L Idi

JZ6J/ '6?

VERSITY LIBRARIES

1....“ ; #1 lllllulllllHll'lllllll1l l w OIL}

LID-u «as f 3 1293 005814

Michigan State
University

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ll,

 

 

 

This is to certify that the

dissertation entitled

THE RELATIONAL SELF IN WOMEN:
PERSPECTIVES ON SELF AND OTHERS

presented by
Lisa Sherron Blank

has been accepted towards fulfillment
of the requirements for

Ph.D Psychology

p/ ' degree in

 

 

5m 61%

Major professor 1

Ellen A. Strommen

Date 7/31/87

 

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

 

“I,” ‘AAM"A-__~.~ . _.
.- - _ . __ ks AA..___ ~d‘ ~-_
_ 4—.w’ H~_...“-_A—-___ w__-4 vs .- _. _._. ,.. _-,‘ ¢
— 4% MW‘ ..._.4-
‘IH' ._.. - o“_ 'Hfl&fl 4 -fi - -—-

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE

 

 

 

.0
291

arm
/1,:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmative ActloNEquel Opportunity Institution

 

THE RELATIONAL SELF IN WOMEN:

PERSPECTIVES ON SELF AND OTHERS

by

Lisa Sherron Blank

A DISSERTATION

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

DOCTOR.OF PHILOSOPHY

Department of Psychology

1988

ABSTRACT

THE RELATIONAL SELF IN W:
PERSPECTIVES (1‘1 SELF AND OTHERS

By
Lisa Sherron Blank

Gilligan (1982) and others (Jordan, 1964; Surrey, 1985; Kaplan,
1986) emphasize the role of connection and interdependence in women's
identity formation and moral development, proposing a relational self-
concept based on connection with others and oriented to activities of
care. The Revised Relationship Self Inventory (RRSI) was developed to
measure Gilligan's distinction between separate/objective and
relational/connected self orientations for women and men, as well as
different manifestations of the relational/ connected self corresponding
to Gilligan's differentiations of the meaning of care for self and
others. This research used the RRSI to identify women describing
themselves as relational/connected selves, but differing in their
understanding of the role of self and others in activities of care, as
reflected in their differential endorsements of the Primacy of Other
Care (POC) and Self and Other Care Chosen Freely (SOCCF) scales. The
results of this study described the relationships between these
relational identity styles and variables of traditional interest for the
psychology of mnen: depression, self-esteem, anger and power.

Participants were 61 female undergraduates at a large midwestern
university and 262 adult women attending an on-campus enrichment program

at the same university, who voluntarily completed the following

Lisa Sherron Blank
instruments: RRSI, center for Epidemiological Studies Depression Scale,
Rosenberg Self-Esteem Inventory, the anger subscale of the EASI-III
Temperament Survey and sentence completion items about anger and power.
Correlational analyses explored relationships between FCC and SOCCF and
depression, self-esteem, and anger. Fbur groups were created using
median scores on FCC and SOCCF, with subgroups of high.POC/low SOOCF and
high SOCCF/low POC women of particular interest for comparisons on these
variables.

Hypothesized relationships between relational style and depression
were not supported. The relationships of self-esteem with FCC and with
SOCCF were significantly different and in the predicted directions, with
POC significantly related to lower self-esteem; the relationship between
SOCCF and self-esteem was positive but not significant. FCC and SOCCF
related differently to anger, with a significant negative relationship
between SOCCF and anger. The two subgroups of interest did not differ
significantly on depression, self-esteem, responses to anger and
experiences of power. Measurement issues for the RRSI scales were

discussed in relation to these results.

PKTTKHHEIEEPENHS

writing a dissertation is a long, absorbing, often frustrating
process which takes tremendous energy, self-discipline, and perspective.
Now, at the end of that process and of graduate education itself, it is
immensely satisfying to recognize and thank-those people who supported,
encouraged, badgered, nudged, advised, befriended, and helped me to
reach this point. It is often the odd, small human gestures that
sustain most: the blunt directive from a friend to have a drink, go to
bed and think about it in the morning, the shared story of a faculty
member's own dissertation struggles, companionable assistance at the
computer center, a celebratory lunch after orals. As someone who values
interdependence and relationships for her own identity development, it
is a special joy to acknowledge the varied and rich connections with
friends, family, faculty, and fellow researchers, who have nurtured my
personal and professional development.

