered symptomatology.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
3. Insecure attachment will predict higher levels of interpersonal guilt.
In the current study, the proposed full model was compared with a
model specifying direct links between parental alcoholism, parental divorce,
and eating disordered symptomatology to determine if inclusion of
Attachment and Interpersonal Guilt as intervening processes would create a
better fitting model.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
31
CHAPTER 2
METHODOLOGY AND PROCEDURES
This section describes the methods and procedures used in the study.
It includes summary of research design, recruitment of the sample, selection
and description of participants, description of the instruments used (including
discussion of their reliability and validity), and analysis procedures.
Research Design
This was a validation study with an exploratory component. It
attempted to evaluate from a psychodynamic, attachment and evolutionary
psychological perspective, some possible origins of development of
attachment, interpersonal guilt and eating disordered symptomatology.
The accompanying diagram, presented in Figure 1, represents the
proposed full model, depicting the hypothesized relationships among the
factors described in this study. The model represents one way to organize
and conceptualize internal processes. Family factors represent the external
antecedents that are believed to potentially contribute to the development of
attachment problems and interpersonal guilt. “ Attachment” is a mediating
template that represents the quality of the bonds between parent and child
from the adult child’s perspective. The construct “Interpersonal Guilt”
represents an intervening cognitive-affective process that is believed to be
related to attachment and to influence development of pathological
behaviors, one of which may be eating disordered symptomatology. The
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
c ose
depend
anxiety
Attachment
parental alcohol
abuse
Eating
Disordered
Symptomatology
parental divorce
Interpersonal
Guilt
self-hate
separation
oral control
dieting
bulimia and food
preoccupation
survivor omnipotent
responsibility
Figure 1. Original proposed model depicting the hypothesized relationships among the variables
33
theoretical variables “Interpersonal Guilt", “ Attachment” and “Eating Disorder
Symptomatology” are represented as circles because they are constructs
that are not directly observable. They are termed “latent variables”
(Newcomb, 1992).
According to Newcomb (1990,1994) one of the conditions in which it
is appropriate to utilize latent factors is when similar measured variables are
indicators of the latent construct. One should be able to hypothesize that
there are unique aspects of these indicator variables that can be
differentiated from the common part that is shared with the other indicators of
the latent variable and that is not truly error variance. Newcomb (1990)
stated,
Multiple-item, unidimensional instruments can be divided into three or
more subscales to reflect a latent construct of that scale. The
correlations among these subscales would reveal the latent factor that
is hypothesized to underlie the responses to the subscales’ items, (p.
32)
For the purposes of this study, three latent variables were created
from a total of 10 measured variables. Four measured variables: Survivor
Guilt, Separation Guilt, Omnipotent Responsibility Guilt and Self-Hate Guilt
are hypothesized as indicators of the general construct, Interpersonal Guilt
(IG). Three measured variables: Close Attachment, Dependent Attachment
and Anxious Attachment are indicators of the general construct, Attachment
(ATT). Three measured variables: Dieting, Oral Control and Bulimia-Food
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
34
Preoccupation are indicators of the general construct, Eating Disordered
Symptomatology.
The logic of factor analysis allows use of several measures of a
theoretical variable and allows extraction of that portion of each measure
specifically related to the theoretical variable (Fassinger, 1986). Latent-
variable models are utilized to determine the relationships of the measured
guilt variables with other variables. The directional relationships among
Interpersonal Guilt, Attachment and the other factors are speculative, but
based on the preceding theoretical considerations. The use of Structural
Equation Modeling (SEM) allows investigation of the associations of the
combined aspects of Interpersonal Guilt, Attachment and Eating Disorder
Symptomatology and allows for the inclusion of the unique aspects of
specific experiences.
Selection of the Sample
The participants were 201 female undergraduate and graduate
students drawn from a variety of classes in two colleges in Los Angeles
County and two colleges in the San Francisco Bay Area. All participants were
over the age of 18 and spoke English. Ages ranged from 18 to 49 (M = 21).
The ethnic breakdown of the sample was as follows: 20% Euro-American,
15% Asian American, 10% African-American, 36% Latina, 4% Middle-
Eastern American, 13% Mixed race.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
35
A passive consent form, which delineated the purpose of the study
and explained participants’ rights, was utilized. Participants were informed in
written and oral format that return of the completed sun/ey indicated their
consent. Participation in the project was entirely voluntary and students
received no course credit or payment for participation in the study.
Instrumentation
All participants received the same questionnaire. The questionnaire is
a combination of items from the Interpersonal Guilt Questionnaire (IGQ-67),
the Adult Attachment Scale (AAS), the Eating Attitudes Test (EAT-26), and a
demographic questionnaire.
Demographic Questionnaire
All respondents completed a brief questionnaire soliciting information
on their age, gender, sexual orientation, year in college, race and ethnicity,
religious affiliation, parents’ marital status and current description of
relationship with parents. This information may be found in Table 1.
Children of Alcoholics Screening Test (CAST)
The CAST is a widely used 30-item inventory that evaluates
perceptions, feelings, attitudes and experiences related to parents’ drinking
behavior. The questionnaire distinguishes children of alcoholics in both
clinical and nonclinical samples (J. W. Jones, 1981). Items are arranged in a
yes or no format. The total number of “ yes” answers establishes whether a
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
36
Table 1
Description of the Women in the Study
Characteristic n %
Age
18-22 162 80.0
23-27 22 11.0
28-32 5 3.0
33-49 12 12.0
Ethnicity
African-American 21 10.0
Latina 73 36.0
Caucasian 40 20.0
Asian-American 31 15.0
Middle-Eastern American 7 4.0
Mixed 27 14.0
Religious/Spiritual Affiliation
Catholic 128 64.0
Protestant 5 3.0
Buddhist 6 3.0
Jewish 4 2.0
Islamic 5 3.0
Episcopal 3 2.0
Other 45 23.0
Educational Level
First year college 50 25.0
Second year college 42 21.0
Third year college 55 27.0
Fourth year college 44 22.0
Graduate student 10 5.0
Sexual Identity
Heterosexual 191 97.0
Lesbian 1 .5
Bisexual 6 3.0
Current Relationship
Single 155 77.0
Married 13 7.0
Partnered 30 15.0
Separated/Divorced 3 2.0
Parents' Relationship
Married 125 63.0
Divorced/Separated 73 37.0
Note. Because of missing data, not ail categories total 201.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
37
person may be classified as a child of an alcoholic. Continuous scores can
be used to measure distress concerning parental drinking (Clair & Genest,
1992; Dinning & Berk, 1989; Roosa, Sandler, Gehring, Beals, & Capp, 1988).
The reliability and validity of the CAST has been studied in adolescent
(Dinning & Berk, 1989), adult (Pilat & J. W. Jones, 1985) and psychiatric
populations (Staley & el-Guebaly, 1991). It has been found to discriminate
between the offspring of alcoholic parents and the offspring of non-alcoholic
parents. In one study, it reliably identified 100% of the children of both
clinically diagnosed and self-reported alcoholics (J. W. Jones, 1981).
However 23% of these children with no known history of parental alcoholism
also scored above the cut-off point, indicating the possibility of false positives
when used with a sample of children. Additionally, the drinking behavior of
the parents of children in the control group was not assessed, so there was
no way to determine the true rate of alcoholism in parents of the control
group. The current study utilized a sample of young adults. Therefore fewer
false positives are expected, for the comprehension level of the adults is
expected to be greater than that of the children used in the aforementioned
sample.
Examples of items from the CAST include: “Did you ever encourage
one of your parents to quit drinking?” And “ Did you ever think your mother
was an alcoholic?” CAST reliability for the current study was .96. The range
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
38
of possible scores is 0 to 30, with higher scores indicating higher level of
distress.
Eating Disorder Symptomatology
The Eating Disorder Symptomatology construct was represented by
three measured variables that comprise the subscales of The Eating Attitude
Test (Gamer, Olmsted, Bohr, & Garfinkel, 1982). The EAT-26 is a shortened
version of the original, 40-item EAT. This self-report measure consists of 26
items and measures cognitions, emotions, and behaviors associated with
anorexia nervosa and bulimia on three scales: Oral control, Dieting, and
Bulimia-Food Preoccupation (Gamer & Garfinkel, 1979). The EAT-26 may
be used as a continuous measure of a range of eating disorder
symptomatology in nonclinical populations (Koslowsky et al., 1992;
O’Halloran & Mintz, 1997). A 6-point frequency scale is used to rate items,
with responses ranging from 6 (always) to 1 (never) with higher scores
indicating more symptomatology (Mazzeo, 1999).
For the original EAT, Gamer and Garfinkel (1979) report internal
consistencies of .79 for a sample of eating disordered individuals and .94 for
a pooled sample of eating and non-eating disordered individuals. Other
research supports the reliability and validity of the EAT-26 (e.g., Button &
Whitehouse, 1981). The three subscales are:
Bulimia and food preoccupation. This subscale contains 6 items that
reflect thoughts about food as well as those that indicating bulimia. Sample
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
39
items for this scale include “I feel that food controls my life” and “I have the
impulse to vomit after eating.” Reliability for the current study: .81.
Dieting. This subscale contains 13 items that relate to an avoidance
of fattening foods and a preoccupation with being thinner. Sample items for
this scale include “I like my stomach to be empty” and “I am preoccupied with
the thought of having fat on my body.” Reliability for the current study: .89.
Oral control. This subscale contains 7 items that measure self-control
of eating and the perceived pressure from others to gain weight. Sample
items for this scale include “I avoid eating when I’m hungry” and “ Other
people think I’m too thin.” Reliability for the current study is: .64.
Attachment
Three subscales from the 18-item Collins and Read (1990)
Attachment Scale reflected the latent construct of Insecure Attachment. This
measure is comprised of six statements for each of three factor-analytically
derived scales: Close; Depend; and Anxiety. Items are in a 5-point Likert-
type format ranging from “Not at all characteristic of me” to “ Very
characteristic of me.” Scores on each scale can range from 6 to 30. High
scores are indicative of more of that attachment dimension.
Anxious attachment. The Anxiety subscale measures fears of
abandonment. High scores on the anxiety scale characterize those who are
fearful of being rejected or abandoned by others. Examples of items from
this scale include “I find that others are reluctant to get as close as I would
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
40
like” and “I want to get close to people, but I worry about being hurt”.
Anxious Attachment reliability for the current study is .87.
