.3:
s...

.2: ‘
20:. 1..
:1

 

 

 

#25,».

x; f

 

 

THESIS
I
9(\(>l

 

LIBRARY
Michigan State
University

rm -In.

 

 

This is to certify that the
thesis entitled
ADULT ROMANTIC ATTACHMENT STYLE AS A MEDIATOR

FOR THE EFFECTS OF DOMESTIC VIOLENCE
ON MATERNAL-FETAL ATTACHMENT BEHAVIOR

presented by

Robin Pierce Weatherill

has been accepted towards fulfillment
of the requirements for

M.A. degree in Psychglggy

 

 

Date W

0.7639 MS U is an Afﬁrmative Action/Equal Opportunity Institution

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

 

DATE DUE I DATE DUE DATE DUE
SEP 3&0“;ng D

I pf" ' s r" h
In.) 1' ”CA: Lu, 1 L94 :_7

 

 

 

’02‘0 9.5;?

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/ClRC/DatoDue.p65—p.15

 

ADULT ROMANTIC ATTACHMENT STYLE AS A MEDIATOR
FOR THE EFFECTS OF DOMESTIC VIOLENCE
ON MATERNAL-FETAL ATTACHMENT BEHAVIOR
By

Robin Pierce Weatherill

AN ABSTRACT OF A THESIS
Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

MASTER OF ARTS

Department of Psychology
2001

Professor Alytia A. Levendosky

ABSTRACT
ADULT ROMANTIC ATTACHMENT STYLE AS A MEDIATOR
FOR THE EFFECTS OF DOMESTIC VIOLENCE
ON MATERNAL-FETAL ATTACHMENT BEHAVIOR
By

Robin Pierce Weatherill

Previous research has shown that domestic violence is associated with numerous
negative outcomes in children whose mothers are abused. Domestic violence may
impact the lives of children by interfering with the development of a secure attachment
relationship between mother and child. Few studies have looked at the prospective
effects of domestic violence on adult attachment or on the attachment of mothers to their
unborn children. This study examined the effects of domestic violence on maternal
attachment to the fetus in women who had experienced domestic violence with either a
previous or current partner. It was hypothesized that women who were physically abused
by a partner during pregnancy would be less engaged in matemal-attachment behaviors
than women who had experienced Violence with a previous partner but who were not
currently battered. It was also hypothesized that adult romantic attachment would
mediate the effects of domestic violence on matemal-fetal attachment behavior, and that
high relationship satisfaction in a non-violent relationship would buffer the effects of
domestic violence on adult romantic attachment. Results revealed a mediated link
between domestic violence and matemal-fetal attachment behaviors, providing
preliminary support for the hypothesis that the negative effects of domestic violence on

the mother-child relationship may begin before birth.

Dedicated to the women who so generously shared their stories, and to my parents and

my brother, for their love and support.

iii

ACKNOWLEDGMENTS

I would like to thank my advisor and mentor, Alytia Levendosky, for her
guidance, support, and generous faith in me. She read endless drafts of my thesis,
challenged and encouraged me at every step, and was as attentive to my personal growth
as to my writing. I also want to thank my other committee members: Alexander von Eye,
for his assistance with the analyses and articulating the hypotheses, and G. Anne Bogat,
for her comments and encouragement to enjoy the process.

In addition, I am grateful to my partners on the project, Alissa Huth-Bocks and
Sally Theran, who helped me in more ways than I can say, both with the details of the
project and with maintaining perspective. I could not have done this without their daily
collaboration and support. I am also grateful to all of our dedicated research assistants
who have given so many hours of their time to the project, whom I was lucky enough to

supervise, and who taught me a great deal.

TABLE OF CONTENTS

List of Tables ......................................................................................... vi
List of Figures .............. A ........................................................................ .vii
Introduction ............................................................................................ 1
Attachment Theory ......................................................................................................... 5
Adult Internal Working Models ............................................................................... 6

Adult State of Mind Regarding Early Childhood Relationships.......................... 9
Adult Romantic Attachment Style 10

Maternal Representations ................................................................... l 1
Intergenerational Transmission of Attachment .................................................. 12
Romantic Pairs as Attachment Relationships... .. 15
Intimate Relationships and Attachment In Adults: Potential fOr Harm and Healing .19
The Effects of Trauma on Adult, Maternal, and Infant Attachment ............................ 20
Domestic Violence and Parenting ................................................................................ 22
Hypotheses and Rationale ............................................................................................ 25
Method .............................................................................................................................. 29
Research Participants ................................................................................................... 29
Measures ....................................................................................................................... 34
Procedures .................................................................................................................... 37
Results ............................................................................................................................... 39
Level of Engagement in Matemal-Fetal Attachment Behavior ................................... 40
Romantic Attachment as a Mediator for the Effects of Domestic Violence
on Matemal-Fetal Attachment Behavior ................................................. 41
Relationship Satisfaction as a Moderator for the Effects of Domestic Violence
on Adult Romantic Attachment Style ...................................................... 44
Discussion ......................................................................................................................... 46
Appendix A: Consent Form .............................................................................................. 56
Appendix B: Fliers .................................................................................. 57
Appendix C: Demographic Questions ............................................................................... 59
Appendix D: Matemal-Fetal Attachment Scale ................................................................ 63
Appendix E: Attachment Style Questionnaire .................................................................. 64
Appendix F: Dyadic Attachment Scale ............................................................................. 66
Appendix GzViolence Against Women Scales ............................................................ 70
References ......................................................................................................................... 76

LIST OF TABLES

q

u

Table 1. Demographic Information on Study Participants .................................... 3
Table 2. Summary Statistics for Variables ...................................................... 39

Table 3. Correlations Between Calculated Romantic Attachment Style Score
and ASQ Subscale Scores: Both Groups ................................................ 40

Table 4. Results of t-test Comparing Mean MFAS Scores for Groups 1 and 2 . . . .. ..41
Table 5. Bivariate Correlations for Severity of Domestic Violence (DV), Relationship
Satisfaction, Adult Romantic Attachment Style, and Matemal-Fetal Attachment

Style ........................................................................................... 43

Table 6. Moderating Effects of Relationship Satisfaction on Adult
Romantic Attachment Style in Group 1: Current Partner DV ....................... 45

Table 7. Moderating Effects of Relationship Satisfaction on Adult
Romantic Attachment Style in Group 2: Previous Partner .......................... 45

vi

LIST OF FIGURES

Figure 1. Adult Romantic Attachment as a Mediator for the Effects
of Domestic Violence on Matemal-Fetal Attachment Behaviors ....................... 72

Figure 2. Relationship Satisfaction as a Moderator for the Effects of Domestic
Violence on Adult Romantic Attachment Style ........................................... 73

Figure 3. Path-Analytic model for Groups 1 and 2 .............................................. 74

vii

INTRODUCTION

Researchers on domestic violence have been expanding the focus of their studies tO
include the effects of partner violence on children and the family system (Campbell &
Lewandowski, 1997; Fantuzzo et al., 1991; Stemberg et al., 1993). There are multiple
and pervasive ways in which domestic violence may affect children. The effects may be
direct, in the form of injuries, or indirect, through witnessing or repercussions to the
mother, beginning even prior to birth. With lifetime prevalence estimated at 21—34%,
domestic violence constitutes a substantial threat to women’s physical and psychological
health, and, by extension, to the health of children (Browne, 1993). A study of the
prevalence of domestic violence in ﬁve US. cities found that households in which
domestic violence occurred were signiﬁcantly more likely to be households with
children, particularly young children under the age of 5 years (Fantuzzo, Boruch,
Beriama, Atkins, & Marcus, 1997). Research suggests that being pregnant and raising
children is associated with increased risk of domestic violence occurring in the home
(Fantuzzo et al., 1991; Gazmararian et al., 1996; Peterson et al., 1997). A review of
studies on the prevalence of violence during pregnancy concluded that between 3.9% and
8.3% of pregnant woman experience partner violence (Peterson et al., 1997). More
recently, a study of 6,718 pregnant women in South Carolina found that 10.9% of the
women reported being physically hurt by a partner or being in a physical fight with the
partner (Cokkinides & Coker, 1998).

The psychological harm that women and children experience when living with

domestic violence has been well documented (Campbell & Lewandowski, 1997;

Fantuzzo et al., 1991; Stemberg et al., 1993). Women and children who live with
domestic violence have been found to be at increased risk for low self-esteem,
depression, anxiety, and PT SD (Campbell & Lewandowski, 1997; Graham-Berman &
Levendosky, 1998a). Children who live with domestic violence are also more likely to
have internalizing and externalizing problems, reduced social competence, and school
problems (e. g. Campbell & Lewandowski, 1997; Edleson, 2000; Grahman-Berman &
Levendosky, 1998b; McCloskey, Figueredo & Koss, 1995). Studies have found a
signiﬁcant overlap of child abuse and wife abuse (Campbell & Lewandowski, 1997), and
even children who witness abuse but are not physically abused themselves Show levels Of
psychological damage similar to physically abused children (Stemberg et al., 1993).

Studies have also examined the impact that experiencing domestic violence may
have on women’s parenting behaviors (Holden & Ritchie, 1991; Levendosky & Graham-
Berrnann, 1998; Levendosky & Graham-Bermann, 2000; Levendosky & Graham-
Berrnann, in press; McCloskey, Figueredo, & Koss, 1995). However, these studies
examined families where domestic violence is or has been part of the children’s own
experience. They have not examined the effects of a mother’s past experience of
violence on her relationship with her children. Researchers estimate that between 21%
and 34% of women will experience domestic violence during their lifetime (Browne,
1993). These numbers suggest that even women who are currently in non-violent
relationships and raising children may have experienced domestic violence in the past
and may still suffer from its effects.

Although a lifetime incidence of 21-34% of women experiencing domestic

violence suggests that the experience of abuse is alarmingly common, the degree Of

violence women may experience varies widely in severity, from threats to lethal assaults.
Studies of women in shelters have typically focused on the effects of severe forms Of
violence. However, milder forms of violence such as hitting, slapping, or punching may
be more common among women who do not come to the attention of shelters or
emergency rooms (Straus, 1979; Tjaden & Thoennes, 2000). Less is known about the
effects of mild to moderate levels of violence, although this kind of violence may be
more common in the general population. A study which found that 56% of individuals
had experienced partner violence also found that the majority of the violence experienced
fell into the “mild’ category, such as threats, pushing, holding down, shaking, and
slapping (Marshall & Vitanza, 1994). While the effects of severe levels of violence on
women may be seen and measured in injuries, the harm caused by mild to moderate
levels of violence may be less obvious but more pervasive. The present study examines
the effects of mild to moderate violence on women living in the community, including
but not limited to woman living in shelters.

There is evidence that the majority of battered women eventually leave their
battering partners (Campbell & Lewandowski, 1997). There is also evidence that even
those who raise children in the context of domestic violence are perceived by their
children as nurturing, effective mothers (Levendosky, 1995). What are the factors that
enable women to be effective mothers to their children despite past or current experiences
of domestic violence? The present study examines two potential mitigating factors:
positive romantic relationships and secure adult romantic attachment style.

The goal of this study is to examine the potential for a supportive romantic

relationship to buffer the negative effects of past domestic violence on prenatal maternal

attachment behavior. Specifically, this study assessed the effects of high satisfaction in
romantic relationships on adult attachment styles and maternal representations of the
fetus in the context of past and current experiences of domestic violence. Abuse was
defined in this study as male-to-female physical assault or threat of assault. The present
study is an effort to understand whether the healing effects of a positive romantic
relationship subsequent to a battering relationship can affect a woman’s internal working
models to a degree that extends to her mental representations of her unborn child.
Studies suggest that adult attachment style is related to current relationship status
(Feeney & Noller, 1990; Main, 1996; van Uzendoom, 1995). Attachment style has been
assumed to be relatively stable over time, however, there is evidence that an individual’s
attachment style may change in response to life changes or new relationships (Davila,
Burge, & Hammen, 1997). Infant attachment has also been shown to change when the
mother’s life circumstances change (Egeland & Sroufe, 1981). The mother’s
relationships may thus inﬂuence both her own attachment style as well as her infant’s
attachment to her, either to make it more secure or more insecure. It is hypothesized that
experiences of domestic violence would be associated with insecure adult attachment,
and that this would also lead to more negative maternal representations of the unborn
child. However, a supportive romantic relationship might counteract the effects of past
violence, leading to secure adult attachment and more positive maternal representations.
Central to this is the possibility that adult attachment style can change in response to new

models of relationships.

Attachment Theory

Attachment theory predicts that infants coordinate their behavior with that of their
caregivers to maintain life—preserving contact with their primary source of nurturance,
modifying their behavior to receive the maximum contact that their caregivers can
tolerate (Slade, Belsky, Aber, & Phelps, 1999b). Bowlby (1988) conceived of attachment
as a system of infant behaviors closely coordinated with the mother’s own behavior,
which functions to evoke certain responses in her: specifically, to maintain her proximity
and prevent abandonment. Such a system of behavior would necessarily involve acute
sensitivity to potential abandonment, and thus would develop in such a way that it would
be activated by the mere threat of separation. The infant would then adapt its behavior to
preserve the relationship with the primary caregiver (Bowlby, 1988).

The Strange Situation was created as an experimental way of eliciting and
observing such separation behavior. Mary Ainsworth (1979) Observed that infants tended
to use their mother as a “secure base” from which to explore the world. As long as the
mother was available, the child would explore, periodically looking back or returning to
the mother for reassurance. Using the Strange Situation to Observe separation behavior in
a laboratory setting, Ainsworth categorized the babies’ style Of separation and reunion
into three types. Secure children would explore as long as their mother was present,
become upset when she left, and would reach to her for comfort when she returned.
Avoidant babies seemed to explore independently whether their mother was there or not,
and to be indifferent to her return. Ambivalent children would cling to their mOther but
appear not to be comforted by her; they would become upset when she left, but even

more upset when she returned.

