.‘ﬁzxrf. 2.. ) I
3‘ V G

:56... . “a
up.

. 0a,... 5 1.1 s. ..
. .L LQVWWW

. 1..."-
9

1:1! 1.

..$$

, . z
1.. ~
. . . .
ix: .3 . ‘
'11: u 2 u
. 5.5.1? . «@313
:33}. .u
. . :bﬁuzz
. {as

2a).». .15
f1. .3...

.ﬁﬂuﬁwﬁ.

3%.

5%..

.IE. )
.ﬂhhoﬁtmul... a
{grazing
2

I1

(1'57... v}... . 5|»... .

.3... (it , ...l.$”!:5.knw

5}. sl
. 3.1133.»
ﬁbultllﬁlﬁtxxgns

. Lclil3ull, . Y’r

 

 

TM lllllglllllllllllIlllllllllllllllllllllllllllllllllllilllll

f, . 7 193 02048 61
L’CCO —LlBRARY
Michigan State
University

 

 

 

This is to certify that the

dissertation entitled

A COMPARISON STUDY OF THE NORMAL AND CLINICAL DIVORCE: STYLE
OF ATTACHMENT, LEVEL OF CONFLICT, AND PARENT-CHILD
RELATIONSHIP

presented by
Mary Bremer Barron
has been accepted towards fulfillment

of the requirements for

Ph . D . degree in Social Science

 

 

422,54,“sz

Major professor

 

Date 8/17/00

 

MSU i: an Afﬁrmative Action/Equal Opportunity Institution 0-12771

 

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

DATE DUE

DATE DUE

 

DEC 3 0 2003

 

_IZ j' Z :3 531.1!
i02404

 

 

 

 

 

 

 

 

 

 

 

 

11/00 mm.“

 

A COMPARISON STUDY OF THE NORMAL AND CLINICAL DIVORCE: STYLE
OF ATTACHMENT, LEVEL OF CONFLICT, AND PARENT-CHILD
RELATIONSHIP

By

Mary Bremer Barron

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Social Work

2000

ABSTRACT

A COMPARISON STUDY OF THE NORMAL AND CLINICAL DIVORCE: STYLE
OF ATTACHMENT, LEVEL OF CONFLICT AND PARENT-CHILD
RELATIONSHIP

By

Mary Bremer Barron

The major thrust of the study was to determine if there were significant
group differences between a normal and a clinical sample of divorced or
divorcing parents regarding their attachment style, level of conflict, and the
amount of parenting time. For this study, “clinical" is defined as an enduring
pattern of thinking, feeling, and behaving that interferes with relationships in two
or more domains, e.g., cognition, affectivity, interpersonal functioning, and
impulse control, while “normal” refers to an adaptive pattern of relating that does
not interfere with interpersonal functioning in more than one area. Attachment
theory was the theoretical basis for the research inquiry in an attempt to
investigate the connection between early childhood experiences, adult
attachment styles and the current relationships between ex-spouses.

The study used a combination of quantitative and qualitative methods in
an effort to capture interpersonal differences between a normal sample of 63 and
a clinical sample of 62. A questionnaire was developed and used with the total
sample and an additional interview was completed with a small sub-sample of 10

persons, five drawn from each group. Participants came from two different

settings: persons attending a local divorce orientation program and persons
mandated by the court into service with a local community mental health agency.

The results from the attachment instruments indicated a lack of difference
between the two groups regarding security of attachment. However, the findings
from the combined sub-scales from two attachment instruments suggest that the
clinical sample had a more dismissing and avoiding reaction to the loss of the
spouse while the normal group had a more preoccupied and fearful reaction.
The face-to-face interviews supported the notion that these two small sub-groups
differed in regard to their early childhood relationship with their parents. The
clinical group had high levels of exposure to marital instability, more instances of
family violence and more distant and rejecting experiences with parents when
compared to the normal group. There were clear differences between the normal
and clinical groups in level of conﬂict. The findings indicated differences in
description of communication, frequency of communication, style of negotiation,
resolution of custody and conflict in the prior year. The results from the study did
not discern any differences in the amount of parenting time granted.

This investigation was a preliminary study completed without the use of a
control group. As such, this study has limitations regarding the generalizability of
the findings to populations beyond the county in which the study occurred. The
results are limited to group differences between divorced couples previously
married 2-19 years with at least one minor child. This study has implications for
research regarding adult attachment styles, practice with couples in high conﬂict,

and policy regarding the division of parenting time.

Copyright by
Mary Bremer Barron
2000

To my daughter Brianna, and step-children Chris, Robin, Becky, and Andy

ACKNOWLEDGMENTS

The completion of a dissertation doesn’t occur without the support and
encouragement of many people who supported this goal as a worthy journey
deserving of time and energy.

I wish to thank Rena Harold, my dissertation chair for being my north star,
whose availability was constant and whose continual guidance provided clarity
and direction. Her words made every task seem feasible and her communication
ability motivated me forward after months of restful inertia.

Committee members, Ellen Whipple, Alytia Levendosky, Tom Luster, and
Jackie Lerner made the dissertation a reality. Each provided their unique part of
the spatial map that added to my thinking about how to conceptualize research
with families.

I wish to thank Genesee County Community Mental Health, Child and
Adolescent Service for allowing me the privilege of working in a clinical
environment that allowed me to be creative and experimental. Through
collaboration with Genesee County Family Court and Friend of the Court I was
allowed access to the interior emotional world of hundreds of families who
experience divorce. Carol Osborn, Cathy Morrison, and Jeanne Sadler were co-
workers who shared with me the exciting challenges of making a difference in the
lives of families in transition. My gratitude for their dedication to the Children of
Divorce Program is immeasurable. Special thanks to Marg Gay as support staff
who made every change and demand seem effortless.

Special recognition goes to Elizabeth Rose, whose graduate studies

vi

intersected with mine. Together we have weathered the storms and disruptions
in life that have temporally taken us off the path of academic pursuits and as a
result share a special bond. Peg Whalen, a fellow graduate student served as
data consultant, and gave generously to the process of analysis and completion.
And most importantly, I wish to recognize my daughter Brianna, who
readily accepted my decision to go to graduate school at a time in her life when
she could have beneﬁted from more parent involvement. To Richard I am
indebted to his faith in my capabilities and for ensuring that our home is a place
of sustenance and good food. Since his arrival in my life, his critical thinking, his
valuable editing, and committed partnership has made for pleasant

companionship on this academic adventure.

vii

TABLE OF CONTENTS

List of Tables ............................................................................................. xi
List of Figures ........................................................................................... xii
Chapter I
INTRODUCTION ....................................................................................... 1
Background of the Study ............................................................................ 8
Problem Statement .................................................................................. 13
Research Question .................................................................................. 14
Professional Signiﬁcance of the Study ..................................................... 16
Overview of Methodology ........................................................................ 20
Limitations ................................................................................................ 21
Chapter II
LITERATURE REVIEW ........................................................................... 23
Attachment Theory ................................................................................... 23
Bowlby ........................................................................................... 24
Ainsworth ...................................................................................... 27
Main .............................................................................................. 32
Security of Attachment in the Mother-Child and Father-Child
Relationship History ...................................................................... 34
Internal Models of Representation ................................................ 35
Disordered Attachment in Infancy ................................................. 42
Insecure Attachment and Child Behavior Problems ...................... 44
Early Insecure Attachment and the Development of Personality
Disorders ....................................................................................... 46
Measurement of Adult Attachment ........................................................... 50
Attachment History and Partner Selection .................................... 60
Adult Attachment and Adjustment to Relationship Dissolution ...... 63
Empirical Studies Regarding Divorce ....................................................... 65
Child Outcome and Parental Divorce ............................................ 65
Quality of Parenting by Custodial Parent ...................................... 71
High Conﬂict lnterparental Relationships ...................................... 72
Post-Divorce Father-Child Relationships ...................................... 80
Chapter III
METHODOLOGY ..................................................................................... 85
Overview .................................................................................................. 85
Research Context .................................................................................... 90
Sampling .................................................................................................. 93
Exclusionary Criteria ................................................................................ 98

viii

Instruments Used for All Participants ....................................................... 99

Procedures ............................................................................................ 107
Chapter IV

QUANTITATIVE RESULTS ................................................................... 112
Overall Results ...................................................................................... 1 12
Age, Gender, and Race ......................................................................... 112
Education, Income and Occupation ....................................................... 113
Additional Family Characteristics ........................................................... 114
Comparison of the Normal and Clinical Samples ................................... 115
Overall Attachment Style ....................................................................... 1 17
Attachment to Spouse ............................................................................ 1 19
Level of Conﬂict ..................................................................................... 121
Parenting Time ....................................................................................... 125
Summary ................................................................................................ 127
Chapter V

QUALITATIVE RESULTS ...................................................................... 129
Brief Vignettes ....................................................................................... 130
Context of Early Childhood Environment ............................................... 139
Relationships to Parents ........................................................................ 143
Childhood Losses and Traumas ............................................................ 148
Marital Partners ..................................................................................... 150
Reaction to Loss of Marital Partner ........................................................ 152
Co-Parental Relationship ....................................................................... 1 55
Relationship to Children ......................................................................... 156
Perceptions of Ongoing Conﬂict ............................................................ 160
Parenting Stress Index ........................................................................... 161
Summary ................................................................................................ 163
Chapter IV

CONCLUSIONS AND IMPLICATIONS .................................................. 166
Restatement of Problem and Methodology ............................................ 166
Summary of the Study ........................................................................... 167
Implications for Research ...................................................................... 175
Implications for Practice ......................................................................... 178
Implications for Policy ............................................................................ 182
Synopsis ................................................................................................ 183
APPENDICES

A: Tables ............................................................................................ 185
B: Figures .......................................................................................... 192
C: Letter to Participate ....................................................................... 197
D: Family Counseling Order ............................................................... 199
E: Divorce Orientation Questionnaire ................................................. 201

F: Family Relationship Interview ........................................................ 215
G: University Committee on Research Involving Human Subjects ...... 219
H: Letter of Consent Normal Sample ................................................. 221
I: Letter of Consent Clinical Sample .................................................. 223
J: Consent to Audiotape ..................................................................... 225
BIBLIOGRAPHY .................................................................................... 227

LIST OF TABLES

Table 1: Gender, Age and Years Married ............................................. 186
Table 2: Race, Income, and Education ................................................. 187
Table 3: Distribution of Reasons Relationship Failed ............................ 188
Table 4: Custody Arrangement and Custody Satisfaction ..................... 189
Table 5: Communication and Level of Conflict by Sample .................... 190
Table 6: Description of Sub-Sample Interviewed ................................... 191

xi

LIST OF FIGURES

Figure 1: Model of the Effect of Childhood Attachment on Adult Attachment
Style, Divorce Experience, and Current Parent-Child Relationship ....... 193

Figure 2: Comparison of Respondents’ Incomes .................................. 194
Figure 3: Parenting Stress Index Normal Sub-Sample ......................... 195
Figure 4: Parenting Stress Index Clinical Sub-Sample ......................... 196

xii

Chapter 1

INTRODUCTION

Divorce is a life transition that requires adaptation from all family
members. Divorce is not a single, brief event that ’ occurs within one’s
development, but is a psychological, social, legal, and economic process that
requires ongoing adaptation (Wallerstein & Kelly, 1980). Over time, many
divorced parents are able to successfully reorganize and lessen their emotional
tie to their former marital partner while maintaining and nurturing a relationship
with their children (Ahrons & Miller, 1993; Maccoby & Mnookin, 1991). Other
parents have difﬁculty either with psychologically letting go of their marital partner
and/or with sustaining contact with their children post-divorce (Arendell, 1992;
Grief, 1995; Hoffman, 1995). While most families are able to master the tasks of
divorce, a signiﬁcant percentage of families, as evidenced by protracted litigation,
remain unadjusted or conﬂicted years after the divorce (Johnston, Kline &
Tschann. 1989; Kline, Johnston, & Tschann, 1991; Maccoby & Mnookin, 1991;
Tschann, Johnson, Kline, 8. Wallerstein, 1989)

Over the last twenty years, divorce research has reﬁned the ﬁndings
regarding divorce outcomes to produce a foundation of convergent results. For

example, there is consensus that how children adapt to divorce is associated

with gender, with age or stage of development at the time of the marital rupture,
and with whether or not a child has a regular ongoing relationship with the non-
resident parent (Allison, 1989; Heatherington, 1989; Kelly, 1993). Other
important factors in determining children’s adjustment are found in studies that
highlight variables such as (1) parental adjustment; (Maccoby, Buchanan,
Mnookin, & Dombush, 1993); (2) the relationship between the parents (Johnson,
Kline & Tschann, 1989); (3) the frequency of access to the nonresidential parent,
(Camera & Resnick, 1989; Hetherington, 1993); (4) interparental hostility (Amato
& Rezac, 1994; Kelly, 1993); and (5) the effects of unresolved custody litigation
(Kitzman & Emery, 1994). Divorce research, like other family studies, also
recognizes the important role of the availability of social support (Garvin, Kalter
8. Hansell, 1993; Hoffman, 1995) and the effect of socioeconomic status (SES)
(McLanahan, 1994; Morrison, 1995) on adjustment outcome.

Because so much of children’s adjustment is dependent upon the
adjustment of their parents, how the custodial parent responds and copes with
divorce is central to child adjustment (Kalter, Kloner, Schreier, & Okla, 1989). Of
particular interest are the internal processes by which both parents cope with
issues of abandonment and rejection. Weiss (1975) was the ﬁrst to study the
incongruent feelings and behavior that exist between former marital partners.
Through interviews with divorced individuals, Weiss (1975) used attachment
theory to explain the persistence of negative attachment that continues for some
individuals for years following divorce. Weiss (1975) used Bowlby's attachment

theory, as a way to describe how the termination of marriage parallels some of

the protest, despair, and acceptance that infants display in response to long term
parental separation.

Attachment behavior is thought to explain some adult responses to
divorce. Recently, interest has developed in studying the link between adult
attachment style and one’s coping behavior when a romantic relationship is
terminated (Bakermans-Kranenburg & van IJzendorn, 1997; Bartholomew &
Horowitz, 1991; Hazan and Shaver, 1987; Masheter, 1991 ). Researchers in a
variety of ﬁelds are also looking at early attachment in retrospective studies as
means of explaining adult psychopathology (Kessler, Davis, & Kendler, 1997;
van IJzendoorn & Bakermans-Kranenburg, 1996) and the development of
personality disorders (Patrick, Hobson, Castle, Howard, & Maughan, 1994).

This dissertation is a preliminary study designed to deﬁne areas of
differences in adult adjustment between groups of individuals who are able to
emotionally disengage from their partner, to arrive at a parenting plan, and to
communicate on behalf of their children and those individuals who are not able to
do so. Although the divorce process itself takes years rather than months to
resolve, some adults are able to respond to their interpersonal crisis by moving
forward emotionally while other adults remain emotionally stuck or negatively
attached to their former spouse.

Clinicians providing outpatient treatment and mediators assigned to help
the divorcing parents recognize the difference in dynamics found between the
normal and abnormal or high conﬂict divorce. Gray and Shields (1992), working

with adults in a clinical setting who did not initiate the divorce, described the

difﬁculty some persons have with divorce adjustment. They describe lack of
divorce adjustment by the development of a defensive stance of ‘moral
superiority’ and as having notable delay in the formation of a new identity up to
four years post-divorce. Mediating disputes in high conﬂict divorces exposes
mental health professions to “couples from hell” (Mathis, 1998) who make
extreme demands on the professional to gain alignment with their perception of
blame.

There are other descriptive differences that differentiate a normal divorce
from a high conﬂict divorce. In a high conﬂict divorce, the adults are not able to
tolerate positive interaction with their former spouse. They remain avoidant or
engaged in a negative competitive battle with the other parent for years following
the divorce and seem to have little conscious awareness of the effect of their
behavior on their children. Frequently, one or both of such high conﬂict parents
depend on the legal system to determine how custody or parenting time is to be
shared and have the expectation that the court will afford them the “justice” that
they deserve. One estimate from Johnson and Campbell’s (1988) California
study was that 15-30 percent of couples experiencing divorce met the criteria for
a high conﬂict divorce. Support for 15% as an indication of the amount of Intense
conﬂict regarding unresolved custody and parenting time was also found in the
Maccoby and Mnookin (1992) results.

For the purposes of this study, the clinical sample refers to the Diagnostic
and Statistical Manual of Mental Disorders' (DSM IV) deﬁnition of personality

disorders: adults who have an enduring pattern of thinking, feeling and behaving

that is relatively stable over time (APA, 1994). This is an enduring pattern of
inner experience and behavior that deviates markedly from the expectations of
the individual’s culture in two or more areas: 1) cognition; 2) affectivity; 3)
interpersonal functioning; and 4) impulse control (APA, 1994). Brennan and
Shaver (1998), quoting the APA, added that these enduring patterns were
pervasive and inﬂexible, and had onset in adolescence or early adulthood and
were stable over time and led to distress or impairment in relationships.
Regardless of the kind of personality disorder, all were characterized by
persistent difﬁculties with interpersonal relations.

In this study, level of interpersonal difﬁculty is differentiated by certain
criteria for the clinical and the normal samples. Those classiﬁed as clinical have
two or more clinical indicators in regard to interpersonal functioning. Those
include: mandated treatment intervention by the family court judge, history of
domestic violence by self-report or spousal report, history of substance abuse by
self-report or spousal report, prior history of court mandated treatment, the
presence of 4 or more psychosocial stressors on the DSM—IV, Axis IV, a high
number of court appearances regarding the divorce, and other legal difﬁculties in
addition to the divorce action, eg. violations of PPO’s (personal protection
orders).

Those classiﬁed as normal will have an absence of markers in the above
clinical areas. However, because of the divorce process "is its expected that
individuals may have at least one clinical indicator in regard to Interpersonal

functioning. The normal sample would not be court mandated into treatment

services. They would have fewer court appearances regarding the divorce action
and report three or fewer psychosocial stressors on the DSM IV Axis IV and have

no other legal difﬁculties in addition to the divorce.

Background of the Study

This ﬁrst chapter of the dissertation is an initial presentation of the
potential link between an individual’s type of attachment relationship to their
former spouse and that adult’s subsequent divorce adjustment. It is also an

introductionuto the possibility that one’s childhood attachment history, whether

W

_~

“DI", """"" k“

attachmenthnﬂ assumption of this study is that individuals bring to a marriage
relationship a pre-existing style of relating that has been inﬂuenced by their
childhood attachment experience with primary caregivers. Thus, a secure or
insecure childhood attachment history can inﬂuence whether or not an individual
chooses a partner with a similar or complementary attachment style. Whether
one has a secure or insecure attachment history may also inﬂuence reaction to
divorce and howhwell one copes with the loss of a partner.

In contemporary literature, the term attachment has been used to describe
attachment behaviors, attachment bonds, attachment systems, and attachment
relationships (Zeanah, Mammen, 8. Lieberman, 1993). According to Bowlby,
attachment was deﬁned as an enduring affectional tie or bond, enduring in its

nature and speciﬁc in its focus (Lamb, 1976b). Ainsworth (1989) deﬁned

‘attachment’ as a particular kind of affectional bond that was relatively long

lasting and in which the partner was important as a unique individual, who was
not interchangeable or replaceable with another. According to Ainsworth (1989),
affectional bonds are not synonymous with relationships. Affectional bonds
include the characteristics of the individual and encompass the internal
organization of the individual, rather than the characteristics or history of a dyadic
relationship. In an affectional tie, there is the desire to maintain closeness to the
partner and a need to maintain proximity upon distress, or pleasure, joy upon
reunion, and grief at loss. West and Sheldon (1988) deﬁne three forms of
attachment in the lives of most adults: residual attachment to parents, care-giving
attachment to dependent children, and reciprocal attachment to a signiﬁcant
other. Borrowing from West and Sheldon, for the purposes of this study, an
attachment ﬁgure is deﬁned as a peer who is not a member of the family of
origin, with whom there was a sexual relationship, and with whom there has been
an exclusive relationship for longer than one year.

As a clinician working with the divorced and divorcing population over the
course of many years, one poignant personal observation is that the divorce
experience appears to be much worse for those adults who report a childhood
history of abuse and neglect. Adults for whom divorce is one of a series of
losses or traumas that involve abandonment or rejection and perhaps leave them
feeling insecure have more difﬁculty adjusting to the loss of a marital partner.
How insecure attachment to parental ﬁgures becomes associated with poor peer

relationships and character formation in the adult personality will be addressed

later in the dissertation. This chapter concludes by noting the limitations of the
study.

The divorce literature has begun to study adult attachment as a factor in
post-divorce adjustment within the last decade. For example, there has been
interest in determining whether romantic partners’ style of attachment is
associated with the partner selection process (Bartholomew & Horowitz, 1991;
Collins & Read, 1991; Hazen & Shaver, 1987). These researchers used infant
attachment concepts as a framework for examining how adult love relationships
were related to earlier parent-child interactions. Studies of romantic relationships
suggest that an adult’s attachment style is predictive of the selection of a partner
with a similar or a complementary style of attachment (Collins & Read, 1990).
For example, adults with a secure attachment frequently chose a partner with a
secure attachment style, and adults with an anxious attachment style tended to
choose others who have an anxious or dismissing style (Hazan & Shaver, 1987)

Infant mental health studies based on attachment theory suggest an
association between one’s attachment history and behavior problems in
childhood and later personality organization (Brennen & Shaver, 1998;Goodman
& Brunley, 1990; Lyons-Ruth, Alpem & Repacholi, 1993; Toth & Cicchetti, 1996).
An infant’s primary attachment relationship is thought to inﬂuence the early
aspects of personality organization (Sroufe, 1985) through reciprocal
experiences between self and other. Attachment behaviors developed in
childhood are thought to be carried forward as patterns of expectations and

levels of trust. Attachment theorists suggest that there may be an association

between childhood attachment history and the kind of attachment expectations
that get directed at partners and children later in life. Crittenden, Partridge and
Clausen (1992) suggest that previously learned expectations and behaviors
affect not only the selection of partners, but also affect the perception and
organization of information about relationships.

lntemal mental representations of the self and other are thought to help
explain how an attachment pattern in early life gets transferred onto other
signiﬁcant relationships during one’s life. The infant mental health literature
suggests that how infants and young children operationalize and mentally
incorporate their view of their caregiver and their self becomes transferable
through “internal mental representations” (Crittenden, 1990; Fish, 1993; Fonagy,
Steele, Steele, Moran & Higgett, 1991; Main, Kaplan, & Cassidy, 1985; Stern,
1991; van lJzendooom, Goldberg, Kroonenberg, & Frenkel, 1992). These
memory and cognitive-affective processes remain constant or change in
response to one’s appraisal of the relationship to a primary attachment ﬁgure.

A parent’s internal representation is believed to be the mechanism by
which she/he is or is not in synchrony and sensitive to a child’s verbal and non-
verbal cues. In a 1996 study by van IJzendoorn and Bakermans-Kranenburg,
parents of disturbed children were found to have had more insecure
representations in their attachment experiences with their own parents. A parent
with a secure style of attachment is more open, ﬂexible, and responsive to a
child’s behavior. Parents with an insecure style of attachment are more closed,

more rigid, and insensitive to cues from their child. As a result of the disparity

between caregiver and infant, the infant view of the self and view of others takes
on a defensive strategy to maintain the tie to the caregiver. The relationship
between a parent’s internal representation from their childhood and current
parent-child interactions is the focus of many recent studies (Cowan, Cohn,
Cowan, and Pearson, 1996; Crandell, Fitzgerald, & Whipple, 1997; Patrick,
Hobson, Peter, Castle, Howard & Maughan, 1994) which help explain the
relationship dynamics observed in parent-child relationships during childhood.
Less is understood about how internal representations change over time, and
what kind of attachments beside primary attachments inﬂuence one’s style of
attachment throughout the life span.

During the last thirty years, research regarding divorce outcome for adults
and children has primarily focused on white middle class families to control for
confounding variables. However, the divorce population includes conditions of
low socioeconomic status, mental illness, inadequate interpersonal skills and
other variables which compound the effects of divorce on family members.
Recognizing that these conditions co-exist with the divorce process adds
complexity to the analysis of ﬁndings. Embedded within the research based
upon large household surveys, it is highly probable that some of the negative
deleterious effects of divorce on children are, in fact, attributable to other existing
conditions within the family. In other words, it is expected that a proportion of
adults has entered into the marriage with a prior history of disordered attachment
and/or mental illness and that their divorce experience will not be similar to that

of others without those conditions.

10

In a retrospective study, the difference between adequate and inadequate
parent-child relationships was associated with insecure attachment between
adult partners and the adults’ attachment history during childhood (Crittenden,
Partridge, 8 Claussen, 1991). Adults with a history of secure attachment were
more likely to select a partner with a secure attachment, whereas insecure
attachment history was more likely to result in the selection of an insecure
partner (Collins 8. Read, 1990; Crittenden, et al., 1991). The studies that
measure adult attachment style link marital relationship quality with child
behavioral problems. Similarly adult attachment style is being studied in terms of
adult divorce adjustment and the presence of ongoing conﬂict (Masheter, 1991)

Generally, normal adults have the emotional and psychological resources
to make interpersonal adaptations and over time, psychologically disengage from
their former partner. For the practitioner, there are marked clinical differences in
the relational dynamics between couples who fall within the normal and those
within clinical guidelines. The clinical adults are more likely to become
developmentally stuck and behaviorally remain preoccupied, embittered, and

conﬂicted in their relationship with their former spouse. The adult with clinical

......,..-
W 4”

symptoms has more difﬁculty managing affect and tends to express rage and
blame that does not dissipate with time. For the adult who meets clinical criteria,
divorce is experienced as a more intense, emotionally humiliating experience
with the source of that internal pain projected on to the action or conduct of the
former partner. For a small percentage of adults, the divorce experience is so

devastating, years following divorce they are at an emotional impasse, wishing to

11

regain access to the lost object, and can neither disengage nor resolve the
relationship with their spouse (Gray & Shields. 1992; Rossiter, 1991).

An assumption that underlies this study is that childhood parent-child
relationships that were secure support adult attachment styles that are also
secure, whereas insecure attachment in childhood are predictive of insecure,

,3 Eavoidant, and ambivalent attachments in adulthood. Several longitudinal studies
\ ,5 \which followed infants, insecurely attached at age one, found that these same
children developed severe behavior problems in childhood. These difﬁculties
with socialization were later associated with adult personality disorder. Toth and
Cicchetti (1996) found an association between early maltreatment and later
development of personality disorder. In a study that focused on relational deﬁcits
in severely depressed and mentally ill parents of infants, Goodman and Brumley
(1990) observed infants who displayed one of three behaviors: 1) withdrawal
from social interactions; 2) actions evidencing deﬁciencies in the development of
social skills; or 3) engaging in coercive exchanges in a persistent effort to elicit
the desired parental response.

Johnson and Campbell (1988) found that nearly all the adults in the high
conﬂict families they studied had experienced traumatic or ambivalent
separations in childhood or during the marital rupture. In the Johnson and
Campbell study, high conﬂict families were estimated to comprise about 10-15%

of all families who divorce. Their study focused on a clinical population that used . I“
/ . .

g “I

conﬂict to manage intolerable feelings of loss, humiliation, and helplessness}

They found that divorcing families in high conﬂict are more likely to use excessive

12

litigation to resolve long-standing intrapsychic, interpersonal and other
environmental issues as part of the divorce process. Such high conﬂictfamilies
who litigate excessively are well known to the court system. However, it was not
clear in their study whether their high conﬂict families originated from the clinical
population or if other factors determined the intensity or duration of litigation.

In summary, when couples with minor children divorce, how well children
fare depends on many factors, not the least of which is the degree of hostility and ‘
conﬂict that continues to exist in the interparental relationship post-divorce. ‘
Within the past ten years, adult attachment style has been found to be a predictor ’3, «=9 i
of adults’ adjustment following the loss of a romantic partner. This study involves r
the application of attachment theory and adult attachment style with parents who
divorce to determine if adult attachment style is also associated with adult
divorce adjustment and indireme with patterns in post-divorce parent-child

relationships.

Problem Statement

This study developed from the provision of clinical intervention to families
in high conﬂict who were litigating custody or parenting time issues. The ﬁrst aim
of this exploratory study was to explain why some adults are unable to
emotionally separate from their former spouses and have difﬁculty in relating with
their children post-divorce. Using attachment theory as a theoretical background,
this study compared two different samples, a normal sample with a clinical

sample. In addition to the criteria listed previously, for the purposes of this study

13

a normal sample refers to adults who experience divorce with a history of few if
any interpersonal problems in several psychosocial domains, who engage in low
levels of litigation to resolve custody or parenting time, and have relatively low
levels of conﬂict with their former spouse regarding access to minor children. A
clinical sample refers to adults who experience divorce with a history of poor
interpersonal relationships in several psychosocial domains, who engage in a
high rate of litigation to resolve custody or parenting time, and who have high
levels of conﬂict with their former spouse regarding access to minor children.

The second aim of this research was to determine if there was an
association between three variables for each sample: the way a parent describes
their style of attachment to the other parent, the level of conﬂict in the co-parental
relationship, and parents’ perceptions of the relational closeness between each
parent and child. The third aim was to determine if there were descriptive
differences in the childhood experiences of a sub-sample within the study that
supported the notion that an adult’s early child-parent history inﬂuences the style
of attachment to the marital partner and that person’s view of their parenting

relationship to each child post-separation.

Research question

The premise of this dissertation is that there are notable mean differences
between normal and clinical groups who divorce, as measured by current style of
couple attachment, level of expressed conﬂict, and pattern of contact for minor

children and the non-resident parent, post separation.

14

Hypothesisl

When compared to the normal sample, the clinical sample will
demonstrate differences in attachment style as measured by pattern and
description of adult relationships in general.
Hypothesis II

When compared to the normal sample, the clinical group will report a
different pattern of coping in regard to loss of marital partner that will indicate a
higher frequency of dismissing, preoccupied, or fearful attachment.
Hypothesis Ill

When compared to the normal sample, the clinical sample will
demonstrate higher levels of conflict in the co-parental relationship as measured
by style of negotiation, frequency of communication, level of disagreement
regarding resolution of custody/parenting time.
Hypothesis IV

When compared to the normal sample, the non-resident parents within the
clinical group will report less parenting time granted, will have more limited or
restricted access to their minor children, and of the court-ordered parenting time
granted, the non-resident parents will report a pattern of contact that will be less
predictable and more infrequent.
Hypothesis V

In the qualitative interviews, the clinical sub-sample when compared to the
normal sub-sample will report more difﬁculty with attachment to primary

caregivers, as evidenced by disengaged, abusive, or highly controlled

15

relationships and will report a higher frequency of losses or traumatic events
during their own childhood that suggest an insecure pattern of relating existed
prior to the marital relationship.

The research question follows a model of inquiry that has been depicted in
Figure 1. The model is an effort to describe an intergenerational pattem of
secure and insecure attachment relationships over time. It demonstrates how
the three major variables under consideration interact and the model also
includes some factors that were not studied. The proposed model suggests that
an individual’s early childhood attachment experience with mother and father
affects that adult’s later style of attachment. One’s established adult attachment
style is thought to inﬂuence how an adult copes with the loss of a marital partner.
Individuals who are more securely attached cope with the loss of the marital
partner better than do individuals who are insecurely attached. Some of the
factors in the co-parental relationship which include attachment style are thought
to inﬂuence how the couples negotiate differences and ultimately whether they
engage in a negotiation style of high or low conﬂict. High conﬂict in the co-
parental relationship is viewed as interfering with the child’s relationship with

each parent and with the child’s felt level of security with each of their parents.

Professional Signiﬁcance of the Study
Within the social science community, there is a need for investigative
studies that are inclusive of the clinical population because it has ramiﬁcations as

to how divorce outcome is interpreted, and it has treatment implications for

16

clinicians providing intervention to divorced/divorcing families. Over twenty years
ago Trafford (1982) attempted to describe some of the psychological and
behavioral changes that occurred between normal divorcing couples. However
there is an absence of investigative studies that address the relationship
dynamics between couples who meet clinical criteria and divorce. Exceptions to
this practice are studies that compare high conﬂict parental relationships in intact
families with high conﬂict parental relationships that result in divorce (Amato,
Loomis, & Booth, 1995; Hanson, 1999; Jekielek, 1997) Although there has been
extensive research focused on family members’ adjustment to divorce over the
last thirty years, there has been little consideration of how the clinical sample
resembles or is different from a normal sample. Certainly antecedent conditions
such as prior attachment history inﬂuence divorce outcome, as does
socioeconomic status, (McLanahan, 1994) and availability of social support
(Garvin, et al., 1993; Ladd & Zvonkovic, 1995). The question becomes one of
degree and in which areas.

Adults who meet clinical criteria and engage in high conﬂict divorce
suggest more instability In the parental relationship and in the parent-child
relationship post-divorce. However, what is not clear is whether this is a function
of attachment history or other conditions such as lower SES. This exploratory
study is an initial attempt to explain some of the differences in dynamics between
couples who meet normal or clinical criteria by integrating concepts gleaned from
attachment theory and infant mental health studies with concepts from divorce

family research. It builds upon the work of other researchers who have found

17

level of conﬂict to be a critical variable in regard to child adjustment post-divorce,
and seeks to explain why level of conﬂict is so prevalent in the sub-population of
couples who divorce.

There has been an established association between mental disorders and
instability of marital relations. In a national co-morbidity study in 1990-92, results
indicated that men and women with an onset of mental illness prior to, or during
the marriage, have a higher rate of divorce, 48.2 % compared to 35.9% who
have no disorder (Kessler, Walter, & Forthhofer, 1998). This study estimated

that 5.9% of divorces by men and 10.3 % of divorces by women were attributable

to psychiatric disorders. Absent from the divorce literature are studies that

compare differences between adults who meet normal or clinical criteria and how
they perceive and interpret family relationships during and following the process
of divorce. Given that the clinical population Is embedded within with general
population, it appears highly likely that most national studies regarding negative
divorce effects include adults who meet clinical criteria.

It is important to study the clinical population who divorce to learn whether
relationship patterns between partners and children are the same or different
from the normal samples. From the clinical perspective, adults with mental
illness and/or personality disorders appear to have had experiences of
abandonment and rejection embedded in their childhood history (Patrick, et al.,
1994). However, these adults are usually unable to articulate any connection
between their prior trauma and the current pain or vulnerability that accompanies

the rejection regarding divorce. Adults with personality disorders are generally

18

non-responsive to traditional methods of outpatient treatment. However, the
clinical population appears to be associated with the proportion of adults who
have high conﬂict divorces. It is important to determine not only the source of the
conﬂict but also to have clinical knowledge about how to intervene.

A comparison study of a normal and clinical sample is also important for
gaining knowledge about the correlates between current adult interpersonal
functioning and decisions regarding maintaining a relationship to the child. The
divorce research suggests that ﬁve years post-divorce, between 30% to 50% of
children lose the active participation of a non-resident parent (Furstenberg, Nord,
Peterson, & Zill, 1983; Mott, 1990). This dramatic break in parent-child"
involvement has been associated with the level of conﬂict in the co-parental
relationship (Ahrons & Miller, 1993; Amato, Spencer, 8 Booth, 1995; Camera &
Resnick, 1989; Johnson, Kline, 8. Tschann, 1989), but it has not been addressed
from the perspective of the parent’s attachment history.

If a divorcing parent had an insecure attachment to their parents, and if
they have had an insecure style of relating to their marital partner, then it seems
probable that their children will also have difﬁculty developing a secure
attachment with parents who have such disordered attachment styles. Studying
parental attachment during the divorce process may help demonstrate a link
between that parent’s style of attachment and the quality of the post-divorce
parent-child relationship.

Finding correlates between attachment theory and divorce is also

important because it has implications as to how childhood experiences get

19

transformed into intergenerational patterns. Infant mental health and early
childhood development studies suggest that attachment to one’s early caregivers
may form an lntemal method of processing experiences that serves as a guide to
perception of later relationships through the use of internal working models.
Theoretically it seems plausible that children who experience divorce are
at higher risk when their parents’ have a pattern of insecure attachment and
when the parents do not protect the child from serious problems in interpersonal
functioning, e.g., mental illness, substance abuse, domestic violence, and/or
excessive litigation. Although the divorce literature is increasingly descriptive
about the normative developmental impact of divorce on the children, pre-school
through young adults, there is an absence of knowledge regarding the causes of
relationship breakdowns between children and their parents other than the

cessation of involvement.

Overview of Methodology

This study consists of a combination of quantitative and qualitative
methods of data collection without the use of a control group. The non-random
samples of adults who were in process of divorce and who had children under
the age of 18 were drawn from a large, primarily urban county in southeastern
Michigan. The participants came from two different settings: divorcing parents
who attended a county required two-hour divorce orientation and education
program for all persons who divorce and those parents who were court mandated

into specialized mental heath services as a result of child custody or parenting

20

time litigation. How individuals were placed into the normal or clinical sample will
be explained in more detail in Chapter III. This study looks at sample differences
comparing how the parents manage relationship conﬂict, both in regard to
intensity of conﬂict and how conﬂict is expressed, and whether or not the
conﬂictual behavior between the parents is a factor in the kind of custodial
arrangement and frequency of access that occurs.

This study obtained a majority of data from the participants through a self-
report questionnaire, which included demographics, information about the
marriage and divorce, the Conﬂict Tactics Scale, and two scales regarding
attachment behaviors. From the larger sample of 62 clinical and 63 normal
adults, a smaller sub-sample of 10 also participated in a semi-structured
qualitative interview that focused on parent—child relationships during their own
childhood, a description of their marital relationship, and current information

about relationships to their children.

Limitations

This study is preliminary in nature and it does not include a control group.
As a result it has limitations regarding its generalizability. Any conclusions
reached are considered preliminary and limited to the samples under
consideration.

