Perceived social support, social skills, and quality of relationships in bulimic women

MISSING IMAGE

Material Information

Title:
Perceived social support, social skills, and quality of relationships in bulimic women
Physical Description:
viii, 115 leaves : ill. ; 29 cm.
Language:
English
Creator:
Grissett, Nadine I., 1964-
Publication Date:

Subjects

Subjects / Keywords:
Bulimia -- psychology   ( mesh )
Interpersonal Relations   ( mesh )
Social Support   ( mesh )
Women   ( mesh )
Clinical and Health Psychology thesis Ph.D   ( mesh )
Dissertations, Academic -- Clinical and Health Psychology -- UF   ( mesh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1991.
Bibliography:
Bibliography: leaves 103-114.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Nadine I. Grissett.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 025734222
oclc - 25487823
System ID:
AA00022538:00001

Table of Contents
    Title Page
        Page i
        Page ii
    Dedication
        Page iii
    Acknowledgement
        Page iv
    Table of Contents
        Page v
        Page vi
    Abstract
        Page vii
        Page viii
    Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
    Method
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
    Results
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
    Discussion
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
    Appendix A. Questionnaires
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
    Appendix B. Dyadic effectiveness scale
        Page 98
        Page 99
    Appendix C. Videotaped interaction
        Page 100
        Page 101
        Page 102
    References
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
    Biographical sketch
        Page 115
        Page 116
        Page 117
Full Text























PERCEIVED SOCIAL SUPPORT, SOCIAL SKILLS, AND
QUALITY OF RELATIONSHIPS IN BULIMIC WOMEN








BY

NADINE I. GRISSETT


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


1991

































Copyright 1991

by

Nadine I. Grissett


























To my husband, Sean McCallum,

whose love, support, and encouragement

made these past several years

much more worthwhile.


iii














ACKNOWLEDGMENTS


I would first like to express my gratitude to my

chairperson, Dr. Nancy K. Norvell, who has provided me

immeasurable inspiration and encouragement throughout my

graduate career. Her support and faith in my abilities kept

me going at times when my own confidence ebbed, and

challenged me to even higher goals, both personally and

professionally. I would also like to thank my committee

members, Dr. James Johnson, Dr. Anthony Greene, and Dr.

Jaquelin Goldman, for their valuable suggestions, support,

and continued interest in my professional development. In

addition, I acknowledge Dr. John Kuldau for his time and

input into this project. Last, but definitely not least, I

would like to thank my husband, Dr. Sean McCallum, for his

listening ear and continued support.















TABLE OF CONTENTS



PaQe

ACKNOWLEDGMENTS ......................................... iv

ABSTRACT ................................................ vii
INTRODUCTION ............................................... 1

Factors Contributing to the Etiology and
Maintenance of Bulimia ................................. 6
Social Maladjustment ................................... 13
Social Support and Bulimia ........ .................. 14
Recent Developments in the Social Support Literature .... 16
Social Support and the Stress Process ................... 23
The Relationship of Stress and Social Support
in Bulimia ............................................ 25
Research Aims and Hypotheses ............................ 32

METHOD ...................................... 35

Subjects ......................................... ....... 35
Measures ............................................... 36
Procedure ..................................... 43
RESULTS ............ *.......................*................ 45

Approach to Data Analysis ........................ 45
Comparison of Groups .................................. 47
Demographic and Descriptive Information .............. 47
Correlations ............................ 47
Perceived Social Support and Social Interactions ..... 50
Quality of Relationships ............................. 51
Social Effectiveness and Self-Reported
Social Competence ............................. 52
Psychopathology ...................................... 54
Covariate Analyses ...................................... 54

DISCUSSION ........................ ......... ... .... 65

The Social Network: Quality and Type of Interactions...66
Social Competence and Effectiveness ..... ........... 68
Implications and Conclusions......................... 70

APPENDIX A QUESTIONNAIRES ............ .... ... ........ 80










APPENDIX B DYADIC EFFECTIVENESS SCALE ................... 98

APPENDIX C VIDEOTAPED INTERACTION .............................. 100

REFERENCES ..................... ............................................... 103













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

PERCEIVED SOCIAL SUPPORT, SOCIAL SKILLS, AND
QUALITY OF RELATIONSHIPS IN BULIMIC WOMEN

By

Nadine I. Grissett

May 1991

Chairperson: Nancy K. Norvell, Ph.D.
Major Department: Clinical and Health Psychology

The emerging consensus among investigators of bulimia

nervosa suggests that this is a multidetermined disorder.

Biological, sociocultural, personality, and family factors

appear to contribute to the development and maintenance of

the bulimic individual's symptoms and psychopathology.

Several studies have suggested that the relationship between

bulimics and their environment is impaired. Although social

maladjustment, lack of perceived social support, and

distressed interpersonal relationships seem to be important

risk factors for bulimia, little research has addressed this

directly. The present study explored specific aspects of

the bulimic's social support network, as well as individual

difference variables which might mediate her ability to

obtain support or to perceive this as adequate.

Twenty-one bulimic women were matched with twenty-one

normal controls and completed a number of self-report

vii







questionnaires assessing perceived social support, the

quality of interactions and relationships, and social

competence, as well as psychopathology. They also

participated in a videotaped interaction which was rated for

social effectiveness by observers. It was hypothesized that

bulimics would report significantly less perceived social

support, significantly more negative social interactions and

poorer quality of relationships, and would demonstrate

significantly poorer social skills.

Results strongly supported all three hypotheses in that

bulimic women, as compared to non-eating disordered women,

reported significantly less perceived social support from

both friends and family. In addition, they reported

experiencing less positive interactions, and more negative

interactions and conflict, particularly with family members.

Finally, bulimics reported feeling less socially competent

in a variety of situations and were rated as less socially

effective by observers unaware of their group membership.

These differences were not due to the subjects' differing

levels of psychopathology, although this variable did affect

the report of perceived social support. These results have

implications for treatment, suggesting that learning

communication, coping, and problem-solving skills may be

particularly important for bulimic women. Future research

should explore the bulimic's relationships in more detail,

particularly elements of interpersonal dysfunction,

conflict, and other aspects of her social support system.

viii














INTRODUCTION


During the past 20 years, bulimia, literally

meaning "ox hunger," has become an increasingly well-

known psychophysiologic disorder. This syndrome refers

to episodes of uncontrollable binge eating followed by

purging methods such as self-induced vomiting,

excessive use of laxatives and/or diuretics, fasting,

and excessive exercise. Although bulimia is an eating

disorder that is widely believed to be of recent

origin, attempts to understand and conceptualize

bulimia date back several hundred years (Stein &

Laakso, 1988). Nevertheless, this disorder has

increased greatly in prevalence during recent years,

and changes have been made in the symptoms seen as

constituting the syndrome.

The contemporary concept of bulimia began in the

mid-1950's with descriptions of patients with excessive

appetite for food and exaggerated hunger, and

Stunkard's (1959) construct of binge eating among obese

patients. It was not until the 1970's, however, that

bulimia came to be recognized as a distinct clinical

entity. Researchers and clinicians used a number of

terms as they sought for a way to describe the symptoms

of the bulimic syndrome -- an abnormal increase in the
1








sensation of hunger, compulsive eating (Rau & Green,

1975), the dietary chaos syndrome (Palmer, 1979),

bulimia nervosa (Russell, 1979), and bulimarexia

(Boskind-White & White, 1983).

These different terms reflect the course of

developing knowledge about bulimia in the last two

decades. Bulimic characteristics were initially

investigated during this time by researchers who noted

their presence in a number of anorectics (Beumont,

George, & Smart, 1976; Casper, Eckert, Halmi, Goldberg,

& Davis, 1980; Pyle, Mitchell, & Eckert, 1981; Russell,

1979). Gradually it became obvious that bulimia was in

many cases a separate disorder, occurring with greater

frequency among individuals with no prior history of

eating difficulties (Halmi, Falk, & Schwartz, 1981;

Hawkins & Clement, 1980; Pyle et al., 1981).

The third edition of the Diagnostic and

Statistical Manual was the first to classify bulimia as

a distinct eating disorder (DSM-III, American

Psychiatric Association, 1980). The symptoms required

for a diagnosis of bulimia were basically consistent

with the historical concept, including recurrent

episodes of binge eating, purging, lack of control, and

affective disturbance. Subsequent research indicated

that bulimic symptoms are common in both student and

nonstudent populations, so frequency criteria became

necessary to distinguish between bulimic symptoms and








the syndrome of bulimia. As a result, the Work Group

on Eating Disorders for the DSM-III-R (1987) proposed a

minimum frequency of binging and purging of twice per

week for at least three months. This criterion has not

yet been empirically validated.

Diagnostic criteria for bulimia in the DSM-III-R

include:

A. Recurrent episodes of binge eating (rapid
consumption of a large amount of food in a
discrete period of time).

B. A feeling of lack of control over eating
behavior during the eating binges.

C. The person regularly engages in either self-
induced vomiting, use of laxatives or diuretics,
strict dieting or fasting, or vigorous exercise in
order to prevent weight gain.

D. A minimum average of two binge eating episodes
a week for at least three months.
E. Persistent overconcern with body shape and
weight.

Most published epidemiological work has surveyed

college or high school students, with the resulting

problems of definition and other problems inherent in

the use of questionnaires (Mitchell & Eckert, 1987).

Early prevalence estimates reported that bulimia

affects between 8 and 19% of college women (Halmi et

al., 1981; Pyle, Halvorson, Neuman, & Mitchell, 1986),

but recent studies based on more restrictive criteria

suggest that rates of clinically significant bulimia in

this population are only 1 5% (Cooper, Charnock, &

Taylor, 1987; Drewnowski, Yee, & Krahn, 1988; Hart &








Ollendick, 1985; Mitchell & Eckert, 1987; Schotte &

Stunkard, 1987). The disorder is much more common

among female students than it is among working women

(Hart & Ollendick, 1985) or males (Halmi et al., 1981),

and appears to occur more frequently in whites than in

blacks, perhaps due to their higher socioeconomic

status (Mitchell & Eckert, 1987). Descriptive studies

up to this point indicate that bulimics are generally

single, white, well-educated young women in their

twenties who begin binge eating in their late teens

(Boskind-Lodahl & White, 1978; Herzog, 1982; Johnson,

Stuckey, Lewis, & Schwartz, 1982; Pyle et al., 1981).

However, further epidemiological studies are needed

which include samples of both urban and rural groups,

as well as multiple racial and ethnic groups (Mitchell

& Eckert, 1987). Bulimic symptoms often follow a

period of dieting which may have been prompted by the

suggestion of friends or family, traumatic events,

weight gain, increased interest in the opposite sex, or

identity confusion (Abraham & Beumont, 1982; Gandour,

1984; Johnson et al., 1982; Pyle et al., 1981). The

frequency of binge eating and purging may vary

considerably (Fairburn, 1980; Halmi et al., 1981; Pyle

et al., 1981; Russell, 1979), although DSM III-R

criteria now require that an individual engage in this

behavior at least twice per week to be diagnosed as

bulimic.








Bulimia occurs among all weight groups, although

most bulimics are of normal weight for their height and

age, or slightly above or below this average (Fairburn,

1981; Herzog, 1982; Johnson et al., 1982). The

majority also report a large discrepancy between this

average or healthy weight, with the desired weight

significantly lower than the healthy weight (Pyle et

al., 1981; Russell et al., 1979). In general, bulimia

is accompanied by negative emotions such as guilt,

anxiety, and depression (Abraham & Beumont, 1982; Pyle

et al., 1981). Serious medical complications may also

arise, including gastrointestinal disturbances,

hypokalemia (potassium deficiency), dental decay,

electrolyte imbalances, dehydration, menstrual

irregularities, and neurological and cardiac

abormalities (Abraham & Beumont, 1982; Goode, 1985;

Pyle et al., 1981; Russell, 1979).

A large body of research has been generated by

researchers investigating factors contributing to the

onset and perpetuation of bulimic behavior. Although

this literature continues to grow, there is an emerging

consensus among investigators that bulimia is a

multidetermined disorder. In order to gain a thorough

understanding of our knowledge thus far, a number of

factors must be considered. These include biological

components, sociocultural factors, and personality and

family characteristics.