These infant attachment styles are initially specific to the relationship with the
primary caregiver, but gradually become internalized as a set of beliefs and expectations
about relationships in general. These internal working models become the basis of the
children’s interactions with others, inﬂuencing how they see themselves in relationships
and how they interpret others’ behavior (Bowlby, 1973). Relationship experiences shape
internal working models, which, in turn, influence future relationship behavior. For
example, a child with a secure attachment history is likely to expect that Others are
trustworthy, and so will behave in an open way and interpret Situations in this light. A
child who expects others to disappoint or abandon them may interpret others’ behaviors
accordingly. Longitudinal studies have found that infant attachment styles are predictive
of behavior outcomes in later childhood; children who were securely attached as infants
tend to be more conﬁdent, have better problem solving skills, and better peer relations
than those who were insecurely attached as infants (Main, 1996).

Adult Internal Working Models

While infant attachment style is behaviorally assessed, measures of adult
attachment style purport to directly tap the internal working models of self and others,
rather than measuring attachment behavior. These internalized representations, or
internal working models, develop in the context of specific relationships and may be
assessed in regard to those relationships. For example, the Adult Attachment Interview
(AAI) (Main & Goldwyn, 1984) assesses working models of the early childhood
relationship with the primary caregiver. The Working Model of the Child Interview
(Zeanah & Benoit, 1995) assesses maternal representations of the child, while measures

of adult romantic attachment style assess internal representations of romantic

relationships. Adult attachment style also differs from infant attachment in that it is
generalized to other relationships. Adults carry with them the memory of the early care-
giving relationship, and also develop a new attachment relationship with their romantic
partner. Measures of adult romantic attachment are intended to tap generalized beliefs
about relationships. However, the theory behind these measures is that the romantic bond
between partners is the primary adult attachment, in Bowlby’s sense of the word, and the
only one that is analogous to the parent-infant bond (Hazan & Zeifman, 1999).

Bowlby deﬁned attachment behavior as any behavior that ensured the individual’s
proximity to another person perceived as better able to cope with the world, as a mother
would be perceived by her infant (although, theoretically, this would apply to an
individual at any age) (Bowlby, 1988). Such behavior would be most evident when the
individual was feeling unable to cope with the world themselves, such as when they were
ill, tired, or frightened. Bowlby Observed that, even in healthy infants, attachment
behavior could be evoked by separation or threat of separation from the primary
caregiver. He hypothesized that in this instance the infant’s behavior was a response to
their perception of potential danger (being alone in the world) rather than actual danger.
The behavior is also a manifestation of the infant’s internal working model of the
relationship with the primary caregiver, because the behavior reﬂects what the infant
believes is most likely to ensure the caregiver’s proximity. For example, if the infant
expects the mother to recoil if he clings to her, his act of indifference will reﬂect this.
Adult attachment researchers have studied the effects of this “threat of separation”

on individuals in romantic relationships by assessing how the individuals react to conﬂict

with their partner (Collins & Read, 1990; Simpson, Rholes, & Phillips, 1996). In

addition to assessing conﬂict as a “threat of separation,” they have also examined the
effects of actual separation due to death or the break-up of a relationship (Hazan &
Zeifman, 1999). The results of these studies suggest that the romantic bond is unique
among adult relationships (Hazan & Zeifman, 1999). Researchers have found that, when
separated, partners in romantic pairs describe levels of stress and proximity seeking
behavior that are greater than those described when separated from siblings and friends
(Hazan & Shaver, 1987). Consistent with attachment theory, individuals also differ in
their behavioral reaction to the threat of separation, with some becoming aloof and others
becoming more clingy. In other words, it would appear that attachment behavior is
invoked within adult romantic relationships in a way that is not evident in other
relationships.

The generalization of internal working models continues when individuals
become parents: they bring their mental representations Of others to this new relationship,
and become the partner for the attachment behavior of the child. Researchers interested
in continuity of attachment style within individuals over time have used the Adult
Attachment Interview to compare parents’ early childhood memories of their primary
attachment relationship (internal working models) with their infants’ attachment behavior
in the Strange Situation (van Uzendoom, 1995).

Recently researchers have begun looking at an intermediate step in the
transmission of attachment: the parent’s working model of the child (Zeanah & Benoit,
1995). Adult memories of early childhood relationships, adult romantic relationships,
and maternal representations of the child have all been the subject of attachment studies.

Theoretically, they should all be related as generalizations of the original working model

internalized by the individual from the early primary caregiving relationship. While
researchers have found that parents’ adult AAI styles are predictive of their infants’
attachment classiﬁcations (van Uzendoom, 1995), no research has been done on the
relationship between these various kinds of adult attachment. It is not clear from the
literature whether these are different manifestations of one working model, or if
individuals have specific working models for speciﬁc relationships, just as children may
have different attachment styles with each parent. The continuity of working models
within individuals over time was an important component of Bowlby’s theory (Bowlby,
1988), yet much of the literature on adult attachment has compartmentalized working
models by assessing only one area of adult relationships. Attachment researchers have
examined either romantic relationships or adult memories of early relationships, but not
both.
Adult State of Mind Rgarding Early Childhood Relationships

Working models of early childhood attachment ﬁgures may be assessed with a
semi-structured interview, such as the AAI, which is coded by trained coders, and yields
three major adult attachment classiﬁcations: autonomous, preoccupied, and dismissing.
Autonomous is considered analogous to the secure attachment classiﬁcation in infants,
while preoccupied and dismissing are considered to be forms of insecure attachment and
are thought to be analogous to the anxious/ambivalent and avoidant forms of infant
attachment, respectively (van Uzendoom, 1995). The AAI also yields a fourth
classiﬁcation; unresolved/disorganized with respect to experiences of trauma or loss.
Individuals classiﬁed as unresolved also receive a classiﬁcation according to the

autonomous/preoccupied/dismissing categories. The AAI assesses adult representations

of early childhood relationships and is used primarily in research that focuses on the
transmission of attachment from parent to child (Beckwith, Cohen, & Hamilton, 1999;
Benoit & Parker, 1994; Main & Hesse, 1990; Schuengel, Bakermans-Kranenburg, & Van
Uzendoom, 1999; Slade & Aber, 1992; van Uzendoom, 1992; van Uzendoom, 1995). A
meta-analysis found that parental AAI classiﬁcation is predictive of infant attachment as
assessed using the Strange Situation; secure parents are more likely to have secure
children, while preoccupied and dismissing parents are more likely to have insecure
children (van Uzendoom, 1995).
Adult Romantic Attachment Style

Adult romantic attachment has been most frequently measured using self—report
questionnaires which require less training to administer than the AAI, making it feasible
to do studies with larger samples (Bartholomew & Horowitz, 1991; Collins & Read,
1990; Feeney & Noller, 1990; Feeney, Noller, & Hanrahan, 1994; Kirkpatrick & Davis,
1994; Kobak & Hazan, 1991; Simpson et al., 1996). These questionnaires differ from the
AAI in that they measure representations of current relationships rather than
representations of memories (Feeney, 1999). Attachment researchers initially
hypothesized categories of romantic attachment style analogous to the infant categories
developed by Ainsworth and the adult categories yielded by the AAI: secure, avoidant
(AAI “dismissing”), and anxious/ambivalent (AAI “preoccupied”) (Feeney & Noller,
1990). Early measures were designed as forced choice questionnaires in which
participants were given three or four paragraphs describing relationship styles and asked
to choose the one that best described themselves (Bartholomew & Horowitz, 1991;

Feeney & Noller, 1990; Hazan & Shaver, 1987; Kobak & Hazan, 1991). Later studies

10

rated romantic attachment style dimensionally rather than categorically, based on the
hypothesis that the three classifications might not be mutually exclusive, and that
individuals might lie along dimensions within each category (Collins & Read, 1990;
Davila et al., 1997; Feeney et al., 1994; George & West, 1999). This hypothesis was
supported by findings that indicated that secure romantic attachment was negatively
correlated with insecure romantic attachment, but that the two insecure styles were
independent of each other (Feeney et al., 1994). In other words, insecurely attached
adults might rate themselves highly on dimensions of ambivalence as well as dimensions
of avoidance. In developing a dimensional measure, Feeney (1994) proposed ﬁve
dimensions of romantic attachment: Conﬁdence, Discomfort with Closeness, Need for
Approval, Preoccupation, and Relationships as Secondary. Discomfort with Closeness
and Relationships as Secondary correlated with the avoidant or dismissing category.
while Need for Approval and Preoccupation correlated with the anxious/ambivalent or
preoccupied category. In the present study, individuals’ scores on the different
dimensions will be converted into a categorical variable: with a secure and an insecure
group.

Maternal Representations

 

In addition to mental representations of their parents and romantic partners,
parents also have internal working models of the parent-child relationships, even before
the child is born. Interest in parents’ working models of the infant grew out of research
on infant temperament. Infant temperament was initially assessed using parental report
measures. When it became clear that the parents’ report said more about their

perceptions and expectations than about the infant per se, temperament researchers began

11

using direct observation to assess infants, while other researchers focused on the parents’
perceptions (Zeanah & Benoit, 1995). Zeanah & Benoit (1995) developed an interview
similar to the AAI to assess maternal representations of the child. The Working Model of
the Child Interview is a semi-structured interview which assesses maternal
representations of the child. It may be administered either during pregnancy or after
birth. The Matemal—Fetal Attachment Scale (MFAS) is a questionnaire that was
developed to assess the mother’s engagement in behaviors thought to contribute to the
development of the infant attachment bond (Cranly, 1981). These maternal behaviors
are not “attachment behaviors” in Bowlby’s sense: attachment behaviors in the truest
sense refer to infant behaviors that serve to ensure the mother’s proximity to the infant.
Rather, the maternal behaviors tapped by this measure are analogous to the observable
parental behaviors that have been associated with different types of infant attachment
(Ainsworth, 1978; van Uzendoom, 1995). This study hypothesizes that the maternal
behaviors measured by the MFAS reﬂect the mother’s working model of the relationship
between herself and her unborn child.
Intergenerational Transmission of Attachment

Studies of adult romantic attachment and research on the transmission of
attachment from parent to child have proceeded largely independently of each other.
Studies of the predictive relationship between parental and infant attachment style have
used adult memories of early childhood attachment as the predictive factor rather than
adult representations of current romantic relationships. The relationship between adult
AAI status and infant Strange Situation classiﬁcation has been well established (van

Uzendoom, 1995). Studies have found between 68% - 75% correspondence between

12

secure/insecure adult attachment representations (dismissing, autonomous/secure,
preoccupied, and disorganized) and infant status (respectively, avoidant, secure, anxious-
ambivalent, and disorganized) (Benoit & Parker, 1994; Van Uzendoom, 1995). In a
meta-analysis, van Ijzendoom (1995) noted that correspondence rates varied depending
whether the study used a classification system of three groups, (autonomous, dismissing,
preoccupied) or if a fourth group (unresolved) was added. When using three groups,
there was an overall correspondence of 75 % between parental AAI classiﬁcations and
infant Strange Situation status for the secure/insecure split (when the two insecure
categories were collapsed into one group) and a 70% correspondence rate for the three-
way classification. When using four groups, the predictive correspondence was less, at
63%. Van Ijzendoom also notes that there is a difference in predictability between the
groups, with the secure group (also the largest) having the strongest predictive power,
and the preoccupied group (the smallest in mostvsamples) having the least predictive
power.

Despite studies supporting the predictive validity of parental AAI status with
regard to infant Strange Situation classification, the mechanism of transmission is only
partly understood (van Uzendoom, 1995). It is hypothesized that parental representations
of past and present attachment figures inﬂuence the degree of sensitivity and
responsiveness with which parents react to infant attachment signals (van Uzendoom,
1995). Parents who are confident in their relationships with others may be better able to
respond to their infants. Insecure parents, who may be uncomfortable with intimacy or
preoccupied with their own attachment experiences, may be less available to their infants

or may have distorted perceptions of their infant which interfere with their ability to

13

interpret infant attachment signals. Insecure attachment in infants has been shown to be
associated with an insecure adult attachment style and maternal experience of unresolved
trauma or loss (Main & Hesse, 1990; van Uzendoom, 1995).

The administration and coding requirements of the AAI make it difficult to use in
large studies. In addition, it has been used primarily with white, middle class samples
(Schuengel et al., 1999; van Uzendoom, 1995), with some exceptions (Main & Hesse,
1990). The self-report design used in most measures of adult romantic Style has made it
possible to use with larger samples (Collins & Read, 1990; Simpson et al., 1996).

According to attachment theory, internal working models that were developed in
the context of a specific relationship are generalized to other relationships and may
inﬂuence the individual’s expectations of and behavior toward others. This study
hypothesizes that women’s internal working models of romantic others will inﬂuence
their working model of their infant. There is evidence for such generalization:
attachment researchers have found that the mother’s way of being with her infant is
associated with varying outcomes in infant attachment (Ainsworth, 1978; Main &
Cassidy, 1988; Main & Hesse, 1990; van Uzendoom, 1995), and part of this variance has
been attributed to the mother’s state of mind with regard to her own early attachments
(Benoit & Parker, 1994; Main & Hesse, 1990; van Uzendoom, 1992). For the present
study, it was hypothesized that the mother’s “way of being” with her fetus reﬂects her
internal representations of her unborn child. The present study tests the hypothesis that
adult romantic working models inﬂuence maternal working models, much as research has
already shown that parents’ working models of their own childhood appear to inﬂuence

their infants’ developing working models of self and others.

14

Romantic Pairs as Attachment Relationships

Following the premise that internal working models are carried through into
adulthood, attachment dynamics will continue in adult relationships. As children develop,
they shift some of their attachment behaviors to seek support from peers, but the parent
remains the central attachment figure until the attachment bond is transferred to a
romantic partner in adulthood (Hazan & Zeifman, 1999). Hazan and Zeifman (1999)
argue that pair bonds are unique among adult relationships in their resemblance to the
infant attachment relationship. They analyzed adolescent and adult relationships with
friends and with romantic partners for components that would be psychologically and
functionally similar to attachment behavior in infants: proximity maintenance, safe‘
haven, separation distress, and secure base. They found that proximity maintenance and
safe-haven behavior was present both in romantic relationships and friendships, but that
separation distress and secure base behavior was found exclusively in romantic
relationships. They concluded that, in adults, “full blown attachment relationships” (p.
350) were seen in only in romantic relationships.