Close attachment. The Close subscale assesses level of comfort with
intimacy. High scores on the close scale endorse feeling comfortable getting
emotionally close to others. Examples of items from this scale include “I find
it relatively easy to get close to people” and “I am comfortable developing
close relationships with others” . Close Attachment reliability for the current
study is .83.
Dependent attachment. The Depend subscale measures the extent to
which participants trust others and depend on them to be available. High
scores on the Depend scale characterize those who feel that they cannot
depend on others when in need. Examples of items from this scale include “I
find it difficult to allow myself to depend on others” and “I find that people are
never there when you need them”. Dependent Attachment reliability for the
current study is .82.
Interpersonal Guilt
The Interpersonal Guilt latent construct was reflected in four
moderately correlated subscales (Separation, Survivor, Omnipotent
Responsibility, and Self-Hate) that were factor analytically derived from the
Interpersonal Guilt Questionnaire (IGQ-67) (L. E. O’Connor, Berry, Weiss,
Bush and Sampson, 1997). The IGQ-67 is a 67 item questionnaire
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
41
assessing proneness to guilt related to the fear of harming others: the four
subscales of the IGQ-67 describe four distinct but related types of
interpersonal guilt: Survivor Guilt, Separation Guilt, Omnipotent
Responsibility Guilt and Self-Hate. For each of the 67 items, respondents
rate themselves on a 5 point Likert scale with 1 = very untrue of me OR
strongly disagree and 5 = very true of me OR strongly agree.
Cronbach’s alpha coefficients determined from previous studies have
ranged from .76 to .85 for Survivor Guilt, from .73 to .83 for Separation Guilt,
from .71 to .83 for Omnipotent responsibility Guilt, and from .84 to .89 for
Self-hate Guilt (L. E. O’Connor, Berry, Weiss, & Gilbert, 1999; L. E.
O’Connor, Berry, Weiss, Bush, & Sampson, 1997; Menaker, 1995). The
construct validity of the IGQ-67 has been established through correlations
with other measures of guilt and with a variety of measures of
psychopathology (L. E. O’Connor, Berry, Weiss, Herbold etal., 1996; L. E.
O’Connor, Berry, Weiss, Bush et al., 1997; L. E. O’Connor, Berry, Weiss, &
Gilbert, 1999).
Omnipotent responsibility guilt. Omnipotent Responsibility Guilt was
represented by a single subscale containing 14 items. This type of guilt is
characterized by exaggerated sense of responsibility and concern for well
being of others. Examples of items from this scale are: “I often find myself
doing what someone else wants me to do rather than doing what I would
most enjoy” and “I can’t stand the idea of hurting someone else.”
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
42
Omnipotent Responsibility Guilt reliability for the current study is .66.
Self-hate guilt. Self-Hate Guilt is represented by a single subscale containing
16 items. This type of guilt may occur as a result of compliance with
punishing, neglectful or rejecting parents and may occur in people who see
themselves through the eyes of someone they believe hates them. This
extreme negative opinion of the self is adapted in order to maintain
connection to parents or caregivers (L. E. O’Connor, Berry, Weiss, Bush et
al., 1997). Examples of items from this scale are: “I do not deserve other
people’s respect or admiration” and “I always assume I am at fault when
something goes wrong.” Self-Hate Guilt reliability for the current study is .92.
Separation guilt. Separation was represented by a single subscale
containing 15 items. Separation guilt assesses belief that if one separates
from loved ones or is different than they are, one is being disloyal and will
thus harm others as a result of pursuing ones' goals (Bush, 1989; Weiss,
1986). Examples of items from this scale are: “I feel that bad things may
happen to my family if I do not stay in close contact with them” and “I feel
uncomfortable when I do things differently than my parents did them.”
Separation guilt reliability for this study is .77.
Survivor guilt. Survivor guilt was represented by a single subscale
containing 22 items. This scale assesses the belief that by endorsing one’s
own cause or goals is injurious to others (Weiss, 1986). Examples of items
from this scale are: “I conceal or minimize my success” and “I am
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
43
uncomfortable talking about my achievements in social situations”. Survivor
guilt reliability for this study is .77.
Procedures
All data were collected on-site at the universities involved. The author
collected all data, ensured that each group received the same instructions,
and that directions were accurate and consistent.
Test Administration
Students were instructed to provide the data requested. Each student
was reminded that her participation in this study was voluntary and that
withdrawal at any time during testing was permitted. Following completion of
the test instrument, participants were given a written debriefing form that
explained the objectives of the study and included information about on-
campus and off-campus resources for assistance with any concerns
regarding family issues, eating disordered symptoms, and relationships.
Completed questionnaires were collected by this researcher. No one except
this author has access to the completed questionnaires.
Data Analyses
The method of analysis utilized in the present study was structural
equation modeling (SEM). A full structural equation model is made up of two
parts: a path model that specifies the hypothesized structure among latent
variables (theoretical constructs not directly observable) and a measurement
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
44
model that defines relations between indicators (directly observable
variables) and the latent variables that they approximate (Newcomb, 1990).
Measured variables with high factor loadings are presumed to
represent an underlying trait. For example, four measured variables—
Survivor Guilt, Separation Guilt, Omnipotent Responsibility Guilt, and Self-
hate Guilt— are hypothesized to represent the general construct
Interpersonal Guilt and the level of correlation among these selected
measured variables is reflected in the latent construct.
When properly identified, latent constructs are more powerful
variables than measured scales and they are more likely to identify subtle
effects that might be hidden by random measurement error. SEM allows
examination of relationships among unobserved (latent) constructs or factors
that are not influenced by errors of measurement (Newcomb & Bentler,
1988). Additionally, SEM allows testing of relationships among various types
of variables including: latent variables, measured variables, residuals of
variables, and mediating/moderating variables (Newcomb, 1990).
All models and their respective interpretative statistics were conducted
with AMOS (Arbuckle, 1997). Data management, including descriptive
statistics was conducted with SPSS (SPSS, Inc., 2000).
First, the measurement models were examined for poorly fitted path
coefficients. Any poor path coefficient will necessarily have a negative
impact on model fit statistics, and thus may impede interpretation of model fit.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
45
Second, “ goodness of fit” was measured with the ratios between
maximum likelihood chi-square estimates and the degrees of freedom (x2 /df).
Chi-square/degrees of freedom ratios of less than 2:1 indicate well-fitting
models (Newcomb, 1994). R. E. Schumacker and Lomax (1996)
recommend using multiple indices for ease of interpretation. The most
common indices used are the Goodness of Fit (GFI) index and Adjusted
Goodness of Fit (AGFI) (Joreskog & Sorbom, 1989). These indices were
selected for evaluation of all models except Model 1 (direct effects model).
Model 1 only runs properly if AMOS is allowed to estimate means and
variances. This is a fairly common occurrence when utilizing SEM, however
it precludes calculation of GFI and AGFI for this model. Thus, only chi-
square and Root Mean Square Error of Approximation (RMSEA) will be
reported for Model 1.
Both GFI and AGFI are similar to correlations between the specified
model and the obtained data set. They each range from 1.0 (perfect relation)
to 0.0 (no relation— values close to .9 are considered a “ good fit” (R. E.
Schumacker & Lomax, 1996). Chi-square, AGFI and GFI cannot be used to
compare competing models, only to determine if the specific model explains
the observed data well.
In order to determine the better of the models, the third step was
utilization of the Root Mean Square Error of Approximation (RMSEA)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
46
(Browne & Cudeck, 1989; Steiger & Lind, 1980). This statistic provides an
index of fit, with 90% confidence intervals, that can be used to compare
competing models (McDonald & Marsh, 1990; R. E. Schumacker & Lomax,
1996). Scores on RMSEA range from 0.0 (perfect fit) to 1.0 (no fit). It is
predicted that model #2 (indirect effects model with path from Attachment to
Interpersonal Guilt) will score significantly better on this index. An
acceptable RMSEA is .05 with an upper bound of no greater than .08
(Browne & Cudeck, 1989).
Power Analysis
Numerous studies conclude that 100 to 150 subjects is the minimum
satisfactory sample size when conducting structural equation models (Ding,
Velicer, and Harlow, 1995; Anderson & Gerbing, 1988). Additionally, Bentler
and Chou (1987) report that five subjects per variable is sufficient for normal
distributions when the latent variables have multiple indicators and 10
subjects per variable is sufficient for other distributions.
Taking into account these suggested sample sizes and the proposed
fit criteria; the power analysis indicated that 200 subjects would be sufficient
to run the models for this study.
Missing-Data Imputation
Data were discarded for all respondents who completed less than
50% of the items on any given scale or subscale. Using this approach, two
participants were eliminated from the study because of missing data.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
47
Consequently, the original sample size of 203 was reduced to 201. For
participants with modest amounts of missing data (n = 22), the mean
replacement method wherein item means (rounded to their integer value) are
used to replace missing values was utilized (Afifi & Elashoff, 1966).
Path Constraints
Some of the paths in the model had to be constrained initially in order
to yield a just identified model— necessary to provide a unique set of
parameters for each latent variable relationship (R. E. Schumacker & Lomax,
1996). Thus, one path for each latent variable must be preset The following
path coefficients were set: Negative Affect to item 6 (.936); Positive Affect to
item 12 (.876); Somaticism to item 7 (.727); Interpersonal to item 9 (.879);
and, for the hierarchical model only, the variance of Depression was set to
1.0.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
48
CHAPTER 3
RESULTS
Correlations Between the Variables
Correlations were run between all manifest variables (Table 2). These
correlations do not imply any direction of effects of causality. Correlations
between variables of interest are as follows: Parental Divorce was
significantly correlated with several variables: Survival Guilt (.14, £ < .05),
Self-Hate Guilt (.20, p < .01), (less) Dependent Attachment (-.37, p < .001),
(more) Anxious Attachment (.21, p < .01) and (less) Separation Guilt (-.25, p
< .001). Parental Alcoholism was only significantly correlated with Survival
Guilt (. 18, p < .05). Survival Guilt was significantly correlated with (less)
Dependent Attachment (-.35, p < .001), (more) Anxious Attachment (.33, p <
.001), Dieting (.15, p < .05), and Oral Control (.17, p < .05). Separation Guilt
was significantly correlated with (more) Dependent Attachment (.21, p < .01).