I am very grateful for the caring and support of friends who
endured this process (and met): for Suzy Cock, Susan Burns, David van
HOuten, Robert Hill, Matthew'Engel, and my sister Leslie. I am
especially grateful for the computer assistance and expertise of Jeff
vancouver and Fred Rogosch, with whom I shared many graduate school
adventures. The two~woman dissertation support group dinners*with Diane
Trebilcock.were fun, and the patient encouragement of WES NOvak critical

in moments of panic and despair.

iv

I am also indebted to the people for whom and with whom I worked in
the Department of Psychiatry during my dissertation year: Terry Stein,
Diane Singleton, Carol Schwartz, Robert Hill, and especially Karen
Miske. I appreciate their loving and patient support of ne through all
the normative aspects of a dissertation experience- exhilaration and
anxiety attacks, avoidance rituals, and obscenity-laced wranglings with
the computer and printer.

In my own department, I have enjoyed working with John Hurley, Bert
Karon, and Elaine Donelson for conprehensive exams. John has been
important to me in many roles: faculty advisor, director of my masters
thesis research, dissertation committee member, and interested, caring
friend. I am also fortunate in knowing my chairperson, Ellen Strcmren,
and other members of my dissertation comnittee, Elaine Donelson and Mary
Ann Reinhart, as fellow members of a research group which has net for
several years. Being a part of this group effort has been the best
possible setting for tie to learn how to design and carry out
psychological research on topics that are personally and professionally
meaningful. This collaborative, cooperative endeavor, with all its
challenges, benefits and problems, was a lived-experience of the values
inplicit in the intellectual model we struggled to articulate and
explore: the importance of interdependence and commnion for human
development. I learned a lot from faculty, graduate and undergraduate
menbers of this group, and was privileged to have Ellen, Elaine, and
Mary Ann as cannittee members.

azzy Pavick also deserves nuch thanks and fond affection for her
helpfulness as a departmental ”guide", conpetent typist, and cheerful,

concerned presence in the department. Graduate students are the

fortunate recipients of her consistent ability to negotiate the
confusing maze of deadlines, program requirements, student loan
communications, and other critical issues of graduate education. I am
grateful for her help and support.

Finally, even with the limitations of this particular researdh, I
have been glad for the chance to explore important aspects of human
experience: relationships, identity, depression, self-esteem, anger, and
power. These are the common themes and threads of countless
conversations with woman friends and clients, as we share experiences,
strengths and problems. I am gratefully indebted to those friends,
clients, therapists and supervisors who have struggled with me towards
better clinical and personal understanding of these basic issues. This
particular project is finished, admittedly with joyful celebration and

grateful relief, but the conversation will go on.

vi

TABLE OF CONTENTS

LIST OF TABIES. O O O O O O O O O O O O O O O O O O O O
IMMIWO O O O C O O O O O O O O O O O O O O O O O
NewModelsofDevelopment. . . . . . . . . . . . .

The Relational Self. . . ...... _ .......
The Relational Self in Adolescence . . .

Gilligan's Model . . . . . . . . . . . . . . . . .
The Revised Relationship Self Inventory. . . . . .
Depression . . . . . . . . . . . . . . . . . . . .
Self-Esteem. . . . . . . . . . . . . . . . . . . .
Anger. . . . . . . . . . . . . . . . . . . . . . .
Power. . . . . . . . . . . . . . . . . . . . . . .
fimmary of Model and Research Hypotheses . . . . .

mm C O O O O O O O O O O O O O O O 0

Participants . . .
wama....
Ethical Review
Instruments. . . . . . . . . . . .
Revised Relationship Self Inventory. . . . .
center for Epidemiological Studies Depression
Scale. . . . . . . . . . . . . . . . . . .

Rosenberg Self-Esteem Inventory. . . . . . . .
Anger Scale. . . . . . . . . . . . . . . . . .
Sentence completions . . . . . . . . . . . . .
Sentence completion codes for Powerful . . . .
Sentence completion codes fOr Anger. . . . . .
Data coding. . . . . . . . . . . . . . . . . . . .
Sentence completion coding . . . . . . . . . .
Analyses . . . . . . . . . . . . . . . . . . . . .