Studies have shown that romantic attachment styles are relatively stable within
individuals (Feeney, 1999; van Uzendoom, 1995). However, there is also evidence of a
reciprocal relationship between relationships and romantic attachment styles. As
individuals shift their primary attachment from parent to romantic partner, they bring the
attachment style from the former relationship into the new romantic relationship. As with
any adaptive behavior developed in response to a speciﬁc context, the individual’s

romantic attachment style may be maladaptive in the new (romantic) context. Each

partner brings his or her own working models to the situation, and each perceives the
other’s behavior through this ﬁlter.

In a 1991 study, Kobak and Hazan hypothesized that, in order to be adaptive,
working models must be ﬂexible enough to change in response to new circumstances, to
assimilate new information about the other into their working model. They speculated
that secure individuals would be more adaptive, more ﬂexible in response to new
information, and thus more accurate in attributing intentions to their partner’s behavior.
Therefore, in addition to assessing the romantic attachment Styles and observing the
couple engaged in problem solving, the researchers assessed the accuracy of each
partner’s perceptions of the interaction. The results suggested that internal working
models did inﬂuence perceptions of a stressful interpersonal interaction (Hammen et al.,
1995). Individuals rated as secure made more accurate attributions about their partner’s
behavior than did individuals rated as insecure. Insecure individuals were less likely to
adapt their working models in response to their partner’s behavior, and this negatively
affected relationship functioning.

The interaction of a couple’s working models can be predictive of the longevity of
the relationship in some instances. Not surprisingly, secure adult romantic attachment
style appears to be associated with high satisfaction as well as stability in romantic
relationships (Collins & Read, 1990). However, couples in which the woman is

classiﬁed as preoccupied and the man as dismissing have also been found to be stable and
long-lasting, despite low relationship satisfaction ratings expressed by both partners
(Simpson et al., 1996). Simpson et a1. hypothesized that these relationships were

Perpetuated because each partner conﬁrmed the expectations of the other; for example,

16

the woman expected her partner to be less interested in intimacy than she was, and her
dismissing male partner confirmed her expectations. The man, on the other hand,
expected his partner to be intrusive and clingy, and his expectations were conﬁrmed by
his preoccupied partner’s efforts to re-engage him in the relationship.

The correlation between adult romantic attachment status and relationship status
may be confounded by a reciprocal relationship between the two. Secure adult romantic
attachment style appears to be associated with high satisfaction in romantic relationships,
as well as with the longevity of the relationship (Collins & Read, 1990). In their study of
married couples, Kobak & Hazan (1991), found that their study contained a high number
of secure couples and speculated that marriage itself might lead to greater security of
attachment. Evidence for reciprocal inﬂuence is provided by a study that found that
individuals classified as insecure who became involved in a romantic relationship were
more likely to be classiﬁed as secure six months later (Davila et al., 1997).
Understanding the reciprocal effects of romantic attachment and intimate relationships
may be further confounded when researchers use a sample of couples, which limits the
data to a self-selected group of people who are able to initiate and sustain (however
brieﬂy) a romantic relationship.

A reciprocal relationship between romantic attachment style and relationship
status suggests that an individual’s romantic attachment style may change in a new

relationship. Davila et a1. (1997) found that the most common change was from insecure
to secure. They hypothesized that this change occurred when an insecure individual
developed a relationship with a supportive partner who disconﬁrrned their expectations of

rejection. On the other hand, the association of insecure romantic attachment and the

17

experience of emotional abuse suggest that change in status from secure to insecure may
also occur (O'Heam & Davis, 1997). In the O’Heam & Davis (1997) study, it was
hypothesized that secure romantic attachment would be negatively correlated with the
experience of emotional abuse, while each of the insecure styles would be positively
correlated with the experience of emotional abuse. Subjects were asked to fill out self-
report questionnaires indicating how often they and their partners had received and
inﬂicted certain abusive behaviors, such as yelling, making the other person feel guilty,
or destroying something belonging to them. Emotional abuse was also assessed in an
interview. Romantic attachment style was assessed with a semi-structured interview and
subjects were categorized according to Bartholomew’s four attachment styles
(Bartholomew & Horowitz, 1991). Sixty-one undergraduates were interviewed, 19 male
and 42 female. Results from the males were inconsistent, so the discussion focused on
data from the female subjects. Preoccupied attachment was positively related to both
experiencing and inﬂicting emotional abuse. Woman with a fearful-avoidant style were
more likely to have received emotional abuse and less likely to have inﬂicted it. Secure
romantic attachment was negatively associated with having received or inﬂicted
emotional abuse. One explanation for this was that adults with secure romantic
attachment styles were less likely to become involved with emotionally abusive partners;
alternatively, the experience of emotional abuse may evoke a preoccupied romantic

attachment style in the partner.

18

Intimate Relationships and Attachment in Adults:
Potential for Ham and Healing

Adult romantic attachment style is both stable and, to varying degrees, adaptable;
it inﬂuences an individual’s romantic relationships but also adapts to new experiences by
assimilating new beliefs about self and others (Davila et al., 1997). The devastating
effects of domestic violence on women’s capacity for relatedness are well documented
(Browne, 1993; Campbell & Lewandowski, 1997). The traumatic repercussions of
violence in intimate relationships include distrust of others, distorted relationship
boundaries, lowered self-esteem, and malevolent beliefs about other’s intentions
(Herman, 1992): all fundamental dimensions of the construct of adult romantic
attachment (Feeney et al., 1994). It would therefore be expected that domestic violence
would be associated with insecure adult romantic attachment, and this is supported by
two of the few studies on this subject (O'Heam & Davis, 1997; Zeanah et al., 1999).

While only two studies have examined changes in adult romantic attachment in
the aftermath of domestic violence, the literature on trauma (such as childhood sexual
abuse) suggests that new relationships with caring therapists or members of a support
group can bring about positive changes in the victims’ beliefs and perceptions of
relationships (Herman, 1992; Saunders & Edelson, 1999). Although there is evidence
that women who have experienced domestic violence are more likely to Show insecure
adult romantic attachment (O'Heam & Davis, 1997), their beliefs about self and others
may be open to change, particularly if they become involved with a new partner whose

behaviors disconﬁrrns their malevolent expectations. One hypothesis of the present study

19

was that, even for women who have experienced violence in the past, a current positive
relationship with a non-violent partner will be associated with a secure attachment style.
The Effects of Trauma on Adult, Maternal, and Infant Attachment

Research has found that trauma that occurs in the context of relationships over
time, such as incest or domestic violence, has particularly devastating effects on the
victim’s capacity for relationships (Herman, 1992). “Where the trauma has been repeated
and prolonged, the patient’s expectations of perverse or malevolent intent can prove
especially resistant to change” (p. 138). From an attachment theory perspective, the
experience of abuse alters the victim’s working models of self and others in a way that
makes the models particularly rigid; in attachment terms, this would be expected to
produce the extremes of dismissing or preoccupied representations of others. One study
that examined adult romantic attachment and partner abuse found that secure adult
romantic attachment was negatively correlated with experiencing abuse, while both
preoccupied and dismissing styles were positively correlated with experiencing abuse
(O'Heam & Davis, 1997).

If abuse is predictive of insecure adult romantic attachment, and the mother’s
attachment style is predictive of her infant’s attachment style (van Uzendoom, 1995),
then it would be expected that maternal experiences of abuse would predict an insecure
attachment style in her infant. In support of this hypothesis, studies have found a
particular type of infant insecure attachment, disorganized attachment, to be associated
with unresolved maternal experiences of trauma (Main & Hesse, 1990; Schuengel et al.,
1999; Zeanah et al., 1999). Disorganized attachment is associated with child

maltreatment (Egeland & Sroufe, 1981) and later child adjustment problems (Lyons-

20

Ruth, 1996). Unresolved trauma is therefore associated with problems in the attachment
relationship and with later child adjustment problems as well. While there is evidence for
a relationship between unresolved maternal trauma and insecure attachment in the child,
the way in which a history of trauma affects the mother-child relationship is still unclear.

One potential explanation for the relationship between maternal trauma and child
attachment style is the theory that “frightened and frightening” maternal behavior toward
the infant contributes to disorganized attachment (Main & Hesse, 1990; Schuengel et al.,
1999). In Main & Goldwyn’s study, frightening behaviors were seen in the majority of
mothers with unresolved loss. However, they were also observed to co-occur with the
sensitive parenting behaviors associated with secure attachment. Secure mothers with
unresolved loss showed fewer frightening behaviors, and unresolved loss in secure
mothers did not predict disorganized attachment status in their infants. Secure
attachment representations in these mothers may have acted as a protective factor in these
cases. The ﬁndings were replicated in another study which found that unresolved loss
predicted frightening behavior and disorganized infant attachment in insecure mothers
but not in secure mothers (Schuengel et al., 1999).

Thus, adult attachment style appears to inﬂuence infant attachment in at least two
ways. Secure and insecure adult styles are predictive of infant attachment style, with the
infant style correlating with the analogous adult style. Secure adult attachment also
appears to buffer infant attachment in the context of unresolved maternal trauma
(Schuengel et al., 1999). However, there is evidence that the experience of trauma itself
is associated with insecure adult attachment styles, and thus might inﬂuence infant

attachment directly (O'Heam & Davis, 1997).

21

 

Domestic Violence and Parenting

One form of trauma that appears to be common among women and mothers is
domestic violence (Cokkinides & Coker, 1998). Recent research has focused on the
impact domestic violence has on the family, including parenting behavior and child
adjustment (Levendosky & Graham-Bermann, 2000; Levendosky & Graham—Bermann,
in press; McCloskey et al., 1995; Stemberg et al., 1993). However, no studies have
measured the prospective effects of domestic violence, past or current, on maternal
representations of the fetus during pregnancy. This is an important area to consider,
given what is known about the negative effects of domestic violence on parenting
behavior and child adjustment (Fantuzzo et al., 1991; Levendosky & Graham-Bermann,
in press; Stemberg et al., 1993). The consequences of violence may begin before the
child is born, at the source of the attachment relationship, the mother’s bond with her
unborn child.

Children who are victims or witnesses of domestic violence are more likely to be
depressed, to have both internalizing and externalizing behavior problems, as well as
symptoms associated with posttraumatic stress (Fantuzzo et al., 1991; Graham-Bermann
& Levendosky, 1998; Stemberg et al., 1993). Marital distress has been shown to have
negative effects on the parent-child relationship (Erel & Burman, 1995). There is also
evidence that witnessing violence may have direct negative effects on children
(Kilpatrick & Williams, 1998). It may be doubling frightening to a child to witness a
physical or emotional assault on their mother, who is presumably seen by the child as a
protective ﬁgure: not only is the child in close proximity to the violence, but the person

who might protect them from it is a victim herself.

22

Recent research has attempted to identify how these outcomes may be moderated
or mediated through parenting, usually maternal parenting. Studies have found that
domestic violence predicts higher levels of parenting stress, which in turn negatively
affected child adjustment (Levendosky & Graham-Bermann, 1998; Wolfe, J affe, Wilson,
& Zak, 1985). Women interviewed in a qualitative study on mothers who had
experienced partner violence reported that physical health problems and single parenting.
both sequellae of their violent relationships, made parenting more difficult for them
(Levendosky, Lynch, & Graham-Hermann, 2000). Children may also be affected by
violence through its effect on their mothers’ psychological functioning (Levendosky &
Graham-Bermann, in press). For example, depression is a common symptom among
victims of domestic violence, and children of depressed parents are at increased risk for
adjustment problems (Downey & Coyne, 1990). The toll of living with violence and the
fear of violence, including depression, physical health problems, isolation, and single
parenting, may mean that these mothers are less available and able to respond to their
children’s needs.

In one study of observed parenting and child behavior, domestic violence was
shown to affect maternal warmth, an important aspect of parenting (Levendosky &
Graham-Bermann, 2000). In this study, mothers who had experienced physical abuse
were observed to be less warm in an interaction with their children, while psychological
abuse to the mother was associated with increased observed antisocial behavior on the

part of the child. The ﬁnding that the experience of domestic violence decreased
observed warmth in mother-child interaction is particularly relevant to the possible

relationship between domestic violence and matemal-fetal attachment. Warmth was

23

defined in this study as “The degree to which the mother is positive to the child, nice to
the child, enjoys being with the child, and is supportive of the child” (Hetherington,
Hagan, & Eisenberg, 1992; as cited in Levendosky and Graham-Berman, 2000).
Maternal positive affect and expressed pleasure with the child were factors used in
another study to assess quality of maternal attachment representations (Slade et al.,
1999b). By extension, if domestic violence negatively affected warmth in observed
mothering, it is possible that this would also be reﬂected in the attachment relationship,
as measured by maternal representations of the infant.

Studies on domestic violence and parenting usually focus on women who are
currently in or have recently been in a violent relationship. Less is known about how a
past history of domestic violence might affect a woman’s parenting. Given that a history
of abuse in childhood has been shown to affect later psychological functioning, which in
turn affects parenting (Levendosky & Graham-Bermann, in press), it would be expected
that experiences of violence in intimate relationships would have repercussions long after
the relationship is over. This would be consistent with research ﬁndings that the effects
of unresolved maternal trauma carry over to the infant’s attachment to the mother. (Main
& Hesse, 1990; Schuengel et al., 1999). If the effects are seen in the infant’s attachment
behavior, it is likely that they were inﬂuencing the mother’s perception of the child even
prenatally, given the evidence for the correlation between maternal representations and
infant attachment (Slade, et al, 1999b). To understand the transmission of attachment, it

would be important to trace the development of the attachment relationship back to its

earliest beginnings, during pregnancy-

24

Hypotheses and Rationale

One of the ways in which domestic violence may negatively affect women and
children is by affecting the mother’s ability to nurture and parent her child. Studies have
documented the negative effects that the presence of domestic violence may have on the
parent—child relationship and child functioning (Fantuzzo et al., 1991; McCloskey et al..
1995; Stemberg et al., 1993). One measure of the parent-child relationship is assessment
of child attachment style. Evidence that unresolved maternal trauma is associated with
disorganized attachment in children suggests that violence may have important
consequences for the attachment relationship (Schuengel et al., 1999), and that the
damage may occur at a very early stage in the mother-child relationship, i.e. prenatally.
If domestic violence does affect the mother-fetal relationship, it is likely that the effects
would be seen in the maternal attachment behaviors toward the unborn child, as this is the
ﬁrst manifestation of the mother-child relationship. Yet no studies have examined the
effects of domestic violence at this point in the attachment relationship. This study
compared women who were currently experiencing violence in a romantic relationship
with those who had experienced violence with a previous romantic partner in order to
understand 1) the effects of partner violence on the matemal-fetal relationship and 2) the
potential for a healing relationship to mitigate those effects.