The measurement of attachment in relationships is not a single construct
and there is no general agreement about how to conceptualize it or how to

measure it at different stages in one’s lifespan. Within couple’s research, adult

21

attachment is a construct that remains nebulous and difﬁcult to deﬁne. Although
there are many measurements of adult attachment, each instrument
conceptualizes different aspects of adult attachment in different ways (Crowell &
Treboux, 1995; Garbarino, 1996). Without agreement about how attachment
behaviors are manifested and carried forward from childhood to adulthood, this
study, like others, will not be able to conclude that its ﬁndings represent the total
attachment relationship between divorced or divorcing parents. Nor will it be able
to state what other relationships in addition to primary caregivers potentially
inﬂuence the development of one’s attachment style. What is an exact deﬁnition
of attachment and how it is captured in an instrument remains beyond the scope
of this study.

Because of the depth and breadth of research on attachment theory, the
description of attachment theory offered in this study should be considered only a
cursory summary of some of the assumptions and prominent features of this

theoretical model.

22

Chapter 2

LITERATURE REVIEW

To determine if there are differences between different groups of adults
who divorce, it is necessary to reveal both the theoretical framework from which
the researcher is approaching the questions, and to provide the empirical context
in which the study is lodged. The following review of the attachment literature
provides the foundation for understanding the connection between early parent-
child attachment relationships and the quality of family relationships post-divorce.
Attachment theory is presented as the transmitter through which secure
relationships evolve during an individual’s life span and from one generation to
the other. Traditionally, the infant mental health literature and the divorce
literature have been approached as two distinct ﬁelds of study. However, this
study applies attachment theory to divorce research as a means of explaining
how adult attachment style is associated with adjustment to the loss of a partner,
and with ongoing relationships with minor children post-divorce. The study
examines differences between clinical divorcing adults who have a history of
interpersonal difﬁculties with normal divorcing adults who do not have a history of

interpersonal difﬁculty.

Attachment Theory

The ﬁrst section of the literature review describes primary concepts related

to attachment theory, emphasizing the major tenets and assumptions as

23

 

developed by the selected early contributors. Additional underpinnings of
attachment theory are summarized under the following headings: parent-child
history and the security of attachment in the mother-child and father-child
relationship, internal models of representation and the transmission of
attachment, disordered attachment in infancy, insecure attachment and
childhood behavioral problems. The mid section of the literature review
addresses how some researchers have conceptualized the measurement of
adult attachment. Following the presentation of the factors that measure adult
attachment, the concept of attachment style is applied both to the selection of
romantic partners and to how adults react to the loss of a partner, such as during
divorce. The ﬁnal section addresses speciﬁc areas of divorce research that
provides the basis for the selection of variables chosen in this study.

Any discussion of attachment theory begins with recognition of its founder,
John Bowlby, who was the product of an ethnological-evolutionary perspective.
Although there have been numerous contributions by many other signiﬁcant
researchers in the ﬁeld of infant mental health and developmental psychology
during the last twenty years, the following discussion will primarily highlight the

work of Bowlby, Ainsworth, and Main.

Bowlby
Bowlby viewed an infant’s attachment behavior directed at the caregiver
as biologically driven and necessary for survival of the species. Basic to

Bowlby’s attachment theory was the assumption that infants are born with a

24

repertoire of behavioral tendencies that have survival value from the point of view
of evolutionary adaptiveness (Campos, Barrett, Lamb, Goldsmith & Steinberg
1983). Attachment theory developed from the study of primates and includes
recognition that a behavioral system develops between a primate infant and its
caregiver (Kraemer, 1992). Attachment theory proposes that infants actively
elicit and respond to their caregivers in instinctual ways and are active
participants in the parent-child relationship.

Bowlby distinguished attachment relationships from attachment behaviors.
Attachment behaviors refered to the signaling mechanisms that an infant or
caregiver uses to mediate the formation and maintenance of attachment bonds
(Bing-Hall & Stevenson-Hinde, 1991; Lamb, Thompson, Garden, Charhov &
Estes, 1984). Attachment behaviors are the signaling mechanisms and the
approach behaviors that motivate the caregiver to stay in close proximity to an
infant. The behavioral pattern or signaling system that develops can be altered
or improved by assisting the parent with a better interpretation of the infant’s
cues. Attachment behavior refers to any form of behavior that predictably results
in a person attaining or maintaining proximity to or communication with some
other preferred individual, underconditioned-stress (Bing-Hall & Stevenson-
Hindel991). \ I , q

According to Kraemer (1992), Bowlby stated that the human infants, like
other primates, are born with an innate mechanism for eliciting attachment from a

caregiver. Human babies are thought to have a built in preference for looking at

patterns, especially the human face, and with repeated exposure are able to

25

distinguish the familiar from the strange. With advancing development, infants
are able to make alterations in their behavior based upon their internal and
external experience. In other words, there exists within infants an lntemal
appraisal system that processes external cues. This system of appraisal allows
for the initiation of new behaviors based upon the reading of social cues from the
primary caregiver.

There were two central hypotheses in Bowlby’s work: (1) that the quality
of anynaﬁttachment relationship depends on the quality of care experienced with
that partner, and (2) that the quality of the primary attachment relationship
strongly inﬂuences early personality organization (Sroufe, 1986). Thus, infants
become attached to individuals who consistently and appropriately respond to
the infant’s proximity-promoting signals and behaviors (Lamb et al., 1984). Two
central assumptions of the ﬁrst hypothesis are that an infant’s attachment
behavior (1) is based on each caregiver’s responsiveness (Bridges, Connell, &
Belsky 1988; Sroufe, 1985) and (2) is dependent upon that child’s learned
experience and stored memory of that experience (Stern, 1995). In Bowlby’s
view, the behavioral characteristics of the caregiver are more inﬂuential than are
the characteristics of the infant in maintaining the attachment relationship.
However, there has been recognition in recent years that some infant
characteristics, such as temperament, affect the dynamics within the parent—child
relationship (lzard, Haynes, Chrisholm & Baak, 1991 ).

Bowlby emphasized that attachment figures served two interrelated

functions for infants. One is the provision of a secure base for exploration in their

26

environment and the other is the assistance with the regulation of lntemal levels
of stress (Campos et al., 1983). There is agreement among attachment theorists
that there is a strong relationship between maternal sensitivity to the infant’s
signals and the quality of the infant-parent attachment. Parental behavior is
considered central to the development of attachment behaviors in the infant
because interactional patterns that develop between parent and infant determine
the infant’s expectations for reciprocity in communication. The pattern of
attunement or lack of attunement affects physiological and emotional reactions in
the infant (Schore, 1994). Parents who are emotionally unavailable, insensitive to
an infant's or toddler’s cues, or who engage in incongruous or inappropriate
behaviors, are thought to predispose the child to set unrealistic goals regarding

relationships (Goodman & Brumley, 1990).

Ainsworth

Ainsworth has also inﬂuenced the ﬁeld of developmental psychology.
Ainsworth, a collaborator with Bowlby over many years, advanced attachment
theory in two signiﬁcant ways (Lamb, 1979). First, she pioneered the empirical
study of individual differences by developing the hypothesis that variations in
maternal sensitivity accounted for the development of individual differences in the
quality or security of infant-mother attachment. Secondly, she developed a
method, the Strange Situation, by which individual differences in mother-child
relationships were assessed. Ainsworth’s work was important because it

standardized a paradigm for assessing infant-mother attachment and it resulted

27

in the development of a classiﬁcation system that described the observed
attachment behaviors. Ainsworth’s work distinguishes between the precursor
attachment behaviors and patterns of attachment (Lamb, 1974). The 13 patterns
of attachment behavior which Ainsworth proposed as signiﬁcant in maintaining
contact with a mregiver were: (1) crying, (2) smiling, (3) vocalization, (4) visual-
motor orientation, (5) crying when attachment figure leaves, (6) following, (7)
scrambling, (8) burying face in lap, (9) exploration from a secure base, (10)
clinging, (11) lifting arms In greeting, (12) clapping hands in greeting, and (13)
approach through locomotion. Ainsworth and her colleges recognized that these
behavioral patterns did not appear to form consistent clusters in all infants
(Lamb, 1974), but rather demonstrated that individual differences were based on
the quality of the mother-infant interaction during the ﬁrst year. Ainsworth was
interested in establishing a measurement of the strength and quality of the
infant’s attachment (Lamb et al., 1984). Her work resulted in the development of
criteria for a classiﬁcation system that measured attachment behaviors.

The Strange Situation procedure was used to differentiate types of
security of attachment between parent and child. Over the course of years, it has
been administered to both mothers and fathers to measure infants’ attachment
behavior with each parent. Through the use of the Strange Situation, Ainsworth
identiﬁed three types or qualities of attachment relationships: secure, anxious,
and defended insecure (Crittenden, 1990). These three categories of attachment
behavior included additional gradations in attachment for a total of eight possible

subgroups whose main headings were labeled: (A) insecure avoidant, (B)

28

secure, and (C) insecure resistant. A forth classiﬁcation was added by Main and
Solomon (1986), which was labeled (D) for disorganized. The Insecure-
disorganized (D) classiﬁcation was used for those infants who did not fall within
the other three categories and in some clinical samples, described infants with
behavioral characteristics of both avoidance and resistance. Infants with the
disorganized (D) category lacked a coherent strategy for coping with tension or
stress. According to van lJzendoom, et al., ( 1992) this pattern occurred with high
frequency with maltreated infants as well as infants of depressed mothers.

Infants classiﬁed as secure (8) in their attachment used the primary
caregiverias— a reference or secure base from which to explore. When a child’s.
attachment relationship was secure, it reﬂected a pattern of strategies that the
child developed to meet the conﬂicting goals or needs between security and
exploration (Davidson, 1991). Infants with a secure classiﬁcation compose about
two-thirds of infants in normal samples (van lJzendoom et al. 1992).

An insecure avoidant infant (A) explores with minimum reference to a
caregiver, was minimally distressed with parental absence, and Ignored or
avoided a parent upon reunion. From a cross-cultural analysis of 32 studies in
eight countries, 20-24% of babies exhibited this pattern (van lJzendoom et al.).
Infants classiﬁed as insecure-ambivalentlresistant (C) were described as
restricted in exploration, as highly distressed by separation from the caregiver,
and as difﬁcult to comfort upon reunion. The insecure resistant classiﬁcation of

infants comprised approximately 14% of the total sample in the van lJzendoom

et al. (1996) study.

29

The Strange Situation as a measurement of attachment, did not come into
use without controversy. Some criticism included temporal variability (Lamb et
al., 1984) cross-cultural differences (van lJzendoom 8 Kroonenberg, 1988) and
individual variation within the same child-parent relationship (Crittenden, et al.,
1992). Consequently some of these challenges to the measurement of
attachment in infancy have resulted in a stronger conceptualization of the
components of attachment. For example, Sroufe and Waters (1977) challenged
the idea that infant attachment behaviors that determine level of security were
stable over time. They asserted that it was the organization of behavior, the
patterns of adaptation, and the quality of the affectional bonds that were stable,
not the infant’s expression of behavior. Sroufe (1985) argued that children
securely attached to caregivers are behaviorally quite heterogeneous. It was
Sroufe’s (1985) position that individual difference in temperament may explain
some differences in classiﬁcation in the Strange Situation. He asserted that the
critical factor in the development of attachment behaviors was the development
of caregiver responsiveness. Taking this a step further, he suggested that the
quality of attachment changes, depending upon the environmental conditions
placed on the caregiver. This perspective took the position that an infant can
have a secure attachment to one parent and an insecure attachment to the other
parent.

Bridges, et al., (1988) sought to determine whether or not the
measurement of attachment in the Strange Situation was relationally based or a

characteristic of the infant. The Strange Situation procedure was given twice,

30

alternatively with mother or father at 12 and 13 months of age. For some infants
the ﬁndings resulted in some similarity between an infant’s reaction to separation
from mother and father, and for other infants there were clear differences in
reaction to each parent. Some of these infants were secure with one parent and
insecure with the other. They concluded that the Strange Situation measured
properties of a relationship rather than a stable characteristic of the infant
(Bridges, et al., 1988).

In a later study, Bridges and Connell (1991) compared infants’
responsiveness to mothers and fathers in the Strange Situation and during free
play. With this sample of 59 infants, they found less variability between parents,
and considered the possibility that attachment behaviors may develop into a
stable characteristic by the age of one. However, this sample was comprised of
families of middle to high SES, and it was not clear whether a high level of
parental care contributed to less variability in this study

The use of the Strange Situation to measure attachment behavior appears
to be most reliable for infants between 12 and 18 months of age (Waters 8
Deane, 1985). Toddlers, two and three years of age, have advanced cognitive
functions that inﬂuence their reaction to separation from a caregiver. As toddlers
begin to freely move about, they enter into a different kind of goal directed
partnership and begin to consider new aspects of caregivers behavior such as,
motivation and intention. Being left by a caregiver takes on a different meaning
at later developmental stages than it does for an infant. As an alternative,

Waters and Deane (1985) developed a Q-sort insthment for capturing

31

attachment in preschool children.
Main

Understanding how attachment patterns were transmitted from one
generation to the other was the research aim of Main and her colleagues at
Berkeley. Main, et al., (1985) expanded the ﬁeld of attachment research by
applying the theoretical concepts of object relations and internal models of
representation to behavioral analysis with adults. Main and her associates
developed a model to explain how an adult’s style of relating was repeated and
transferred to an adult’s own children by developing a protocol to assess adult
memories and perceptions about one’s own childhood.

The protocol they developed was a semi-structured qualitative research
tool called the Adult Attachment Interview (AAI). The AAI consists of an interview
designed to probe for descriptions of early relationships with parents or early
caregivers. Applying the attachment classiﬁcations identiﬁed by Ainsworth, Main
and her associates developed a similar system for classifying adult patterns of
attachment classiﬁcations. By administering the AAI to adults, and by
administering the Strange Situation procedure to their respective infants, it
became possible to compare parent and offspring type of security of attachment.
Main’s work established a correlation between infant attachment patterns and
their mother’s own childhood attachment patterns. Through the analysis of the
narrative reports of parents’ responses to questions, it was believed that aspects
of the adult’s intemal working model were revealed. Main proposed that the

actual events of childhood were less important than were the adults” memories

32

and feelings about their childhood experience. The AAI asks respondents to give
information about relationships to both parents past and present, to provide ﬁve
descriptive adjectives to describe each parent and to cite supportive evidence as
to the degree of parental acceptance or rejection that they recalled during their
childhood. They also ask the participant to analyze why their parents behaved as
they did. Main et al. (1985) looked at the content as well as contradictions and
inconsistencies in the linguistic responses. Through analysis of the AAI they
found four kinds of attachment relationships. Adults with a secure attachment
were found to value relationships; they also exhibited readiness of memory and
ease of discussion. What became apparent with securely attached adults was

the lack of idealization in the description of parents or description of childhood

experience. Adults in the second pattern, had a dismissing pattern of insecure
attachment, which was characterized by considering their parents as having little
value, or inﬂuence on their adult personalities. In a third pattern of insecure adult
attachment, some respondents remained preoccupied and dependent on their
parents for approval and were still struggling to please them. A fourth group of
insecure adults was unresolved in regard to parental attachment, and was
labeled “disorganized” in their attachment pattern. van IJzendoorn et al. (1992)
analyzed the work of Main and Goldwyn and found that in about 80% of the
cases studied, infant-mother attachment could be predicted on the basis of the
mother's lntemal working model of attachment, that resulted from her attachment

history with her own mother and father.

33

Security of Attachment in the Mother-Child and Father-Child Relationship History

The contribution from Main and her colleagues has been the primary
mechanism to extrapolate from childhood history the salient memories that depict
the various patterns of interaction between parent and child. As stated in
Chapter I the AAI qualitative interview has been widely used both for infant
mental health research and for retrospective studies looking at the antecedents
of adult psychopathology. The AAI has also inﬂuenced how other researchers
have conceptualized the measurement of adult attachment in romantic
relationships as will explained later in the chapter.

The second section of the literature review continues to build on the
importance of the AAI as a qualitative narrative of whether an individual has
experienced a secure or insecure relationship with a mother and father.
lntrinsically linked to the AAI is the concept of lntemal models of representation,
which is taken from both attachment theory and infant mental health research.
After presenting how a secure parent-child attachment develops, this literature
review will then look at the implications of disordered attachment in infancy as a
precursor to childhood behavior problems and later difﬁculty in adult
interpersonal relationships.

Initially the AAI was only given to mothers. However, in recent years, the
AAI has been administered to fathers, and the ﬁndings have proven equally
useful (Chornesky, 1992; Cohn, Silver, Cowan, Cowan 8 Pearson, 1992; Cowan
et al., 1996; van lJzendoom 8 Bakersmans-Kranerburg, 1996). Similarly,

fathers’ lntemal representations were of equal importance to mothers’ when

 

analyzing parent-child attachment. Some researchers (Chornesky, 1992; Stern,
1991; Trad, 1992) were interested in knowing whether fathers’ internal
representations of their infants were predictive of later father-child attachment as
it was of mother-child attachment. Expectant fathers who are ‘attuned’ or ‘out of
touch’ were found to have internal fantasies about their infants similar to mothers
(Crittenden, 1990; Stern, 1991).

Chornesky’s study (1992) focused on fathers of different ages and their
relationships with their sons. This study, like others Using the AAI, found a
correlation between the father’s security of attachment with his own parents and
the father’s current relationship to his male offspring. The outcome of father-son
attachment was dependent on whether the father had updated or modiﬁed his
view of his father as he reached adulthood. Without conscious self-awareness or
clinical intervention the insecurely attached adult was more likely to have an
obstruction in memory as evidence by the defensive exclusion of certain personal
experience and information. Fathers, who were able to update their view of their
primary attachment ﬁgures from childhood, were more likely to have the capacity
to modify their own relationships with their sons. The ﬁndings indicate that
intervention with these fathers changed not only their perceptions of their
parents, but also changed the amount of awareness they directed at their own

children (Chomesky ,1992).

lntemal Models of Representation and the Transmission of Attachment

The intergenerational transmission of relational patterns was thought to be

35

modulated by the parents’ lntemal working models, mental representations of
themselves and their child, which get transmitted through interactions with
infants. According to several researchers (Fonagy, et al.,, 1991; Levine, Tuber,
Slade, 8 Ward, 1991; Stern, 1995), the parent’s mental representation of his or
her own parents may be the single best predictor of the pattern of attachment
that parents will establish with their own infants at 12 months of age. The
parental lntemal representational model reveals the amount and kind of distortion
in a parent’s image of the infant. Through analysis of the parent’s
representational model, a researcher can track the dominant themes regarding a
parent’s view of their infant and can discern the narrative coherence between
parental childhood experience and evaluation of that experience (Stern, 1995).
Much is to be learned about how these lntemal representations get acted out
with the infant. However, personal expectations, cultural inﬂuences, external
events, and infant characteristics are all believed to interact and form patterns of
behavior.

In other words, the kind of attachment behavior developed in their own
childhood is likely to be unconsciously repeated by an adult with the next
generation by utilizing the same pattern of attachment behaviors with her/his own
child (Minde 8 Hesse, 1996). lntemal representation models are thought to be
the mechanism through which patterns of attachment formulate a template for
later relationships. Some attachment researchers like Sroufe and Waters (1977)
disagree that early relationships form a template for later relationships. They

view attachment, as ﬂuid, as changing with time and circumstances, and as an

ongoing developmental process. For example, an insecure attachment
experience with a primary caregiver does not always result in the replication of
the same pattern of behavior with their children. When parents with insecure
attachment histories had access to early memory, they also had more conscious
awareness of the difﬁculties in the early parent-child relationship. As a result,
they were highly motivated to parent differently with their own children and were
less likely to repeat those same behaviors. What matters is how parents
conceptualize and perceive the caregiving they received. It is the cognitive
decisions about one’s prior experience that guides interaction with children.

Intergenerational patterns appear to be repeated through selective
memory and unconscious mental processes. The existence of an internal
working model suggested that parents have both mental representations of
themselves and their child, which may be consciously or unconsciously active
during the initial process of attachment, as well as within the ongoing parent-child
relationship (Arnmanti, 1991). This concept suggests that an individual forms two
representational models of each relationship: one of the self and one of the other
person in the relationship (Crittenden, 1990).

Stern (1991) explained that the interpersonal world of the infant is
remembered and translated from one generation to another through a highly
selective interpretation of one’s own selective memory. More speciﬁcally, a
caregiver’s view of a baby is not exclusively based upon external reality, but also
consists of lntemal mental processes, which encompass idealized images,

fantasies, and perceptions of the self and other. All parents are believed to have

37

some distortion about themselves and their infants. To clarify, a positively
attached parent may inﬂate or exaggerate positive afﬁrmations of their infant’s
capabilities whereas a negatively attached parent may perceive the infant’s
behavior as purposefully bad and characterize them with negative attributes. In
this way, the positive or the negative view of the self gets projected on to the
infant as a perception of the other in the relationship.

By the end of the ﬁrst year, the infant begins to develop his or her own
lntemal working models of self and other, based on his or her experience of the
care received and a generalized expectation of the caregiver (Main et al., 1985;
Sroufe, 1986). A child’s working model of attachment develops out of the quality
of the relationship between parent and child and the history over time of the
child’s adaptation to the caregiving environment. Attachment relationships are
believed to modify over time with alterations in cognition due to the onset of new
developmental stages.

Through interactions with parents and a construction of an lntemal
working model of the self and of the other, young children begin to incorporate
their parents’ style of relating. Attachment literature suggests that the lntemal
working model ﬁrst emerges at about 12 months, (Schore, 1994) the early
practicing period. It is at this developmental stage when there is the
simultaneous anatomical maturation of the orbital frontal cortex that allows for a
developmental advance in memory and cognitive functions. More speciﬁcally,
Schore (1994) asserts that lntemal representations encode both the infant’s own

physiological affective responses but also tracks the emotionally expressive face

38

 

of the attachment ﬁgure. lntemal working models are understood to enable an
infant to form expectations and to evaluate the interactions that regulate his/her
attachment system.

Schore’s contribution to the conceptualization of the lntemal working
model was demonstrated through an explanation of what happens in the
autonomic physiological-affective processes, as infants use the emotional
responsiveness of the caregiver to regulate themselves. For example, at about
12 months of age when parental demands for socialization increase, infants
experience parental disapproval in tone of voice and facial expression. When
being shamed or scolded, infants switch from a sympathetic state of high arousal
and exploration to a parasympathetic state of low energy and decreased mobility.
Through repeated experience with parental demands for socialization,
socioemotional components of the infant-parent relationship inﬂuence changes in
the infant's developing orbital frontal cortex (Schore, 1994).

lntemal working models are large order memory structures that contain all
of the processes involved in the individual’s subjective experience of others in
social relationships (Zeanah et. al, 1993). They include processes of attention
and perception, of affect selection, of memory evocation, and of behavioral
responses to others in important relationships (Zeanah, et al., 1993). Object
constancy and the ability to have a representational model of the primary
attachment ﬁgure are intrinsically linked. When an infant has the cognitive ability
to know that an object exists continuously in time and space, an infant can guide

his/her behavior by initiating certain behavior and inhibiting others, through

39

stored information, and through internalized representational memory (Schore,
1994).

As patterns of interaction and affective responses are repeated in close
relationships over time, infants build expectations about future interaction with
parents that guide their interpretations and behaviors in new situations (Lyons-
Ruth 8 Zeanah, 1993). According to this viewpoint, cognitive and affective
constructs become incorporated into the personality structure of the individual
and become relatively stable over time. Intergenerational transmission involves
not only organized patterns of behavior, but organized ways of thinking and
feeling in relationships that accompany these interactional patterns (Lyons-Ruth
8 Zeanah, 1993).

During the second year of infancy, Schore (1994) asserted that working
models are further organized into highly complex, symbolic, and enduring
working models that contain auditory, as well as visual, tactile, and olfactory
components. Schore suggested that while the toddler was separating and
individuating from his caregiver the toddler used the facial cues and affective
regulating qualities of the caretaker to establish his/her system of lntemal
regulation. When an infant or toddler was unable to re-establish psychobiological
attunement, or reciprocity with the caregiver after periods of high tension or
frustration, the young child developed an internal working model of the caregiver
that was rejecting or unavailable and held a concordant view the self as unworthy
of comfort.

In summary, when a caregiver controls their own affect, they are more

40

likely to have infants who also learn how to modulate their own feelings. This is
accomplished through a complex interaction between neurological development
and social-emotional experiences that impacts how the brain responds to
environmental stressors (Schore, 1994). Infants who are able to satisfactorily
internalize infantile experiences are able to develop a balanced mental
representational world, which is experienced and borrowed from the experience
with primary caregivers (Ammanti, 1991; Stern, 1995).

As a child matures, perceptions of the self and other are believed to
change over time through the normal developmental stages and increasing
cognitive abilities. Through remembered experiences, toddlers learn to expect
certain responses from their parents. Preschool children begin to grasp the
meaning behind parental behaviors and intention. Through improved language
skills and lntemal strategies, the child begins to more directly inﬂuence parental
behavior through alterations in their own behavior. (Ainsworth, 1989). Because
of the inherent difﬁculty of measuring attachment relationships at different
developmental stages, the theoretical constructs without supporting empirical
evidence are insufﬁcient to explain how attachment behaviors change and adapt
with the emotionally, cognitively and socially developing school-age child.
Nevertheless, it is agreed that by late adolescence or early adulthood, children
are expected to be able to function autonomously from their caregivers and to
begin to pair with others to form their own afffectional bonds.

One argument is that although the early caregiving environment may

inﬂuence attachment relationships, security of attachment is considered

41

amenable to change through clinical treatment, and/or a strong desire to
transcend one’s childhood experience. For this reason, however, attachment
theory is considered an important theoretical bridge between early child
development research and the development of clinical social work intervention
(McMillen, 1992). According to McMillen, attachment theory’s use of internal
working models and patterns of attachment may serve to integrate
developmental theory with both family therapy and cognitive approaches to

clinical practice.

Disordered Attachment in Infancy

To understand the behavioral differences between secure and insecure
infants and toddlers some researchers have studied the behavior of caregivers
and the respective adaptation in the child over time. Goodman and Brumley
(1990) were particularly interested discovering how disordered attachment
develops during infancy. They found that mentally ill parents tend to provide
overly harsh or punitive care and exhibit behavior that is difﬁcult for an infant to
interpret. Mentally ill mothers generally engage in more punishment and
discipline and depressed mothers tend to avoid punishment and discipline and
provide a lower quality physical environment than do mothers in the normal
population. Mental illness or pervasive depression in the caregiver signiﬁcantly
negatively inﬂuence an infant’s mental performance as evidenced by scores on
the Bayley Scales of Infant Development or McCarthy Scales of Children’s
Abilities (Goodman 8 Brumley, 1990).

42

Other researchers were interested in the treatment of disordered
attachment. When infants with depressed and socially disadvantaged mothers
and caregivers receive home-visiting intervention services, there is improvement
in the infant’s cognitive development (Lyons-Ruth, Connell, Grunebaum 8 Botein,
1990). This was found to be true especially for infants with depressed mothers.
When mother and infant receive services, there are improved infant scores on
the Bayley when compared to nonserved infants. (Lyons-Ruth et al., 1990). High
risk mothers use more controlling teaching strategies, use less physical
distancing, and give fewer attributions of child competence than do low risk
mothers (Diaz, Neal, 8 Vachio, 1991).

Attachment disorders are deﬁned as diagnostic categories based on either
the DSM-IV or the lCD-10 that describe persistent deﬁcits in a child’s ability to
socially relate to others that begin before age 5 (Zeanah,et al., 1993). These
authors identiﬁed two general kinds of attachment disorders with the onset in
childhood. One type was labeled inhibited, and comprised ambivalent, inhibited
or hypervigilant responses that were centered on more than one adult. The other
type was labeled disinhibited, and encompassed indiscriminate oversociability, a
failure to discriminate when seeking comfort, and poorly modulated social
interactions. Disorders of attachment were thought to represent profound and
pervasive disturbances in the child’s feelings of safety and security, as a result of
grossly inadequate or repeated changes in caretakers (Zeanah, et al.,1993).

Zeanah, et al.(1993) suggested when considering disordered attachment

as a diagnostic category, several opposing areas of child behavior be evaluated.

43

These areas are lack of affection or promiscuous affection; absent, odd or
ambivalent comfort-seeking from the caregiver; excessive dependence or failure
to use the supportive presence of the caregiver when needed; noncompliance or
over compliance; excessively inhibited exploratory behavior or exploration
without checking back; over solicitous and inappropriate caregiving or the
excessively bossy and punitive attempts by the child to control the parent's
behavior; and failure to reestablish affective contact after brief separations
including ignoring, angry, or unaffectionate responses. Using case examples,
these researchers described ﬁve types of attachment disorders: nonattached,
indiscriminate, inhibited, aggressive, and role-reversed. They proposed that
attachment problems become psychiatric disorders when emotions and other
behavior displayed in connection with the caregiver relationship are so disturbed

that the infant is in persistent distress.

Insecure Attachment and Child Behavioral Problems

Infant mental health research ﬁndings offer support for the hypothesis that
early attachment disorders contribute to poor social skills in young children.
More speciﬁcally, insecure attachment in infancy is associated with hostile
behavior in preschool children (Lyons-Ruth, et al., 1993). Secure attachment in
infancy has been studied as a predictor of childhood social competence with
peers, self-esteem, curiosity, coping with novelty, coping with failure, enthusiasm
and persistence in problem solving, independence, and frequency of behavior

problems among other things (Sroufe, 1985).

Aggressive behavior and poor social skills in childhood were associated
with early harsh punitive care (Weiss 8 Dodge 1992). In a study of infants and
their caregivers living under conditions of poverty, maternal depression, and
caregiving inadequacy, 71% of children with unusually high levels of hostile
behavior at age 5 had been disorganized in their attachment behavior in infancy.
This compares to only 33% of children not exhibiting hostile behavior (Lyons-
Ruth et al., 1993). Other longitudinal studies support the association between
type of security of attachment in Infancy and later problems with aggression
(Lyons-Ruth, 1996). In a cross sectional study, risk factors predicting later
aggression in children include: family adversity, parental hostility, parental
depression, and child cognitive deﬁcits (Lyons-Ruth, 1996).

The kind of attachment relationship that exists between child and parent
has application for a broad range of children who have severely disturbed
relationships with their primary caregivers (Zeanah, 1996) but especially for
maltreated children. In a comparative study of maltreated and non maltreated
children ages 8-12 years, Toth and Cicchetti (1996) studied the existence of
depressive symptomatology and perceived competence in developing children.
This study suggested that the experience of abuse and neglect during childhood
may lead to the development of a negative representational model of primary
attachment ﬁgures, a negative view of the self, as well as a negative expectation
in relationship to others. Maltreated, insecurely attached young children are
likely to have a rigid pattern of relating and are closed to adapting to new ways of

relating (Crittenden, 1990). They are likely to exhibit severe behavior patterns

45

and to tolerate punishment, without recognizing the need for an lntemal
adjustment to external demands from their caregivers. Toth and Cicchetti (1996)
concluded that within their study of maltreated children, children who had
experienced sexual abuse have a higher incidence of depressive symptomology
and of impaired behavioral conduct than do children who are maltreated but are
not sexually abused.

Not all researchers agree that early childhood problems of aggression are
the result of early attachment difﬁculties. Fagot and Kavanaugh (1990) caution
researchers who infer that clinical observations and early attachment problems
are directly associated. However, Fagot and Kavanaugh (1990) do agree that in
infancy some child behavioral signals are thought to be indicative of distress in
the mother-infant relationship. Speciﬁcally, three disorganized attachment
indicators that are directed toward the caregiver are fear or dysphoria,
irresolvable conﬂict between approach and avoidant behavior, and elevated
cortisol levels after separation (Lyons-Ruth 1996). All types of symptoms of

disorganization suggest distress in the infant-caregiver relationship.

Early Insecure Attachment and the Development of Personality Disorders
Additional support for disordered attachment affecting the formation of
personality disorders also comes from retrospective studies of the adult
psychiatric population. Attachment research is currently one of the most
promising avenues for understanding the psychological antecedents of

disordered behavior in adults (Jones, 1996). Fonagy, et al., (1996) applied Main

46

and Goldwyn’s revised classiﬁcation of the AAI, to a psychiatric sample to
determine if early childhood history was correlated with psychiatric status.
According to Fonagy et al., (1996) the new Main and Goldwyn system provided
for four slightly different major classiﬁcations of attachment: (F) free-autonomous,
coherent valuing of attachment; (D) dismissive idealizing, derogatory, and cut off
from attachment experiences; and (E) preoccupied passive, angry, and
entangled by past relationships. A fourth category (CC) was used for unable to
classify.

In the Fonagy 1996 study the psychiatric sample was recruited from an
inpatient facility and included patients who had a chronic or failed history of
treatment with an average hospital treatment stay of 9.4 months with a high
proportion (45%) having an Axis I diagnosis of substance abuse. The control
group, matched on demographics, was recruited from an outpatient department
of a university teaching hospital. This study found overwhelming support for an
association between psychiatric disorder and unresolved difﬁcult relationships in
early childhood. In this particular study, the psychiatric sample was weighted
toward the more extreme end of the continuum of the severe chronically mentally
ill where childhood difﬁculty would not be an unexpected ﬁnding. A more
interesting aspect of the study was the attempt to connect a speciﬁc type of
attachment difﬁculty, and history of abuse, with a certain diagnostic category, i.e.,
Borderline Personality Disorder.

In an effort to prove that insecure attachment and later personality

disorder have common antecedents, Brennan and Shaver (1998), studied a large

47

nonclinical sample of young adults. The researchers used a self-report measure
developed by Bartholomew and Horowitz (1991) to measure attachment style
and a scale to measure perceived quality of treatment by mother and father
during childhood, along with the Personality Diagnostic Questionnaire (PDQ-R).
Their results suggested that their sample of college age adults met their criteria
for having at least one personality disorder. The rate of prevalence was 75%,
which was unusually high for a non-clinical population. The authors admit that
their deﬁnition of personality disorder did not meet criteria for clinical diagnosis.
Using a second more rigorous scoring method in this research, the prevalence
rate fell to 14% which was closer to the 16.6 % rate found in the clustering of
individuals who have several different kinds of psychiatric disorders in the United
States (Kessler, 1994).

Within the Brennan and Shaver (1998) study, the attachment style rating
fell into the following ranges: 47.9% were secure, 21% were fearful, 15.2 were
preoccupied, and 15.9% were dismissing. This ﬁnding generally compares with
the breakdown in attachment classiﬁcation for infants (van IJzendoorn et al.,
1992). Childhood relationship to parents was measured by self-report and
covered three domains: level of acceptance or rejection, degree of independence
verses over protectiveness, and defensive idealization of parental ﬁgures.

From their 1998 ﬁndings, Brennen and Shaver ordered personality
disorders into three categories: (1) General Pathology (schizotypical, paranoid,
avoidant, obsessive-compulsive, self-defeating, and borderline) was depicted by

low sociability, distrust, low self-esteem, and disordered thinking; (2) Excessive

48

self-reliance or dependence characterized counter-dependence (schizoid,
hystrionic, and dependent); and (3) Interpersonal aggression and callousness
illustrated psychopathology (anti-social, sadistic, passive-aggressive and
narcissistic). Thinking about personality disorders with this kind of typology
ampliﬁes how different attachment experiences may inﬂuence the defensive style
or personality characteristics of an individual, particularly for those with an early
history of harsh, inconsistent, or punitive care from one or more caregivers.

Despite the strong conceptual arguments in this study and the interesting
manner in which personality disorders were organized, one weakness of the
Brennan and Shaver ﬁndings was that all of the results were determined by self-
report without any secondary sources or observations for validation. They admit
the instrument used to depict personality disorder was not a sufﬁcient match with
a clinical diagnosis of personality disorder. What their research did provide was
a replication of the Bartholomew and Horowitz’s ﬁnding and thus adds credence
to the description of the defensive style an individual may use to relate to an
adult partner.

In their instrument, West and Sheldon (1988) organized pathological
attachment patterns, based on clinical experience and relevant clinical literature,
according to behavioral difficulty with an adult attachment ﬁgure. According to
their instrument which was tested on a college-aged sample, there were four
factors or descriptions of attachment difﬁculty: (1) compulsive self-reliance;
avoidance of the caregiver for help, avoidance of affection or closeness, and

uncomfortable with the partner’s need for him/her; (2) compulsive caregiving;

49

consistently placing high priority on the needs of the other, feelings of self-
sacriﬁce and martyrdom, and provision of care whether or not it was requested;
(3) compulsive care seeking; the deﬁnition of life according to problems with
which one needs assistance, the deﬁnition of attachment relationship in terms of
receiving care, and the expectation that partner will assume major responsibility
in all areas of life; and (4) angry withdrawal; negative reaction to perceived
unavailability of other, negative view of perceived lack of responsiveness in
partner, and generalized anger directed toward the attachment ﬁgure (West 8
Sheldon, 1988). This study was an attempt to measure some of the anxiety,
anger and ambivalence in attachment behaviors as ﬁrst described by Bowlby.
Results indicated that the lntemal reliability for compulsive self-reliance,
compulsive care—giving, and angry withdrawal were acceptable whereas
compulsive care giving did not meet expectations. They conceptualized that
attachment behaviors were on a continuum from distant and detached patterns to

close and enmeshed patterns.

Measurement of Adult Attachment

The following section of the literature review emphasizes the contributions
of some researchers whose work has been focused on developing instruments to
capture the adult attachment style in romantic relationships. As part of the
presentation and initiation of new measures of adult attachment, each researcher
shares his or her ideas about important variables within the adult relationships.