REWTS O O O O C O C O O O O O O O C O O O O O O O O 0

Descriptive Statistics . . . . . . . . . . . . . .
Relationships among Relational Perspectives
andOtherVariables............ .
comparisons of High FCC/low SOOCF and High SOOCF/
low Poc women on Depression, Self-esteem,
and Anger . . . . . . . . . . . . . . . . . .

vii

11
13
18
21
30
34
38
42

46

46
48
48
48
49

50
51
51
52
54
57
61
61
62

66
66

70

74

Comparisons of High POC/low SOCCF and High SOCCF/
low POC women on Anger and Power Sentence

m1 etion mta O O C O C O O O O O O O O 77

Mger. O O O O I O O C O O O O O O O O O O O O O O 78
mr. O O O O O O O O O O 0 O O O O O O O O O O O 83
Summary of Results Related to Hypotheses . . . . . 87

DISCUSSIme O O O O O O O O O O O O O O O O O O O O O O O 92

Depression . . 92

Self-esteem 95
Ammn.... ... ... ... ... ... ... .. 96
Power. . . . . . . . . . . . . . . . . . . . . . . . 100
Relationships among Depression, Self-esteem,
and Anger . . . . . . . . . . . . . . . . . . . 102
Measurement Considerations . . . . .-. . . . . . . . 103
Other Measurement Issues . . . . . . . . . . . . . . 108
Summary. . . . . . . . . . . . . . . . . . . . . . . 109
APPENDICES. . . . . . . . . . . . . . . . . . . . . . . . 110
Appendix A . . . . . . . . . . . . . . . . . . . . . 110
Appendix B . . . . . . . . . . . . . . . . . . . . . 112
Appendix c . . . . . . . . . . . . . . . . . . . . . 122
Appendix D . . . . . . . . . . . . . . . . . . . . . 133

W8. 0 O O O O O O O O O O O O O O O O O O O O O O 140

viii

LIST OF TABLES

Table Page
Tables in Body

 

1. Means and Standard Deviations for Sample . . . . . . . 68

2. Comparisons of undergraduate and college week women
in Research Sample. . . . . . . . . . . . . . . . 69

3. Intercorrelations Among All Measures . . . . . . . . . 71
4. Means and Standard Deviations for Groups . . . . . . . 76

5. Frequencies and Percents for Anger Categories
by GI-oups O O O O O O O O O O O O O O O O O O O O 7 9

6. Anger Categories Organized from Most to Least
Frequently Used by Groups . . . . . . . . . . . . 81

7. Frequencies and Percents for Power Categories
by Groups . . . . . . . . . . . . . . . . . . . . 84

8. Power Categories Organized from Most to Least
Frequently Used by Groups . . . . . . . . . . . . 85

Tables in Appendix C

 

C1. REliabilities (alma)e e e e e e e e e e e e e e e e e 126
C2. Scale Intercorrelations. . . . . . . . . . . . . . . . 126

c3. ItemrScale Total correlations and Scale Reliabilites
of the Revised Relationship Self Inventory. . . . 127

INTRODUCTION

Recent exploration of the phenomenology, psychology, and
development of women's lives across the life cycle has yielded new ways
of describing and understanding the experiences of women. The central
role and significance of attachment and interdependence for women's
identity formation and moral development has emerged as an important
theme. Recognition and articulation of a relational self-concept (where
identity is developed through relationships) informs the psychology of
women, validates women's individual experiences and personal integrity,
and shapes clinical interventions with women. It also presents a
challenge to more general theories of human development.

The failure of women to fit existing models of human growth, often
interpreted as problems or inadequacies of female development,
increasingly suggests limitations in the models themselves. These
limitations are the result of a bias of perspective that occurs when
male experience is assumed normative for all human experience.
Traditional theories of identity development emphasize the role of
separation from others , individuation, and autonomy. These theories nay
be enriched and balanced by continued exploration of central themes in
women's lives, thereby moving towards a more encompassing vision of
human health and development. This research is one such exploration of
women's identity development, as it relates to experiences of

depression, selfeesteem, anger, and power.

2

As a result of listening to wonen describe their views of
themselves, their relationships, and their experiences of moral dilemmas
and choices, Gilligan (1982) has proposed two distinct modes of
organizing the self in relation to others. The separate/objective self
is based on an autonomous experience and perception of the self, and
stresses individual achievement and objective reciprocity in
relationships. This View of the self is associated primarily with men,
and is accurately described by most traditional theories of identity
developuent across the life span (e.g., Erikson, 1950; Kolhberg, 1981;
Vaillant, 1977; Levinson, 1978, etc.)