Factors ContributinQ to the Etiology
and Maintenance of Bulimia

Although the contribution of organic factors to

the onset and maintenance of bulimia is unclear,

several lines of evidence suggest that bulimia may be

closely related to biologically-mediated affective

disorders (Johnson & Maddi, 1986). First, many bulimic

patients report symptoms characteristic of affective

illness, including fluctuating mood states, low

frustration tolerance, anxiety, and suicidal ideation

(Glassman & Walsh, 1983; Hudson, Pope, Jonas, &

Yergelun-Todd, 1983; Johnson & Larson, 1982; Pyle et

al., 1981). Second, several studies indicate a high

incidence of major affective disorder among first- and

second-degree relatives of bulimic patients (Hudson,

Laffer, & Pope, 1982; Hudson et al., 1983). In

addition, biological factors involved in bulimia are

suggested by the fact that two biological markers for

depression (the dexamethasone suppression test and the

thyroid-releasing hormone stimulating test) have

yielded positive results in bulimic patients with the

same frequency as in patients with major depression

(Gwirtsman, Roy-Byrne, & Yager, 1983; Hudson et al.,

1982). Finally, several double-blind placebo-

controlled studies of antidepressant pharmacotherapy

have indicated that this treatment may be effective in

reducing bulimic behaviors, further supporting

biological hypotheses (Brotman, Herzog, & Woods, 1984;








Pope & Hudson, 1982; Pope et al., 1983; Sabine, Yonace,

Farrington, Barratt, & Wakeling, 1983; Walsh, Stewart,

Wright, Harrison, Roose, & Glassman, 1982). However,

although physiological mechanisms appear to play an

important role in the pathogenesis of bulimia, research

results have not been consistent, suggesting that this

disorder results from a number of environmental and

personality variables.

Many researchers have implicated sociocultural

factors in the etiology and maintenance of bulimia.

During the past several decades there has been an

increasing emphasis on the importance and social

desirability of attractiveness in general and thinness

in particular. In recent years, social standards for

women have moved towards an increasingly thin ideal,

with the mass media placing much more emphasis on what

an acceptable body should look like, and how to attain

it through dieting and fitness (Garner, Garfinkel,

Schwartz, & Thompson, 1980; Striegel-Moore,

Silberstein, & Rodin, 1986). Bulimic women seem

especially susceptible to this cultural ideal, and have

difficulty distancing their self-expectation from

society's ideal, often with unhealthy consequences

(Steiner-Adair, 1986; Striegel-Moore et al., 1986).

For example, results of one study of adolescent girls

(Steiner-Adair, 1986) indicated that while all girls

had a similar ideal of "superwoman," only those who








were eating-disordered saw this ideal as consistent

with their own goals. Females without eating disorders

reported more modest goals.

Beauty and thinness are often linked with

femininity, as is dieting behavior (Gillen, 1981;

Striegel-Moore et al., 1986). Thinness may also be

associated with success or personal achievement, and

for some women, being thin may serve to further their

success in the professional world and give them a

competitive edge (Striegel-Moore et al., 1986). The

pursuit of thinness may be one way for a young woman to

compete, prove her success and personal accomplishment,

and demonstrate self-control (Johnson & Maddi, 1986;

Striegel-Moore et al., 1986). This is intensified by

the fact that young women today are raised in a world

of shifting cultural norms and as such are faced with

many ambiguous and sometimes conflicting role

expectations (Garner, Garfinkel, & Olmsted, 1983).

Research indicates that bulimics have difficulty

establishing a good self-concept, identifying and

asserting their needs, and developing personal autonomy

and independence (Baird & Sights, 1986; Dunn &

Ondercin, 1981). They tend to feel undifferentiated

and have low self-esteem, and as such may be especially

unable to cope with the complex sex role expectations

of our culture (Grissett & Norvell, 1987; Lewis &

Johnson, 1985; Timko, Striegel-Moore, Silberstein, &








Rodin, 1987). A recent study of female undergraduates

(Timko et al., 1987) indicated that women who deemed

socially desirable masculine traits as important for

themselves, and who felt that many roles were central

to their sense of self, reported significantly more

eating disorder symptoms.

Despite the importance of cultural variables in

the etiology and maintenance of bulimia, many young

women today do not develop eating disorders, and as

such it appears that individual variables must also be

taken into consideration. Many studies have

investigated psychopathological and personality

variables of bulimics which might make them more prone

to bulimic behavior. Using standardized assessment

instruments, the psychological profiles of bulimics

have been compared to normal controls and other patient

populations such as obese individuals and substance

abusers. In general, results are fairly consistent, in

that bulimics frequently obtain elevated scores on a

number of scales measuring psychiatric disturbance

(Hatsukami, Owen, Pyle, & Mitchell, 1982; Johnson et

al., 1982; Pyle et al., 1981; Williamson, Kelly, Davis,

Ruggiero, & Blouin, 1985). Generally, they report

feeling more tense, anxious, depressed, compulsive,

alienated, and more impaired on measures of life

adjustment (Dunn & Ondercin, 1981; Johnson & Larson,

1982; Pyle et al., 1981; Williamson, et al., 1985).








Results of studies investigating the personality

characteristics of bulimics have been quite variable,

but two factors seem to emerge consistently. First,

bulimics experience considerable affective instability,

as evidenced by depression, fluctuating moods, anxiety,

impulsive behavior, and a general feeling of being out

of control (Dunn & Ondercin, 1981; Johnson & Larson,

1982; Johnson & Maddi, 1986). It is not clear whether

the affective instability precedes or follows the onset

of bulimic symptoms, but it appears that these

difficulties are long-standing, and result from both

biogenetic vulnerabilities and maladaptive parenting

styles (Johnson & Maddi, 1986).

A second prominent personality trait among

bulimics is low self-esteem (Baird & Sights, 1986;

Boskind-Lodahl, 1976). In bulimics, this includes

several distinctive features. First, they seem to have

difficulty identifying and expressing internal states,

which leads them to feel undifferentiated, ineffective,

and helpless to control these internal states (Bruch,

1973; Lewis & Johnson, 1984). In addition, bulimics

are very sensitive to rejection, non-assertive, and

feel uncomfortable socially (Boskind-Lodahl, 1976;

Johnson et al., 1982; Pyle et al., 1981; Schneider &

Agras, 1985). Finally, bulimics have very high

expectations of themselves, expriencing shame and guilt

because of the discrepancy they feel between their








actual and ideal selves (Goodsitt, 1984; Kohut, 1971)

which is exacerbated by their bulimic behavior.

In the search for possible origins of some of the

bulimics' psychopathology and related difficulties, a

few studies have investigated family characteristics

among bulimic patients, most using self-report measures

of family interaction style. In general, findings have

been fairly consistent. Compared with normal control

families, the families of normal weight bulimics use

more indirect patterns of communication, place less

emphasis on assertiveness and autonomy, and express

higher achievement expectations, although they are at

the same time less interested in political, social,

cultural, and recreational events (Johnson & Flach,

1985; Ordman & Kirschenbaum, 1984). In addition, they

express more aggression, anger, and conflict, and give

each other less support and commitment (Johnson &

Flach, 1985; Ordman & Kirschenbaum, 1984). Compared

with the families of restricting anorexics, bulimic

families report greater overall psychopathology, as

reflected by a higher degree of problems in many areas

of family interaction, including communication,

affective expression and involvement, control, and

social desirability (Garner, Garfinkel, & Olmsted,

1983). Several investigators using direct

observational measures of bulimics' family interaction

style have reported that compared with normal control








families, families of bulimic-anorexics were less

helpful, trusting, nurturing, and approaching, and gave

more belittling, negative, and contradictory messages

(Humphrey, Apple, & Kirschenbaum, 1985).

Thus it appears that a number of biological,

sociocultural, personality, and family factors may

contribute to the etiology and maintenance of bulimia.

Bulimics seem to have a long history of difficulty

identifying and modulating their internal affective

states, which contributes to feelings of helplessness,

ineffectiveness, and lack of confidence interpersonally

(Johnson & Maddi, 1986). In addition, bulimics'

families demonstrate significant psychopathology, are

disengaged and chaotic, and experience a high degree of

conflict and life stress. Research findings suggest

that compared to normal families, bulimic families

communicate in indirect and contradictory ways, have

less problem-solving skills, and are less supportive,

while having higher achievement expectations (Johnson &

Maddi, 1986). Growing up in this type of environment

may exacerbate the bulimic's psychopathology and

difficulty dealing with her own thoughts and feelings,

and she is likely to feel increasingly unstable,

lonely, and unable to cope with life stressors. In

addition, she probably fails to learn adequate skills

needed to interact comfortably and confidently with

others while satisfying her own needs.








Social Maladjustment

In light of the research reported above, it is not

surprising that several researchers have investigated

the life adjustment of bulimics. Most authors have

utilized the Social Adjustment Scale-Self Report (SAS-

SR) (Weissman, Prusoff, & Thompson, 1976), which

measures performance over the past two weeks in six

major areas (work, social and leisure activities,

relationship with extended family, role as a spouse,

role as a parent, and membership in the family unit).

In a preliminary investigation, Johnson and Berndt

(1983) found that compared to a community sample,

bulimics showed significantly poorer adjustment in all

areas, and their scores were most similar to those of a

group of alcoholic women. Norman and Herzog (1984)

found similar results at initial evaluation of bulimics

and at a one-year follow-up.

Likewise, in a study comparing bulimic graduate

students and their non-eating disordered colleagues,

Herzog, Norman, Rigotti, and Pepose (1986) found that

bulimics reported significantly more social

maladjustment in the student, social/leisure, and

family spheres. Frequency of binge eating and purging

was associated with degree of social impairment, with

significant social dysfunction noted on the overall

scale at a minimum of binge eating/purging frequency of

once per week. A second study (Herzog, Keller, Lavori,








& Ott, 1987) comparing bulimic women to matched

controls on the same measure of social maladjustment

found very similar results. Sixty-eight percent of the

bulimic subjects and only 13% of the controls scored

within the impaired range on one or more of the

subscales (Herzog et al., 1987). Thus it appears

bulimic women are significantly impaired across a

number of areas of social interaction.

Social Support and Bulimia

It seems then that the bulimic's difficulties in

social interactions stem in part from conflicted and

pathological family relationships, which in turn result

in social maladjustment in many areas. These findings

of social maladjustment and the sense of isolation

reported by bulimics (Silberstein et al., 1986) suggest

that the interaction between the bulimic and her social

environment is significantly impaired. One might

hypothesize that the bulimic individual's social

difficulties affect her ability to receive adequate

social supports to cope with stress. However, very few

studies have directly addressed the role of social

support in the onset and perpetuation of bulimia.

In a retrospective study of bulimics and

anorexics, Slater (1988) explored the relationship

between ideal and perceived support as reported in

eating disordered women. Results indicated that larger

discrepancies between ideal and perceived social








support from parents and a significant other were

related to increased eating disorder symptomatology in

bulimics. Both groups reported receiving less social

support than they desired from either parent. Bulimic

subjects also demonstrated strong positive correlations

between ideal social support and seven of the eight

subscales on the Eating Disorders Inventory (Garner &

Olmsted, 1984).

In a recent study designed to further investigate

the relationships among bulimic symptoms, social

support, and social anxiety and distress, 15 bulimics

and 15 matched controls were examined for differences

on measures of social support, psychopathology, and

social-evaluative anxiety (Slater, Grissett, & Norvell,

1988). Results indicated that bulimic women harbor a

pronounced fear of negative evaluation in social

situations, and exhibit significantly more

psychopathology, in that they feel more depressed,

anxious, inadequate, and alienated from others. In

addition, although the actual reported amount of social

support did not differ between groups, bulimic women

reported significantly lower satisfaction with their

social support. In fact, this dissatisfaction proved

to be the best predictor of severity of bulimic

behaviors.

These studies suggest that the lack of adequate

perceived social support and distressed interpersonal








relationships appear to be important risk factors in

the development and/or maintenance of bulimia.

Although bulimic women may have access to a similar

amount of social support as normals, they are

nonetheless dissatisfied with this. At this point, a

number of hypotheses could be proposed as we attempt to

explore the relationship between social support and the

bulimic syndrome. However, in order to formulate

meaningful hypotheses regarding the relationship of

social support to bulimia, it is first important to

understand relevant social support literature.

Recent Developments
in the Social Support Literature

The social support literature has grown

considerably in the past two decades, with a great deal

of emphasis on the relationship between social support

and physical and emotional health (Cohen, 1988; Cohen &

Hoberman, 1983; Kessler & McLeod, 1985). Lack of

social support has been implicated in the etiology of

physical illness and prospectively associated with

higher mortality rates in both healthy and unhealthy

individuals (Berkman, 1985; Cohen & Wills, 1985;

Kessler & McLeod, 1985; Wallston, Alagna, DeVellis, &

DeVellis, 1983). Perceived availability of support has

also been shown to protect individuals from the

psychological impacts of exposure to stressful life

events and chronic life strains (Cohen & Hoberman,








1983; Cohen & Wills, 1985; Kessler & McLeod; 1985;

Wilcox, 1981).