This second section of the literature review will summarize ﬁve measurements of

50

adult attachment by reviewing the work of: (1) Hazan and Shaver, (2) Collins and
Read, (3) Simpson, (4) Bartholomew and Horowitz, and (5) Grifﬁn and

Bartholomew.

Hazan and Shaver.

In 1987, Hazan and Shaver developed a three-category measure to
capture adult attachment styles that involve romantic relationships. In
accordance with Ainsworth’s infant classiﬁcation system from the Strange
Situation, Hazan and Shaver, classiﬁed adult attachment into one of three types;
secure, anxious-ambivalent and avoidant. Adults with secure attachment found it
easy to get close to others and to be comfortable in depending on others.
Secure adults were not worried about being abandoned, nor were they worried
about someone getting too close. An anxious-ambivalent style of attachment
was typiﬁed by relationships where others were reluctant to get as close as the
individual would like, and by worry whether the partner would stay and worry
whether the partner really loved them. An avoidant style of attachment was
described as uncomfortable being close to others, difﬁculty in trusting, and
difﬁculty in allowing themselves to be dependent on others. The avoidant adult
was uncomfortable with too much closeness or intimacy in a relationship.

The Hazan and Shaver (1987) study offered a new way to look at adults in
love relationships. Although it was initially tested on a college age sample, it has
been widely used on other adult populations, and has been widely used as a

research instrument, with other researchers making modiﬁcations (Bartholomew

51

8 Horowitz, 1991; Collins 8 Read, 1990; Simpson, 1990). Hazan and Shaver
(1987) proposed that adults with different attachment styles, hold different beliefs
about romantic love. Secure adults were most likely to view their partner as a
trustworthy friend. Anxious-ambivalent adults were most likely to fall in love, and
then long for the partners’ reciprocation. Avoidant adults were least likely to
accept their partners’ faults and were highly defended regarding their own

dependency on others.

Collins and Read

Collins and Read (1990) developed an 18-item scale, Adult Attachment
Scale, that was normed on a college age sample and was based upon Hazan
and Shaver's measurement of attachment style. Factor analysis of their scale
revealed three dimensions of attachment: comfort with closeness, ability to
depend on others, and anxious or fearful about being abandoned or unloved.
Whereas the original Hazan and Shaver measure asked respondents to read a
description of attachment behaviors and pick the one description most like
themselves, the Collins and Read measure was a self-administered instrument.
Collins and Read added questions to indicate whether or not the attachment
ﬁgure was available and responsive and also how the person coped with
separations from the romantic partner.

In summary, Collins and Read’s deﬁnitions of attachment style were: a)
depend, the extent to which subjects could trust others and depend on them to

be available; b) anxious, the degree of anxiety in the relationship, e.g., the fear of

52

being abandoned; and c) close, the extent to which the subject felt comfortable
with closeness and intimacy. Because Collins and Read’s factor I (depend) and
factor 3 (close), included both secure and avoidant descriptions, and factor 2
(anxiety) included items from anxious and secure descriptions, the scales did not
correspond with Hazan and Shaver’s three styles of attachment: secure, anxious
and avoidant. What their scales did depict were the characteristics that underlie
the three styles.

When analyzing subjects who took both the Hazan and Shaver instrument
and the Collins and Read scale, their scale correctly classiﬁed about 73% of the
subjects into similar styles of attachment. In doing so, they found that most of
the adults who scored secure on one measure, also scored secure on the other.
They had more difﬁculty with reconciling anxious and avoidant on the two
measures. However after additional cluster analyses, they found a better degree
of match with Hazan and Shaver. A secure style was indicated by a high score
on close and depend and a low score on anxiety. Anxious styles were implied by
high scores on anxiety and with moderate scores on dimensions of close and
depend. The avoidant style was associated with low scores on the close, depend

and anxiety categories.

Simpson
Simpson (1990) followed 144 young adult, dating couples longitudinally to
study differences among secure, anxious, and avoidant pairs along with the kinds

of emotions expressed and the level of distress following dissolution of the

53

relationship. Simpson extrapolated questions from the original Hazan and
Shaver (1987) attachment vignettes and developed 13 questions on a Likert
scale and administered other instruments to measure interdependence, trust,
commitment, range of emotions, and satisfaction with the relationship. Six
months later, measures were repeated by telephone interview. Thirty-four
members of the sample were no longer dating. Couples, who scored higher on
the security of attachment index, also scored higher on level of interdependence,
displayed greater commitment, greater trust and greater satisfaction with the
dating relationship. Men and women with an avoidant attachment index were
found to be in dating relationships with lesser amounts of independence,
commitment, trust and satisfaction. Men and women with an anxious attachment
index were involved in dating relationships with less trust and satisfaction.
Commitment and interdependence were not reliable indices for this last category
of attachment.

Emotional reaction to relationship dissolution did not yield any strong
results in this Simpson (1990) study. There are several reasons why the
measurement of affect following the termination of the relationship was
negligible. The results were based on a small sub-sample of 46 couples, or 36%
of the total, which may have resulted in insufﬁcient power for division into four
categories of discrete responses by gender and by attachment style.
Additionally, given the mean age, 19 years, and life stage of college-age
students, it follows that some of the couples who ended their relationship were

likely to be in dating relationships that were transitory or experimental. Couples

actively involved in the process of partner selection and in relationships of
relatively short duration would be expected to have differences in emotional
reaction to relationship termination compared to the couples who chose to marry

or who have spent years together.

Bartholomew and Horowitz

Bartholomew and Horowitz (1991) approached the determination of
attachment style from a different conceptualization of how to define security
within relationships. Their sample consisted of well-functioning, college-aged
adults from two-parent upper-middle class families. Their four-category model
included positive and negative valances of the adults’ view of other and view of
the self, which coincided nicely with how internal models of representation get
operationalized. From this abstraction they developed a model of attachment
style that weighed degree of dependence with degree of avoidance in adult
relationships. Their typology of attachment style coincides with similar deﬁnitions
of attachment behavior found in the infant mental health and developmental
psychology literature. Individuals with a secure style were comfortable with
intimacy and autonomy, and had a positive view of the self and other. Adults
with a preoccupied style were preoccupied with relationships and a positive view
of the other and a negative view of the self. Those with a dismissing style were
counter—dependent and had a negative view of the other and positive view of the
self. Those with a fearful-avoidant style of attachment had difﬁculty with

intimacy, and were socially avoidant, and the view of the self and the view of the

55

other were both negative.

Bartholomew and Horowitz’s measurement of attachment occurred
through the use of a semi-structured interview as well as self-report and friend
report. Three independent raters evaluated each subject’s tape-recorded
responses and placed them on four 9-point scales. A secure prototype was
characterized by valuing friendships, the capacity to maintain close friendships
without losing personal autonomy, and coherence and thoughtfulness in
discussing relationships. A dismissing prototype was characterized by the
devaluing of the importance of close relationships, restricted emotionality, and a
emphasis on independence and self-reliance and lack of credibility or clarity in
discussing relationships. The preoccupied prototype is characterized by over
involvement in close relationships, dependence on others for self-validation, and
a tendency to idealize others, and incoherence and exaggerated emotionality
when discussing relationships. The fearful prototype is characterized by an
avoidance of close relationships because of a fear of rejection, a sense of
personal insecurity, and distrust of others. In the Bartholomew and Horowitz
(1991) study, 47% were considered secure, 18% were dismissing, 14% were
preoccupied, and 21 % as fearful.

In a second study with the same sample, Bartholomew and Horowitz
(1991) looked at subjects’ family of origin and peer relationships as it related to
interpersonal problems related to warmth and dominance. Questions were
asked about their family backgrounds, the quality of their relationships with each

parent, their reactions to parental separation, feelings about parental rejection,

and parental comfort during childhood distress. They found that corresponding
family ratings and peer relationships were signiﬁcantly correlated with each other
as well as with the presence of personal problems. The preoccupied (anxious)
and the fearful (avoidant) had the highest levels of interpersonal distress.
However, Bartholomew and Horowitz could not conclude from their hierarchical
regression analysis that family relationships were the dominant feature in
determining individual differences regarding attachment style. Peer relationships
were more correlated than were family relationships. They concluded that their
model of attachment style was not reducible to early childhood experiences.
They also stated that none of the subjects clearly ﬁt into one attachment
prototype, as subjects had different attachment outcomes across time and
relationships. In sum Bartholomew and Horowitz’s major contribution was the
creation of a four-category model of attachment, that has been accepted by other

attachment researchers including Brennan and Shaver (1998).

Brennen and Shaver

In 1995, Brennan and Shaver, sought to improve upon Hazan and
Shaver's (1987; 1990) categorical rating of the measurement of attachment style
and sought to extend Collins and Read’s work by correlating it with 7 factor
based scales. The seven scales included questions about frustration with
partners, proximity seeking, self-reliance, ambivalence, trust/conﬁdence in
others, jealousy/tear of abandonment, and anxious clinging to partners. From

these 7 scales, they were able to describe in more details some of the

57

characteristics of different attachment styles. Both anxious-ambivalent and
avoidant adults were considered to be insecure types. Brennen and Shaver
(1995) reported that both anxious-ambivalent and avoidant individuals are more
frustrated with previous partners than are secure individuals. Avoidant people
were found to differ from secure adults in the areas of proximity seeking and self-
reliance. Unlike secure individuals, anxious-ambivalent individuals were more
jealous and ‘clingy’.

In discussing their results with other researchers of adult attachment
ﬁndings, they concluded that an anxious/ambivalent attachment style was
marked by simultaneous fear of abandonment, and anger at partners who seem
insufﬁciently available. The anxious/ambivalent individual remains preoccupied
with potential loss, expressing anger and fear about being abandoned. An
avoidant style involves denial of need for the relationship, and distraction through
work, failure to focus on feelings, and an inability to self-disclose. Those with a
secure attachment style were described as the opposite behaviors of the
avoidant and preoccupied styles and as having the capacity to trust and share

with partners appropriately.

Grifﬁn and Bartholomew

Grifﬁn and Bartholomew (1994) sought to validate the two dimensions that
underlies the Bartholomew and Horowitz’s (1991) four category model of adult
attachment through reanalysis of the original data. In study 1, they found

convergent validity by the moderately high correlation within each attachment

58

dimension (self and other) and across instruments, they found discriminate
validity within the three reports. They reported that the two-dimensional
structure, self and other, was a signiﬁcantly better ﬁt.

In a third study, Grifﬁn and Bartholomew (1994) used the two dimensional
model with committed couples to determine if gender was related to attachment
style. They were interested in how variations of Hazan and Shaver’s three
attachment patterns, Simpson’s two attachment sub-scales, and Collins and
Read’s thee attachment sub-scales compared with their own attachment
measure. They expected that Collins and Read’s sub-scales, anxiety, comfort
with closeness and comfort with depending on others and Simpson’s sub-scales
of anxiety and avoidance would correspond with their self and other two
dimensional attachment. For inclusion in the study, couples had to be in a
relationship for at least two years, have no children and be under the age of 35.
Of the 78 heterosexual couples in the study, 28% were married, 44% were co-
habitating, and 25% were living separately. They found that for both men and
women, the Collins and Read and Simpson measures were structurally similar to
their two dimensional attachment variables when compared to qualities of self
and other. They were surprised to ﬁnd that the Hazan and Shaver model also
had a goodness of ﬁt. Grifﬁn and Bartholomew concluded that their own
measure captures the dynamics between self and other as ﬁrst described by
Bowlby. They asserted that the other attachment measures developed by Hazan
and Shaver, and Bartholomew and Horowitz, and Simpson, were more

descriptive of the strategies for maintaining felt security as ﬁrst identiﬁed by

59

Ainsworth.

Attachment History and Partner Selection

Although there has been extensive sociological research on marriage,
much is still unknown about the attachment dynamics involved mate selection.
Attachment research has primarily focused on the formative years with some
recognition by researchers and clinicians that early attachment relationships are
connected to adult adaptation throughout the lifespan. Ainsworth (1989) viewed
infant attachment behaviors as life enduring because they are rooted in
neurophysiological processes of the individual which are inﬂuenced by genetic
and environmental factors. In comparison to parent-child attachment, she
viewed sexual pair bonds as not necessarily symmetrical or reciprocal, but more
complementary in nature. However, other researchers, doing retrospective
studies with adults, connect childhood experience with current adult relationship
difﬁculty and clinical psychopathology (Karen, 1994). For some adults, marital
partners can become primary attachments, for which one experiences persistent
longing, not unlike an infant’s attachment to a primary caregiver (Weiss ,1975).

From the infancy studies, it was inferred that adult expectations of how we
get our emotional needs met, are associated with earlier childhood expectations
formed in relationship to our caregivers. Adults with a secure history of
attachment are more open to trusting and depending on others, whereas adults
with an insecure history develop anxious or avoidant strategies to maintain

relationships. Historically, security of attachment seems to inﬂuence both how

60

one interprets the intention behind others’ behavior and how one unconsciously
replicates the security experience from early childhood. Although a direct
association between childhood attachment history and adult attachment to a
signiﬁcant partner has not been made, there does seem to be some theoretical
and empirical support for the notion that whether an adult has experienced a
secure or insecure attachment pattern will, to some extent, determine partner
selection and reaction to loss when that relationship dissolves.

The following studies address in more detail how attachment style is
associated with partner selection and reaction to loss under circumstances of
divorce. In a 1994 study, Hill, Young and Nord, recruited 20 adult men and 20
adult women whose parents had participated in a larger research study. They
administered several measures: The Home Environment Interview to assess the
nurturing behavior of parents during early life; the Hazan and Shaver Adult
Attachment Questionnaire to determine style of attachment; and a Social
Adjustment scale to measure individual levels of functioning in a variety of areas.
Their ﬁndings suggested that attachment security is affected by early childhood
conditions, especially whether a parent was invested in child rearing and whether
the parent was or was not nurturing. Non-secure adults reported more difﬁculty
in their own mother-child relationship, and held the perception that one parent
was “particularly hard on 1119’. Secure adults differed from the non-secure in the
number of multiple risks present in early childhood, and they were less likely to

have grown up with perceived poverty.

61

In this Hill et al. (1994) study there were differences in the onset, duration
and success of maintaining a co-habitating or marital relationships according to
attachment pattern of the adults. Secure adults were more likely to be in existing
relationships and to score high on personal adjustment. The non-secure adults,
who did have current adult relationships, began those relationships at an earlier
age and after a briefer courtship when compared to the secure adults. This study
was interesting because it included environmental early childhood risk factors,
such as SES, and abusive punishment, and intervals between sibling birth as
possible factors that affect parent-child attachment beyond the relationship itself.

Collins and Read (1990), in another study, examined whether there was a
correlation between a young adults’ style of attachment and the attachment
dimensions of the subjects’ partner. They were interested in adult relationship
quality. They hypothesized that relationships with parents would inﬂuence
expectations about our partners and beliefs about oneself as a love object. They
expected that those with secure attachment style would select partners with a
similar secure attachment pattern. Their ﬁndings indicated that for young adults
who were comfortable with closeness, they usually had a partner who also
enjoyed closeness. There was less matching on the dimension of the ability to
depend on others. There was also less matching between couples who scored
high on avoidance and anxiety. An anxious subject sometimes paired with an
avoidant partner and vice a versa.

One of the stronger associations found in this study was that women with

a history of a warm and responsive father and men with a warm and responsive

62

mother were more likely to be dating partners with a secure attachment style.
Whereas men who described their mother as cold, distant, or inconsistent, and
women who described their father as cold, distant or inconsistent tended to have
partners who were anxious. Both men’s and women’s styles of attachment were
related to their partners’ satisfaction with the relationship. A female subject who
had a partner who scored high on closeness was more likely to give the
relationship high marks. The strongest predictor for male subjects’

dissatisfaction was having an anxious female partner fearful of abandonment.

Adult Attachment and Adjustment to Relationship Dissolution
Bakermans-Kranenburg and van lJzendoom (1997), as part of a larger
study in the Netherlands were interested in determining if there was an
association between an adults’ classiﬁcation on the Adult Attachment Interview
(AAI) and how they perceived the termination of a romantic relationship.
Following the administration of the AAI, the adults were placed into one of three
categories: autonomous, preoccupied, or dismissing. The sample consisted of
46 autonomous, 17 preoccupied subjects and 20 dismissing subjects. Of those
who had experienced the ending of a relationship with an adult partner, 24 were
considered autonomous, 6 were considered preoccupied, and 8 were considered
dismissing. Once classiﬁed into an attachment type, the women were given a
score as to degree of being resolved or unresolved regarding the romantic
relationship. The ﬁndings in the study revealed a high proportion of autonomous

women reported being unresolved which was contrary to the anticipated results.

63

Because this study of young women asked about all ended romantic
relationships, it was not clear if the frequency or kind of losses confounded
results. Additionally, the study did not specify how many were marriages, how
many were living together, or how many were brief romantic liaisons.

Masheter (1991) studied the attachment relationships between previously
married couples. She considered the post-divorce relationships of 265 men and
women to determine the role of attachment in Interpersonal conﬂict. Contact
frequency for couples without children was 63% and those with children was 86%
for couples having occasional contact 2 —2 and 1A. years following divorce. Her
questionnaire measured friendly verses hostile feelings toward the ex-spouse,
and determined that 43% were in the friendly range, 36% were in the neutral
range, and 21 % were in the hostile range. Signiﬁcantly larger proportions of men
(45%) were preoccupied and 60% had friendly feelings toward their spouse,
whereas women were slightly less preoccupied (34%) and less friendly (46%).
Masheter concluded that there were some gender differences in regard to coping
with loss of a partner with women having less difﬁculty. Those adults who had
remarried had lower preoccupation with the former spouse when compared to
those who remained single. Divorced partners with weekly contact were more
preoccupied than were those who reported no contact. Adult well-being was not
related to kind of affect directed at the former partner; however, frequent verbal

argument and preoccupation were associated with poorer well being.

Empirical studies regarding divorce

To differentiate between high conﬂict and low conﬂict divorce relationship
and the accompanying effect on parent—child relationships post-divorce, it is
important to understand some of the expected reactions to divorce by children in
a variety of stages of development. The following section of the literature review
is an attempt to highlight some of the general ﬁndings representative of the larger
body of divorce literature. It is also an introduction to the impact of high conﬂict
on children’s well being in intact and divorced families. The ﬁrst purpose of
discussing both of these areas is to bridge how early attachment history, adult
attachment style and post-divorce parent-child relationships are being viewed
within the context of this study. The second reason for including these particular
ﬁndings is to provide the foundation for the selection of the variables used in this

study.

Child Outcome and Parental Divorce

Studies of the consequences of divorce on children have resulted in
inconsistent ﬁndings. To examine the effects by age and gender, Allison and
Fursternberg (1989) used a large sample from a National Survey of Children to
reexamine divorce effects by comparing children from intact families with children
from divorced families. Data were collected from the custodial parent, the child’s
teacher and the child. Face-to-face interviews were conducted initially and
follow-up interviews were held by telephone. In summary, Allison and
Furstenberg reported an overall reduced well-being for children that experienced

marital disruption in regard to problem behavior, psychological distress, and

65

academic performance. However, children living in stepfamilies fared no better
than children did living with a single parent. The Allison and Furstenberg study
(1989) found no support of the hypothesis that divorce has a greater impact on
males than females. They did ﬁnd support that preschool children were more
negatively affected by divorce than were adolescents. However, this ﬁnding has
limitations, as age at time of divorce was not closely controlled for in this
comparison.

A Johnston, et al., (1989) study looked at child adjustment according to
type of custody arrangement. Secondarily, they explored some of the process
variables that occurred when parents contested custody. They hypothesized that
parents who had frequent contact with each other would learn to communicate
and problem solve on behalf of their children. The sample used were litigating
parents with unresolved custody and parenting time disputes who were referred
for counseling or mediation services by the Family Court. This study included a
high percentage of minorities, which was representative of the California county
community from which the sample was drawn. Families were considered low
SES at $10,000-18,000 per year per couple. Parents ﬁlled out questionnaires
regarding custody as well as information about the number of weekly transitions
between households and amount of parenting time by days per month. The
measure used to assess child adjustment was the Child Behavior Checklist
(CBCL) which was used to place children into normal and clinical groups.
Parents’ level of conﬂict was measured by the Conﬂict Tactics Scale (Strauss,

1979). Families were seen at base line or point of referral and again at 29

months later when parents had been separated for a mean of 4.5 years.

Of the 100 children in the Johnson et al. (1989) study, both parents rated
16% of their children as meeting clinical criteria. These clinicians found an
association between the clinical rating of the child’s behavior and the parents’
level of verbal and physical aggression with each other. The clinical group of
children was discovered to have parents who were more verbally and physical
aggressive with each other and the children were found to make more frequent
transitions between both homes. This study was important because it focused
on divorcing parents who were known to be in conﬂict. What was not established
was whether the children with the behavioral difﬁculty also had a family
background of high conﬂict that preceded the divorce or whether there were pre-
existing behavior problems. Children from the clinical group may have been
reacting to the parents’ style of conﬂict and not to the frequency of contact. The
implication from this research was that court mandated joint custody was
contraindicated when the parents were in high conﬂict and the children were
exposed to ongoing unresolved conﬂict post-divorce.

The Stanford Custody Project begun by Maccoby and Mnookin (1992)
was a longitudinal study undertaken to examine several aspects of divorce.
Study l objectives were: 1) to determine the Impact of joint and sole custody
arrangements on children and their respective relationships to parents; 2) to
discern how much conﬂict occurred between parents in reaching a custody
decision; 3) to understand more about the co-parental relationship when children

have frequent access to each parent; and 4) to identify what factors maintain

67

cooperative, conﬂictual or disengaged styles of parenting in the co-parental
relationship. This study involved an initial collection of data during the process of
separation and follow-up one year later and 3 years later.

The Stanford project was a large study of approximately 1,000 families,
approximately 50% of whom rated themselves as having little difﬁculty in coming
to agreement regarding physical custody with 25% of the families as having
substantial legal conﬂict. Increased parental conﬂict was considered to be the
result of three factors: an indication of unresolved hostility in the parental
relationship, differences in perception as to how much each parent had
contributed to previous child care during the marriage, and concern about the
child when in the care of an incompetent parent or lack of parenting skill. In the
Stanford study physical custody awards resulted in mothers’ receiving custody
66% of the time with fathers receiving custody 9% of the time and joint custody
occurring 20% of the time.

Over time there was considerable movement in the custody
arrangements, although the percentage of mother custody, father custody, and
joint custody arrangements were fairly stable. The area of most movement
involved joint custody. The study by Maccoby et al., (1993) found that after 4.5
years after separation, only a small percentage (13%) of children had not seen
their non-resident parent in the prior year. These researchers believed that the
high rate of paternal participation was related to factors related to motivation to
participate in the study. However, this study found only a marginal correlation

between amount of parental conﬂict and whether or not parent-child contact was

68

maintained, increased or diminished. (Maccoby et al., 1993).

Although parent child contact was maintained, communication in the
coparental relationship appeared to diminish over time. The Stanford study
classiﬁed parents into three different coparenting styles: cooperative, conﬂicted
and disengaged. Cooperative parents were able to talk with each other
frequently and rarely argued. The conflictual parents maintained contact, but
argued frequently regarding child access issues. With the disengaged style of
relating co-parents rarely communicated and made decisions independently of
the other. They found that one parent in half of the sample would refuse to
communicate with the other parent. The cooperative coparenting style occurred
most frequently when neither parent had begun a relationship with a new partner.
In this sample the disengaged pattern of relating became the norm, but they
noticed that some parents, after a period of conﬂict, became more cooperative,
and other parents who began with a cooperative style became more disengaged.
Some of the factors that inﬂuenced the style of coparenting in the Stanford study
were age of child, family size, unresolved interparental hostility, the presence of
legal conﬂict, discrepant perceptions regarding pre-separation parenting roles,
concern about child well-being in the other household, and parent’s new adult
relationship.

The Stanford longitudinal research with a large sample of families was
noteworthy because of their delineation of three kinds of coparenting styles,
cooperative, conﬂicted, disengaged and mixed, which conceptually coincides

with attachment research regarding the three attachments style secure,

69

preoccuppied and dismissing (Bartholomew 8 Horowitz, 1991; Hazan 8 Shaver,
1987; Simpson, 1990). Other studies, which will be described later in this
section, refer to differences in post-divorce interparental hostility. One of the
underlying hypotheses of this study is that parental adjustment following divorce,
of which high conﬂict is a marker, is associated with early attachment
relationships, and their adult attachment style. Those persons with a prior history
of insecure attachment history or insecure attachment style will exhibit more
conﬂict than will persons with a secure attachment style.

Study 2 of the Stanford project focused on adolescent outcome at the last
post-measure. Maccoby et al. (1993) found an interaction between how close an
adolescent (ages 10-18 years) felt towards their parents and the corresponding
level of hostility and discordant parenting between the two households: the more
conﬂicted the parents, the more likely the adolescents felt ’caught in the middle’,
and the lower was the child’s overall divorce adjustment. However, their ﬁndings
also infer that the parenting ability of custodial parents to a large degree
contributes to the child’s post-divorce resiliency. Children’s long term adjustment
was positively associated with (a) having a close relationship with the residential
parent (b) monitoring by the resident parent (c) joint-decision making between
parents (d) low conﬂict with the resident parent and (e) the organization of the
resident parent’s household (Maccoby, et al., 1993). This study in addition to
describing the different styles of post-divorce coparental relationships also lends
support to other studies that indicate that a child relationship with the custodial

parent is a resiliency factor, especially when that parent has a warm consistent

70

approach to parenting.

Quality of Parenting by the Custodial Parent

How well children cope following divorce is moderated by the quality of
parenting in the custodial home. Custodial parents who are authoritative,
characterized by high levels of warmth, support, monitoring, communication, and
ﬁrm consistent control and low levels of punishment and coherence, support
improved adjustment outcome for children (Bray. 8 Hetherington, 1993;
Heatherington 8 Stanley-Hagen, 1997). In the Virginia Longitudinal Study the
cluster of children from single parent and stepfamilies who fared the worst, and
had the most severe behavior problems, came from homes with overt conﬂict
and unsatisfactory conﬂict resolution styles. These children had parents whose
parenting style was not authoritative and they did not have a close relationship
with an adult inside or outside the home (Hetherington, 1993). Kalter, et al.,
(1989) used a small Michigan sample to test six different hypotheses as to which
factors are most predictive of child adjustment post-divorce. Data were collected
on 56 white, well-educated custodial mothers and their school-aged children with
fairly low income. Adjustment measures were administered to mothers and
children. The strongest predictor of child adjustment in this sample was the
custodial parent’s adjustment. Father absence, interparental hostility, and
demand of multiple life stress did not receive support as a factor in child
adjustment. What surfaced in this study was the predictors for male child
adjustment did not match predictors for female child adjustment. Upon closer

examination it was noticed that the research sample included custodial remarried

71

mothers along with divorced single mothers in this sample. Given that this was a
small sample and 19 of 56 mothers had remarried, it appears that the transition
to an another marital status might have different effects on the adjustment of
male and female children. Child outcomes depend upon factors such as the age
and sex of the child, length of time in a single-parent family, family economic
resources, and parental conﬂict before and after divorce. (Barber 8 Eccles,
1992). The Barber and Eccles review of the literature addressed and highlighted
some of the economic and single parenting factors that contribute to adolescent
resilience following divorce. They conclude that having an employed single
mother, may support some normal adolescent strivings, particulaity in regard to
the values connected to work, during this developmental stage, which positively

contribute to adolescent outcome post divorce.

High Conﬂict lnterparental Relationships

Conﬂict between divorced parents has been associated with increased
behavioral problems with children and poorer adjustment. The central aspect of
parental conﬂict is the degree to which the child has been exposed or caught in
the middle of parental differences. One of the difﬁculties in determining the effect
of divorce on child adjustment is separating out the negative effects of parental
relationships in high conﬂict from unresolved attachments of the marital
relationship. Many divorce studies have been cross-sectional in design and
result in outcomes that suggest children who experience divorce consistently
have poorer outcomes than children raised in intact families. However, cross-

sectional designs do not explain the causal direction of the variables nor do they

72

explain developmental effects on children. The following longitudinal studies
were reviewed to sort out some of the differences between exposure to parental
conﬂict from the effects of parental divorce.

Whether the post-divorce parental relationship affected the pattern of
contact between fathers and children was the research question in Ahrons and
Miller’s (1993) study. They followed 64 non-clinical cooperative couples in a
Midwest community, who were primarily white, educated, employed and whose
visitation patterns were stable at the time of divorce. They followed these
couples at three and ﬁve year intervals post divorce. Split custody and father
custody situations were excluded because the focus of the study was in the
maintenance of paternal involvement. Data were collected from both mothers
and fathers with a semi-structured interview lasting 1 ‘25 hours or more. Paternal
involvement was measured in terms of frequency and duration of visitations, and
paternal participation in parenting activities. The relationship with the former
spouse was measured by ten questions that addressed areas of conﬂict and
cooperation.

Seventy-ﬁve percent of fathers in the Ahrons and Miller (1993) study
maintained moderate contact twice per month with their children at the ﬁve-year
mark. Although there was conﬂict in the parental relationship at the time of the
divorce, the initial pattern of conﬂict did not seem to dissuade fathers from
maintaining contact with children. Mothers and fathers held different perceptions
about how the quality of their relationship affected father’s involvement, but

basically contact between children and fathers stabilized following divorce. The

73

age and gender of the child, remarriage, father’s education and geographic
proximity did not affect level of paternal involvement. The existence of such a
high level of paternal involvement has not been replicated in other studies, and
the criterion used to select the sample appeared to be biased toward couples
who had some basic agreement about custody and parenting time from the
onset. The study did not specify what differentiated high conﬂict parents from low
conﬂict parents in the communication instrument used.

In a 1994 study, Amato and Rezac, looked at child contact with the non-
resident parent as a factor in child adjustment. Using a sub-sample of the
National Survey of Households, they looked at child behavior problems in two
groups of children, middle childhood and adolescence. Their hypothesis was
that child behavior problems post-divorce were the result of an interaction
between the non-resident’s parent-child level of contact and the amount of
conﬂict between the divorced parents. Their hypothesis was supported for male
children but not female children. One explanation for the gender difference in this
study was that information was gathered only from the custodial parent who may
have had more difﬁculty with male children. Additionally the deﬁnition of
behavioral problems was weighted heavily on cognitive and social adjustment in
the school environment and on externalizing behaviors, which is more descriptive
of male socialization difﬁculty than it is of female difﬁculty.

When parental conﬂict was low, boys who had contact with the
nonresident parent had fewer behavior problems, but when parental conﬂict was

high, regular contact with the nonresident parent resulted in more behavioral

74

problems. Findings from this study indicate that frequent contact under
conditions of high conﬂict increases behavioral problems for some children.
Other ﬁndings from Amato and Rezac (1994) infer that higher SES was a factor
in the maintenance of parent-child contact post divorce. Contact frequency with
the non-resident parent was positively associated with parental education, age,
and household income. However, as other studies have indicted, visitation rates
generally drop to a lower level overtime.

Healy, Malley 8 Stewart (1990) followed the divorce adjustment of 121
young children (age 5-12 yr.) in mother custody homes. This study utilized a
non-clinical sample with a range of custodial parent SES that was recruited from
court records. They speciﬁcally focused on child adjustment by age and gender,
and ﬁrst took measurements during the marital separation, then repeated
measures about 12 months later when the divorce was ﬁnal. Information was
primarily gathered from the mothers, and children, regarding the frequency and
duration of parenting time and closeness within the father-child relationship.
Additionally, legal conﬂict was considered both a predictor and moderator
variable. Legal conﬂict was deﬁned as dispute over major aspects of the divorce,
such as custody/parenting time as evidenced by number of petitions ﬁled in
court.

The Healy et al., (1990) study found support for an association between
legal conﬂict and behavior problems in children. However, their ﬁndings were
overly general, and do not clarify how many ﬁlings of petitions during the process

of divorce would be standard and ordinary, and or what number of legal

75

complaints separates high conﬂict from low conﬂict in the spousal relationship. It
did not specify how the legal dispute affected the interaction between the parents
or if it resulted in the denial of access to the children. Results from the Healy et
al. (1990) study give an indication that child high self-esteem was more likely to
occur when legal conﬂict was low, and that child low self-esteem was more likely
to occur under conditions of high conﬂict. Regular and frequent contact was
associated with in the fewest behavior problems. More behavior problems were
apparent when parents were in legal conﬂict and when the father contact was not
frequent or regular.

One weakness of the study was that the time I and time ll comparisons
both occurred early in the divorce process when family members’ individual
adjustment was not likely to have stabilized. As a result of the time of the post
measure, it is difﬁcult to know whether the children’s behavioral difﬁculties were a
response to the marital rupture, or an indication of divorce adjustment. Other
studies suggest that two years post-divorce was a more predicable time to
measure behavioral differences in children.

Amato, Loomis and Booth (1995) utilized a 12-year longitudinal study to
sort out the impact of parental marital conﬂict and divorce on offspring well being.
Adult child well being was measured by presence of psychological distress and
satisfaction with family relationships. This study was begun in 1980 and the
median age at follow-up was 23.5 years. In this telephone study, the sample size
of children who experienced divorce was quite small (42) and only contained

children who were age 10 or above. Their ﬁndings suggested those children in

76

high conﬂict families whose parents divorce do better than children in high
conﬂict families when the parents do not divorce. However, children from low
conﬂict families whose parents divorce do worse than do children from low
conﬂict families who remain together. One explanation was that in a majority of
low conﬂict divorces disengagement, boredom, and dissatisfaction typiﬁed the
spousal relationship and as a result the child was surprised by the divorce. With
these low conﬂict stable households the child beneﬁted from access to both
parents. The ﬁnding in this study was important beCause it lent support to the
notion that exposure to intense, chronic and unresolved parental conﬂict is more
detrimental to child well being whether in an intact or divorced family. Additionally
despite losses associated with divorce, some children do better when they are
removed from high conﬂict. This study also gave credence to the idea that some
of the long-term negative consequence of divorce was associated with level of
conﬂict prior to the separation.

Amato’s et al. (1995) ﬁndings have been replicated by Jekielek (1997) and
Hanson (1999). Using the National Longitudinal Survey of Youth, Jekielek was
interested in examining if children were better off if they remained in a high
conﬂict intact family or if their parents divorced. Jekielek followed 1,640 children
6-14 years in 1992, of whom 241 experienced parental divorce or separation
since 1988. She was particularly interested in measuring child-well being using
internalizing behaviors, anxiety and depression/withdrawal rather than overt
behavioral problems. The ﬁndings indicate that children who are being raised in

an intact family have lower levels of anxiety and depression than do children who

77

experience divorce. However, for children whose prior home environment
included a high level of conﬂict, the condition of divorce lessened a child’s
internalizing behaviors especially for those who were 2 or more years post-
divorce. The Jekielek study found no gender difference in children coping with
high conﬂict.

Hanson (1999) used data from the National Survey of Families and
Households to examine whether parental conﬂict prior to divorce could explain
why children with divorced parents exhibit more academic and adjustment
difﬁculties than children with parents who stay together. Hanson was particularly
interested in sorting out the effects of exposure to parents in high conﬂict from
the effects of marital separation, on children’s outcome. Child well being was
measured by (1) school performance and behavior, (2) delinquency, (3) health
and health behavior, and (4) psychological well being. Parental conﬂict was
measured by responses from both wives and husbands, frequency of
disagreement and method of resolving disagreements, including physical
altercations. Interestingly, the Hanson study notes that high conﬂict does not
necessarily predict divorce. At the ﬁve year follow-up about half (48.1%) of
parents who divorced came from the high tier of conﬂict 4-6 years earlier, but
about 78.3% of the couples identiﬁed as having the highest level of conﬂict were
still married. Clearly, in this study the report of high conﬂict does not necessary
lead to divorce. Hanson (1999) did ﬁnd a slight rise in level of conﬂict as couples
move toward divorce, but he did not ﬁnd a direct association between the divorce

process and conﬂict for the majority of couples who divorce.

78

Similar to Amato and Jekielek, Hanson, (1999) concluded that conﬂict and
divorce have independent effects on children’s well-being. Children exposed to
long term high conﬂict appeared to show some improvement in overall well-being
when parents divorced, whereas children from low-conﬂict families had a decline
in overall adjustment when parents divorce. The results in the Hanson (1999)
study were helpful in identifying and measuring parental conﬂict 4—6 years prior to
marital separation and In offering an explanation of how conﬂict and divorce
interact but do not necessarily effect child outcome in the same way. Exposure
to high conﬂict in Hanson’s view accounts for about 11% of the effects of divorce
on child well being. The study consisted of a large sample (1,907) and included
approximately 293 focal children or 15% who experienced divorce between the
Time I and Time II waves. However, what was missing was a demographic
description of the sample that indicates to whom the results could apply.

Although the above research tells us about family conﬂict prior to divorce,
it does not tell us about the nature of parental conﬂict following divorce, or how
issues between parents change with marital status. In the parental survey, child
well-being encompassed a variety of areas in the questionnaire but the study
does not give us information about current quality of parent-child relationships.
For example, it was reported that children did better when high conﬂict parents
divorced, but this does not tell us anything about causation. Children exposed to
high conﬂict might do better because one parent became disengaged and
dropped out of their life, or they might be doing better because they reached a

different developmental stage and have more resources to deal with parental

79

difference.

The Virginia Longitudinal Study of Divorce and Remarriage (Hetherington,
1989) also focused on the adolescent adjustment. This research compared the
adjustment of parents and children under conditions of non-divorce, divorce, and
remarriage. Multifactors that were thought to strongly inﬂuence divorce outcome
were included: income, single parenting stress, child gender, and level of conﬂict
in the parental relationship. The year following divorce for both male and female
children showed more anxious, demanding, noncompliance, and aggressive and
dependent behavior with peers and adults. When these preschool children were
studied at adolescence, there was no clear gender difference in child adjustment.
Early adolescence was a developmental period in which all three family types
experienced an increase in child behavior problems, but the increase was more
prevalent for children with remarried and divorced parents. Mothers, fathers and
stepparents rated adolescent children on the Child Behavior checklist. According
to maternal reports, 10% of adolescents from intact families, 28% of males and
16% of females in remarried families, and 26% of males and 34% of females in
divorced families scored in the clinical range. Paternal reports did not result in
gender differences, however, 7% of adolescents in intact families, 37% in
remarried families and 30% in divorced families were described as meeting

clinical criteria.