The relational/connected self is based on a view of the self as
interdependent and interconnected with others, and stresses activities
of care as a means of creating and sustaining connection. This
perspective is associated primarily with wonen, and has not been
described or included in psychological theories of the life cycle,
except as pathological. Among wouen for whom identity is developed
through connection with others, Gilligan has traced a developmental
sequence of three perspectives representing increasingly conplex and
differentiated understandings of care of self and others: 1) care of
self is necessary because others will not care; 2) care of others has
priority over care of self; and 3) care of self is chosen freely since -
care for all, including self, is inportant.

Gilligan has based mch of her theory on small and highly educated
samples of women facing various dilamas and life transitions, using a
methodology of lengthy semi-structured interviews. Other theorists
using data drawn primarily from their clinical experience with mnen

also note the prevalence of a relational self-concept in women's

3
accounts of their lives. Their work attests to the value of the
selfein-relation construct as a cornerstone for a theory of women's
development (Surrey, 1985). Self-in-relation theory also highlights the
role of empathy in relational development (JOrdan, 1984), and provides
an analysis of depression as a distortion of relational development
(Kaplan, 1986). However, with limited research evidence, time-consuming
methodology, and few instruments available to measure new constructs,
these models have not been validated with larger and more representative
samples of women and men.

The deveIOpment of an inventory to measure relationship orientation
which can be easily administered and used with large samples of
adolescents and adults has been the goal of an ongoing research program.
The Revised Relationship Self Inventory (RRSI) is a paper—and-pencil,
self-report instrument which is proposed to reliably measure Gilligan's
distinction between separate/objective and relational/connected self
orientations for both men and women (Strommen, Reinhart, Pearson,
Barnes, Blank, cebollero, cornwell, Donelson, & Kamptner, 1987). It
also includes measures of different manifestations of the relationall
connected self, corresponding to Gilligan's differentiations of the
meaning of care for self and others (Primacy of Other Care, and Self and
Other care Chosen Freely.) validation studies examining hypothesized
relationships between style of relationship self and other processes are
in progress.

Ruth the RRSI it becomes possible to explore and expand these
concepts of identity and relationship orientation using large samples of
men and women. The focus of my researCh, however, is limited to

exploration of the relational/connected self as it illuminates the

4
development of women, and to expanding current understandings of the
differences among wmen in terms of their perspectives on the meanings
of care of self and others. looking at women who are similar in their
descriptions of themselves as relational 1y oriented, but who differ in
their understanding of the role of self and other in activities of care,
may clarify the relationship of these differentiating perspectives and
their psychological correlates for women. Hypotheses linking these
differentiating perspectives to more traditional issues of women's
psychology, particularly depression, self-esteem, anger, and power, test
the value of these perspectives in accounting for differences among
women in areas recognized as conflictual or generally problematic for
women. By describing the relationships between relational identity
style and depression, self-esteem, anger, and power, this research
attempts to expand the model of the relational self for women, and to
demonstrate the value of the self-in-relation theory for understanding

and guiding mmen's development.

New Models of Development

 

A new model of development emerging from the work of several
theorists describes the development of the self in the context of
relationships, emphasizing the continuing role of attachment,
correction, and interdependence across the life span. Historical
antecedents of this view of the self as inseparable from interpersonal
interaction, necessarily involving reference to others, include Cooley
(1902), Mead (1934), and Sullivan (1953). As currently formulated, this
model explains the development of a communal or relational identity

consistent with the life experiences of many women. It also offers a

5
framework for considering developmental differences of men and women,
and highlights a line of development for both sexes previously
overshadowed in psychological literature and theory by other accounts of
development.

Gilligan (1982) and others criticize many of the foremost theories
of identity development for their emphasis on the role of separation and
autonomy in development. Some theories do describe the role of
attachment and the contribution of nurturing relationships to ego
development and achievement of basic trust (Erikson, 1950) in early
infancy. The tasks of subsequent psychosocial development through
childhood, adolescence, and young adulthood, however, are described
largely in terms of individuation, autonomy, initiative, and industry.
These terms stress the independent, achievement-oriented focus of human
activity. In Bakan's terminology, this describes the development of the
agentic self (Bakan, 1966). The importance of separation from others
for the development of an individualized identity is stressed: from
mother in early childhood (Mahler, 1975), from the family at adolescence
(Erikson, 1963), and from mentors in adulthood (Ieyinson, 1978).