A number of studies have directly linked the

social environment to disease and mortality, but these

provide little information about the processes by which

this occurs (Cohen, 1988). Cohen (1988) reviews

several psychosocial process models which rely on

hypothesized links between social support and

psychosocial and biological processes. Main-effect

models have focused primarily on links between social

integration (a structural index of social ties) and

health, while stress-buffering models have focused on

the perceived availability of support. Although a

relatively small amount of literature has been

concerned with the direct effects of social support on

illness onset, the majority of research has

investigated the hypothesis that social support

protects individuals from the negative consequences of

stressors (Wallston et al., 1983). Although the

relationship between social support and life events is

quite complex, perceived support has been found to

result in stress-buffering effects (Cohen, 1988).

Further complicating the literature are the many

conceptual, methodological, and theoretical problems

involved in the study of social support. Thoits (1982)

suggests that the concept of social support has often

been poorly conceptualized and operationalized, perhaps








leading to confounds between life events and social

support measures. This may have caused researchers to

underestimate the value of the main effects of social

support. Similarly, Abbey, Abramis, and Caplan (1985)

emphasize the importance of considering the effects of

both social support and social conflict.

The vague nature of the social support concept has

also been heightened when different researchers have

used similar terms to refer to a disparate set of

processes, or when others have used different terms

that refer to basically the same dimensions (Jung,

1984). Despite the fact that several researchers have

proposed taxonomies of the components of social support

(House, 1981), studies often fail to assess the

influence of these different components separately and

use broad definitions that combine several elements

(Jung, 1984).

The difficulty which researchers have encountered

in conceptualizing and operationalizing social support

is evident in the literature by the variability of

indicators that have been used to measure this

construct (Barrera, 1986). In defining social support,

it is important to realize that the amount of social

support is not necessarily equal to number of social

contacts or the size of one's network. Many

quantitative and structural aspects of social support

have been investigated, including number of social








relationships, composition of the social network,

patterns of interconnectedness among network members,

and accessibility of network members (Hall & Wellman,

1985; Henderson, Duncan-Jones, McAuley, & Ritchie,

1978; Silberfeld, 1978; Tolsdorf, 1976). However, it

seems that other qualitative aspects may be equally, or

perhaps more, important in affecting the facilitation

and interpretation of supportive behaviors and

contributing to the perception or psychological sense

of support (Cutrona, 1986; Gottlieb, 1984). Some of

the qualitative aspects that have proven to be

important include such factors as the influence of

expectations (Gottlieb, 1984), positive beliefs in the

benefits of help-seeking (Eckenrode, 1983),

environmental factors (Cutrona, 1986), sociodemographic

variables (Riley & Eckenrode, 1986), and personality

factors such as self-esteem, hardiness, locus of

control, coping skills, affiliation and autonomy needs,

and pre-existing levels of social support (Cohen,

Mermelstein, Kamarck, & Hoberman, 1985; Cohen & Syme,

1985; Dunkel-Schetter, Folkman, & Lazarus, 1987;

Eckenrode, 1983; Kobasa & Pucetti, 1983; Lefcourt,

Martin, & Saleh, 1984; I.G. Sarason, Levine, Basham, &

B.R. Sarason, 1983; I.G. Sarason, B.R. Sarason, &

Shearin, 1986).

Researchers also differ on whether they assess the

support an individual actually receives, or their








perception of the support available to them,

emphasizing the individual's subjective cognitive

appraisal of their connections to others rather than

simply the number of supporters or amount of social

contact (Barrera, 1986; Cohen et al., 1985; I.G.

Sarason et al., 1983). This distinction is important

because perceived and received social support

instruments often demonstrate different associations

with other measures such as indices of negative life

events or mortality risk (B.R. Sarason, Shearin,

Pierce, & I.G. Sarason, 1987). It appears that

perceived available support may be an important

qualitative aspect to consider when researching this

area, as it is often a more significant predictor of

symptomatology than merely quantitative measures (B.R.

Sarason et al., 1987). The perceived availability and

adequacy of social support is thus an important element

to be assessed, as it has been consistently linked to

positive mental and physical health outcomes (Cutrona,

1986), and to more positive personal adjustment

(Pierce, I.G. Sarason, & B.R. Sarason, 1988).

Another important consideration which has

developed recently in the social support literature

concerns the quality of the relationships which provide

support. Up until the early 1980's, researchers had

almost exclusively studied the social network in terms

of its positive influences. However, recently there








has been an explicit recognition that an individual's

social network often consists of conflicted

relationships which may be a source of both positive

and negative interactions (Barrera, 1981; Eckenrode &

Gore, 1981). This is congruent with social exchange

theorists who have long emphasized the fact that social

relations entail both costs and rewards (Thibaut &

Kelley, 1959).

In a study examining the relative impact of

positive and negative social interactions on older

women's well-being, Rook (1984) found that negative

social interactions were more consistently and more

strongly related to well-being than positive social

interactions. Fiore, Becker, and Coppel (1983)

proposed that when individuals rate their satisfaction

with their social support, they are actually responding

with summary assessments made up of both positive and

negative perceptions of the network. These researchers

suggested, as did Gore (1978), that individuals

reporting low satisfaction are experiencing more unmet

support expectations and are therefore more stressed

and more symptomatic. Likewise, Brenner and Norvell

(in press) found that the presence of at least one

source of consistent problems in the individual's

network was more predictive of life satisfaction than

the presence of consistent positive supports.








Pagel, Erdly, and Becker (1987) confirmed these

findings in a longitudinal study investigating both the

helpful (positive) and the upsetting (negative) aspects

of social networks of spouses caring for a husband or

wife with Alzheimer's disease. Results showed that the

care givers' degree of upset with their networks was

strongly associated with lower network satisfaction and

increased depression over time. Helpful aspects of the

network interacted with network upset in predicting

satisfaction and depression (Pagel et al., 1987). In

addition, Pierce et al. (1988) found that the quality

of relationships (perceived positivity and importance

of personal relationships, as well as conflict) was

correlated with the perceived availability and adequacy

of social support. In fact, the quality of

relationships made a significant contribution to

personal adjustment which was independent of that made

by perceived social support (Pierce et al., 1988).

Some investigators (Henderson et al., 1978) have

considered the possibility that since perceptions of

support adequacy are subjectively determined, they may

merely reflect the individual's level of adjustment or

depression. Vinokur, Schul, and Caplan (1987) found

that perception of support was moderately determined by

the recipients' negative outlook bias and only weakly

determined by poor mental health (anxiety and

depression). However, findings of several other








studies (Fiore et al., 1983; Pagel et al., 1987) are

conflicting and have failed to support the hypothesis

that the perception of support adequacy simply reflects

the individual's psychological adjustment. These

researchers found that perceptions of support were not

merely a function of level of depression (Fiore et al.,

1983). In fact, after controlling for initial

depression and initial level of upset with one's social

network, changes in perception over time predicted

changes in depression. That is, level of depression

increased as the degree of upset and dissatisfaction

with the social network increased (Pagel et al., 1987).

Social Support and the Stress Process

As previously stated, perceived availability of

social support appears to moderate the impact of life

events on mental and physical health (Cohen, 1988), and

may have a direct effect as well (Thoits, 1982). The

stress-buffering effect of social support has received

considerable attention in the literature, and in many

cases, perceived support appears to be an important

factor in a complex and interactive stress process.

Pearlin, Menaghan, Lieberman, and Mullan (1981) propose

a process of stress including life events, chronic life

strains, self-concepts, coping, and social supports.

They hypothesize that life events adversely affect

enduring role strains, which in turn erode positive

self-concepts such as self-esteem and mastery. The








individual is then left especially vulnerable to

experiencing symptoms of stress, often including

depression. Thus, according to this model, coping and

social supports have an indirect effect in that they

minimize the elevation of depression by preventing the

deterioration of self-concepts. In other words,

psychological variables such as personal control and

self-esteem mediate the stress-buffering effects of

social support (Pearlin et al., 1981).

Lazarus and Launier (1978) propose a transactional

model describing stress as the discrepancy between the

demands on a person and that person's appraisal and

evaluations of his or her potential responses to these

demands. Elliott and Eisdorfer (1982) conceptualize

the stress process as a series of interactions between

the individual and the environment, including four

components (potential stressors, reactions to a

particular stressor, consequences of the reactions, and

mediators at each stage of the process). Other

researchers (Shinn, Lehmann, & Wong, 1984) have also

suggested that typical research models of social

support are overly simple and should consider the

influences of stressors, psychological distress,

personal characteristics of recipients, and

environmental constraints on support, as well as the

negative consequences of social interactions.








The Relationship of Stress
and Social Support in Bulimia

In the past several years, researchers have begun

to investigate how stress is related to the etiology

and maintenance of bulimia. Shatford and Evans (1986),

using linear structural relations analysis (LISREL),

developed a causal model of bulimia based on a stress

process comprised of the sources, mediators, and

manifestations of stress. The sources of stress they

considered included environmental stressors (life

events and daily hassles) and psychological status

(depression, low self-esteem, external locus of

control, and general mental health). Mediators of

stress included methods of coping (active-cognitive,

active-behavioral, and avoidance) and focuses of coping

(problem-focused and emotion-focused). Based on

previous research reporting behavioral expressions of

stress such as increased alcohol use, eating, and

smoking (Pearlin & Schooler, 1978; Billings & Moos,

1981), Shatford and Evans (1986) considered bulimia to

be a manifestation of stress in the vulnerable bulimic

individual. Their model thus attempted to describe the

relationships between environmental stressors,

depression, psychological status, and stress mediators.

Results indicated that coping skills are an important

mediator of stress, and that having a high frequency of

environmental stressors and/or the presence of

depression or risk for depression, may lead an








individual to use ineffective coping mechanisms, which

may in turn result in bulimic behavior (Shatford &

Evans, 1986).

Cattanach and Rodin (1988) recently suggested the

importance of assessing the role of psychosocial stress

in bulimia by viewing it as a process which includes

stimulus and response, as well as appraisal, coping

processes, control, social supports, personality

factors, and other intervening variables predisposing

an individual to experience more stressors or to be

more reactive to potential stressors. These authors

present several internal and external mediators which

affect the nature of an individual's reactions to

stressors. Internal mediators include such things as

coping abilities, expectations, and prior experience.

A person's perception of the environment and his or her

appraisal of probable response outcomes and available

response options are important because they influence

the selection of a coping response (Lazarus, 1966).

Coping styles include problem-focused coping, which is

intended to manage the situation, and emotion-focused,

which is aimed at the resulting emotions. Recent

research indicates that active coping styles reduce the

effects of potential stressors, and result in better

adjustment and less depression, while passive coping

styles are less effective and are associated with

increased depression and physical illness (Billings &








Moos, 1984; Coyne, Aldwin, & Lazarus, 1981; Pearlin &

Schooler, 1978). In addition, perceived lack of

control over events has been identified by several

investigators as an important variable associated with

increased illness and psychological distress (Cattanach

& Rodin, 1988).

Cattanach and Rodin (1988) also briefly mention

several variables which have been investigated as

external mediators between the individual and the

environment. One of these mediators is social support,

which may function in a variety of ways, including

providing protection from the full impact of potential

stressors and facilitating coping and adaptation

(Cattanach & Rodin, 1988). Personality characteristics

may also mediate the relationship between the

individual and the environment, and persons with

certain styles of perceiving their environment may be

especially vulnerable to certain kinds of stress and

may respond to stressors differently then others

(Cattanach & Rodin, 1988).

Several of these elements of the stress process

have been investigated in bulimic patients, and might

play an important role in the etiology and maintenance

of bulimia. Studies investigating the number and types

of potential stressors experienced by bulimics are

often confounded by their retrospective self-report

nature, and in general indicate that the events and








conditions reported are not highly unusual. As such,

it seems likely that individual intervening variables

are more important in determining the relationship

between potential stressors and bulimic symptoms

(Cattanach and Rodin, 1988).

One variable which may mediate between stressors

and bulimic behavior is an inaccurate perception and

appraisal of the environment. Several studies have

indicated that bulimics may have difficulty appraising

situations accurately, perhaps causing the perceived

effect of these events to be exacerbated (Cattanach &

Rodin, 1988; Heilbrun & Bloomfield, 1986). In

addition, bulimics may perceive themselves as less able

to cope with stressors (Lehman & Rodin, 1986). Some

researchers have suggested that bulimics may lack a

full repertoire of coping responses from which to

select (Hawkins & Clement, 1980), while others report

that adequate coping strategies may be available, but

bulimics are unable to use them skillfully and

effectively to cope with difficult situations (Katzman

& Wolchik, 1985). Katzman and Wolchik (1985) found

that bulimics, as well as binge eaters and depressed

subjects, generally used passive coping styles (e.g.,

avoid actively confronting problems, manage resulting

emotions rather than situations) and were unable to

express their feelings, a combination which has been

associated with poorer adjustment (Billings & Moos,








1981). Shatford and Evans (1986) found that bulimic

women tend to use avoidance and emotion-focused coping

responses which are less effective than the problem-

focused coping responses used by nonbulimic women.