Post-Divorce Father-Child Relationships.
Neither the quality nor the frequency of contact between the nonresident

father and the child could be predicted from the pie-divorce parent-child

80

relationship. Over time non-resident fathers tended to become more permissive
and disengaged following divorce. Some previously active fathers became less
involved and some inactive fathers became more involved. This kind of pattern
variation was previously documented in the Wallerstein and Kelly (1980) ﬁndings.
Similar to other studies, this Virginia study noticed a significant drop in parental
involvement, less than 25% had weekly contact, at the two-year post-divorce
follow-up. After 11 years, 50% of fathers had no direct contact with their
adolescent children in the preceding year.

The ﬁndings from this study found inconsistent results regarding child
adjustment associated with nonresident parent contact. This study found no
main effect for frequency of visitation. Contact with the nonresident parent was
modiﬁed by the quality of the relationship between the divorced parents and by
the attributes and behavior of the non-custodial parent. When conﬂict was high
and the visiting father was incompetent or antisocial, children were more likely to
have an increase in behavior problems, a decrease in academic achievement,
and lower self—esteem. Low conﬂict and 000peration between the parents and
frequent visitation resulted in fewer behavior problems, higher academic
achievement, and higher self—esteem, especially for male children. Only about
20% of divorced parents were considered to be cooperative. It was more
frequent for parents to parent independently of the other, and to have low levels
of communication.

In contrast to the parents who are very disinterested in contact there are

some couples who continue to be engaged in high conﬂict years following

81

divorce. Johnson and Campbell’s (1993) work regarding divorce and violence
utilized two separate samples of post-divorce couples to clinically describe the
typologies of physical altercation that are associated with violence that
accompanies relationship breakup. The ﬁrst sample consisted of couples
entrenched in parenting time and custody disputes two years after
divorce/breakup, 30% of whom were involved in post-decree litigation. This
sample primarily encapsulated families of low-middle income, with 36% of the
participants included in racial groups other than Caucasian.

The second sample also primarily involved divorced couples, with 20%
minority representation and of diverse socioeconomic status. In the second
sample the rate of post-divorce litigation was 50%. The family court for
counseling and mediation referred both groups, with each family receiving up to
20-30 hours of clinical intervention per family. Johnson and Campbell (1993)
clearly stated that their conclusions were based on clinical inference and not on
statistiwl inference. They used documentation in case notes regarding the ﬁrst
sample and the administration of the Conﬂict Tactic Scales in the second sample
by two clinicians to classify the primary aggressor into four major types of violent
behavior. The theoretical classiﬁcation included information about whether the
violent behavior accompanied separation trauma or not. The four typologies
were 1) ongoing or episodic male battering; 2) female initiated violence; 3) male-
controlled interactive violence; and 4) separation and post-divorce violence.
Separation and post-divorce violence involved behaviors that were not evidenced

during the marriage but that appear triggered by emotional trauma connected to

82

humiliation, helplessness and abandonment.

This framework for looking at high conﬂict families by Johnson and
Campbell is unique because it comes close to describing some behavioral
differences in the divorced population who appear to have underlying
psychological adjustment difﬁculty at the intra-psychic, interactional and external
factors, that gets expressed in the ongoing relationship with the other parent.
Because of the limitations in the study, the authors were not able to state to what
degree these typologies were representative of the physical aggression that
occurs before and after highly stressed relationships terminate.

In summary, Chapter II has covered three areas: attachment theory,
measurement of attachment, and child adjustment to parental divorce. The
literature review began with a description of the major tenets and assumptions of
attachment theory by highlighting the work of Bowlby, Ainsworth and Main. A
discussion of an infant’s early attachment relationship to the primary caregiver
served as the basis for the explanation of differences between securely and
insecurely attached children. Internal models of representation were highlighted
as the mechanism through which attachment relationships get transmitted from
parent to child and through which an individual’s style of relating gets repeated in
other signiﬁcant relationships.

The second area of the literature review addressed some of the self-
report instruments used to measure adult attachment. Five researchers’ ideas
on adult attachment measurement were compared. During the process of

comparison both conceptual differences and similarities as to what constitutes a

83

secure or an insecure attachment style were discussed. Included in this chapter
was an examination of the AAI, a qualitative instrument. The AAI has been
widely used in infant studies to explain the parental role in determining security of
attachment, and it has been more recently used in adult retrospective studies to
explain the antecedents of personality disorder.

The third portion of this chapter has focused on divorce research with
special emphasis on longitudinal studies. The chosen studies looked at
children’s adjustment to divorce under conditions of both low and high conﬂict.
Children are seen to do better when parental divorce results in a resolution of
conﬂict, and children fare worse when such divorce results in the continuation of
parental hostility in which the child is in the middle. The co-parental relationship
was, therefore, recognized as a critical variable in a child’s divorce adjustment.

The review of the literature was an attempt to link attachment theory with
the divorce literature. This comparison study of two groups of adults who divorce
it is an attempt to link an adult’s style of attachment to their current relationship
with their former marital partner and the emotional reaction to the loss of that
marital partner. To address differences in security of attachment, a normal
sample is compared with a clinical sample. The question to be answered in this
study is this: there are group differences between a normal sample and a clinical
sample of adults who divorce regarding attachment style, level of conﬂict, and the

non-resident-parents’ access to minor children.

84

Chapter 3

METHODOLOGY

Overview

This study utilized a combination of quantitative and qualitative methods
for the purposes of data collection. Measurement of group differences was
obtained primarily through self-report instruments. Some of the areas of inquiry
within the quantitative survey were based upon other researchers’ approach to
the measurement of adult attachment as well as other researchers use of
variables borrowed from the divorce literature. The chosen concepts were
extrapolated from the literature to support the explanation of clinical phenomena
between groups of individuals who divorce.

A survey design is commonly used for social science research. It is
typically used to study demographics and/or the characteristics of a large group
of people (Cherry, 2000). Self-report surveys are particularly useful for
exploratory research that is seeking to determine if there is a relationship
between two or more variables. The researcher collects the data from a group of
people in order to describe some aspect or characteristic of the population of
which that group is a part (Fraenkel 8 Wallen, 1996). The self-report instrument,
by asking the same question from every respondent, has the advantage of
indicating numerical differences between groups of persons. Numerical
differences help determine whether each hypothesis under consideration is

statistically supported. However, one limitation of quantitative research is that

85

 

 

(n

results do not describe individual differences or contextual variation that is unique
to each individual or family.

In contrast, according to Cherry (2000), qualitative methods are used to
gain knowledge regarding the meaning of behavior, understanding social process
or human interaction, and identiﬁcation of patterns or themes over time.
Qualitative research is a process where systematic analysis of human behavior is
used to identify underlying meanings and patterns of relationships. Grounded
theory is one of four main approaches used in qualitative research (Cherry,
2000).

Grounded theory was originally developed by sociologists Glaser and
Strauss (Strauss 8 Corbin, 1990). The purpose of grounded theory is to
systematically observe a behavior or social process and identify trends and
patterns that suggest a tentative theory. Because the tentative theory is derived
from the data, the theory is said to be grounded in the data (Cherry, 2000).
Analysis of the data continues until there are no new insights about the
behavioral patterns. Grounded theory meets the criteria for “good" science
(Corbin 8 Strauss, 1990) if rigorous procedures are followed. The analytic
process includes: 1) data collection; 2) categorization; 3) memoing (the ideas that
occur to the researcher); 4) movement toward parsimony; and 5) writing a theory.

The qualitative interview looks to the participants as primary sources to
help explain how he or she makes sense out of their lives, and their relationship
to others. According to Corbin and Strauss (1990), a grounded theory approach

requires a strong reliance on the review of the literature, and the use of concepts.

86

The concepts represent statements of relationship, which are organized
according to themes. Selections of cases are chosen for their degree of
similarity or difference from other previous cases. With grounded theory
research the purpose is to discover relevant categories and relationships among
variables rather than testing relationships among variables (Straus 8 Corbin,
1990). The data analysis is used to explain or expand a theoretical model in a
new or creative way. The data analysis can used as supplementary validation of
other ﬁndings pointing out similarities and differences. The participants are
chosen because they offer a unique contribution to the development of theory.
Each case is compared to the next, allowing the researcher to creatively decide
why case to case differences occur. This process is achieved through inductive
reasoning where one uses his or her observations to propose generalizations to
the whole.

With qualitative interviewing, assumptions, motives, reasons, goals, and
values are likely to be the focus of the researchers questions (Fraenkel 8
Wallen, 1996). Structured questions are used to obtain information that can be
compared and contrasted, whereas prompts are helpful in understanding the
thinking behind the response. When constmcting the questions for the face to
face interviews in this study, the researcher was aware of the importance of the
narrative stories. To capture some aspects of the individual’s life story, questions
regarding the context of the early family environment and the marital relationship
were designed to be shared in narrative form.

When more than one method of data collection is used, there is a beneﬁt

87

when ﬁndings point to similar interactions between variables. Under this
circumstance, the research results have more credibility and utility. In other
words, combining quantitative and qualitative methods allows for the
simultaneous data collection of information about the subject of the study and the
context in which they are embedded (Reinhartz, 1992). The use of two or more
methods of data collection to measure similar variables is referred to as
triangulation (Cherry, 2000; Reinhartz, 1992). For this reason, both qualitative
and quantitative methods were employed in this study.

A small sub-sample of respondents who completed the divorce
questionnaire from each of the two groups, “normal” and “clinical" divorcing
persons, was recniited to complete a qualitative interview that was based, in part,
upon the Adult Attachment Interview. The instrument the researcher developed
was called the Family Attachment Interview. It was designed to enrich and
enliven the results of the self-report instruments through description of
relationships to parents, spouses and children. One aspect of the qualitative
interview was to determine whether there were similarities or differences in the
kinds of relationships these adults had with their own primary caregivers. It was
also a means of testing whether an insecure or secure attachment history was
carried forward into spousal selection and into the relationship with their children.
In reporting data from these face to face interviews, themes and concepts
derived from the analysis are illustrated in summary form as well as by
quotations and paraphrases.

Attachment theory was the foundation from which the hypotheses in this

88

study were developed. Based upon the researchers clinical background and
review of the literature, three aspects of the divorce process were selected:
relationship style, level of conﬂict, and parent-child contact. The process of
conceptualization began by studying each of the above variables in reverse order
by moving backward through the divorce process. Inspection of the factors in the
divorce literature that inﬂuenced whether or not a nonresident parent stayed
involved with children led to a strong association between the quality of the
coparental relationship and amount and frequency of contact. Studying the level
of conﬂict in the co-parental relationship yielded information about the possible
role that conﬂict played in the co-parental relationship as a means of maintaining
an ambivalent attachment to a former spouse. Finally, the degree of felt security
with the spouse was associated with other signiﬁcant attachment experiences
during childhood. Thus, qualitative interviews with the sub-sample were an
attempt to describe in more detail how a secure or insecure attachment history in
childhood affects divorce adjustment as evidenced by the quality of co-parental
relationship, post-separation, and the parent’s relationship with minor children.

As previously stated, to compare group differences between normal and
clinical couples who divorce, this study examined three clusters of variables:
style of attachment, level of conﬂict, and degree of non-resident parent-child
contact. The normal sample was recruited from a countywide two-hour divorce
orientation program and the clinical sample was recruited from a local community
mental health specialized treatment program for clients who are mandated for

care. As stated in Chapter I, the normal sample represents adults who

89

experienced divorce with a fairly positive interpersonal history as indicated by
having no more than one problem indicator in three major domains of functioning.
The assignment of clinical status referred to adults who were court mandated into
mental health services and who experienced divorce with a poor interpersonal
history in two or more problem indicators in three domains of functioning. For the
purpose of this study, when a parent was engaged in domestic violence, had
prior involvement with protective service, had a history of substance abuse
dependence, or had an active personal protection order then both spouses were
considered to be part of the clinical sample. Further explanation about the
sorting into normal and clinical sample groups will be addressed later in the

procedure section of this chapter.

Research Context

This study took place in a large, mostly urban county in lower
southeastern Michigan. Each year, approximately 1300 couples with minor
children ﬁle for divorce in this county. The recruitment for the two nonrandom
samples was drawn from two pools of divorcing and divorced couples who have
minor children. One pool consisted of a countywide divorce orientation program.
On any given month, between 50-80 individuals attend this evening divorce
orientation program. Each parent receives a letter from a family court judge
mandating him or her to attend this meeting approximately 60 days after the
petition for divorce has been ﬁled. The letter essentially serves as an invitation to
participate, as there are no sanctions for not attending (see Appendix C). Some

divorcing couples who attend this brief educational program have already been

90

separated for several months, whereas some are still residing in the marital
household.

The divorce orientation program consists of three different presenters and
a short video. First, a family court judge states his or her philosophy of the
divorce process from the bench. Second, the county Friend of the Court
provides information about how to access services within that agency. Third, a
social worker from the local community mental health (CMH) agency talks about
family members’ emotional reaction to divorce with emphasis on the
developmental needs of children. The social worker encourages parents to be
reasonable and cooperative on behalf of their children and highlights the skills
necessary to be successful co-parents. A short video entitled “Listen to the
Children” highlights the thoughts and feelings of a group of children who have
experienced marital rupture. In the ﬁlm the children discuss their parent’s
behavior during the process of divorce as well as their respective feelings. The
program includes a written curriculum regarding divorce for parents to keep with
references to local community resources and services. At the end of the evening
participants are encouraged to direct questions to any of the three presenters.

The second recruitment pool consisted of families who had been court
ordered to participate in a specialized clinical intervention program associated
with the local community mental health. Parents are typically referred for
unresolved custody and parenting time conflicts by way of a Family Counseling
Order. Referrals include couples in process of divorce as well as couples who

have obtained a divorce, but have post-divorce litigation. Often, disputing

91

families come to the attention of the court because of serious allegations of
substance abuse, domestic violence, child endangerment due to neglect or
abuse, or lack of parental access to a child. Intervention services include an
assessment and evaluation, mediation, or family counseling (See Appendix D).
Unlike the mandated participation to the Divorce Orientation program, the family
counseling order has potential repercussions for nonparticipation. Refusal to
participate could result in the postponement of a hearing, the dismissal of a
petition, the judge refusing to make alterations in the custody or parenting time,
or a contempt of court citation.

Most frequently, parents have opposing perceptions about the origin of
difﬁculties and have opposing goals for service delivery. The parents referred to
the program tend to be in high conﬂict, and have rather ingrained patterns of
relating to the other parent. Frequently one parent views the goal of service as
changing some aspect of the other parent rather then any condition regarding
them. It is not unusual for one or both spouses to openly exert pressure on the
therapist to align with his or her view of the desired outcome of service.

In the clinical intervention program, a medical record is opened for each
parent and each parent meets alone with the therapist for the initial assessment.
Length of service varies, but typically, lasts 8-10 sessions. Children are seen
with each parent for at least one session. Additionally each parent is asked to
invite other important collateral contacts, eg. new spouses, live-together
partners, or grandparents to the treatment process. To establish new ground

rules for behavior, facilitate communication, and address allegations, it is often

92

necessary to hold conjoint sessions with the parents. Depending on the court
order, one session is used for administering the MMPI 2 (Minnesota Multiphasic
Personality Inventory). Other sessions are scheduled as determined by the
therapist or by consumer request. Following completion of service or at a legal
impasse, the treating professional is expected to submit a lengthy written report
to the court with his or her ﬁndings and recommendations.

This unique specialized treatment program was begun as a Michigan
mental health prevention project in 1983. Following the completion of the
research grant, the project resulted in a collaborative delivery of services
between the local community mental health and the family court. Funding for the
program comes from the CMH budget, a contract with the court, third party
insurance, and family’s ability to pay. The mental health professionals providing
the interventions are masters degree trained social workers, with 18, 15, 10, and
8 years of experience, respectively, working with families in conﬂict regarding the
care and custody of minor children. The professional staff has a strong
background in the diagnosis and treatment of both adults and children. They
have received additional credentials and training in family and divorce mediation.
As a result of working exclusively with a population of families in conﬂict, the
therapists have become quite proﬁcient in a variety of skills, which include
mediation, clinical treatment, collaboration with legal professionals, and

innovative ways to support parent-child relationships.

Sampling

From survey information collected prior to data collection for this study,

93

demographic information on the couples participating in the divorce orientation
program for the months of June, July, and August, 1999, indicated that attendees
consisted of 42% males and 58% females. The number of children of the
marriage was as follows: 26% had one child, 51% had two children, 16% had
three children and remaining 7% had four or more children. Number of years
married ranged from less than 1 yr. to 30 years, with a mean of 10 years. A
majority was early in the divorce process with nearly all petitions for divorce
having been ﬁled in the last 6 months. The annual income ranged from less that
$5,000 per year, to over $55,000 with fairly equal distribution for each $10,000
category. However, overall, the mean and median income were quite modest,
falling between $25,000-$35,000 annual income. When income is compared
with US Census (1995), for the county, it is somewhat less than the $38,905
median household income estimate. This lower reporting of income makes
sense as when couples separate they no longer report joint household income.
The preliminary surveys collected on the divorce orientation attendees
were designed to measure client satisfaction and did not include demographics
regarding race. In a pre-instrument trial in September and October of 1999,
information on race was collected. This two month sampling indicated 53.3%
were European-American, 6.7% African-American, 2.2% Hispanic, 15.6% listed
other and 22% failed to specify. The 1990 US Bureau of Census indicates
county distribution by race as: 78.2% White, 19.5% Black, 1.9 % Hispanic origin,
.07% American Indian, .07% Asian or Paciﬁc Islander, and .09% as other. It is

not known whether the racial composition for the county has shifted signiﬁcantly

within the last ten years; however, this is the most recent comparison available.
The primary clinical sample was known to consist of a high proportion of
those with an Axis II diagnosis of personality disorder, and an undiagnosed
history of mental illness, substance abuse, and long term interpersonal
relationship difﬁculty. In a review of case records from the previous year, it was
noted that there were usually three or more areas on the DSM IV Axis IV, e.g.,
ﬁnancial problems, underemployment, or health problems as well as contested
litigation regarding child custody or parenting time that indicated difﬁculty with
overall interpersonal adjustment. Because the clinical sample included parents
with post-divorce litigation, it was expected that this group would be different from
the normal sample in regard to average time since the marital rupture. The
majority of families continue to reside in the county where the divorce case was
ﬁled. However, residency was not a factor that was considered in this study.
Based on annual demographic information prepared for the 1998-99 CMH
annual report, the clinical sample was expected to consist of almost equal
numbers of men and women (56% female, 44% male) with an approximate age
of 35 years with a range of 20—50 year. Racial composition for clients served for
ﬁscal year 1998-99 consisted of 86.2 % White; 10.7% Black; 1.2% Hispanic;
.004% American Indian, and 2% other. Tracking racial composition of individuals
for nine months of ﬁscal year 1999-2000 Indicates a similar pattern of 84.4%
White, 11.9 Black, 0% Hispanic, 0% Native American and .005 % other. The US
Bureau of Census (1990) describes the county household income as ranging

from less than $5,000 to over $100,000. There was no reliable data available

95

prior to the study that illustrated range of income for persons who received
clinical services, but from a cursory review of ﬁnancial information from active
cases, it was predicted that income would be quite modest and close to that of
the normal sample.

Recruitment for the normal respondents was collected in the following
manner. During ﬁve monthly divorce orientation meetings individuals in
attendance were asked to participate in the study. Data was used only from
those individuals who volunteered to participate. There was no requirement that
only one or both spouses take part in the study. Those who refused consent and
those with extensive missing data were removed from the potential sample.
Approximately 104 surveys were returned, granting consent to participate in the
study. However, 19 surveys were removed from the pool of respondents
because did not meet the criteria for number of years married in this study. Next,
the predetermined clinical criterion was applied to each batch of the normal
sample, creating two sub-groups, “normal normals” and “clinical normals”.
Because 21 “clinical normals” met criteria for the clinical sample they were added
to the clinical group which left a pool of 66 respondents. Lastly, birthdays and
family descriptions were compared in each sample to insure that each sample
consisted of an unduplicated count. There were three individuals that agreed to
participate at the divorce orientation program that later appeared for service at
CMH. Because of this duplication the ﬁnal number for the normal sample was 63
respondents.

After ﬁnal sorting the clinical sample was comprised of 18 respondents

from the divorce orientation program that met clinical criteria and 44 respondents
that were mandated into care by the court. To determine if these two sub-
samples were different according to demographics the Mann Whitney U test and
the Chi—square analysis was done. The combined results of this cross-tabulation
revealed no signiﬁcant difference regarding age, gender, race, number of years
married, age relationship began with the spouse, who ﬁled for divorce or
employment status. There was one area of signiﬁcance in that the normal
clinicals had a higher number of children (m=2.56) when compared to the
mandated clinical group (mean=1.59) t=2.34, g<.05, two tailed. Another minor
difference involved whether or not the divorce was ﬁnal and whether or not child
support was ordered. This within sample difference was anticipated because the
divorce orientation sample primarily consisted of couples recently separated
whereas more of the couples mandated into care had divorce decrees. By virtue
of meeting clinical criteria and by having very similar demographics the two sub-
samples that comprised the clinical sample were not signiﬁcantly different, except
for number of children.

Recruitment for the clinical sample occurred over a period of ﬁve months.
As individuals presented for their intake appointment, they were asked to
participate in a study by a secretary. Data was used only from those individuals
who agreed to participate. There was a potential for a higher number of couples
in the clinical as each was mandated to participate, however in this study there
was no attempt to match the responses of spouses. Of approximately 66

POtential respondents 10 refused to participate. Another 10 respondents were

97

removed from the sample because they had never been married to the other
parent, and 2 surveys were not used because of extensive missing information.
The 18 subjects from the normal clinicals were added to the 44 respondents for a
total of 62.

Sampling for the face-to-face interview occurred by reviewing completed
questionnaires from each respective sample and determining if they expressed
interest in being contacted. Of 104 normal respondents 26 indicated willingness
to be contacted and of the 44 clinical respondents 26 indicated willingness to be
contacted for an interview. Prior to the contact, the researcher scanned each
survey to insure that the respondent met criteria of the study. There was no
tracking of failed attempts to make contact, although participants were most
responsive to agreeing when they were at home to take a call. Once a subject
agreed, there was an attempt to continue with that person and reschedule
appointments if necessary. However, one subject from each sample did not

come in to be interviewed despite outreach efforts.

Exclusionary Criteria

Couples, mandated to the divorce orientation program and mandated to
the mental health intervention service, were not all married. In view of possible
attachment differences between couples who marry and couples who do not
marry, it was decided to study only those couples who had married and who had
at least one minor child in common. The study also excluded couples who were
married one year or less and those who were married for 20 years or more. The

lationale for eliminating newly formed couples was that these individuals may not

98

emotionally react to the loss of a marital partner with similar dynamics when
compared to couples who have struggled together for a number of years.
Conversely, those who have been together 20 or more years may have a strong
emotional investment in the marital relationship, and therefore have more
difﬁculty mastering the tasks of loss and grief associated with divorce. Twenty
years was selected as the division point to include more families with minor
children, and to focus on couples parenting young and latency age children.
Divorcing couples with adolescent or adult children also have additional custody
issues to consider, as older adolescents are more likely to exert inﬂuence

regarding their custody preference.

Instruments Used for All Participants

The Divorce Questionnaire (see Appendix E) is a self-report survey that
was developed by the researcher. In its original form, it was designed as a pre-
service instrument as a means of providing information to therapists who were
intervening with outpatient clinical families. It was adapted and reﬁned for the
purposes of this study. The divorce questionnaire was designed to collect six
clusters of descriptive data: demographic background, current household
members, relationship dynamics within the separated or divorced family,
attachment behaviors, level of conﬂict, and adult interpersonal functioning.
Embedded within the questionnaire are modiﬁcations of two attachment surveys
developed by other researchers (Simpson, 1990; Bartholomew 8 Horowitz,
1992), the Conﬂict Tactic Scale (Strauss, 1979), and a measurement of

Interpersonal functioning borrowed from the DSM IV (American Psychiatric

99

Association, 1994), Axis IV scales. Additional information regarding the selection
and utility of these instruments will be addressed later in this chapter. All
respondents who participated in the divorce orientation program and who were
referred for clinical intervention were asked to complete the same questionnaire.
Basic demographic information included the number of children of the marriage,
which spouse petitioned for divorce, number of years married, age relationship
began, and perception of each parents’ shared responsibility for child care in the
year prior to the marital rupture. Other demographic information was collected
speciﬁc to each parent e.g., zip code, birth date, gender, race, employment,
income, household composition, amount of child support, type of
custody/parenting time arrangement.

The current household constellation was gathered through a grid asking
respondents to list themselves, household members and minor children affected
by the pending court action. Across each grid are labels to identify, ﬁrst name,
relationship, date of birth, gender, education, and custody. Asking for a
description of the current household was an attempt to allow for a creative way to
recognize the conﬂuence of a variety of relationships impacted by divorce which
might include grandparents, live-together-partners, step-children, children from
prior relationships, or friends.

To discern the relationship dynamics in the co-parental relationship and in
the parent-child relationship, information was collected regarding the kind of
custody arrangement, the frequency and regularity of parenting time, and

communication patterns between the parents. Additionally, each parent was

100

asked to rate the relationship between themselves and each child, and to also
rate the other parent’s relationship to each child. A deﬁnition of close, friendly,
neutral and distant was provided as a key for answering this question. Another
area that used a grid-like response was having each parent rate how childcare
tasks were shared between parents when the family resided together. Father
and mother participation was divided into the following categories: physical care,
supervision of children, discipline, medical, education, leisurelplay/recreation, on
a graduated scale from 100% to 0%. This area of inquiry was added to see if
there was any association between parent participation with children care and
the kind of custody arrangement that was established. However this was not

considered a major question for this study.

Overall Measurement of Attachment

After reviewing the attachment literature and understanding that different
adult attachment instruments measure different aspects of adult relationships,
two instruments were selected for modiﬁcation. The ﬁrst was based upon
Simpson’s (1990) self-report instrument that was developed to measure overall
attachment patterns: secure, avoidant or anxious/ambivalent. Simpson’s 13
original questions were used. However the Likert scale format was decreased
from seven to ﬁve choices, strongly agree, agree, not sure, disagree and strongly
disagree. This instrument focuses on measurement of comfortableness with
closeness and dependence as well as worry about intimacy and abandonment.
Simpson (1990) revised Hazan 8 Shaver (1987) attachment vignettes into a

scale to measure overall attachment style. Simpson’s measure was used to

101

assess adult attachment style not speciﬁc to the marital partner, but rather
attachment in general.

A secure Attachment Style Index was found by aggregating items a
through e. Higher scores reﬂect greater security. The Avoidant Attachment Style
Index was developed by aggregating items f through I. Higher scores indicate
greater avoidance. The Anxious Attachment Style Index was created by
aggregating items j through in. Higher scores reﬂect greater anxiousness.
Despite the low alphas on two of the three scales, this measure was selected for
two reasons. One, there is an absence of a more reliable instrument that
captures the complexity of adult attachment, and two Simpson’s concepts were
most adaptable to a self-report instrument.

As part of this study instrument reliability was performed on each
attachment measure. The reliability of the sub-scales of secure, avoidant and
anxious/avoidant on the ﬁrst instrument were as follows: 1) There was limited
reliability revealed on the Chronbach’s alpha = .58 for the sub-scale of secure.
By dropping one item (I rarely worry about being abandoned by others) the alpha
increased to an acceptable level, alpha = .60. 2) The sub-scale avoidant had an
acceptable level of reliability, alpha = .83. 3) The sub-score for anxious/avoidant
was found to have limited reliability, Cronbach’s alpha = .27. By removing one
item (I rarely worry about my partner(s) leaving me) the reliability was increased

to alpha = .49.

Attachment to Spouse

The second attachment instrument, based on Bartholomew and

102

Horowitz’s four-category model, looked at dimensions of dependency and of
closeness in the self and in the other was modiﬁed. It was modiﬁed into a forced
choice, true-false option, to measure coping behavior regarding loss of marital
partner. It has four sub-scales that indicate secure, preoccupied, dismissing and
fearful styles of attachment. It was selected because of its theoretical
compatibility with internal models of representation, and because it recognized
attachment relationships as having both components of the self and of the other.
The reliability was calculated for each of the four sub-scales on the
instrument with the following results: 1) The sub-scale for secure did not reach an
acceptable level of reliability. After dropping one item from the sub-scale the
alpha was still inadequate; 2) The sub-scale for dismissing did not meet an
acceptable level of reliability, however by removing one item (Being alone hasn’t
been that bad, I am surprised how easily I have adjusted), reliability reached an
acceptable level, alpha = .63; 3) The sub-scale for preoccupied did not meet an
acceptable level of reliability. By removing one item (I ﬁnd myself driving by my
former partner's residence or calling and just hanging up) reliability reached an
acceptable level, alpha = .68; and 4) The sub-scale fearful met criteria for

reliability, alpha =. 73.

Conﬂict Tactics Scales

To establish the overall variable of level of conﬂict, several items within the
Divorce Questionnaire were combined, including the sub-scores for the Conﬂict
Tactics Scale (Straus, 1978; 1990a). More speciﬁcally, level of conﬂict was

determined by the scoring of eight questions: degree of difﬁculty resolving

103

custody, frequency of court appearances, dissatisfaction with custody
arrangement, communication pattern, communication frequency, decision
making style, and a current rating of verbal conﬂict. The original Conﬂict Tactics
Scales (Straus 1978, 1990a) was used to measure whether partners in a dating,
cohabiting or marital relationship engage in psychological and/or physical attacks
on each other during the past year. It measures the respondent and the
partner’s behavior in regard to three scales: reasoning, verbal aggression, and
physical assault. The CTS has been broadly usedto discern physical violence
between couples and has evolved into revised CTSZ and a parent-child version,
(PCCTS). The theoretical basis for the CTS is conﬂict theory. The CTS was a
widely used instrument with strong evidence of validity and reliability (Barling et
al., 1987; Straus, 1990a). The coefﬁcient of reliability for the Conﬂict Tactics
Scales, form n (n=2143) for couples score, was .76 for sub-scale reasoning, .88
for sub-scale verbal aggression, and .88 for sub-scale violence (Straus & Gelles,
1990).

However, the CTS has been criticized for not accounting for the context in
which the conﬂict occurs and for a lack of construct validity in one of its sub-
scales. It was originally intended to be used in conjunction with other measures
and during a face-to-face interview rather than as self-administered
questionnaire. Despite the criticism of the CTS, for the purposes of this study the
original version was chosen because of its brevity, 38 items versus 78, and
because the theoretical basis and mode of operation were fundamentally the

same. Following some of the recommended improvements in the formatting for

104

the CTS, layout changes were made in the organization of CST so that it could
be used as self-report measure and for ease in scoring (The CST Manual is
available from the Family Research Laboratory, University of New Hampshire,

Durham, NH 03824).

Instruments Used for the Sub-Sample

The qualitative interview, named Family Relationship Interview, (See
Appendix F) was developed for this study by the researcher and was based upon
some general concepts extracted from Main’s Adult Attachment Interview (Main,
et al., 1985), e.g., what adjectives would you use to describe your mother, and
your father. Unfortunately, access for training in administration and interpretation
of the AAI is not widely available. Therefore it was necessary to develop a
qualitative instrument based upon this researcher’s clinical knowledge, and
familiarity with the attachment literature. Bartholomew and Horowitz (1991)
created their own attachment interview. Their attachment interview consisted of
historical questions about an individual’s relationship to mother and father from
earliest memory to present as well as a comparison of those attachment patterns
with additional questions regarding style of attachment to romantic partners. In
their study, they used tape-recorded interviews to classify adults into one of four
attachment styles: secure, dismissing, preoccupied and fearful.

The questions chosen for the Family Relationship Interview covered three
distinct areas: relationship to parents, relationship to spouse and relationship to
children. As stated previously, this area of inquiry was grounded in attachment

theory and the research was begun with the idea of ﬁnding connections between

105

childhood history and low and high conﬂict divorce. This investigator approached
the interview with theoretical awareness of the subtleties and meaning of the
data (Strauss & Corbin, 1990). Although the research followed a predetermined
schedule of questions for comparison of responses, the researcher allowed for
considerable ﬂexibility in regard to the respondent decisions about how best to
answer. Occasionally it was necessary to clarify a response, but otherwise the
researcher did not attempt to interfere with the ﬂow of speech once a subject

began to talk.

Parenting Stress Index

A second instrument was given to the small sub-sample of parents who
participated in the qualitative interviews. This instrument was not administered to
the entire sample due do to resource limitations. The Parenting Stress Index
was given to provide an indication of the relationship dynamics between parents
and children. The Parenting Stress Index (PSI) was developed by Abidin (1995)
to assist in the assessment of parent-child relationships under stress (The PSI is
available from Psychological Assessment Resources, Inc).

This instrument was normed on 2,633 mothers and 600 fathers with
children ages 1-12 years. This instrument includes characteristics of the parent
and characteristics of the child as two separate domains. The PSI has been
used to detect temperament or behavioral problems in children and stress in the
parenting role for a number of different environmental conditions, e.g., chronic

illness, child abuse and neglect. It has been used to screen for parental

106

attachment difﬁculties and families at risk. From derivatives of the original
version of 120 questions, a 36 question PSI-Short Form was developed. It has
ﬁve sub-scales that measure: a) total stress; b) parental distress; c) parent-child
dysfunctional interaction; d) difﬁcult child, and e) defended responding. At the
present time, the PSI-Short Form does not possess a body of independent
research that supports its validity. However, correlation between the full length
PSI and the short version sub-scales are as follows: r=.94 total stress; r=.92
parent distress, r=.73 parent-child dysfunctional interaction; r=.87 difﬁcult child;
r=.73 child domain and r=.50 parent domain on the parent-child dysfunctional

interaction.

Procedures

Prior to data collection, the researcher obtained written permission from
Michigan State University, University Committee on Research Involving Human
Subjects (UCRHIS) (See Appendix G). Additional permission was obtained from
the executive director of the community mental health center from which the
majority of the clinical sample was drawn. The local Friend of the Court director
also gave approval for the collection of data at the divorce orientation program.
Data collection began with the monthly divorce orientation program. Parents
were asked to participate in a study regarding low and high conﬂict divorce by the
researcher, who was one of presenters. The Divorce Questionnaire was
included in the packet of information distributed to each adult in attendance.

lnfonnation was shared about the voluntary nature of participation, conﬁdentiality,

107

and anonymity. A consent form (See Appendix H) became the first page of the
questionnaire, and only information from those who signed the consent was
used. Several individuals completed the questionnaire but did not sign the
consent. The questionnaire took most participants about 20 minutes to
complete. At the completion of the questionnaire, respondents were asked if
they would be willing to participate in an additional 1-hr interview. If selected,
they were told that they would be compensated for their time. Willingness to be
contacted was indicated by completing a request for their name and address.
This process was repeated for the months of March and April of 2000. During
the month of January, the Family Court coordinator of the program inadvertently
omitted the questionnaire from the packet, and February’s surveys were
considered invalid because of a copying error.

Similarly, individuals who were receiving clinical service were asked to
participate in the study when they arrived for their intake appointment. A
secretary approached individuals while they were signing other permission forms
regarding their willingness to participate. The letters of consent (See Appendix I)
promised conﬁdentiality and anonymity in reported results and also mentioned
that the information disclosed in the study would not become part of their medical
record or assist them with legal proceedings. As was done with the other
subjects, respondents were asked if they would be interested in participating in
an additional interview.

For the sub-sample, 5 respondents from the clinical group and 5

respondents from the normal group who had children between 5 and 12 years

108

were selected to provide additional qualitative and quantitative data. In selecting
this sub-sample, the researcher began by sorting 47 from the normal sample and
26 from the clinical sample who were willing to be interviewed. From that group,
she tried to recruit equal numbers of men and women with at least one child from
that marriage between the ages of 5 and 12. The primary reason for this
selection was to highlight parents with children who were old enough to display
an emotional and behavioral reaction to divorce, and yet not old enough to
participate in custody and parenting time decisions. A secondary reason was to
select a group of parents who have a high rate of divorce, and who are likely to
have parenting differences. Other factors in case selection were race,
occupation, and kind of custody arrangement. Most participants were contacted
by telephone and some were contacted by letter. When a contact was made,
further details of the face-to-face interview were shared. They were informed
that the interview would be audiotaped and they were told they would receive a
$40 gift certiﬁcate to a local mall for their time. Prospective interviewees were
told that the content of the interview consisted of questions about the family they
grew up in, their relationship to the marital partner and their current relationship
to their children.

All interviews were conducted in a professional ofﬁce at the community
mental health clinic where the researcher was employed. None of the clinical
sub-sample had a treatment relationship with the researcher, as the sample was
drawn from the caseloads of other therapists. Because the clinical sub-sample

was familiar with coming to the clinic, persons from this sub-sample were quite

109

cooperative about being seen, and seemed highly motivated to “tell their family
story”. Care was taken to make appointments at convenient times, and when the
clinic was open. As might be expected, it was slightly more difﬁcult to recruit
persons from the normal group, who were less familiar with the clinic. When
personal contact was made, agreement to participate was good. However, when
contact was by voice mail or answering machine, it was rare for a potential
interviewee to make contact. Among this group there was also a more frequent
rate of “no shows” or a need to reschedule appointments. However, those who
did come seemed genuinely interested in taking part.