‘ Although adult capacities for intimacy and generativity (Erikson,
1950) are assumed to follow from the development of a secure sense of
identity, there is little description of how these capacities are
developed and nurtured before fruition in adulthood. Other theories of
adult development articulate the role of work in continued development
(Ievinson, 1978; vaillant, 1977), but are relatively silent about other
important areas of development and functioning: friendships, marital,
and parent-Child relationships. The relational capacities of the

nurturing parent deemed so critical in the creation of an optimal

6
environment for early development (”good enough" mothering responsible
for the stable "holding environment” for the child— Winnecott, 1971)
are usually discussed only in terms of their outcomes for the child.
They become a focus of attention themselves usually in the context of
responsibility for production of psychopathology.

In these accounts, then, development itself becomes identified with
separation, and maturity with the depiction of independent, autonomous ,
achievement-oriented striving. Against this standard, concerns with and
skills related to connection with others appear regressive or are
devalued. The importance of affiliation and the fusion of intimacy with
identity for women may be acknowledged as sex differences (Erikson,
1968) , but do not alter the basic models of identity development.
Inability of these theories to account for women' 5 development suggests
that important truths about women, and about those aspects of human
functioning women are said to "carry” for our society (Miller, 1976) ,

may be missing from our accounts of development.

The Relational Self

 

Theorists studying the experiences of women, particularly
mother—daughter relationships , describe a new developmental pathway that
begins with attachment and proceeds through increasing differentiation
within relationships. Here emphasis shifts from separation to
relationship as the basis for self-experience and development, and other
aspects of self-development (autonomy, assertion, creativity,
competence, etc.) are assumed to emerge in the context of relationships.
Relational needs and realthy relationships thus propel psychological

growth (Surrey, 1985). Chodorow (1974, 1976) decribes the early

7
mother-daughter relationship as it shapes feminine personality through
the fusion of the experience of attachment with the process of identity
formation.

The double identification a woman makes with her own mother and
with her child (as herself) is likely to be strengthened with a female
child. It is thus easier for mothers to identify with and experience
their daughters as themselves, and for daughters to identify with their
major caretaker as a same-sex figure. Societal values also support the
early mother—daughter attachment and allow greater self-other boundary
flexibility in girls. In contrast, there is pressure in and on the
mother-son relationship against primary identification and towards more
rigid self-other differentiation in boys. Chodorow notes the possible
pathological outcomes of developmental distortions for each sex:
narcissistic projection of the mother onto the daughter, with resulting
boundary confusion for the daughter , and seductive behavior tmards sons
viewed as objects for the mother.

Jordan, alrrey and Kaplan (1983) relate later sex differences in
empathy (with girls more motivated than boys to attend to affect in
others— Hoffman, 1977) to the affectively—tuned mirroring, mutual
identification, and empathic interplay between mother and daughter .

This early experience may strengthen the girl's sense of relatedness and
her feeling of being directly and emotionally understood, as well as
encourage her development of empathic skills (Jordan, 1984) .

Female gender identity is also facilitated through relationships.
Chodorow (1974) describes the development of a girl's gender identity as
continuous with her earliest identification and attachment to mother.

Her gender role identification is leaned within and mediated by her

8

affective relationship with someone who is consistently and familiarly
available to her as a real person, allowing for a personal
identification with mother's character traits and values. Feminine role
activities are carried out in her daily world, directly and immediately
apprehensible to her. Thus, feminine identification is based on ”the
gradual learning of a way of being familiar in everday life, and
exemplified by the person... with whom she has been most involved...
continuous with her early childhood identification and attachments"
(Chodorow, 1974, p.51).