If the abuse occurred in a past relationship, it is unclear what effects the
experience will have in the context of later parenting; some studies suggest that
unresolved trauma has long lasting effects (Benoit & Parker, 1994; Main & Hesse, 1990:
Schuengel et al., 1999), while other studies indicate that the current partner relationship is
the most predictive of current parenting or maternal attachment style (Lindahl, Clements,

& Markman, 1997; Zeanah et al., 1999). This study hypothesized that women who were

25

experiencing violence in a current relationship would be less engaged in matemal-fetal
attachment behaviors than women who had experienced violence in the past but were not
experiencing it with their current partner.

Furthermore, the consequences of domestic violence may begin prior to the
matemal-fetal relationship by negatively affecting the mother’s adult romantic
attachment style. There is some evidence for the inﬂuence of trauma on adult attachment
status (O'Heam & Davis, 1997). Women who have experienced violence in a romantic
relationship may have more insecure working models of others, and their romantic
working models might then inﬂuence their working model, or maternal representation, of
their unborn child. The insecure quality of their maternal representation would be
manifest in their behavior, and they would therefore be less engaged in matemal-fetal
attachment behaviors than women who have experienced partner violence in the past but
who are not currently experiencing it in their present relationship.

Adult attachment style in regard to early childhood memories (as measured by the
AAI) has been shown to account for some of the variance in differences in parenting
behavior (Slade et al., 1999b; van Uzendoom, 1995). According to attachment theory,
adult romantic attachment style would be inﬂuenced by the same internal working
models that are manifested in the Adult Attachment Interview. If adult attachment style
with regard to early childhood memories and adult romantic attachment style are thus
connected, adult romantic attachment style should also be associated with variation in
parenting behavior. Therefore, the second hypothesis of this study was that secure adult
romantic attachment would mediate the relationship between domestic violence and

matemal-fetal attachment behaviors, whether the violence was experienced with a

26

previous or current romantic partner (see Figure 1). One way in which this study differs
from other studies on the effects of domestic violence on women and children is that it
examined the effects of past experiences of domestic violence with a former partner as
well as the presence of domestic violence in a current relationship. No prior studies have
examined the romantic attachment styles of women with battering histories who are
currently in non-battering relationships. If adult romantic attachment style does mediate
the effects of domestic violence on matemal-fetal attachment behavior, it would be
important to examine whether it was a mediator for past experiences as well as current
ones.

Given the evidence for a reciprocal relationship between adult relationship and
attachment status (Davila et al., 1997), it would be expected that the negative effects of
the abuse on the woman’s working models of others might be ameliorated if the new
partner no longer conﬁrms her negative working models of others. If the new
relationship is experienced as a satisfying one, then it may serve as a healing relationship,
i.e. it may moderate the effects of past experiences of domestic violence on current adult
romantic attachment status. Therefore, the third hypothesis of this study was that, within
the group of women who had experienced violence in the past but were not currently
battered, high relationship satisfaction in the current relationship would buffer the effects
of past domestic violence on adult romantic attachment style. It was also hypothesized
that in relationships where there was on-going domestic violence, the effects of the
violence would overcome any potential mitigating effects that high relationship

satisfaction might have, and the buffering effect would not be seen (see Figure 2).

27

Much of the research on the transmission of attachment has used samples from
middle-class, married, White/Caucasian populations. Less is known about the connection
between adult romantic attachment and maternal attachment behavior in the context of
poverty and trauma, and among single mothers and minorities. There is evidence for a
higher incidence of unresolved trauma in mothers from low socioeconomic groups (van
Uzendoom, 1996), yet only one study has examined infant attachment in the context of
domestic violence (Zeanah et al., 1999), and this study did not measure maternal
attachment behaviors, only infant attachment style. The present study examines the
relationship between adult romantic attachment style and maternal prenatal attachment
behavior in the context of past and current experiences of domestic violence, providing a
prospective assessment of the effects of domestic violence on unborn children. The
participants in this study came from a range of backgrounds, but were primarily from low
socioeconomic groups. This was important in evaluating whether existing hypotheses
about the transmission of attachment are applicable in the context of poverty and

domestic violence and among an economically and ethnically diverse population.

28

METHOD

Research Participants

This study was done in conjunction with a larger study conducted by Alytia
Levendosky, G. Anne Bogat, William Davidson and Alexander von Eye at Michigan
State University. Participants included 106 pregnant women recruited to participate in
the Mother-Infant Study, a longitudinal study on the effects of domestic violence on
mother-infant relationships. The 106 women selected for this study from the total sample
of 207 were divided into two groups: those who had been battered by a previous partner
but not by their current one (n=45, 43%) and those who had been battered by their current
partner (n=61, 57%). Women who reported that they had never experienced domestic
violence or who reported that they had experienced domestic violence with their current
partner in the year prior to but not during their pregnancy were excluded from the total
sample of 106. Participants were recruited by posting ﬂiers in public areas and
businesses in Lansing and surrounding urban and rural counties in Southeast Michigan.
In addition, ﬂiers were posted in and referrals obtained from agencies, hospitals and
programs that provided financial, legal, and health services to pregnant women or victims
of domestic violence.

Women called the project office, where they were screened for eligibility, which
was determined by their age, current relationship status, and history of battering. Women
who were not involved in a relationship during their pregnancy or who were under the
age of 18 or over the age of 40 were excluded. The women were told the study was about

women’s relationships with the important people in their life, including partners, family

29

members, and children, and that if they participated in the study they would be asked
about their thoughts and feeling about their relationships and recent life events, including
domestic violence. They were also told that they did not necessarily need to have
experienced domestic violence in order to be eligible for the study. If they were
interested in participating, they were then told that they would be asked some questions
about themselves and their relationships over the telephone in a five-minute interview.
The screen was explained as an effort to ensure that the study included all different kinds
of women from the community, so that if they answered the questions in a similar way to
the women already enrolled in the study, they might not be eligible to participate. The
purpose of this telephone screen, which included the Conﬂict Tactics Scale (Straus,
1979), was to enroll equal numbers of battered and non-battered women in the study. For
the purpose of the study, women were categorized as “battered” if they had experienced
violence during pregnancy. Women who were categorized as “non-battered” by the
telephone screen were then asked if they had ever experienced violence in any romantic
relationship. Women were excluded if they had experienced violence in the past but
were not battered during pregnancy, in order to ensure that non-battered group would be
more likely to include women who had never experienced domestic violence. The
woman’s battering status for the purpose of this study was determined by her answers on
the SVAWS during the in-person interview; while this often was consistent with their
battering status according to the telephone screen, this was not true all of the time (e. g. a
woman who was categorized as “battered” on the telephone screen may not have met the
criteria for the “battered” category in the interview). Therefore, careful count of the

battering status of the completed interviewees was kept, and the telephone screen was

30

used to screen out women once there were enough “non—battered” participants to make up
the control group. Toward the end of the study, many women who were otherwise
eligible but did not screen as “battered” were excluded. Once women were determined to
be eligible, they were scheduled for an interview during the third trimester of their
pregnancy.

In the group of women who had experienced domestic violence with a previous
partner but not their current partner (n=45), over half the women in the study identiﬁed
themselves as White/Caucasian (65.2 %), with 21.7% identifying themselves as
Black/African American, 4.3% Latina, 3% Biracial, and 1% other. The average age of
women in this group was 25 years. In this group, 22% had a high school diploma or the
equivalent, and 18% had some high school education, 46.7% of the women had some
college education, and 11.1 % had a college or graduate degree; 43.5% were working
outside of the home at the time of the interview, and 37% had worked outside the home
in the past year; the mean monthly income was $2012. Half of the women who were not
currently experiencing domestic violence were married (50%); 32.6% were single, never
married, and 8% were either separated or divorced.

In the group of women who experienced domestic violence during their
pregnancy (n=61), similarly to the ﬁrst group, over half identified themselves as
White/Caucasian (65.6 %), with 24.6% identifying themselves as Black/African
American, 6.6% Biracial, and 3.3% Latina. The average age was 25 years. In this group,
34% had a high school diploma or equivalent, 25% had some high school education, 36%
had at least some college education, and 3.2 % had a college of graduate degree; 31%

were working out of the home at the time of the interview and 52% had worked outside

3l

of the home in the past year; the mean monthly income was $1173. Three quarters of the
women who were currently experiencing domestic violence were single, never married
(75.4%); 13.1% were married, and 7% were either separated, divorced, or widowed.

In summary, the two groups were demographically similar except on the variables
of income and marital status. Women in the group that had experienced domestic
violence during pregnancy were more likely to have a lower household income (p < .001)
and more likely to be single, never married, (p < .001) than women in the group that had

not experienced violence during the current pregnancy (see Table l).

32

Table l

Demogramtic Information on Study Participants

 

 

 

Characteristics Current Partner DV Previous Partner DV
n=61 n=45
Racial/Ethnic Group
Caucasian/White 65.6 % 64.4 %
African American/Black 24.6 % 22.2 %
Latina/Hispanic/Chicana 3.3 % 4.4 %
Biracial 6.6 % 6.7 %
Other 0 % 2.2 %
Marital Status
Single, never married 75.4 %* 31.1 %*
Manied 13.1 % 51.1%
Separated 8.2 % 8.9 %
Divorced 1.6 % 8.9 %
Widowed 1.6 % O %
Educational Status

Grades 7-12 59.0 % 40.0 %
Some College 36.1 % 46.7 %
Bachelor’s Degree 1.6 % 4.4 %
Graduate Degree 1.6 % 6.7 %
Other 1.6 % 2.2 %
*p < .001

33

 

Characteristics Current Partner DV Previous Partner DV

 

 

n=6l n=45
Mean
Income $1 173* $2012*
Age 25 25
*p < .001
Measures
Demggraphics.

 

A brief questionnaire was administered to obtain basic demographic information
such as marital or relationship status, ethnicity, parental education, parental employment,
number of children, and family income.

Dyadic Adjustment Scale. (DAS) (Spanier, 1976)

This is a 32-item, self-report scale that assesses the quality of a romantic
relationship. Only the current or most recent relationship were assessed with this
measure. The overall quality of dyadic adjustment scale were used. Examples of items
include “How often do you discuss or have you considered divorce, separation, or
terminating your relationship,” “Do you kiss your mate,” and “How often do you and
your mate engage in outside interests together.” Participants responded on a number of
different scales. In this measure, participants chose their level of agreement, based on a
6-point scale; they rated the amount of common outside interests on a 5-point scale; they

chose one of six statements that best described the future of their relationship, and they

34

rated the happiness of their relationship based on a 7-point scale. A coefficient alpha of
.95 was obtained in this study.

Maternal Fetal Attachment Scale. (MFAS) (Cranly, 1981)

The MFAS is a 24—item questionnaire developed to measure matemal-fetal
attachment behavior during pregnancy. It yields ﬁve sub-scales scores and one overall
score. For the purpose of his study, the overall score of matemal-fetal attachment was
used. Examples of items include “I talk to my unborn baby,” “I do things to take care of
myself that I would not do if I were not pregnant,” and “I try to picture what the baby will
look like.” Participants responded on a 5-item scale, ranging from “Deﬁnitely Yes” to
“Definitely No.” A coefﬁcient alpha of .74 was obtained in this study.

Attachment Style Questionnaire. (ASQ) (Feeney et al., 1994)

Adult romantic attachment style was assessed by using the Attachment Style
Questionnaire, a 40-item measure assessing attitudes about self and others in the context
of relationships. There are ﬁve sub-scales: 1) Conﬁdence; 2) Discomfort with Closeness;
3) Need for Approval; 4) Preoccupation with Relationships and 5) Relationships as
Secondary. Examples of items include “I find it hard to trust other people,” and “I am
too busy with other activities to put much time into relationships.” Respondents rated
their level of agreement on a 6-point scale ranging from “Totally Disagree” to “Totally
Agree.” Internal reliabilities ranging from .76-.84 have been reported for the individual
scales, with test-retest reliabilities of .67-.78 (Feeney et al., 1994). In this study, a
coefﬁcient alpha of .80 was obtained.

Romantic attachment style is rated on a dimensional scale, according the

individual’s combined ratings on the four scales. A “secure” proﬁle consists of high

35

scores on the Confidence subscale and low scores on everything else. An “insecure”
proﬁle indicates high scores on Discomfort, Approval, Preoccupation, and Relationships
as Secondary, and low scores on Confidence. For the purpose of this study, each
individual’s scores on the five dimensions was converted into a single score by
subtracting the higher of the scores on the two insecure dimensions from the individual’s
score on the Conﬁdence dimension.

Severity of Violence Against Women Scales. (SVAWS) (Marshall, 1992)

The type and severity of domestic violence, deﬁned here as male-to-female
violence by a romantic partner, was measured by the Severity of Violence Against
Women Scales. This is a 46-item questionnaire assessing violent behaviors and threats
the woman may have experienced. There are nine categories of abuse: symbolic
violence, threats of mild violence, threats of moderate violence, threats of serious
violence, mild violence, minor violence, moderate violence, serious violence, and sexual
violence. Respondents rated their experiences of abuse on a 4-point scale ranging from
“Never” to “Many Times.” Women were asked to complete the scale for the period of
current pregnancy, the year before pregnancy, and most recent previous relationship. In
this study, coefﬁcient alphas of 94-98 for the three time periods were demonstrated.