Omnipotent Responsibility Guilt was significantly correlated with (less)
Dependent Attachment (-.19, p < .01) (more) Anxious Attachment (.27, p <
.001) and Dieting (.16, p < .05). Self-Hate Guilt was significantly correlated
with (less) Dependent Attachment (-.40, p < .001), (more) Anxious
Attachment (.48, p < .001), Dieting (.32, p < .001), Bulimia/Food
Preoccupation (.44, p < .001) and Oral Control (.22, p < .01). Close
Attachment showed no significant relationship with any of the indicators of
Eating Disorder Symptomatology. (Less) Dependent Attachment correlated
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49
significantly with Bulimia and Food Preoccupation (-.20, p < .01). (More)
Anxious Attachment correlated significantly with Bulimia and Food
Preoccupation (.18, p < .01). In sum, the strongest correlations were
between the Attachment variables and the Interpersonal Guilt variables.
Additionally, a relatively strong correlation between Self-Hate Guilt and
Bulimia and Food Preoccupation was noted.
Measurement Models
Evaluation of the full model must include an examination of the relative
strength of the measurement model. The manifest variables were expected
to be good indicators of their respective latent variables, as they were
derived directly from subscales that measured the respective latent
constructs. All 10 of the manifest variables appeared to be relatively good
indicators of the hypothesized latent constructs. Specifically, the loadings for
the indicator (manifest) variables ranged from a low of .31 for the Separation
Guilt scale on the Interpersonal Guilt latent variable to a high of .89 for the
Dieting subscale on the Eating Disordered Symptomatology latent variable.
Additionally, all of the paths between the latent and manifest variables were
significant at the .001 level. Prior use of these variables within structural
equation models has found values approximating those for the measurement
models utilized in the current study (see Hodson, Newcomb, Locke &
Goodyear, 2002; Mazzeo & Espelage, 2002).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
Path Models
Following inspection of the measurement models, the path models
were examined. A summary of model fit indexes can be found in Table 3.
Model 1 (Direct Effects Model)
Results for Model 1 (Figure 2) yielded a x2 = 49.47 (df = 7), p < .001
and RMSEA = .17 (90% confidence interval was .13 to .22). According to
criteria previously specified in this study, the data did not adequately fit this
model (as RMSEA upper bound was greater than .08 and x2was significant).
Model 2 (Hypothesized Model)
Results for Model 2 (Figures 3 and 4) yielded a x2 = 188.03 (df = 49),
g < .001, GFI = .87, AGFI = .79, and RMSEA = .12 (90% confidence interval
was .10 to .14). According to criteria previously specified in this study, the
data did not fit this model (as RMSEA upper bound was greater than .08 and
X 2was significant).
Model Comparisons
There was a difference between Model 1 and Model 2, indicating that
Model 2 was a better fit for the data. However the difference was not
significant. No overlap among the RMSEA confidence intervals would be
indicative of a significant difference (See Figure 5). Neither model fit the
goodness of fit criteria specified previously. As Model 2 was the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
51
Table 3
Summary of Model Fit Indexes for Models 1-4
Model
x2
df
X 2 /df £
G FI3 AGFIb RMSEAC
#1 49.473 7 7.07 < .001 .174
#2
188.03 49 3.84 < .001 .869 .792 .119
#3
20.430 19 1.07 <.370 .975 .952 .019
#4 116.14 26 4.47 < .001 .886 .803 .132
Note. Dashes indicate that the data were not applicable.
a GFI = Goodness of Fit Index. b AGFI = Adjusted Goodness of Fit Index.
C RMSEA = Root Mean Squared Error of Approximation.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
52
Figure 2a. Model 1 — Measurement model depicting first-order latent constructs.
Eating
Disordered
Symptomatology
.62
oral control
bulimia and food
Dreoccuoation
dieting
Figure 2b Model 1-Path model.
.13
Eating
Disordered
Symptomatology
.05
parental divorce
parental alcohol
abuse
*g < .05. **2 < 01. ***2 < 001.
Figure 2. Model 1— Measurement model depicting first-order latent
constructs and path model.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
53
.33
Eating
Disordered
Symptomatology
.89
68
-.71
Attachment
-.83
.61
.83
31
Interpersonal
Guilt .63
.73
.54
.49
.46
.20
.32
.90
.62
anxiety
survivor
depend
close
bulimia and food
oreoccuDation
separation
dieting
oral control
self-hate
Figure 3. Model 2— Measurement model depicting first-order latent constructs.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission.
parental divorce
parental alcohol
abuse
38***
-.09
.16'
00
*J 3 < .05. **£><.01. ***£<.001.
Figure 4. Model 2--Path model.
Eating
Disordered
Symptomatology
Interpersonal
Guilt
R M S E A V alue Range
55
0. 26
0. 24
0. 22
0. 2
0. 18 ■
0. 16
0.14-
0. 1 2 -
0. 1 -
0. 08 -
0. 06 -
0 .0 4 -
0. 02 -
1 1 ------
M odel 1 Model 2 Model 3
RMSEA Ranges Per Model
M odel 4
Figure 5. RMSEA model comparison.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
56
hypothesized model and did not meet the goodness of fit criteria, the source
of poor fit was sought in several ways.
Model Modifications
Given the problems with model specification, Model 2 was examined
to determine potential sources of the problems. First, it was noted that the
Attachment to Eating Disordered Symptomatology path coefficient was much
lower than expected based on theory and substantial empirical research
linking the two. This particular sample was ethnically diverse with
approximately one third of the sample consisting of women of Latin origin.
Much of the literature examining eating disorders and attachment is based on
Euro-American individuals (e.g., O’Keamy, 1996; Ward et al., 2000).
Additionally, it has been suggested that different ethnicities might present
differently in terms of eating disordered symptomatology (Altabe, 1998; le
Grange, Stone & Brownell, 1998) and may score differently on the EAT-26
(Abood & Chandler, 1997). Due to these considerations, t-tests were run to
determine if there were significant differences between ethnic groups on
these variables. No significant differences were found.
Next, the Interpersonal Guilt construct was examined. The
Interpersonal Guilt construct, although linked to Eating Disordered
Symptomatology in the clinical literature, has relatively limited support as a
correlate of Eating Disordered Symptomatology in empirical literature,
compared with the support for other hypothesized associations in the model.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
57
Consequently, hypotheses regarding Interpersonal Guilt were more
exploratory in nature than were other associations tested in the model.
However, Interpersonal Guilt seemed to be an important part of the model.
For as much as one can interpret “ significant” paths in models with
unacceptable fit: In Model 2, the Attachment construct had a significant
effect on Interpersonal Guilt and in turn, Interpersonal Guilt had a significant
effect on Eating Disordered Symptomatology. Additionally, several of the
Interpersonal Guilt indicators correlated significantly with the Eating Disorder
Symptomatology indicators. Based on these results as well as the
theoretical underpinnings of the model, and in order to represent the potential
importance of Interpersonal Guilt within the model, another model was
considered that featured a reciprocal relationship between Interpersonal Guilt
and Attachment and between Interpersonal Guilt and Eating Disordered
Symptomatology. This proposal contrasted with the original full model that
proposed Interpersonal Guilt as affected by Attachment and in turn affecting
Eating Disordered Symptomology (as represented by the single-headed
straight arrow in Model 2). However, it was discovered that this proposal
was not viable because it created a non-recursive model.
Therefore, a third model was proposed wherein Interpersonal Guilt
was dropped from the model and Attachment was retained in the model.
This decision was based partially on the lesser theoretical basis for retaining
Interpersonal Guilt and the greater theoretical basis for retaining Attachment,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
58
with the additional notion that because the two appear to be highly related,
they might be canceling each others’ effects. Following are the results.
Model 3 (Interpersonal Guilt Deleted)
Results for Model 3 (Figures 6 and 7) yielded a x2 = 20.43 (df = 19), £
< .37, GFI = .98, AGFI = .95, and RMSEA = .02 (90% confidence interval
was .00 to .07). According to the criteria previously specified in this study,
the data had an excellent fit for this model. However, as the path from
Attachment to Eating Disordered Symptomatology remained statistically
nonsignificant, it seemed prudent to test a fourth model to rule out the
possibility that Interpersonal Guilt had more of an effect on Eating Disorder
Symptomatology than Attachment. Following are the results from this fourth
and final model.
Model 4 (Attachment Deleted)
Results for Model 4 (Figures 8 and 9) yielded a x2 = 116.138 (df = 26),
2 < .001, GFI = .89, AGFI = .80, and RMSEA = .13 (90% confidence interval
was .11 to .16). According to the criteria previously specified in this study,
the data did not fit this model.
Model 3 Comparison and Interpretation
Models 1, 2, and 4 were rejected and Model 3 was accepted as a
reasonable reflection of the data. The lack of overlap between Model 3 and
any of the other models, as evidenced in Figure 5, indicates that Model 3 fits
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
59
.89
oral control
.32
Eating
Disordered
Symptomatology
.93
dieting
.65
bulimia and food
preoccupation
close
-.68
.21
/Attachment
depend
.56
anxiety
Figure 6. Model 3— Measurement model depicting first-order latent
constructs.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
60
C O
C O
o >
o
- C
o
o
a )
C O
_
C D.a
c
C D
a )
t o
Q _
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
61
bulimia and food
preoccupation
oral control
dieting
.33
Eating
Disordered
Symptomatology
.91
.66
.84
.37
interpersonal
Guilt .65
survivor
separation
omnipotent
responsibility
self-hate
.29
<3
.57
<3
Figure 8. Model A — Measurement model depicting first-order latent
constructs.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
62
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
63
the data significantly better than any of the other models. The original model
(Model 2) included the hypothesis that the effects of parental alcoholism and
divorce would indirectly affect eating disordered symptomatology through the
constructs of attachment and interpersonal guilt. An indirect effect is
operationalized as a relationship between the predictor and an intermediate
target variable, followed by a relationship between the intermediate target
variable and the criterion; no direct relationship exists between the predictor
and criterion (Holmbeck, 1997). Thus, Model 3 revealed indirect effects of
parental alcoholism on eating disordered symptomatology and parental
divorce on eating disordered symptomatology through attachment.
Additionally, the magnitude of the effect of parental divorce was stronger
than the magnitude of the effect of parental alcoholism.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
64
CHAPTER 4
DISCUSSION
In the last 20 years, research on the correlates of eating disordered
behaviors and symptomatology has increased remarkably. In spite of this,
questions about the etiology of these behaviors and symptoms still linger.