Their model also suggested that environmental stressors

and/or depression might lead an individual to use

ineffective coping mechanisms (Shatford & Evans, 1986)

and that the bulimic's lack of perceived control may

lead to binging and purging (Cattanach & Rodin, 1988).

Individual difference variables such as

personality characteristics or mental health may also

be important mediators in the stress process for

bulimics. For example, many bulimics evidence

significant levels of depression (Hudson, Laffer, &

Pope, 1982; Katzman & Wolchik, 1984; Williamson et al.,

1985) which could affect their situational appraisals

and may interfere with effective coping responses

(Katzman & Wolchik, 1984; Lehman & Rodin, 1986).

Perhaps the bulimic's dysphoric mood impairs her

appraisal, causing her to perceive more stress than

others in a similar environment, and thus leading to

bulimic behavior (Cattanach & Rodin, 1988). Variables

such as depression and self-esteem have previously been

implicated as important factors in the stress process

(Pearlin et al., 1981). In addition to being a

precursor of bulimic behavior, these variables may be

increased by the bulimic symptoms, or may be stressors








in themselves (Cattanach, & Rodin, 1988; Shatford &

Evans, 1986).

One factor in the stress process which has

received limited attention in the bulimic literature is

that of social support. As reported earlier, a number

of studies indicate that bulimic women are often

socially maladjusted and feel isolated from others

(Johnson & Berndt, 1983; Norman & Herzog, 1984; Herzog

et al., 1987), suggesting that they lack adequate

social supports to cope with stress. In addition,

bulimics express significant dissatisfaction with their

perceived social support, which is strongly related to

the severity of bulimic symptoms (Slater et al., 1988).

However, little is known about specific aspects of the

bulimic's social support system which may be helpful or

problematic, or which might contribute to her

dissatisfaction and maladjustment.

It seems apparent from such research that

disturbed interpersonal relationships and the lack of

perceived social support are important components of

the bulimic syndrome. At this point a number of

hypotheses might be considered. For example, perhaps

the bulimic lacks the skills or competence necessary to

take advantage of available support. Previous research

has demonstrated the bulimic's considerable fear of

negative evaluation in social situations, interpersonal

sensitivity, low self-esteem, and affective instability








(Johnson & Maddi, 1986; Slater et al., 1988). In

addition, bulimics appear to be socially maladjusted

(Herzog et al., 1987; Johnson & Berndt, 1983),

suggesting that these personality and

psychopathological characteristics may interfere with

her ability to use the available social support network

in a helpful and adaptive manner. On the other hand,

perhaps certain aspects of the bulimic's social support

network (e.g., quality of relationships) are

problematic and result in the failure of this network

to provide adequate positive support. This hypothesis

is based on previous research which has demonstrated

conflicted and chaotic relationships in bulimic

families (Garner et al., 1983; Humphrey et al., 1984;

Johnson & Flach, 1985) and the importance of

considering negative interactions when assessing social

support (Abbey et al., 1985; Rook, 1984). It may also

be possible that the bulimic's symptoms and

psychopathology play an important role in her social

difficulties, and that this is simply reflected by her

social maladjustment and dissatisfaction with social

support. This hypothesis stems from previous research

suggesting that perhaps perceptions of support adequacy

are determined by the individual's level of adjustment

or depression (Henderson, et al., 1978; Vinokur, et

al., 1987).








Research Aims and Hypotheses

Although many of the factors discussed above have

been suggested as important components of the stress

process, research investigating these factors with

bulimic subjects has been minimal. Inadequate social

support appears to be part of the bulimic's environment

and may be important in perpetuating the bulimic cycle,

but virtually no information is available about

specific aspects of the bulimic's social support

network or about other individual difference variables

which may affect the degree to which adequate social

support is received. The present study explored

specific aspects of the bulimic's social support

network, the quality of her interactions within this

network, and individual difference variables which

might mediate her ability to obtain support or to

perceive this as adequate.

In light of previous research concerning the

importance of considering different sources of social

support, the first aim was to investigate the bulimic's

perceived social support from both family and friends.

Several researchers have indicated that these sources

of support are related but separate, valid, and useful

constructs (Procidano & Heller, 1983; Sarason et al.,

1987). Procidano and Heller (1983) suggest that the

distinction between perceived support from family and

friends is important, in that these two sources appear








to be differentially related to symptoms of distress

and psychopathology, as well as to personality

characteristics such as social competence, anxiety, and

mood state (Procidano & Heller, 1983).

The second aim of the present study was to explore

both positive and negative aspects of social

interaction within the bulimic's social support

network. Social conflict seems to be strongly related

to psychopathology and network dissatisfaction (Fiore

et al., 1983; Pagel et al., 1987; Pierce et al., 1988),

while experiencing positive and important relationships

is associated with the perception of higher levels of

support (Pierce et al., 1988). It was hypothesized

that increased level of negative interactions and poor

quality of relationships would be highly related to

bulimic symptoms and thus important to consider when

exploring the relationship between the social network

and bulimia.

In light of previous research emphasizing the

importance of individual difference variables on the

perception of social support, this study also aimed to

explore the relationship of social competence to the

perception of support and to reported conflicting and

supportive interactions. Several studies have

demonstrated a relationship between social skills and

social support (Cohen et al., 1986; I.G. Sarason et

al., 1985). Sarason et al. (1985) found that when








compared to individuals with low levels of reported

social support, those high in social support were

significantly greater in self-described and

experimenter-rated social skills. Similarly, in a

study by Cohen, Sherrod, and Clark (1986), social

skills were found to be prospectively predictive of the

development of social support and friendship formation.

Therefore it is important to investigate the

relationship between social competence and the

perception of the adequacy and nature of social

support.

With these aims in mind, the study investigated

the following:

1. It was hypothesized that bulimics would report

significantly less perceived social support, as

compared to normal controls.

2. It was hypothesized that bulimics would report

significantly more negative social interactions

(conflict) and poorer quality of relationships than

normal controls.

3. It was hypothesized that bulimics would

demonstrate significantly poorer social skills than

controls, as assessed by both self-reported competence

and observer ratings of social effectiveness.














METHOD

Subjects

Subjects were 42 female undergraduates at the

University of Florida who participated in the study as

part of a class requirement or for a payment of $5.00.

Approximately 800 undergraduates were screened for

bulimic symptoms using the Bulimia Test (BULIT) (Smith

& Thelen, 1984). A research cut-off score of 88 was

used to identify women endorsing behaviors and

psychological characteristics similar to those of

clinically diagnosed bulimics. These women then

participated in a structured clinical interview, and

twenty-one who were diagnosed as bulimic by DSM-III-R

criteria made up the experimental group. Those who did

not meet DSM-III-R criteria for bulimia nervosa were

excluded from the study. The 21 women in the

experimental group were then matched on the variables

of height and weight with 21 women who served as the

control group. Matched pairs were within two inches

and ten pounds of each other. Mean height and weight

for bulimics were 65 inches and 133 pounds, while

controls averaged 66 inches and 132 pounds.








Measures

The Bulimia Test (BULIT)

The Bulimia Test (Smith & Thelen, 1984) was used

to screen subjects for the normal and bulimic samples.

This 36-item, multiple-choice self-report scale was

specifically designed to assess bulimic symptoms.

Construction of the BULIT was based on DSM-III criteria

and was initially conducted by comparing responses of

clinically identified female bulimics with non-eating

disordered female college students on preliminary test

items. The scale proved to be a good predictor of

bulimia in both the initial and replication samples.

Cross validation was then performed using samples of

bulimic and normal control subjects, and the measure

was subsequently administered and validated with

nonclinical populations of undergraduate college women.

Results indicated that the BULIT is a reliable and

valid predictor of bulimia in nonclinical populations

as well. Evidence for construct validity has been

demonstrated by the BULIT's high correlation (r=.93,

p<.0001) with the Binge Scale (Hawkins & Clement,

1980), another measure of binging behavior.

Discriminative validity has been demonstrated by

significant differences between bulimics (n=20) and

normal control subjects (n=94) (M=124.0 and M=60.3,

respectively), t(i12)=15.25, p<.0001. In addition, in

the cross validation studies, the BULIT demonstrated








high predictive ability, with sensitivity, specificity,

and positive and negative predictive values all above

.90. Predictive ability in the nonclinical sample used

in further validation studies was lower, presumably

because these subjects' scores were less extreme and

thus more difficult to classify. In these studies,

sensitivity and specificity were .64 and .89,

respectively, while the positive and negative

predictive values were .74 and .84, respectively.

A cut-off score of 88 was used to screen for those

women endorsing behavioral and psychological

characteristics similar to bulimics. Using this

criterion with a nonclinical population, Smith and

Thelen report a false negative rate of 0.0 (Smith &

Thelen, 1984). Similarly, Slater et al. (1988) found

that 15 of 18 women screened in this manner were

subsequently diagnosed as clinically bulimic. This

cut-off provides an efficient way of identifying women

who display bulimic behaviors which can then be

verified in a subsequent structured clinical interview

designed to determine whether they meet DSM III-R

criteria for bulimia.

The Perceived Support Scale (PSS)

The Perceived Support Scale (PSS) (Procidano &

Heller, 1983) is comprised of two 20-item subscales

with a dichotomous response (yes-no) format, designed

to measure the extent to which an individual perceives








that his or her needs for support, information, and

feedback are fulfilled by friends (PSS-Fr) and by

family (PSS-Fa). In a validation study with 222

undergraduates, the PSS measures proved to be

internally consistent (Cronbach's alpha = .88 and .90,

respectively), and appeared to measure valid constructs

that were separate from each other and from network

measures (Procidano & Heller, 1983). Separate factor

analyses with orthogonal factor rotation domonstrated

each scale to be composed of a single factor (B.R.

Sarason et al., 1987). Test-retest reliability over a

1-month period was estimated to be .83 (Procidano &

Heller, 1983).

Evidence for construct validity was found, in that

these two measures were shown to be better predictors

of psychiatric symptomatology, as measured by the short

form of the MMPI (Faschinghauer, 1974), than life

events or structural characteristics of support

networks. Subsequent studies supported the

independence of these constructs by demonstrating their

differing relationships with various measures of mood

state, anxiety, psychopathology, and verbal inhibition

(Procidano & Heller, 1983). Other findings also

suggest it is important to distinguish between friends

and family in the provision of social support (B.R.

Sarason et al., 1987).








The Quality of Relationships Inventory (ORI)

The Quality of Relationships Inventory (QRI;

Pierce et al., 1988; Pierce, B.R. Sarason, & I.G.

Sarason, 1989) is a recently developed scale designed

to assess the quality of the relationships which

provide social support. The revised version of this

inventory consists of three scales measuring the

perceived positivity and importance of primary

relationships (Depth), the extent to which the

relationship is a source of conflict and ambivalence

(Conflict), and the perceived availability of social

support from specific relationships (Support). Factor

analysis indicated that these three aspects of

relationships are independent. Results of a validity

study with 360 undergraduates indicated that the QRI

scales significantly contribute to personal adjustment

independently of the contribution made by social

support. The QRI was consistently related to perceived

social support and adjustment measures, and

discriminated between several categories of

relationships (Pierce et al., 1989). In the present

study, subjects completed the QRI for their mother,

father, closest same-sex friend, and an individual with

whom they have a romantic relationship (or in absence

of this, their closest male friend).








Social Interactions Scale (SIS)

A measure similar to that used by Abbey et al.

(1985) was utilized to assess the qualitative

perception of negative and positive interactions.

Abbey et al. (1985) found that their measure of social

support was related to quality of life, negative

affect, and psychological well-being, while social

conflict demonstrated a strong relationship with

anxiety and depression. In addition, the existence of

social conflict appeared to be different and more

distressing than the absence of social support.

Questions for the SIS were developed to measure

social support (7 items) and social conflict (10

items), and were in the following format: "In the past

seven days, how much have people in your personal

life..." (e.g., acted in ways that show you they

appreciate what you do, treated you with respect,

argued with you about something, gotten on your

nerves). In addition, for each conflict question,

respondents rated two aspects on a 7-point Likert-type

scale: 1) how much these occurrences bothered them,

and 2) how they would explain why these interactions

occurred. These ratings were designed to explore the

possible influences of perceived impact and personal

attributions on the effects of negative social

interactions.








Social Competence and Effectiveness

Self-report. The Social Competence Questionnaire

(Com-Q) (I.G. Sarason et al., 1985) is a 10-item self-

report scale designed to tap responses reflecting the

degree of discomfort in various social situations.

Com-Q items were rated by the subject on a 4-point

scale ranging from "not at all like me" to "a great

deal like me." Example items are "have trouble getting

to know someone" and "feel confident of my social

behavior." This measure has demonstrated desirable

psychometric properties (I.G. Sarason et al., 1985).