A second consent form regarding permission to audiotape (See Appendix
J) was reviewed and signed when subjects arrived at the researchers ofﬁce.
Prior to the interview, the researcher disclosed that she was a graduate student
as well as a therapist. The tape recorder was turned on at the point all
preliminary information was clariﬁed. Either prior to, or following the semi-
structured interview, the parents also completed the Parenting Stress Index, on
the child of the marriage. When there was more than one child, the researcher
selected the child in closest age range to other focal children for closer
developmental similarity. The face-to-face interviews generally lasted about 1
hour and 15 minutes. One question was added to the interview schedule after
the ﬁrst two interviews. The question was whether or not the subject thought they
were in a high or low conﬂict divorce, and related to that question was their
explanation about why some couples had difﬁculty adjusting to divorce. As

happens with qualitative interviewing, the researcher became aware that the

110

subject’s own thoughts and ideas about the question under study were an
important consideration. Because of the interactive nature or qualitative research,
the format was adjusted to include subjects’ ideas about low and high conﬂict
divorces.

During the interview process, the researcher attempted not to respond to
the disclosed information as a therapist for two reasons. First, it was important
not to interfere with the current treatment relationship and clinical direction of the
case and secondly, it was important to stay focused on the process of the
interview from the perspective of a family researcher. At the completion of the
interview, some of the sub-sample asked for advice, and this was handled by the
telling of a metaphorical story, by recognizing their individual strengths and
efforts in making their future life better, or by referring back to their therapist
when appropriate. None of the normal sample asked for a referral to treatment
or other community resources. However, two parents had questions about their
children’s adjustment and used a few minutes at the end of the interview to
express their concern. The interviewees were thanked for coming and given the
gift certiﬁcate.

This concludes the explanation of the methods used in the study. Chapter
IV is focused on the quantitative results and Chapter V is devoted to the

qualitative results.

111

Chapter 4

QUANTITATIVE RESULTS

The divorced or divorcing respondents in this study through the
completion of the questionnaire revealed a detailed picture of themselves and
their relationships to their former spouses and children. Much was learned about
their areas of conﬂict and agreement and what kinds of spousal dynamics
accompany the process of divorce. The ﬁrst section of the results highlights the
descriptive statistics of the overall sample as well as a comparison of the normal
and clinical sample. The second section of the chapter addresses the
quantitative results of the ﬁrst four hypotheses. The results of the qualitative

interview with the sub-sample will be discussed in Chapter V.

Descriptive Statistics
Age, Gender, and Race

The 125 respondents who agreed to participate and who met criteria for
the study ranged in age from 22 to 47 years with mean age of 33.8 years. This
study comprised almost equal numbers of men, 59, (47.2%) and women, 66,
(52.8%). Number of years married ranged from 1.5 to 18 years with a mean of
7.9 years (See Table 1). Because of the criteria for participation in the study all
had minor children with an average of 1.7 children with a range of 1 to 6 children.
About 80% of the parents had two or fewer children from the marriage. The

racial composition of the respondents approximated the county census with 71 %

112

European-American, 6.4% African-American, 4.8% Native American, 4%
Hispanic, .08%, Asian American, and 1.6% other. About 11% of the respondents
failed to provide any lnfonnation regarding race. There were a small number of
paired couples who had been married to each other in the normal sample and
the clinical sample, however the major interest of the study was to look at group

differences.

Education, Income and Occupation

For the overall sample the mean years of education was 13.24 years
(s.d.= 1.97) or approximately one year beyond high school. Level of education
ranged from a low of 8‘" grade to a high of master’s degree (18 yr. of education).
Correspondingly, median income for the total sample fell between categories of
$15,000 and $25,000 with a low of less than $5,000 to a high range of over
$100,000 (See Table 2). The large majority of respondents, 81.6%, indicated
full-time employment and part-time employment, was speciﬁed by 11.2%.
Following a US Census pattern of grouping occupations, the overall sample
included: 16.8% professional, managerial, and specialty occupations; 21.6%
precision labor, craft and repair; 25.6% technical, sales, administrative support,
5.6% service occupations, and 26% operators, fabricators and general labor.
2.4% were not in the labor market and this group included students and retired

persons.

113

Additional Family Characteristics.

The age of the respondent when the relationship began with the spouse
varied from under the age of 18 to over age 35. The proportions were as follows:
24.8% under 18 yr., 36.8% age 19-24 yr., 27.2% age 25-30, 8.8% age 30—35,
and 1.6% over 35 yr. Custody of the minor children included a variety of
arrangements. Forty-six percent were in the mother’s physical custody, 3.2%
were in the father's physical custody, 21.6% involved shared joint physical
custody, 4.8% had children divided between two- households, 8.8% were still
residing in the same household with physical custody undetermined, and another
8% described some other custody plan. The amount of parenting time awarded
the non-resident parent was 21.6% for joint custody, 23.2% liberal access or
several days per month, 16% with every other weekend and a mid-week contact,
and 23.25% had no established court order. Only 2.4% of the respondents listed
parenting time as limited or restricted to supervised contact of less than 2 days
per month. The nonresidents’ pattern of contact with minor children was listed as
listed as 31.2% predictable, 24% regular, 9.6% unpredictable and 15.2% no
pattern of contact. Another 12.8% were unable to provide an answer, as there
was no pattern of contact established.

The respondents listed a variety of living arrangements: 9.6% were still in
the same household, 10.9% were living alone, 46.4% were living with minor
children, 11.2% were living with relatives, .08% were living with friends, 16%

were in a new live-together relationship, and 4.8% were remarried. The overall

114

sample was geographically distributed throughout the county with the largest

proportion, 52%, coming from the urban area in the center of the county.

Comparison of Normal and Cliniwl Samples

The comparison of the two samples revealed a high degree of
homogeneity in demographics. There was a signiﬁcant differences in age
between the normal sample (m = 32.7 yr., s._d. =6.26) and the clinical sample (m
= 34.9 yr., ﬂ = 5.67). This slight difference in age makes sense when it is
considered with time since the marital separation. More of the clinical sample is
divorced whereas the majority of the normal sample is recently separated. There
were no signiﬁcant differences in gender. The normal sample consisted of 30
males (47.6%) and 33 females (52.4%) and the clinical sample consisted of n=
29 males (46%) and n= 34 (54%), Chi-sq. (1,63)=. 925, Q>. 05 two-tailed. No
signiﬁcant differences were reported for race. It should be noted that the overall
size of minorities was so small that proportionate differences did not emerge as
statistiml difference (See Table 2).

Ordinal level variables, such as income, communication patterns, plans to
change custody, and parenting time were analyzed using the nonparametric
Mann Whitney U test statistic and the Chi-square test for proportionate
differences. This test choice for statistical comparison was based on the analysis
decision model offered by Sprinthall (1987, p 399-402). There were no
signiﬁcant differences in education between the normal sample, (m: 13.1 yr., §-_<L

=1.89) and the clinical sample, (__n_1= 13.4 yr., s3, = 2.03). There was no

115

signiﬁcant difference in how the normal and clinical samples ranked in income.
The normal sample indicated (n=61, mean rank = 66.83) and the clinical sample
(n=62, mean rank = 57.25), U= 1596.50, Q> .05 two tailed. However, by
reviewing a graph on income (Figure 2) it is evident that a certain proportion of
the clinical sample earned less than $15,000 per year. Correspondingly, there
was no signiﬁcant difference in occupational ranking in the normal sample and
clinical sample, Chi-sq. (4, n=121) = .299, g>. 05).

A t-test revealed that there were no signiﬁcant differences in the mean
number of years married between the normal sample, (m=7.8 yr., ﬂ. = 4.37)
and in the clinical sample, (m=7.9 yr., ﬂ. = 4.00). Likewise, the t-test revealed
no signiﬁcant difference in the mean number of children from the marriage, for
the normal sample (__= 1.7 children _s_._<_:l_,=. 71) and for the clinical sample (in:
1.84, _s_._d,=1.31) (See Table 1).

There was no signiﬁcant difference between the normal and clinical group
in regards to who initiated the divorce petition. Chi-square analysis revealed 37
(54.4%) of the normal sample and 31 (45.6%) of the clinical sample ﬁled. As
expected there was signiﬁcant sample difference in regard to whether or not the
divorce was ﬁnal. More clinical respondents, n=22 (36.1%), were divorced
compared to normal respondents, n=3 (4.8%) Chi-sq. (1, n=124)=18.867, Q <.
05, two tailed). Time since the marital separation also determined whether or not
there was an order for child support. There was signiﬁcant difference between
normal respondents and clinical respondents in regard to the establishment of

child support. More clinical subjects, n=45 (73%) compared to normal subjects

116

n=29 (46%) had an order for child support Chi-sq. (1, n=124) = 9.11, g<. 05 two
tailed.

All respondents were asked to list the top three reasons that the marriage
relationship failed (see Table 3). Because of the opportunity to select several
reasons for the divorce, the percentages exceed 100%. After visual inspection of
the percentage differences, speciﬁc reasons were analyzed for possible group
difference. There was signiﬁcant difference in regard to involvement in another
relationship between the two samples of respondents. A Chi square analysis
revealed signiﬁcantly more normal respondents n=28 (75.7%) compared to
clinical respondents n=30 (44.6%) listed involvement in another relationship as
contributing to the decision to divorce. Chi sq. (1, n=91) =3.846, g< .05, two
tailed. A second area of difference involved domestic violence. The Chi square
analysis indicated that there were signiﬁcant areas of difference for the clinical
subjects n=20 (41.7%) compared to the normal subjects n=5 (17.2%) in listing
domestic violence as a reason for the decision to divorce Chi sq. (1, n=77)

=4.919, g <. 05, two tailed.

Normal and Clinical Differences
Overall Attachment Style

Hypothesis I stated when compared to the normal sample, the clinical
sample will demonstrate differences in attachment style as measured by the

pattern and description of adult relationships in general.

117

The ﬁrst attachment scale was administered to gain information about an
adult's overall attachment pattern. As was noted in Chapter III, an item was
dropped from both the secure and anxious scales to improve this instrument’s
reliability. The analysis of the sub-scales on this attachment instrument did not
result in a conﬁrmation of the hypothesis regarding group differences for overall
attachment style. Only one of the three sub-scores supported the hypothesis as
stated, which was that the clinical sample indicated more avoidant responses to
attachment.

A t-test was used to determine sample differences for the sub—scale,
Secure. There was no signiﬁcant difference between the normal sample (m:
3.35, §_._c_l. = .656) and the clinical sample (m: 3.42, g. = .647), t (122) = .558.
This ﬁnding suggests that the sub-scale for capturing a relationship style
characteristic of a secure pattern of attachment may have been insufﬁciently
sensitive. This potential problem of internal reliability was discussed in the
methods section. The lack of difference in respondents may also imply that
some respondents overstated their ease in establishing comfortableness with
others in adult relationships. For the sub-scale Avoidant there was signiﬁcant
difference between the samples. A t test analysis revealed that the clinical
sample responded with a higher level of avoidant behavior (m: 2.75, g = .933)
as compared to the normal sample (m= 2.39, s_.d_. = .799), t (122) =2.28, g, < .05,
one tailed. Likewise, a t test showed that there was no signiﬁcant difference
between the normal and clinical respondents on the sub-scale of Anxious

attachment. The clinical sample (m: 2.45, s_.d_.: .559) responded to this sub-

118

scale similarly to the normal sample (m=2.50, s._d. = .629), t (120)= .511.
Because two of the three sub-scales failed to indicate large group differences,
the hypothesis regarding normal and clinical differences in overall attachment

was not conﬁrmed.

Attachment to Spouse

Hypothesis II stated, when compared to the normal sample, the clinical
group will report a different pattern of coping in regard to loss of marital partner
that will indicate a higher frequency of dismissing, preoccupied, or fearful
attachment.

A second modiﬁed attachment scale with four sub-scales was
administered as an indication of how well an individual was coping with the loss
of their marital partner. As was noted in Chapter III, an item was dropped from
both the secure and anxious sub-scales to improve this instrument’s reliability.
The results from the sub-scale secure on this attachment instrument did not
result in a conﬁrmation of the hypothesis regarding group differences on the
indices of secure attachment. However, it did highlight group differences on
three of the four sub-scales.

A t-test showed that there was no signiﬁcant difference for the sub-scale
of Secure in the two groups of respondents. The clinical sample (m: 2.13, g, =.
74) responded similarly to the normal sample (m=2.03, £1 =. 88), t (123)=. 55.
There was a signiﬁcant difference for the sub-scale of Dismissing. The clinical

sample (m: 2.10, s_.d = .92) endorsed more items dismissing the importance of

119

the spouse, as compared to the normal sample (m=1.17, s._d. = .96), t
(123)=5.49, Q'—' <. 05, one tailed.

There was a signiﬁcant difference for the sub-scale Preoccupied as
revealed by the t-test. The normal sample endorsed more items of
preoccupation with the spouse (Ln_= .63, s._d. = .89) as compared to the clinical
sample (m= .40, _s_._g.= .71), t (123)=1.66. This ﬁnding was not anticipated.
However, a similar response on the sub-scale Fearful suggests the possibility of
group difference regarding how the respondents approached the instrument.
There was signiﬁcant difference for the normal respondents endorsement of
responses of Fearful attachment (m: .89, _s._d=1.09) as compared to the clinical
sample (m: .48, s._d.: 84) t (123)=2.41, g. <. 05 one-tailed. This higher
endorsement of fearful reaction by the normal subjects may indicate some of the
expected emotional uncertainty of being in the process of divorce, whereas the
lack of emotional response by the clinical subjects may be the result of a later
stage of divorce adjustment. Alternatively, the lack of emotional reaction may
also be an indication of a higher degree of defensiveness regarding the loss of a
partner.

This second measurement of attachment regarding coping with loss of
partner did not result in expected ﬁndings. Only the sub-scale Dismissing
resulted in support for the hypothesis in the expected direction. As was
suggested in the ﬁrst attachment style measurement, the lack of endorsement of
the insecure sub—scales by the clinical subjects, may indicate group differences in

the degree of defensive responding. The normal respondents higher

120

endorsement of difﬁculty may indicate a more realistic appraisal of the meaning
of the loss of the partner or it may indicate a greater willingness to self disclose.
Because the sub-scale for secure attachment did not result in a difference
between the two groups and because two of the four sub-scales had differences
in the opposite to the hypothesized direction, the hypothesis was only partially

conﬁrmed.

Level of Conﬂict

Hypothesis I stated: When compared to the normal sample, the clinical
sample will demonstrate higher levels of conﬂict in the co-parental relationship as
measured by style of negotiation, frequency of communication, level of
disagreement regarding custody/parenting time and scores on the Conﬂict
Tactics Scale.

All of the variables that contributed to the construct of co-parental conﬂict
were conﬁrmed as supporting the hypothesis of expected differences between
the normal and clinical respondents. These ﬁndings indicate that there are
signiﬁcant behavioral differences between couples who divorce and meet clinical
criteria, and couples who do not meet clinical criteria. Differences in the
expression of conﬂict were evident in each sample’s description of
communication, the frequency of communication, and the style of negotiation
when communicating. Due to the basis of the sample selection it was expected
that the clinical sample would have more difﬁculty with the resolution of custody.

This expectation was conﬁrmed by the respondents’ perception of difﬁculty

121

resolving custody and number of court appearances. The sub—scales for the
CTS revealed some interesting group differences, especially on how each
sample rated the responses of the spouse.

The many of the factors that contributed to the overall variable level of
conﬂict were ordinal level variables. As such, they were analyzed by the Mann
Whitney U test and the Chi square test to determine proportional differences.
Each variable was analyzed individually as a separate contribution to the
hypothesis. There were signiﬁcant differences in the normal and clinical
respondents regarding how they described communication with the other parent.
The clinical sample described a higher rate of difﬁculty with communication
(n=60, mean rank=82.00) as compared to the normal sample (n=62, mean rank =
41.70), U= 630.0, g< .001, one tailed. There was signiﬁcant difference in the
respective normal and clinical samples regarding the frequency of
communication in a typical month. The clinical sample (n=59, mean rank=66.04)
reported a higher frequency of little or no communication when compared to the
normal sample (n=55, mean rank=48.34), U= 1118.05, Q< .05, one tailed. The
normal sample was more likely to engage in telephone contact on a weekly basis
than was the clinical sample. The clinical sample was more likely to have very
little telephone contact or to have no direct means of communicating (See table
4).

There were distinctive differences between the normal and clinical groups
for style of negotiation. The clinical subjects reported signiﬁcantly higher levels

of aggressive attempts to negotiate (n=57, mean rank=76.21) as compared to the

122

normal sample (n=55, mean rank = 41.39.), U=672.0, g< .05, one tailed.
Additional analysis of the responses on the variable style of communication
provided clariﬁcation regarding sample differences. A sample comparison by
cross tabulation for individual items revealed the following (Cramer’s V=.595,
p<.05); 1) A signiﬁcantly higher percentage of the normal sample (n=29, 49.2%)
reported using negotiation style that was typiﬁed by listening to the other and
making a decision based upon the reasonableness of the request whereas only
(n=6,10%) of the clinical sample reported using a similar style. 2) There was a
proportional difference between the clinical sample (n= 23, 39.7%) and the
normal sample (n=6,10.3%) regarding refusal to be available and commit to a
decision. 3) A higher proportion of the clinical subjects (n=15, 25.9%) used
verbal attacks, name calling, and swearing as a style of negotiation compared to
the normal subjects (n=3, 5.1%). 4) A higher proportion (5.1%) of the cliniml
sample indicated that verbal threats of harm, or physical attacks were method of
decision making, where none (0 %) of the normal sample indicated this method.
There was a signiﬁcant difference in the normal and clinical respondents
in regard to expectation regarding difﬁculty of resolving custody/parenting time.
Clinical subjects rated relative ease of resolution of custody as singiﬁcantly lower
(n=62, mean rank = 40.43) as compared to the normal sample (n=61, mean rank
= 83.93), U= 553.5, g< .05, one tailed. Related to the question of custody were
how many times the respondents had appeared in court. The normal sample
reported a signiﬁcantly lower number of court appearances (n=60, mean

rank=43.67) whereas the clinical sample reported a higher level of court

123

appearances (n=62, mean rank = 78.75), U=790.5, g< .05, one tailed. Support
for this ﬁnding also came from responses to a question about whether conﬂict
had increased, deceased or stayed the same since the physical separation. The
clinical respondents (n=57, mean rank=68.59) reported a signiﬁcantly higher
increase in conﬂict after the separation as compared to the normal respondents
(n=60, mean rank = 49.89), U=1163.5, g< .05, one tailed (see Table 4).

Each of the respondents was given the Conﬂict Tactic Scale (CTS) which
has three sub-scales, reasoning, verbal aggression, and violence. Each of the
questions required a rating of the self and a rating of the partner. Interestingly,
each sub-scale revealed one difference between the samples. A t-test analysis
revealed that there were signiﬁcant differences between the normal and clinical
subjects for the sub-scale of reasoning. The normal respondents were
signiﬁcantly higher on the rating of the other person’s reasoning (m: 7.96, s._d.:
4.04) as compared to the clinical respondents (m_= 4.811, ii: 3.95), t = (110)
4.17, g< .05, one tailed. There was no difference between the normal and
clinical respondents rating of self on reasoning. For the sub-scale of verbal
aggression , the normal subjects were signiﬁcantly higher (m = 11.17, a: 6.60)
as compared to the clinical sample (_r_n_= 7.00, i9; = 6.26), t (110): 3.431, 9, <.05
two tailed. A possible explanation for this difference is that either the normal
sample was more verbally aggressive in their conﬂicts or they were more open
about reporting the kind of conﬂicts that had occurred between themselves and
the spouse during the last year than was the clinical sample. As hypothesized

for the sub-scale of violence by the other person, there was signiﬁcant difference

124

for lower indications of violence in the normal respondents (_rr_1_= 1.98, g = 4.02)
as compared to the clinical respondents (m =4.02, ii =7.83), t (109) = 1.745, g,
< .05, one-tailed. The difference in rating of violent acts in combination with the
above variables supports the notion that there are behavioral differences
between the normal and clinical respondents in the use of verbal and physical
tactics used to deal with conﬂict. In sum Hypothesis III was strongly supported in
that all of the factors that contributed to the variable level of conﬂict demonstrated

higher level of conﬂict in the clinical sample.

Parenting Time

Hypothesis IV stated that when compared to the normal sample, the non-
resident parents within the clinical group will report less parenting time granted,
will have more limited or restricted access to their minor children, and of the
court-ordered parenting time granted, the non-resident parents will report a
pattern of contact that will be less predictable and more infrequent.

The assumption underlying this hypothesis was that the clinical
respondents would represent more conﬂicted co-parental relationships, and as a
result, there would be signiﬁcant differences in both the amount of parenting time
for the non-resident parent and signiﬁcant differences regarding the pattern of
contact. Surprisingly, this assumption was not conﬁrmed. A Mann Whitney U
test revealed that there was no signiﬁcant difference in the amount of parenting
time awarded to the nonresident parent for the normal respondents (n=41, mean

rank=45.23) as compared to the clinical respondents (n=55, mean rank 50.94)

125

U=993.50. Similarly, there was no signiﬁcant difference between the normal
respondents (n=51, mean rank=53.12) and the clinical respondents (n=50, mean
rank 48.84) for the pattern of contact. It suggests that both groups have variation
in the regularity of parenting time with some non-resident parents from both
samples being predictable and unpredictable. This study, by looking only at
between group differences, was not able to discern any variability regarding a
non-resident’s access to a child. It was expected that the nonresident parent of
the high conﬂict families would be litigating bemuse ofrestricted access to minor
children. However, this expectation was not conﬁrmed. A comparison of how
much of the parenting time was used by each of the respective samples revealed
additional similarity.

There was no signiﬁcant difference for the normal subjects (m=7.23, s.d.
=4.28), as compared to the clinical subjects (_rr_l= 6.98, 3d =6.33), regarding the
amount of parenting time used, t (63)= .180. As was indicated in the descriptive
statistics, the amount of parenting time for the non-resident parent for both the
normal and clinical sample includes a high number of joint custody (21 .6%) and
liberal access (23.2%) arrangements. Based on these percentages it appears
that an every other weekend arrangement was fairly uncommon as 7.10 days
was the average for the total sample. There were only 2.4 % of the respondents
who had limited or restricted contact of less than two days per month. There are
two possible explanations for the lack of difference in these ﬁndings. One
possibility is that the respondents who do not have a court order for parenting

time represent a percentage of non-resident parents who are being denied

126

access to children. A second possibility is that the clinical sample is strongly

defending their access to their children through active litigation.

Summary

In summary, there was support for the hypothesis that normal and clinical
samples differ in regard to conﬂict in the co-parental relationship. Group
differences were noted in the kind, style, and frequency of communication.
Clearly, the clinical respondents were more representative of divorced or
divorcing couples who have expectations that resolving custody will be difﬁcult.
Further evidence for this is supported by the frequency of court hearings and
intentions to change matters in court. The sub-scales on the CTS revealed
sample difference in all three areas: reasoning, verbal aggression and violence.

The two measurements of attachment did not result in a clear delineation
of secure and insecure attachment styles between the normal and clinical
respondents, and for this reason, Hypotheses l and II were not conﬁrmed as
written. However, the analysis of the sub-scales within each attachment
instrument provided some interesting questions as to why the two samples
responded so differently. It appears that the normal respondents were more apt
to indicate difﬁculty in coping with the loss of the marital partner, whereas the
clinical respondents were more likely to be avoidant in adult relationship and to
dismiss the importance of the loss of a spouse.

There was no support for a distinction of sample differences regarding

amount of parenting time for the non-resident parent. It is possible that

127

comparing amount of time granted with parenting time used was not a good
index of access to minor children. The lack of discrimination between groups
may be also be an indication of hidden problems with how the subjects were
asked to document parenting time.

The quantitative results highlighted that the normal group and the clinical
groups were very similar in terms of background characteristics, such as, age,
gender, race, education and income. There was strong evidence to support that
the normal group and the clinical group engage in different levels of conﬂict with
the other parent. The two attachment scales indicated some problems in deﬁning
secure attachment but because of noticeable group differences on the combined
sub-scales of two measures, there was reason to suspect that there were group
differences about how the samples responded to the question. To further
determine group differences, the qualitative data were collected to address the
context of the early childhood relationships with parents, which is the focus of the

next chapter.

128

Chapter 5

QUALITATIVE RESULTS

To explore the question of whether or not there was an attachment
difference originating from childhood, ﬁve respondents from the normal subjects
and ﬁve respondents from the clinical subjects were recruited. The interviews
with the respondents covered three distinct areas of attachment: their early
relationship with their mother and father, their relationShip with their spouse prior
to the divorce, and their relationship with their children. The purpose of exploring
these three areas was to establish whether or not there were individual patterns
of attachment that were carried forward from childhood that might offer some
insight as to individual differences in reactions to divorce. The underlying
assumption based on attachment literature and clinical practice was that the
clinical sample would contain a higher percentage of individuals with an insecure
style of relating whereas the normal sample would contain a higher percentage of
individuals with a secure style of relating. The participants were not selected at
random and there was no attempt to match them according to any demographic
or personal characteristics (see Table 4). The names of the respondents have
been changed to pseudonyms.

Results from each group will be presented by ﬁrst providing a general
background of each subject through the use of a vignette. The analysis for each
group begins with presenting the context of their early family life. The focus then

shifts to a description of early childhood relationships to mothers and fathers, as

129

well an accounting of some of the traumas or losses that occurred while growing
up. The next section addresses the respondent’s appraisal of the relationship to
the spouse and their perception of the dynamics of their marriage. Lastly, each
parent’s view of their relationship to their children is highlighted. The method of
reporting the results includes summary, paraphrase, and direct quotation. For
purposes of clarity, each topic will be divided into responses from the normal and

clinical groups.

Brief Vignettes of the Sub-Sample
Normal sample

Josh, age 26, is a white male employed with a skilled trade in a small
company. Following high school he entered into a skilled trade apprenticeship
and has been employed in the same job for 8 years. He is in process of divorce
after a 4-year marriage to his high school sweetheart. He and Kristen married
due to pregnancy. Together they have two sons, Tyler and Troy, ages 5 and 2
years. Both Josh and Kristen continue to live in the same household upon the
advice of their respective attorneys. At the present time, he and Kristen work
hours that overlap so they share parenting responsibility and also rely on day
care. Josh is hoping to have joint legal custody of his sons and he speaks about
his desire to remain in an active role as their father. He is of the opinion that the
sons, particularly the toddler, are not receiving adequate attention when they are
with the mother.

Josh is the second of two sons raised in a two-parent household in

130

another state. At age twelve, his parents moved to Michigan due to a job
transfer. Josh states that he did not initiate the divorce and indicated that it came
as a “complete shock”. He views his wife as wanting her freedom as she has
begun a new relationship with a bouncer at a bar and has placed drinking and
partying above parenting. He admits that there were ﬁnancial problems in the
marriage and he is now considering leaving his current job for higher income. He
states that initially he was quite depressed about the termination of the marriage,
and decided to enter brief therapy. He has the support of his parents who live
nearby and assist with caring for his children. He states that both his therapist
and attorney have given him good advice.

Kelsey, age 26, is a white female employed as general labor for a small
factory. She has a high school diploma. She is in the process of divorce after a
4-yr. marriage. At age 19, shortly into a dating relationship, she became
pregnant. She and Charlie decided to live together. They did not marry until
their daughter, Haley was about two years of age. Kelsey talks freely about how
she and Haley’s father love their daughter very much. At the present time, she
and Charlie share in a joint physical custody arrangement of every other week.

Kelsey grew up as the eldest of ﬁve in a middle class neighborhood.
Despite the appearances of an intact family, Kelsey recalled intense ﬁghting
between her parents that lasted well into her emancipation. She described a
rather turbulent adolescence having been inﬂuenced by a boyfriend who was
involved with dmgs and alcohol. Kelsey initiated the divorce, stating that despite

their love for their daughter, the marital relationship never developed. In her

131

view, they never socialized as a couple, and her husband continued to go
drinking with his friends while she was home with Haley. Kelsey has begun a
new relationship with a co-worker, who is the father of a 3-yr.old daughter.
Kelsey is very happy about this new relationship and indicates that her family is
quite judgmental about her decision to divorce.

Dan, age 32, is a white male who works as a computer programmer for a
large corporation. His divorce was ﬁnalized in about 60 days after an eight and
one-half year marriage. He resides in an apartment in the same neighborhood
as his ex-wife, Ashley, so that it was easier on their son, Scott, age 7, to adjust.
They share joint physical custody, and Scott spends every evening during the
week with Dad. Dan stated there were times in the marriage when things were
very good, and there were other times when he couldn’t ﬁgure out the problem.
Ashley is a nurse and has been diagnosed as having manic-depressive illness.
According to Dan, Ashley did well on medication, but she is non-compliant.
Since Scott was born, he and Ashley have worked different shifts. He said in
recent years, the weekends were difﬁcult because sometimes things were great
and sometimes she was too depressed to get out of bed. Ashley ﬁled for divorce
and the 6-rnonth waiting period was waived because they were in agreement
regarding all property and custody. Dan says, with Ashley “everything is quick
and snappy.” Ashley has begun a new relationship, which the father reports is
problematic for their son.

Dan now believes that Ashley is following a pattern of divorce and a

pattern of serial relationships like her mother. He states that Ashley’s parents

132

divorced when she was a teenager and that her mother is continually changing
jobs, changing where she is living, and changing boyfriends. He has heard that
Ashley is planning to sell the marital home and may move out of state. If so, they
may have a problem with custody.

Dan is the oldest of two. He was raised in a two-parent home under rather
ideal circumstances. The family ate meals together and had a family night once
per week. Although his parents worked different shifts, the parents focused on
their own marital relationship as well as family life. Dan mentioned that his
parents ate lunch together nearly every day. Weekends were for the family and
extended family events. Dan’s parent’s live nearby on a lake, and they help Dan
out with daycare.

Greg, age 42, is a white male, employed as a supervisor of a
manufacturing company, second shift. He is in the process of divorce after a 12-
year marriage. This was his ﬁrst marriage and his spouse’s second. He and
Carla have three sons, ages 11, 8, and 6. At the present time, he resides in the
marital home and has full custody of the boys. The children’s mother moved out
of the home to pursue a relationship with a man, ten years her senior whom she
met over the lntemet. Greg recognizes that the boys are very sad and upset
about the divorce but because of his work hours, he has to rely on help from
members of his church and extended family to help with childcare. He is
expecting that he and Carla will probably end up with a joint custody
arrangement, if she decides to marry the man with whom she is involved.

Greg is the fourth of ﬁve siblings raised in an intact supportive family in a

133

suburban community outside of the Detroit area. Within the past year, his father
has died, he lost his job, and his marriage ended abruptly. Greg has found it
difﬁcult to meet the demands of a new job and the demands of single parenting.
He is very reﬂective regarding the dissolution of the marriage and recognizes that
his wife was probably very lonely, while he was busy working long hours to
support the family.

Amber is an African-American female, age 29, who works in a local
hospital in a clerical role. In addition to working full time, she is also a full time
nursing student, near graduation. She is the mother of four children, and she is
the primary custodial parent. Amber has been separated from her husband for
eight years and she is expecting the divorce to be ﬁnal soon. Her two oldest
sons, Anton, 12 years, and Dequon, 10 years, are children of her marriage. A
daughter, age 6, and a 10—month-old baby boy are children in common with her
current partner. Amber married at age 18 years due to pregnancy. Her husband
joined the service, and she lived with his parents in his absence. At age 21, she
learned that he was having a child with another woman, and she left his parents’
home and sought assistance from her grandmother in caring for her two young
children. Her husband had very little contact with Anton and Dequon until he was
discharged from the service about four years ago.

Amber is the third of four siblings. Her parents were divorced when she
was about 10 years old. She described a very difﬁcult, messy divorce in which
custody was a big issue. Amber lived with her mother and had little contact with

her father. Amber states she has a close relationship with her mother, who lives

134

out of state and a distant relationship with her father. Although Amber does not
agree with her husband’s lifestyle that includes going to clubs and drinking, she
does not interfere with his access to their sons every other weekend. Amber is

very happy with her current adult relationship and hopes to be married someday.

Clinical Sample

Alex, age 36, is a Native American male who works as a general laborer
in home construction. He is the father of one son, Ian, age seven. Alex sees Ian
every other weekend and once during the week. Currently, Alex lives with his
ﬁancee, Sheryl, and her two children, ages 14 and 8 years. He has been
divorced for two years. Alex met his ex-wife, Mandy, in a bar, and they were
married for four years. He decided to end the marriage after two previous
separations and an inability to get along. He described sexual problems and
frequent ﬁghting for dominance and control as the cause of the marital
breakdown. Alex denies that his drinking behavior was related to the marital
conﬂict, but he mentioned that he stopped drinking following the separation. In
his opinion he and Mandy are in high conﬂict having been to court at least six
times in the last ﬁve months.

Alex is the second youngest of ﬁve children born to his mother, and the
ﬁrst child born to his father. He has three older half siblings, one of whom is
deceased. His parents were not married at the time of his birth, and his father
returned from the service sometime in the ﬁrst year of his life. Until six years ago

he carried the last name of his mother’s former spouse. Alex spent the early part

135

of his childhood in a city where he and his siblings were a minority in primarily a
black neighborhood. About fourth grade the family moved to a rural community
several miles from the city. He describes having a stay at home mother until she
went to work full time when he was about 10 years old. Alex spoke very
generally about his childhood relationships, but he remained close to his
grandparents who lived in the city.

Renee, age 36, is a white female who is employed as a nurse in a doctors
ofﬁce. She is the mother of one son, Adam, age six. She and Walt have been
divorced for three years, and by court mandate, they share joint physical custody
of Adam every other week. Both Renee and Walt have been previously married,
but this is the only child for each parent. Walt is very litigious, and they have
been in court so many times that the judge has taken it upon himself to give
many parenting directives. Renee would like to leave Michigan but jeopardizes
losing joint custody with the court if she does so. Renee does not date and has
no plans to do so in the future.

Renee met Walt following their respective divorces. She had little contact
with his family prior to their marriage, and it was his desire to move to Michigan
to be near his family. She described a very highly controlled marriage whereby
her autonomy was restricted by her spouse and his extended family. The
marriage seemed to deteriorate rapidly following the birth of Adam and following
her disclosure her own history of sexual abuse during marriage counseling.

Renee is the middle of three siblings and is the only female. She grew up

in a rural area outside of Michigan. She reports a very unhappy childhood,

136

having been sexually abused by her father for a number of years. Her mother
was quite ineffectual at protecting her or her brothers from physical abuse. Her
parents divorced after she left for college.

Courtney is a white male, age 37, in process of divorce. He works as a
motor technician in the transportation ﬁeld. He has two children from this
marriage: Gabriella, age 8, and Joshua, age 5. Both Courtney and his spouse,
Robin, brought children into the marriage. She has a 16-year-old son, for whom
she has physical custody. Courtney adopted this son during the marriage.
Courtney has a 15 year old daughter who came to live with this family ﬁve years
ago when her life with her mother became unstable. This daughter now lives
with her paternal grandmother.

Courtney met Robin in a bar. He stated that the marital conﬂict was
related to her smothering and possessiveness. He does admit that he did seek
comfort from a female friend and that caused the ﬁnal separation. They have
been separated for two years and during that time they have been in high conﬂict
regarding his access to the children including the stepson whom he adopted.

Courtney is the second oldest of ﬁve children. He describes growing up in
a family with many problems. His parents had intense physical altercations.
During his early growing up they separated and reunited and eventually divorced
when he was about 8 years old. He describes his mother as physically abusive
and his father as rejecting. He was very happy to be adopted by his stepfather at
age twelve.

Kyle, age 38, is a white male who works in a position of authority in the

137

penal system. He is the father of one child of the marriage, Nicholas, age eight.
He has been divorced for over six years. His was in a very brief marriage lasting
about one year. He and the mother, Jodi, share joint physical custody of Nick,
however mother’s home is considered primary and he has liberal access which
includes the right to have more time whenever she is not available. Kyle has
another child, a daughter whom he rarely sees. He reports that he was never
informed that he was the father until the matter was conﬁrmed by paternity
testing when the girl was three years old. Kyle’s relationship to Jodi was his only
marriage although he has had several long-term relationships with women.

Kyle met Jodi on the rebound after the termination of an eight-year
relationship. They had one separation prior to marriage and married one week
after Nick was born. Kyle and Jodi have had protracted legal disputes since their
ﬁnal separation that have run into excess of thirty thousand of dollars. He
describes his relationship with Jodi as volatile. He is currently in a live-together
relationship with another woman whom he refers to as his ﬁancee.

Kyle is the second oldest of four children. He grew up in an intact family
until the age of 16 when his parents divorced. He describes a family with very
traditional roles for men and women, and he indicated that because of his
father’s long work hours in an auto factory much of the parenting responsibility
fell on his mother. His parents divorce was made worse by the fact that a house
ﬁre destroyed the family home.

Jodi, age 36, is a white female who currently works in a job involving

commissioned sales. She is the former spouse of Kyle and they have been

138

divorced for six years. As indicated above, they share joint physical custody of
Nick and they follow a rather elaborate schedule of parenting that has been the
result of litigation in court. Nick is Jodi’s only child and she was not previously
married. She does not date and has no plans to do so in the near future.

Jodi is the third of four children. Both parents raised her until the age of
12, when her parents divorced. She describes a very permissive atmosphere in
her mother’s home following the divorce with little parental supervision or
guidance. When her father was in the home, he was very physically abusive to
her brother, and she knew she was her father's favorite. Her relationship with
her father terminated at the point in her life when she told him that she was
pregnant and not married.