For the purpose of this study, a woman met criteria for being battered if she
endorsed any item in the following categories: threats of serious violence, mild violence,
minor violence, moderate violence, serious violence, and sexual violence. Women were
included in the “past battered but not current” group if they indicated that they had
experienced domestic violence in a previous romantic relationship but not in their current

relationship (as measured during their current pregnancy or the year prior to pregnancy).

36

Women were included in the “currently battered” group if they indicated they had
experienced domestic violence during their current or pregnancy, regardless of
experiences in previous relationships. Women were excluded who indicated that they
had experienced violence in their current relationship during the year prior to pregnancy
but not during their pregnancy. Each item in the SVAWS has a physical harm impact
weight according to the level of seriousness, abusiveness, and aggressiveness of the act.
Severity of violence was calculated by multiplying each item’s score by its physical harm
impact weight and summing the products (Marshall, 1992).
Procedures

Research assistants, both undergraduate and graduate level, were trained to
administer the questionnaires and to conduct a semi-structured interview. The training
period lasted approximately three months, and the trainees did several interviews under
supervision until they could conduct the interview according to a standard of 95% inter-
rater reliability. The interviewers were trained to maintain a neutral, non-judgmental
stance throughout the interview. They were also trained in how to maintain
conﬁdentiality and handle difﬁcult situations, such as the intrusion of partners or family
members who were unhappy about the subject’s participation in the study. The majority
of the interviews were conducted by a single interviewer, but a second interviewer would
accompany and observe the primary interviewer if the interview was to be conducted in
an unsafe neighborhood. The interviewer would begin by explaining the procedure,
telling the woman that the interview was completely conﬁdential, and obtaining the
participant’s written consent to take part in the interview. Participants were also told that

they could withdraw from the study at any time without any penalty or negative

37

consequences. The interviewers read all questions aloud and marked down the women’s
responses. The questionnaires were administrated orally to control for any variation in
the level of literacy among participants. Interviewers were blind to the battering status of
the woman; this was ensured by administering the questionnaires on domestic violence at
the end of the interview. Confidentiality was maintained by assigning all participants an
identiﬁcation number which was placed on all data rather than the participant’s name,
and the participant list was kept apart from the data. At the end of the interview,
participants were paid $50.00. All participants received a list of community resources

available for women and children.

38

RESULTS

This study tested three hypotheses about the effects of domestic violence on
attachment styles and the relationship between romantic and maternal attachment. Table
2 includes the characteristics of the variables measured by the Severity of Violence
Against Women Scales (SVAWS), the Dyadic Adjustment Scale (DAS), the Attachment
Style Questionnaire (ASQ), and the Matemal—Fetal Attachment Scale (MFAS). Table 3
shows the correlation for the data reduction done on the ASQ. The results for each of the
hypotheses are then presented.

Table 2

Summary Statistics for Variables

 

 

Variable Alpha Mean SD Possible Range in
range this study

SVAWS: Current .94 6.41 l 1.97 0-99 0-47

partner

SVAWS”: .98 19.24 28.56 0-99 0-80

Previous partner

DAS .95 99.05 26.06 0-151 20-140
ASQ .80 4.84 10.37 -54 to 48 -32 to 25
MFAS .74 97.66 8.26 28-1 16 79-115

 

39

Data Reduction

Because a single romantic attachment style score for each woman was calculated
from a combination of her scores on the five subscales of the Attachment Style
Questionnaire (ASQ), an analysis was done to assess whether the calculated attachment
style score correlated in the expected ways with the original ASQ subscale scores. As
can be seen in Table 3, correlations between the calculated score and the five subscales
were signiﬁcant and in the expected direction in each case, suggesting that the calculated
score was a valid indicator‘of the individual’s attachment style proﬁle as measured by the
ASQ.
Table 3
Correlations Between Calculated Romantic Attachment Style Score and ASQ Subscale

Scores: Both Groups (N=106)

 

Relationships Need for Discomfort Preoccupation Conﬁdence
as Secondary Approval with with

Closeness Relationships

 

Calculated —.560*** -.599*** -.818*** -.641*** .91 l***
romantic
attachment

style score

 

*** p< .001

Level of Engagement in Matemal-Fetal Attachment Behavior

The results of this study did not support the first hypothesis, that domestic

violence would be associated with a lower level of engagement in maternal fetal

40

attachment behaviors. A t-test was used to test for mean differences on the Maternal-
Fetal Attachment Scale (MFAS) between the two groups of women. AS can be seen in
Table 4, there was no difference in mean MFAS score between the group of women who
had experienced violence with a current partner during pregnancy and those who had
experienced violence with a previous partner.

Table 4

Results of t-test Comparing Mean MFAS Scores for Women Who Experienced Domestic

Violence (DV) With a Current Partner or With a Previous Partner

 

 

 

Group Mean MFAS score Std deviation
Group 1:

Current partner DV (n=61) 97.7 8.3

Group 2:

Previous partner DV (n=45) 97.6 8.3

t=.041

Romantic Attachment as a Mediator for the Effects of Domestic Violence on Maternal
Fetal Attachment Behavior

The hypothesis that adult romantic attachment style would mediate the effects of
domestic violence on matemal-fetal attachment behavior was partially supported by a
path-analytic model. See Tables 2 and 5 for variable characteristics and bivariate
correlations. All SEM analyses were conducted using Lisrel 8.3 (JOreSkog & SOrbom,
1999). Analyses were based on observed variables. A variety of Goodness of Fit indices

can be examined in determining model ﬁt; for the purposes of this study, the chi-square

statistic (X2) and the Goodness of Fit Index (GFI) were used, as recommended by Hoyle

41

and Panter (1995). The chi-square is a direct derivation of the fitting function which
indicates how much the model deviates from a perfect fit. The GFI is analogous to R2
used in multiple regression analyses, indicating the amount of variance accounted for by
the model. Overall, a small chi-square is desired, because it indicates non-significant
data-model discrepancies, and the CPI should be greater than .90.

Path models were tested separately for each group. The parameters associated
with the model for each group are shown in Figure 3, with 32 values shown in each
endogenous variable. The model ﬁt for Group 1 (Current DV) was X2(1, N=61) = .37, p
> .05, GFI = 1.00. The path coefficient for the effects of domestic violence on adult
romantic attachment style was significant ([3 = -.25, p = 0.05, one-tailed), and the path
coefﬁcient between adult romantic attachment style and matemal-fetal attachment
behavior was also signiﬁcant ([3 = .44, p =.001, one-tailed). The overall fit of the model
was good, supporting the hypothesis that adult romantic attachment serves as a mediator
for the effects of domestic violence on matemal-fetal attachment behavior. When the
indirect effect of domestic violence on matemal-fetal attachment behavior was tested, the
path coefﬁcient was insigniﬁcant ([3 = -.07), and the model was saturated so that it did not
yield statistics on the ﬁt of the data.

The model fit for Group 2 (Previous DV) was x2( 1, 18:45) = .79, 2 > .05, GFI
.99). The path coefficients were non-significant (see Figure 3) but the overall fit of the
model was good, suggesting that for this group adult romantic attachment does not serve
as a mediator between domestic violence and matemal-fetal attachment behavior. The
indirect correlation for severity of DV on matemal-fetal attachment behavior was .032.

As with Group 1, when the indirect effect of domestic violence on matemal-fetal

42

attachment behavior was tested, the path coefficient was insignificant ([3 = .13), and the
model was saturated so that it did not yield statistics on the ﬁt of the data.

Table 5

Bivariate Correlations for Severity of Domestic Violence (DV), Relationship

Satisfaction, Adult Romantic Attachment Style, and Matemal-Fetal Attachment Style

 

Variable Severity of violence Romantic Matemal—fetal

attachment style attachment Style

 

Group 1: Current partner DV (n=61)

Severity of violence
Romantic -.23
attachment style
Matemal— fetal -.16 .42***
attachment style
Group 2: Previous partner DV (n=45)
Severity of violence
Romantic -.20
attachment style
Matemal-fetal .08 .20

attachment style

 

*** p< .001

43

Relationship Satisfaction as a Moderator for the Effects of Domestic Violence on Adult
Romantic Attachment Style

The third hypothesis was that high relationship satisfaction would modify the
effects of domestic violence on adult romantic attachment only for women who had not
experienced domestic violence with their current partner. This was tested with
hierarchical multiple regressions, which were performed separately for each group.
Relationship satisfaction, severity of battering, and the interaction of these terms were
entered as predictors and adult attachment style as the outcome. The results offer partial
support for the hypothesis. For Group 1, as predicted, the beta weight for the interaction
term was not signiﬁcant (see Table 6). It was predicted that there would be a signiﬁcant
moderating effect of relationship satisfaction on the effects domestic violence on the
level of engagement in matemal-fetal attachment behavior for Group 2, however, this
was not supported by the analyses (see Table 7). Furthermore, although relationship
satisfaction was highly correlated with adult attachment style in both groups, the
moderator hypothesis was not supported because the interaction between relationship
satisfaction and severity of domestic violence accounted for little of the variance in adult
attachment style (see Tables 6 and 7). The analysis was also run using standardized

scores for all variables, with no difference in the results.

Table 6

Moderating Effects of Relationship Satisfaction on Adult Romantic Attachment Style

in Group 1: Current Partner DV

 

 

B AR’ P
Step 1 .35 .000
Severity of DV: Current partner -.097 .39
Relationship Satisfaction .557*** .000
Step 2 .003 .61
Severity of Current DV x .171 .61

Relationship Satisfaction

 

***p <.001.

Table 7

Moderating Effects of Relationship Satisfaction on Adult Romantic Attachment Style

in Group A Previous Partner DV

 

 

13 AR“ p
Step 1 .34 .000
Severity of DV: Previous partner —.132 .31
Relationship Satisfaction .550*** .000
Step 2 .003 .34
Severity of Previous DV x 1.48 .34

Relationship Satisfaction

 

***p <.001.

45

DISCUSSION

The results of this study provide limited support for the theory that internal
working models in adults are consistent within the individual across relationships, so that
a woman’s romantic attachment style may be associated with the development of her
attachment relationship with her child, even before the child is born. A mediated link
between domestic violence and the mother’s attachment to her unborn child was
demonstrated, providing preliminary support for the hypothesis that the negative effects
of domestic violence on the mother-child relationship may begin before birth.

Contrary to the ﬁrst hypothesis, this study found no difference in the level of
engagement in matemal-fetal attachment behaviors between the group of women who
had experienced domestic violence with a current partner during pregnancy and the
women who had experienced domestic violence with a previous partner. There are
several possible explanations for this finding. It may be that even in the context of
domestic violence experienced during pregnancy a woman may nonetheless preserve
hope and warm feelings for her unborn child and maintain a positive image of her child’s
future. However, other studies of parenting with older children have found that domestic
violence negatively affects maternal expression of warmth (Levendosky & Graham-
Berrnann, 2000), and is associated with an increase in maternal depression and parenting
stress (Levendosky & Graham-Bermann, 1998; Wolfe et al., 1985).

Another possible explanation for the difference between these findings and those
of other studies is that in this study matemal-fetal attachment was calculated as a linear
variable reﬂecting the level of engagement in certain behaviors as opposed to the

categorical variable yielded by the Strange Situation procedure. The endorsement of a

46

high level of engagement in attachment behaviors by this sample is contrary to the
findings of one of the few studies that has been done on domestic violence and infant
attachment style in a hi gh-risk sample. Zeanah, et al. (1999) found a high rate of
disorganized infant attachment (56.9%) in a sample that included low-income women,
over 60% of whom had experienced domestic violence with a current or a previous
partner . If level of engagement in matemal-fetal attachment behavior is related to infant
attachment style, it would be expected that a high-risk sample, such as the one used for
this study, would contain a number of women who would have low scores on the MFAS,
corresponding with insecure infant attachment.

A restricted range on the level of violence experienced during pregnancy may
have contributed to the similarity of the two groups’ MFAS scores. The mean score on
the SVAWS for the group that had experienced violence during pregnancy was relatively
low (see Table 2), so that perhaps the level of violence that Group I experienced during
pregnancy was mild enough not to significantly impact that group’s level of engagement
in attachment behaviors. The mean score of severity of violence for Group 2 was
approximately three times the mean severity of violence experienced by Group 1 (see
Table 2). Perhaps experiencing severe violence in the past and experiencing mild
violence during pregnancy has a similar impact on MFAS score. There may be an
interaction between the severity of the violence and the immediacy or duration of the
violent relationship. In other words, the lingering effects of severe violence with an ex-
partner may be similar to the effects of milder but more recent violence with a current
partner. Future studies could assess this by using a comparison group of women who had

never experienced domestic violence. By comparing the MFAS scores of women who

47

had experienced current and past partner violence with the scores of women who had
never experienced violence, it would be possible to assess whether the experience of
domestic violence at any time made a difference in the MFAS score.

Finally, the lack of difference between the two groups scores may be a reﬂection
of their common experiences. Both groups of women shared a history of domestic
violence. Perhaps the signiﬁcant factor was that they had experienced domestic violence
at some point, rather the degree of severity or whether it was with a current or previous
partner. However, it is difﬁcult to assess this from these results, given the restricted
range in both the SVAWS and the MFAS.

The second hypothesis, that adult romantic attachment style would mediate the
relationship between domestic violence and matemal-fetal attachment behavior, was
partially supported by the results of this study. A path-analysis model of Group 1 found
that adult romantic attachment mediated the relationship between domestic violence and
matemal-fetal attachment behaviors. For women in Group 2, who had only experienced
violence with a previous partner, the path analysis did not show romantic attachment as a
mediator. This may be because violence with an ex-partner may have less impact than
current violence on a woman’s psychological functioning, and by extension, on her
parenting. A recent study on domestic violence and parenting found that the effects of
violence did not directly impact the women’s parenting but affected it indirectly,
mediated through the woman’s psychological functioning (Levendosky & Graham-
Berrnann, in press). It may be the case in this study that adult romantic attachment is a
component of psychological functioning, and therefore mediates the effects of current

domestic violence on matemal-fetal attachment. It may not mediate the relationship

48

between past experiences of domestic violence because the current non-violent
relationship has more effect on the woman’s adult romantic attachment style than past
experiences of violence. This would be consistent with the findings of a study on marital
relationships and parenting that indicated that the parents’ relationship at the time of the
birth of their child was more predictive of their parenting style than their relationship
style prior to the birth (Lindahl et al., 1997).