Videotaped interactions. Observer ratings of

social skills were obtained on videotaped interactions

in a procedure similar to that used in a study by

Sarason, Sarason, and Shearin (1986). In the

interaction, each subject participated in a 5-minute

role-play with a female confederate who was unaware of

the subject's group membership. During this

interaction, the subject and confederate discussed how

they might improve their living situation with regard

to a troublesome third female roommate. The

confederate was trained to interact in a standardized

manner with each subject. (See Appendix C for a

summary of the training instructions).

Rating of videotaped interactions. Each tape was

rated by 3 male and 3 female psychology research

assistants from the same undergraduate subject pool








from which the experimental and control subjects were

drawn. Observers rated subjects using the Dyadic

Effectiveness Scale (I.G. Sarason et al., 1986) (see

Appendix B), which consists of 10 qualities rated on a

scale from 1 to 6 ("not at all" to "very, very much").

(See Appendix C for instructions given to raters.) In

the original validation study, these ten items as a

single scale had a reliability of .95 (Cronbach's

alpha). The items contribute to three correlated

subscales: leadership, consideration, and

attractiveness, which together accounted for 87% of the

variance and had reliabilities of .96 (Cronbach's

alpha), .92 (Cronbach's alpha), and .80 (Pearson

correlation), respectively (Sarason et al., 1986).

In the present study, observer raters were unaware

of the subjects' group membership. As in the study by

Sarason et al. (1986), specific uidelines or training

were not given to the raters since the purpose was to

obtain their subjective reactions to the subjects

rather than to force agreement. Interrater

reliabilities were computed for all pairs of raters.

All correlations among raters for total DES scores were

positive and significant (ranging from .39 to .73).

Cronbach's alpha for the six raters averaged across the

10 rated questions was .89; the range was from .88 to

.96. This was considered more than adequate, given the








subjective nature of the items' content and the lack of

intensive training in the rating system.

Psychopathology

The Symptom Checklist-90 (SCL-90) (Derogatis,

Rickels, & Rock, 1976) was used to assess

psychopathology. This measure was developed to examine

psychiatric symptomatology in outpatients. Each item

of the SCL-90 is rated on a 5-point scale of distress

ranging from "not at all" (0) to "extremely" (4). The

SCL-90 yields nine subscales of primary symptom

dimensions and three overall indices of distress. This

measure has established psychometric qualities, and has

been used extensively in previous research. Several

previous studies comparing bulimic women with other

normal women have indicated significant differences

between these groups on several scales of the SCL-90

(Slater et al., 1988; Williamson et al., 1985).

Procedure
Once they were screened for group membership

according to their scores on the BULIT, subjects were

contacted to arrange a one and one-half hour lab

session. Upon arrival at the lab, the subject was told

that this was a study about social support and the

quality of her social interactions, and was asked to

sign an informed consent if she wished to participate.

Height and weight were then verified by the

experimenter.








Subjects then completed a packet of questionnaires

consisting of the Perceived Support Scale (PSS), the

Social Support Questionnaire (SSQ), the Quality of

Relationships Index (QRI), the Social Interactions

Scale (SIS), Social Competence Questionnaire (COM-Q),

and the Symptom Checklist-90 (SCL-90). Each subject

then participated in the five-minute videotaped

interaction. Finally, a diagnostic interview was

conducted to determine whether or not the subject met

DSM-III-R criteria for bulimia. Following completion

of this interview, subjects were debriefed as to the

nature of the study, given personal feedback regarding

their scores on the BULIT, and provided with possible

psychotherapy referrals if the subject desired this

information.














RESULTS

Approach to Data Analysis

In order to prepare the data for analysis, several

preliminary data analyses were conducted. Subjects had

completed the Quality of Relationships Inventory (QRI)

(including subscales of Support, Depth, and Conflict)

for four individuals (mother, father, closest same-sex

friend, and romantic relationship/closest male friend),

resulting in 12 separate QRI variables. In order to

facilitate further analyses to assess group differences

on the QRI, mother and father ratings on the three QRI

subscales were collapsed to form "family" variables,

while ratings of female friends and romantic

relationships were combined to form "other" variables.

Thus six combination variables were calculated and

utilized in further analyses of QRI data: Family

Support, Family Depth, Family Conflict, Other Support,

Other Depth, and Other Conflict. Mean values of these

six variables for the two groups are presented in Table

1.

Several other measures were also collapsed to form

combination variables for use in certain analyses.

First, the Percieved Support Scale (PSS) measures which








had been completed for both family and friends were

combined to create the variable of Total Perceived

Support. The QRI measures were also further collapsed

to form the variables of Total Conflict, Total Support,

and Total Depth. Finally, the Dyadic Effectiveness

Scale (DES) total was broken down into the three

subscales of Leadership, Consideration, and

Attractiveness, which were used in several analyses.

The initial approach to data analysis involved a

series of correlational analyses to explore the

relationships among variables. Several MANOVAs were

then utilized to test the major hypotheses of group

differences on measures of perceived support, negative

interactions, and quality of relationships, as well as

both self-reported social competence and observer

ratings of social effectiveness. A MANOVA was also

used to test for differences between bulimics and

controls on the measure of psychopathology.

A final goal of the present study was to further

understand the role of psychopathology in the bulimic's

perception of social support, social interactions, and

quality of relationships. As a result, it was

desirable to investigate differences between groups on

the various measures after removing the effects of

psychopathology. Therefore, group scores on the

Perceived Support Scale (PSS), the Social Interactions

Scale (SIS), and the Quality of Relationship Inventory








(QRI) were compared using Multivariate Analysis of

Covariance (MANCOVA) with overall psychopathology (as

assessed by the Global Severity Index of the SCL-90) as

the covariate. Similar MANCOVAs were also utilized to

test for differences between groups on the measures of

social competence and effectiveness.

Comparison of Groups

Demographic and Descriptive Information

Results of a Multivariate Analysis of Variance

(MANOVA) revealed that the mean scores on the Bulimia

Test (BULIT) for the bulimic and normal control groups

were significantly different (F(1,35) = 380.73, R <

.001). The mean BULIT score for the bulimic group was

109.8 (SD = 13.39), while the mean score for the normal

group was 43.8 (SD = 5.5). The two groups did not

differ significantly on any of the demographic

variables (age, height, weight, grade point average, or

SAT scores). Means and standard deviations of these

variables are presented in Table 2.

Correlations

As an initial step toward understanding the

relationships among the variables in the present study,

a number of Pearsons' product-moment correlations were

computed. In order to investigate the severity of

bulimia in relation to the other variables,

correlations were computed for subjects' scores on the

BULIT, PSS-Friends and PSS-Family, and the SIS








variables (see Table 3). Higher BULIT scores were

significantly negatively associated with amount of

positive interactions (r = -.38, R < .05), and

positively associated with increased negative

interactions (r = .50, R < .001) and perceived impact

of this conflict (r = .47, p < .01). In addition,

perceived social support from family and friends was

negatively correlated with BULIT scores (r = -.50, R <

.001 and r = -.45, R < .01, respectively), indicating

that women scoring higher on the BULIT reported

significantly less perceived social support from both

of these sources.

Correlations were also computed between the BULIT,

social competence, and the Dyadic Effectiveness Scale

(DES) total, as well as its three subscales (see Table

4). The relationship between BULIT scores and social

competence approached significance (r = -.29, p = .06),

as did the correlation between BULIT scores and the DES

total (r = -.30, R = .06). The relationship between

self-reported social competence and overall observer

ratings of social effectiveness also approached

significance (r = .31, R = .06).

Finally, correlations were computed between the

BULIT and the QRI variables (see Table 5). Significant

positive relationships were found between BULIT scores

and measures of conflict with family (r = .60, R <

.001) and others (r = .31, p = .05), while there was a








significant negative relationship between severity of

bulimic symptoms and family support (r = -.33, p <

.05).
Pearsons' correlations were also utilized to

explore the relationships among the PSS and the SIS

variables. As shown in Table 3, perceived social

support from both friends and family was significantly

positively related to positive interactions (r = .61, p

< .001 and r = .43, p < .01, respectively), and

negatively related to negative interactions (r = -.52,

p < .001 and r = -.32, P < .05, respectively),

indicating that individuals reporting higher levels of

social support also report more positive interactions

and less negative interactions. Additional

correlations indicated that increased social competence

was related to greater amounts of perceived social

support from friends and family (r = .44, R < .01, and

r = .37, R < .05, respectively), as well as more

positive interactions (r = .38, p < .05) and less

negative interactions (1 = -.36, p < .05).

In order to better understand the relationship

between psychopathology and the other measures,

Pearsons' correlations were also computed between the

Global Severity Index of the SCL-90 and the BULIT, PSS,

SIS, and QRI variables (see Table 3). Higher BULIT

scores were significantly positively associated with

increased psychopathology, as indicated by higher








scores on the GSI (r = .72, R < .001) (see Table 3),
as well as every subscale. Significant relationships

were also found between severity of psychopathology and

several other measures. Women reporting more severe

psychopathology reported less perceived social support

from family and friends (r = -.47, R < .01 and r -

-.40, p < .01, respectively), fewer positive

interactions (r = -.45, R < .01), more negative

interactions (r = .46, R < .01), and a stronger impact

of this conflict (r = .46, R < .01). In addition,

increased psychopathology was related to increased

family conflict (r = .60, R < .001).

Perceived Social Support and Social Interactions

A MANOVA was used to compare the bulimic and

control groups on the Perceived Support Scale for both

friends and family (PSS-FR and PSS-FA) and the Social

Interactions Scale (SIS) (including measures of

positive interactions, negative interactions, the

impact of negative interactions, and attributions for

these interactions). Results of this MANOVA

demonstrated a significant overall group effect

(E(6,32) = 3.82, R<.01) (see Table 6). Subsequent

examination of the univariate analyses indicated that

the bulimic and control groups demonstrated significant

differences on a number of dimensions. Compared to

controls, bulimic women reported significantly less

perceived social support from friends and family








(F(1,37) = 7.39, R < .01 and F(1,37) = 5.28, R < .05,

respectively). Although the two groups did not differ

on the amount of positive interactions experienced,

bulimics reported significantly more negative

interactions (f(1,37) = 15.70, R < .001) and indicated

that this conflict had a greater impact on them

(F(1,37) = 12.42, p < .01). In addition, bulimics

demonstrated a stronger tendency than controls to

attribute these negative interactions to themselves

more than to others or to the situation (f(1,37) =

3.02, R = .09), although this result did not attain

significance.

Quality of Relationships

In order to investigate differences between groups

on the QRI, the three subscales were analyzed using 3

two (groups: bulimics and controls) by two (sources:

family and other) ANOVAs. Analysis of the mean support

scales yielded a significant main effect for source

(family or other) (F(1,37) = 20.16, R < .001). The

main effect for the groups approached significance

(F(1,37) = 3.75, R = .06), and there was no group by

source interaction. In light of the mean support

scores, these results indicate that both groups

reported receiving significantly more support from

others than from their family, and that there was a

trend for controls to report more support, although the

groups did not differ significantly on this variable.








Examination of the univariate results indicates that

this trend is primarily due to differences on Family

Support, which approached significance (f(1,37) = 3.61,

R = .06).

The analyses of the Depth and Conflict scales were

conducted in the same manner. For depth scores, there

were no significant differences for source or group

main effects or for the group by source interaction.

This indicates that both groups responded similarly in

their depth ratings of family and others, and did not

differ in the amount of depth they reported for these

relationships.

Results of Conflict scales analyses revealed a

significant group by source interaction (f(1,36) =

4.55, R < .05). These results demonstrated that while

all subjects reported more conflict with family than

with others, bulimic women did so to a greater extent

(hence the interaction effect). Analysis of univariate

results indicated significant differences between

groups on the measure of Family Conflict (F(1,36) =

20.68, p < .001), indicating that the bulimic women

reported a much greater amount of family conflict.

Social Effectivess and Self-Reported Social Comvetence

Reliability analysis for the ten Dyadic

Effectiveness Scale (DES) items showed that as a single

scale, they had a reliability of .98 (Cronbach's

alpha). The three subscales of Leadership (Items 1, 3,








4, 5, and 7), Consideration (Items 2, 6, and 8), and

Attractiveness (Items 9 and 10), had reliabilities of

.98 (Cronbach's alpha), .95 (Cronbach's alpha), and .74

(Pearson's r). The subscales were highly

intercorrelated (.75 to .87), and as a result, analyses

were first performed on the DES total and then repeated

on the three subscales.

Multivariate Analysis of Variance (MANOVA) was

utilized on the Social Competence Questionnaire and the

DES total in order to examine possible differences

between groups on self-reported social competence and

observer ratings of social effectiveness during the

videotaped interaction. Results demonstrated a

significant overall group effect (f(4,32) = 2.86, p <

.05) (see Table 7). Examination of the univariate

analyses indicated that bulimic women reported

significantly less social competence than normal

controls (F(1,35) = 7.52, p < .01), and were rated by

observers as less socially effective overall (F(1,35) =

4.58, p < .05). Univariate anayses of the subscales

indicated that bulimics were rated significantly lower

than normal controls on the Consideration subscale

(F(1,35) = 5.83, p <.05), while differences between

groups on the Leadership scale approached significance

(F(1,35) = 3.7, R = .06). The means of these measures

are presented in Table 7.