Jodi indicated that she suffers from depression, and entered into therapy
about four years ago to deal with a host of interpersonal problems including
perceived rejection by her mother. She has been on medication to help manage
her depression for the past four years. Nicholas, the son, has also been seen for
psychiatric treatment, and he has been diagnosed with ADHD and depression.
The mother views the child’s difﬁculty as exacerbated by the excessive litigation

and the tension surrounding their conﬂict that does not seem to end.

Context of Early Childhood Environment.
Normal Sample
Four of the ﬁve respondents from the normal sample described growing

up in a two-parent household and all have a close relationship with at least one

139

parent at the present time. What bemme notable in the responses from these
men and women was the clarity of their stories. When they began to talk, it was
without hesitation or doubt. All added detailed comments that supported their
perception of family life. Also evident during the interviews was the positive
affective state of the respondents. Most had smiles on their faces and each
shared some aspect of their family of origin of which they were very proud. In
response to a question about happy events growing up, all mentioned holidays,
vacations, birthdays, and relationships with extended family as signiﬁcant family
events. Their descriptions of their family appeared congruent with existing
feelings about their parents. All the respondents recruited from the normal
sample mentioned a strong ongoing emotional connection to one or both parents.

Evidence for family stability is contained in the following comments by the
subjects: Kelsey reported: “My mom and dad’s been married all my life. I grew
up in a very stable home, went to the same school all of my life.” Dan described
his household as very family-oriented: “We had family night, we did everything
together, weekends were up north at the cottage. We all hugged each other and
kissed each other and it was very warm.” Greg also described a home with
affection: “My mom was the one that gave us the love and nurturing. And we
were always very open with each other. Just everybody got along. We didn’t
ﬁght and there was no pettiness”. Josh described family closeness following his
family's move, “And we just grew that much closer, ‘cause that’s all we had.”
Amber indicated an exception to family stability by indicating that her early life

included frequent moves that were confusing to her: “We’d always move; that’s

140

all I remember when l was little, is that we moved all the time and I didn’t ever
know why.“ Later she added that she thought the moves were related to her

father’s startup businesses, but she didn’t know for sure.

Clinical Sample

The parental stability noted in the normal sample was not evident in the
clinical sample. Four of the ﬁve respondents drawn from the clinical sample had
parents who divorced and the other subject’s parents were not married at the
time of his birth. Within the clinical group, there were disturbingly high incidences
of violence and childhood abuse. One male respondent recalled that during his
early years his parents had “knock-down, drag-out ﬁghts”, and the physiml
abuse by his mother continued after their divorce. As was previously mentioned,
her father sexually abused Renee. Three of the ﬁve openly talked about either
witnessing or being the victim of child abuse. Jodi recalled witnessing her
father’s brutal whipping of her brother to the degree that he could not sit. Renee
was witness to her brothers being “beat“.

The verbal detail that was evident in the normal sample was not apparent
in the clinical sample. In the telling of their stories, there were more breaks in
thought, switching of topics and responding without answering the question.
There was a higher degree of dependence on global terms or cliches to express
meaning for the male respondents. Alex described his family in this way: “Oh, I
got a good family. Always has to work for what I had. And, you know, I never

had no problem with it.” The second male subject also stated, “I grew up in a

141

traditional family. We were close, we were a very close family.” However, each
of these two positive presentations did not match with later descriptions of
intense conﬂicts and the severing of ties with siblings. Overall, the male subjects
presented with a high degree of anxiety as evidenced clinically by their pressured
speech and the intensity in which they discussed connections to parents at the
present time. All the males mentioned having weekly contact with their mother
as proof of their closeness.

The two female subjects from the clinical grOup clearly recognized that
they had endured a great deal of sadness and depression during their childhood
and they talked about the deﬁcits in each of their parents’ ability to parent. For
example, Renee shared that what she emotionally received from her parents was
insufﬁcient. “You know I, I felt like I missed out a lot, not having two, two loving
parents.” Both women saw their mothers as preoccupied and unable to
emotionally give to them. Renee and Jodi experienced rejection from their
fathers associated with their emancipation and assertion of independent thoughts
and feelings. Both told very salient stories about the last conversations they had
with their fathers. The two females have contact with their mothers. However,
both expressed concerned with their mother’s current welfare. As adults, they
have struggled to understand their mother’s lack of connection to them during

their childhood.

142

Relationships to Parents
Normal Subjects

Both of the females from the normal group were very clear about sharing
negative aspects of their parents’ marriages and both women shared that their
primary attachment was and continues to be with their mother. One female
respondent recognized that her father was not affectionate, and she longed for
more emotional connection to him. The other female respondent, whose parents
were divorced, described her father as “in and out, in and out, and in and out,”
showing up only occasionally. Despite not having a close relationship with her
father, she mentioned that he always ﬁnancially supported the family to the
degree that her mother did not work and the family did not receive ﬁnancial
assistance. The males from the normal group all described a relationship with
their father that included admiration or some type of positive role model. One of
the three recognized there was some emotional distance in the relationship
between him and his father while growing up, despite spending time together.

All the men described a warm and supportive relationship with their
mother. Mothers were described in very positive terms that included emotional
support, warmth, and emotional availability. Greg states, “She was always there
for us, and all the love that I have comes from her.” This same male respondent’s
earliest memory was of his mother singing to him. Another male respondent
indicated, “My mom is a very caring person and will do anything for us. Some
people say she hasn’t cut the umbilical cord yet.” The male respondents all

mentioned food as part of their nurturance from mother and made references to

143

regular family dinners.

One female respondent was aware of her mother’s perfectionist control,
and her own struggle for autonomy that became a theme throughout her growing
up. “I always tried to be good in my mom’s eyes. I wanted to be good for my
mom, yet I so much wanted to do what I liked or what I wanted to do.” This
respondent, along with all the others, recognized that her mother had the largest
share of responsibility for child rearing. “My mom had a lot of the responsibility of
raising us ﬁve kids.”

When asked to use four adjectives to describe their mothers, respondents’
answers from the normal group were very succinct and clear. “ Well, she’s loving,
very caring, responsible and a teacher." Other descriptions were very similar:
“deﬁnitely loving, open-hearted, affectionate and caring,” and “very loving,
honest, sincere, sweet", and “caring, hard-working, ambitious, healthy, and
sweet“. In summary this group of subjects viewed their mothers as emotionally
available and attentive throughout their growing up years.

The respondents from the normal sample described relationships to
fathers with more variation. There were uniform references regarding the father’s
contribution of providing economic support for the family by all ﬁve subjects. One
male subject stated, “the way I described my mom is how I would describe my
dad, except my dad’s a worker; he’s always putting into the community, being a
member of a board, an association, such as the Jaycees.” The fathers were
described by the male respondents as teaching speciﬁc concrete skills. Greg

described his father’s involvement in photography, woodworking, and mechanics

144

and the mentoring role that occurred. “As I grew up I followed him, and he’d
explain the things that he did and why. He always studied everything before he
did it and then when he did it, he did it just right.” Building things together was a
common descriptor by the male subjects. The female subjects were less
complementary regarding their father’s relationship with them and they seemed
open to revealing their father’s deﬁcits. “My dad didn’t show affection.” She
continued, “You knew dad loved you, but you never ran up and gave dad a hug
and kiss and I wanted that.” The other female subject described not having a
relationship with her dad because he wasn’t physically or emotionally available,
“We didn’t really have a relationship."

Some adjectives used to describe fathers were: “methodical,
disciplinarian, and frugal” and “loving, caring, very work-oriented, and a role
model.“ Less complementary adjectives were balanced with an appraisal of a
parent-child relationship by one female respondent who stated, “My dad’s not
understanding, he needs to listen more, and not be so stubborn, but I would say
that he’s very caring and loving, too.” Another female subject described her
father as “awful, mean, and a bully” but recognized that he always indulged her
for her birthday and Christmas, “That’s the one thing that he did do”.

Clinical Sample

The subjects interviewed from the clinical group were less able to recall
speciﬁc memories between themselves and each parent at different
developmental stages. They were more likely to respond by talking about what

they did as a family rather than by describing feelings. As with the other

145

respondents these subjects made references to birthdays and family holidays
with extended family as special family times. One noticeable pattern among the
clinical respondents was the common use of the pronoun “we“ instead of a
references to themselves. Sometimes the pronoun ‘we’ seemed to represent the
family unit and other times, the ‘we’ was an inclusion of siblings in their response
to relationships to parents. Two subjects said they had difﬁculty recalling any
individual happy times with their parents. The clinical group included more
descriptions of love from a parent that also included emotional distance. In
addition to Renee and Jodi, whose fathers refused to have any thing further to do
with them when they confronted their fathers, Courtney also indiwted strong
feelings of rejection when his father terminated his parental rights as means to be
released from his child support obligation.

Respondents from the clinical group characterized their mothers in the
following ways: Kyle stated, “ I’m closer to my mom than I am to my dad. I’m
close to them both, but my mom is, you know, my mom.“ Alex used these words
to describe his mother, “As a young child, see, we, we, always done stuff
together. She never worked until I was 10 years. We always done stuff together,
oh, we’d go out to eat. When she’d go to the store, I’d go with her, do grocery
shopping, go on family vacations. We always done stuff together.“ Courtney
described his mother in this way, “My relationship to my mother was a close one,
but yet a distant one. The distance was because of the abusiveness.” He further
states, “Our mother was the only one we could cling to.” Renee spoke of her

mother in these terms: “As a kid I thought I was close to mom, and as I look back

146

on it now, I don’t think it was as close as I thought it was. I think she spent a lot
of her time trying to keep peace in the family, so she didn’t have a lot of time or
energy to do much else.” Jodi also speaks of having little closeness with her
mother. “It wasn’t very close, I don’t recall much of my childhood, so I couldn’t
really tell you a whole lot. She didn’t get us involved in anything. She was really
caught up in her own life, so you know we never felt love.” Both of the females
from the clinical group talked about blocks in their memory.

Alex used these adjectives to describe his mother, ”Uh, kind. And she
always wants to spend money. Always. Even though she don’t have to, she’ll do
it. She's very loving. And affectionate.” Kyle described his mother as “very
warm, very caring. My mom’s a huggy-feely, warm-fuzzy, gives me a kiss kinda
person.” Courtney describes his mother with ambivalence, by stating, “Quick to
anger, not necessarily quick to forgive, and a lasting love, and educator.” Jodi
struggled with adjectives to use for her mother. “Gosh, she’s a tough one to
describe, outgoing, lost, selﬁsh, and unconcerned with her future.” Renee also
struggled to come up with words, “I didn’t think that would be so hard.” She
came up with the words “fearful” and “hard working,” and changed the subject
without naming two other adjectives.

Three of the male respondents described their father as the disciplinarian
who taught them right from wrong through lessons learned by way of
punishment. Alex, whose family lived in a racially mixed community, credits his
father for teaching him how to ﬁght. Courtney, who was adopted by his

stepfather at age 12, credits this man as teaching him how to react, “He was the

147

one who said, ‘Yeah, you did wrong and you will be punished, but I am not going
to be mad at you forever.” Kyle’s father never attended any of his sporting
events growing up, and minimizes this by stating “I was a little resentful.” Despite
no evidence of emotional involvement with his father growing Kyle describes his
father “as one of my best friends.” The female subjects have little difﬁculty
describing fathers in negative terms. Renee has no memory of any kind words
from her father, “There just wasn’t a time that we didn’t feel fear he was going to
be angry about something." She uses the adjectives, “mean,” “angry," and “very
selﬁsh” to describe her dad. She indicated that, nowadays, he would have been
arrested for abusing her brothers. Jodi, at this point in her life, views her dad as
a perfectionist, a jerk, controlling and mean. Three of the ﬁve subjects concluded
that that they were neither nurtured nor protected growing up, whereas two hold

the perception that they were loved.

Childhood Losses and Traumas

Nearly all of the respondents from each group mentioned signiﬁcant
losses in the middle of childhood. Particularly salient were deaths of
grandparents who played important active roles in their lives. The losses and
injuries described by the clinical respondents were not described with the same
level of resilience as was described by the normal subjects who seemed to have
processed their loss and accepted loss as part of life. The clinical subjects talked
about the traumas with more poignancy and with either blunted or with an

absence of affect.

148

Normal Subjects

Some of the losses and traumas described by the normal sample were as
follows: Dan mentions the loss of a great uncle with whom he was very close,
“He’s passed away but my aunt is still alive.” Josh, at age 9, witnessed the car
accident in which his grandmother died. He and his brother were the ﬁrst
persons on the scene. Josh described his grandmother’s death as “having a big
impact on me because it was the ﬁrst time I’ve ever experienced a tragedy.”
Amber recalls growing up learning that her brother had been shot in the leg. She
was quite philosophical about it, recognizing that he was somewhere on the
street where he wasn’t supposed to be. She also talked about the unexpected
death of a cousin, who died shortly after the birth of twins.
Clinical Subjects

In contrast, the traumas in the clinical group were more dramatic and
cryptically presented. Renee talked about the death of her grandfather with
these words; ”He gave the praise and the love and the affection that we didn’t get
at all from my dad. I’m sure the saddest I can remember ever being is when he
died.” Alex indicated that the deaths of his grandparents were very difﬁcult for
him. “The death of my grandma and grandpa just devastated me.” With further
probes it was discovered that these were not childhood events, but events that
occurred recently in adulthood. Courtney shared that at age nine, he was alone
with his great—grandfather when he died. “I remember him sitting at the table and
he fell off the chair in the kitchen. When [he] hit the ﬂoor he was dead. I did not

know this. When he fell, his arms were outstretched. What I did was I went and

149

lay on his arm and went to sleep... But I was not afraid.” Courtney’s response to
this traumatic event was highly unusual, but more interesting is his need to
portray that he was not afraid. Kyle, too, mentions with much remorse the ﬁre
that destroyed the family home shortly after his parents’ separation. He was sent
to live with an aunt and uncle to ﬁnish high school because there wasn’t room in

his mother’s apartment where she had custody of his two younger siblings.

Marital Partners

Among the ten subjects selected for the interview, there was a high
incidence of brief courtship and pre-marital pregnancy. There did not seem to be
any pattern of courtship difference that stood out between the two groups such
as length of relationship, or age at marriage. Three of the ﬁve subjects from the
normal sample met during high school, and the remaining two met through work
or a common interest. Three of ﬁve from the clinical sample said they met in
bars. Subjects from both groups made comments about differences between
themselves and their partners regarding the use of drugs and alcohol, particularly
as part of marital separation. Although alcohol was not recognized as a
difference during courtship, it was a recognized behavioral difference that
accompanying the process of divorce. None of the respondents from the normal
sample were previously married, whereas two respondents from the clinical
sample were divorcing for the second time. All of the subjects from the normal
sample were in low conﬂict with the other parent and had some positive regard or

positive appraisal of the relationship with the spouse at some point in time. In

150

contrast, all of the subjects from the clinical sample were in high conﬂict with the
other parent and were less likely to attribute any positive value to their former
spouse. However, this difference in conﬂict was expected as it reﬂects one
criteria for being selected into the clinical group.
Normal Sample

Kelsey describes her early relationship in this way, “It was good. It was
fast, real fast. We were going together for one month when I got pregnant for our
daughter.” One male subject described the love he felt for his wife in very
idealistic terms. “She was the woman I always dreamed of and had the biggest
crush on [in] school”. Dan describes his courtship as taking a little while to get
started but then as “taking off”. He and his wife lived together for a year and a
half prior to marriage and they helped each other with educational expenses for
their current careers. Amber describes her marriage as a shotgun wedding
because of family pressure to marry.
Clinical Sample

Renee met Walt about six months after her divorce and they dated about
one year. She stated, “ It was good, and l, we, did a lot of things, and had a lotta
fun and traveled some. I could see having a future with somebody that I thought
was very well-grounded in his work, and he was.” Renee indicated that she was
fooled by his dedication to his family, and overlooked the fact that his mother is
schizophrenic. Alex said the marriage was ﬁne in the beginning but stated there
was an abrupt change that accompanied the birth of their son, “ It’s like putting

your ﬁnger in a light socket and switching to a different person. She was totally

151

changed.” Alex described their competition for dominance in this way, she
became “the dragon queen or what ever you want to call it.” Jodi recognizes that
there were severe problems in her relationship with Kyle prior to marriage.
“Basically, he just started pushing me over the edge one night and started
beating me up and kicking me, knocked me down and I just said that’s it. Two
weeks before I knew I was pregnant I had no intention of ever taking him back
again because I saw what a selﬁsh person he was." Kyle described the early
relationship to Jodi in this manner, “Things went positively for four or ﬁve months,

and then suddenly it changed.”

Reaction to Loss of Marital Partner
Normal Sample

There were emotional differences between the normal and clinical
respondents regarding acknowledging ambivalence about ending the marriage.
Three of the male respondents from the normal group were able to share their
difﬁculty in accepting divorce despite the recognition that there were problems in
the marital relationship. In discussing his ex-wife’s current partner one male
stated, “if she gets married to this guy, then that puts a ﬁnal end to it, then I could
pick up and move on.” A second male respondent stated, “I’ve come a long
ways. I’ve had a lot of family support, a lot of friend support, and I’m doing much
better. I guess a lot of it is the comfort of my kids.” A third male, when asked
how he will know he is over the relationship put it very directly, “I may never get

over it. I mean, my mindset was, I got married for life. When I said, ‘I do, till

152

death do us part, better or worse’, I meant all of it.” One female respondent
shared that her husband apologized to her regarding his behavior while in the
military, and that seemed to be very important to her. “I was so proud of him, I
never seen something more wonderful. He knew he has messed up and he just
left me alone. He never argued with me, we never argued, “we just agreed
maybe we shouldn’t have been forced to be married by everybody.” She
considers herself and the children’s father as friends. The other female
respondent recognized that the relationship happened too fast and that because
of their young ages, they didn’t recognize their individual differences. Her
ambivalence is related to her daughter, “Because of my daughter, I wonder if I
could have tried harder, and in my heart, I know it’s not going anywhere. I feel
like I owe him something even though he owes me a lot.”
Clinical Subjects

For the clinical subjects, the termination of the marriage involved more
abrupt separations and fewer attempts to live together while working on the
relationship. Following the separation, there was more behavioral acting out for
the clinical sample. His brother-in-Iaw assaulted Courtney after being called to
come to his spouse’s home for an emergency with the children. Jodi indicated
that Kyle would show up at her work and try to get her ﬁred. He was arrested
and taken to jail overnight on two occasions following violence at her home. Jodi
has also been charged with assault by his girlfriend, and the matter is going to
trial. Generally, the clinical subjects placed a higher degree of blame on the

other spouse for causing the divorce. Jodi places 90% of the blame on her ex-

153

spouse, Kyle, and states “My husband was an angry, hateful man when I met
him and he hasn’t changed.” While discussing her father she mentions that her
husband is controlling like her father only ten times worse. Alex blames his ex-
wife for 90% of the divorce because she just totally changed.

Within the clinical sample, the descriptions of getting over the divorce
varied widely, but most contained words of disrespect masking hurt. In response
to the question “What tells you that you are over the divorce”? One respondent
stated, “The fact that I hate him. The fact that I wish he would disappear.” Alex
denied having any feelings of loss regarding the marriage. He stated he had
zero difﬁculty coping, “I’ve had no problem with it.” Renee placed her level of
difﬁculty, on a ten-point scale at “2". It wasn’t hard to live without a husband. I
think the conﬂict that was there, was always between us, and that’s why it is so
easy to let that go.” Kyle states that he is very indifferent about his ex-wife, and
that he wishes her well; however, in another part of the interview he reported
having extreme difﬁculty managing being in the same room with her. Courtney
attributes the marriage breakdown to his wife’s smothering and possessiveness.
Although there has been a two-year separation with many daily conﬂicts,
Courtney is one of the ﬁve clinical subjects to admit that he isn’t emotionally over
the marital partner, “Well, to this point I do not think that I am over it. It will take
awhile because part of me still says yes, I will love her. So I think it is going to
take some time. I am still connected to her because of the children.” Courtney

has not ruled out the possibility of reconciliation.

154

Co-parental Relationship
Normal Sample

There were distinct differences between the two samples regarding
recognition of the child’s right to have a relationship with the other parent or
recognition that they didn’t want to deprive the child from beneﬁting from the
other parent. Most of the normal respondents could separate their personal need
from the developmental needs of the child. “When it comes to our daughter, we
both love her very much, and she is my main purpose for everything.“ Another
female subject recognized the need to get over her hurt and to let go of the
marital relationship. This subject’s ability to get on with her life is demonstrated
by her acceptance of her role as single-parent. “Because he was in the military, I
knew I would always raise my kids by myself so, he never gave me a conﬂict
about them and if he ever wanted to see them, I never gave him a conﬂict. We’re
friends now.” Another male subject stated his struggle in this way. “I want to
give the time that she needs with the kids, but dealing with her betrayal is very
hard.” Another father stated it this way, “I know she loves the kids and is very
good with them.” These subjects seem to have clear understanding of what was
required of them as parents separate from their feelings about the marital
relationship
Clinical Sample

Because the level of conﬂict between the parents in the clinical group was
very high there was also more competitiveness regarding time with the children.

One mother who has a joint custody arrangement doubts that their son is

155

beneﬁting from time with his father: “If I felt that Nick was beneﬁting from his
situation with his dad, I would never try to change it. But I don’t feel like he’s
beneﬁting at all”. Kyle, the other parent of Nick, states, “She interferes with
everything, it’s unbelievable. Everything’s a ﬁght. I get a court order and my
court order ain’t worth toilet paper.” Alex refuses to talk to the other parent and
rationalizes his behavior by stating, “I don’t even try to communicate with her. I
just stop, pick up my son, and leave. I don’t want to communicate with her
‘cause every time I do, it ends up in an argument.” Renee expects the conﬂict to

go on with her son’s father until Adam is 18 and possibility beyond.

Relationship to Children
Normal Sample

All the parents interviewed expressed love for their children. What was
notable between the two groups was whether the parent viewed the child as an
individual separate from himself or herself. More of the normal subjects,
compared to the clinical subjects, recognized the child’s loyalty struggle with
each parent. Kelsey knows that with the joint custody arrangement, she is giving
up some of her daughter’s dependence and loyalty to her. “So now I have to
share my time with her dad and it bothers me because it before it was just ‘mom
and me.” She goes on to explain why she doesn’t want to interfere with her
daughter’s emotional connection to her dad. Greg is very ambivalent about
whether or not to allow the boys to have ovemights with their mother, because

morally he thinks it is wrong for them to be exposed to a live-together

156

relationship, and yet he knows that his sons miss being with their mother.

The normal subjects were more likely to make references to the child’s
behaviors that were problematic to them whereas the clinical sample made no
mention of having any difﬁculty with parenting. In talking about her children,
Amber recognizes that she needs to work on her oldest son’s attitude, whereas
her other son and daughter are easy-going and laid-back. Greg was also able to
discuss the emotional struggles that his sons were experiencing. “ My oldest son
is a lot more selﬁsh than I’d like him to be. He has some serious bouts of anger
and talks back to me more than I ever talked back to my dad.”

Subjects from the normal sample were more focused on the future and
recognized that what was occurring in their life now had an impact on their
children’s future. Amber states she is role-modeling the need for education by
going to school and working. She feels strongly that she needs to instill in her
children, while they are growing up, the need for education beyond high school.
This mother clearly saw the connection between her own behavior and her
children’s later expectations, “They are constantly watching what you do, “cause
they know if you do it, its acceptable and it’s OK.” Josh also shares his future
expectations for his children by describing their characteristics and how he views
their personalities as helping them adapt in the future. Josh was also quite open
about sharing the meaningfulness of being a father". “ I live for everyday for them.
I love to come home and see their faces. It’s deﬁnitely an experience that is
irreplaceable. I guess it’s always what I wanted.”

Dan also shared high enthusiasm for parenting. “My son and I have a very

157

strong relationship.” This dad says he is able to focus on the child’s interests
70% of the time while he is with his son, and the child has to adapt to his needs
about 30% of the time. He describes his parenting in this way, “I’m pretty much,
Mr. Mom and Mr. Dad rolled into one.” What became apparent from the parents
from the normal group was their strong identiﬁcation with parenting and their
overall general positive regard for their children.
Clinical Sample

In contrast, more parents in the clinical sample were likely to see their
children as an extension of themselves, rather than being autonomous
individuals who were impacted by their behavior. There was generally a lack of
sensitivity to issues between the parents. They were more likely to equate
talking to or being with their children as evidence of closeness. One father
reports how he talks to his son, “When I know something is bothering him, I just
sit him down and ask him, “Well what’s bothering you?’ ‘Oh nothing,’ I goes,
‘something is bothering you, and you tell me what it is.’” Another father reports,
“They love to be with me and love to be around me, and they love it when I come
to pick them up and take them to my mom’s and get them away.”

During the interview, Kyle talked about the daughter whom he rarely sees.
He stated, “I hate to say it, but I just don’t have the bond with her like when my
son was born.” In response to a question to describe his relationship to his
children, Courtney revealed his favoritism for the two children from the current
marriage contrasting his feelings with his adopted son and his oldest daughter.

He believes that he has treated his adopted son very well, and believes his

158

treatment of him has been equal to his children. He accounts for his lack of
attachment to his daughter with these words, “I was not able to bond with her and
be with her like I would want to be as a father.” Because of intense conﬂict
between stepmother and stepdaughter, this adolescent who came to live with
him at age 10, now lives with her paternal grandmother, and he sees her on the
weekends. This father used the same words uses and the same terms to
describe his relationship with his daughter that he used to describe his
relationship with his mother, “close but distant.” * Courtney does not seem
bothered by his identiﬁcation with the two children from his second marriage as
being his favorite, “because they look like me and they are built like me.” For
these two fathers, not being ‘bonded’ was an acceptable rationale for not being
active in their child’s current life. At the same time, they are engaged in a highly
competitive legal struggle to get as much time as they possibly can with other
children so that they can have inﬂuence over their lives.

The mothers described their sons by talking about the level of affection
between them. One mother stated, “My son knows how much I love him. We’re
huggy, kissy, touchy, feely, you know, kinda family. We like to snuggle.” The
other mother described her relationship to her son, in this manner, “It’s close. We
have a lot of fun. I’m very loving, although now he’s telling me he’s too old to give
me kisses and we got into that.“ Two fathers see time with the child as critical to
their role as a parent. The ﬁrst father states, “I wanna be a dad to my son, and I
want to be with him as much as possible. I’ve got a responsibility so he can grow

up to be a good person.” The second describes his relationship to his son: “We

159

get along great. Y’know, l, l’d set him right from wrong.”

These parents believe they are satisfactory parents. Although they
recognize that the conﬂict between them and the other parent is not good for the
child, none of the ﬁve really directly talked about the impact of a child caught in
the middle. Renee came close recognizing the child’s struggle, but seemed to
hope the child will ultimately pick her. “He’s pulled in two separate directions,

and you just hope he takes the right one”.

Perceptions About Ongoing Conﬂict

Amber offered this as the explanation of why some couples continue to
fight and be involved with each other post-divorce: “Either the woman don’t
wanna let go or the man don’t wanna let go, and the reason is they too selﬁsh to
tell each other that they still want ‘em.’ In her opinion divorce is too easy and
she thinks more effort should be placed on keeping couples together. Dan
mentioned two possible issues about divorce conﬂict: One was men’s anger
related to the changing roles of working women’s and their lack of ﬁnancial
dependence on men. He said that ﬁnancial freedom increased expectations for
companionship and raising children together. The second issue for Dan was
about couples living together and choosing not to many. He thinks it is because
of an overall lack of trust between couples. For Josh it is important that he stay
focused on solving problems and controlling his anger appears very important to
him. “I don’t raise my voice often, and I don’t argue, l, I might talk a little more

aggressively, but I just don’t raise may voice.” Josh believes strongly in using

160

someone outside the family to help resolve issues when there is conﬂict.

Renee believes the ongoing conﬂict with her ex-husband is a way to
interfere with her happiness. “The goal is to keep me from being really happy,
“cause I, in, I think, in his eyes, he doesn’t think I deserve to be happy for
whatever reason. I didn’t conform to the way that he wanted me to be.” She see
herself as being overly sensitive to issues of control because of the family in
which she was raised. Greg sees couples as having decreased motivation to
cooperate once it is clear the relationship is dissolved. “ The main reason for
wanting to work it out was to stay together. Well, that’s gone, so why do you
wanna work it out?” He adds, “The anger that you develop just gets focused on

the other person.”

Parenting Stress Index

To determine if their were notable differences in the subjects’ approach to
parenting, each of the parents who were interviewed were asked to complete the
short form of the Parenting Stress Index (PSI) on a focal child. Although the
sample size is too small to make any generalizations, there were interesting
differences on two of ﬁve sub-scales of the instrument: defended responding;
parental distress; parent-child dysfunction, difﬁcult child; and total stress.

On the PSI one father from the clinical group met criteria for defended
parenting because his scores were so low. The father's lack of scoring items is
inconsistent with the child’s reported history of psychiatric treatment and

medication. Extremely low scores have three possible interpretations: 1) The

161

parent is trying to portray the image of a very competent individual who is free of
the emotional stresses normally associated with parenting, 2) The parent is not
invested in the role of parent and therefore is not experiencing the normal
stresses in caring for a child, 3) The parent is, in fact, a very competent individual
who handles the responsibilities of parenting very well and has excellent
relationship with others, including his or her spouse. (Abidin, 1995). The ﬁrst
interpretation is used when there are known facts about the parent and child
relationship. Abidin suggests that the defended responding is a common
response of an overcontrolled individual who denies the reality that parenting is
difﬁcult work. With this exception the other four parents from the clinical group
had elevated scores for difﬁcult child at or near the clinical range. The parents’
rating of difﬁculty with the child was in sharp contrast to the interview where the
clinical sample did not talk about behavioral difﬁculties with their child.

In comparison, the parents from the normal group had average to low
scores regarding the sub-scale, “difﬁcult child”. As was previously mentioned,
the parents from this group spontaneously talked in the interview about their
concern regarding their children’s behavior. Within the normal sample, one
mother scored high on defended parenting. The PSI manual suggests that in
some instances a low score it can be an indication of successful parenting. One
father from the normal group scored very high on overall parenting stress, which
was consistent with the early stage of divorce and sharing in the care of two

young children. Additional details, regarding the PSI, are found on the graphs

162

comparing the ranges of responses on the sub-scales by group (See Figures 4 &

5).

Summary

The results from the interviews support Hypotheses V as proposed. The
subjects from the normal sample appeared to have a secure attachment
relationship with one or both parents whereas the subjects from the clinical
sample had more insecure or ambivalent attachments to one or both parents. In
terms of degree of felt security, most of the normal sample grew up in two parent
homes in which they felt assured of at least one parent’s love and emotional
support. Whereas, the clinical sample was comprised of adults who were more
insecurely attached as evidenced by childhood abuse and neglect or ambivalent
attachment to one or both parents. The attachment literature indicated that
children who are securely or insecurely attached to caregivers are more likely to
develop a defensive style that perpetuates these styles of relating. The adults
from the normal group were able to react to the divorce with recognition of their
feelings of loss and grief, whereas the clinical group was not able to recognize
feelings of loss and denied being strongly affected by the departure of the
spouse.

Other important attachment differences became evident in the qualitative
analysis. The normal sample either described their parents in very positive
supportive terms or their parents were appraised with deﬁcits. All have an

ongoing connection with their parents at the present time. The normal subjects

163

approached the divorce from their partner with low conﬂict. Their effective
response to the loss of the marital partner included more feelings of loss and
sadness. The parents of the normal sample were more likely to recognize their
child’s independent relationship with the other parent and to not interfere.

The clinical sample’s families of origin included high incidence of parental
divorce, and occurrence of violent behavior. Within this sample, there was wider
variation in descriptions of parents that included negative qualities as well as
evidence of parental rejection. Two of the ﬁve clearly recognized that they were
not close to either parent while growing up, while the remaining three perceived
that they were close to at least one parent. Three of the ﬁve were estranged
from their fathers for several years, but all had some contact with their mother
into the present. The subjects from the clinical sample approached the divorce
with high conﬂict. The loss of the martial partner was presented, as occurring
with little difﬁculty or sadness yet the ongoing legal conﬂict with the other parent
was typiﬁed by intense feelings of anger and hostility. The clinical sample was
less able to identify with their child’s loyalty to the other parent and were more
likely to deﬁne the relationship with their child from a perception of how much the
child loved them.

Of signiﬁcance were the clear affective differences regarding the reaction
to divorce and difference in the level of conﬂict. The two groups displayed very
different reactions to the marital dissolution with the adults from the normal group
recognizing feelings of ambivalence and difﬁculty accepting divorce. The clinical

sample had a defensive style that indicated they had little difﬁculty coping with

164

the loss of the marriage partner. These ﬁndings regarding difference in affect
lend credence to the ﬁndings in Chapter IV. The quantitative results regarding
attachment style differences indicated that the clinical sample was more
dismissing and avoidant of the marital partner after marital separation. The
ﬁndings from the qualitative interview support this difference as well. The results
from the PSI provide some indication that more children in the clinical sub-group
were having behavioral difﬁculty whereas children of the normal sub-group were
having average or lower behavioral difﬁculty. This may suggest that the children
from the clinical sample already may be having attachment difﬁculties.

In sum, the qualitative data provide anecdotal support for several of the
issues investigated with quantitative methods. In addition they bring life to the
stories of these two groups of subjects by giving voice to some of their speciﬁc

experiences in negotiating the process of divorce.

165

Chapter 6

CONCLUSIONS AND IMPLICATIONS

This ﬁnal chapter begins with a restatement of the problem and a brief
review of the methodology. The main body of the chapter consists of a summary
of the ﬁndings by hypothesis. Because some hypotheses were conﬁrmed and
others were not, the summaries also include the researcher’s proposed
explanations. Following the summary section, the implications for future
research and implications for social work practice are addressed. The chapter

concludes with a brief ﬁnal synopsis of the study.

Restatement of the Problem and Methodology

This study developed from the provision of clinical intervention to families
in high conﬂict who were in litigation over custody or parenting-time issues. The
major thrust of the investigation was to determine if there were notable group
differences between a normal sample and a clinical sample of adults who
divorce, regarding their style of attachment, their level of conﬂict, and the non-
resident parent’s access to minor children. To answer this major question ﬁve
hypotheses were posed.

The study used a combination of quantitative and qualitative methods in
an effort to capture interpersonal differences between the two groups. A
questionnaire was developed and used with a total sample of 125 respondents

and an additional interview was completed with a small sub-sample of 10

166

persons that was drawn from both larger groups. This investigation was a
preliminary study using a non-random sample. As such, this study has
limitations regarding the generalizability of the ﬁndings to populations beyond the
county in which the study occurred.

Participants came from two different settings, persons attending a local
divorce orientation program and persons mandated into service with a local CMH
agency. The study was comprised of a normal sample of 63 and a clinical
sample of 62. The two groups were surprisingly comparable in regard to age,
gender, race, income, occupation, the number of children and the length of
marriage. From the demographics, it was determined that the average age of the
participant was 33.8 years with an average length of marriage of 8 years. The
majority of the participants were European-American, with a fairly small
representation by ethnic minorities. Years of education averaged one year
beyond high school and average income was quite modest, ranging between
$15,000-25,000 per year. As anticipated, more of the clinical group (31%)

compared to the normal group (4.8%) had divorces that were ﬁnalized.

Summary of the study

Some of the hypotheses in this study were conﬁrmed and others
were not. The following is an overview of the conclusions of the study organized
by the variables. Because Hypothesis l, and II, and V involved the measurement
of attachment they will be discussed together. The instrument used to measure

an individual’s overall level of attachment was not completely useful for

167

distinguishing secure and insecure attachment styles between the two sample
groups. On the ﬁrst instrument, only one of the three sub-scales proved to be
helpful in discerning sample difference. More of the clinical sample, when
compared to the normal sample indicated an avoidant pattern of relating. There
was a similar problem of lack of discrimination on the sub-scale of secure
attachment style in the second attachment measure. This second attachment
measure focused on evaluating a person’s reaction to the loss of the spouse.
Three of the four sub-scales did yield differences between the two sample
groups; however, only one was in the expected direction. The clinical group
indicated more dismissing reaction to the loss of a spouse, and the normal group
indicated more preoccupied and fearful reactions. Vlﬁth this instrument the items
that made up the sub-scale dismissing included responses of no reaction, which
seemed to coincide with the clinical sample’s effort to minimize difﬁculty, whereas
the two other sub-scales required some acknowledgement of their affective
response. In the study it appeared that the normal sample was more likely to
acknowledge personal feelings about the marital partner.

The results from the attachment instruments indicate several possible
problems with the measurement of attachment. The ﬁrst is the matter of external
reliability. As documented in the literature, Crowell and Treboux (1995) indicate
that each adult attachment instniment has a slightly different conceptualization of
the construct of “secure attachment”. For some researchers, secure attachment

is related to comfort with closeness and comfort with dependency (Collins &

168

Read, 1990), whereas for others it is symbolized by a positive view of the self
and a positive view of others (Bartholomew & Horowitz, 1991).

The second possible problem was that of lntemal reliability. As was noted
in Chapter III, both of the instruments used in this study had some problem with
reliability. The questions used to measure the security of attachment did not
have strong internal consistency between items for some scales and the scales
were further weakened by the removal of items. As a result of both of these
factors, it is possible that the use of other instruments may have better
demonstrated group differences in attachment style. Because each of the two
adult attachment instruments did display some differences between the normal
and clinical groups, it also suggests the possibility that other factors inﬂuenced
how the respondents approached responding to the attachment questions. More
speciﬁcally, it suggests that the clinical sample was either more highly defended,
or viewed themselves as having little difﬁculty in adult relationships, when in fact
serious interpersonal relationship problems existed. Conversely, the normal
sample may have been less defensive about admitting problems and approached
the instrument with an earnest attempt to respond according to their present
evaluation of themselves in relationship to the other spouse. The combined
ﬁndings in Hypotheses I and II resulted in the conclusion that the respondents in
the clinical sample were more likely to indicate an avoidant and dismissing style
of attachment whereas the respondents in the normal sample were more likely to
indicate preoccupied and fearful attachment. Because of the mixed results

Hypotheses l and II were only partially conﬁrmed.