The idea that a woman’s romantic attachment style may reﬂect the inﬂuence of a
current positive relationship over the inﬂuence of past experiences of violence suggests a
resilience in adult attachment style. In the context of a new, non-violent relationship the
woman may rework her internal working models. This is consistent with a recent
longitudinal study on adult romantic attachment, which found that attachment style was
related to relationship status and changed from insecure to secure after the beginning of a
new relationship (Davila et al., 1997).

The third and ﬁnal hypothesis of this study was that relationship satisfaction with
a current partner would moderate the effects of ex-partner violence on adult romantic
attachment style if the woman was not experiencing violence with that current partner.
However, the results indicated that relationship satisfaction did not moderate the effects
of domestic violence on adult romantic attachment style for either group. As can be seen
in Tables 6 and 7, in both groups relationship satisfaction accounted for much of the
variance in romantic attachment style. Domestic violence, whether with a current or a
previous partner, accounted for very little of the variance, making it difﬁcult to test

whether relationship satisfaction might be acting as a moderator.

49

Other studies of adult romantic attachment have found that adult attachment is
closely related to relationship satisfaction. A study by Feeney and Noller (1990) found
that adult romantic attachment style correlated highly with subjects’ ratings of attitudes
towards love and romantic partners, while Collins & Read (1990) found that attachment
style was strongly related to each partner’s perceptions of the quality of their relationship.
The ﬁnding in this and other studies that relationship satisfaction and adult romantic
attachment style are highly correlated poses the question that they might be measuring
the same construct. Is adult romantic attachment just another measure of current
relationship functioning rather than an assessment of enduring beliefs about self and
relationships? If the DAS and the ASQ were both assessing positive or negative attitudes
toward romantic relationships, it would be difﬁcult for one to moderate the other.

Despite their close association, there is evidence that relationship satisfaction and
adult attachment style are not necessarily measuring the same construct. For example,
Feeney & Noller (1990) found that although adult romantic attachment was closely
associated with relationship quality, it was more highly correlated with generalized views
of the self and relationships than were the subjects’ ratings of attitudes toward romantic
love. Kirkpatrick & Davis (1994) found that attachment style predicted relationship
stability even between partners who gave low ratings of satisfaction with the quality of
their relationship, supporting the idea that attachment style is something different than a
measure of relationship satisfaction. In this study, when an analysis was done using the
whole sample, the DAS and the ASQ correlated in a similar way with many variables
such as anxiety, depression, and self-esteem, as would be expected if they were

measuring the same construct. However, they differed in their correlation with another

50

attachment measure: a significant relationship was found between romantic attachment
and matemal-fetal attachment but go_t between relationship satisfaction and matemal-fetal
attachment.

Measuring attachment style with a self-report instrument may be problematic, as
attachment style is an unconscious process (Bowlby, 1973), a difﬁcult thing to ask
someone to report on themselves. This is a concern with much of the research on adult
romantic attachment styles, which has primarily relied upon self-report measures
(Bowlby, 1973; Collins & Read, 1990; Davila et al., 1997; Feeney et al., 1994; Hazan &
Shaver, 1987). Because studies of adult romantic attachment style have primarily used
self-report questionnaires, it is difﬁcult to compare romantic attachment to the extensive
literature on infant attachment and internal working models, which utilize behavior
observation and extensive clinical interviews respectively, and are coded according to a
complex system which takes into account internal processes and representations
(Ainsworth, 1978; Main & Goldwyn, 1984; Schuengel et al., 1999; van Uzendoom,
1995). Another difﬁculty faced by this study in comparing these two types of attachment
is that the comparison is between attitudes toward relationships (romantic attachment)
and specific behaviors (matemal-fetal attachment).

The ﬁndings in this study contribute to the growing evidence that domestic
violence may affect the mother-child relationship through its impact on the mother, even
before the birth of the child. The results support the hypothesis that for women who are
currently experiencing domestic violence, adult romantic attachment mediates the impact
of the violence on the developing matemal-fetal relationship. These findings provide

another way of understanding the implications of domestic violence on children. In

51

addition to what is already known about the direct effects that domestic violence may
have on the health of the mother and her fetus, it may indirectly affect the child through
the mother’s parenting. It will be important for future researchers to consider the indirect
effects of domestic violence that are mediated by the mother’s romantic attachment to her
partner and transmitted through her attachment to the child.

The results of this study provide support for the theoretical link between adult
romantic attachment and maternal attachment to the fetus, however, the limited range
yielded by some measures is indicative of some of the problems inherent in assessing
attachment with self—report measures. Social desirability may affect the results, and
unconscious processes may be missed. More research is needed to clarify the ways in
which adult romantic attachment style differs from relationship satisfaction.
Furthermore, it is difﬁcult to test the theory that an individual’s attachment style endures
throughout their development if attachment is measured differently at each life stage.
The self-report measures that dominate the ﬁeld of adult attachment may not be the best
way to address these issues.

One of the strengths of this study was that it utilized a community sample of
women from a broad range of socioeconomic and ethnic backgrounds, making the results
more generalizable than those of studies whose samples are limited to women in shelters.
The women in this study experienced primarily mild to moderate violence. While this
makes it more difﬁcult to assess the impact of domestic violence, what is lost in statistical
power is made up for by the possibility that the results are more relevant to the
experiences of women in the community, as mild to moderate violence has been found to

be more common than severe domestic violence (Straus & Gelles, 1990; Tjaden &

52

Thoennes, 2000). The nature of the community sample makes the results more clinically
relevant as well. Future research in attachment and domestic violence should continue to
include a broad based community sample to best understand the effects of domestic
violence on the lives of women and children.

The clinical implications of these findings apply to interventions with domestic
violence and with attachment problems. First, it is clear that in order to protect children
from the damaging effects of domestic violence, it is necessary to protect women from
these effects even before the child is born. Interventions aimed at preserving the mother-
child relationship should consider the mother’s adult attachment style and environmental
factors that may protect or harm it. The results of this study suggest that violent romantic
relationships are a factor in both adult romantic and matemal-fetal attachment. However,
the results from the control group of women who experienced violence with a previous
partner indicate that the impact of domestic violence on attachment style may be
diminished when the woman leaves that relationship and enters a non-violent romantic
relationship. This suggests some resilience in adult attachment style and offers a hopeful
note for interventions aimed at helping women recover from domestic violence.

As with the harmful effects of domestic violence, the results of this study suggest
that the effects of attachment problems on the mother-child relationship may begin before
the child is born or even conceived, as the maternal attachment is inﬂuenced by adult
romantic attachment style. Future research into attachment problems should consider
further assessment of the connection between adult romantic attachment and matemal-

fetal attachment. If adult romantic attachment is responsive to changes in the

53

environment, specifically changes in relationship status, this would be an important area

in which to intervene to prevent future mother-child attachment difﬁculties.

54

APPENDICES

55

Appendix A

MOTHER-INF ANT STUDY
CONSENT FORM

This study is part of a survey of women in Michigan, some of whom may be experiencing
domestic violence. We hope to learn about the strengths that you bring to your situation, your feelings and
perceptions of your baby during pregnancy, and your relationships with others. including family members.
partners, and friends. We hope to use this information to help plan better programs for families
experiencing domestic violence.

If you decide to take part in the survey, you will be asked questions about how you have been
feeling recently, events that have happened to you, your feelings about pregnancy and your baby, the
people in your life who provide support to you, and your memories of your childhood. This will take a total
of 2-3 hours.

All information will be kept strictly conﬁdential. Your name will be removed from all
questionnaires and an identiﬁcation number will be put on them instead. All questionnaires will be kept in
locked ﬁle cabinets in a locked ofﬁce. Your identity will not be revealed in any reports written about this
study. We will summarize information from all study participants and will not report information about
individuals. The only exception is in the case of ongoing child abuse. If you indicate that child abuse is
occurring in your household, we are required to make a report to Protective Services.

You have the right to withdraw from this study at any point during the interview with no penalty
or negative consequences. Your decision about whether to participate or not will not affect your
relationship with any agencies or Michigan State University. If you have any questions, please ask us. If
you have questions later, you can contact Dr. Anne Bogat or Dr. Alytia Levendosky at (517) 432-1447.

We are also interested in recontacting you about 2 months after the birth of your baby by
telephone and then we would like to meet with you and your baby at 12 months after the birth of your baby.
50 at the end of the interview today, we will ask you for information that will help us keep in contact with
you. Your participation at @ time does not obligate you to participate in the second telephone
appointment, or the third interview. You will be paid $50 for the ﬁrst interview, mailed a baby gift after
the telephone interview, and $75 and a baby gift for the third interview, if you wish to participate.
*****It****************************************************************#*******

I have read and understood the above statements. I understand that my participation in this study is
completely voluntary and that I can withdraw from this study at any time without penalty or negative
consequences.

Signature of Participant Date

Witness Date

Anne Bogat, Ph.D. Alytia Levendosky, Ph.D.
Michigan State University Michigan State University
Department of Psychology Department of Psychology
123 Snyder Hall 121 Snyder Hall

East Lansing, MI48824-1117 East Lansing, MI 48824-1 117

56

Appendix B

Fliers

      

HAVE YOU BEEN HURT BY SOMEONE YOU LOVE?

ARE YOU PREGNANT AND HAVE YOU BEEN PUSHED
OR GRABBED OR HIT OR SLAPPED OR KICKED (OR
WORSE) BY A PARTNER OR BOYFRIEND DURING
YOUR PREGNANCY?

 
   
      
    

We need women to take part in an interview about their lives and
their pregnancies.

 
   

  

- Interview can be done at MSU or at your home.

0 You will be paid $50.00 in cash.

' All information is kept completely confidential.
!! $50.00 !!

********************************************

 
   
     
 

If you are interested or would like more information,
please call 432-1447 and ask for

Mother-Infant Study

      
 

 

 

 

 

 

 

MOTHER- MOTHER- MOTHER- MOTHER- MOTHER-
INFANT INFANT INFANT INFANT INFANT
STUDY STUDY STUDY STUDY STUDY
432-1447 432-1447 432-1447 432-1447 432-1447

 

57

 

ARE YOU PREGNANT?

YOU MAY BE ELIGIBLE TO PARTICIPATE IN A STUDY
ABOUT
MOTHER-INFAN T RELATIONSHIPS

!! $50.00 I!
We are looking for pregnant women due before April 1, 2000 to

participate in a research study at Michigan State University. You
will be asked about experiences and feelings during pregnancy,

perceptions of your infants, and recent life events.

- Interview can be done at MSU or at your home.
- You will be paid $50.00 in cash.
0 All information is kept completely confidential.

!! $50.00 !!

*************>l<**>l<***************************

If you are interested or would like more information,
please call 432-1447 and ask for

Dr. Anne Bogat’s Mother-Infant Study

 

58

7.

Appendix C

Pregnancy Interview
Demographic Questionnaire

Your date of birth: /_ /

(m0) (dy) (yr)

Your baby’s due date: / /

(mo) (dy) (yr)
Have you been pregnant before? (Circle one)
1 = YES 2 = NO (If NO. go to Question 7)
If YES, to Question 3:
4. How many times?

5. Have you had any miscarriages, still births, or abortions? (Circle one)
1 2 YES
2 = NO

6. How many biological children do you currently have?

How many people, including yourself, live in your household?
(If participant is living in a shelter, questions 7 & 8 refer to household composition before

moving into shelter.)

8.

Please list these: (Write in speciﬁc relationship to mother. Be speciﬁc--is the person (for ex.) a

husband, stepfather, biological child, foster child, or partner's child?)

9.

 

 

 

 

 

 

 

 

 

Choose the one that best describes your current marital/relationship status (choose only one):

(a) single, never married (see below)

(b) married For how long? __ (in months)

(c) separated For how long? __ (in months)

((1) divorced For how long? __ (in months)

(e) widowed For how long? _ (in months)

If (a) is circled: Are you currently in a relationship? YES NO

If YES, go to Question 10.

59

If NO. were you in a relationship that lasted at least

 

 

 

 

6 weeks during your current pregnancy? YES NO
10. First name of your current partner or the partner you were with for at least 6 weeks during your
pregnancy:
1 1. Are you currently living with your partner/spouse? (Circle one)
1 = YES
2 = NO
12. If yes to Question 1 1. how long have you been doing so? (Circle one)
1 = less than 1 year
2 = 1-3 years
3 = 4-6 years
4 = 7-9 years
5 = 10-12 years
6 = 13-15 years
7 = 16 - 18 years
8 = 19 - 21 years
9 = 22 - 24 years
10 = 25 or more years
13. Prior to your current romantic relationship, speciﬁed in Question #10
(a) were you ever married? 1 = YES 2 = NO
(b) did you ever live with a partner? 1 = YES 2 = NO
(c) were you ever separated? 1 = YES 2 = NO
((1) were you ever divorced? l = YES 2 = NO
(e) were you ever widowed? 1 = YES 2 = NO
14. What is your current relationship with the father of your baby? (Circle one)
1 = spouse
2 = ex-spouse
3 = partner
4 = ex-partner
5 = friend
6 = acquaintance
7 = stranger
8 = other Please specify:
15. What is your racial or ethnic group? (Circle one)
1 = Native American
2 = Asian American/Paciﬁc Islander
3 = Black, African American
4 = Latino, Hispanic, Chicano
5 = Biracial (mixed): Specify
6 = Caucasian, White
7 = Other:
16. What is the baby's father's racial or ethnic group? (Circle one)

1 = Native American

2 = Asian American/Paciﬁc Islander
3 = Black, African American

4 = Latino, Hispanic, Chicano

5 = Biracial (mixed): Specify
6 = Caucasian, White

7 = Other:

 

 

60

 

 

 

 

 

 

 

 

17. What is the highest level of education you have completed? (Circle one)
1 = grades 1, 2, 3, 4, 5, or 6 (circle specific grade)
2 = grades 7, 8. 9, 10, 11, 12, GED (circle specific grade)
3 = some college Where?
4 = AA degree Where?
5 = BA/BS Where?
6 = some grad school Where?
7 = graduate degree Where?
MA?
Ph.D.?
Law?
MD?
8 = other; Specify (e.g., Beauty School, nursing school)
18. Do you currently work outside the home? YES NO

19. If YES to either part of Question 18, what is/was your occupation?

IfNO, did you work outside the home during the last year? YES NO

 

0 Please be speciﬁc. For example, bookkeeper, cashier, computer programmer.
0 If there were two jobs/occupations, have participant choose the one that she feels best
represents her occupation.