Many variables have been correlated with disordered eating, yet particular
processes through which these variables potentially affect disordered eating
remain largely unexplained. This is partially the result of reliance upon
univariate statistical approaches. The complexity of clinical and sub clinical
eating disorders demands multivariate analysis. This study utilized SEM to
explore some of the potential indirect relationships among variables
previously found to be associated with disordered eating.
Specifically, this study was an attempt to confirm attachment as an
intervening process for women with eating disordered symptoms, as well as
examine interpersonal guilt as an effect of insecure attachment and an
additional intervening process. The effect of parental divorce and parental
alcoholism on all of these variables was examined. To the knowledge of this
researcher, this was the first study to examine the family variables of parental
alcoholism and divorce, eating disordered symptomatology and attachment
utilizing structural equation modeling.
The major finding of this study is that parental divorce and parental
alcoholism were indirectly associated with disordered eating through
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
65
attachment and not through interpersonal guilt However, due to the
statistically nonsignificant Attachment to Eating Disordered Symptomatology
path, this must be considered a preliminary conclusion, in need of replication.
Additionally, based on the correlational analysis, Interpersonal Guilt appears
to have a relationship with Eating Disordered Symptomatology and a
relationship with Attachment. However, it is unclear how or if Interpersonal
Guilt belongs in this model. Therefore, the bulk of the conclusions will cover
the current best-fitting model (Model 3) from which Interpersonal Guilt was
deleted.
Summary of the Study
Correlations were run between Eating Disordered Symptomatology
and the other 9 variables in the study. There were no significant correlations
between Parental Divorce and Parental Alcoholism and any of the Eating
Disordered Symptomatology variables. Three of the Interpersonal Guilt
variables correlated significantly with Eating Disordered Symptomatology.
The strongest association was between Self-Hate Guilt and Bulimia and
Food Preoccupation. The Attachment variables showed a significant
relationship with Bulimia and Food Preoccupation. Additionally, all
Attachment variables were significantly correlated with Parental Divorce, but
only Dependent Attachment was significantly correlated with Parental
Alcoholism. Finally, many of the Attachment variables were associated with
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
66
the four aspects of Interpersonal Guilt, indicating a general relationship
between these two latent variables.
No direction of effects can be inferred from the correlational analysis.
Therefore, structural modeling was next used to determine directional effects
and indirect properties of Attachment and Interpersonal Guilt.
After a satisfactory measurement model was confirmed, two path
models were run and compared to determine the better fitting model. As
expected, Model 1, which proposed direct links between the family variables
of Parental Alcoholism and Parental Divorce and Eating Disordered
Symptomatology did not fit the data. Model 2 included all of the variables
and proposed that Parental Alcoholism and Parental Divorce have an indirect
effect on Eating Disordered Symptomatology through Attachment and
Interpersonal Guilt. Model 2 did not fit the data and was not significantly
different than Model 1, although it was a better fit.
Because the model of interest (Model 2) did not fit the data at
acceptable levels, this model was inspected to determine where the
problems might lie. Due to the low path coefficient between Attachment and
Eating Disordered Symptomatology, t-tests were run on ethnicity, attachment
and eating disordered symptomatology to see if ethnic differences were
present and possibly contributing to this low association. No differences
were found. Next, a model linking Interpersonal Guilt to Attachment and
Eating Disorder Symptomatology with curved arrows (indicating a reciprocal
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
67
relationship rather than unidirectional effects) was considered. This proposal
was rejected as it would have rendered the model non-recursive.
Another model was proposed wherein Interpersonal Guilt was deleted
from the model. Theoretical considerations were taken into account when
making the decision to delete Interpersonal Guilt. Interpersonal Guilt had the
least theoretical and empirical evidence for inclusion in the model.
Additionally, because of the high correlations between Attachment and
Interpersonal Guilt, it was believed that the effects of both might confound
the other. This model (Model 3) provided a very good fit for the data. To rule
out the possibility that Interpersonal Guilt might have more of an overall
effect than Attachment, a fourth and final model (Model 4) was run wherein
Attachment was deleted from the model. This model was not a good fit for
the data and was therefore rejected.
Model 3 Discussion
The results indicated that Model 3 fit the data well and that the
hypothesized relationships among the variables were in the expected
directions. Parental Divorce and Parental Alcoholism had an indirect effect
on Eating Disorder Symptomatology through Attachment. Although the path
from attachment to eating disorder symptomatology approached significance
in this model, it did not quite reach statistical significance. A discussion of
the clinical usefulness of Model 3 as a whole and speculation regarding
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
68
potential reasons for the low value of the path between Attachment and
Eating Disordered Symptomatology follows.
Clinical Implications
The type of model examined in the present study may be especially
relevant for practitioners. Research that attempts to link parental alcoholism
or parental divorce directly to eating disordered symptomatology gives little
direction for clinical intervention. However, conceptualizations that include
attachment as a mediating process can be powerful tools for work with
individuals with eating disorder symptoms.
Utilization of attachment-related conceptualizations aids practitioners
in understanding client history, anticipating client reactions and behaviors,
informing therapeutic stance, and formulating interventions. In these ways,
therapists may help clients perceive and change dysfunctional working
attachment models that potentially have an impact on eating disorder
symptoms.
Although the most effective treatment for individuals with clinical and
sub-clinical eating disorders appears to be multi-modal (Richards et al.,
2000), it is widely recognized that a critical factor for the success of any
therapy is the formation of a basic relationship between client and therapist.
A crucial aspect of this relationship is the adoption of an empathic stance on
the part of the therapist (Rogers, 1951). An attachment perspective can
help the therapist to alter stance based upon knowledge and understanding
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
69
of what the client expects and needs in order to have an experience that is
healing or emotionally corrective. Thus, evaluation and consideration of a
client’s attachment style can provide the grounding for a multi-modal format.
Other important treatment components for clinical and sub-clinical eating
disorders such as cognitive-behavioral interventions, psychodynamic work,
family therapy, and group therapy can be utilized within the overarching
attachment framework.
As previously noted, those with eating disorders and sub-clinical
eating disorders are often difficult to treat, sometimes eliciting negative
transference reactions from therapists. An attachment perspective promotes
empathy on the part of the therapist because it reminds the therapist that
although it might be challenging, it is especially critical to form a strong bond
with these particular clients. Given a strong bonding experience with the
therapist, the client may in turn be able to be more accepting and empathic
toward herself. Rather than accepting a view of her problem as a rebellion,
(which is how eating disorders sometimes come to be perceived by families),
superficial attempt to meet societal norms (which may be how these
individuals are perceived and judged in the world outside of therapy), or
disease, the client may begin to view her eating disorder as a way in which
she has tried to maintain important attachment bonds with others.
Use of this perspective may help the client evaluate the quality of her
attachment bonds while simultaneously providing a person to whom she can
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
70
become securely attached.. Thus a healing context and a place of safety for
continued work on other aspects of the problem is provided.
Unexpected Findings
Overall, Attachment in this model had a small and nonsignificant effect
on Eating Disorder Symptomatology. This finding is at odds with much of the
current literature that reveals a strong relationship between insecure
attachment styles and eating disordered symptoms and behaviors.
In exploring potential reasons for this unexpected finding, it should be
noted that the sample for this study was particularly ethnically diverse and
much of the prior research on eating disorders and correlates has been
conducted with Euro-American individuals (Streigel-Moore & Cachelin,
2001). Specifically, to the best knowledge of this researcher, no eating
disorder research evaluating ethnicity, attachment and eating disorders has
been conducted. Due to the dearth in this area, it is somewhat difficult to
speculate about the results of the current study with relation to ethnicity.
However, some theoretical and empirical considerations regarding
attachment and ethnicity are available and important to take into account
when interpreting the results of this study. These will provide the basis for
the following speculations.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
71
Attachment and Ethnicity
It has been suggested that some ethnic minority individuals may hold
some particular beliefs and values that might impact perception and
experience of attachment relationships. For example, respect for hierarchy
within intergenerational relationships is an important value for many ethnic
minority individuals (Rastogi & Wampler, 1999). The traditional Western
conceptualization of attachment neglects to measure aspects such as this,
and the attachment measure utilized in this study did not measure values
relating to hierarchy. Components such as this may have bearing on
insecure versus secure attachment patterns that in turn might affect
development of eating disorder symptomatology in unexplored ways.
Additionally, it has been suggested that minority students on mostly
White college campuses may experience an adaptive “ cultural paranoia”
which reflects a heightened sense of cultural mistrust (Terrell & Terrell,
1981). This may elicit an orientation toward greater avoidance of intimacy in
relations. The avoidant aspect of attachment is generally considered
“insecure,” but may reflect adaptive functioning in some ethnic minorities,
given certain contexts. Thus, factors such as this may be less likely to show
association with pathology (in this case, eating disorder symptomatology).
Parental Divorce and Religion
Another unexpected finding was that Parental Divorce had a greater
overall effect on Eating Disordered Symptomatology through Attachment
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
72
than did Parental Alcoholism. The sample was largely Catholic which could
partially account for this finding. Parental divorce could have a greater
impact on those with Catholic upbringing and values versus those without.
Not only are relationships with parents disrupted, an entire value system
might be thrown into question. This could influence development of eating
disordered symptomatology which sometimes may emerge as a reaction to
chaos and attempt to regain control (Gordon, 1990).
Unfortunately, due to sample size, a multiple groups analysis of the
models was not possible. Therefore, the potential effects of ethnicity and
religion on the full models were not examined.
Limitations
These findings should be interpreted with caution. Given the single
time-point assessments of this study, no direct cause-effect inferences
concerning the key variables are warranted. The results may only be
considered as a tentative statement describing the reality of the moment.
Because the sample was restricted to female students attending
Western colleges, generalization of the findings to other groups, institutions,
or regions must necessarily be tentative. Additionally, validity is limited by
the subjects’ honesty in the self-report measures, and internal and external
validity was limited to the instruments used in the study.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
73
Last, a family systems theoretical framework was used for this study,
thus many potentially important aspects of other theories regarding eating
disordered symptomatology were not covered.
Future Directions
There are several considerations for future research in this area.
First, to reduce the probability of specification error, future investigations
should include other variables that are particularly relevant to the criteria
(Kline, 1998). These variables might include additional potential loss and
attachment-related variables, such as death of a parent or mental illness of a
parent. Depression, family dysfunction and abuse history may be other
important variables for inclusion.
Second, although the sample was quite diverse in terms of race and
ethnicity, the data analysis did not look at different racial/ethnic groups. Due
to sample size, a multiple group analysis by ethnic group was not possible.
The study would benefit from a larger sample size and an analysis of data by
ethnic groups, particularly since little research could be found in the area of
attachment, eating disorders and ethnicity. If this task is undertaken,
variables that have shown a correlation with eating disordered
symptomatology in ethnic minority samples and that might also impact
attachment style should be included. These may include acculturation,
discrimination, and immigration issues.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(A
Third, the parental divorce variable should include salient aspects of
the divorce experience. It may be that aspects of the experience of parental
divorce are the critical elements, rather than simply the fact that one’s
parents have divorced. For example, contact with nonresidential parent,
relationship with both parents, and age of the individual when parents
divorced have been shown to be important aspects pertaining to the divorce
experience for children (Laumann-Billings & Emery, 2000). These might be
included in future studies.
Fourth, expansion of the parental alcoholism variable might yield
important information. In particular, gender of the alcoholic parent may be an
important consideration, based on much theoretical work implicating
maternal relationships as critical to the development of eating disorders (e.g.,
Chemin, 1985). Other types of parental substance abuse might also be
important factors to consider.
Fifth, the study should be replicated in a clinical population. The use
of a nonclinical sample population in this research is both an asset and a
limitation. Use of a nonclinical population was important because eating
disordered symptomatology occurs on a continuum (Stice, Killen, Hayward, &
Taylor, 1996; Tylka & Subich, 1999) and it is clear that those with sublinical
eating disorders may experience significant amounts of distress (Mintz &
Betz, 1988; Dancyger & Garfinkel, 1995; Tylka & Subich, 1999). In addition,
investigation of problematic eating behaviors is consistent with the goals of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
75
counseling psychology that include working with people at all levels of
functioning. However, the findings of this study cannot be generalized to
individuals with clinical eating disorders as these individuals may represent a
different group. Therefore, future research with a model such as this should
examine clinical samples.
Sixth, although structural equation modeling is often called “ causal
modeling,0 no causal inferences may be drawn from the cross-sectional data
obtained in this study. Carefully designed longitudinal studies may replicate
and extend the results of this study so that causal inferences may be made.
Finally, although the Interpersonal Guilt variable was removed from
the final model, it may still be an important concept connected to eating
disordered symptomatology. Several of the Interpersonal Guilt variables
correlated significantly with several of the indicators of Eating Disordered
Symptomatology. Additionally, Attachment and Interpersonal Guilt appeared
to be related. Perhaps the development of disordered eating promotes and
perpetuates the development of interpersonal guilt. Future studies may
explore this relationship.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
76
BIBLIOGRAPHY
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
77
BIBLIOGRAPHY
Abood, D. A., & Chandler, S. B. (1997). Race and the role of weight,
weight change, and body dissatisfaction in eating disorders. American
Journal of Health Behavior. 21 (1). 21-25.
Afifi, A., & Elashoff, R. M. (1966). Missing Observations in
multivariate statistics. Part 1. Review of the literature. Journal of the
American Statistical Association. 61. 595-604.
Ainsworth, M., Blehar, M., Waters, E., & Wall, S. (1978). Patterns of
attachment: A psychological study of the Strange Situation. Hilldale, NJ:
Lawrence Erlbaum.
Altabe, M. (1998). Ethnicity and body image: Quantitative and
qualitative analysis. International Journal of Eating Disorders, 23(2). 153-
159.
American Psychiatric Association. (1994). Diagnostic and statistical
manual of mental disorders (4th ed.). Washington, DC: Author.
Anderson, J. C., & Gerbing, D. W. (1988). Structural equation
modeling in practice: A review and recommended two step approach.
Psychological Bulletin. 103. 411 -423.
Arbuckle, J. L. (1997). Amos (Version 3.61-W 32). Chicago: Small
Waters Corporation.
Ausubel, D. P. (1955). Relationships between shame and guilt in the
socializing process. Psychological Review. 62. 378-390.
Balk, D. (1983). How teenagers cope with sibling death: Some
implications for school counselors. School Counselor. 31 (2). 150-158.
Banks, C. G. (1992). “ Culture” in culture-bound syndromes: The
case of anorexia nervosa. Social Science & Medicine. 34(8). 867-884.
Bartholomew, K., & Horowitz, L. M. (1991). Attachment styles among
young adults: A test of a four-category model. Journal of Personality and
Social Psychology. 61(2), 226-244.
Baumeister, R. F., Stillwell, A M., & Heatherton, T. F. (1994). Guilt:
An interpersonal approach. Psychological Bulletin. 115. 243-267.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
78
Bentler, P. M., & Chou, C. (1987). Practical issues in structural
modeling. Sociological Methods and Research. 16. 78-117.
Boszormenyi-Nagy, I., & Spark, G. M. (1973). Invisible loyalties:
Reciprocity in interqenerational family therapy. New York: Harper & Row.
Boumann, C. E., & Yates, W. R. (1994). Risk factors for bulimia
nervosa: A controlled study of parental psychiatric illness and divorce.
Addictive Behaviors. 19(6). 667-675.
Bowlby, J. (1969). Attachment: Vol 1. Attachment and loss.
London: Penguin.
Bowlby, J. (1973). Separation: Vol. 2. Attachment and loss.
London: Penguin.
Bowlby, J. (1980). Loss, sadness and depression: Vol. 3.
Attachment and loss. New York: Basic Books.
Bowlby, J. (1988). A secure base: Parent-child attachment and
healthy human development. New York: Basic Books.
Breen, J. (1985). Children of alcoholics: The subterranean grieving
process. Psychotherapy Patient 2(1), 85-94.
Brennan, K. A., & Shaver, P. R. (1993). Attachment styles and
parental divorce. Journal of Divorce and Remarriage. 21. 161 -175.
Brennan, K. A., & Shaver, P. R. (1998). Attachment styles and
personality disorders: Their connections to each other and to parental
divorce, parental death, and perceptions of parental caregiving. Journal of
Personality. 66(5). 835-878.
Brennan, K. A., Shaver, P., &Tobey, A. (1991). Attachment styles,
gender and parental problem drinking. Journal of Social and Personal
Relationships, 8. 451-466.
Bretherton, I. (1985). Attachment theory: Retrospect and prospect.
In I. Bretherton & E. Waters (Eds.), Growing points in attachment theory and
research. Monographs of the Society for Research in Child Development. 50
(1-2, Serial No. 209), 3-35.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
79
Browne, M. W., & Cudeck, R. (1989). Single sample cross-validation
indices for covariance structures. Multivariate Behavioral Research. 24. 445-
455.
Bruch, H. (1978). The golden cage. Cambridge, MA: Harvard
University Press.
Bulik, C. M. (1987). Drug and alcohol use by bulimic women and
their families. American Journal of Psychiatry. 144(121. 1604-1606.
Bush, M. (1989). The role of unconscious guilt in psychopathology
and psychotherapy. Bulletin of the Menninaer Clinic. 53(21. 97-107.
Button, E. J., & Whitehouse, A. (1981). Sub-clinical anorexia
nervosa. Psychological Medicine. 11. 509-516.
Bybee, J., Zigler, E., Berliner, D., & Merisca, R. (1996). Guilt, guilt-
evoking events, depression and eating disorders. Current Psychology:
Developmental. Learning. Personality. Social. 15(2). 113-127.
Candelori, C., & Ciocca, A. (1998). Attachment and eating disorders.
In Psychotherapeutic issues in eating disorders: Models, methods, and
results (139-153). Rome: Societa Editrice Universo.
Canziani F. (1996). Possible children’s reactions to their parents’
divorce. Giomale di Neuropsichiatria dell’Eta Evolutiva. 16(4). 255-282.
Carpenter, D. R. (1995). Adult children of alcoholics: CAQ profiles.
Alcoholism Treatment Quarterly. 13(2). 63-70.
Carrol, S. (1991). Personality characteristics of adult children of
alcoholics (Doctoral dissertation, Southern Illinois University at Carbondale,
1989). Dissertation Abstracts International. 51. 3556B.
Chandy, J. M., Harris, L., Blum, R., & Resnick, M. D. (1994).
Disordered eating among adolescents whose parents misuse alcohol:
Protective and risk factors. International Journal of Addictions. 29(4). 505-
516.
Chassler, L. (1997). Understanding anorexia nervosa and bulimia
nervosa from an attachment perspective. Clinical Social Work Journal. 25.
407-423.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
80
Chemin, K. (1985). The hungry self: Women, eating and identity.
New York: Time Books.
Clair, K. J., & Genest, M. (1992). The children of alcoholics screening
test: Reliability and relationship to family environment, adjustment, and
alcohol-related stressors of adolescent offspring of alcoholics. Journal of
Clinical Psychology. 48(3). 414-420.
Claydon, P. (1987). Self-reported alcohol, drug, and eating-disorder
problems among male and female collegiate children of alcoholics. Journal
of American College Health. 36(2). 111-116.
Clopton, J. R., Haas, H. L., Kent, J. S., & Robert-McComb, J. J.
(2001). Eating disorders in women and children: Prevention, stress
management, and treatment. Boca Raton, FL: CRC Press, Inc.
Collins, N. L., & Read, S. J. (1990). Adult attachment, working
models, and relationship quality in dating couples. Journal of Personality and
Social Psychology. 58(4). 644-663.
Crago, M., Shisslak, C. M., & Estes, L. S. (1996). Eating
disturbances among American minority groups. International Journal of
Eating Disorders. 19. 239-248.
Crittenden, P. M. (1988). Relationships at risk. In J. Belsky & T.
Nezworski (Eds.), Clinical implications of attachment. Hillsdale, NJ:
Lawrence Erlbaum.
Cuijpers, P., Langendoen, Y., & Bijl, R. V. 1999). Psychiatric
disorders in adult children of problem drinkers: Prevalence, first onset and
comparison with other risk factors. Addiction. 94 (10). 1489-1498.
Dancyger, I. F., & Garfinkel, P. E. (1995). The relationship of partial
syndrome eating disorders to anorexia nervosa and bulimia nervosa.
Psychological Medicine. 25. 1019-1025.
Diehl, M., Elnick, A. B., Bourbeau, L. S., & Labouvie-Vief, G. (1998).
Adult attachment styles: Their relations to family context and personality.
Journal of Personality & Social Psychology, 74(6L 1656-1669.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
81
Ding, L., Velicer, W. F., & Harlow, L. L. (1995). Effects of estimation
methods, number of indicators per factor, and improper solutions on
structural equation modeling fit indices. Structural Equation Modeling. 2.
119-143.
Dinning, W. D., & Berk, L. A. (1989). The children of alcoholics
screening test: Relatioship to sex, family environment, and social adjustment
in adolescents. Journal of Clinical Psychology. 45. 335-339.
Egeland, B., & Sroufe, L. A. (1981). Attachment and early
maltreatment. Child Development. 52(1), 44-52.
el-Guebaly, N., West, M., Maticka-Tyndale, E., & Cohn, D. A. (1993).
Attachment among adult children of alcoholics. Addiction. 8 8 .1405-1411.
el-lslam, F. M. (1969). Depression and guilt: A study at an Arab
psychiatric clinic. Social Psychiatry. 4(2). 56-58.
Engel, B. (1990). The emotionally abused woman: Overcoming
destructive patterns and reclaiming yourself. New York: Ballantine Books.
Escherick, M., O’Connor, L. E., Berry, J., & Weiss, J. (1999). The
role of quilt in obsessions and compulsions. Unpublished manuscript.
Evans, J. J., & Bloom, B. L. (1996). Effects of parental divorce
among college undergraduates. Journal of Divorce and Remarriage. 26. 69-
91.
Evans, L., ScWertheim, E.H. (1998). Intimacy patterns and
relationship satisfaction of women with eating problems and the mediating
effects of depression, train anxiety and social anxiety. Journal of
Psychosomatic Research. 44. 355-365.
Fassinger, R. E. (1986). Use of structural equation modeling in
counseling psychology research. Journal of Counseling Psychology. 34(4).
425-436.
Feeney, J. A., & Noller, P. (1990). Attachment style as a predictor of
adult romantic relationships. Journal of Personality and Social Psychology.
58, 281-291.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
82
Friedberg, N. L., & Lyddon, W. J. (1996). Self-other working models
and eating disorders. Journal of Cognitive Psychotherapy: An International
Quarterly. 1 0 .193-203.
Friedman, M. (1985). Survivor guilt in the pathogenesis of anorexia
nervosa. Psychiatry. 48, 25-39.
Gandour, M. J. (1984). Bulimia: Clinical description, assessment,
etiology, and treatment. International Journal of Eating Disorders. 3. 3-38.
Gamer, D. M., & Garfinkel, P. (1979). The Eating Attitudes Test: An
index of the symptoms of anorexia nervosa. Psychological Medicine. 9. 273-
279.
Gamer, D. M., Garfinkel, P., & O’Shaughnessy, M. (1985). The
validity of the distinction between bulimics with and without anorexia.
American Journal of Psychiatry, 142, 581 -587.
Gamer, D. M., Olmsted, M. P., Bohr, Y., & Garfinkel, P. E. (1982).
The Eating Attitudes Test: Psychometric features and clinical correlates.
Psychological Medicine. 12. 871 -878.
Glover, G. (1994). The hero child in the alcoholic home:
Recommendations for counselors. School Counselor. 41(3). 185-190.
Goodsitt, A. (1997). Eating disorders: A self-psychological
perspective. In D. M. Gamer & P. E. Garfinkel (Eds.), Handbook of
Treatment for Eating Disorders (pp.205-228). New York: The Guilford
Press.
Gordon, R. A. (1990). Anorexia and bulimia: Anatomy of a social
epidemic. Cambridge, MA: Basil Blackwell, Inc.
Gremillion, H. (1992). Psychiatry as social ordering: Anorexia
nervosa, a paradigm. Social Science Medicine. 35(1). 57-71.
Griffin, D. W., & Bartholomew, K. (1994). Models of the self and
other Fundamental dimensions underlying measures of adult attachment.
Journal of Personality & Social Psychology. 67(3). 430-445.
Halmi, K. A , Falk, J. R., & Schwartz, E. (1981). Binge-eating and
vomiting: A survey of a college population. Psychological Medicine. 11. 697-
706.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
83
Harder, D. W .t Cutler, L., & Rockart, L. (1992). Assessment of shame
and guilt and their relationships to psychopathology. Journal of Personality
Assessment. 59(3). 584-604.
Harder, D. W., & Greenwald, D. F. (1999). Further validation of the
shame and guilt scales of the Harder Personal Feelings Questionnaire-2.
Psychological Reports. 85(1). 271-281.
Hart, K.J., & Ollendick, T. H. (1985). Prevalence of bulimia in working
and university women. American Journal of Psychiatry. 142. 851-854.
Hazen, C., & Shaver, P. R. (1987). Romantic love conceptualized as
an attachment process. Journal of Personality and Social Psychology. 59.
270-280.
Hodson, C., Newcomb, M. D., Locke, T. F., & Goodyear, R. K. (2002)
Polv-substance problems and disordered eating in maltreated adolescent
Latinas: Mediated and indirect paths in a community sample. Manuscript
submitted for publication.
Holden, N. L., & Robinson, P. H. (1988). Anorexia and bulimia
nervosa in British Blacks. British Journal of Psychiatry. 152. 544-549.
Holmbeck, G. N. (1997). Toward terminological, conceptual, and
statistical clarity in the study of mediators and moderators: Examples from
the child-clinical and pediatric psychology literatures. Journal of Consulting &
Clinical Psychology. 65(4). 599-610.
Holmbeck, G. N., & Hill, J. (1986). A path-analytic approach to the
relations between parental traits and acceptance and adolescence
adjustment. Sex Roles. 14(5-6). 315-334.
Hudson, J. I., Pope, H. G., Jonas, J. M., & Yurgelun-Todd, D. (1983).
Family history study of anorexia and bulimia. British Journal of Psychiatry.
142. 133-138.
Humphrey, L. L. (1986). Structural analysis of parent-child
relationships in eating disorders. Journal of Abnormal Psychology. 95. 395-
402.
Humphrey, L. L. (1988). Relationships within subtypes of anorexic,
bulimic, and normal families. Journal of the American Academy of Child and
Adolescent Psychiatry. 27, 544-551.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
84
Humphrey, L. L. (1989). Observed family interactions among
subtypes of eating disorders using structural analysis of social behavior.
Journal of Consulting and Clinical Psychology. 57. 206-214.
Igoin-Apfelbaum, L. & Apfelbaum, M. (1990). Incidence of broken
home family background in bulimia. British Reiview of Bulimia & Anorexia
Nerygsa,_4(2)! 55-59
Jones, A. D. (1985). Anorexia nervosa, bulimia and alcoholism:
Association of eating disorder and alcohol. Journal of Psychiatric Research.
19(2-3), 377-380.
Jones, J. W. (1981). The Children of Alcoholics Screening Test
(CAST). Chicago: Family Recovery Press.
Jones, J. W. (1986). The Children of Alcoholics Screening Test
(CAST) manual. Chicago: Camelot Unlimited.
Joreskog, K. G., & Sorbom, D. (1989). LISREL 7 user's reference
guide. Mooresville, IN: Scientific Software, Inc.
Kanter, R. A., Williams, B. E., & Cummings, C. (1992). Personal and
parental alcohol abuse, and victimization in obese binge eaters and
nonbingeing obese. Addictive Behaviors. 17(5) 439-445.
Kassett, J. A., Gershon, E. S., Maxwell, M. E., Guroff, J. J., Kazuba,
D. M., Smith, A. L., Brandt, H. A., & Jimerson, D. C. (1989). Psychiatric
disorders in the first-degree relatives of probands with bulimia nervosa.
American Journal of Psychiatry. 146. 1468-1471.
Kenny, M. E. (1987). The extent and function of parental attachment
among first-year college students. Journal of Youth and Adolescence. 16.
17-29.
Kenny, M. E. (1990). College seniors’ perceptions of parental
attachments: The value and stability of family ties. Journal of College
Student Development. 31. 39-46.
Kenny, M. E., & Donaldson, G. A. (1991). Contribution of parental
attachment and family structure to the social and psychological functioning of
first-year college students. Journal of Counseling Psychology, 38. 479-486.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
85
Killer), J. D., Taylor, C. B., Hayward, C., Haydel, K. F., Wilson, D. M.,
Hammer, L., Kraemer, H., Blair-Greiner, A., & Strachowski, D. 1996).
Weight concerns influence the development of eating disorders: A 4-year
prospective study. Journal of Consulting and Clinical Psychology. 6 4 (5).
936-940.
Kline, R. B. (1998). Principles and practice of structural equation
modeling. New York: Guilford Press.
Klosinski, G. (1993). Psychological maltreatment in the context of
separation and divorce. Child Abuse & Neglect. 17(4). 557-563.
Koslowsky, M., Scheinberg, Z., Bleich, A., Mark, M., Apter, A., Danon,
Y., & Solomon, Z. (1992). The factor structure and criterion validity of the
short form of the Eating Attitudes Test. Journal of Personality Assessment.
58,27-35.
Laumann-Billings, L„ & Emery, R. E. (2000). Distress among young
adults from divorced families. Journal of Family Psychology. 14(4). 671-687.
le Grange, D., Stone, A. A., & Brownell, K. D. (1998). Eating
disturbances in White and minority female dieters. International Journal of
Eating Disorders. 24(4). 395-403.
Lee, S. (1995). Reconsidering the status of anorexia nervosa as a
western culture-bound syndrome. Social Science & Medicine, 42(1). 21-34.
Leondari, A., & Kiosseoglou, G. (2000). The relationship of parental
attachment and psychological separation to the psychological functioning of
young adults. Journal of Social Psychology. 140(4). 451-464.
Levine, M. P. (1987). Student eating disorders: Anorexia nervosa
and bulimia. Washington, DC: National Education Association.
Lopez, F. G. (1991). The impact of parental divorce on college
students. New Directions for Student Services. 54, 19-33.
Lopez, F. G., Gover, M. R., Leskela, J., Sauer, E. M., Schirmer, L., &
Wyssmann, J. (1997). Attachment styles, shame, guilt, and collaborative
problem-solving orientations. Personal Relationships. 4. 187-199.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
86
Mahon, J., Bradley, N. S., Harvey, P. K., Winston, A. P. & Palmer, R.
L. (2001). Childhood trauma has dose-effect relationship with dropping out
from psychotherapeutic treatment for bulimia nervosa: A replication.
International Journal of Eating Disorders. 30. 138-148.
Main, M., & Cassidy, J. (1988). Categories of response with the
parent at age six: Predicted from infant attachment classifications and stable
over a one-month period. Developmental Psychology, 24. 415-426.
Main, M., & Solomon, J. (1990). Procedures for identifying infants as
disorganized/disoriented during the Ainsworth Strange Situation. In M. T.
Greenburg, D. Cicchetti, & E.M. Cummings (Eds.), Attachment in the
preschool years: Theory, research, and intervention. Chicago: University of
Chicago Press.
Marcus, A M., & Tisne, S. (1987). Perception of maternal behavior
by elementary school children of alcoholic mothers. International Journal of
Addictions. 22(6L 543-555.
Maxwell, S. (1985). Review of Children of Alcoholics Screening Test.
In J. V. Mitchell, Jr. (Ed.), The ninth mental measurement yearbook (Vol. 1,
pp. 307-308). Lincoln: University of Nebraska Press.
Mazzeo, S. E. (1999). Modification of an existing measure of body
image preoccupation and its relationship to disordered eating in female
college students. Journal of Counseling Psychology. 49(1). 42-50.
Mazzeo, S. E., & Espelage, D. L. (2002). Association between
childhood physical and emotional abuse and disordered eating behaviors in
female undergraduates: An investigation of the mediating role of alexithymia
and depression. Journal of Counseling Psychology. 49(1), 86-100.
McCarthy, G., & Taylor, A. (1999). Avoidant/ambivalent attachment
style as a mediator between abusive childhood experiences and adult
relationship difficulties. Journal of Child Psychology and Psychiatry. 40(3).
465-477.
McDonald, R. P., & Marsh, H. W. (1990). Choosing a multivariate
model: Noncentrality and goodness of fit. Psychological Bulletin. 107. 247-
255.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
87
Meehan, W., O’Connor, L. E., Berry, J. W., Morrison, A., & Alcompora,
A. (1996). Guilt, shame and depression in clients in recovery from addiction.
Journal of Psvchoactive Drugs. 28(21 125-134.
Menaker, A. (1995). The relationship between attributional style and
interpersonal quilt. Unpublished doctoral dissertation, California School of
Professional Psychology, Alameda, CA.
Meyer, D. F. (1997). Codependency as a mediator between stressful
events and eating disorders. Journal of Clinical Psychology. 53(2). 107-116.
Meyer, D. F., & Russell, R. K. (1998). Caretaking, separation from
parents, and the development of eating disorders. Journal of Counseling and
Development. 76(2)166-173.
Mickelson, K. D., Kessler, R. C., & Shaver, P. R. (1997). Adult
attachment in a nationally representative sample. Journal of Personality and
Social Psychology. 73. 1092-1106.
Mikulincer, M., & Florian, V. (1999). The association between
parental reports of attachment style and family dynamics, and offspring’s
reports of adult attachment style. Family Process. 38(2). 243-257.
Mintz, L. B., & Betz, N. E. (1988). Prevalence and correlates of eating
disordered behaviors among undergraduate women. Journal of Counseling
Psychology. 44. 63-79.
Mintz, L. B., Kashubeck, S., & Tracy, L. S. (1995). Relations among
parental alcoholism, eating disorders, and substance abuse in nonclinical
college women: Additional evidence against the uniformity myth. Journal of
Counseling Psychology. 42(1). 65-70.
Minuchin, S. (1974). Families and family therapy. Cambridge, MA:
Harvard University Press.
Minuchin, S., Rosman, B. L., & Baker, L. (1978). Psychosomatic
families: Anorexia nervosa in context. Harvard University Press.
Mitchell, J. E., Hatsukami, D., Eckert, E. D., & Pyle, R. L. (1985).
Characteristics of 275 patients with bulimia. American Journal of Psychiatry.
142(4), 482-485.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
88
Modell, A. H. (1971). The origin of certain forms of pre-oedipal guilt
and the implications for a psychoanalytic theory of affects. International
Journal of Psychoanalysis. 52. 337-346.
Mothersead, P. K., Kivlighan, D. M., & Wynkoop, T. F. (1998).
Attachment, family dysfunction, parental alcoholism, and interpersonal
distress in late adolescence: A structural model. Journal of Counseling
Psychology. 45(21. 196-203.
Mullholland, A. M., & Mintz, L. B. (2001). Prevalence of eating
disorders among African American women. Journal of Counseling
Psychology. 48(1). 111-116.
Murphy, J. (1984). Substance abuse and the family. Journal for
Specialists in Group Work. 9(2). 106-112.
Murray, S. H., Touyz, S. W., & Beumont, P. J. V. (1996). Awareness
and perceived influence of body ideals in the media: A comparison of eating
disorder patients and the general community. Eating Disorders. 4(1), 33-46.
Mussell, M. P., Binford, R. B., & Fulkerson, J. A. (2000). Eating
disorders: Summary of risk factors, prevention programming, and prevention
research. The Counseling Psychologist. 28. 764-796
Neiderland, W. G. (1981). The survivor syndrome: Further
observations and dimensions. Journal of American Psychoanalytic
Association. 29(2). 413-425.
Neumaerker, K. (2000). Mortality rates and causes of death.
European Eating Disorders Review. 8(2). 181-187.
Newcomb, M. D. (1990). What structural modeling techniques can
tell us about social support. In I. G. Sarason, B. R. Sarason, & G. R. Pierce
(Eds.), Social Support: An International View (pp. 26-62). New York: Wiley.
Newcomb, M. D. (1992). Understanding the multidimensional nature
of drug use and abuse: The role of consumption, risk factors, and protective
factors. In M. D. Glantz & R. Pickens (Eds.), Vulnerability to drug abuse (pp.
255-297). Washington DC: American Psychological Association.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
89
Newcomb, M. D. (1994). Drug use and intimate relationships among
women and men: Separating specific from general effects in prospective data
using structural equation models. Journal of Consulting and Clinical
Psychology. 62(3). 463-476.
Newcomb, M. D., & Bentler, P. M. (1988). Consequences of
Adolescent Drug Use: Impact on the Lives of Young Adults. Beverly Hills,
CA: Sage.
O’Connor, L. E., & Berry, J. W. (1999). Interpersonal guilt, shame,
and psychological problems. Journal of Social and Clinical Psychology.
18(2), 181-203.
O’Connor, L. E., Berry, J. W., Inaba, D., Weiss, J. & Morrison, A.
(1994). Shame, guilt, and depression in men and women in recovery from
addiction. Journal of Substance Abuse Treatment. 11(6). 503-510.
O’Connor, L. E., Berry, J. W., Weiss, J., Bush, M., & Sampson, H.
(1997). Interpersonal guilt The development of a new measure. Journal of
Clinical Psychology. 53(1), 73-89.
O’Connor, L. E., Berry, J. W., Weiss, J., & Gilbert, P. (1999). Guilt,
fear, submission and empathy in college students and clinically depressed
patients. Unpublished manuscript.
O’Connor, L. E., Berry, J. W., Weiss, J., Herbold, J., Meehan, W., &
Webster, R. (1996). Interpersonal quilt and psychopathology: Development
of a new measure. Poster presented at the meetings of the California
Psychological Association, San Diego, California.
O’Connor, M. J., Sigman, M., & Brill, N. (1987). Disorganization of
attachment in relation to maternal alcohol consumption. Journal of
Consulting and Clinical Psychology. 55. 831-836.
O’Halloran, M. S., & Mintz, L. B. (1997). Evaluating EAT diagnosis of
eating disorders in a nonclinical sample. Paper presented at the 105
Annual Convention of the American Psychological Association, Chicago.
O’Keamey, R. (1996). Attachment disruption in anorexia nervosa and
bulimia nervosa: A review of theory and empirical research. International
Journal of Eating Disorders. 20(2). 115-127.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
90
Orenstein, A., Davis, R. B., & Wolfe, H. (1993). Parental substance
abuse treatment and adolescent problems. Journal of Alcohol & Drug
Education. 38(2). 5-61.
Pfaller, J., Gerstein, L., & Kiselica, M. (1998). Attachment style and
family dynamics in young adults. Journal of Counseling Psychology. 45(3).
353-357.
Pike, K. M., & Rodin, J. (1991). Mothers, daughters, and disordered
eating. Journal of Abnormal Psychology, 100(2). 198-204.
Pilat, J. M., & Jones, J. W. (1985). Identification of children of
alcoholics: Two empirical studies. Alcohol Health and Research World. 9.
27-33.
Pook, R., & Conti, T., Lester, D. (1996). Eating disorders,
depression and suicidal preoccupation in a nonclinical sample.
Psychological Reports. 79(1), 302.
Rastogi, M., & Wampler, K. (1999). Adult daughters’ perceptions of
the mother-daughter relation cross-cultural comparison. Family Relations:
Interdisciplinary Journal of Applied Family Studies. 48. 336.
Richards, P. S., Baldwin, B. M., Frost, H. A., Clark-Sly, J. B., Berrett,
M. E. & Hardman, R. K. (2000). Eating disorders. The Journal of Treatment
and Prevention. 8(3). 189-206.
Rieger, E., Touyz, S. W., Swain, T., 8c Beumont, P. J. V. (2001).
Cross-cultural research on anorexia nervosa: Assumptions regarding the
role of body weight. International Journal of Eating Disorders. 29(2). 205-
215.
Robinson, P., 8c Andersen, A E. (1985). Anorexia nervosa in
American Blacks. Journal of Psychiatric Resarch. 19(2-3). 183-188.
Rodin, J., Silberstein, L. R., 8c Striegel-Moore, R. H. (1985). Women
and weight: A normative discontent. In T. B. Sonderegger (Ed.), Nebraska
Symposium on Motivation: Psychology and Gender (Vol. 32, pp. 267-307).
Lincoln, NE: University of Nebraska Press.
Rogers, C. (1951). Client-centered therapy: Its current practice.
implications, and theory. Oxford, England: Houghton Mifflin.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
91
Roosa, M. W ., Sandler, I. N., Gehring, M., Beals, J., & Capp, L.
1988). The Children of Alcoholics Life-Events Schedule: A stress scale for
children pf alcoholic-abusing parents. Journal of Studies on Alcohol. 49.
422-429.
Sands, R., Tricker, J., Sherman, C., Armatas, C., & Maschette, W.
(1997). Disordered eating patterns, body image, self-esteem, and physical
activity in preadolescent school children. International Journal of Eating
Disorders. 21(2). 159-166.
Schneider-Rosen, K., Braunwald, K. G., Carlson, V., & Cicchetti, D.
(1985). Current perspectives in attachment tfiepry: Illustration from the study
of maltreated infants. Monographs of tf]e ^ocietv for Research in Child
Development. 50(1-2). 194-210.
Schumaker, R. E., & Lomax, R. G. (1996). A beginner's guide to
structural eguation modeling. Mahwah, NJ: L a w re n c e Erlbaum.
Schumaker, J. F., Warren, W. G., Carr, S. C., Schreiber, G. S., &
Jackson, C. C. (1995). Dissociation and expression in eating disorders.
Social Behavior and Personality. 23(1). 53-58.
Sharpe, T. M., Ki|len, J. D., Bryson, S. W., Shisslak, C. M., Estes, L.
S., Gray, N., Crago, M., & Taylor, C. B. (1998). Attachment style and weight
concerns in preadolescent and adolescent girls. International Journal of
Eating Disorders. 23. 39-44.
Shisslak, C. M., Crago, M., McKnight, K. M., Estes, L. S., Gray, N., &
Pamaby, 0 . G. (1998). Potential risk factors associated with weight control
behaviors in elementary and middle school girls. Journal of Psychosomatic
Research. 44. 301 -313.
Silber, T. J. (1986). Anorexia nervosa in blacks and Hispanics.
International Journal of Eating Disorders. 5(1). 121-128.
Smolak, L., & Levine, M. P. (1993). Separation-individuation
difficulties and the distinction between bulimia nervosa and anorexia nervosa
in college women. International Journal of Eating Disorders, 14(1), 33-41.
Soler, M., Bargada, M., Molina, M., Bassas, N., & Vilaltella, J. (1996).
Psychopathological aspects of divorce. Revista de Psiguiatria Infanto-
Juvenil. 4. 253-262.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
92
Sprecher, S., Cate, R., & Levin, L. (1998). Parental divorce and
young adults’ beliefs about love. Journal of Divorce and Remarriage. 28(3-
4), 107-120.
SPSS, Inc. (2000). SPSS for Windows (Version 10.00). New York:
Author.
Staley D, & el-Guebaly N: (1991). Psychometric evaluation of the
Children of Alcoholics Screening Test (CAST) in a psychiatric sample.
International Journal of Addictions. 26. 657-668.
Steiger, J. H., & Lind, J. M. (1980). Statistically-based models tests
for the number of common factors. Paper presented at the Psychometric
Society Meeting, Iowa City, IA.
Stice, E., Killen, J. D., Hayward, C., & Taylor, C. B. (1998). Support
for the continuity hypothesis of bulimic pathology. Journal of Consulting &
Clinical Psychology. 66(5). 784-790.
Stice, E., Ziemba, C., Margolis, J., & Flick, P. (1996). The dual
pathway model differentiates bulimics, sub-clinical bulimics, and controls:
Testing the continuity hypothesis. Behavior Therapy. 27. 531-549.
Streigel-Moore, R. H. (1995). Psychological factors in the etiology of
binge eating. Addictive Behaviors. 20.(6). 713-723.
Streigel-Moore, R. H. & Cachelin, F. M. (2001). Etiology of eating
disorders in women. The Counseling Psychologist. 29(5). 635-661.
Striegel-Moore, R. H., Silberstein, L. R., Frensch, P., & Rodin, J.
(1989). International Journal of Eating Disorders. 8(5). 499-509.
Streigel-Moore, R. H., & Smolak, L. (2000). The influence of ethnicity
on eating disorders in women. In R. M. Eisler & M. Hersen (Eds.), Handbook
of gender, culture, and health (pp. 227-254). Mawhaw, NJ: Lawrence
Erlbaum.
Strober, M. (1981). The significance of bulimia in juvenile anorexia
nervosa: An exploration of possible etiological factors. International Journal
of Eating Disorders. 1(1). 28-43.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
93
Summers, P., Forehand, R., Armistead, L., &Tannenbaum, L. (1998).
Parental divorce during early adolescence in Caucasian families: The role of
family process variables in predicting the long-term consequences of early
adult psychosocial adjustment. Journal of Consulting and Clinical
Psychology. 66. 327-336.
Tangney, J. P. (1990). Assessing individual differences in proneness
to shame and guilt: Development of the self-conscious affect and attribution
inventory. Journal of Personality and Social Psychology. 5 9 .102-111.
Tangney, J. P. (1991). Moral affect: The good, the bad, and the ugly.
Journal of Personality and Social Psychology. 61. 598-607.
Terril, F., & Terrell, S. L. (1981). An inventory to measure cultural
mistrust among Blacks. Western Journal of Black Studies. 5 .180-184.
Tiggemann, M., & Pickering, A. S. (1996). Role of television in
adolescent women’s body dissatisfaction and drive for thinness.
International Journal of Eating Disorders. 20(2). 199-203.
Tomori, M. (1994). Personality characteristics of adolescents with
alcoholic parents. Adolescence. 29(116). 949-959.
Tylka, T. L., & Subich, L. M. (1999). Exploring the construct validity of
the eating disorder continuum. Journal of Counseling Psychology. 46. 268-
276.
Vilas, N. A. (1097). Interpersonal guilt and eating disorder
symptomatology in college women. Dissertation Abstracts International:
Section B : The Sciences & Engineering, 57(7-B)4730.
Vitiello, B., & Lederhendler, I. (2000). Research on eating disorders:
Current status and future prospects. Biological Psychiatry. 47(9). 777-786.
Ward, A., Ramsay, R., & Treasure, J. (2000). Attachment research in
eating disorders. British Journal of Medical Psychology. 73. 35-51.
Weiss, J. (1986). Unconscious guilt. In J. Weiss & H. Sampson
(Eds ), The psychoanalytic process: Theory, clinical observation and
empirical research (pp. 43-67). New York: Guilford Press.
Weiss, J. (1993). How psychotherapy works: Process and
technigue. New York: Guilford.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
94
Wiilcox, M., & Sattler, D. N. (1996). The relationship between eating
disorders and depression. The Journal of Social Psychology. 136(2). 269-
271.
Zahn-Waxler, C., & Kochanska, G. (1990). The origins of guilt. In R.
Thomson (Ed.), 36th annual Nebraska symposium on motivation: Socio-
emotional development (pp. 183-258). Lincoln: University of Nebraska
Press.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 2
Means. Standard Deviations, and Intercorrelations Among the Variables
Variable 1 2 3 4 5- 6 7 8 9 10 11 12
1. PAa - .12 -.06 -.17* .18* - -.04 12 .06 .05
. . . Q8
.06
2. PDb - .21** -23** -.37*** .14* -.25*** .00 .20“ .04 .00 .08
3. AAC - -.43 -.49** .33*** -.10 .27*** .48*** -.02 .08 .18“
■ C k
o
>
a .
- .60** -.35** .06 -.11 -.42** .01 -.02 -.15*
5. DAe
- -.35*** .21** -.19**
. 40***
-.09 -.11 -.20“
6. Sur. G.f
- .28** .55“ .60“ .17* .15* .13
7. Sep. G.9
- .50“ .13 .08 .11 .02
8. ORGh
- .39“ .10 .16* .09
9. S-HG1
- .22“ .32“
44**
10. OCj
- .32“ .16*
11. Dk
- .61“
12. B/FP'
-
Mean
Standard Deviation
4.97
7.63
,36
.48
16.70
6.01
21.15
5.22
18.22
5.26
63.34
9.43___
40.24
7.42
47.02
6.77
32.41
11.29
16.37
5.33
35.90
12.80
11.80
5.08
a PA = Parental Alcoholism. b PD = Parental Divorce. C AA = Anxious Attachment. U CA = Close Attachment. e DA = = Dependent Attachment. 'Sur. G.
= Survivor Guilt. 9 Sep. G. = Separation Guilt. b ORG = Omnipotent Responsibility Guilt. ‘S-HG = Self-Hate Guilt. f°OC = Oral Control. k D = Dieting.
'B/FP = Bulimia/Food Preoccupation
Two-tailed level of significance: *p <05. “ p < .01. *“ p < 001
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
An investigation of a new diagnostic sub-type: Post traumatic stress disorder with psychotic features
PDF
Gender differences in motivation for sexual intercourse: Implications for risky sexual behavior and substance use in a university and community sample
PDF
Integration of science and practice: A collective case study of scientist -practitioner programs in counseling psychology
PDF
Gender differences in symptom presentation of depression in primary care settings
PDF
Effects of test interpretation style and need for cognition on elaboration, favorability, and recall
PDF
Intergenerational conflict, family functioning, and acculturation experienced by Asian American community college students
PDF
Development and validation of the Cooper Quality of Imagery Scale: A measure of vividness of sporting mental imagery
PDF
Assessment of racial identity and self -esteem in an Armenian American population
PDF
Gender role conflict, personality, and help -seeking in adult men
PDF
Gay men with eating disorders and food, body image, and exercise concerns: A group treatment approach
PDF
A daily diary approach to compare the accuracy of depressed and nondepressed participants' estimation of positive and negative mood: A test of the depressive realism hypothesis
PDF
Family functioning, intergenerational conflict, and psychological symptomology of Asian American college students
PDF
Heterosexuals' attitudes toward lesbians and gay men and the attitude functions they serve: Correlates, stability and gender differences
PDF
Alcohol treatment entry and refusal in a sample of older veterans
PDF
An adolescent -parent conflict resolution skills training program for ethnically diverse families: A program evaluation study
PDF
Anger expression in men with elevated blood pressure: Relationships to assertiveness
PDF
Considering the impact of culture on the therapeutic relationship: A look at Latino school-age children in a special education setting
PDF
An integrative analysis of racism and quality of life: A comparison of multidimensional moderators in an ethnically diverse sample
PDF
A Positive Psychology goal intervention using strengths and values to enhance goals and increase well -being
PDF
Anorexia nervosa: A developmental perspective
Asset Metadata
Creator
Reichardt, Gretchen Roberta
(author)
Core Title
Attachment style, interpersonal guilt, parental alcoholism, parental divorce and eating disordered symptomatology in college women
School
Graduate School
Degree
Doctor of Philosophy
Degree Program
Education (Counseling Psychology)
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
OAI-PMH Harvest,psychology, clinical
Language
English
Contributor
Digitized by ProQuest
(provenance)
Advisor
Goodyear, Rodney (
committee chair
), [illegible] (
committee member
)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c16-272798
Unique identifier
UC11335009
Identifier
3093974.pdf (filename),usctheses-c16-272798 (legacy record id)
Legacy Identifier
3093974.pdf
Dmrecord
272798
Document Type
Dissertation
Rights
Reichardt, Gretchen Roberta
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA
Tags
psychology, clinical