169

That there were group differences in regard to defensive responding was
further indicated by group differences on the CTS and attachment history that
were revealed in the face-to-face interviews with the sub-sample. The sub-score
of verbal aggression was higher in the normal sample than it was for the clinical
sample for the CTS. Because it is counter intuitive that those in the normal
sample engaged in more verbal aggression than the clinical sample, the disparity
suggests a differences in willingness to expose negative behaviors and/or a
higher degree of defensiveness about admitting aggressive behavior.

Results from the face-to-face interviews supported the hypothesis that
there were sample differences regarding security of attachment to primary
caregivers and differences in response to reaction to loss of the marital partner.
The collective ﬁndings from the ﬁve respondents in the clinical sub-sample
indicated more insecure parent-child relationships while growing up as evidenced
by exposure to parents who did not emotionally protect or nurture them, more
instances of family violence and more descriptions of distance or rejection in the
relationship with one or both parents. In contrast, the collective ﬁndings from the
ﬁve respondents in the normal sample indicated more secure attachment in
parent-child relationships while growing up, more stable marriages of parents, no
reported incidences of domestic violence or child abuse, and descriptions of a
warm, emotionally responsive relationship with at least one and sometimes both
parents. The results from the interview also suggest that there are attitudinal and
behavioral differences regarding the reaction to the loss of the marital partner.

The clinical sub-sample was more likely to deny any emotional difﬁculty with the

170

termination of the marriage and to have lower expectations about how long it
would take to adjust. In contrast, the normal sample was able to discuss some of
their ambivalence or emotional struggle in letting go of the spouse. This
difference in reaction to loss of marital partner supports the concept that
individuals who were securely attached to their spouses at one point in the
marriage were able to express more of the accompanying affect that coexists
with divorce, whereas those who were insecurely attached had a stronger need
to deny the importance of the spouse and, consequently, deny having affective
responses to loss of a relationship.

This problem of measurement of adult attachment by self-report
instruments with individuals who have personality disorders has been touched
upon in the literature. Although there has been interest in using attachment
instruments with the clinical population, Patrick et al. (1994) recognized that
persons with personality disorders might not be good candidates for self-report
instruments. According to Crowell and Treboux (1995), self-report instruments
work well with individuals’ conscious feelings and perceptions about
relationships, but many individuals have limited direct awareness of their
attachment representations and strategies. In these authors’ opinion, narrative
techniques better access factors outside of the individuals’ awareness. Patrick,
et al. (1994), also supports the use of direct methods of observation in
combination with self-report instruments to discern attachment style. However,
they add that one of the challenges of direct observations is still establishing

criteria for measuring external behavior. Two recommendations for change would

171

be these: to include a behavioral measurement of children’s adjustment and to
more carefully conceptualize how to measure attachment styles with the clinical
population. Given some of the limitations of self-report instruments for
measuring adult attachment, it appears that the qualitative methods provide more
information when working with the clinical population. The AAI, as a reliable
method, continues to receive a high level of support as an attachment instrument
(Crowell & Treboux 1995).

Another issue found in the literature that may confound the measurement
of adult attachment is the possibility that adult attachment is relationship speciﬁc.
Baldwin and Fehr (1995) argue that one cannot conclude that adult style of
attachment exists across relationships. In their review of adult attachment
instruments, they assert that there is an inherent problem of test and re-test
reliability. In their opinion, the self-report measures are doing an adequate job of
deﬁning adult attachment but they are unconvinced that attachment style is
stable over time. They assert instead that attachment style is relationship
speciﬁc.

Certainly it is agreed that adult relationships are very complex and contain
many variables that inﬂuence one’s reaction to divorce. Each person brings to
relationships their own history of attachment which potentially inﬂuences the
marital relationship and adds to the difﬁculty of analysis. Bartholomew (1994)
proposes that each adult attachment style in a particular relationship is a
component of that relationship and suggests that attachment style may not be an

individual characteristic or behavior system that is stable across the life span.

172

As was indicated in Chapter IV, there was very strong support for
Hypothesis III, which measured level of conﬂict in the co-parental relationship.
There were clear differences in regard to how the normal and clinical samples
communicated as co-parents and ultimately how they negotiated their
differences. All of the factors that encompassed the variable “level of conﬂict“
were conﬁrmed as supporting the hypothesis. These were differences in the
description of communication, frequency of communication, style of negotiation
when communicating, resolution of custody and the sub-scales on the CTS.
These ﬁndings were expected bemuse, by deﬁnition, clinical persons have more
difﬁculty with interpersonal relationships. Based on clinical observation, they are
also less likely to depend on the socially acceptable use of words and are more
likely to depend upon more aggressive techniques for negotiating. However, the
ﬁndings are important because the results support the work of other researchers
regarding the effect of high conﬂict on child well-being (Ahrons & Miller, 1993;
Amato et al., 1995; Hanson, 1999; and Jekielek; 1997).

The lack of difference between the normal and clinical groups on
Hypotheses IV, which measured parenting time, came as a surprise. Based
upon clinical experience with high conﬂict families, one of the most frequent
complaints is lack of access to minor children. However, the results from this
study indicated no difference in the kind of custody arrangement and no
difference in regard to amount of parenting time. The overall amount parenting
time granted exceed expectations given the percentage of joint physical custody

(21 .6%) and liberal access parenting plans (23.2%).

173

Some possible explanations for the lack of difference in parenting time are
as follows. First, it is possible that the questions on the survey were not written
with enough speciﬁc detail to discern differences between parenting time granted
and parenting time used. A better approach to measuring parenting time would
be to inquire about the amount of missed visitation over a period of months and
to determine the cause of those missed parenting times. Although this study
considered the possibility of comparing mothers’ and fathers’ responses, there
was insufﬁcient matching of paired couples to enter into such an analysis.

A second possibility is that those parents without existing court orders are
most typical of parents in conﬂict, and because the inquiry was only made about
existing orders for parenting time, some parents in conﬂict may have been
overlooked. A possible third factor that may affect amount of parenting time is
how determined a non-resident parent was willing to ﬁght in court for more
parenting time. Support of the use of litigation as a means to increase parenting
time was investigated in an adjacent county. Fox and Kelly (1995), who looked
at how custody decisions get made, suggest that fathers who want physical
custody have to exert legal effort to strengthen their claims.

In this study, it is difﬁcult to conclude with certainty what the ﬁndings
regarding parenting time mean. It does suggest, however, whether by mutual
agreement or by court order, more non—resident parents are being granted
parenting time beyond every other weekend with a mid week contact. It is equally
difﬁcult to know whether the high number of shared custody arrangements found

in this study indicates a current trend or is an anomaly. This higher percentage

174

of shared custody arrangement will need to be conﬁrmed by other current
studies.

In Chapter I a model was proposed that described how the three major
variables in this study style of attachment, level of conﬂict in the co-parental
relationship interacted (See Figure 1). The ﬁndings, particularly from the
qualitative interviews, support the notion that security of attachment with parents
is carried forward in the degree of defensiveness with which one deals with loss.
It suggests that those with less secure attachment histories are more likely to
react to the loss of the marital partner with more of the protest and despair that
Weiss (1975) ﬁrst described. The model as written provides a basic framework
to think about generational patterns of attachment. To be more meaningful,
some additional information about the mate selection process would need to be
obtained from both partners in the relationship well as information about the
behavioral characteristics of the child. This application of this model seems best
suited to qualitative methods when applying attachment concepts to the clinical

population.

Implications for Research

The comparison of two samples of adults who experience divorce
provides some interesting ideas for future research. First, this study approached
the inquiry into co-parental relationship by considering some important tenets of
attachment theory as a method of explaining differences between couples in high

and low conﬂict. The study sought to determine if prior security of attachment

175

affects how adults cope with the loss of the spouse. However, to determine if
this is so, the model as proposed would need to be replicated.

This study only focused on group differences and there was no attempt to
match couples’ responses to each of the variables within the study. Although
there were some couples in both the normal and clinical samples, the results
were not analyzed by matched pairs because the numbers were insufﬁcient.
However, to do so in the future would add the dimension of gender and, perhaps,
more detail about attachment as it pertains to mate selection. It would also offer
the opportunity to compare perceptions of the relationship and the dissolution of
the marriage.

This study only looked at couples who were once married. However,
sociologically, many adults have children in common but do not marry. In the
clinical population from which the clinical sample was drawn study about 34% of
couples referred for services were unmarried partners. It would be interesting to
compare conﬂict in the co-parental relationship between the once married and
never the married parents. From clinical experience, it appears that never
married parents have more interpersonal conﬂict difﬁculty than do married
parents who are divorced. How couples interpret and deal with conﬂict according
to marital status would be another interesting study.

To broaden attachment theory’s application to adult attachment, more
knowledge is needed about the mate selection process. Although there are a
few studies that have looked at secure and insecure attachment with the college-

age population, less is understood about how adult attachment inﬂuences partner

176

selection, or how partner selection inﬂuences adult attachment. For example,
individual attachment style may be seen as a predictor of a couple’s relationship
satisfaction. Some current ﬁndings suggest that an adult with a history of secure
attachment is more likely to seek a relationship with another adult with a similar
style of attachment. Other studies suggest that some couples balance each
other’s insecure attachment style when one partner, for example, is avoidant and
the other is anxious. If more was understood about the partner selection
process, then more could also be understood about the decision to divorce as
related to secure or insecure couple attachment.

At the present time, there is no clear evidence as to what role gender
plays in attachment style. What became apparent in analyzing the qualitative
interviews was that females in both the normal and clinical sup-groups appeared
better able to recognize and talk about deﬁcits in the parenting they received.
The male subjects, particularly in the clinical sub-sample, appeared to be
defensive about exposing deﬁcits. Bartholomew and Horowitz (1991) did not ﬁnd
an association between gender and attachment style in their study. However,
the Brenner & Shaver (1998) study regarding attachment and personality
disorder suggests that certain diagnostic categories of personality disorders are
more prevalent with one gender than the other. The role of gender in the
development of attachment style is likely to be another interesting aspect of
attachment research for the future.

Within the last ﬁve years, there has been interest in determining the role of

security of adult attachment with the affect regulating process that accompanies

177

domestic violence (Brennan 8 Shaver, 1995). Dutton, Sanders, Starzomski, and
Bartholomew (1994) have also begun to explore domestic assault from the
perspective of males with an insecure attachment style. Just as the
measurement of attachment in infancy has inﬂuenced research regarding affect
regulation (Schore, 1994), it now appears that affect regulation will be addressed
in the adult population. Affect regulation, like lntemal models of representation,
may be another intergenerational pathway that affects how children are impacted

by parents’ prior experience with their own caregivers.

Implications for Practice

This study has many implications for practice. However, only three major
areas will be discussed. They are: presence of adults who met clinical criteria in
the population that divorces, attachment theory’s relevance to social work
practice and the need for clinical intervention with parents in high conﬂict.

This study began with the recognition that the general population contains
adults who meet clinical criteria. Although the prevalence rate varies from study
to study, a conservative estimate is that between 9.6% of males and 10.3 % of
females are diagnosed with at least one personality disorder (Maier,
Lichtennann, Klinger, Heun & Hallmayer, 1992). It does not seem coincidental
that the Maacoby and Mnookin (1992) and Johnston (1994) estimates from their
studies indicate that about 15% of all couples who divorce use excessive

litigation to resolve disputes and present to the court as couples in high conﬂict.

178

Although there was no attempt to correlate meeting clinical criteria and being in
high conﬂict, this study does suggest such an association.

When working with families who divorce, it is important to recognizing that
there is a proportion of parents with minor children who also have accompanying
personality disorders or psychiatric symptoms. Individuals with personality
disorders have different reactions to divorces, and how to help them and their
spouses and children is not well documented. Because adults with psychiatric
symptoms may not respond to loss of the marital partner with the same kind of
anger and sadness that others who divorce express, alternative methods of
intervention are needed to contain their affect. Adults with personality disorders
are more likely to engage in behaviors that express intense rage, and they are
more likely to project blame on the other parent and to have less conscious
awareness of how the conﬂict between them and the other parent is detrimental
to their children. Given the number of parents who have children in common and
the number of divorces that occur each year, it is important for the ﬁelds of social
work and psychology to establish specialized interventions for parents in high
conﬂict.

The second major implication for practice is to consider the origin of
interpersonal relationship difﬁculties that accompany some couples who divorce.
This study approached the problem of co-parental conﬂict as symbolic of other
kinds of experiences with rejection and abandonment in early childhood.
Attachment theory offers one explanation of how patterns of attachment get

transmitted from one generation to the other through that parent’s internal

179

working model. Attachment theory suggests that with intervention, lntemal
working models can be updated and altered when an individual is able to access
childhood experiences through memory. Attachment theory implies that early
intervention with parents of young children, when there are known attachment
difﬁculties, is necessary for prevention. The studies of neglected and abused
children indicate that disorders of attachment in childhood are connected to more
aggressive behavior problems in childhood and the development of personality
disorders in adulthood. Mc Milllan (1992) views attachment theory as extremely
valuable to the ﬁeld of clinical social work. In her view, attachment theory
provides an important theoretical framework for providing treatment for couples
and families.

When interpreting ﬁndings about the effects divorce has on children, one
needs to consider the interpersonal functioning of the parent prior to divorce as
part of that interpretation. As Amato et al. (1995) have clearly deﬁned in their
work, some children exposed to high conﬂict do better when parents divorce.
The ﬁndings regarding high conﬂict and the number of non-resident parents with
joint and liberal access (40% combined) suggest that many children are making
several transitions per month between two homes. Although a joint custody
arrangement may satisfy one or both parents, under conditions of high conﬂict,
joint custody may not serve the child’s emotional or psychological attachment
needs. Knowledge of how to intervene with parents in high conﬂict and how to
make recommendations regarding the minor children is especially relevant to the

ﬁeld of practice. This is particularly so since mental health professionals are

180

often relied upon to participate in legal proceedings regarding custody
determination.

Within the legal arena, custody can be used as a bargaining tool to reduce
parental conﬂict when, in fact, parents continue to have ongoing disputes and
have widely disparate views about how to parent. In discussing therapeutic
intervention from the perspective of the child, Johnson (1994), a clinician,
suggests that where there is concern about the capacity of both parents to
protect the child from interparental conﬂict and from their own disturbed attitudes
and behavior, it may be appropriate to consider custody by other supportive
caregivers. Under circumstances of ongoing parental conﬂict, Johnson
recognizes the need for the child to have access to their own counselor, who can
intervene on their behalf with both parents.

Johnson and Roseby (1997) describe in much clinical detail what happens
to children who are psychologically “caught in the middle” and are exposed to the
highly conﬂicted and violent divorce. The practice implications of children’s
symptomology related to divorce include the need to ﬁnd better ways of
intervening with parents in conﬂict. Mediation and family treatment are two
means of intervention. The experience and the knowledge of the mediator or
therapist are critical factors in helping divorcing parents with multiple issues. In
Kelly’s (1995) review of a decade of mediation research, she states:

More effective mediators intervene more quickly when conﬂict is
high; have greater communication competence, shape communication in

productive agreement oriented directions and focus on interests. Couples

181

with serious relationship problems are less likely to reach agreement,
particularly when there is no opportunity in mediation to deal with the
problem, or when the style of the mediator is to deal only with facts.
Agreement rates in custody disputes are affected by the number of and
severity of relationship problems, rather than the number of disputes

(p.382).

Synopsis of Study

This study began with the clinical recognition that not all couples who
divorce are able to emotionally disengage from their former spouse. Of the
couples who divorce there are a percentage of couples with children who engage
in high conﬂict behaviors. Divorcing couples in high conﬂict who use extensive
litigation also appear to be representative of the clinical sample with a prior
history of interpersonal relationship difﬁculty. This study approached the
research question of why couples are in high or low conﬂict by applying concepts
of attachment theory and adult attachment to the body of literature regarding
divorce. This approach was utilized because of the clinical knowledge that the
population of persons who divorce includes those who meet DSM IV diagnostic
criteria.

This was a preliminary study to explore how normal and clinical samples
differed regarding their style of attachment, level of conﬂict, and parenting time.
Both quantitative and qualitative methods were used. The quantitative results

strongly supported the ﬁnding that the two samples differed in expressed level of

182

conﬂict. Although the ﬁndings regarding attachment style had some mixed
ﬁndings for each sample, the qualitative interview supported the notion that the
clinical sample is more likely to come from an unstable family life and have
distant relationships to at least one parent. No differences were detected in
regard to the amount of parenting time granted to the non-resident parent. This
study has many implications for future research as to how clinicians should
approach working with parents who have children in common. Most importantly
it suggests that the adult attachment style and how an adult copes with divorce

merits further research.

183

APPENDICES

184

APPENDIX A

TABLES

185

 

 

 

 

 

 

 

 

 

 

Table 1 Gender, Age, and Years Married
Normal Clinical Statistic Probability
Gender n % n %
Male 30 47.6 29 46
Female 33 52.4 34 54
Chi Sq.=.032 n.s.
Age in s.d m s.d
32.723 A 34.953 5.67
t=2.10 *
Years Married m s.d m s.d
7.89 4.37 7.98 4
t=.1 3 n.s.
Number of Children m s.d. m s.d
1.7 0.71 1.84 1.13
t=.81 n.s.
Judgement Granted n % n %
Yes 3 4.8 22 36.1
No 60 95.2 39 83.9
Chi sq =18.8 *‘*
Child Support Ordered n % n %
Yes 29 46 45 73
No 34 54 17 27
Chi Sq.=9.11 *
Note: *p=<.05
“p=<.01
*“p=<.001

n.s.= not signiﬁcant

186

Table 2 Race, Income, and Education

 

 

 

 

 

 

Normal CliWcal
Race/Ethnicity f % f % Statistic Probabilim
Caucasian/White 48 78.3 44 81 .5
Afro American 7 12.1 1 1.9
Native American 2 3.4 4 7.4
Asian 0 0 1 1.9
Hispanic 1 1.7 4 7.4
Other 2 3.2 0 0 Cramer's
V= .30 n.s.
Income Normal Clinical
f % f %
Less than $5,000 7 11.1 12 19.0
$5,001 -1 5,000 7 11.1 14 22.2
$15,001-25,000 20 31.7 11 17.5
$25,001-35,000 8 12.7 10 15.9
$35,001-45,000 4 6.3 4 8.3
“5,001-55,000 6 9.5 4 8.3
$55,001-65,000 4 6.3 4 6.3
65,001 -75,000 1 1.6 1 1 .6
$75,001 -85,000 3 4.8 1 1 .6
$85,001 -95,000 1 1.6 1 1 .6
missing 2 3.2 1 1 .6
Mann Whitney
mean rank 66.83 57.25 U= 1596.50 n.s.
Education
f % - f %
Less High School 4 8.7 3 6.0
High School/GED 22 47 18 36.0
Some college/Assoc 1 1 25.9 19 36.0
B.AJB.S. 7 15.2 7 14.0
M.AJM.S. 2 4.3 3 6
mean s.d mean s.d
13.08 1.89 13.37 2.03 t= .74 n.s.
Note: *p=<.05
“p=<.01
*“p=<.001

n.s.= not signiﬁcant

187

Table 3 Distribution of Reasons Relationship Failed

 

 

 

 

 

 

 

 

Normal Clinical

n % n %
Differences in Parenting 20 32.8 24 40
Difﬁculty in Communication 44 72.1 51 85
Involvement in Another Relationship 28 45.9 30 50
Substance Abuse 7 11.5 14 23.3
Domestic Violence 5 8.2 20 33.3
Mental Illness 5 8.2 10 16.7
Other 20 32.8 13 21.7

 

188

Table 4 Custody Arrangement and Custody Satisfaction

 

 

 

 

 

 

 

 

Normals Clincals Statistic Probability

Custody n % n %
Mother Custody 35 62.5 30 53.5
Father Custody 2 5.6 1 1.8
Joint Custody 16 28.8 1 1 19.8
Split Custody 1 1.8 5 8.9
Other Custody 1 1.8 9 18.1

Cramer's V= .324 n.s
Satisfaction with custody n % n %
Very satisﬁed 20 32.8 6 9.7
Satisﬁed 20 32.8 9 14.5
Not sure 9 14.8 9 14.5
Dissatiﬁed 4 8.6 14 22.6
Very dissatisﬁed 8 13.1 24 38.7
mean 2.34 3.66 t=5.35 ***
Difﬁculty Resolving Custody n % n %
Very difﬁcult 5 7.9 36 58.1
Somewhat difﬁcult 7 11.1 15 24.2
Somewhat easy 24 38.1 8 12.9
Very easy 24 38.1 3 4.8

Mann Whitney

mean rank 83.93 40.43 U= 553.50 “*
Plans to Change_Custody n % n %
yes 13 21.7 34 60.7
no 46 76.7 21 37.5

Cramer's V = .421 *‘*
Parening‘ﬁme Used m s.d. m s.d

7.23 4.28 6.98 6.33 t-test=.180
Note: *p=5_.05
“pas .01
“*p= 5.001

n.s.= not signiﬁcant

189

Table 5 Commnication and Level of Conﬂict by Sample

 

 

 

 

 

 

 

 

Normal Clinical Statistic Probability
Communlation Pattem n % n %
Respequ talk 29 48.8 8 13.3
Occassional differences 20 32.3 5 8.3
Independent decisions 8 9.7 4 6.7
Frequent differences 6 9.7 20 33.3
Unwilling to speak 0 0 8 13.3
Restricted communication 1 1.6 14 23.3
Mann-Whitney
mean rank 41.7 82 U=630.00 ***
Communication Frequency n % n %
1-2 times per week 29 52.7 18 27.1
1-2 times per month 15 27.3 17 ' 28.8
rarely 5 9.1 13 22
do not talk 5 9.1 13 2
Mann Whitney
mean rank 48.34 66.04 U=118.05 **
Style of Negotiation n % n %
Listen to each other 29 49.2 6 10.5
Each give up something 17 28.8 6 10.5
Refuses to be available 6 10.2 23 40.4
One parent gives in 4 6.8 5 8.8
Verbal attacks, swearing 3 5.1 14 24.6
Physical attacks 0 0 3 5.3
Mann Whitney
mean rank 41.39 76.21 U=672.0 ***
Conﬂict Intensity n % n %
Decreased 35 58.3 16 28.1
Unchanged 13 21.7 19 33.3
Increased 12 20 22 38.6
Chi sq =11 .05 **
Conﬂict Tactics Scale m s.d m s.d t Test
Reasoning self 9.44 4.09 8.71 3.92 t=0.97 n.s.
Reasoning other 7.96 4.04 4.81 3.95 t=4.17 *
Verbal aggression self 11.17 6.6 7.00 6.26 t=3.43 *‘*
Verbal aggression other 14.59 7.89 14.52 9.19 t=0.05 n.s.
Violence self 0.53 1.19 0.58 1.35 t=0.21 n.s.
Violence other 1.98 4.02 4.02 7.83 t=1.75 *
Note: 'p=<.05
*‘p=<.01
“*p=<.001

n.s.= not signiﬁcant

190

 

 

 

 

 

9.8 some 2 359: . 902
:2: as. o :8 F a: o: R 9 89 3...; 232 mm .62.
:2: 2.0.. o :8 F 8.. o: 8 9 2082.8 2...; 22: 8 .05.
ca... 550:. F .3 no F 63 F 8.. 8.. R m :u.o_::ootouoE 2...; 0.9: mm 55500
go... 2.0.. o :3 F o: 8.» mm m 35: 2...? 29:2 mm 35:.

so... 352: n :3 F o: a: mu m 89m. :onogmcoo E< 96:2 0.9: on 53%
26. .992: 3 .FF «:3 N 8.. o: 3 9 «5:35.20 29> .:< 50:? Ease.— mm :3E<
so. 859 m .m. FF 23 m o: 0: mm NF >389 .omansw 9...; 29: av no.0
26. 2.0. x. :3 F o: 0: cu m.» .9552: :9:an0 2...; 0.9: mm :3
.so. «50.. m .9339. F o: 0: FN v Spa. .933 223 29:9 om >23.
26. 6963:: N . m 28 N o: 0: mm v ova: 62.2w 9...; 0.9: mm 50..

.cE.oz

6.580 .6830 once 5.2.5 522...". 3:9: 09:2: 3:9: 5:883 88 .85.. our: 95: 29:5

02 .o .353: 550 226.35 Fa on. 83..
.333285 anuwinam F0 agatonoo - o 0.9:...

191

APPENDIX B

Figures

192

320.35.".

®

2.58:8:

55:35.: 3.5.2.2....

 

 

0.0m
28 2.5
32>an

 

 

.

 

 

132—Eco
02.22.80

 

 

 

.

 

5.828
8a:

 

 

 

 

0.239.:
.52
25:39...

V

 

 

 

 

.

 

23582
no
25 .53.

 

 

 

 

 

FOE—boo
F0 .o>04

 

 

 

coocctcnxul.5co.uc.cc 85.3..

E253... .0 25

8.35:5 82.2.5

4.22.252

2.5.2.29. «c250 FE- .cocctcnxc 022.... 6.?- E2558? =3.- :0 «555-:- 39—330 .0 Soto a... .0 .cuos. AIM—5.2m—

193

 

 

.x. .wo.:__0 B
. 30.5.0 n.
..x. .9502 I
. _mE..oZ I

 

 

0E8...

 

 

 

 

 

 

 

 

 

 

 

   

 

 

 

 

   

 

   

 

  
 

 

 

 

 

 

 

  

 

 

388... 3:02.273. .6 5.3.3.50

a 2.5.“.

 

 

QF 0F 3 3 no no 3 3F 3F man 0.? e.._.8_c__o.u
F F F F v v F. S FF «F NF ._8.c__o.u
Nm QF m... mF Om 3 3 RF 5m FFF F.FF $5.502-
N F n F v m .V m cm F F :méozﬂ,
9.3:. 89mm 8......» 80.3 80.3 83m 80.? 08.8 80.8 08.2 08.3

. . -893» -533 -898 -Foo.m3 -833 -Foo.m$ -F.8 .mm» +8..me -.F8 .3 $53.3, o

- m

- oF

, mF

- on

mm

8

- mm

auapuodsau p aﬁauamad

194

Raw Scale Scores

100
9° ‘ 89
l
80 1
70 69 f
60 j
i 58
i
!
50 i 51
40 41 ‘
. 36 ;
30 !
: !
$ 23 \ i
20 ‘ J!!! ‘—-I-'_1£L 19 i
l / i
i 4,
4—2 .3 12 E
= I
l
o L ,_, F ,, r , ____ _— .0
Defensive Parental Distress Parent-Child Difﬁcult Child Total Stress
Responding Dysfunctional
Interaction

Parenting Stress - Normal Sample Informant:

 

 

 

 

 

 

 

 

 

 

 

t_ _____, , _-_.__.._ "ﬁ

119—" “Ii—03h + " 'KerséJA-t—‘M‘AriibeF—4‘ 'oén -l—Gregﬁ}

 

 

Figure 3: Parenting Stress Index Normal Sub-Sample

195

Raw Scale Score

80m

704

40

 

.... _._-.... _~_._.,_ .. .- w

20%

10~

 

Parentlng Stress - Clinical Sample Informant:

 

 

 

' l

Defensive Parental Distress Parent-Child Difﬁcult Child

Responding Dysfunctional
Interaction

 

 

 

E—O—WAlex +__-_ Renee +Jodiﬂ—O-“5—kyg +Cou‘rtney 1i

 

Figure 4: Parenting Stress Index Clinical Sub-Sample

196

 

83

Total Stress

 

APPENDIX C

LETTER TO PARTICIPATE IN DIVORCE ORIENTATION

197

Seventh Judicial Circuit of Michigan

Genesee County Friend of the Court

Jame Kama
Friend of the Court.
I"? ma IO ZCCO .5. i;

' 3 . I
. I
. g I ‘ ...I ’
I .

_ ,,..._-._....- .J

 

IAUQUSt 7. 2000

.m

,“~.om.-..

Case No.

John Doe
1234 Anywhere St.
Anytown. Ml 48532

Re: Divorce Orientation and Education Program
Dear MrlMs Doe:

The Judges of the Family Division of the Genesee County Circuit Court require divorcing parents

who have minor children to attend the Divorce Orientation and Education Program (see attached
order.)

Your scheduled date is:

Date: March 23, 1999
Time: 6:00 pm. - 8:00 pm.
Place: Willard P. Harris Auditorium - 3rd Floor
County Administration Building
1101 Beach St., Flint, MI 48502

In the event you are unable to attend the workshob listed above you can attend a video taped
session the first Friday of every month from noon to 2:00 p.m. at the same location. There is no
charge or fee. Children may not attend and there are no child care services avaiiable at this
time. A certificate of attendance by the Genesee County Friend of the Court will be placed in
your court records verifying that you did in fact participate in this program.

Divorce can be a time when both parents and children feel overwhelmed by the losses and
changes they are experiencing. Every family member must adapt to a new way of living. The
more parents know about divorce. the better they are able to cope with the changes and help
their children adjust. It is in the best interest of your child(ren) that you attend this program.
Please call Katrina Weier at 257-3300, ext. 2098 with any questions or concerns.

The Honorable Duncan M. Beagle

Family Court Judge
Attachments

198

APPENDIX D

Family Counseling Order

199

 

am or nouns FAMILY COUNSELING can so.:

 

ORDER

 

 

Court 'relqhone tlo. :

 

 

 

 

 

 

 

 

Plaintitt Detuudant
Dhone he. : rhone lo. :
I Attorney: Attorney:

 

 

 

 

Atasusiceotsudcurtheldinthecourthouuinthecityornint,mmtyen
the dayet

 

 

.mucomm

this utter having cone betcre the Court upon presentment ct this Order, the
subjectotvhichhubeenagradtobynspecuwcoumelaMitappeuingtotheOmrt
that the parties hereto are eligible {or referral through the Circuit Court taeily
Counseling Act, being Act lo. 1‘ of Public Acts of 1900 and that there exists a
controversy regarding custody and/or visitation and that said controversy nay reasonably

beresolvedasaresultotthisOrderandtheCourtbeingmlyadvisedinthepre-ises:
mmnumommmt:

Both parties in this latter are hereby type of service requested:
referred to:

(Please check one)
(Please check one)

— Children or Divorce Program -— Nation/Parenting Agree-ant
1102 Machin Road
257-3676 Assessment 5 Waluation tor

Physical Custody or Parenting Tine
-— Private Psychologist] Social worker

friend ot the Cast Therapist/Individual or Ianily

--— Other Agenq —— Divorce Orientation i lducation

_ Specialized Service to be
Determined

— 0th.:

 

A copy of this Order shall be given to the Family Court Administrator Office.

both parties are hereby instructed to contact the above service/organisation torthvith
to arrange their initial appointunt. Payment for counseling shall be nade wrsuant to
this Act and/or the ia-ily's third party insurance or the family's ability to pay. there
is a

lisit per fully, paid tor by the Cirmit Court, unless othervise
given additional mroval.

 

Next hearing date or report needed by

 

 

Date

 

PHIL! COURT.| owes

200

APPENDIX E

DIVORCE QUESTIONNAIRE

201

DIVORCE ORIENTATION QUESTIONNAIRE

 

Court Case Hunber

INSTRUCTIONS: PLEASE ANSWER ALL QUESTIONS. IE A QUESTION
DOES NOT DIRECTLY APPLY TO YOUR SITUATION, BRIEFLY EXPLAIN.

 

 

 

1) bars: 2) crrrlzrp cont:
3) YOUR EIRTEDATE: 4) YOUR SEX:
Male Female
5) CHECK ONE: European American __ Afro American _
'American Indian __ Hispanic _ Asian __ Other (specify)
Bi-racial _ (Specify)

 

6) Court ordered to participate? (please circle) Yes No

7) Have you previously received service £rcm.the Divorce

Program? (please circle) Yes No

8) DESCRIPTION OF FAMILY

Please list yourself, household.meﬂbers, and.ninor children

(include half siblings and stepchildren) affected.by your
court action.
FIRST RELATIONSHIP D.O.B.

 

9) number of children from this relationship:

 

10).At what age did you begin a relationship with the other
parent? Under 18 19-24 25-30 30-35 over 35

11) Number of years married:

12) When was the petition for divorce filed?

 

Month Year

13) Are you the plaintiff (person who filed)?
‘ Yes No

14) Has judgment been granted? Yes___No

15) If yes: month year

202

16)

17)

18)

19)
20)

21)

22)

Has child support been ordered?

 

Yes No

If yes, amount per week per child:

 

Child support is being paid: (please check)

Regularly on time
Occasionally late
Irregularly

Not being paid and in severe arrearage

What.is your occupation?

 

Pull-time: _Part-time: (please check)

.At the present time, my income is: (please check)

_____ Less than $5,000 _____ $55,001 - $65,000
__ $5,001 - $15,000 __ $65,001 - $75,000
__ $15,001 - $25,000 __ $75,001 - $85,000
__ $25,001 - $35,000 __ $85,001 - $95,000
__ $35,001 - $45,000 __ $95,001 - $105,000
_ $45,001 - $55,000 __ Over $105,000

Primary reason why relationship with other parent
failed, in order of importance to you:

(1 8 most important reason, 2 = second.most reason,
3 - third.most reason)

Difference in parenting values
Difficulty with communication
Involvement in another relationship
Substance abuse

Domestic violence

Mental illness

Other

 

203

23)

Using the guide above each answer, please describe how you

generally respond in relationships with other adults. Check
one box for each question.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SA Strongly Agree
.A _Agree
NS Not Sure
D Disagree
SD Strongly
Disagree
SA .A NS I) SD
(a) I find it relatively easy to get
close to others.
(b) I am not very comfortable having to
depend on people.
(c) I am comfortable having others depend
on me.
(d) I rarely worry about being abandoned
by others.
(e) I do not like people getting too
close to me.
(f) I am somewhat uncomfortable being too
close to others.
(9) I find it difficult to trust people
completely.
(h) I am nervous whenever anyone gets too
close to me.
(i) Others often want me to be more
intimate than I feel.
(j) Others often are reluctant to get as
close as I would like.
(k) I often worry that my partner(s) do
not really love me.
(1) I rarely worry about my partner(s)
Leaving me.
(m) I often want to merge completely with
others, and this desire sometimes
scares them away.

 

 

 

 

204

 

 

24) Please answer the following items that best des

cribe how

you are coping with divorce or separation by checking

either True or False:

(a) Divorce was the only option left.
We tried for a very long time to
keep the relationship alive.

(b) Being alone has not been that bad.
I am surprised how easily I have
adjusted.

.(c) Despite being very angry and upset
about the divorce/break-up, I find
myself constantly thinking about my
former partner. .

(d) On most days, I 51:111 cannot believe
that he/she has left.

(e) I am the kind of person who trusted
and depended upon my partner.

(f) It would be better if I just did not
ever have to see or deal with my
partner again.

(g) I find myself driving by my former
partner's residence, or calling and
just hanging up.

(h) Some days I want to reconcile.
Other days I am mean and cruel to
the other parent.

(1) Although I do not totally agree with
divorce/Splitting up, I can imagine

the future when living apart will

work out better for everybody.

(j) As a person, I can honestly say that
he/she did not contribute that much
to family life.

(k) Even though the relationship has
ended, I still have feelings of
jealousy and resentment about the

loss of my partner.

(1) Even though I have suffered a lot of
emotional pain and abuse, I cannot stop
thinking that if we had only tried

harder we would still be together.

205

True

False

25) Current living arrangements: (please check)

Still in same household with other parent
Single, living alone

Self/children

Living with relatives

Living with friends

Live together relationship

Remarried

26) How difficult has it_been, or do you expect it to be, to
resolve custody and parenting time with the other parent.
(please check the answer that best describes your situation)

Very difficult. Possible legal battle, with hearing and
trial, with lots of people involved.

Somewhat difficult. Have some partial agreements, but
differ strongly regarding custody or parenting time.

Somewhat easy. Some differences, but expect to settle
issues regarding child(ren).

Very easy. Have reached or expect to reach agreement with

the other parent without attorneys, or without much court
involvement by the court or Friend of the Court.

27) How many times have you been in court for a hearing
regarding your divorce/parenting action?
(please check one)

Many, more than 16-20+ times Actual number, if known
Frequently, 10-15 times‘
Occasionally, 5-9 times

Rarely, 1-4 times

Never, 0 times

 

28) How’many more times do you expect to be in Court to
resolve issues?

206

29) Which physical custody arrangement best describes your
situation? (please check one)

Mother has physical custody with reasonable rights to father
Father has physical custody with reasonable rights to mother

Joint physical custody with both parents active involvement
in child(ren) care

Mother has physical custody without regular contact by
father

Father has physical custody without regular contact by
mother

, Split custody, children divided between two households

Other custody arrangement

 

Both parents in household, custody not yet determined

 

30) Bow satisfied are you with this arrangement?

(please
check one)

Very satisfied

 

Satisfied

Not sure

Dissatisfied

Very dissatisfied

 

31) Do you have plans to change this in Court? ___Yes No

32) In an average month, what is the non-resident’s amount
of parenting time awarded by the Court? (please check one)

______ Joint custody up to 50% of available time

_____ Liberal access, between 6-10 days per month

Reasonable rights, 4 days per month

_____ Approximately 1-2 day visits per month with no overnights

_____ Supervised parenting by friend/family less than 8 hours
Both parents in same household

Other unusual circumstances, please explain:

 

There is no Court Order

207

33) In an average month, how'much does the non-resident
parent actually use their parenting time? (fill in all that
IP91!)