20. What is the highest level of education your partner/spouse has completed? (Circle one)

21.

22.

23.

24.

25.

1 = grades 1, 2, 3. 4, 5, or 6 (circle specific grade)
2 = grades 7, 8, 9, 10, 1 l, 12, GED (circle speciﬁc grade)

 

 

 

 

 

 

3 = some college Where?
4 = AA degree Where?
5 = BA/BS Where?
6 = some grad school Where?
7 = graduate degree Where?

MA?

Ph.D.?

Law?

MD?

 

8 = other; Specify (e.g., Beauty School, nursing school)

 

Does S/he work outside the home? (Circle one)
1 = YES
2 = NO

If yes to Question 20, what is his/her occupation?

 

(Please be speciﬁc)

What is your total family income per month (estimate)?

 

Do you currently receive any public assistance? (Circle one)
1 = YES
2 = NO

Are you currently residing in a shelter for battered women? YES NO # days?

61

26. Have you ever stayed in a shelter for battered women before . . .
(a)Because of your experience of abuse? YES NO
(b)Because of your mother's/guardian’s

experience of abuse? YES NO

27. Have you ever stayed in a homeless shelter before . . .

 

(a)Because of your experiences? YES NO

(b )Because of your parents’lguardians’
experiences? YES NO

62

# days?

# days?

# days?

# days?

Appendix D

Maternal-Fetal Attachment Scale

Please respond to the following items about yourself and the baby you are expecting. There are no
right or wrong answers. Your ﬁrst impression is usually the best reﬂection of your feelings.
Make sure you mark only one answer per sentence.

I think or do the following: Deﬁniter Yes Yes Uncertain N2 Deﬁniter No
2

1.

PWﬂQVPP’N

1 3 4 5
I talk to my unborn baby.

I feel all the trouble of being pregnant is worth it.

I enjoy watching my tummy jiggle as the baby kicks inside.

1 picture myself feeding the baby.

I'm really looking forward to seeing what the baby looks like.

I wonder if the baby feels cramped in there.

I refer to my baby by a nickname.

I imagine myself taking care of the baby.

I can almost guess what my baby’s personality will be from the way she/he moves around.

. I have decided on a name for a girl baby.

. I do things to try to stay healthy that I would not do if I were not pregnant.
. I wonder if the baby can hear inside of me.

. I have decided on a name for a boy baby.

. I wonder if the baby thinks and feels “things" inside of me.

. I eat meat and vegetables to be sure my baby gets a good diet.

. It seems my baby kicks and moves to tell me it’s eating time.

. I poke my baby to get him/her to poke back.

. I can hardly wait to hold the baby.

. I try to picture what the baby will look like.

. I stroke my tummy to quiet the baby when there is too much kicking.
. I can tell that the baby has hiccoughs (hiccups).

. I feel my body is ugly.

. I give up doing certain things because I want to help my baby.

. I grasp my baby’s foot through my tummy to move it around.

63

Appendix E

Attachment Style Questionnaire

Show how much you agree with each of the following items by rating them on this scale:

Totally Disagree Strongly Disagree Slightly Disagree Slightly Agree Strongly Agree Totally Agree
2

1

l.

2.

10.
ll.
12.
l3.
14.
15.
16.
I7.

18.

3 4 5 6
Overall, I am a worthwhile person.
I am easier to get to know than most people.
I feel conﬁdent that other people will be there when I need them.
I prefer to depend on myself rather than other people.
I prefer to keep to myself.
To ask for help is to admit that you're a failure.
People’s worth should be judged by what they achieve.
Achieving things is more important than building relationships.
Doing your best is more important than getting on with others.
If you've got a job to do, you should do it no matter who gets hurt.
It's important to me that others like me.
It's important to me to avoid doing things that others won't like.
I ﬁnd it hard to make a decision unless I know what other people think.
My relationships with others are generally superﬁcial.
Sometimes I think I am no good at all.
I ﬁnd it hard to trust other people.
I ﬁnd it difﬁcult to depend on others.

I ﬁnd that others are reluctant to get as close as I would like.

Totally Disagree Strongly Disagree Slightly Disagree Slightly Agree Strongly Agree Totally Agree

1

2 3 4 5 6

I ﬁnd it relatively easy to get close to other people.

I ﬁnd it easy to trust others.

I feel comfortable depending on other people.

I worry that others won’t care about me as much as I care about them.

I worry about people getting too close.

I worry that I won't measure up to other people.

I have mixed feelings about being close to others.

While I want to get close to others, I feel uneasy about it.

I wonder why people would want to be involved with me.

It's very important to me to have a close relationship.

I worry a lot about my relationships.

I wonder how I would cope without someone to love me.

I feel conﬁdent about relating to others.

I often feel left out or alone.

I often worry that I do not really ﬁt in with other people.

Other people have their own problems, so I don't bother them with mine.
When I talk over my problems with others, I generally feel ashamed or foolish.
I am too busy with other activities to put much time into relationships.
If something is bothering me, others are generally aware and concerned.
I am conﬁdent that other people will like and respect me.

I get frustrated when others are not available when I need them.

Other people often disappoint me.

65

Appendix F

Dyadic Adjustment Scale

*****This questionnaire refers to [NAME, see Page 2, Question 10].*****
Most people have disagreements in their relationships. Please indicate below the approximate extent of
agreement or disagreement between you and NAME for each item on the list by placing an “X” in the
appropriate box.

 

 

 

 

 

 

 

 

 

 

 

 

 

Please give the dates of this relationship: / to /
(mo) Ur) (m0) (yr)
If on-going relationship, leave 2'“l mo/yr blank.
3‘ >.
"a 8 E E a. E
w -. 2a as --3 as
3‘3 833 3.52 gas 8350 —.a
.E a E .2 a 8 D a 0 .5. 2 .2 < 0
<< <<< O <<o
1. Handling family ﬁnances 5 4 2 I 0
2. Matters of recreation 5 4 3 2 t 0
3. Religious matters 5 4 3 2 1 0
4. Demonstration of affection 5 4 3 2 I 0
5. Friends 5 4 3 2 1 0
6. Sex relations 5 4 3 2 1 0
7. Conventionality (correct or 5 4 3 2 l 0
proper behavior)
8. Philosophy of life 5 4 3 2 l 0
9. Ways of dealing with parents 5 4 3 2 l 0
or in-laws
10. Aims, goals, and things 5 4 3 2 l 0
believed important

 

 

 

 

 

 

 

 

 

66

ll.

12.

13.

14.

15.

16.

I7.

18.

19.

20.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

>5
— >4
:8 8:3 5% ea 8:3 2%
3321i 2381 ._,., so 83cc am
—< --—< {Ge we ——.9 —.e
< < < 8 a E a < < a < a
0
Amount of time spent together 5 4 2 l 0
Making major decisions 5 4 3 2 l 0
Household tasks 5 4 3 2 I 0
Leisure time interests and 5 4 3 2 l 0
activities
Career decisions 5 4 3 2 l 0
How often do you and your partner engage in the following activities:
0 >4
5 E 5 .. =5
'12 - -- «1 ° = b 33
o 8 E; 3 g .9 E5 5
5 2 5 ‘5 2 § a: Z
= u— 2
< 0 5
How often do you discuss or have you 0 1 2 4 5
considered divorce, separation, or
terminating your relationship?
How often do you or your mate leave 0 l 2 3 4 5
the house after a ﬁght?
In general, how often do you think that 5 4 3 2 l 0
things between you and your partner are
going well?
Do you conﬁde in your mate? 5 4 3 2 l 0
Do you ever regret that you got 0 1 2 3 4 5
married? (Or lived together?)

 

 

 

 

 

 

 

 

67

 

 

21.

22.

23.

24.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

How often do you and your partner 2 3 4 S
quarrel?
How often do you and your mate “get on 2 3 4 5
each other’s nerves?”
>4 >4 Tu
CU ... <6
0 g a 8 2.: b
I... __ - a)
‘>’ < 2 8 a: z
m LL] 0
Do you kiss your mate? 4 3 2 l 0
h— '4— 3 s...
o
’5 E .‘3 E ... E e E 3 E
= o m o E 0 >45 t: o
< ‘5 O 5 o 5 33' s... O '5
2 VJ > 0 Z
Do you and your mate engage in outside 4 3 2 I 0

interests together?

 

 

 

 

 

 

These are some things about which couples sometimes agree, and sometimes disagree. Indicate if either
item below caused a difference of opinions or were problems in your relationship during the past few
weeks. (Check yes or no)

25.

26.

 

 

 

 

Yes No
Being too tired for sex 0 1
Not showing love 0 I

 

 

 

 

How often would you say the following events occur between you and your mate?

27.

 

 

 

J: O Q)
a E .2 .2 >- r:
... as o 3 5 3 x '8 cg
O .r: E "' r: "’ 0 o
> H l— o I- 4) CU
o w m 0 E C 3 o 0
Z w o 0 0 U S
3 Q U «I 0 G3 =
r: = = O 2
O O 0
Have a stimulating exchange of 0 l 2 3 4 5

ideas

 

 

 

 

 

 

 

68

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

28. Laugh together 0 l 2 3 4 5
29. Calmly discuss something 0 l 2 3 4 5
30. Work together on a project 0 l 2 3 4 5
31. Which of the following statements best describes how you feel about the future of your
relationship? (Check only one)
5 I want desperately for my relationship to succeed, and would go to almost any length to see that
it does.
4 I want very much for my relationship to succeed, and will do all that I can to see that it does.
3 I want very much for my relationship to succeed, and will do my fair share to see that it does.
2 It would be nice if my relationship succeeded, but I can’t do much more than I am doing now to
help it succeed.
I It would be nice if it succeeded, but I refuse to do any more than I am doing now to keep the
relationship going.
0 My relationship can never succeed, and there is no more that I can do to keep the relationship
going.
32. The boxes below represent different degrees of happiness in your relationship. The middle
point, ‘happy.’ represents the degree of happiness of most relationships. PLEASE CHECK
THE BOX WHICH BEST DESCRIBES THE DEGREE OF HAPPINESS, ALL THINGS
CONSIDERED, OF YOUR RELATIONSHIP.
Extremely Fairly A little Happy Very Happy Extremely Perfect
_U_nhappy Unhappy mhappy Happy
0 1 2 3 4 5 6

 

 

 

 

 

 

 

69

 

Appendix G

VAW Scales--Pregnancy Interview, Pt. I
””‘This questionnaire refers to (NAME, see Page 2, Question 10].“"‘
You and have probably experienced anger or conﬂict. Below is a list of behaviors he may have done.
Describe how often he has done each behavior at 2 different times (during your current pregnancy and the year before
you became pregnant) by choosing a letter from the following scale. (Interviewer.- If participant was not lilt'olt ed m a
relationship with during the year before she became pregnant. code E /

 

 

A BB C D E
never once a few times many times not applicable
Du ring your current pregnancy: During your current pregnancy:
The year before you became pregnant: The year before you became pregnant:
Hit or kicked a wall. door or furniture Spanked you
_‘I'hrew, smashed or broke an object ___Bit you

Driven dangerously with you in the car Slapped you with the palm of his/her hand

_Threw an object at you Slapped you with the back of his/her hand

_ _5h00k a ﬁ"8“ 3‘ you Slapped you around your face and head
Made threatening gestures or faces at you Kicked you
_ _5h00k 8 f1“ 3‘ you Hit you with an object
_Acted like a bully toward you Stomped on you
Destroyed something belonging to you Choked you
_ _‘l'hreatened to harm or damage things __ _Punched you

you care about
_Burned you with something

Threat: edt d tro
n 0 CS 1’ property Used a club-like object on you

__ _Threatened someone you care about

__ _Beat you up
— _Threatened to hurt you _ _Used a knife or gun on you
—‘ _Threatened to kill himself __ _Demanded sex whether you wanted
_ __Threatened you with a club-like object 3° 0’ “0‘
__ _Threatened you with a knife or gun _ _Made you have oral sex against your will
_ _Threatened ‘0 kl" you _ _Made you have sexual intercourse

against your will

_ _Threatened You With a weapon Physically forced you to have sex

Aer dl'k h want dt kill
e i e e e 0 you _Made you have anal sex against your will
__ _Held you down. pinning you in place
_Used an object on you in a sexual way
_ _Pushed or shoved you
__ _Shook or roughly handjod you __ _Grabbed you suddenly or forcefully

Scratched you
Pulled your hair

Twisted your arm

VAWS--Pregnancy Interview, Pt. II
How often did your most recent previous partner (the person before [NAMED engage in each of these activities with you?
I lntenrewer: Relationship with previous partner must have lasted 6 weeks or longer in order to complete questionnaire. 1

 

Please give the dates of this relationship: I to /
(m0) (yr) (mo) (yr)
A BB C D E
never once a few times many times not applicable/no previous partner

How often did vour previous partn_e£
_Hit or kicked a wall, door or furniture
_Threw, smashed or broke an object
___Driven dangerously with you in the car
_Threw an object at you

_Shook a ﬁnger at you

Made threatening gestures or faces at you

Shook a ﬁst at you

 

Acted like a bully toward you

 

 

Destroyed something belonging to you

Threatened to harm or damage things you
care about

_Threatened to destroy property
_Threatened someone you care about
_Threatened to hurt you
_Threatened to kill himself
____Threatened you with a club-like object
_Threatened you with a knife or gun

_Threatened to kill you

Threatened you with a weapon

 

Acted like he wanted to kill you

 

Held you down, pinning you in place
_Pushed or shoved you
Shook or roughly handled you

Scratched you

 

Pulled your hair

 

Twisted your arm

71

How often di_d your previous partner:

Spanked you

 

____Bit you
_Slappcd you with the palm of his/her hand

Slapped you with the back of his/her hand

 

Slapped you around your face and head

 

Kicked you

 

_Hit you with an object

 

Stomped on you
Choked you

 

Punched you

 

Burned you with something
Used at club-like object on you

Beat you up

 

_Used a knife or gun on you
_Demanded sex whether you wanted to or not
_Made you have oral sex against your will

Made you have sexual intercourse

against your will

Physically forced you to have sex
_Made you have anal sex against your will
_Used an object on you in a sexual way

_Grabbed you suddenly or forcefully

Were you ever pregnant during the time that any of these
events occurred? ( l ) yes (2) no (5) n/a

Did your mother/guardian ever experience
any of these events with one of her partners?
(1) yes (2) no (5) don tknow

.muogazon EoEcomzm _Sﬂ-_mE2aE .5 8:20; 05088 no ﬂooto 2: he 8662: a 3 .58588 2:588 :35. ._ ouswi

 

8320mm Eon—€32
30935032

 

 

 

 

 

25m acoEnoaz<
0:52:3— :33

 

 

 

 

8535 2.8809
“.0 bto>om

 

 

 

 

72

0.5m EoEcomzm 352:8 :35 :o 8533 052:8 .3 38:0 2: .8 5:20on a mm 553me qucocﬂom .N 233m

 

25m Eoﬁnoﬁz
ozcanm :35~

 

 

 

 

 

 

552E £5395
”ooze—o5
osmoEoQ

h.o bto>om

 

 

 

 

:ozoﬂmcmm
Enwcozﬂom

 

 

 

25m Eo65§<
0:583— 253..

 

 

 

 

 

 

555E €250

“coco—o; 058.com

mo bto>om

 

 

 

 

sesame—mtg
9:20:23—

 

 

73

.3238. .52.: ..ovat

$2233 E05583

38-3589: can £73m Ens—cogs 2.52:9 56a .8220; 0:358 .8 553m ”N 58 _ 3:20 .8 385 gingham .m 053m

 

Snag

cesium 595—052
Eon—45582

 

w_.

 

 

 

 

 

swam

532$ 28.5.22
agent—$5532

 

ozua

“wow—VV.

 

comm

2.3m 5.25.22?
0:55: :53

w_.-

 

 

 

 

 

 

 

 

mod nu

 

3...?

2.9m E05582
05:25: ==c<

_N._-H~

*_WN.I

855$
£5395 “>0 ..o bto>om

 

 

 

74

 

 

 

 

8..- no

355$
555.0 ”>0 ‘8 5.55m

 

 

 

REFERENCES

75

REFERENCES

Ainsworth, M. D. S., Blehar, M.C., Waters, E., & Wall, S. (1978). Patterns of
attachment: A psychological study of the Strange Situation. Hillsdale, NJ: Erlbaum.

Bartholomew, K., & Horowitz, L. M. (1991 ). Attachment styles among young
adults: A test of a four-category model. Journal of Personality and Social Psychology,
_6_1_(2), 226-244.

Beckwith, L., Cohen, S. E., & Hamilton, C. E. (1999). Maternal sensitivity
during infancy and subsequent life events relate to attachment representation at early
adulthood. Developmental Psychology, 35(3), 693-700.

Benoit, D., & Parker, K. (1994). Stability and transmission of attachment across
three generations. Child Development, 65, 1444-1456.

Bowlby, J. (1973). Attachment and loss: Vol. 3. Separation, anxiety, and anger.
New York: Basic Books.

Bowlby, J. (1988). A secure base: Parent-child attachment and healthy human
development. New York: BasicBooks.

Browne, A. (1993). Violence against women by male partners. American
Psychologist, 48, 1077-1087.

Campbell, J. C., & Lewandowski, L. A. (1997). Mental and physical effects of
intimate partner violence on women and children. The Psychiatric Clinics of North
America 20(2), 353-374.

 

 

Cokkinides, V. E., & Coker, A. L. (1998). Experiencing physical violence
during pregnancy: Prevalence and correlates. Family Community Health, 20(4), 19-37.

Collins, N. L., & Read, S. J. (1990). Adult attachment, working models, and
relationship quality in dating couples. Journal of Personality and Social Psychology
QM), 644-663.

Cranly, M. S. (1981). Development of a tool for the measurement of maternal
attachment during pregnancy. Nursing Reserach, 30(5), 281-284.

Davila, J ., Burge, D., & Hammen, C. (1997). Why does attachment style
change? J oumal of Personality and Social Psychology, 73(4), 826-838.

Downey, G., & Coyne, J. C. (1990). Children of depressed parents: and
integrative review. Psychological Bulletin, 108(1), 50-76.

76

Egeland, B., & Sroufe, L. A. (1981). Attachment and early maltreatment. Child
Development, 52, 44-52.

Ere], 0., & Burman, B. (1995). Interrelatedness of marital relations and parent-
child relations: A meta-analytic review. Psychological Bullitan, 118(1), 108-132.

Fantuzzo, J., Boruch, R., Beriama, A., Atkins, M., & Marcus, S. (1997).
Domestic violence and children: Prevalence and risk in five major US. cities. Journal of
the American Academy of Child and Adolescent Psychiatry, 36(1), 1 16-122.

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

competencies of young children. Journal of Consulting and Clinical Psychology, 59(2),
258-265.

Feeney, J. (1999). Adult romantic attachment and couple relationships. In J.
Cassidy & P. R. Shaver (Eds.), Handbook of attachmentzTheory. reserach, and clinical
applications (pp. 355-377). New York: The Guilford Press.

Feeney, J. A., & Noller, P. (1990). Attachment style as a predictor of adult
romantic relationships. Journal of Personality and Social Psychology, 58(2), 281-291.

Feeney, J. A., Noller, P., & Hanrahan, M. (1994). Assessing adult attachment.
In M. B. Sperling, W. (Ed.), Attachment in Adults . New York: the Guilford Press.

 

Gazmararian, J. A., Lazorick, S., Spitz, A., M., Ballard, T. J ., Saltzman, L., &
Marks, J. S. (1996). Prevalence of violence against women. Journal of the American
Medical Asscﬂttion, 225(24), 1915-1125.

 

George, C., & West, M. (1999). Developmental vs. social personality models of
adult attachment and mental ill health. British Houmal of Medical Psychology, 72(3),
285-303.

Graham-Bermann, S. A., & Levendosky, A. A. (1998). Traumatic stress
symptoms in children of battered women. Journal of Interpersonal violence, 13(1), 111-
128.

 

Hammen, C. L., Borge, D., Daley, S., B., Davila, J., Paley, B., & Rudolph, K.,
D. (1995). Interpersonal attachment cognitions and predictions of symptomatic responses
to interpersonal stress. Journal of Abnormyalﬁ Psychology, 104(3), 436-443.

Hazan, C., & Shaver, P. P. (1987). Romantic love conceptualized as an

attachment process. Journal of Social Personality and Personal Relationshi+p34987(52),
51 1-524.

77

Hazan, C., & Zeifman, D. (1999). Pair bonds as attachments. In J. Cassidy & P.
R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications
(pp. 336-354). New York: The Guilford Press.

Herman, J. L. (1992). Trauma and I‘CCOVCﬂ. New York: Basic Books.

Holden, G. W., & Ritchie, K. L. (1991). Linking extreme marital discord, child
rearing, and child behavior problems: evidence from battered women. Child
Developmentg62(2), 31 1-327.

Kilpatrick, K. L., & Williams, L. M. (1998). Potential mediators of post-
traumatic stress disorder in child witnesses to domestic violence. Child Abuse and
Neglect, 22(4), 319-330.

Kirkpatrick, L. A., & Davis, K. E. (1994). Attachment style, gender, and
relationship stability: A longitudinal analysis. Journal of Personality and Social
Psychology, 66(3), 502-512.

 

Kobak, R. R., & Hazan, C. (1991). Attachment in marriage: Effects of security
and accuracy of working models. J oumal of Personality and Social Psychology, 60(6),
861-869.

Levendosky, A. A. (1995). Parenting in battered women: The effects of domestic
violence on women and children. : Unpublished dissertation, University of Michigan.

Levendosky, A. A., & Graham-Bermann, S. A. (1998). The moderating effects
of parenting stress on children's adjustment in woman-abusing families. Journal of
Interpersonal Violence, 13(3), 383-397.

Levendosky, A. A., & Graham-Bermann, S. A. (2000). Behavioral observations
of parenting in battered women. Journal of Family Psychology, 14(1), 1-15.

Levendosky, A. A., & Graham-Bermann, S. A. (in press). Parenting in battered
women: the effects of domestic violence on women and their children. Journal of Family
Violence.

Levendosky, A. A., Lynch, S. L., & Graham-Bermann, S. A. (2000). Mothers'
perceptions of the impact of woman abuse on their parenting. Violence against women.

Lindahl, K., M., Clements, M., & Markman, H. (1997). Predicting marital and
parent functioning in dyads and triads: A longitudinal investigation of marital processes.
Journal of Family Psychology, 11(2), 139-151.

Lyons-Ruth, K. (1996). Attachment relationships among children with

aggressive behavior problems, the role of disorganized early attachment patterns. J oumal
of Consulting and Clinical Psychology, 64(1), 64-73.

78

Main, M. (1996). Introduction to the special section of attachment and
psychopathology: 2. Overview of the ﬁeld of attachment. J oumal of Consulting and
Clinical Psychology, 64(2), 237-243.

Main, M., & Cassidy, J. (1988). Catagories of response to reunion 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., & Goldwyn, R. (1984). Adult attachment scoring and classiﬁcation
systems. (Second ed.): Unpublished manuscript, University of California at Berkeley.

Main, M., & Hesse, E. (1990). Parent’s unresolved traumatic experiences are
related to infant disorganized attachment status: Is frightened and/or frightening parental
behavior the linking mechanism? In M. T. Greenberg, D. Cicchetti, & E. M. Cummings
(Eds), Attachment in preschool years: theory, research, and intervention (pp. 161-184).
Chicago: University of Chicago Press. '

Marshall, L. L., & Vitanza, S. (1994). Physical violence in close relationships:
Myths and realities. In A. L. Weber (Ed.), Perspectives on Close Relationships . Boston:
Allyn & Bacon.

McCloskey, L. A., Figueredo, A. J ., & Koss, M. P. (1995). The effects of
systemic family violence on children's mental health. Chld Development, 66(5), 1239-
1261.

O'Heam, R. B., & Davis, K. E. (1997). Women's experience of giving and
receiving emotional abuse: An attachment perspective. Journal of Interpersonal Violence,
1_2(3), 375-391.

Peterson, R., Gazmararian, J. A., Spitz, A., M., Rowley, D. L., Goodwin, M. M.,
Saltzman, L., & Marks, J. S. (1997). Violence and adverse pregnancy outcomes: A
review of the the literature and directions for future research. American Journal of
Preventive Medicine, 13(5), 366-373.

Saunders, E. A., & Edelson, J. A. (1999). Attachment style, traumatic bonding,
and developing relational capacities in a long-term trauma group for women.
Intematiorgtl Journal of Group Psychotherapy, 49(4), 465-485.

Schuengel, C., Bakermans-Kranenburg, M. J ., & Van Uzendoom, M. H. (1999).
Frightening maternal behavior linking unresolved loss and disorganized infant
attachment. Journal of Consulting and Clinical Psychology, 67(1), 54-63.

Simpson, J. A., Rholes, W. S., & Phillips, D. (1996). Conﬂict in close
relationships: An attachment perspective. Journal of Personaly and Social Psychology,
_‘Z_l_(5), 899-914.

79

Slade, A., & Aber, J. L. (1992). Attachment, drives and development: Conflicts
and convergences . In M. E. J. Barron, & D. Wolitsky (Ed.), Interface of psychoanalysis
and psychology (pp. 154-185). Washington, DC: American Psychological Association.

Slade, A., Belsky, J ., Aber, J. L., & Phelps, J. L. (1999b). Mothers'
representations of their relationships with their toddlers: Links to adult attachment and
observed mothering. Developmental Psychology, 35(3), 611-619.

Stemberg, K. J ., Lamb, M. E., Greenbaum, C., Ciccetti, D., Dawud, S., Cortes,
R. M., Krispin, 0., & Lorey, F. (1993). Effects of domestic violence on children's
behavior problems and depression. Developmental Psychology, 29(1), 44-52.

Straus, M. A. (1979). Measuring intrafamily conﬂict and violence: The Conflict
Tactics (CT)Scales. Journal of Marriage and the Family, 41, 75-88.

Tjaden, P., & Thoennes, N. (2000). Prevalence and consequences of male-to-
female and female-to-male intimate partner violence as measured by the National
Violence Against Women Survey. Violence Against Women, 6(2), 142-161.

van Uzendoom, M. H. (1992). Intergenerational transmission of parenting: a
review of studies in nonclinical populations. Developmental Review, 12, 76-99.

van Uzendoom, M. H. (1995). Adult attachment representations, parental
responsiveness and infant attachment: a meta-analysis on the predictive validity of the
Adult Attachment Interview. Psychological Bulletin, 117(3), 387—403.

van Uzendoom, M. H. (1996). Attachment representations in mothers, fathers,
adolescents, and clinical groups. Journal of Consulting and Clinical Psychology, 64(1), 8-
21.

Wolfe, D. A., Jaffe, P., Wilson, S. K., & Zak, L. (1985). Children of battered
women: The relation of child behavior to family violence and maternal stress. Journal of
Consulting and Clinical Psychology, 53(5), 657-665.

Zeanah, C. H., & Benoit, D. (1995). Clinical applications of a parent perception
interview in infant mental health. Infant Psychiatry, 4(3), 539-554.

Zeanah, C. H., Danis, B., Hirshberg, L., Benoit, D., Miller, D., & Heller, S. S.

(1999). Disorganized attachment associated with partner violence: A research note. Infant
Mental Health Journal, 20(1), 77-86.

80

1111L1‘Uljlﬂlﬂillfﬂllﬂlllyjlltl