PsychopatholoQy

One aim of the present study was to explore the

relationship of psychopathology to bulimic symptoms, as

well as to the measures of perceived social support,

positive and negative interactions, and the quality of

relationships. Multivariate Analysis of Variance

(MANOVA) was used to compare the bulimic and normal

control groups on the SCL-90 subscales (see Table 8).

Results demonstrated a significant difference between

groups on the Global Severity Index (GSI), a composite

score for the Symptom Checklist-90 (SCL-90) which

measures severity of psychopathology (F(1,40) = 36.71,

R < .001). Significant differences between the bulimic

and control groups were also apparent on the Positive

Symptom Distress Index (PSDI) of the SCL-90, a measure

of symptomatology intensity (f(1,40) = 33.36, p <

.001). Univariate analyses of subscale scores

indicated significant differences between groups on all

subscales (see Table 8).

Covariate Analyses

One MANCOVA was utilized to compare the bulimic

and normal control groups on the PSS (including PSS-

Friends and PSS-Family) and the SIS (including positive

and negative interactions, impact of negative

interactions, and attributions for these). Results

demonstrated that after controlling for the effects of

psychopathology, the differences between groups on








reported negative interactions was still significantly

different (f(2,36) = 3.96, R = .05). These results

indicate that the amount of negative interactions made

a significant contribution to the model over and above

that of psychopathology. No other significant

differences were revealed.

MANCOVAs were also utilized to compare differences

between groups on the QRI variables (Family Support,

Other Support, Family Depth, Other Depth, Family

Conflict, and Other Conflict). Results of these

analyses indicated no significant differences on

Support or Depth scores after controlling for

psychopathology. However, a significant difference was

demonstrated between groups in their overall report of

Conflict (F(1,35) = 4.31, R < .05). Examination of the

univariate analyses revealed that this difference was

primarily due to the measure of family conflict. After

considering the effects of degree of psychopathology,

group differences on reported family conflict still

provided a contribution which nearly attained

significance (F(1,35) = 3.85, R=.058).

A MANCOVA was also used to test for differences on

the COMQ and DES total and subscales. Results

indicated that after controlling for psychopathology,

the differences between groups on the DES total

approached significance (F(2,34) = 3.41, p = .07).

Group differences on the Consideration subscale were








still significant (F(2,34) = 4.46, R < .05), while the

Leadership subscale difference approached significance

(F(2,34) = 2.97. p = .09).

Finally, it was desirable to investigate possible

moderating effects of the variables on the impact of

psychopathology in predicting whether or not a woman was

bulimic. In order to test this hypothesis, a series of

stepwise discriminant analyses were performed, utilizing

interaction terms formed by multiplying each respective

variable (PSS-Family, PSS-Friends, Negative

Interactions, Positive Interactions, Family Support,

Other Support, Family Conflict, Other Conflict, Family

Depth, Other Depth, Social Competence, and DES Total) by

the measure of overall psychopathology (GSI). These

were all entered into equations with no more than three

variables per model (Variable, GSI, and Variable x GSI).

Results revealed the presence of two interaction terms

which significantly predicted group membership better

than either variable alone: Family Conflict x GSI

(F(1,35) = 33.4, R<.001), and Other Conflict x GSI

(F(1,39) = 39.7, R<.001). Thus, it appears that the

self-reported level of conflict exerted a moderating

effect on the level of psychopathology in predicting

group membership. Consequently, in the present sample,

for two subjects reporting equal levels of

psychopathology, the one with a higher level of conflict

would more likely be classified as bulimic.









Table 1

Univariate Analyses of Repeated Measures ANOVAs on
Combination Quality of Relationship Inventory
Variables.


Family Support

Family Depth

Family Conflict

Other Support

Other Depth

Other Conflict


Bul imics

M SD

40.8 7.6

47.8 8.2

67.3 11.9

47.8 5.3

51.7 7.3

51.6 9.6


Normals

M SD

44.8 7.6

51.8 7.4

53.3 8.7

49.2 2.9

51.8 6.1

46.4 11.3


*22<.001


F

3.61

2.10

20.68*

0.69

0.00

1.93









Table 2

Univariate Analyses of DemoQraphic Variables and
Bulimia Test (BULIT) Scores.


MANOVA Overall
Group Effect


57. 1*


Bulimia Test

Age

Height

Weight

SAT Score

Grade Point
Average


Bulimics

M SD

109.8 13.4

20.3 2.8

65.0 4.2

132.5 18.7

1063.3 114.9


3.1


0.5


Normals

M SD

43.8 5.5

20.3 2.7

65.7 2.6

131.7 20.7

1072.1 115.4


3.0


0.5


*p<- 001


F

380.73*

0.40

0.12

0.07

0.05


0.73









Table 3

Pearson Correlations of the Bulimia Test (BULIT),
Psychopathology, Perceived Social Support, and Social
Interactions.


BULIT GSI PSS-FR PSS-FA POSINT NEGINT IMPACT

GSI .72***

PSS-FR -.45** -.40**


PSS-FA -.50*** -.47** .39**


POSINT -.38*

NEGINT .50***

IMPACT .47**

ATTRIB -.26


-.45** .61*** .43**

.46** -.52*** -.32* -.59***

.46** -.42** -.34* -.52*** .88***


-.28


.06


.12


.04 -.18 -.10


*R<.05
**R<.01
***R<.001

Note:
GSI=SCL-90 Global Severity Index;
PSS=Perceived Support Scale (Friends and Family);
POSINT=Positive Interactions from Social Interactions
Scale (SIS);
NEGINT=Negative Interactions from SIS;
IMPACT=Impact of Negative Interactions;
ATTRIB=Attribution for Negative Interactions.









Table 4

Pearson Correlations of the Bulimia Test (BULIT),
Psychopathology, Self-Reported Social Competence, and
Observer Ratings of Videotaped Interactions.


BULIT GSI COMQ DES-TOT


DES-L DES-C


GSI

COMo


DES-TOT


.72*

-.29 -.51*

-.30 -.24


DES-LEAD -.28

DES-CONS -.31

DES-ATTR -.28


-.22

-.22

-.28


.31

.26

.31


.97*

.93*

.89*


.87*

.86*


.75*


*R<.001

Note:
GSI=SCL-90 Global Severity Index;
COMQ=Social Competence Questionnaire;
DES-TOT=Dyadic Effectiveness Scale (DES) Total;
DES-LEAD=Leadership Subscale of DES;
DES-CONS=Consideration Subscale of DES;
DES-ATTR=Attractiveness Subscale of DES.









Table 5

Pearson Correlations of the Bulimia Test (BULIT),
Psychopathology, and the Quality of Relationships.


BULIT GSI FAMSUP FAMDEP FAMCONF OTHSUP OTHDEP


GSI


FAMSUP -.34*

FAMDEP -.26

FAMCONF .60**

OTHSUP -.06

OTHDEP .05

OTHCONF .31


-.16

-.14


.60** -.66** -.63**


-.16


.18


-.07 -.07 -.17


.17 -.07


.11


.11 -.07


.11 .69**

.14 -.12


*p<.05
**p<.001

Note:
GSI=SCL-90 Global Severity Index;
FAMSUP=Family Support from Quality of Relationships
Inventory (QRI);
FAMDEP=Family Depth from QRI;
FAMCONF=Family Conflict from QRI;
OTHSUP=Other Support from QRI;
OTHDEP=Other Depth from QRI;
OTHCONF=Other Conflict from QRI.


-.15








Table 6

Univariate Analyses of Perceived Social Support and
Social Interaction Variables.


F
MANOVA Overall Group
Effect (PSS & SIS) 3.82**

Bulimics Normals

M SD M SD F

Perceived Social
Support Friends 13.1 5.0 17.1 3.4 7.39**
Perceived Social
Support Family 10.9 5.7 15.5 4.9 5.28*

Positive Interactions 34.4 9.6 39.1 6.8 2.00

Negative Interactions 34.6 10.5 24.2 9.4 15.70***

Impact of Negative
Interactions 39.1 12.1 27.9 11.6 12.42**

Attribution for
Negative
Interactions 39.8 8.2 44.7 9.7 3.02


*R<.05
**R<.01
***R<.001

Note:
PSS=Perceived Support Scale
SIS=Social Interactions Scale








Table 7

Univariate Analyses of Self-Reported Social Competence
and Observer Ratings of Videotaped Interactions.


F
MANOVA Overall
Group Effect 2.86*

Bulimics Normals

M SD M SD F

Social Competence
Questionnaire 25.8 6.8 30.1 4.5 7.52**
Dyadic Effectiveness
Scale (DES) Total 223.1 39.4 250.2 31.6 4.58*

DES Leadership 102.5 24.7 117.6 18.5 3.70

DES Consideration 72.9 11.2 81.0 8.6 5.83*

DES Attractiveness 47.6 5.5 51.5 6.0 3.29


*R<.05
**P<.Ol









Table 8

Univariate Analyses of SCL-90 Variables.

---------------------------------------------------
F
MANOVA Overall Group
Effect (SCL-90) 5.51**


Bulimics


Global Severity Index

Positive Symptom
Distress Index

Somaticism

Obsessive-
Compulsive
Interpersonal
Sensitivity

Depression

Anxiety

Hostility

Phobia

Paranoia

Psychoticism


M

1.26


1.97

.81


SD

.61


.51

.48


1.51 .86


1.93

1.63

1.00

1.03

.50

1.35


.84

.87

.68

.80

.46

1.05


Normals

M SD


.39


1.28

.38


.62


.55

.59

.28

.30

.11

.33


1.07 .65 .20


.24 36.71**


.19 33.36**

.31 12.07*


.36 19.24**


.40

.45

.23

.18

.14

.38


45.82**

23.83**

21. 05**

16.52**

13.51**

17.20**


.25 32.18**


*R<.01
**R<.001














DISCUSSION


The present study addressed several hypotheses

concerning the social network and interactions of

individuals with bulimia. Specifically, the study

explored various aspects of the bulimic individual's

social support network, the quality of her interactions

within this network, and individual difference

variables which might mediate her ability to obtain

support or perceive this as adequate. It was

hypothesized that compared to normal control women,

bulimic women would report less perceived social

support, more negative social interactions (conflict),

and poorer quality of relationships. In addition, it

was proposed that they would demonstrate poorer social

skills, both in terms of self-reported social

competence and observer ratings of social

effectiveness. Results support these three hypotheses,

emphasizing in particular the high level of conflict in

bulimic individual's relationships and their lack of

social effectiveness, differences which are significant

even after considering their level of psychopathology.








The Social Network: Quality and Type of Interactions

Results strongly indicate a number of significant

differences between the social networks and

interpersonal interactions of women with bulimia and

non-eating disordered women. Bulimic women perceive

much less social support in their environment,

providing additional confirmation of previous research

findings suggesting that bulimic individuals are

dissatisfied with their social support network (Slater

et al., 1989), as well as studies documenting the

bulimic's sense of isolation and social maladjustment

(Johnson & Berndt, 1983; Norman & Herzog, 1984; Herzog

et al., 1987). In order to distinguish between sources

of support, the present study utilized the Perceived

Support Scale, an instrument which allows for separate

examination of support from both family and friends.

Compared to controls, bulimic individuals reported much

less support from both of these sources.

Previous research has indicated that social

support may function as a mediator between the

individual and the environment (Cattanach & Rodin,

1988). Indeed, the perceived availability and adequacy

of support is strongly linked to positive mental and

physical health and personal adjustment (Cutrona, 1986;

Pierce et al., 1988). In fact, these perceived

qualities of relationships, rather than objective

features of the social environment, seem to be the most








important aspect of social support (Sandler & Barrera,

1984). In the present study, overall perceived support

was strongly related to severity of bulimia and proved

to be an important predictor of group membership.

Although it is difficult to determine causal direction

in the relationship between bulimia and the lack of

support, this perceived lack may make the bulimic woman

particularly vulnerable to certain types of stress and

contribute to the development and maintenance of

bulimic symptoms.

Results of the present study also revealed

significant overall differences between bulimic and

control women in the type and quality of their

interactions. First, bulimic women report a much

higher occurrence of negative interactions. In

addition, the quality of their relationships is poorer,

in that they appear to experience much more conflict

overall, particularly with their family (parents). The

level of self-reported conflict was significantly

higher among bulimic women than controls, and was

strongly related to the severity of bulimic symptoms.