Full days per month and/or partial days

34) Describe the non-resident's pattern of contact:
(please check one)

Predictable - rarely misses or reschedules

Regular - some monthly adjustment but child(ren) generally
seen on a consistent basis

Unpredictable - frequent unexpected changes, cancellations,
or misses being with child(ren)

No schedule - parent calls when he/she wants to see the
child(ren)

35)‘Which of the following situations best describes your
communication with the other parent? (please check one)

We are able to talk respectfully face to face or by phone,
and work out a plan to decide who is going to do what
regarding the child(ren), and are usually successful

We make an effort to communicate regarding some of our
child(ren)'s basic needs, such as medical care and school
events, but have occasional differences

We occasionally talk to each other but when we do not
agree, we each make independent decisions

We have frequent differences, rarely get issues resolved by
talking, one or both of us gets very angry or upset

One or both parents is/are unwilling to speak to other
parent, we use child(ren)/others as go between

Restricted communication, or no communication, one of us has

a P.P.O. or court order that forbids telephone contact
access regarding the child(ren)

36) How often do one of you attempt to communicate in
writing or by telephone regarding issues affecting or

related to the children in a typical month? (please check
one)

Often, 1-2 times per week
Somewhat, 1—2 times per month
Rarely, less than 1 time per month

Never, we do not talk at all

208

37) Which of the following styles best describes the way
decisions get.made between you and the other parent:
(please check one)

Parents listen to each other, make a decision based upon the

reasonableness of the situation and needs/wishes of the
child(ren)

Both parent give up some of what they want in order to reach
a conclusion

One parent refuses to make his/her self available, does not
commit to a decision, refuses to discuss issues

One parent decides, and the other parent gives in
Verbal attacks, name calling, put downs, swearing

Physical attacks or verbal threats of harm

38) Compared to when you and the other parent were together,
how would you rate the level of verbal conflict with the
other parent now? (please check one)

Level of conflict has decreased, less frequent, less intense

Level of conflict is unchanged

Level of conflict has increased, more intense and frequent

39) Describe the home atmosphere for the year prior to the
physical separation: (please check one)

Frequent arguments, continual hostility, negative tense
atmosphere

Intermittent periods of calm with eruptions in emotions
Indifferent, left each other alone, led separate lives
Relatively calm, lived as a typical family

Never lived together

209

40) How would you describe the other parent's ability to
care for the child(ren)? (please check one)

Very capable, no concerns regarding the ohi1d(ren)'s well
being while in his/her presence

Capable parent with most aspects of childcare

Noticeable inadequacies regarding ability to relate to
child(ren), but meets children's basic needs

Very poor, apparent inadequacies, I worry about child(ren)'s

health, safety, and parent's ability to protect child from
harm when in their care

 

Has never parented child, unable to answer question

41) For the next two questions use this guide to select your
answer: '

Close, share a positive relationship that is satisfying to both

Friendly, occasional difference in opinions and feelings between
child and parent

Neutral, share in activities, but do not discuss feelings

Distant, uncomfortable, have difficulty communicating

.A. MOTHER’S relationship to children (check one description
for each child)

 

CLOSE FRIENDLY NEUTRAL DISTANT

 

Oldest child
Second child
Third child
Fourth child

 

 

 

 

 

 

 

 

 

 

B. FATHER'S relationship to children (check one description
for each child)

 

CLOSE FRIENDLY NEUTRAL DISTANT

 

Oldest child
Second child
Third child
Fourth child

 

 

 

 

 

 

 

 

 

 

210

42)‘While living together, check the percentage that.best

describes how child care tasks were shared.between.parents
during an average week?

FATHER 100% 75% 50% 25% 0%

 

Physical Care
Supervision of
Children
Discipline
Medical
Education
Leisure/Play/
Recreation

 

 

 

 

 

 

 

 

 

 

 

 

 

MOTHER

 

Physical Care
Supervision of
Children
‘Discipline
Medical
Education
Leisure/Play/
Recreation .1

 

 

 

 

 

 

 

 

 

 

 

 

 

211

43) Please check if any of the following problemm has
affected you or the other parent in the preceding year:

(Eggicnmutnmumr
(a) Problems getting along "““—‘—'
with parents, child,
sibling, other parent

 

(b) Problems with friends,
coworkers, neighbors,
support system

(c) .Problems with education

(d) Problems at work

 

(e) Problems with housing

(f) Problems getting access
to health care

 

(9) Problems with legal
system/crime

 

(h) Other social or environmental
problems

 

(1) Been a victim of domestic violence

 

(j) Been investigated for child abuse
or neglect

 

(k) Been dependent on alcohol,
or other substances

 

(1) Had a Personal Protection
Order filed against them

 

212

44) Are you currently under a doctor' s care for a serious

medical or mental condition? Yes No

 

45) For what condition(s) are you being seen? (circle all

that apply)
(1) Heart condition (9) Anxiety
(2) Diabetes (10) Depression
(3) Back problems (11) Addictions
(4) Cancer (12) Other

(5) Nervous condition

(6) Chronic pain

(7) Digestive/bowel problems
(8) Surgeries

 

 

 

46) ' Have you ever been court ordered to receive mental
health services for yourself? __ Yes No

47) If ygg, by whom? (e.g., judge, probation officer, etc.)

48) Have you ever voluntarily chosen to participate in
outpatient counseling? Yes No

If yes: Approx. number of sessions

Place/Agency Year

 

 

 

 

 

 

49) If you and the other parent do not agree on
custody/visitation, which service do you think would be
needed to resolve the situation? (please check one)

Litigation in court, or other hearing or trial

Referee hearing

Referral to mental health professional/expert for evaluation

Negotiation between attorneys
Use of Friend of the Court
Mediation

Other

 

213

Thank you for your help in completing this questionnaire;
Professionals working with families in transition

interested in learning more about family relationships .

would you be willing to participate in an additional hour

interview? If selected, you would be compensated for you:
time .

tour responses would be confidential and would be
used for research, and not be a part of your court case or
your medical record.

are always

If so, please indicate your willingness to be contacted:

 

Address:

 

Phone number:

 

Person we can contact, and number we can call if you move:

 

214

APPENDIX F

FAMILY RELATIONSHIP SURVEY

215

Family Relationship Survey

Say something about my role not as a clinician but as a researcher trying learn
more about what happens to adults when they experience divorce. We will be
covering three areas, the family you grew-up in, the marriage that ended and
your relationship with your children now.

Tell me about the family you grew up in. Where did you grow up? How many
siblings did you have?

Where did you ﬁt in by age with your siblings? Were you raised by anyone
other than your biological mother and father? Do you consider them your parent
now? ,

Describe your relationship to your mother as a young child? As a school-age
child? As a teenage?

What made it close?

What made it distant?

What four adjectives do you think describe her as a person?
A.
B.
C.
D.

Describe your relationship to your father. As a young child? As a school-age
child? As a teenager?

What made it close?

What made it distant?

What four adjectives do you think describe him as a person?

new»

216

How accepted did you feel by your mother?

Did you feel she had your best interests in mind?
How accepted did you feel by your father?

Did you feel she had your best interests in mind?

How did your mother and father spend time with you? About how much time
would say each week? Was that amount of time satisfactory to you?

When you think about very happy or sad events growing up, what major life
events impacted you? ‘

Tell me about any separations or losses you experienced as a child.
Deaths of important people?

Moves to new towns?

Health problems or physical injuries?

What was your parent’s method of child rearing/discipline?

Why do you think they choose that method?

How has your relationship with your parents changed over time?
How easy is it to talk to your mother now? Your father?

What parts of your upbringing would you like to repeat for your children?
What parts do you hope to do different?

How did your family experiences shape your family?

Describe your relationship with each of your children?

On a scale from one to ten with ten being very strict and one being very
permissive, where do you fit between strict and easy going?

Do you feel you can effect your son or daughter development and how they turn
out? What might make it possible to be more influential?

217

Tell me about your relationship to your wife/husband when you ﬁrst met.

How did the relationship change over time?

In your opinion what lead to the breakdown in your marriage?

What things do you think added stress to the relationship?

What blame to place on him or her. What blame do you place on yourself?
Who do you think most wanted out of the relationship?

On a scale from one to ten how difﬁcult has it been for you to cope with the loss
of the marriage partner with ten being very difficult and one being very very
easy.

What made it difﬁcult, what made it easy.

How long will it take or how did it take to emotionally get over your current
feelings regarding your divorce? Give me an estimate in amount of time?

What tells you, your are over your husband/wife?

Is there anything else you would like to share with me that I haven’t asked you
about that would help me understand more about your divorce?

218

APPENDIX G

UNIVERSITY COMMITTEE ON RESEARCH INVOLVING HUMAN SUBJECTS
APPROVAL

219

MICHIGAN STATE

UNIVERSITY

January 14. 2000

TO: Rena HAROLD
254 Baker Hall

RE: IRB # 98285 CATEGORY: 2-F
TITLE: DIVORCING PARENTS: A STUDY OF LEVEL OF CONFLICT AND AMOUNT OF

PARENTING TIME
ANNUAL APPROVAL DATE: May 25. 1999
REVISION REQUESTED: October 6. 1999

REVISION APPROVAL DATE: January 14, 2000

The University Committee on Research Involving Human Subjects' (UCRIHS) review of this project is
complete and I am pleased to advise that the rights and welfare Of the human subjects appear to be
adequately protected and methods to obtain Informed consent are appropriate. Therefore, the
UCRIHS APPROVED THIS PROJECT'S REVISION.

This letter approves the additional Instruments and audio-taped Interviews.

RENEWALS: UCRIHS approval ls valid for one calendar year, beginning with the approval date
shown above. Projects continuing beyond one year must be renewed with the green renewal form.
A maximum of four such expedited renewal are possible. Investigators wishing to continue a
project beyond that time need to submit it again for a complete review.

REVISIONS: UCRIHS must review any changes in procedures involving human subjects. prior to

UCRIHS Chair. requesting revised approval and referencing the project's IRB# and title. Include in

your request a description of the change and any revised instruments, consent forms or
advertisements that are applicable.

PROBLEMS/CHANGES: Should either of the following arise during the course of the work. notify
OFFICE OF UCRIHS promptly: 1) problems (unexpected side effects, complaints, etc.) involving human
subjects or 2) changes in the research environment or new lnfonnation Indicating greater risk to the

RESEARCH human subjects than existed when the protocol was previously reviewed and approved.

If we can be of further assistance. please contact us at 517 355-2180 or via email:
GRADUATE UCRIHS@pilot.msu.edu.

STUDIES

 

Michigan State University

245 11¢me Buildim David E. Wright. Ph.D.

EastLansing. Michigan UCRIHS Chair
48824-1046
517/355218)
FAX 517/353—2976
Web: www.mucdulumrlwihs
E-Mall. ansueOu DE , d
: Mary Campbell
2296 N. McKinley
Flushing, Ml 48433

220

APPENDIX H

LETTER OF CONSENT NORMAL SAMPLE

221

LETTER OF CONSENT

As a participant in the Genesee County Divorce Orientation Program, you
are being asked to participate in a study. The purpose Of the study is to
determine what is different between those parents who are able to communicate
and reach agreement regarding the process of divorce from those parents who
are in high conﬂict, and make frequent use of the Family Court. The information
collected may help determine, in the future, which families may beneﬁt from
additional services during the divorce process, and what kind of service best
helps families come to agreement.

As a participant, we will be asking you to ﬁll out a questionnaire during
today’s presentation. Your commitment will involve approximately ﬁfteen to
twenty minutes Of your time. Your participation is entirely voluntary, you may
choose not to participate, or choose not to answer a particular question. You can
withdraw your participation at any time.

The ﬁndings in the information will not contain identiﬁable information
about you and/or your children. The information collected will be conﬁdential.
The information collected will be summarized by the group who complete the
questionnaire, rather than by families. Names will be changed tO an assigned
number. The information disclosed in the questionnaire will not be given to any
other agency without your written consent. You understand that your
participation in this study will not involve any ﬁnancial costs to you or effect the
outcome of your involvement with the Court

I have read the above letter of consent, and my signature on this letter
indicates my voluntary agreement to participate

I am interested in Ieaming about results of the study. For the contact
person, please contact Mary L. Campbell, A.C.S.W., C.S.W., A.B.D.,
Divorce Program Supervisor, 1102 Mackin Road, Flint, Michigan
48503 (810) 257-3676.

 

 

Name Date

 

 

Witness Date

222

APPENDIX I

LETTER OF CONSENT CLINICAL SAMPLE

223

JUBTTEEIIJF CINNSERTP

.As a participant in the Children of Divorce Program, you
are being asked to participate in a study. The purpose of the
study is to determine what is different between those parents
who are able to communicate and reach agreement regarding the
process Of divorce from those parents who are in high conflict,
and make frequent use of the Family Court. The information
collected may help determine, in the future, which families may
benefit from additional services during the divorce process, and
what kind Of service best helps families come to agreement.

.As a participant, you will be asked to fill out a
questionnaire prior to service delivery. Your commitment will
involve approximately fifteen to twenty minutes Of your time.
Your participation is entirely voluntary, you may choose not to
participate. If you dO'not participate, it will not effect the
outcome Of your treatment or the outcome of court litigation.

The findings in the information will not contain
identifiable information about you and/or your children. The
questionnaire will be confidential and.will not be a part of
your medical record. The information collected will be
summarized by the group who complete the questionnaire, rather
than by families. Names will be changed to an assigned number.

You understand that your participation in this study will not
involve any financial costs to you.

I have read the above letter Of consent, and my

signature on this letter indicates my voluntary
agreement.

I am interested in learning about results Of the study.
For the contact person, please contact Mary L. Campbell,
.A.C.S.W., C.S.W., A.B.D., Divorce Program Supervisor,
1102 Mackin Road, Flint, Michigan 48503 (810) 257-3676.

 

 

Name Date

 

 

Witness Date

224

APPENDIX J

CONSENT TO AUDIO TAPE

225

CONSENT FOR USE OF AUDIOTAPE AND INTERVIEW

Thank you for volunteering to participate in an
additional interview. The interview will consist of
questions about growing up with your parents, questions
about details of your divorce, and questions about you and
your children. .Along with the interview, you will be asked
to fill out three questionnaires that measure conflict, ’

parenting stress, and how you generally are in
relationships.

The information you share about yourself will be
confidential. It will not effect the outcome of your legal
case, nor will it effect the outcome of your participation
in the Children of Divorce Program. The information
collected will not become a part of any medical record.

Information that you share may be used to explain the
findings of the previous questionnaire. An effort will be
made to disguise the identity of you and your family, but
because of unique features of your particular story, total
anonymity may not be possible. Your name will not be used.

My signature indicates my agreement to be audiotaped.

My signature indicates my willingness to complete
three additional questionnaires.

You will be compensated for your time upon completion
of the interview.

 

 

Signature Witness

 

 

Date Date

226

Bibliography

(1994). Diagnostic and Statistical Manual of Mental Disorders. (Fourth ed.).
Washington, DC: American Psychiatric Association.

Abidin, R. R. (1995). Parenting Stress Index. (3rd ed). FL: Odessa:
Psychological Assessment Resources.

Ahrons, C., & Miller, R. (1993). The effect of the postdivoroe relationship
on paternal involvement: A longitudinal analysis. Journal of American

Orthogsychiatg, 63(3), 441-50.

Ainsworth, M. (1989). Attachments beyond infancy. American
Psychologist, 44(4), 709-716.

Allison, P., & Furstenberg, F. (1989). How marital dissolution affects
children: Variations by age and sex. Developmental Psychology, 25(4), 540-549.

Amato, P., & Rezac, S. (1994). Contact with nonresident parents,
interparental conﬂict, and children’s behavior. Journal of Family Issues, 15(2),
191-207.

Amato, P., Loomis, L., 8. Booth, A. (1995). Parental divorce, marital
conﬂict, and offspring well being during early adulthood. Social Forces, 73(3),
895-915.

Ammanti, M. (1991). Maternal representations during pregnancy and early
infant-mother interactions. Infant Mental Health Journal 12(3), 246-255.

 

Arendell, T. (1992). After divorce: Investigations into father absence.
Gender and Society, 6(4), 562-586.

Bakermans-Kranenburg, M., & van lJzendoom, M., H. (1997). Adult
attachment and the break-up of romantic relationships. Journal of Divorce &
Remarriage, 27(3/4), 121-1 39.

Baldwin, M., W., & Fehr, B. (1995). On the instability of attachment style
ratings. Personal Relationships, 2, 247-261.

Barber, 8., & Eccles, J. (1992). Long-term inﬂuence of divorce and single
parenting on adolescent family-and work related values, behaviors and

aspirations. Psychological Bulletin, 111(1), 108-126.

227

Barling, J., O'Leary, K, Daniel, Jouriles, D., Vivian, & MacEwen, K, E.
(1987). Factor similarity of the Conﬂict Tactics Scales across samples, spouses,
and sites: Issues and implications. Journal of Family Violence, 2(1), 34—54.

Bartholomew, K, & Horowitz, L., M. (1991). Attachment styles among
young adults: A test of a four-category model. Journal of Personality and Social
Psychology, 61(2), 226—244.

 

Bartholomew, K. (1994). Assessment of individual differences in adult
attachment. Psychological Inguiry, 5(1), 23—67.

Bing-Hall, J., & Stevenson-Hinde, J. (1991). Attachment relationships
within a family system. Infant Mental Health Journal 12(3), 187-200.

 

Black, A., & Pedro-Carroll, J. (1993). Role of parent-child relationships in
mediating the effects of marital disruption. Joumal of American Academy of Child
& AdLolescent Psmhiatm, 32(5), 1019-1 027.

Bray, J. H., & Heatherington, E. M. (1993). Families in Transition:
Introduction and Overview. Journal of Finilv ngchoquv. 7(1), 3-8.

Brennan, K, A., & Shaver, P., R. (1995). Dimensions of adult attachment,
affect regulation, and romantic relationship functioning. Society for Personally

and Social Psychology, 21(3), 267—83.

Brennan, K. A., 8. Shaver, P. R. (1998). Attachment styles and personality
disorders: Their connections to each other and to parental divorce, parental
death, and perceptions of parental caregiving. Journal of Personality, 66(5), 835-
878.

Bridges, L., Connell, J., & Belsky, J. (1988). Similarities and differences in
infant-mother and infant-father interaction in the strange situation: A component
process analysis. Developmental Psychology, 24(1), 92-100.

Buehler, C., & Legg, B. (1992). Selected aspects of parenting and
children's social competence post-separation: The effects of child's sex, age, and
family economic hardship. Journal ofﬂorce and Remarriage, 18(3-4), 177-195.

Camara, K, & Resnick, G. (1989). Styles of conflict resolution and
cooperation between divorced parents: Effects on child behavior and adjustment.
Journal of American Orthopsvchigtrv. 59(4), 560-575.

Campos, J., Barrett, K, Lamb, M., Goldsmith, H. H., & Stenberg, C.

(Eds.). (1983). Socioemotional Development (4th ed). (Vol. Infant
Developmental Psychobiology, II). NY: John Wiley & Sons.

228

Cherry, A. L, Jr. (2000). A reseimﬂmwor the helping professions:
Methods, statistics and writing. Belmont,: Brooks/Colefl’homas Learning.

Chomesky, A. (1992). Updating paternal images in fathers-adult son
relationships. Smith Qollpge Studies in Social Work, 62(2), 169-87.

Cohn, D., A., Silver, 0., L., Cowan, C., P., Cowan, P., A., & Pearson, J.
(1992). Working models of childhood attachment and couple relationships.
Journal of ﬁmilv Issues 13(4), 432-449.

 

Collins, N., L., & Read, 8., J. (1990). Adult attachment, working models,
and relationship quality in dating couples. Journal of Personality and Social
Psychology, 58(4), 644-663.

Cowan, P. A., Cohn, D., A., Cowan, C. P., & Pearson, J. (1996). Parent's
attachment histories and children's externalizing and internalizing behaviors
Exploring family systems models of linkage. Journal of Consulting and Clinical

Psychology, 64(1), 53—63.

Crandell, L., Fitzgerald, H., & Whipple, E. (1997). Dyadic synchrony in
parent-child interactions: A link with maternal representations of attachment
relationships. Infant Menta_l Health Journal 18(3), 247-264.

 

Crittenden, P. (1990). Internal representation models of attachment
relationships. mm Mental Health Journal 11(3), 259-277.

 

Crittenden, P., M., Partridge, M., F., & Claussen, A., H. (1992). Family
patterns of relationship in normative and dysfunctional families. Development and

Psychopathology, 3, 491-512.

Crowell, J. A., & Treboux, D. (1995). A review of adult attachment

measures: Implications for theory and research. Social Development, 4(3), 294-
327.

Davidson, C. E. (1991). Attachment issues and the cocaine exposed dyad.
Child and Adolescent sml Work. 8(4), 269-294.

Diaz, R., M., Neal, 0., J., & Vachio, A. (1991). Maternal teaching in the
zone of proximal development: A comparison of low-and high risk dyads. Merrill-

Palmer Quarterly,, 37(1), 83-108.

Dutton, D., G., Sauders, K, Starzomski, A., & Bartholomew. (1994).
Intimacy-anger and insecure attachment as precursors of abuse in intimate

relationships. Journal of Applied Psychology, 24(15), 1367-1386.

 

229

Fagot, B. I., & Kavanaugh, K (1990). The prediction of antisocial behavior
from avoidant attachment classiﬁcations. Child Development, 61, 864-873.

Fish, B. (1993). Meaning and attachment in mothers and toddlers. Child
and Adolescent Social Work Journal, 10(3), 177-188.

Fonagy, P., Steele, M., Steele, H., Moran, G., & Higgett, A. (1991). The
capacity for understanding mental states: The reﬂective self in parent and child
and its signiﬁcance for security of attachment. Infant Mental Health Journal
1_2_(3), 201-218.

 

Fonagy, P., Leigh, T., Steele, M., Steele, H., Kennedy, R., & Mattoon, G.
(1996). The relation of attachment status, psychiatric classiﬁcation, and response

to psychotherapy. Journal of Consulting and Clinical Psychology, 64(1), 22-31.

Fox, 6., & Kelly, R. (1995). Determinants of child custody. Journal of
Marriage and the Familv. 51, 693-708.

Fraenkel, J., R., & Wallen, N., E. (1996). How to design and evaluate
research in education. (Third ed.). New York: McGraw-Hill, Inc.

Furstenberg, F., Jr., Morgan, S. P., & Allison, P. D. (1987). Paternal
participation and children’s well-being after marital dissolution. American
Sociological Review, 52, 695-701.

Garbarino, J. J. ,( January 26, 1996). A Review of the Psychometric
Properties of Selected Attachment Instruments. Paper presented at the Annual
Meeting of the Southwest Educational Research Association, New Orleans, LA.

Garvin, V., Kalter, N., & Hansell, J. (1993). Divorced women: Individual
differences in stressors, mediating factors, adjustment outcome. American
Journal of OrthopsvchLaLrv. 63(2), 232-240.

Goodman, S., & Brumley, H. E. (1990). Schizophrenic and depressed
mothers: Relational deﬁcits in parenting. Developmental Psychology, 26(1), 31-
39. .

Gray, 0., & Shields, J., (1992). The development of an instrument to
measure the psychological response to separation and divorce. Journal of

Divorce and Remarriage, 17(3/4), 43-56.

Greif, G. (1995). When divorced fathers want no contact with their
children: A preliminary analysis. Journal of Qvorce & Remarriage, 23(1/2), 75-85.

 

230

Griffin, 0., & Bartholomew, K (1994). Models of the self and other:
Fundamental dimensions underlying measures of adult attachment. Journal of
Personalig and Social Psvcholomr. 67(3), 430-445.

Hanson, T. L. (1999). Does parental conﬂict explain why divorce is
negatively associated with child welfare? Socia_l Forces. 77(4), 1283-1315.

Hartman, C., & Reynolds, D. (1987). Resistant clients: Confrontation,
interpretation, and alliance. Social Casework: The Journal of ContemporaLy
Social Work 68(4), 205-213.

 

Hazan, C., & Shaver, P. (1987). Conceptualizing romantic love as an
attachment process. Journal of Personality and Social Psychology, 52, 511-524.

Healy, J., Malley, J., & Stewart, A. (1990). Children and their fathers after
parental separation. American Journal of Qrthopsychiatg, 60(4), 531-543.

Heatherington, E. M. (1989). Coping with family transitions: Winners,
losers, and survivors. Child Development. 60, 1-14.

Heatherington, E. M. (1993). An overview of the Virginia longitudinal study
of divorce and remarriage with a focus on early adolescence. Journal of Fa_milv

Psycholggy, 7(1), 39-56.

Heatherington, E. M., 8. Stanley-Hagan, M. (1997). The effects of divorce
on fathers and their children. In M. E. Lamb (Ed), The Role of the Father in Child
Development . New York: John \Mley 8. Sons, Inc.

Hill, E. M., Young, J., P., & Nord, J., L. (1994). Childhood adversity,
attachment security, and adult relationships: A preliminary study. Ethnolpgy and

Sociobiolpgy, 15, 323-338.

Hoffman, C. (1995). Pre-and post-divorce father-child relationships and
child adjustment: Noncustodial fathers’ perspectives. Journal of Divorce and

Remarriage, 23(1/2), 3-20.

Holloway, S. D., & Machida, S. (1991). Child-rearing effectiveness of
divorced mothers: Relationship to coping strategies and social support. Journal

of Divorce and Remarriage, 14(3-4), 179-201.

Izard, C., Haynes, O. M., Chisholm, G., & Baak, K. (1991). Emotional
determinants of infant-mother attachment. Child Development. 6_2_, 906-917.

Jekielek, S. M. (1998). Parental conﬂict, marital disruption and children's
emotional well-being. Social Forces 76(3), 905-35.

 

231

Johnson, J. R., & Campbell, L. E. G. (1988). lmpasses of divorce: The
gynamics and resolution of family conﬂict. NY: The Free Press.

Johnson, J., Kline, M., & Tschann, J. (1989). Ongoing postdivoroe conﬂict:
Effects on children of joint custody and frequent access to children. Journal of

American Orthopsychiatm, 59(4), 576-592.

Johnson, J., & Campbell, L. (1993). A clinical typology of interparental

violence in disputed-custody divorces. American Journal of 'Orthopsychiag,
63(2), 190-199.

Johnson, J. R. (1994). High-conﬂict divorce, Future of Children (Vol. 4,):
Center for the Future of Children, The David and Lucile Packard Foundation.

Johnston, J., R., & Roseby, V. (1997). In the name of the child: A

gevelomental approach to understanding and helping children of Qnﬂicted and
violent divorce. New York: The Free Press.

 

Jones, E. (1996). Introduction to the special section on attachment and
psychopathology: Part I. Journal of Consulting and Clinical Psychology, 64(1), 5-
6.

Kalter, N., Kloner, A., Schreier, S., & Okla, K. (1989). Predictors of

children’s postdivoroe adjustment. American Journal of Orthopsychiagy, 59(4),
605-617.

Karen, R. (1994). Becoming attached: unfolding the mysterv of the infar_1_t_-
motherpond and its impact on later life. New York: Warner Books, Inc.

Keith, S. (1997). Two dimensions of adult attachment: Further
validation. Journal of Social and Personal Relationships, 14(1), 133-143.

Kelly, J. (1993). Current research on children’s post divorce adjustment:
No simple answers. Family and Conciliation Coarts Review. 31(1), 29—49.

Kelly, J. B. (1994). The determination of child custody, The Future of
Children (Vol. 5, pp. 121-142): Center for the Future of Children, The David and
Lucile Packard Foundation.

Kelly, J. B. (1996). A decade of divorce mediation research: Some
answers and questions. Family and Conciliation Courts Review. 34(3), 373-385.

Kessler, R. C. (1994). The national comorbidity survey of the United
States. International Review of Psvchiatm, 6, 365-376.

232

Kessler, R. C., Davis, C. G., 8. Kendler, K S. (1997). Childhood adversity
and adult psychiatric disorder in the US National Comorbidity Survey.

Psycholggical Medicine, 27(5), 1 101-11 19.

Kessler, R. C., Walters, E. E., 8. Forthofer, M. S. (1998). The social
consequences of psychiatric disorders, III: Probability of marital stability.

American Journal of Psychiatg, 155(8), 1092-1096.

Kitzmann, K, & Emery, R. E. (1994). Child and family coping one year
after mediated and litigated child custody disputes. Journal of Family Psychology,
8(2), 150-159.

Kraemer, G. (1992). A psychobiological theory of attachment. Behavioral
and Brain Sciences 15, 493-541.

 

Ladd, L., D., 8. Zvonkovic, A (1995). Single mothers with custody
following divorce. Marriage and Famihr Review,g_g(1a), 189-210.

Lamb, M. E. (1974). A defense of the concept of attachment. Human
Development, 17, 376-385.

 

Lamb, M. E. (1977b). The development of mother-infant and father-infant
attachments in the second year of life. Developmental Psychology, 13(6), 637-
648.

Lamb, M. E. (1979). Social development in infancy: Reﬂections on a
theme. Human DeveloLmentag(1),68-72.

Lamb, M. E., Thompson, R., Gardner, W., Chamov, E., 8. Estes, E.
(1984). Security of infantile attachment as assessed in the “strange situation”: Its

study and biological interpretation. The Behavioral and Brain Sciences 7, 127-
171.

 

Lyons-Ruth, K, Connell, D., Grunebaum, H., 8. Botein, S. (1990). Infants
at social risk: Maternal depression and family support services as mediators of
infant development and security of attachment. Child Development, 61, 85-98.

Lyons-Ruth, K, Alpern, L., 8. Repacholi, B. (1993). Disorganized infant
attachment classiﬁcation and maternal psychosocial problems as predictors of

hostile-aggressive behavior in the preschool child classroom. Child Development,
64;, 572-585.

Lyons-Ruth, K, 8 Zeanah, 0., Jr. (Eds). (1993). The family context of

infant mentaL health: Aﬁedivaaevelgment in the primapy caregiving relationship.
NY: Guilford Press.

233

Lyons-Ruth, K (1996). Attachment relationships among children with
aggressive behavior problems: The role of disorganized early attachment

patterns. Joumal of Consulting and Clinical Psychology, 64(1), 64-73.

Maccoby, E. E., 8. Mnookin, R. H. (1992). Dividing the child: Social and
legal dilemas of custody. Mk Harvard University Press.

Maccoby, E., Buchanan, C., Mnoookin, R., 8. Dombusch, S. (1993).
Postdivorce roles of mothers and fathers in the lives of their children. Journal of

Family Psychology, 7(1), 24-38.

Maier, W., Lichten'nann, D., Klinger, T., Heun, R., 8. Hallmayer, J. (1992).
Prevalence of personality disorders (DSM III R) in the community. Journal of

Personalig Disorders, 6(3), 187-1 96.

Main, M., Kaplan, N., & Cassidy, J. (Eds); (1985). security in infang,
childhood and adulthoocL: A move to t_he level of rapresentation. Chicago:
University of Chicago Press.

 

Main, M., 8. Solomon, J. (1986). Discovem of an insecure-
disorganizedldisoriented attachment pattern. NJ: Ablex Press.

Masheter, C. (1991). Postdivorce relationships between ex-spouses: The
roles of attachment and interpersonal conﬂict. Journal of Marriage and Family,
Q, 103-110.

Mathis, R., D. (1998). Couples from Hell: Undifferentiated spouses in
divorce mediation. Mediation Quarterly, 16(1), 37-49.

McLanahan, S. S. (1994). The Consequences of single motherhood. Ih_a
American Prospect. 18. 48-58.

McMillen, J. C. (1992). Attachment theory and clinical social work. Clinical
Social Work Journal 20(2), 205-218.

 

McNeaI, C., & Amato, P. (1998). Parents' Marital Violence. Journal of
F_amil\Llssy_as. 19(2), 123-139.

Minde, K, 8. Hesse, E. (1996). The role of the adult attachment interview
in parent-infant psychotherapy: A case presentation. Infant MenLaL Healt_h_
Journal 17(2), 115-26.

 

Morrison, 0., & Cherlin, A. (1995). The divorce process and young
children's well-being: a prospective analysis. Journal of Marriage and the Family,
_5_Z, 800-812.

234

Mott, F. (1990). When is a father really gone? Parental-child contact in
father-absent homes. Demography, 27(4), 499-517.

Patrick, M., Hobson, R., Peter, Castle, 0., Howard, R., 8. Maughan, B.
(1994). Personality disorder and the mental representation of early social

experience. Development and Psychopathology, 6, 375-388.

Reinhartz, S. (1992). Feminist methods in social research. NY: Oxford
University Press.

Rossiter, A. B. (1991). Initiator status and separation adjustment. Journal
of Divorce and Remarriage, 15, 141-155.

Schore, A. (1994). Affect Regulation and the Origin of meéelf. NJ:
Hillside: Erlbaum.

Simpson, J., A. (1990). Inﬂuence of attachment styles on romantic

relationships. Journal of Personality and Social Psychology, 56(5), 971-980.

Sprinthall, R. (1987). Basic Statistical Analysis (2"d Ed.). NJ: Prentice-
Hall, Inc.
Sroufe, L. A, 8. Waters, E. (1977). Attachment as an organizational

construct. Child Development, 48, 1184-1 199.

Sroufe, L. A. (1985). Attachment classiﬁcation from the perspective of
infant-caregiver relationships and infant temperament. Child Development. 56, 1-
14.

Sroufe, L. A. (1986). Appraisal: Bowlby’s contribution to psychoanalytic
theory and developmental psychology; attachment: separation: loss. Journal of

Child Psycholng and Psychiatry and Allied Disciplinesgzw), 841 -849.

Stein, H., Jacobs, N., J., Ferguson, K, 8., Allen, J., G., & Fonagy, P.
(1998). What do adult attachment scales measure? Bulletin of the Menninger
Clinic 62(1), 33-82.

 

Stern, D. (1991). Maternal representations: A clinical and subjective
phenomenological view. Infant Mental Health 12(3), 174—86.

 

Stern, D. (1995). The Motherhood Constellation: A uniﬁeg view of parent-
infant psychotherapy. New York: Basic Books.

Straus, M. A. (1979). Measuring intrafamily conﬂict and violence. Journal
of Marriage and the Familv. 41, 75—88.

235

Straus, M. A. (1990). Physical violence in American families: risk factors
and adaptations to violence in 8,145 families. New Brunswick: Transaction
Publications.

Straus, M. A., 8. Smith, C. (Eds.). (1990a). The Conﬂict Tactics Scales and
its critics: An evaluation and new data on validity and reliabiliy. New Brunswick:

Transaction Publishing.

Strauss, A., 8. Corbin, J. (1990). Basics of gualitative resaarch: Grounded
theom and procedures and technigues. Newbury Park: SAGE, Publications, Inc.

Toth, S., L., 8. Cicchetti, D. (1996). Patterns of relatedness, depressive
symptomology, and perceived competence in maltreated children. Journal of

Consulting and Clinical Psychology, 64(1), 32-41.

Trad, P. V. (1992). The role of parents in infant development: Early
predictor§ of t_heﬂher—infant interaction. New York: John Wiley & Sons, Inc.

Trafford, A. (1982). Crazy time: Surviving divorce: Bantan Books.

Tschann, J. M., Johnston, J. R., Kline, M., & Wallerstein, J. (1989). Family
process and children's functioning during divorce. Journal of Marriage and the
Family, 51, 431-444.

US Bureau of Census. (1990). 1990 US. Census Data .

US Bureau of Census (1995) (1995). Small area income and poverty
estimates. .

van lJzendoom, M., 8. Kroonenberg, P. (1988). Cross-cultural patterns of
attachment: A meta-analysis of the strange situation. Child_pevelopment. 59,
147-156.

van IJzendoorn, M., Goldberg, S., Kroonenberg, P., 8. Frenkel, O. (1992).
The relative effects of maternal and child problems on the quality of attachment:
A meta-analysis of attachment in clinical samples. Child Development, 63, 840-
858.

van IJzendoorn, M., 8. Bakermans-Kranenburg, M. (1996). Attachment
representations in mothers, fathers, adolescents, and clinical groups: A meta-
analytic search for normative data. Journal of Consulting and Clinical
Psvcholqgv. 64(1), 8-21.

Wall, J., 8. Amadio, C. (1994). An integrated approach to child custody

evaluation: Utilizing the “best interest” of the child and family systems
frameworks. Journal of Divorce é Remarriage, 2(3l4), 39-57.

236

Wallerstein, J. S., & Kelly, J. (1980). Surviving the breakup: How children
and parents copa- with divorce. New York: Basic Books.

Waters, E., & Deane, K E. (1985) Deﬁning and assessing individual
differences in attachment rela_tionsmps: g-met hodology and the organization of
behavior In infangj “and early childhood. Chicago. University of Chicago Press.

Weiss, R. S. (1975). Marital Separation. New York: Basic Books, Inc.

Weiss, 3., 8. Dodge, K. A. (1992). Some consequences of early harsh
discipline: Child aggression and a maladaptive social information processing
style. Child Development, 63, 1321 -1 335.

West, M., 8. Sheldon, A. E. R. (1988). ‘ Classiﬁcation of pathological
attachment patterns in adults. Journal of Personality Disorders 2(2), 153-159.

 

West, M., Rose, M., Sarah, 8. Sheldon, A. (1993). Anxious attachment as
a determinant of adult psychology. The Journal of Nervous and Mental Disease
1_8_1(7), 422-427.

 

Zeanah, C. J., Mammen, O., 8. Lieberman, A. (1993). Disorders of
attachment. NY: Guildford Press.

Zeanah, C. (1996). Beyond insecurity: A reconceptualization of

attachment disorders of infancy. Journal of Consulting and Clinical Psyahology,
_6_4(1), 42-52.

237

   

“IIIIIIIIIIIIII: