The impact on interpersonal intimacy of parental divorce and the subsequent father-daughter relationship

MISSING IMAGE

Material Information

Title:
The impact on interpersonal intimacy of parental divorce and the subsequent father-daughter relationship
Physical Description:
viii, 128 leaves : ill. ; 29 cm.
Language:
English
Creator:
Freeman, Diane E., 1947-
Publication Date:

Subjects

Subjects / Keywords:
Divorce -- Psychological aspects   ( lcsh )
Fathers and daughters   ( lcsh )
Children of divorced parents -- Psychology   ( lcsh )
Divorced fathers   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1993.
Bibliography:
Includes bibliographical references (leaves 116-127).
Statement of Responsibility:
by Diane E. Freeman.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001921473
notis - AJZ7237
oclc - 30461504
System ID:
AA00003239:00001

Full Text









THE IMPACT ON INTERPERSONAL INTIMACY OF PARENTAL DIVORCE AND
THE SUBSEQUENT FATHER-DAUGHTER RELATIONSHIP
















By

DIANE E. FREEMAN


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


1993












This is dedicated to the memory of my grandmother,

Mary McKim Turner, who was always my greatest supporter

and friend. She was a survivor, and an inspiration.










ACKNOWLEDGEMENTS


Sincere appreciation goes to Dr. Dorothy Nevill as the

chair of my committee. Her guidance, support and

encouragement have been invaluable throughout this project

and throughout graduate school. Over the last five years,

she has been my teacher, my mentor and my friend. She has

seen my strengths and weaknesses; she has allowed me to cry

on her shoulder; but she has never allowed me to give up.

She is also an excellent role model, helping me to learn to

take care of myself as well as others.

Thanks also go to Drs. Harry Grater, Marty Heesacker,

Phyllis Meek and Larry Severy who generously offered their

time and knowledge. Dr. Grater has been one of my favorite

professors, giving me courage to take risks and grow as a

therapist. Dr. Heesacker helped me to discover some of the

"fun" involved in doing research. Dean Meek gave me a new

understanding and appreciation of feminism, and she was

another strong female role model. Finally, Dr. Severy has

shown me respect, given me both responsibility and freedom,

and supported me in ways that have helped me to feel

confident in my abilities.

Christine Pugleise, my research assistant, has also

been a tremendous help. Without her, data collection would


iii







have been difficult, if not impossible. My friends, Linda,

Sally, and Moseley have also been invaluable. They have

encouraged and supported me both before and during graduate

school, and Linda even helped in the coding of the data.













TABLE OF CONTENTS


page

ACKNOWLEDGEMENTS...................................... iii

ABSTRACT .......... .................................... vii

CHAPTERS

1 INTRODUCTION................................... 1

Divorce as a Process............................. 3
Importance of the Father-Daughter Relationship... 4
Intimacy and Adolescent Development............. 5
Theoretical Foundation ........................... 8
Purpose and Need for the Study................... 13

2 REVIEW OF THE LITERATURE........................ 15

Method of Literature Search....................... 16
Long Term Effects of Divorce..................... 17
Intergenerational Transmission of Divorce........ 27
Father-Child Relationship Following Divorce...... 31
Intimacy and its Measurement..................... 39
Capacity for Intimacy......................... 41
Intimacy Motivation........................... 43
Impact of Parental Divorce on Intimacy.......... 44
Summary.................... .................... 47
Hypotheses...................... .. ... ...... ....... 49

3 METHODS..... .................. ........... ..... 53

Subjects........... .............................. 53
Procedure................................. ....... 56
Measures......... .. ............................ 57
Demographic Questionnaire....................... 57
Parent-Child Relationship Survey (PCRS)........ 58
Miller Social Intimacy Inventory (MSIS)........ 60
Risk in Intimacy Inventory (RII).............. 62
Rubin's Love Scale (RLS)....................... 63
Dyadic Trust Scale (DTS)...................... 65














4 RESULTS ........................................ 68

Frequencies and Univariate Statistics for
Parent-Child Relationships...................... 69
The Chi Square Test of Racial Composition........ 71
Socio-Economic Status........................... 71
Analyses of Variance for Primary Hypotheses...... 72
Relationship with Father...................... 72
Parental Marital Status and Intimacy........... 74
Father-Daughter Relationship and Intimacy....... 75
Interaction of Father-Daughter Relationship
and Parental Marital Status.................. 75
Impact of Having Someone "Like a Father"....... 77
Factor Analyses of Intimacy Scales............... 79

5 DISCUSSION...................................... 83

Conclusions................................... 89
Limitations of the Study......................... 91
Implications for Future Research................. 95

APPENDICES

A INFORMED CONSENT................................ 99

B DEBRIEFING.... .................. ...... ......... 100

C DEMOGRAPHIC QUESTIONNAIRE....................... 101

D PARENT-CHILD RELATIONSHIP SURVEY.................. 104

E RUBIN'S LOVE SCALE ............... .............. 110

F DYADIC TRUST SCALE............................... 112

G RISK IN INTIMACY INVENTORY....................... 114

REFERENCES ............................................. 116

BIOGRAPHICAL SKETCH..................................... 128












Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy


THE IMPACT ON INTERPERSONAL INTIMACY OF PARENTAL DIVORCE AND
THE SUBSEQUENT FATHER-DAUGHTER RELATIONSHIP

By

Diane E. Freeman

August 1993

Chairperson: Dr. Dorothy D. Nevill
Major Department: Counseling Psychology

The present investigation looked first at the impact of

parental divorce on the father-daughter relationship, and

found there to be a significant negative effect. Overall,

fathers were found to be rated more negatively than mothers,

and the relationship was more negatively influenced by

divorce. Next, parental marital status and the relationship

with the father were examined for their impact on

interpersonal intimacy. There appeared to be little

difference between daughters from intact or divorced families

on the measures of intimacy, with the exception that daughters

of divorce perceived more risk associated with intimacy. The

father-daughter relationship contributed little to the

variance in intimacy scores. Finally, the impact of having

another significant father figure was examined. Daughters of

divorce with a poor relationship with their father perceived

vii












higher levels of risk associated with intimacy when there was

no other significant father figure. When someone "like a

father" was present, the level of perceived risk was similar

to that found in daughters from intact families.


viii












CHAPTER 1
INTRODUCTION


Americans have an increasing desire for intimacy, but

seem to have a decreasing capacity to fill that desire

(Reis, 1984). Among the possible contributors to this

incapacity are the divorce rate and the number of children

raised in single parent homes. If divorce is

intergenerationally transmitted as some suggest, then the

deficiencies in intimacy may be perpetuated. Consequently,

it is important to explore further the long-term impact of

parental divorce and the subsequent capacity for

heterosexual intimacy in the children of divorce.

In contemporary society, it is rare to find someone who

has not been touched by divorce. Divorce "affects an

interpersonal system that includes many people other than

spouses as many as five million people will be tied to

the more than one million who divorce" (Rice & Rice, 1986,

p.3) each year. Hetherington, in 1979, predicted that 40-

50% of children born in the 1970's would experience divorce

and spend an average of six years living in a single parent

home. Consequently, a large number of today's college

students either will have been impacted by parental divorce,

or can expect to experience it.








Given the incidence of divorce and the number of

children who spend at least part of their youth in single

parent homes, a number of investigators have been concerned

about the economic, social and emotional consequences of

divorce. Both the short-term, and to a lesser extent, the

long-term sequellae have been examined. Any of the factors

that surround divorce may have profound effects which "are

incorporated within the character, the attitudes, the

relationships, the self-concept, the expectations, and the

world view of the child" (Wallerstein, 1983, p. 233).

However, in every major area of investigation, including

interpersonal relationships, contradictory results exist

(Edwards, 1987).

The ambiguous nature of the literature may be due to

differences in research methodology, including failure to

control for socioeconomic status; differences in subject

samples; small or nonrepresentative samples; outcome

measures, for example, use of projective tests or measures

that are insensitive to changes that might occur over time;

or theoretical perspectives (Edwards, 1987; Kanoy &

Cunningham, 1984). However, contradictions might also occur

because the responses to divorce are dependent upon a number

of variables in the experience of the family. For example,

a number of factors have been identified that might affect

adjustment, including the age of the child at the time of

the disruption; the quality of the parents' marriage before






3

the disruption; remarriage; conflict before, during or after

the divorce; and the relationship with the noncustodial

parent (Kalter, 1987; White, Brinkerhoff, & Booth, 1985).

Divorce as a Process

Obviously, divorce cannot be viewed as a single event,

but must be viewed as a cluster of experiences or as a

process. The tendency to dichotomize subjects as either

from intact or divorced families is an oversimplification

and is one of the problems with methodology, which was

identified by Lopez (1987) in a review of the literature.

The systemic process of divorce begins before the legal

separation and extends well into the post-divorce years

(Lussen, 1988). It is particularly important to note that,

when comparing children of divorce with children from intact

families, it is not necessarily "a comparison of one

category of persons, all of whom had very negative early

influences, with another category of persons, all of whom

were free of very negative early influences" (Glenn &

Kramer, 1985, p. 910).

For some, parental divorce brings relief from tension

and anger, so it is a positive event. For others, divorce

may be perceived negatively, as it brings feelings of loss

or shattered dreams. Whatever the initial experience,

divorce generally produces changes in the parent-child

relationship. While relationships with both mothers and

fathers may change following parental divorce, the quality








of the relationship with both the custodial and the

noncustodial parent is considered to be important in

predicting a child's behavior (Hess & Camara, 1979).

Importance of the Father-daughter Relationship

Following a parental divorce, a child still has two

natural parents, but may be faced with divided loyalties.

Approximately 90% of children who end up living with only

one natural parent are in the custody of their mother

(White, Brinkerhoff, & Booth, 1985). Consequently, "father

absence, or at least decreased father availability, is a

typical concomitant of divorce" (Biller, 1981, p. 489).

Because of changes in the living situation and other

divisive tendencies within families following divorce, the

child often experiences increased closeness and quality of

relations with one parent and reduced contact and quality of

relations with the other, generally, the non-custodial

parent, or father.

White, Brinkerhoff and Booth (1985) point out that the

decrease in attachment to the non-custodial parent reduces

the child's affectional network. Although both sons and

daughters are affected, the reduced affectional network

uniquely impacts the daughter, as father absence seems to

have an effect on her ability to function in interpersonal

and heterosexual relationships (Biller, 1981). The presence

of the father and his influence helps the daughter "to






5

experience herself as a feminine person and helps her relate

to the social world as a female" (Forrest, 1966, p. 21).

Intimacy and Adolescent Development

The possibility that parental divorce may have a long-

term impact on interpersonal and heterosexual relationships

has significant implications for the adolescent and young

adult. The developmental task for this age group is the

achievement of greater psychological separation from the

family and the establishment of an adult identity. Erikson

(1968) describes this developmental task as that of

establishing mutual psychosocial intimacy.

A part of creating an adult identity involves forming

intimate relationships with others, outside of the family,

who show understanding and love. Gilligan (1982) suggested

that women define their identity in the context of human

relationships. The ability to establish these relationships

is related to self-esteem, trust, willingness to take risks,

and willingness to make commitments, and it is rooted in the

parent-child relationship. The ability to establish

relationships does not suddenly emerge in young adulthood,

but builds on skills attained in previous relationships

(Bar-Yam Hassan & Bar-Yam, 1987).

Because there are changes in parent-child relationships

following divorce, and because Hetherington (1979) suggests

that adolescence is a time where earlier unresolved issues

reemerge in disturbances in heterosexual relations, it is








important to explore further the impact of divorce on the

development of intimacy. The concept of intimacy captures

the essence of "shared norms (about communication,

responsibilities); attitudes (liking, loving, trust);

beliefs about the relationship (its uniqueness, importance);

and relations with other persons" (Kelley et al., 1983, p.

39). Intimacy involves mutuality in being able to share

worries and problems, being able to express emotions, having

a genuine interest in others, and lacking defensiveness

(Orlofsky, Marcia & Lesser, 1973). The ability to develop

an intimate relationship with a member of the opposite sex,

is a complex process, and can be influenced by a variety of

factors. For instance, a deterioration in relations with

either parent is associated with increased courtship

activity and decreased satisfaction with interpersonal

relations for both males and females (Booth, Brinkerhoff &

White, 1984).

The break-up of the parental marriage may produce

suffering and a feeling of abandonment that has long lasting

effects, extending well into adult life (Jersild, Brook, &

Brook, 1978). Following parental divorce, the daughter

seems to get more maternal support, but she also gets less

paternal attention and she is more affected by father

absence than is the son (Hetherington, 1979). Leonard

(1966) suggested that the father's unavailability to give

love and to be loved is critical to the daughter's








development. Perhaps because father availability is

decreased, Lopez, Campbell and Watkins (1988) found that

parental divorce might actually accelerate most forms of

father-daughter psychological separation.

This separation may be manifested in a number of ways.

For example, adult female children of divorce are more

likely to have lower levels of well-being as defined by

happiness, health self-ratings, and satisfaction with

health, community, leisure, friendship and family life

(Glenn & Kramer, 1985). They are also more likely to become

sexually active at an earlier age (Booth, Brinkerhoff, &

White, 1984; Newcomer & Udry, 1987), to have a greater

number of sexual partners (Hepworth, Ryder & Dreyer, 1984),

and to marry earlier (Hetherington & Parke, 1979). Some of

these behaviors may be problematic, as making excessively

early commitments to another person may have negative

implications for the ability to negotiate later adult issues

(Franz & White, 1985). There is also some evidence of the

intergenerational transmission of divorce, as children of

divorce tend to marry earlier (McLanahan & Bumpass, 1988)

and to be more likely to divorce than are children from

intact families (Glenn & Shelton, 1983). Some persons with

loss by divorce seem to "seek to demonstrate by moving in

and out of a series of relationships, that the losses do not

hurt and that relationships have diminished value"

(Hepworth, Ryder, & Dreyer, 1984, p. 79).








Biller (1981) suggested that the "lack of opportunity

to observe meaningful male-female relationships in childhood

can make it more difficult for the father-absent female to

develop the interpersonal skills necessary for adequate

heterosexual adjustment" (p. 502). While Biller's

suggestion seems reasonable, it also seems inadequate to

completely explain the impact of parental divorce upon the

subsequent interpersonal relationships. Object relations

theory provides a broader theoretical foundation that

addresses the complexity of the responses to parental

divorce.

Theoretical Foundation

A person "is comprehensible only within [the] tapestry

of relationships, past and present" (Mitchell, 1988, p. 3).

Whereas object relations theory is described as a field

theory that considers the individual as anchored in his/her

environment or matrix of relationships (Antonovsky. 1987),

it provides a good foundation for understanding the

behaviors that are manifest in intimate interpersonal

relationships (Alford, 1990; Horner, 1984). Dicks (1963)

suggests that an individual internalizes relationships with

significant others (mother, father, siblings, etc.) and

because he or she has felt loved, cherished and accepted, he

or she learns to love as an adult.

The "objects" in object relations are human objects,

and the "relations" may be real or fantasied, internal or






9

external interactions with others (Cashdan, 1988). Sullivan

(1953), considered by Kernberg (1976) to be somewhat of an

object relations theorist, stresses the importance "of

interpersonal relationships as determinants of intrapsychic

and interpersonal structures" (p. 122). Early relationships

become internalized as mental representations that later

become manifest in behaviors with others (Lieberman, 1984).

Interactions with objects lead to "significant and lasting

modifications of the personality, usually conceived of as

internal structures, which affect all later experiences with

others" (Antonovsky, 1987, p. 538). Thus, the inner world

of object relations determines the way an individual relates

to the external world through interpersonal relationships.

Object relations evolve over the first three or four

years of life, but they continue to be modified by

experiences throughout life. In fact, the adolescent ego

identity has a foundation in the behavior of meaningful

others toward him or her (Sharf, 1989). The adolescent

integrates these perceptions and experiences into his or her

own changing self-concept.

Although the focus in object relations theory is on the

importance of the mother-child relationship, others also

play a significant role in the life of the child (Applegate,

1990; Lieberman, 1984). It is not the role of mother or

father, determined biologically or legally, that is

critical, but the object function of mothering or fathering








(Rosenberger, 1990). Rutter (1974) suggests that "most

children develop bonds with several people and it appears

likely that these bonds are basically similar" (p. 125).

Consequently, the father, or fathering figure, who is active

in the life of his child, may be almost as important in the

formation of object relations as is the mother, or mothering

figure.

Modification or formation of the internal structures

"is likely to take place in the presence of and in reaction

to strongly experienced affects such as...pain in the

relationship with significant others" (Antonovsky, 1987, p.

539). Obviously, parental divorce and the subsequent

decrease in the amount of contact and the quality of the

relationship with the father can create such pain. Divorce

differs from other loss, i.e. parental death, because the

father is still present. Consequently, the decrease in

contact may be perceived as rejection. This aspect of the

divorce experience might well create more pain and have a

larger impact upon interpersonal behaviors than any other.

"A primary human need is attachment to a caring person

or persons; we develop intense attachments because we crave

relatedness" (Mitchell, 1988, p. 26). In fact, Fairbairn

(1954) stated that the ultimate goal of human behavior is

the establishment of meaningful relationships. In the

family, the child learns a mode of connection, "and these

learned modes are desperately maintained throughout life"








(Mitchell, 1988, p. 27). Events that break or prevent the

necessary attachments, may produce effects that "crop up in

different ways over the years, as the child passes through

progressive stages of development (Hetherington, Cox & Cox,

1978).

According to object relations theorists, "development

is a process which takes place within an interactive matrix

of constitutional endowments, significant relationships, and

critical events" (Nicholson, 1988, p. 26). Divorce is not

only a critical event, but it also impacts on interpersonal

relationships. Parental divorce, almost of necessity,

interrupts or interferes with attachment to the noncustodial

parent, usually the father. Children often experience pain,

fear, anger, or depression following a parental divorce. As

suggested by Antonovsky (1987), this strong affect may lead

to modifications of internal structures.

If divorce results in the perception of abandonment or

rejection by the father, even though this rejection is only

fantasied, and if this perception is internalized, later

relationships may well be affected. The goal of the child

will be to avoid anxiety. To do this, the child may utilize

defenses of splitting or projective identification. For

example, a daughter may see men as all good or all bad; she

might use sexuality to attract men, as if to attract her

father; she may expect to be rejected and act in a way to

elicit rejection; or she might use dominance or control, as







12

if to say that she negates the importance of her father and

can stand alone. In essence, her mind will work to maintain

connections with objects, or parents (Alford, 1990).

"Painful feelings, self-destructive relationships,

self-sabotaging situations, (may be) recreated throughout

life as vehicles for the perpetuation of early ties to

significant others" (Mitchell, 1988, p. 27). In particular,

girls may manifest problems, in relating to males, that

surface when their interest in the opposite sex heightens

(Lynn, 1974). The old ways of maintaining connections may

no longer be appropriate or useful in forming intimate

relationships in young adulthood, but they may be

perpetuated none-the-less.

In conclusion, changes in the intrapsychic structures

"can have a profound effect on an individual's capacity to

enter into mature interpersonal relationships in adult life"

(Nicholson, 1988, p. 26). One may carry unconscious

fantasies along with mental representations of objects that

can color, distort, and affect relations with significant

others (Arlow, 1980). Object relations theory seems to

provide a theoretical base that addresses the complexity of

the impact of divorce on development. Specifically, it

gives a foundation for understanding the impact of father-

daughter relations on the ability to establish meaningful

interpersonal relationships.








Purpose and Need for the Study

To summarize, divorce impacts the entire family,

including the children, and there is some evidence that this

impact extends well into the post-divorce years (Lussen,

1988). In fact, there is some indication that children of

divorce are themselves more likely to end up divorcing. It

seems important to ascertain whether this is due to social

learning or perhaps due to the ability of children of

divorce to establish intimate relationships.

Prior research that has examined the long-term effects

of divorce on children has produced confusing and

contradictory results, generally due to the variability of

factors that might affect the adjustment of individuals

following divorce (Kalter, 1987; White, Brinkerhoff & Booth,

1985). One of the factors that has received some attention

is the child's relationship with the father. Divorce often

produces a decrease in the amount of contact and the quality

of the relationship with the non-custodial parent, typically

the father (Biller, 1981). The daughter's experience

following parental divorce differs from that of the son,

because the daughter is more affected by father absence than

is the son (Hetherington, 1979).

Although some studies have examined the importance of

the father-daughter relationship following divorce, none

have focused upon the impact of this relationship on the

developmentally appropriate task of college aged students,







14

i.e. the formation of intimate relationships outside of the

family. The present investigation seeks to extend the work

of others and to avoid some of the pitfalls of prior work by

using objective measures with good reliability and validity

and by using enough subjects to allow for detection of

significant differences, but not so many as to detect

differences of questionable importance.

The findings of this study may help to guide parental

decisions regarding the importance of a continuing father-

daughter relationship. If there are long-term ramifications

of divorce and subsequent decreases in parental relations,

as suspected, counselors may also utilize this information

in aiding families in transition. At the very least,

information will be gathered that will indicate whether or

not the father-daughter relationship needs to be examined or

controlled in future divorce research.













CHAPTER 2
REVIEW OF THE LITERATURE


The preponderance of research on the impact of parental

divorce has addressed the traumatic effects that occur

shortly after the event or within the period of adjustment,

up to three years following the divorce. However, several

longitudinal studies and other examinations of long-term

effects have been undertaken. Those that have dealt with

the long-term effects of parental divorce on adolescent and

young adult women have focused primarily on psychological

adjustment, family relationships, heterosexual

relationships, and the intergenerational transmission of

divorce. Although this body of literature provided the

foundation for the present investigation, few studies of

divorce have examined "conditions surrounding the divorce

that might affect children's attempts to form rewarding

relationships with the opposite sex" (Booth, Brinkerhoff, &

White, 1984, p. 85).

There are a variety of changes set in motion by

divorce, including changes in the father-daughter

relationship. Because the daughter's relationship with her

father might affect her relations with the opposite sex,

studies of the role of the father in adolescent development,

15









and the quantity and quality of the father-daughter

relationship following divorce were summarized. Also, the

research that focused on intimacy, the developmental task of

adolescence and young adulthood, was reviewed.

Studies that linked divorce and the father-daughter

relationship to interpersonal intimacy would have been

ideal. Unfortunately, such studies were not found.

However, four studies were reviewed that linked either the

father-child relationship or parental divorce to courtship

behaviors, interpersonal relationships, or intimacy (Andrews

& Christensen, 1951; Booth, Brinkerhoff & White, 1984;

Gabardi, 1990; Weiss, 1988).

Method of Literature Search

The literature was sampled via a computerized search of

the Psychlit, Sociofile and Dissertation Abstracts

International data bases. Search terms were gathered from

the literature and from the Thesaurus of Psychological Index

Terms (Walker, 1991). The terms included: "intimacy,"

"object relations," "divorce," "parent child relations,"

"young adults," "adolescents," "interpersonal intimacy," and

"father child relations." The terms were used alone and in

various combinations. Additionally, a search was performed

of the Social Science Citation Index on highly relevant

studies, such as Gabardi (1990). Finally, the reference

sections of useful articles were searched. The studies that

were utilized focused on the impact of parental divorce on









adolescents and young adults, the long-term impact of

divorce, the impact of the father-daughter relationship on

intimacy, and the impact of parental divorce on intimacy.

Studies were excluded that focused on young children or on

the short term consequences of divorce.
Long Term Effects of Divorce

Wallerstein and Kelly (1976, 1980, 1987) conducted a

seminal longitudinal study, which involved 60 families with

131 children. The subjects came from a nonclinical

population, and were referred for anticipatory guidance at

the time of separation. Subjects were seen again at 18

months, 5 years and 10 years. The data were gathered

primarily through lengthy interviews.

At the five year follow-up, the crisis period was over,

and most children had resolved their negative feelings about

their parents' divorce. It was noted that those who had

positive relationships with both parents achieved the best

adjustment. Despite resolution of many of the problems,

there was some evidence of emotional difficulties in one-

third of the children. These children described being

intensely dissatisfied with their post-divorce lives,

depressed and lonely. Only 34% of the children were

considered to be doing especially well, with high self-

esteem.

At the 10-year follow-up, Wallerstein (1987) re-visited

16 girls and 22 boys who had been early latency aged (6-8









years old) at the time of their parents' divorce. Semi-

structured interviews were supplemented by questionnaires.

All of the children were in the custody of their mothers,

but during the ten-year period several had spent some time

living with their fathers. A little more than one-third saw

their fathers regularly, defined as one or more times per

month. Fifty-seven percent of the girls and 44% of the boys

had irregular visits, defined as less than six visits per

year (Wallerstein, 1987).

In terms of school performance, Wallerstein (1987)

estimated that 40% of the subjects were underachieving to a

significant degree. Additionally, their career aspirations

were notably shallow. This group was found to be

significantly less psychologically and socially well

adjusted than the youngest group, who were now 11-15 years

old. Overall, "half of the boys and one fourth of the girls

were considered poorly adjusted and at high risk at the time

of the ten-year follow-up" (Wallerstein, 1987, p. 210).

The profound unhappiness which these subjects

experienced in their current relationships distinguished

them from the children who were either younger or older.

The unhappiness with relationships is particularly

significant as the formation of intimate relationships is

the appropriate developmental task for adolescents and young

adults (Erikson, 1968). Girls were noted to have had three

or more boyfriends during their adolescence, and one quarter







19

of the girls had had abortions. The girls in this age group

also reported more depression and suicide attempts

(Wallerstein, 1987).

The Wallerstein and Kelly studies focused attention on

the long-term ramifications of divorce with negative

findings in educational, psychological and social areas of

functioning which extended over a ten year period. These

negative results seem to have sparked a variety of

subsequent investigations. However, attempts to replicate

the findings have not always been successful. The lack of a

control group of intact families, the small number of

subjects in some of the comparison groups, and the use of

the interview as the primary data gathering method have been

criticized (Levitin, 1979). Nonetheless, the findings have

received considerable attention, in both the research and

popular literature, and Wallerstein and Kelly contributed

much to the study of divorce by delineating outcomes for

children of different ages and developmental levels.

A longitudinal investigation by Guidubaldi and Perry

(1985) focused on long-term psychological adjustment and

utilized a multifactored mental health assessment of 110

children. Children of parental divorce performed more

poorly than children of intact families on 9 out of 30

mental health measures (Guidubaldi & Perry, 1985). In this

study, the average length of time since the divorce was 6.41







20

years, again indicating the continuation of negative effects

beyond the initial adjustment period.

Guidubaldi and Perry (1985) found that divorce seemed

more related to maladjustment in boys than in girls.

However, girls were found to have more social involvement,

unreflectiveness, irrelevant talk, negative feelings,

critical-competitiveness and blaming than boys (Guidubaldi &

Perry, 1985). This investigation utilized parent ratings in

addition to other methods, and the authors suggested that

the daughters of divorce were more likely than the daughters

of intact families to tell their mothers if something good

happened. Thus, mothers' ratings of the daughters of

divorce could have been positively skewed. On the other

hand, the sons of divorce were less likely to tell their

mothers if something good happened than the sons of intact

families, possibly accounting for more negative ratings.

Psychological well-being following divorce was also

examined by Glenn and Kramer (1985). Females whose parents

had divorced, when compared to females from intact families,

had negative coefficients which reached significance on five

out of eight dimensions of psychological well-being. There

was no evidence that the negative effects diminished with

time, and there was evidence that females were more

negatively affected than males. Comparisons were also made

between persons who had lost a parent due to death and those

from intact families. The results indicated that loss of a








parent due to death does not produce the same long-term

negative effects as loss due to divorce (Glenn & Kramer,

1985). While this appears to contradict the theory that

father loss is an significant component of divorce, it

suggests that the reason for the loss is important.

Following parental death, there is no reason for a child to

expect continued contact, but following divorce, the father

is still alive and available. A decrease in the amount and

quality of contact with the father might be perceived as a

rejection and this perception might contribute to negative

effects.

Divorce often involves parental conflict, and this

conflict also affects the adjustment of the child. For

example, Chess et al. (1983) found high parental conflict

associated with poorer adaptation; Long (1986) found self

esteem positively related to parental happiness rather than

to family structure; and Ellison (1983) found a positive

correlation between parental harmony and children's

assessment of their own psychological adjustment.

Slater and Calhoun (1988) examined the influences of

parental marital status and level of conflict during

childhood (high or low) on social functioning of

undergraduate psychology students. "The ability to develop

and maintain supportive friendships and dating relationships

varied as a function of family structure and conflict"

(Slater & Calhoun, 1988, p. 123). Interestingly, within the







22

divorced group, those with high conflict had better indices

of social functioning than did those with low conflict.

However, females in the divorced/high conflict group were

less likely to have a boyfriend. The authors concluded that

students from the high conflict groups might have lower

expectations and reduced skills for the maintenance of

intimate relationships.

Parental conflict also affects the parent-child

relationship. Conflictual parents are more likely to be

self-absorbed, having less energy or capacity for parenting.

The child may end up feeling like a pawn, torn between two

warring parents, and the conflict may make the access to the

non-custodial parent more difficult (Kline, Johnston &

Tschann, 1991). Farber, Felner and Primavera (1985) also

found levels of family cohesion and conflict to be

predictive of adaptive outcome. Where there was greater

conflict and less cohesion, there was an increase in

anxiety.

Hetherington, Cox, & Cox (1985) conducted a

longitudinal study of 144 children and their parents which

began to address family relationships. In the original

investigation, one-half of the children were from divorced

families, and were in the custody of their mothers; the

other half were from intact families. At the six year

follow-up, 42 of the divorced mothers had remarried.

Consequently, some subjects were added so that there were 30









sons and 30 daughters in each of three groups: intact,

divorced/remarried, and divorced.

This investigation utilized interviews, standardized

measures, and in-home observations. Several differences

were observed in family relationships, depending upon family

structure. For example, sons and daughters from divorced

families were allowed more responsibility and independence

than children from intact homes (Hetherington, 1989).

Interestingly, what Hetherington described in positive

terms, Wallerstein (1985) perceived more negatively, as she

felt that children became "overburdened" with various types

of responsibilities. Whether or not the daughters of

divorce perceived their life changes negatively varied with

whether or not their mothers had remarried. Those whose

mothers had remarried saw themselves and were seen as having

more problems than those from intact families and those from

divorced, non-remarried families. Despite many strengths of

this study, the quality of parent-child relationships was

not addressed.

In an attempt to examine the attitudes of children

toward themselves and their parents, Parish and Wigle (1985)

evaluated the attitudes of 639 students in another

longitudinal study. There were three randomly selected

groups of 30 adolescents each: families intact throughout

the study; families divorced at the onset and completion of







24

the study; and families intact at the onset but divorced at

the completion.

Subjects were asked to complete the Personal Attribute

Inventory for Children (Parish & Wigle, 1985) with their

mothers, fathers, and themselves as targets. The

adolescents from intact families evaluated themselves and

their families more positively than adolescents from

divorced families. The ratings were the lowest for those

adolescents whose families experienced parental divorce

during the study. These findings demonstrate the impact of

recent divorce, and suggest that the pain of divorce

diminishes with time (Parish & Wigle, 1985).

Although the evaluations of self and parents became

more positive over time, the evaluations from children of

divorce never became as positive as those from intact

families. This study went beyond family structure to

examine family processes and relationships. Family

processes appear related to the way adolescents evaluate

themselves and their parents, and father absence, in

particular, was strongly associated with negative

evaluations (Parish & Wigle, 1985).

Parental divorce also seems to have an impact on

attitudes and expectations for intimate relationships.

Wallerstein and Blakeslee (1989) interviewed 116 children

ten years after their parents' divorce. At that time, the

children were between 11 and 29 years of age. The findings






25

indicated that these children wanted what their parents had

failed to achieve: a lasting, committed, romantic love and

marriage. However, they felt it was unlikely that they

would achieve these goals. Many of these subjects felt

rejected and feared rejection in future relationships with

the opposite sex.

As a part of the Wallerstein and Kelly investigation,

Kelly (1981) conducted a study of 18 adolescents and young

adults (aged 17-23). Eighteen months following their

parents' divorce, the subjects were categorized as resuming

their developmental agendas or remaining delayed, disrupted

or fixated in their development. After five years, those

who had been considered to be well adjusted, at the eighteen

month interview, had not yet developed appropriate, enduring

relationships. Their relationships tended to be short-lived

and terminated by the subjects. Those adolescents who had

been functioning at a lower level, had rushed into

heterosexual activity and clung to relationships that were

unsatisfying. Both patterns could be indicative of problems

with interpersonal intimacy. However, it should be noted

that these conclusions are based on a small sample; that

gender was not controlled; and that there was no comparison

group of children from intact families.

Kalter et al. (1985) compared the attitudes of 42

female college students from divorced families to the

attitudes of 42 female subjects from intact families. A 19









page questionnaire and 2 Thematic Apperception Test (TAT)

(Murray, 1971) cards were administered. Several findings

emerged: the women in the divorced group began dating later

than those in the intact group; the divorced group saw men

as more unfeeling and less strong; the divorced group saw

females as less sensitive and less mature; and the divorced

group was less hopeful about the future and less certain

about having a lasting marriage.

Using TAT cards, Lussen (1988) compared the stories of

nine girls who experienced parental divorce and ten girls

from intact families. The divorced group wrote stories that

had the same conflicts, themes, and tasks as the stories

from the intact group, but their stories had less resolution

and greater ambivalence. The divorced group also had a less

trusting view of the world than the intact group. Men were

seen as "less supportive, more absent, more pursuing, less

rational, more impulsive, and less understandable" where

women tended to be "more stuck, burdened, worn down, and

trapped by their relationships and cares" (Lussen, 1988, p.

114-115). The author does not indicate that the stories

were judged blind and, if that were not done, it would have

been a major flaw in the research design.

Parental divorce has been found to have significant

negative effects in educational, psychological and social

areas of adjustment. Eleven studies have been reviewed and

have indicated that the negative effects of divorce extend









beyond the three year adjustment period. Even though the

negative effects diminish over time, Parish and Wigle (1985)

found that the evaluations of self and parent by children of

divorce never became as positive as those from children of

intact families. Following divorce, not only does the

structure of the family change, but the family processes and

relationships also change. The absence of the father has

been found to be strongly associated with adolescents'

negative evaluations of themselves and their parents (Parish

& Wigle, 1985). Relationships within the family were not

the only ones affected; Kelly (1981) found that adolescents'

heterosexual relationships tended to be either short-lived

or unsatisfying. Additionally, girls who have experienced

parental divorce have been found to view men less positively

and to have lower hopes and expectations for marriage

(Kalter et al., 1985).

Intergenerational Transmission of Divorce

With parental divorce impacting upon relationships and

expectations for marriage, one of the long-term consequences

of divorce that has received much of the research focus has

been the intergenerational transmission of divorce. Several

investigators have found there to be no significant

differences between intact and divorced adult children on

attitudes toward marriage. For example, adult children of

divorce have been found to be as likely to want to get

married (Black & Sprenkle, 1991; Ganong, Coleman & Brown,









1981) and to perceive the advantages of marriage similarly

to the children from intact families (Amato, 1988).

However, children living in either single parent or

reconstituted families view divorce more favorably than

those from intact families (Coleman & Ganong, 1984;

Greenberg & Nay, 1982).

Wallerstein and Blakeslee (1989) interviewed 116

children ten years after their parents' divorce. At that

time, the children were between 11 and 29 years of age. The

findings indicated that these children wanted what their

parents failed to achieve: a lasting, committed, romantic

love and marriage. However, they felt it was unlikely that

they would achieve these goals. Many of these subjects felt

rejected and feared rejection in relationships with the

opposite sex.

Kulka and Weingarten (1979) utilized data from two

national surveys which were conducted by the Survey Research

Center at the University of Michigan. The initial survey,

in 1957, included 2460 respondents and then 2264 respondents

were surveyed in 1976. The data were collected using ninety

minute interviews, and results indicated that young adults

from divorced homes were less likely than those from intact

homes to report being very happy. Also, children of

parental divorce were more likely to report having their own

marital problems. Other investigators confirmed the

findings that parental divorce is positively related to









lower levels of marital happiness, increased marital

instability, and increased marital disagreements (Booth &

Edwards, 1989; Glenn & Kramer, 1987).

In another study designed to examine the impact of

preadult experiences on behavior and adult well-being, Glenn

and Shelton (1983) found that females whose parents had

divorced had a 59.3% greater divorce rate than those whose

families remained intact. Similarly, Pope and Mueller

(1976) found a higher rate of marital dissolution among

children from disrupted homes. By analyzing data including

parents' and grandparents' marital status, Black and

Sprenkle (1991) found that intergenerational marital

instability was significantly greater for the divorced

group. Adult children who perceived their parents' marriage

to be unhappy had lower levels of psychological, social and

marital well-being, as did the adult children of divorce

(Amato & Booth, 1991).

In addition to a greater likelihood for divorce or for

marital difficulties, parental divorce may influence

children's relationships and their later adult well-being in

other ways. McLanahan and Bumpass (1988) extended previous

studies by examining variables relevant to the formation of

families. Interviews were conducted with 7,969 women who

ranged in age from 15-44 years. The results provided strong

evidence that women who experienced parental divorce and

lived in single parent homes, whether in the custody of









mother or father, were more likely to marry before age

twenty, give birth before age twenty, give birth before

marriage, and to divorce.

Long (1987) had 134 female undergraduates fill out

questionnaires on two occasions in order to evaluate

attitudes toward marriage. The expectations and evaluations

of marriage were lower for daughters who perceived their

parents' marriages to be unhappy. Daughters of broken

marriage also expected to marry later.

Wallerstein (1983) stated that one of the six

psychological tasks of the child after divorce is to achieve

realistic hopes regarding relationships. Generally, the

achievement of this goal is an issue of adolescence and

young adulthood. "In order to trust in the reliability of

relationships and maintain the capacity to love and be

loved, the child of divorce will need to have acquired

confidence in his or her ability and self-worth"

(Wallerstein, 1983, p. 242). A number of Wallerstein's

(1983) subjects indicated on-going problems in achieving

this task. Some said they would never marry, some were

described as promiscuous with low self-esteem, and some

seemed cynical and felt hopeless.

The conclusions of the preceding investigations seem to

indicate that while the adult children of divorce appear as

likely to enter into a marriage (Black & Sprenkle, 1991;

Ganong, Coleman & Brown, 1981), they may be more likely than







31

children from intact families to experience problems within

marriage and to divorce (Kulka & Weingarten, 1979; Glenn &

Shelton, 1983). The eleven studies that were reviewed

provided evidence to support theories of intergenerational

transmission of marital instability. Perhaps some aspects

of parental divorce influence relationships and

consequently impact upon the ability to find satisfaction in

marriage.

Father-Child Relationship Following Divorce

While much of the previous research focused upon the

impact of parental divorce as if it were a single event, a

number of more recent investigations have explored the

impact of components of the divorce process. It may well be

that divorce status does not simply indicate the structure

of the family, but indicates the kinds of interactions or

relationships likely to be found within families. Relations

with parents have been found to be particularly important in

contributing to the successful adjustment of children

following parental divorce (Hetherington, Stanley-Hagan, &

Anderson, 1989). For example, parent-child relationship

variables were found to have a greater impact on social

relations than parental harmony (Hess & Camara, 1979), and

the parent-child relationship was found to have a greater

influence on marital attitudes than was family dissolution

(Coleman & Ganong, 1984).









Whereas relationships with both parents are important,

significant differences have been noted in intimate

attachments to mothers and fathers, with fathers being rated

lower on intimacy (LeCroy, 1988). However, a father who is

warm, involved and accepting contributes to the optimal

development of his child (Weinraub, 1978). Comparisons of

children from intact and divorced families, using measures

of attitudes toward parents and the number of positive

adjectives checked, reveal higher ratings of fathers by

children from intact families than by children from divorced

families (Drill, 1987; Parish, 1981; Parish, 1991). These

ratings may be related to changes in the amount of contact

with the father or in the quality of the father-child

relationship.

Based on a review of the literature, Biller and Weiss

(1970) concluded that the role of the father is important in

the personality development of the daughter and in her

feminine identification. The relationship with the father

also explains a significant amount of the variance in self-

esteem and problem behaviors (LeCroy, 1988). The impact of

the father-daughter relationship on self-esteem is relevant

for the current investigation because a good sense of self

and strong ego are essential for the ability to achieve

emotional and sexual intimacy (Kaslow & Schwartz, 1987).

Young and Parish (1977) specifically explored the

impact of father absence on female college students. There









were three groups of students: daughters who lost their

father because of death; daughters who lost their fathers

because of divorce; and daughters from intact families.

Within each of the "father loss" groups, some of the mothers

had remarried and some remained single. Standardized

measures were administered in order to assess reflection vs.

impulsivity, self-criticism, and feelings of security.

Daughters who had lost a father and whose mothers had not

remarried thought of themselves more negatively and saw

themselves as more insecure than girls from intact families

(Young & Parish, 1977).

Similarly, Parish and Wigle (1985) found father absence

to be strongly associated with negative evaluations of self

and parents by the child. When examining the father-child

relationship, fathers from divorced families were rated

significantly more negatively than those from intact

families (Parish & Osterberg, 1984). Attachment to the non-

custodial parent was significantly lower than attachment to

parents in intact families (White, Brinkerhoff & Booth,

1985).

Following parental divorce, both males and females

typically have less contact with their fathers, but the

amount of contact is considerably less for females than for

males (Amato & Booth, 1991). Southworth and Schwarz (1987)

assessed the frequency, regularity and duration of visits

with the father in a study of 104 female college students,









one-half from divorced families and one-half from intact

families. A composite score indicated that post-divorce

contact decreased over time. The amount of contact during

the 3-5 year period after divorce was significantly lower

than during the first two years (Southworth & Schwarz,

1987). Indicating the impact of a decrease in contact,

Lopez and Watkins (1991) found that those students who

reported a low level of contact with their fathers had

higher scores of functional, emotional, and attitudinal

independence from their fathers.

Although the amount of contact does not necessarily

indicate the quality of the relationship with the father,

there tends to be some correspondence. For example, those

who experienced less frequent contact with their fathers

following divorce perceived their fathers as less accepting

and more inconsistent (Southworth & Schwarz, 1987).

Further, within the divorced group, when there was a

perception of decreased acceptance and consistency, there

was an indication of a decrease in heterosexual trust.

These findings are particularly significant because there is

a correspondence between the capacity for intimacy and

parental acceptance (Finch, 1986).

Overall, quality is considered more important than the

frequency of visitation with the father, as children need to

have confidence in their ties with their father (Hess &

Camara, 1979; Wallerstein & Blakeslee, 1989). Daughters not









only experience a decrease in contact with the father

following divorce, but the nature of the father-daughter

relationship also changes over time (Hetherington, Cox &

Cox, 1978). Fathers were found to become more and more

unavailable to their children over the two year period

following divorce. Fathers became less nurturant and more

detached, and when interactions between fathers and their

children were observed, the divorced fathers ignored their

children more and demonstrated less affection (Hetherington,

Cox & Cox, 1978).

In a study designed to examine the effects of father

absence on personality development in adolescent daughters,

Hetherington (1972) utilized observations of non-verbal

behaviors, interviews and personality measures. There were

three groups of 24 adolescents each; the subjects were

either from intact families, families with father absence

due to death, or families with father absence due to

divorce. Significant differences were found between the

groups in the ways they responded to male interviewers,

while few differences were observed in the responses to

female interviewers. Additionally, daughters of divorce

reported more heterosexual activity, more negative feelings

toward their fathers and more conflict with their fathers

than either of the other groups. Further, the effects were









greater with early separation rather than later separation,

perhaps attesting to the importance of early object

relations.

For daughters, there seem to be different patterns of

effects depending upon the reason for the father absence,

and these effects appear during adolescence and are

manifested primarily as the inability to interact

appropriately and satisfactorily with males (Hetherington,

1972; Chase-Lansdale & Hetherington, 1990). Zaslow (1988)

has described the long-term effects of divorce and father

absence as "sleeper effects", appearing as a child passes

through progressive stages of development. What

Hetherington (1972) described as apprehension and inadequate

skills in relating to males becomes most evident just when

these skills are required.

Hainline and Feig (1978) attempted to replicate

Hetherington's (1972) findings regarding the impact of

father absence in a study of college-aged women. Non-verbal

behavior during an interview and performance on both

standardized and nonstandardized measures were evaluated.

Subjects who experienced father absence before age five

seemed to have more traditional attitudes about some aspects

of sexual behavior than did the subjects who experienced

father absence between ages 5 and 10. However, there were

no significant differences noted between any of the father

loss or control groups on measures of sexual behaviors or








Rubin's Romantic Love Scale (Rubin, 1970). While this

investigation failed to replicate the findings of

Hetherington (1972), it should be noted that there were only

six subjects in each of the cells; there were cultural

differences in the samples, and the subjects in the

Hetherington study were of a lower socioeconomic status.

Vess, Schwebel and Moreland (1983) further explored the

long-term impact of divorce on feminine development.

Undergraduates were utilized whose families were intact,

whose families divorced before the subject was five years of

age, and whose families divorced when the subject was

between five and ten years of age. The girls whose fathers

left before age five selected more feminine characteristics

on the Gough Femininity Scale than those whose parents

divorced later. The authors further concluded, based on the

indications of distrust and insecurity that persisted well

after the parental divorce, that the child's perceived

rejection by one or both parents may result in enduring

problems (Vess, Schwebel & Moreland, 1983).

Kinnaird and Gerrard (1986) examined sexual behavior

and attitudes toward marriage, using adolescents and young

adults whose families were divorced, remarried or intact.

Subjects were administered five questionnaires and the

results indicated that the intact group had more positive

attitudes toward marriage; the daughters from both divorced

and reconstituted families reported more sexual experience;









and subjects from divorced and reconstituted families were

more likely to respond that they would have liked more

contact with their fathers. An effort was made to

investigate the impact of other significant males, including

the stepfather, but the presence of other influential males

was not predictive of any behavioral, attitudinal or

adjustment variables.

Wallerstein (1984), as a part of the ten-year follow-

up, found that, for thirty adolescents who had been between

two and one-half and six at the time of divorce, "the non-

custodial father remained a significant psychological

presence" (p. 454). Even those children whose mothers had

remarried had an intense awareness of their fathers, and the

relationships with fathers and stepfathers were

distinguishable and separate. The need for the father had

not diminished but heightened with the advent of adolescence

(Wallerstein, 1984).

The relationship with the father has been found to be

important in contributing to adjustment following parental

divorce (Hetherington, Stanley-Hagan & Anderson, 1989) and

to the optimal development of the child (Weinraub, 1978).

Following divorce, not only is the amount of contact with

the father decreased, but the nature of the relationship

also changes (Hetherington, Cox & Cox, 1978). Daughters

experience less frequent contact than sons and the amount of

contact decreased over time (Amato & Booth, 1991; Southworth









& Schwarz, 1987). While some authors have suggested that

the presence of another significant male figure might

decrease the impact of changes in the father-daughter

relationship, this has not received much support in the

studies that were reviewed.

A poor relationship with the father has been linked

with apprehension and inadequate skills in relating to males

(Hetherington, 1972); with changes in the personality

development of the daughter and her feminine identification

(Biller & Weiss, 1970; Vess, Schwebel & Moreland, 1983); and

with sexual behaviors and negative attitudes toward marriage

(Kinnaird & Gerrard, 1986). Of all the components of the

divorce process, the relationship with the father may have

the greatest impact upon the ability to have satisfying,

intimate relationships.

Intimacy and its Measurement

Much of the literature about intimacy is descriptive,

attempting to define the construct. This is understandable

as some equate intimacy with love, some equate intimacy with

sexual behavior, and others believe that intimacy involves

primarily the ability to self-disclose. The diverse ways

that intimacy has been defined make it appear more complex

than any of these views indicate. Intimacy involves

affection, the willingness to commit oneself to another, the

ability to trust, the ability to be open and share personal

information, and the willingness to provide support to








another (Levitz-Jones & Orlofsky, 1985; Perlman & Fehr,

1987; Trotter, 1986). Intimacy also requires the ability to

both give and receive love, so it entails mutuality

(McAdams, 1989; Dyk & Adams, 1987).

A study by Roscoe, Kennedy and Pope (1987) further

illustrates the complexity of defining intimacy. When 277

undergraduate students were asked to respond to questions

requesting their thoughts of what distinguishes an intimate

from a non-intimate relationship, the most frequently

mentioned components were sharing, physical/sexual

interaction, trust and faith, openness, and love (Roscoe,

Kennedy & Pope, 1987). There were also significant

differences between the components mentioned most frequently

by men and women. Males were more likely to cite

physical/sexual interaction, and females were more likely to

cite openness and self-abandon.

Intimacy is but one component of love (McAdams, 1989;

Sternberg, 1986). Sternberg's triangular theory of love

proposes that there are three components of love; intimacy,

passion, and commitment. The intimacy component seems to be

at the core of loving relationships (Sternberg & Grajek,

1984). Intimacy is necessary for both friendship and love,

and "the desire for intimacy is a fundamental psychological

need in human lives, one of a few basic needs that organize

our behavior and experience and provide our lives with

meaning" (McAdams, 1989, p. 2).






41

While Sullivan, Erikson and McAdams see intimacy as an

individual capacity, Argyle and Dean (1965) see it as both a

quality of persons and interactions. In the second view, an

individual may have the capacity for intimacy, but not

achieve it because it requires a relationship. There has

also been some debate over whether intimacy is a state or a

process, but Perlman and Fehr (1987) see intimacy as both

dynamic and static, and as both personal and situational.

Perhaps intimacy is best captured by examining both

attitudes and behaviors.

Capacity for Intimacy

The capacity for intimacy depends upon the individual

resolving their own separateness, and it is unlikely that a

person would achieve this level of development until late

adolescence or early adulthood (Erikson, 1968). In a study

of intimate relationships between college women, Rayfield,

Liabre and Stokes (1987) found that self-actualization was

the best predictor of intimate friendships. Whether

intimacy precedes the self-actualized, mentally healthy

state or whether the mentally healthy state precedes

intimacy is unknown.

The capacity for intimacy also requires the ability to

trust another person. Optimism and trust were measured in a

study by Franklin, Janoff-Bulman and Roberts (1990). Groups

of college students from divorced families and intact

families were compared. Although the groups did not differ









significantly in trust or optimism regarding present or

future dating relationships, or close friendships, the

parental divorce subjects were less optimistic about their

own future marriages. Perhaps witnessing the breakdown in

trust in their parents' marriage led to increased pessimism

and caution, or perhaps, as the authors suggest, these

children of divorce have a more realistic assessment of the

dynamics of relationships (Franklin, Janoff-Bulman &

Roberts, 1990).

Craig-Bray and Adams (1986) performed a study which

utilized a variety of measures of intimacy, including the

Orlofsky, Marcia and Lesser (1973) interview. Additionally,

subjects were asked to fill out a record of their social

interactions. The sample included 23 males and 25 females

between the ages of 18 and 22 years. Females with higher

social intimacy were found to have greater self-disclosure

and satisfaction with interactions in opposite sex contexts

(Craig-Bray & Adams, 1986). The interview and self-report

measures which were utilized were found to have strong, but

not perfect convergence, and the measures lacked strong

convergence with daily social interactions. Intimacy status

in same sex relationships paralled intimacy status in

heterosexual relationships (Craig-Bray & Adams, 1986),

suggesting that intimacy is a capacity which may depend upon

a relationship, but not on a specific type of relationship.









Intimacy Motivation

Intimate relationships were found to be preferred over

less intimate relationships, in a study by Caldwell and

Peplau (1982). When 98 college students were asked about

the types of friendships that they prefer, 73% of the men

and 83% of the women preferred having a few intimate friends

over having many less intimate friends. Similarly, Reis and

Shaver (1988) found that five of the most highly rated

friendship goals, based on the ratings of 99 college

students, also fit the definition of intimacy.

In order to study intimacy motivation, McAdams (1989)

designed a way to score the Thematic Apperception Test

(TAT), as he believed that people who were both high and low

in intimacy motivation were likely to rate themselves as

natural, warm, sincere and loving. Utilizing the TAT,

McAdams et al.(1988) explored sex differences in intimacy

motivation. Consistently, women had higher scores than men,

indicating greater desire for intimacy. Although women

seemed to have a disposition to prefer intimacy, overt

behaviors were not being evaluated. There were no

significant differences between women and men on fear of

intimacy (McAdams et al., 1988).

In a study that specifically addressed fear of intimacy

in college women, Lutwak (1985) found 31 out of 107 women to

have a high fear of intimacy. Most of the 31 students were

frightened of being hurt or taking risks. The women feared









marriage and commitment and sought security and safety.

These characteristics fit the descriptions of women in low-

intimacy status, as described in Levitz-Jones and Orlofsky

(1985). Unfortunately, Lutwak (1985) did not provide data

on parental marital status, so no information is available

on whether the students who had a high fear of intimacy came

from any particular family background.

Impact of Parental Divorce on Intimacy

While dating behaviors do not necessarily correspond to

intimacy, either absent or excessive dating may indicate

problems in the capacity for intimacy. Andrews and

Christensen (1951) found that both males and females who had

a father missing from the home had begun dating at an

earlier age, "gone steady" earlier, gotten engaged earlier,

and had more broken engagements than the group that had both

parents present. Thus there seemed to be accelerated

courtship activity when the father was absent, but the

outcomes did not necessarily involve close, committed

relationships.

Booth, Brinkerhoff and White (1984) assessed the long-

term consequences of divorce on courtship. They utilized a

questionnaire designed for their investigation. The sample

of 2538 college students included 228 whose parents'

marriage was broken by death and 365 whose parents had

separated or divorced. The children of divorce were found

to be actively involved in courtship behaviors, and as









likely to form long-term relationships as others. However,

children of divorce were more likely to engage in premarital

sexual intercourse or to be cohabiting. Additionally, when

there was a decline in relations with mother or father,

there was a significantly large percentage of the children

of divorce who reported "difficulty in dating people with

whom they could develop a serious relationship" (Booth,

Brinkerhoff & White, 1984, p. 90). When relations with the

parents were analyzed separately, those students who were

less close to their fathers reported less satisfaction with

their heterosexual relationships.

Following disruption of their parents' marriage, adult

female children of divorce have been found to "become more

solitary within interpersonal relationships" (Weiss, 1988,

p. 148). Additionally, when compared to daughters of intact

families, those from divorced families felt more distant

from their fathers. In this particular investigation, these

findings persisted over a period as long as 17 years, again

indicating that the impact of divorce extends well beyond

the adjustment period.

Gabardi (1990) performed a study of the differences

between college students from divorced and intact families

on measures of intimacy, sexual behaviors and beliefs about

relationships. Subjects were both male and female students

aged 18-25 years old. There were 185 children of intact

families and 115 children of divorce. The results indicated









that the number of sexual partners was significantly

influenced by a number of factors. Parental divorce,

parental unhappiness and parental conflict all predicted a

greater number of sexual partners. Since these same three

factors also affect the quality of the parent child

relationship, it is possible that the nature of the

relationship actually explains the most variance in the

number of sexual partners. No significant differences were

noted between college students from divorced and intact

families on measures of intimacy, relationship beliefs or

self-esteem. Students whose parents had divorced also had a

more negative view of their sociability than did the

students from intact families.

Intimacy is a construct that has been defined and

measured in different ways by different researchers.

However, there do seem to be some commonly accepted

components of intimacy, including the ability to self-

disclose, the ability to trust, and the sharing of support

and love. Intimate relationships have been found to be

preferred to other types of relationships (Caldwell &

Peplau, 1982), and women seem to prefer intimacy more than

men (McAdams, et al., 1988). Behavioral indications of

intimacy suggest that parental divorce and the absence of

the father are negative influences on the ability to

establish close, committed relationships (Andrews &

Christensen, 1951; Booth, Brinkerhoff & White, 1984;









Gabardi, 1990; Weiss, 1988). Because the ability to

establish intimate relationships influences both

heterosexual and same-sex relationships, further exploration

into the impact of divorce on interpersonal intimacy is

warranted.

Summary

The impact of parental divorce has been found to extend

well beyond the period immediately surrounding the legal

separation. In the years following divorce, not only does

the structure of the family change but family processes and

relationships also change. Of the family relationships, the

one between the father and daughter has been found to be

negatively affected, with contact becoming less frequent and

fathers becoming less available and less nurturant (Amato &

Booth, 1991; Southworth & Schwarz, 1987; Hetherington, Cox &

Cox, 1987). The impact of parental divorce also extends

beyond the family; parental divorce affects attitudes and

behaviors related to dating, courtship, and the formation of

long-lasting relationships. The daughters of divorce were

found to have dated earlier (Andrews & Christensen, 1951),

had a greater number of dating relationships (Wallerstein,

1987), had more sexual experience (Kinnaird & Gerrard,

1986), had shorter-lived relationships (Kelly, 1981), and

had lower hopes and expectations for marriage (Kalter et

al., 1985) than did daughters from intact families.

Children of divorce were also more likely to report problems









within their own marriages (Booth & Edwards, 1989; Glenn &

Kramer, 1987).

These behavioral and attitudinal indices suggest that

intimacy, or the ability to commit oneself to another,

trust, interact physically/sexually, be open and share

personal information, and the willingness to provide support

to another (Levitz-Jones & Orlofsky, 1985; Perlman & Fehr,

1987; Roscoe, Kennedy & Pope, 1987), is affected following

parental divorce. However, various measures of intimacy and

behavioral measures have not always had perfect convergence

(Craig-Bray & Adams, 1986) and studies of intimacy following

parental divorce have not always found differences between

children from intact and divorced families (Gabardi, 1990).

Since parental divorce does not necessarily involve

uniformly negative experiences, perhaps some aspect of the

divorce process, for example, the father-daughter

relationship, is more important in impacting intimacy than

is the divorce itself.

It has been suggested that father absence or a poor

relationship with the father has been linked with

apprehension and inadequate skills in relating to males

(Hetherington, 1972); with changes in the personality

development of the daughter and her feminine identification

(Biller & Weiss, 1970; Vess, Schwebel & Moreland, 1983); and

with sexual behaviors and negative attitudes toward marriage

(Kinnaird & Gerrard, 1986). It may be that the failure to








control for the nature of the father-daughter relationship

contributed to the lack of significant results in studies

that used divorce as the predictor of the ability to

establish meaningful, intimate relationships (Franklin,

Janoff-Bulman & Roberts, 1990; Gabardi, 1990).

Object relations theory provides a foundation for the

examination of the father-daughter relationship and its

impact upon intimacy as it suggests that early relationships

with significant others are internalized and impact upon the

ability to love as an adult (Dicks, 1963). Antonovsky

(1985) states that interactions with significant others are

internalized and "affect all later experiences with others"

(p. 538). Further, although object relations are formed

early in the life of a child, they continue to be modified

throughout life. Modification is likely to occur when there

is "pain in relationship with significant others"

(Antonovsky, 1987, p. 538). Parental divorce often

interferes with or interrupts the attachment to the father.

The resultant disturbance of this important object

relationship may be expected to influence interpersonal

intimacy.
Hypotheses

The goals of the current investigation include

confirmation of previous studies, which have concluded that

divorce has a negative effect upon the father-daughter

relationship, and exploration of the impact of the father-







50

daughter relationship, following divorce, upon intimacy. It

is expected that both parental divorce and the quality of

the father-daughter contact will impact the capacity for

intimacy with the opposite sex. Accordingly, the present

investigation tests the following hypotheses:

HYPOTHESIS #1: Daughters of parental divorce will be more

likely to rate their fathers lower on the Parent-Child

Relationship Survey than will the daughters from intact

families. This hypothesis tests a main effect: that the

father-daughter relationship is negatively affected by

divorce. This hypothesis was supported by the findings of

Drill (1987), Parish (1981), Parish (1991), Parish and

Osterberg (1984).

HYPOTHESIS #2: Daughters of parental divorce will score

lower than daughters from intact families on measures of

intimacy. This hypothesis tests the effect of divorce on

heterosexual intimacy irrespective of the quality of the

father-daughter relationship, which is a main effect.

Although the negative effect of divorce on intimacy was not

supported by Gabardi (1990), Weiss (1988) found daughters of

divorce to be "more solitary within interpersonal

relationships" (p.148).

HYPOTHESIS #3: Daughters who have a poor relationship with

their father will score lower on measures of intimacy than

daughters who have a good relationship with their father,

regardless of parental marital status. This hypothesis








tests the main effect of parent-child relationship on

measures of intimacy. This hypothesis was supported by

Biller and Weiss (1970) and Hetherington (1972), but was not

supported by Hainline and Feig (1978).

HYPOTHESIS #4: Daughters of parental divorce who have a poor

relationship with their father will score lower on measures

of intimacy than any other comparison group. This

hypothesis tests an interaction effect. While this

hypothesis and the one following have not yet been tested,

they are predicted based upon the support of their separate

components, as indicated above.

HYPOTHESIS #5: Daughters from intact families who have a

good relationship with their fathers will score higher on

measures of intimacy than daughters from any other

comparison group. This hypothesis tests the interaction of

parental marital status and father-child relationship on

measures of intimacy. Figure 1 illustrates the predicted

directions of the interaction effects for Hypotheses 4 and

5.

HYPOTHESIS #6: Daughters of divorce who have a poor

relationship with their father, but who have someone "like a

father" to them will score higher on measures of intimacy

than will daughters who have a poor relationship with their

father and no other significant father figure. Kinnaird and

Gerrard (1986) found that the presence of another

significant male was not predictive of behavioral,







52

attitudinal or adjustment variables, but they did not assess

the quality of the relationship. Object relations theory

suggests that the fathering figure is important, but it need

not be the biological father (Rosenberger, 1990).

Consequently, it is considered important to examine the

effect of someone who is perceived as filling the fathering

role.




High I
I
D
Intimacy


Low D

Poor Good

Father-child Relationship

Figure 1: Predicted interactions for Hypotheses 4 and 5.

Note. "D" and "I" refer to parental marital status with "D"
being divorced and "I" being intact. Scores on the Risk in
Intimacy Inventory would be reversed.












CHAPTER 3
METHODS

Subjects

Subjects for the current investigation were taken from

a sample of 379 female students enrolled in undergraduate

psychology courses. Some volunteers, particularly those who

were taking general psychology, participated for course

credit; others received no compensation or credit for their

participation. Data were also gathered from 67 male

students who filled out the questionnaires within their

psychology classes, but the data from these subjects were

not used in the tests of the primary hypotheses. When

subjects were solicited from the general psychology pool,

the notices requested females aged 17-25. Subjects were not

selected or solicited based on parental marital status or

the nature of the father-daughter relationship.

Since the purpose of the present investigation was to

ascertain the long term impact on interpersonal intimacy, of

the father-daughter relationship following divorce, only

subjects for whom information was available on parental

marital status, age of the subject at the time of divorce,

and the nature of the father-daughter relationship were

included. Subjects were placed in one of two groups, those

whose parents were divorced and those whose families of

53









origin were intact. Approximately thirty percent of the

total sample were from families of divorce and seventy

percent were from intact families. Further, subjects were

categorized as having either a "good" or a "poor"

relationship with their fathers. This categorization was

determined by the score on the Parent Child Relationship

Survey (Fine, Worley & Schwebel, 1985), as described within

this chapter, and by a single question on the demographic

form: please rate your relationship with your father, on a

scale from 1-7, with 1 being poor and seven being excellent.

Every attempt was made to either eliminate or control

for as many confounding variables as possible, as several

researchers have indicated that there are a number of

variables that might affect adjustment following parental

divorce. These covariates included gender (Kalter & Rembar,

1981); current age (Kalter & Rembar, 1981); length of time

since the divorce (Frost & Pakiz, 1990); age at the time of

divorce (Black & Sprenkle, 1991; Lopez, 1987; White,

Brinkerhoff & Booth, 1985); socioeconomic status (Biller,

1981; Edwards, 1987; Guidubaldi & Perry, 1985; Mueller &

Cooper, 1986); and race (Glenn & Kramer, 1985). In addition

to these variables, presence of another significant male

figure, sexuality and custody arrangements were also

considered important by the author.







55

Of these variables, gender and current age were judged

to be of primary importance and they were controlled by

using only female subjects within a limited age range (17-25

years of age). Another covariate which had to be controlled

was the length of time since the divorce, since only long-

term effects were being examined. Accordingly, in order to

be included in the category of children of parental divorce,

the divorce must have occurred three or more years prior to

the administration of the questionnaires.

Socioeconomic status was utilized as a covariate as it

was found to have a significant effect on one of the

intimacy measures. Information on race was gathered and

analyzed to assure that a balanced composition was

maintained in the cells. Information on the presence of

another significant male figure was collected and examined

in order to determine if it contributed significantly to the

variance in intimacy scores. Data on homosexual and

bisexual students (n=9) were eliminated from the primary

analyses, as it was considered possible that the father-

daughter relationship would have a different impact on

intimacy, depending upon sexual orientation. Finally, data

on subjects who had lost a parent due to death or who were

in the custody of their father following divorce were

excluded from the analyses. After all of these factors were

controlled, approximately 300 subjects remained. A summary

of the variables that were controlled appears in Table 1,









and the number of subjects in each cell is indicated in

Table 2.


Table 1: Description of final sample.


Gender -
Age -
Sexuality -
If parents
are divorced-


Female
Aged 17-25
Heterosexual

Mother had custody
Divorce occurred 3 or more years ago


Table 2: Number of subj .


Parental Marital
Status

Intact


Divorced


Father-Daughter Relationship

Good Poor


Procedure

Subjects were given a demographic/descriptive

questionnaire, a parent-child relationship survey and four

measures of intimacy. Surveys were administered to groups

of approximately 10-30 subjects. Prior to the

administration of the questionnaires, the informed consent

form was read. The subjects were told that the purpose of

this investigation was to learn more about interpersonal

relationships. The whole purpose of the study was not

revealed to the subjects until the debriefing, in order to

prevent any tendency to respond in the expected direction.


140 89
(47%) (30%)

11 58
(4%) (19%)


Table 2: Number. of iubiects in-each categorv






57

Within each packet, the questionnaires were randomly ordered

in an attempt to prevent any response bias, and the forms

were identified only by last four digits of the subjects'

parents' phone number so that anonymity could be maintained.

After completing the instruments, the subjects were given a

debriefing form and were allowed to ask questions.

Measures

Students completed a demographic and descriptive

questionnaire, a parent-child relationship survey, and four

measures of intimacy: the Miller Social Intimacy Scale, the

Risk in Intimacy Inventory, Rubin's Love Scale, and the

Dyadic Trust Scale. Based on the information from the

demographic/descriptive and parent-child relationship

surveys, subjects were placed in appropriate categories:

intact/good relationship; intact/poor relationship;

divorced/good relationship; and divorced/poor relationship.

Demographic Questionnaire The demographic and descriptive

questionnaire was designed to assess the parents' marital

status, the subject's current age and the age at the time of

the divorce (if applicable), some indication of the parent-

child relationship, and whether or not someone else may have

been a significant "object," like a father, in the subject's

life. Additionally, information about current dating status

and sexual orientation was solicited. A copy of the

demographic questionnaire can be found in Appendix C.

Responses to questions about parental occupations and







58

educational levels were used to code the socioeconomic level

according to the socioeconomic index and occupational

classification scheme of Stevens and Cho (1985).

Parent-Child Relationship Survey (PCRS) This paper and

pencil, self-report questionnaire, was designed to "assess

older children's perceptions of the quality of their own

parent-child relationships on a number of dimensions" (Fine,

Worley & Schwebel, 1985, p. 155). A copy of the Parent-

Child Relationship Survey can be found in Appendix D. There

are two parallel subscales, one for the mother and one for

the father. Each uses 24 Likert-type items, scored on a

seven point scale. For the purposes of this investigation,

only the Father Scale was utilized.

The psychometric properties of the Father Scale are

mainly unidimensional, measuring primarily the Positive

Affect of the father-child relationship, as perceived by the

child. Positive Affect accounts for over 50% of the

variance (Fine, Worley & Schwebel, 1985). Other factors,

which accounted for much less of the common variance, were

trust/respect; lack of anger; and father identification

(Fine, Worley & Schwebel, 1985).

Reliability and validity of the PCRS have been examined

in several studies of middle-class, college-aged students

(Fine, Moreland & Schwebel, 1983; Fine, Worley & Schwebel,

1985, 1986). Internal consistency of the Father subscale

was .96. The Father subscale also has discriminative








ability in that it distinguishes between subjects from

divorced and intact families (Fine, Worley & Schwebel,

1986).

While the range of possible scores, on the father

subscale, is 24-268, the means for women from intact and

divorced homes were 128 and 97, respectively in a study by

Fine, Moreland and Schwebel (1983). When daughters of

parental divorce filled out the subscale with their

stepfathers as the target, the mean was 114 (Sauer & Fine,

1988). For the present investigation, the mean for the

total sample (n = 440) was 114 and the median was 123. For

the approximately 300 subjects in the final sample, the mean

was 115, and the median was 126, and for the 51 subjects who

had some one "like a father", the mean for the other

significant father figure was 116 and the median was 122.

For the purpose of determining whether the father

daughter relationship would be considered "good" or "bad",

it was decided to use a median split. Although it was

thought that utilizing either the upper and lower forty

percent or the upper and lower thirds would perhaps give a

greater ability to detect differences, there was a

concomitant loss of power because of the reduction in cell

sizes. Additionally, when the different ways of

categorizing the father daughter relationship were compared,

there appeared to be little difference in the outcomes. The

PCRS was also correlated with the single item on the









demographic questionnaire which dealt with the father-

daughter relationship. That single item captured 86% of the

variance explained by the PCRS.

Miller Social Intimacy Scale (MSIS) The Miller Social

Intimacy Scale (MSIS) was designed to measure intimacy in

the context of interpersonal relationships (Miller &

Lefcourt, 1982). A copy of the Miller Social Intimacy Scale

can be found in Miller and Lefcourt (1982). In order to

develop this scale, an initial pool of thirty items was

generated from interviews with fifty students. Based on

high inter-item and item-total correlations, seventeen

intimacy items were retained for the final scale. The items

are scored on a ten point frequency and intensity scale.

Two hundred and fifty-two subjects participated in a

study designed to assess the validity and reliability of the

scale. There were 72 male and 116 female unmarried

undergraduates, a married sample of 17 couples, and a

married clinic sample of 15 couples who were seeking marital

therapy (Miller & Lefcourt, 1982). The mean MSIS score for

unmarried females was 139.3 and the mean score for married

females was 156.2. The mean of 143.59 for the sample in the

present investigation appears similar and fell within the

range of means found by Miller and Lefcourt (1982).

A Cronbach alpha coefficient of 0.91 lent strong

support to the internal consistency of the measure and

indicated that a single construct was being assessed. Test-









retest reliability was measured over a two month period.

The results indicated that there is some stability in the

maximum level of intimacy experienced (Miller & Lefcourt,

1982).

Convergent validity was explored by having subjects

also complete the Interpersonal Relationship Scale (IRS),

which assesses trust and intimacy in the marital

relationship. Those who scored high on the MSIS also scored

high on the IRS. Another group completed a measure of

loneliness and those with low scores also scored low on the

MSIS, as predicted.

Construct validity was examined by having subjects

complete the MSIS twice, describing a casual friend and then

their closest friend. The intimacy scores were

significantly higher for closest friends than for casual

friends. MSIS scores for married students were also

significantly greater than scores for unmarried students.

Finally, MSIS scores for married students were higher than

for the marreid clinic sample (Miller & Lefcourt, 1982).

These findings suggest that the MSIS is a valid and

reliable instrument for measuring social intimacy. Waring

(1985), in a review of measures of intimacy, suggests that a

problem with this instrument revolves around the issue of

whose intimacy is being measured his, hers, or theirs.

The question of whose intimacy is being measured does not

seem to be of practical importance for this study, but it is








important that this scale utilizes a broad definition of

intimacy. The MSIS is designed to recognize the

multifaceted nature of intimacy, and does not simply

focus on self-disclosure, as other scales have done (Waring,

1985).

Risk in Intimacy Inventory (RII) This inventory, with 10

items scored on a 6-point scale, was designed to measure

differences in the perception of risks associated with

intimacy (Pilkington & Richardson, 1988). A copy of the

Risk in Intimacy Inventory can be found in Appendix G. One

hundred and ninety-five female and 201 male undergraduates

completed the Risk in Intimacy Inventory (RII). Based on

their responses and a principal components factor analysis,

one principal factor was identified, which accounted for 37%

of the variance. A Cronbach's alpha of .80 indicated good

internal consistency of the measure.

Convergent validity was suggested as those who

perceived greater risk also were more likely to have lower

self-esteem and fewer close friends, and were less likely to

be currently involved in a romantic relationship (Pilkington

& Richardson, 1988). Negative correlations were also noted

between high RII scores and measures of interpersonal trust

and dating assertiveness.

A second study supported these findings with one factor

accounting for 45% of the variance and a Cronbach's alpha of

.86. Additionally, sociability and overall extraversion









correlated negatively with RII. "In general, those who

perceive risk in intimacy report attitudes (e.g. low trust)

and behaviors (e.g. low assertiveness) consistent with their

perceptions (Pilkington & Richardson, 1988, p. 507)."

Pilkington & Nezlek (1991) summed each subject's

responses to the 10 items and performed a median split.

Thus, subjects were defined as perceiving higher levels of

risk in intimacy or perceiving lower levels of risk in

intimacy. For the Pilkington and Nezlak (1991)

investigation, the median score was 22. For the present

investigation, the median score was 25. The means in two

studies by Pilkington and Richardson (1988) were 24.31 and

25.68. The mean for the current study was 25.78, which is

considered similar to those found in the prior

investigations by Pilkington and Richardson (1988).

Rubin's Love Scale (RLS) This 13-item scale was developed

to measure three components of romantic love: affiliation

and dependent need; predisposition to help; and

exclusiveness and absorption (Rubin, 1970). The items for

the scale were selected based on separate factor analyses

for responses with reference to "lovers" and "friends." The

items which loaded the highest when "lovers" were the target

formed the love scale. Sternberg and Grajek (1984) found

that the love scale focuses on the intimacy component of

close relationships rather than on the other two components

of love: commitment and passion.






64

The love scale had a high internal consistency, with a

coefficient alpha of .84 for women and .86 for men. Thus,

it seems to primarily tap into a unitary concept. There was

also some support for its discriminative validity, as the

love scale was only moderately correlated with the parallel

liking scale, r = .39 for women.

The love and liking scales were administered to 158

dating couples, and the mean scores for dating partners were

much higher than they were for friends. For women, the mean

love score for a dating partner was 89.46, and the mean love

score for a friend of the same sex was 65.27. These means

compare favorably to the mean of 84.98 found in the present

investigation. Kacerguis and Adams (1980) found the Rubin

Love Scale and the Yufit intimacy scales positively

correlated. The Love Scale also correlates significantly

with the strength scale of the Relationship Closeness

Inventory (Berscheid, Snyder, & Omoto, 1989). These results

lend support to the construct validity of this scale.

Additionally, the scores were uncorrelated with the Marlowe-

Crowne Social Desirability Scale.

Of importance to the present investigation, the Love

scale seems to measure an attitude toward a particular

person rather than a general capacity for love or intimacy.

However, high scores would seem more likely when a person is

more capable of developing intimate relationships.









Dyadic Trust Scale (DTS) This eight-item scale was

constructed from a pool of 57 items taken from previous

trust scales (Larzelere & Huston, 1980). The items which

were selected for this scale, seemed to tap into a

unidimensional construct, based on factor analysis. The

item-total correlations were high, ranging from .72 to .89.

A sample population that was not involved in the item

selection also completed the Dyadic Trust Scale. The

reliability found for that sample was .93. Additionally,

the Dyadic Trust Scale had a low correlation with social

desirability and with other generalized trust scales. Thus,

this measure appears reliable, unaffected by social

desirability, and distinct from generalized trust (Larzelere

& Huston, 1980). The overall mean for the sample in the

present investigation was 48.12

Forty dating couples, 20 newlywed couples, and 20

longer-married couples completed the Dyadic Trust Scale as

well as Rubin's Love Scale. The findings indicated a strong

relationship between dyadic trust and love, for individual

and couple scores. When comparing the longer-married

couples with the unmarried couples, the correlations were

significantly higher for the longer-married couples.

However, when the newlyweds were included in the comparison,

there was no significant difference. Larzelere and Huston

(1980) suggest that the engagement and newlywed period is









one of transition and turbulence, which would account for

lower correlations between love and dyadic trust.

Larzelere and Huston (1980) found that the mean dycdic

trust scores varied by the type of relationship being

measured, but the range was from 43.63 to 49.40 for couples

that were either dating or married. The mean score for the

present investigation was 48.04. This mean compares

favorably to the scores found by Larlelere and Huston

(1980).

Dyadic trust was also noted to be associated with self-

disclosure for both the dating and married participants.

There was some trend for these correlations to increase as

levels of commitment increased (Larzelere & Huston, 1980).

The dyadic trust scores also varied by relationship status.

Separated or divorced partners had significantly lower

scores than any other group, and the newlyweds had the

highest scores of the other groups, being significantly

higher than the dating or separated/divorced groups.

However, the differences between the dating and longer

married couples failed to reach the levels of statistical

difference.

Dyadic trust is considered to be a necessary component

of intimacy. The validity and reliability of this scale, as

well as its brevity make it a desirable addition to other

measures of intimacy. Larzelere and Huston suggest that the

"reciprocity of dyadic trust in intimate relationships is at









least as strong as the reciprocity of self-disclosure, and

apparently stronger than reciprocity of love" (1980, p.

603). The scores were expected to provide some confirmation

of the quality of the dating relationship.

In conclusion, four measures of intimacy were used in

order to assess different aspects of the construct. These

four scales are objective in format, and have good

reliability and validity. The aspects of intimacy that were

measured are summarized in Table 3.



Table 3: Summary of the Instruments and Aspects of Intimacy


Instrument


Aspect of Intimacy Measured


MSIS Social Intimacy intimacy in the context of
interpersonal relationships
RII Differences in the perception of risks
associated with intimacy
RLS Three components of romantic love:
affiliation and dependent need;
predisposition to help; and exclusiveness
and absorption
DTS Dyadic Trust in intimate relationships












CHAPTER 4
RESULTS


The data were analyzed utilizing SAS software (SAS

Institute, 1982) and the statistical procedures described

below. Univariate statistics were calculated so that

frequencies, means, medians and standard deviations could be

examined and compared. The medians on the PCRS were used to

place participants in appropriate categories for the father

daughter relationship, i.e., "good" or "poor." A Chi-Square

test was performed to ascertain whether or not the cells

were similar on racial composition.

Because of the unequal number of participants in each

category, the general linear model was used to test the

primary hypotheses. This procedure called for the

performance of two-way analyses of variance designed for

unbalanced data, with parental marital status and

relationship with the father being the predictor variables

and the scores on the four measures of intimacy as the

criterion variables. Socioeconomic status was examined for

its contribution to the variance in intimacy scores, and the

impact of divorce on socioeconomic status was also explored.

Once statistically significant results were observed in the

general linear model suggesting the contribution of a









significant proportion of the variance by one of the

independent variables, simple effects tests were utilized to

provide a clearer picture of the actual differences.

Finally, a factor analysis was performed on the items in the

measures of intimacy, in order to determine that there were

indeed four factors, corresponding to the four intimacy

scales.

Frequencies and Univariate Statistics for Parent-Child
Relationships

Participants were asked to rate the quality of their

relationship with their father and then to rate the quality

of their relationship with their mother on the demographic

questionnaire. The rating of these parent-child

relationships was on a seven point Likert scale, with 1

being poor and 7 being excellent. Using the entire sample,

responses to these items were examined and the distributions

are shown in Figures 2 and 3. Not only were there apparent

differences in the distribution patterns, there were also

differences in the mean ratings, E ( 1, 438) = 66.1, R <

.0001. The mean for mothers was 6.0 and for fathers was

4.9. Even when parental marital status was not a factor,

the relationship with the mother was rated significantly

higher than the relationship with the father.










180

Frequency 160
(n=443)
140

120

100

80

60

40

20 120-

0-8
1 2 3


Figure 2: Relationship with Mother


4 5 6 7
Rating


Frequency
(n=438)


180

160

140

120

100

80

60

40

20

0


-100-


-110-


-54-

-32- -28-- 27



1 2 3 4 5 6 7
Rating


Figure 3: Relationship with Father


-87-









The Chi-Square Test of Racial Composition

The Chi-Square Test for differences in racial

composition in the cells revealed no significant

differences, F ( 9, 248) = 15.7, R < .074. These findings

indicated that it was not necessary to use racial/ethnic

background as a covariate in further analyses. The

composition of each of the cells is indicated in Table 4.


Table 4: Racial/ethnic composition of cells, by parental
marital status and relationship with father.

Int/Poor Int/Good Div/Poor Div/Good

Caucasian 59 110 48 8
(75.5%)

African American 8 8 5 2
(8%)

Hispanic 8 13 2 1
(8%)

Asian 12 9 3
(8%)

Other 6
(<1%)

Note: "Int" and "div" refer to the parental marital status,
with "int" being intact and "div" being divorced; good and
poor refer to the nature of the father-daughter
relationship.

Socioeconomic Status

Socioeconomic status (SES) was entered into the general

linear model as a covariate in order to see if it

contributed significantly to the variance in intimacy

scores. Scores on the Dyadic Trust Scale were significantly

lower when SES was lower, F ( 1, 281) = 4.2, R < .042.







72

Additionally, findings of an apparent interaction of marital

status and the father-daughter relationship on Rubin's Love

Scale disappeared when SES was included. Participants from

divorced and intact families differed significantly on

socioeconomic status, F (1, 286) = 6.5, R < .012. Intact

families had a significantly higher SES than divorced

families. The mean score for intact families (n = 220) was

58.3 and for divorced families (n = 67) was 50.6. Based on

these observations, SES was included as a covariate in the

analyses of the primary hypotheses.

Analyses of Variance for Primary Hypotheses

Relationship with Father

The first hypothesis predicted that daughters of

parental divorce would rate their fathers lower on the PCRS

than would daughters from intact families. This hypothesis

was supported, F (1,296) = 119.7, R < .0001. The mean for

daughters of divorce (n = 69) was 82.3 and for daughters

from intact families (n = 228) was 126.5. For this sample,

divorce was associated with a reduction in the quality of

the relationship with the father.

Given these findings, it was decided to further examine

the father-child relationship as affected by divorce, and to

determine whether the relationship with the mother was

similarly affected. Because the single item regarding the

relationship with the father on the demographic

questionnaire correlated 86% with the father subscale of the









PCRS, this item was considered to be a reasonable way to

further examine the parent-child relationship.

Additionally, because the mother subscale of the PCRS had

not been utilized, it was decided to use the single question

to compare the impact of divorce on the relationships with

both the father and mother. There was a significant

interaction effect between parental marital status and the

relationships with parents, E (1,296) = 67.8, R < .0001.

The mean ratings of relationships with fathers and mothers,

by parental marital status, are shown in Figure 4.

7

6 --6.0--
Mean -5.4- -- 5.6-
Ratings 5
of
Fathers 4
and -3.4-
Mothers 3

2

1

Father Mother Father Mother
Intact Divorced

Figure 4: Mean ratings of relationships with fathers and
mothers, by parental marital status

Simple effects tests were then conducted to determine

if the observed differences were significant at each level

of marital status. When the parental marriage was intact,

the relationship with the father was rated significantly

lower than the relationship with the mother, E (1, 228) =

43.3, R <.0001. When parents were divorced, the differences







74

were also strong, F (1, 69) = 79.3, R = .0001. The ratings

of mothers from divorced and intact families did not differ

significantly, F (1, 297) = 2.91, p < .089, but fathers from

divorced families were rated significantly lower than

fathers from intact families, F ( 1, 297) = 94.9, R < .0001.

The results regarding fathers' ratings agree with the

findings based on the PCRS.

Parental Marital Status and Intimacy

The second hypothesis predicted that daughters of

divorce would score lower on measures of intimacy than

daughters from intact families. This hypothesis was not

supported on three of the measures of intimacy: the Rubin

Love Scale (RLS), F (1, 293) = .02, R < .88; the Dyadic

Trust Scale (DTS), F (1, 292) = .80, R < .37; and the Miller

Social Intimacy Scale (MSIS), F (1, 294) = 1.1, R < .30.

Findings on the Risk in Intimacy Inventory (RII) did reach

the level of statistical significance, F (1, 297) = 5.0, R <

.026. The means on each of these measures are indicated in

Table 5. It appears that daughters of divorce perceive more

Table 5: Means on intimacy scales, by parental marital
status.


Note. indicates statistically significant results.


RLS DTS

Divorced Intact Divorced Intact
86.0 84.1 48.4 4.8

MSIS RII*

Divorced Intact Divorced Intact
145.5 142.9 28.4 25.0







75

risks associated with intimacy, but daughters from divorced

and intact families do not differ significantly on other

measures of trust and intimacy.

Father-Daughter Relationship and Intimacy

The third hypothesis predicted that daughters who had a

"poor" relationship with their fathers would score lower on

measures on intimacy than daughters who had a "good"

relationship. This hypothesis was not supported by the

data, as the relationship with the father contributed little

to the variance in intimacy scores. The findings for the

four intimacy scales are as follow: Rubin Love Scale (RLS),

F (1, 293) = 2.1, R < .15; Dyadic Trust Scale (DTS), F (1,

292) = 2.2, R < .14; Miller Social Intimacy Scale (MSIS), E

(1, 294) = .77, R < .38; and the Risk in Intimacy Inventory

(RII), F (1, 297) = 0.0, R < .97. The means for each of the

intimacy measures are shown in Table 6.



Table 6: Means on intimacy scales, by father-daughter
relationship.


RLS DTS

Good Poor Good Poor
83.9 85.7 49.7 46.9

MSIS RII

Good Poor Good Poor
144.7 142.3 24.7 26.9









Interaction of Father-Daughter Relationship and Parental
Marital Status

The fourth and fifth hypotheses examined interaction

effects. The fourth Hypothesis predicted that daughters of

divorce with a poor relationship with their fathers would

score the lowest on measures of intimacy. This hypothesis

was not supported by the data as there was no significant

interaction between marital status and father daughter

relationship. The findings for the four intimacy scales

were as follow: Rubin Love Scale (RLS), (1, 293) = 2.5, R

< .11; Dyadic Trust Scale (DTS), F (1, 292) = .03, R < .87;

Miller Social Intimacy Scale (MSIS), (1, 294) = .03, E <

.85; and Risk in Intimacy Inventory (RII), E (1, 297) = 1.2,

R = .28. The means for the tests of interaction are given

in Table 7.



Table 7: Means on intimacy scales, by father daughter
relationship and parental marital status.


RLS DTS

Intact Divorced Intact Divorced
Good Poor Good Poor Good Poor Good Poor
84.3 83.8 78.6 88.4 49.7 46.4 51.1 47.8

MSIS RII

Intact Divorced Intact Divorced
Good Poor Good Poor Good Poor Good Poor
144.4 140.4 147.6 145.0 24.2 26.1 30.1 28.1



The fifth hypothesis also predicted an interaction

effect, where the daughters from intact families who had a







77

good relationship with their fathers would score higher than

any other comparison group. Again, there were no

significant effects noted. The findings and means are noted

above, under Hypothesis 4 and are shown in Table 7. It

appears that the interaction of father daughter relationship

and parental marital status contributes little to

differences in the four measures of intimacy.

Impact of Having Someone "Like a Father"

The sixth hypothesis predicted that, of daughters who

rated their relationship with their father as poor, those

who have "someone like a father" would score higher on the

intimacy scales than those who do not have another

significant father figure. This hypothesis received mixed

support from the data: it was not supported on three of the

measures of intimacy: the Rubin Love Scale (RLS), F (1, 57)

= 1.1, R < .31; the Dyadic Trust Scale (DTS), F (1, 57) =

.28, R < .60; and the Miller Social Intimacy Scale (MSIS), E

(1, 67) = .13, E < .72. Findings on the Risk in Intimacy

Inventory (RII) did reach the level of statistical

significance, F (1, 57) = 5.1, R < .03. Daughters of

divorce with a poor relationship with their father perceived

higher levels of risk associated with intimacy when there

was no other significant father figure. The means for the

test of the sixth hypothesis are indicated in Table 8.

The findings for the RII were further explored by

comparing daughters from intact families with daughters of







78

divorce who had someone else like a father and those who did

not have another significant father figure. A main effect

for group was observed, F (2, 297) = 4.5 p < .01. The

means are indicated in Figure 5.



Table 8: Means on intimacy scales, by the presence or
absence of someone like a father.

RLS DTS

No Yes No Yes
83.2 89.0 49.1 47.2

MSIS RII*

No Yes No Yes
146.9 144.8 30.1 24.0

Note. indicates statistically significant results.


Mean Scores on the
RII
(Higher = More
Perceived Risk)


60 I

32

30

28


Intact Div/None Div/With

Figure 5: Mean scores on the RII by parental marital status
and presence of a significant father figure.


Follow-up comparisons were conducted to isolate the

effect of group. Comparing those participants whose parents

were divorced and who had someone like a father'with those

whose parents were divorced with no other significant father









figure, the differences were statistically significant,

E ( 1, 68) = 4.1, R < .04, with the daughters who lacked

someone like a father perceiving more risk associated with

intimacy. The daughters whose parents were married also

perceived significantly less risk associated with intimacy

than did those whose parents were divorced with no one else

like a father, f ( 1, 262) = 8.9, p < .003. When comparing

those whose parents were married and those whose parents

were divorced with another father figure present, there was

no significant difference noted, F ( 1, 263) = .10, E < .75.

The presence of another significant father figure following

parental divorce seems to reduce the perception of risk

associated with intimacy.

Factor Analyses of Intimacy Scales

The Scree plot, shown in Figure 6, of all items on the

four questionnaires indicated that there were indeed four

factors. The four factor solution was rotated using a

promax rotation because there was some correlation between

the scales, as shown in Table 9. The factor analysis,

including all scores greater than an absolute value of .2,

is shown in Table 10. In large part, each factor

represented one of the intimacy scales. Each item

correlated with one of the scales, with the exception that

the fourteenth item of the MSIS correlated with the second

factor.
















































89012
34567 89012
3456789012 3456789012 3
45678


0 5 10 15 20
Numbe


25 30 35 40 45


Figure 6:


Scree plot of Eigenvalues


TPah1e Q! Correlation


analysis of the intima .


RLS
RLS 1.0
DTS
MSIS
RII


DTS
.35
1.0


MSIS RII
.62 -.22
.47 -.27
1.0 -.38
1.0


- ar--e 6: Scree .....of iaenva u


Table 9: Correlation analvsis of the intimacv scales.









Table 10: Factor Analysis Using a Promax Rotation
Factor 1 Factor 2 Factor 3 Factor 4
RLS 1 .57716 -
RLS 2 .40188 .36969 -
RLS 3 .32448 -
RLS 4 .65962 -
RLS 5 .63370 -
RLS 6 .75504 -
RLS 7 .78488 -
RLS 8 .25155 .59526 -
RLS 9 .69333 -
RLS 10 .62745 -
RLS 11 .56193 -
RLS 12 .54140 -
RLS 13 .69669 -
DTS 1 .70511 -
DTS 2 .72139 -
DTS 3 .79015 -
DTS 4 .84766 -
DTS 5 .84187 -
DTS 6 .71402 -
DTS 7 .83442 -
DTS 8 .77816 -
MSIS 1 .78792 -
MSIS 2 .29523 -.22879
MSIS 3 .81108 -
MSIS 4 .57546 -.25041
MSIS 5 .34903 -
MSIS 6 .76892 .25787 -
MSIS 7 .82179 -
MSIS 8 .69403 -
MSIS 9 .78929 .28110 -
MSIS 10 .75565 -
MSIS 11 .67330 .39775 -
MSIS 12 .83176 -
MSIS 13 .51260 .22130
MSIS 14 .50464 -
MSIS 15 .72773 -
MSIS 16 .75011 -.25509
MSIS 17 .66629 .22474 -
RII 1 .77797
RII 2 -.64279
RII 3 .75524
RII 4 .20291 .56562
RII 5 .78707
RII 6 -.24088 .51288
RII 7 .74167
RII 8 .65331
RII 9 .84575
RII 10 .50640

Note: Only correlations of +/- .20 were shown.













In conclusion, three of the measures of intimacy,

Rubin's Love Scale, the Dyadic Trust Scale, and the Miller

Social Intimacy Scale correlated positively with each other

and correlated negatively with the Risk in Intimacy

Inventory. Also, the intimacy scales seemed to represent

different factors. The correlations and factor analysis

seem to indicate that the scales tap into different aspects

of intimacy, and they support the decision of this author to

utilize all four scales.












CHAPTER 5
DISCUSSION


The present investigation looked first at the impact

that parental marital status has on the father-daughter

relationship. The relationships between fathers and

daughters were found to be generally rated lower than those

between mothers and daughters, regardless of marital status.

Furthermore, relationships between fathers and daughters

were found to be significantly negatively affected by

parental divorce. These findings supported the conclusions

of several other investigations (Drill, 1987; Parish, 1981;

Parish, 1991; Parish & Osterberg, 1984; Parish & Wigle,

1985). Given the expected reduction in the quality of the

father-daughter relationship following divorce, it was not

surprising that daughters of divorced parents who had a good

relationship with their fathers were the least-frequently

found participants in this investigation. Only eleven

daughters (16% of the daughters whose parents were divorced)

rated the relationship with their fathers as good. This

seems a disturbing statistic, but it is important to

remember that all of these participants were in the custody

of their mothers following the divorce of their parents.






84

Because of the custodial arrangement, it may not have

been totally unexpected to find that only 16% of the

daughters did have a good relationship with their fathers,

because children of divorce who are in the custody of their

mothers may have less contact with their fathers. Amato and

Booth (1991) reported that "both males and females from

divorced families had less contact with their fathers, but

the difference was considerably greater for females than for

males" (p. 903), and Southworth and Schwarz (1987) found

that females who had less frequent contact perceived their

fathers as less accepting and more inconsistent. In the

present investigation, all participants whose parents were

divorced and who were included in the final sample were

likely to have had some level of restriction on the amount

of contact with their fathers. Consequently, it is possible

that the eleven fathers who received higher ratings gave

some extra attention to maintaining a good relationship.

More often, children are not that fortunate, so there is

still some concern that, within this sample, 84% of the

female participants whose parents were divorced were not

satisfied with their relationship with their fathers. The

percentage of participants who rated the father-daughter

relationship as poor may not be unexpected, but it still

seems undesirable because the father is important to the

development of the child.








The second area of investigation was the effect of

parental marital status on intimacy. The prediction was

that daughters of divorce would score lower on measures of

intimacy than would the daughters from intact families.

Four intimacy scales, which measured specifically love,

dyadic trust, social intimacy and risk in intimacy, were

used. Four scales were selected in order to examine the

commonly accepted components of intimacy, including the

ability to self-disclose, the ability to trust, the ability

to establish a close committed relationship, and the sharing

of support and love. The prediction that marital status

would affect intimacy was not wholly supported. There were

no significant differences on Rubin's Love Scale, Dyadic

Trust Scale, or Miller Social Intimacy Scale. These results

supported the findings of Greenberg and Nay (1982) that

children from divorced and intact families were not

significantly different in level or quality of dating

behavior or in attitudes to marriage, but they contradict

the findings of Booth, et. al. (1984) and Hetherington

(1972) that parental divorce affected the quality of

courtship relations. The data also did not support the

conclusions of Weiss (1988), who found that daughters of

divorce tended to become "more solitary within interpersonal

relationships" (p. 148).

Although parental divorce did not seem to have an

effect on love, dyadic trust, or social intimacy, the









results did suggest that divorce was related to a higher

perception of risk in intimacy. Lutwak (1985) found that

fear of intimacy was related to fear of marriage and

commitment, supporting the conclusions of Carson, Madison

and Santrock (1987) that adolescents from divorced families

were more apprehensive about entering into marriage. Booth

and Edwards (1989) and Carson, Madison and Santrock (1987)

also indicated that the perception of risk and the fear of

intimacy may be linked to a greater willingness to break off

an unsatisfying relationship. Based on these earlier

findings, the higher level of risk perceived among the

daughters of divorce may be predictive of the

intergenerational transmission of divorce.

Trust is one component of intimacy. Pilkington and

Richardson (1988) found the perception of risk to be

associated with lower levels of trust. The present

investigation also tended to support that association, as

the scores on the Dyadic Trust Scale, which measures dyadic

trust in intimate relationships, were negatively correlated

(r = -.27) with the RII scores. As the perception of risk

increased, the level of dyadic trust decreased. Despite the

correlation, it is important to remember that parental

marital status did not have a statistically significant

impact on the score on the Dyadic Trust Scale.

The impact of the father-daughter relationship on

intimacy was explored next. The relationship with the









father was found to have little impact on the intimacy

measures. This finding contradicts that of Booth,

Brinkerhoff and White (1984) who concluded that students who

were less close to their fathers reported less satisfaction

with heterosexual relationships. The lack of significant

findings regarding the impact of the father-daughter

relationship would tend to contradict the expectation that

pain in the relationship with the father would impact

interpersonal relationships. This would also seem to refute

the importance of the father as a significant object in the

life of the child, but it may be that when the father is

"lost" because of divorce, someone else can fill his role,

as in parental remarriage.

Because a fathering figure may have contributed to the

variance, the presence of someone "like a father" was

examined for its impact on interpersonal intimacy. The

presence of a significant father figure, other than the

biological father, was found to be associated with a lower

perception of risk in intimate relationships, so the

daughters of divorce who had a significant father figure

perceived similar levels of risk similar to those found in

daughters of intact families. Those with no one "like a

father" perceived significantly more risk associated with

intimacy. There were no significant results noted on any of

the other measures of intimacy. In the majority of cases

(68%), the person who was "like a father" was the









stepfather, but grandfathers and others were included as

well. These findings support the suggestion by object

relations theorists that the fathering figure is important,

but that he need not be the biological father (Rosenberger,

1990).

It was also noted that socioeconomic status was

significantly lower in families of divorce than in intact

families. This observation supports the conclusions of

previous investigations that it is important to control for

socioeconomic status (Biller, 1981; Edwards, 1987;

Guidubaldi & Perry, 1985; Mueller & Cooper, 1986). Although

socioeconomic status was not the subject of this study, it

may have a "bearing on socialization practices and parent-

child relationships" (Edwards, 1987, p. 361). For example,

the amount of time spent with children and the nature of

interactions and relationships may vary with socioeconomic

status.

One final observation was that several scales were

needed to capture the construct of intimacy. The data

indicated that there were four factors in the intimacy items

used in the present investigation. This seems to confirm

that intimacy is indeed a complex construct, but it does not

necessarily indicate that the four scales, measuring love,

dyadic trust, social intimacy, and risk in intimacy,

captured all of its components. Alternatively, intimacy may

actually be less complex, and these scales may have been









measuring different factors rather than components of one

factor. While this is not considered to be a likely

explanation, it is possible.

Conclusions

It is certainly clear that divorce has a negative

impact on the relationship with the father, but it is

unclear, at this point, exactly what the ramifications are

for the changes in that significant relationship. The

present investigation sought to explore the impact on

interpersonal intimacy, as the absence of the father from

the home or even from the life of his daughter was thought

to be associated with her feeling less secure or adequate in

intimate interpersonal relationships.

Neither parental divorce nor the father-daughter

relationship were associated with significant differences in

love, dyadic trust, or social intimacy, as measured in this

study. Biller (1981) had concluded that divorce uniquely

affected the daughter, as father absence had an effect on

her ability to function in interpersonal and heterosexual

relationships. The present investigation would tend to

indicate that this is not necessarily the case, as daughters

of divorce appear to be as satisfied in their relationships

as are their counterparts from intact families. The one

significant negative finding relative to intimacy was that

the daughters of divorced parents perceived more risk

associated with intimacy. When this effect was further






90

examined, it was found that those participants whose parents

were divorced and who had no other significant father figure

perceived the highest levels of risk. The presence of

someone "like a father" apparently was associated with a

reduction of the perception of risk in intimacy.

The expectation was that something about the

relationship with the father, perhaps communication,

emotional support, reliability, encouragement, or some other

interpersonal factor, would create differences in the

ability to establish meaningful intimate relationships.

Leonard (1966) had suggested that the father's

unavailability to give love and to be loved was critical to

the daughter's development, but it is also important to

remember that the presence of the father in the home is

likely to provide more economic stability. The provision of

economic stability was evidenced by the significant

differences between participants from divorced and intact

families on socioeconomic status, and by the significant

effect of socioeconomic status on the scores on the Dyadic

Trust Scale. The reason why participants who had someone

"like a father" perceived less risk with intimacy might have

been because his presence provided financial as well as

emotional stability. Further study could indicate whether

economic or interpersonal factors contribute more to the

perception of risk in intimacy.









Limitations of the Study

The study's limitations will be discussed and related

to threats to the following different forms of validity:

internal, statistical conclusion, external, and construct.

Although attempts were made to reduce the threats to

validity to a minimum, no research project is perfect.

Gelso (1979) suggested that all research designs are flawed,

but that the researcher can be cognizant of the flaws in the

chosen design. Accordingly, an attempt will be made to

delineate and describe the possible threats to validity in

the present investigation.

Because the present investigation was a correlational

design, threats to internal validity should be considered.

Internal validity refers to the confidence that can be

placed in the association between the predictor and

criterion variables and to the ability to rule out

alternative explanations for the observed relationships. In

a correlational design, variables are not manipulated, and

there may be extraneous factors that influence the results.

History, or life events or experiences since the divorce,

might have affected the observations in the present study.

For example, an alternative explanation of the results could

have been that subjects experienced depression as a result

of parental divorce and that depression made them more

pessimistic about relationships and more fearful. Although

a number of variables were controlled in this study, it








would have been impossible to eliminate all of the threats

due to history.

Inadequate statistical power may have reduced the

probability of obtaining statistically significant results,

creating a threat to statistical conclusion validity

(Heppner, Kivlighan & Wampold, 1992). Despite attempts to

obtain a large enough sample, only eleven participants were

from divorced families and also had a good relationship with

their fathers. The paucity of participants in this group

may have reduced the power of the current investigation to

detect differences between the groups. There were some

marginally significant results, not always in the expected

direction, that might have reached the level of statistical

significance, if the number of participants in this group

were higher.

One possible way to have increased the number of

participants in the divorced/ good relationship with father

group would have been to include participants who were in

the custody of their fathers (14.6% of the original sample).

These participants would have been more likely to have had a

better relationship with their fathers than those in the

custody of their mothers. However, paternal custody may

have produced different childhood experiences for the

daughters. In addition, the father might have been awarded

custody because of an existing poor mother-child

relationship. In either of these two cases, the random