This supports previous research which has suggested the

importance not only of positive aspects of the social

network, but also highlighted the need to consider the

contribution of negative interactions and conflict to

personal adjustment and psychopathology (Brenner,








Norvell, & Limacher, in press; Pagel et al., 1987;

Pierce et al., 1988; Pierce et al., 1989).

Some researchers have suggested that since

perceptions of support and other aspects of social

interactions are subjectively determined, they may

simply be a function of the individual's level of

psychological adjustment (Henderson et al., 1978).

Although research findings in this area are

conflicting, it seems warranted to consider level of

adjustment when comparing self-report data of this

nature. In addition, in the present study the bulimic

and control groups differed significantly on overall

levels of psychopathology, which suggests that this

variable deserves special consideration when comparing

these two groups on other psychological variables.

Analyses conducted which controlled for the degree of

psychopathology indicated that the experimental and

control groups did not differ in their perception of

support. However, when controlling for

psychopathology, the two groups did differ

significantly on several other aspects of their social

interactions, with bulimic women still reporting higher

levels of negative interactions, total conflict, and

family conflict.

Social Competence and Effectiveness

Results of the present study indicate that bulimic

women feel much less socially competent than normal








control women. They report discomfort and incompetence

in a variety of social situations, including less

confidence in their ability to function well socially

and form close relationships with others, as well as a

decreased likelihood of engaging in behaviors such as

seeking out social encounters. Bulimic women therefore

report both less social support and less social

competence, consistent with previous evidence

establishing the relationship between perceived social

support and self-reported social competence,

assertiveness, and dating skills (Procidano & Heller,

1983; B.R. Sarason et al., 1985).

The results of the present study also suggest that

observers respond differently to eating disordered

women than to non-eating disordered individuals. In

addition to the finding that bulimic women perceived

themselves as much less socially competent, they were

also rated as less socially effective compared to

control group women. Observers rating the taped

behavior of bulimic women engaging in a five-minute

interaction perceived them to be less trustworthy

leaders, worse at problem-solving, and poorer team

members. Bulimic women were also seen as less skilled

in their social interaction and rated as less

considerate and less likely to be a good friend. Thus,

using two methods to assess the subjects' social

competence, results strongly suggest that bulimic women








are less socially competent compared to non-eating

disordered women.

Implications and Conclusions

These results suggest that there are a number of

aspects of the bulimic individual's social network and

interactions which are indeed quite different from

those of non-eating disordered women. Previous studies

have demonstrated that bulimics' families are more

pathological in a number of ways (Garner et al., 1983;

Humphrey et al., 1985; Johnson & Flach, 1985) and that

bulimics themselves are socially maladjusted in many

areas of life, including work, social and leisure

activities, and relationships with family, spouses, and

others (Herzog et al., 1986; Herzog et al., 1987;

Johnson & Berndt, 1983). The present study adds to

these findings by demonstrating that the bulimic

individual's current interactions and relationships are

more negative and conflictual than those of non-eating

disordered women, and that they demonstrate

significantly poorer social skills. In addition,

unlike previous studies, the present findings are based

not only on self-report but also on observer ratings of

a brief interaction between a bulimic and a non-eating

disordered woman. Finally, results indicate that these

differences between bulimic individuals and controls

are significant even after considering the contribution

of bulimics' greater psychopathology.








In light of recent research clarifying the social

support construct, these findings have important

implications for understanding the bulimic individual's

social environment. Several authors have suggested

that the sense of being loved, valued, and accepted may

be the most active ingredient of social support

(Sarason et al., 1987). Pierce (1988) emphasized the

importance of considering the individual's social

matrix, particularly the nature of personal ties and

the level of intimacy or feeling of being understood,

validated, cared for, and closely connected to others.

Assessing the quality of relationships has proven to

add significantly to the prediction of adjustment,

beyond the contribution of general social support

(Pierce, 1988).

The results of the present study support this

concept, indicating that perceived social support and

the quality of relationships make independent

contributions to an individual's level of adjustment.

The quality of relationships particularly influences

the effect of social support and contributes greatly to

bulimic symptomatology. In addition, a high level of

conflict appears to be even more distressing than a low

level of social support. This finding is consistent

with that of Abbey et al. (1985), who found that social

conflict has an active component that may be more

distressing than lack of support. This high level of








conflict and negative interactions in the bulimic

individual's social network seems to be quite

problematic, in that it is strongly related to the

severity of her bulimic symptoms.

The social interactions of women with bulimia also

appear to be greatly affected by their lack of social

effectiveness. Thus, it is plausible that the bulimic

individual's self-reported social maladjustment is

related to her poor social skills. The present study

provides support for this hypothesis, indicating that

not only do bulimic women report less social

competence, but they are perceived as less socially

effective by others. Results of a recent study by Van

Buren and Williamson (1988) provide further evidence

for this finding. These researchers compared married

bulimic couples to maritally distressed couples and

normal control couples on measures of relationship

satisfaction, conflict resolution, and beliefs about

intimate relationships. Compared to normal controls,

bulimics in their study demonstrated several

similarities to the maritally-distressed women who were

seeking couples therapy. Bulimics experienced a high

level of dissatisfaction with their marriages, and

reported deficiencies in conflict resolution skills,

using few problem-solving skills and frequently

withdrawing from conflict. In addition, they endorsed

a belief that their partners, as well as the quality of








their relationships, cannot change; this belief may

result in fewer active attempts to resolve conflicts

(Van Buren & Williamson, 1988).

Unlike the present study, which included

primarily single bulimic women, the bulimic individuals

in Van Buren and Williamson's study were married and

the information reported pertained to their spousal

relationships. However, it seems highly likely that

these patterns of communication and high levels of

conflict are common to all of the bulimic woman's

interactions. Reibel (1989) discusses at length the

poor communication skills of individuals with bulimia,

particularly their evasion of direct messages and their

tendency to do the "right" thing rather than the "real"

thing. She suggests four misconceptions which may

hamper communication for these women. First, the

bulimic individual censors outgoing messages for fear

that her true feelings such as anger, uncertainty, or

resentment will not be tolerated. Second, she believes

that she is "transparent," and as such avoids eye

contact and does not ask for what she wants, assuming

others know and are ignoring her desires. Third, the

bulimic feels the need to protect others from her own

opinions and feelings, assuming others will judge her

or are not strong enough to deal with her feelings.

Finally, she tends to believe that honest communication

will only destroy relationships. As a result of these








beliefs, the bulimic individual mistrusts herself and

others, is prepared for the worst in relationships, and

feels unable to change anything (Riebel, 1989).

These observations suggest that individuals with

bulimia have very disordered communication patterns,

which are likely to contribute to their lack of social

effectiveness, disturbed interpersonal relationships,

and increased conflict. In addition, it appears that

the bulimic's lack of social skills or competence may

indeed interfere with her ability to take advantage of

available social support, leading to her

dissatisfaction and perhaps to the exacerbation of her

bulimic symptoms. Once again, causal direction is

difficult to determine, but it appears that bulimic

symptoms frequently arise and are sustained by the

individual's conflicted, ambivalent relationships and

lack of intimacy and support.

Researchers attempting to identify risk factors

associated with the development of eating disorders

have implicated several salient factors. These include

some demographic characteristics, personality

variables, family dynamics, a constitutional

disposition, sociocultural influences, and the

physiological and psychological consequences of severe

dieting (Shisslak, Crago, Neal, & Swain, 1987).

Garfinkel, Garner, and Goldbloom (1987) have separated

these risk factors into three general areas --








cultural, familial, and individual -- and suggest that

these vary greatly from individual to individual.

Culturally, the thin female form is idealized and women

are pressured to perform and please others. Within the

bulimic individual's family, there may be a family

history of eating or affective disorders, a

magnification of cultural attitudes, or family

relationships which discourage autonomy. The eating

disordered individual herself may then be particularly

vulnerable to developing disturbed self-perceptions, a

lack of autonomy, and personality features and a

cognitive style which contribute to the onset of

bulimic symptoms. Garfinkel et al. (1987) suggest that

these symptoms are then perpetuated by a number of

factors including the effects of starvation on

thoughts, emotions, and behavior, the use of bulimic

symptoms to modulate affect, depression, and secondary

gain such as the power and sense of specialness derived

from the bulimia. In addition, lack of social skills

and friendships may play a powerful role in maintaining

the self-perpetuating cycle of binge eating and purging

(Garfinkel, Garner, & Goldbloom, 1987).

The present study supports this suggestion and

highlights the fact that bulimic individuals experience

poor relationships with many people in their lives.

This appears to be in part a result of their inability

to communicate honestly and effectively, and to resolve








conflict appropriately. Interpersonal stress has been

found to increase the likelihood of binge eating (Van

Buren & Williamson, 1988), and indeed binge eating and

purging may relieve the tension resulting from frequent

unresolved conflict and dissatisfaction. Bulimic

behavior may also provide a means of attempting to

obtain acceptance and intimacy by becoming more

desirable and attractive. In addition, bulimic

symptoms may function as an attempt to fill needs for

self-gratification which are not met by the bulimic

individual's social network. In any case, it appears

that a comprehensive treatment program must address the

bulimic's lack of social effectiveness and poor

communication and interaction skills which contribute

to the perpetuation of her bulimic symptoms.

Limitations of the present study include its

relatively small sample size and demographically

restricted sample, which consisted of undergraduate

women screened for bulimic symptoms and recruited for

research purposes. In addition, the measure utilized

to assess the quality of relationships asked subjects

to report information on four individuals (father,

mother, closest female friend, and romantic

relationship/closest male friend). Although these

individuals were chosen because they are likely to be

important in the bulimic's life, this measure may have

excluded information about relationships with other








significant individuals. A final limitation of the

study involves the measures utilized to assess social

competence and social effectiveness, which were brief

and consisted of somewhat general items. Bulimic women

reported overall social competence in a variety of

situations, but did not directly report their feelings

and perception of competence during the videotaped

interaction. In addition, the scale utilized by

observers to rate bulimics' social effectiveness was

somewhat limited in the depth and breadth of behaviors

and verbalizations considered.

Despite these limitations, results of this study

have important implications for treatment. Teaching

communication, coping, and problem-solving skills may

meet important needs for bulimic women. These skills

can improve daily functioning, increase their sense of

self-efficacy, and gradually improve the quality of

their relationships. This may in turn enable the

bulimic individual to take advantage of available

social support and interact more effectively with

others, thus challenging several factors which

perpetuate bulimic symptoms. Clear listening and

reality testing, beginning to understand and convey her

real messages and needs, learning to comunicate

directly and honestly -- all of these skills could

decrease the bulimic's need to use food in the

maladaptive manner of eating disordered women. Further








treatment studies would be helpful in determining the

impact and effectiveness of these interventions in

decreasing bulimic symptoms.

The results of the present study clearly have

implications for understanding the social network and

interpersonal interactions of individuals with bulimia.

In order to generalize beyond the results from this

sample, this study should be replicated with a larger

sample and with community samples that include male

bulimics, a wider age range, and diversity of

demographic backgrounds. Future studies might also

utilize more extensive and in-depth measures of social

effectiveness to target more specific areas for

treatment. For example, it would be helpful to obtain

both quantitative and qualitative ratings of social

effectiveness in a given situation, including measures

of nonverbal behaviors (e.g., eye contact and speech

duration), the quality of interaction, and the quality

of problem-solving ability. These measures should be

obtained not only from observers, but also from

participants in interactions with bulimics. In

addition, it would be interesting to have the bulimic

individuals rate their own social performance during

the interaction. In this way it would be possible to

identify the presence of particular behaviors which

contribute to the social incompetence and interpersonal

difficulties of eating disordered women, and to learn








more about the bulimic individual's perceptions,

thoughts, and feelings during interactions.

In addition to further exploration of the

dysfunctional nature of the communication patterns of

bulimic individuals, studies are also needed to explore

in more detail the quality of their relationships with

a variety of significant individuals. It will be

important to obtain more information about the

bulimic's intra- and interpersonal conflict and

dysfunction, as well as other aspects of her social

interactions which may not have been tapped in the

present study. This should include further

investigation of the bulimic individual's social

support system and her dissatisfaction with this

support. Only by continuing to increase our

understanding of these complex factors will it be

possible to formulate more comprehensive and fine-tuned

theory, as well as plan more appropriate and effective

approaches to the treatment of bulimia nervosa.














APPENDIX A

QUESTIONNAIRES









BULIT


Answer each question on the following pages by filling
in the appropriate circles on the computer answer
sheet. Please respond to each item as honestly as
possible; remember, all of the information you provide
will be kept strictly confidential.

1. Do you ever eat uncontrollably to the point of
stuffing yourself (i.e., going on eating binges)?
(a) Once a month or less (or never)
(b) 2-3 times a month
(c) Once or twice a week
(d) 3-6 times a week
+(e) Once a day or more

2. I am satisfied with my eating patterns.
(a) Agree
(b) Neutral
(c) Disagree a little
(d) Disagree
+(e) Disagree strongly

3. Have you ever kept eating until you thought you'd
explode?
+(a) Practically every time I eat
(b) Very frequently
(c) Often
(d) Sometimes
(e) Seldom or never

4. Would you presently call yourself a "binge eater"?
+(a) Yes, absolutely
(b) Yes
(c) Yes, probably
(d) Yes, possibly
(e) No, probably not

5. I prefer to eat:
+(a) At home alone
(b) At home with others
(c) In a public restaurant
(d) At a friend's house
(e) Doesn't matter

6. Do you feel you have control over the amount of food
you consume?
(a) Most or all of the time
(b) A lot of the time
(c) Occasionally
(d) Rarely
+(e) Never








X 7. I use laxatives or suppositories to help control
my weight.
(a) Once a day or more
(b) 3-6 times a week
(c) Once or twice a week
(d) 2-3 times a month
(e) Once a month or less (or never)

8. I eat until I feel too tired to continue.
+(a) At least once a day
(b) 3-6 times a week
(c) Once or twice a week
(d) 2-3 times a month
(e) Once a month or less (or never)

9. How often do you prefer eating ice cream, milk
shakes, or
puddings during a binge?
+(a) Always
(b) Frequently
(c) Sometimes
(d) Seldom or never
(e) I don't binge

10. How much are you concerned about your eating
binges?
(a) I don't binge
(b) Bothers me a little
(c) Moderate concern
(d) Major concern
+(e) Probably the biggest concern of my life.

11. Most people would be amazed if they knew how much
food I can consume at one sitting.
+(a) Without a doubt
(b) Very probably
(c) Probably
(d) Possibly
(e) No

12. Do you ever eat to the point of feeling sick?
+(a) Very frequently
(b) Frequently
(c) Fairly often
(d) Occasionally
(e) Rarely or never

13. I am afraid to eat anything for fear that I won't
be able to stop.
+(a) Always
(b) Almost always
(c) Frequently
(d) Sometimes
(e) Seldom or never








14. I don't like myself after I eat too much.
+(a) Always
(b) Frequently
(c) Sometimes
(d) Seldom or never
(e) I don't eat too much

15. How often do you intentionally vomit after eating?
+(a) 2 or more times a week
(b) Once a week
(c) 2-3 times a month
(d) Once a month
(e) Less than once a month (or never)

16. Which of the following describes your feelings
after binge eating?
(a) I don't binge eat
(b) I feel O.K.
(c) I feel mildly upset with myself
(d) I feel quite upset with myself
+(e) I hate myself

17. I eat a lot of food when I'm not even hungry.
+(a) Very frequently
(b) Frequently
(c) Occasionally
(d) Sometimes
(e) Seldom or never

18. My eating patterns are quite different from eating
patterns of most people.
+(a) Always
(b) Almost always
(c) Frequently
(d) Sometimes
(e) Seldom or never

19. I have tried to lose weight by fasting or going on
"crash" diets.
(a) Not in the past year
(b) Once in the past year
(c) 2-3 times in the past year
(d) 4-5 times in the past year
+(e) More than 5 times in the past year

20. I feel sad or blue after eating more than I'd
planned to eat.
+(a) Always
(b) Almost always
(c) Frequently
(d) Sometimes
(e) Seldom, never, or not applicable








21. When engaged in an eating binge, I tend to eat
foods that are high in carbohydrates (sweets and
starches).
+(a) Always
(b) Almost always
(c) Frequently
(d) Sometimes
(e) Seldom, or I don't binge

22. Compared to most people, my ability to control my
eating behavior seems to be:
(a) Greater than other's ability
(b) About the same
(c) Less
(d) Much less
+(e) I have absolutely no control

23. One of your best friends suddenly suggests that you
both eat at a new restaurant buffet that night.
Although you'd planned on eating something light at
home, you go ahead and eat out, eating quite a lot and
feeling uncomfortably full. How would you feel about
yourself on the ride home?
(a) Fine, glad that I'd tried the new restaurant
(b) A little regretful that I'd eaten so much
(c) Somewhat disappointed in myself
(d) Upset with myself
+(e) Totally disgusted with myself

24. I would presently label myself a "compulsive eater"
(one who engages in episodes of uncontrolled eating).
+(a) Absolutely
(b) Yes
(c) Yes, probably
(d) Yes, possibly
(e) No, probably not

25. What is the most weight you've ever lost in one
month?
+(a) Over 20 pounds
(b) 12-20 pounds
(c) 8-11 pounds
(d) 4-7 pounds
(e) Less than 4 pounds

26. If I eat too much at night I feel depressed the
next morning.
+(a) Always
(b) Frequently
(c) Sometimes
(d) Seldom or never
(e) I don't eat too much at night








27. Do you believe that it is easier for you to vomit
than it is for most people?
+(a) Yes, it's no problem at all for me
(b) Yes, it's easier
(c) Yes, it's a little easier
(d) About the same
(e) No, it's less easy

28. I feel that food controls my life.
+(a) Always
(b) Almost always
(c) Frequently
(d) Sometimes
(e) Seldom or never

29. I feel depressed immediately after I eat too much.
+(a) Always
(b) Frequently
(c) Sometimes
(d) Seldom or never
(e) I don't eat too much

30. How often do you vomit after eating in order to
lose weight?
(a) Less than once a month (or never)
(b) Once a month
(c) 2-3 times a month
(d) Once a week
+(e) 2 or more times a week

31. When consuming a large quantity of food, at what
rate of speed do you usually eat?
+(a) More rapidly than most people have ever eaten in
their lives
(b) A lot more rapidly than most people
(c) A little more rapidly than most people
(d) About the same rate as most people
(e) More slowly than most people (or not applicable)

32. What is the most weight you've ever gained in one
month?
+(a) Over 20 pounds
(b) 12-20 pounds
(c) 8-11 pounds
(d) 4-7 pounds
(e) Less than 4 pounds

X 33. Females only. My last menstrual period was
(a) Within the past month
(b) Within the past 2 months
(c) Within the past 4 months
(d) Within the past 6 months
(e) Not within in the past 6 months









X 34. I use diuretics (water pills) to help control my
weight.
(a) Once a day or more
(b) 3-6 times a week
(c) Once or twice a week
(d) 2-3 times a month
(e) Once a month or less (or never)

35. How do you think your appetite compares with that
of most people you know?
+(a) Many times larger than most
(b) Much larger
(c) A little larger
(d) About the same
(e) Smaller than most

X 36. Females only. My menstrual cycles occur once a
month:
(a) Always
(b) Usually
(c) Sometimes
(d) Seldom
(e) Never


+ represents the most 'symptomatic' response and
receives a score of 5 points

X denotes questions for which responses are not
included in the summed BULIT score








PSS-Fr

The statements which follow refer to feelings and
experiences which occur to most people at one time or
another in their relationships with friends. For each
statement there are three possible answers: Yes, No,
and Don't know. Please use the scale provided when
answering each question, and mark your answers on the
answer sheet.
1 2 3
Yes No Don't know

1. My friends give me the moral support I need.
2. Most other people are closer to their friends than
I am.
3. My friends enjoy hearing about what I think.
4. Certain friends come to me when they have problems
or need advice.
5. I rely on my friends for emotional support.
6. If I felt that one or more of my friends were upset
with me, I'd just keep it to myself.
7. I feel that I'm on the fringe in my circle of
friends.
8. There is a friend I could go to if I were just
feeling down, without feeling funny about it later.
9. My friends and I are very open about what we think
about things.
10. My friends are sensitive to my personal needs.
11. My friends come to me for emotional support.
12. My friends are good at helping me solve my
problems.
13. I have a deep sharing relationship with a number of
friends.
14. My friends get good ideas about how to do things or
make things from me.
15. When I confide in friends, it makes me feel
uncomfortable.
16. My friends seek me out for companionship.
17. I think that my friends feel that I'm good at
helping them solve problems.
18. I don't have a relationship with a friend that is
as intimate as other people's relationships with
friends.
19. I've recently gotten a good idea about how to do
something from a friend.
20. I wish my friends were much different.








PSS-Fa

The statements which follow refer to feelings and
experiences which occur to most people at one time or
another in their relationships with their families.
For each statement there are three possible answers:
Yes, No, and Don't know. Please use the scale provided
when answering each question, and mark your answers on
the answer sheet.
1 2 3
Yes No Don't know

1. My family gives me the moral support I need.
2. I get good ideas about how to do things or make
things from my family.
3. Most people are closer to their families than I am.
4. When I confide in the members of my family who are
closest to me, I get the idea that it makes them
uncomfortable.
5. My family enjoys hearing about what I think.
6. Members of my family share many of my interests.
7. Certain members of my family come to me when they
have problems or need advice.
8. I rely on my family for emotional support.
9. There is a member of my family I could go to if I
were just feeling down, without feeling funny about
it later.
10. My family and I are very open about what we think
about things.
11. My family is sensitive to my personal needs.
12. Members of my family come to me for emotional
support.
13. Members of my family are good at helping me solve
problems.
14. I have a deep sharing relationship with a number of
members of my family.
15. Members of my family get good ideas about how to do
things or make things from me.
16. When I confide in members of my family, it makes me
uncomfortable.
17. Members of my family seek me out for companionship.
18. I think that my family feels that I'm good at
helping them solve problems.
19. I don't have a relationship with my family that is
as close as other people's relationships with
family members.
20. I wish my family were much different.





89


Social Interactions Scale

In the course of daily living, people's interactions
with others can be pleasant, helpful, and supportive,
or they can be a source of conflict and negative
feelings. Please read each question below, and using
the answer sheet, mark the number that best reflects
the social interactions you have had in the past week.

1. In the past seven days, how much have people in your
personal life
acted in ways that show they appreciate you?

1 2 3 4 5 6 7
not at all a great deal

2. In the past seven days, how much have people in your
personal life treated you with respect?

1 2 3 4 5 6 7
not at all a great deal

3. In the past seven days, how much have people in your
personal life shown you that they cared about you as
a person?

1 2 3 4 5 6 7
not at all a great deal

4. In the past seven days, how much have people in your
personal life given you useful information and
advice when you wanted it?

1 2 3 4 5 6 7
not at all a great deal

5. In the past seven days, how much have people in your
personal life helped out when too many things needed
to get done or you couldn't do them yourself?

1 2 3 4 5 6 7
not at all a great deal

6. In the past seven days, how much have people in your
personal life listened when you wanted to confide
about things that were important to you?

1 2 3 4 5 6 7
not at all a great deal








7. In the past seven days, how much have people in your
personal life visited with you?


1
not at all


a great deal


8. In the past seven days, how much have people in your
personal life argued with you about something?


1
not at all


a great deal


9. How much did these occurrences bother you,
including your emotions, your thoughts, and your
behavior?


1
no impact


a great deal
of impact


10. How would you explain why these interactions
occurred?


1 2
something to
do with me


something to do
with others or
the situation


11. In the past seven days, how much have people gotten
on your nerves?


1
not at all


a great deal


12. How much did these occurrences bother you,
including your emotions, your thoughts, and your
behavior?


1
no impact


a great deal
of impact


13. How would you explain why these interactions
occurred?


1 2
something to
do with me


something to do
with others or
the situation





91


14. In the past seven days, how much have people in
your personal life misunderstood the way you
thought and felt about things?


1
not at all


6 7
a great deal


15. How much did these occurrences bother you,
including your emotions, your thoughts, and your
behavior?


1
no impact


a great deal
of impact


16. How would you explain why these interactions
occurred?


something to
do with me


something to do
with others or
the situation


17. In the past seven days, how much have people in
your personal life done things that conflicted with
your own sense of what should be done?


1
not at all


6 7
a great deal


18. How much did these occurrences bother you,
including your emotions, your thoughts, and your
behavior?


1
no impact


6 7
a great deal
of impact


19. How would you explain why these interactions
occurred?


1 2
something to
do with me


something to do
with others or
the situation


20. In the past seven days, how much have people in
your personal life acted in an unpleasant or angry
manner toward you?


a great deal


not at all








21. How much did these occurrences bother you,
including your emotions, your thoughts, and your
behavior?


1
no impact


a great deal
of impact


22. How would you explain why these interactions
occurred?


something to
do with me


something to do
with others or
the situation


23. In the past seven days, how much have people in
your personal life invaded your privacy?


1
not at all


6 7
a great deal


24. How much did these occurrences bother you,
including your emotions, your thoughts, and your
behavior?


1
no impact


a great deal
of impact


25. How would you explain why these interactions
occurred?


1 2
something to
do with me


something to do
with others or
the situation


26. In the past seven days, how much have people in
your personal life treated you as though they did
not respect or value you as a person?


1
not at all


6 7
a great deal


27. How much did these occurrences bother you,
including your emotions, your thoughts, and your
behavior?


1
no impact


a great deal
of impact