The need for girls to transfer primary sexual object choice from
mother/ females to father/males was originally considered by Freud and

early psychoanalysts to create a major discontinuity in female

 

development. This shift in object choice was said to make female
development tortuous, difficult, and ultimately vulnerable to incomplete
resolution of the "oedipal crisis." Chodorow (1974) reinterprets this
view. She acknowledges that the "oedipal crisis” is not resolved for
girls in the same absolute way hypothesized for boys, because a girl
develops her relationship to her father wi_th_ip_ the context of her
alliance with her mother, and the strength and quality of this new
relationship is related to the quality of the earlier relationship.
Chodorow challenges the view of the incomplete rejection of mother as
inadequate resolution of an important conflict, with negative
consequences for personality development. She suggests, instead, "that
a girl's internalized and externalized real object relations become and
remain more complex, and at the same time more defining of her, than

those of a boy,” (1974, p. 53) , resulting in general genital

9
heterosexuality and the broad and rich variety of interpersonal
relationships characteristic of adult women.

Stone center theorists (Surrey, 1985; JOrdan, Surrey & Kaplan,
1983; Miller, 1984) also describe early developmental precursors of
women's relational self-structure in the mother-daughter relationship.
The mother's empathic mirroring of the child as described by'wunnecott
(1971), Kohut (1971), and others supports the child's early development
of a self. This also begins a reciprocal process of learning about the
self through mutual sharing, of learning to orient and attune to others
through feelings, of complex cognitive and affective operations which
continue to be practiced and developed in later relationships with other
significant people. The sharing process fosters experiences of mutual
understanding and connection, which form the framework for increasing
relational development for the daughter and support her expectation that
self-growth is facilitated through psychological connection. Mothers
also experience increased connection and enhancement of their own
self-awareness through this process of mutual empathy, often reporting
deepening self-understanding in their ongoing experience of relating to
a growing child: learning in tandem about themselves and their daughters
from infancy onward (Surrey, 1985).

continued identification with her mother enables the girl to
develop increasingly sophisticated empathic and relational skills which
she practices in the mother-daughter relationship, as both she and
mother are motivated to care for and take care of the relationship
between them. Thus mothers help daughters to experience validation of
their own developing empathic competence by allowing themlto feel

successful at understanding and giving support at whatever level is

10
appropriate for their development. The mother-daughter relationship is
mutually empowering of both mother and daughter's abilities to perceive,
respond, and relate to the needs and feelings of others; relational
competence becomes an important source of self-esteem for women. Thus
the process of relational development is one of mutual empathy,
sensitivity, and responsibility, resulting in mutual empowerment and
self-knowledge: a two—way interactional model where it is as important
to understand the other as it is to be understood.

Surrey (1985) describes the pathway of relational development by
two constructs: "relationship—differentiation" (as a contrast to
separationrindividuation) and "relationshipvauthenticity.' Here
”differentiation" is not a developmental goal of encouraging
separateness through the assertion of difference, but is used to
describe a process, similar to embryological development, of "increasing
levels of complexity, choice, fluidity, and articulation within the
context of human relationship" (Surrey, 1985; p. 8.). "Relationship-
authenticity” refers to the ongoing challenge of remaining genuinely
connected and emotionally "real" in changing relationships, and
necessitates risk, conflict, expression of a full range of affect
(including anger and other negative affects), and willingness to alter
patterns of relating to adapt to growth in self and others. The
characteristics of the persons in relationship are acknowledged in their
impact on the development of the relationship, as is the power of the
relationship to define the continued growth of its participants. The
goal in development is ”toward more relatedness, not less; toward
better relatedness, not separation" (Surrey, 1985, p.9), with better

relatedness insuring greater flexibility, range, and choice for the

11
individuals and the relationship itself. Thus relationships and

identity develop in synchrony .

The Relational Self in Adolescence

 

For the adolescent girl , development continues through more
articulated and expanded relational experiences. The goal is not
necessarily separation from her parents/family, but changing and
adapting those important relationships to reflect, acknowledge, and
affirm her own developmental changes and expanded relational needs.
Instead of viewing conflict between the adolescent daughter and her
parents as the key dynamic of separation and individuation (Blos, 1980) ,
conflict is seen as one mode of intense and continued engagement whereby
differences are confronted within relationship. The ability to engage
others in conflict without disrupting underlying relational ties is a
valuable skill for later adult relationships, and serves the adolescent
girl '3 desire to preserve an ongoing and evolving relationship with her
parents, particularly with her mother (Kaplan, 1985; Gleason, 1985).

" The expanding world of the adolescent girl opens up new
possibilities for identifications, new self-images in relationships, and
growth through relationships beyond the primary childhood bonds. The
importance of relationships and relational competence remains central to
her self-development, however, and a new concept and term
”response/ability” is suggested (alrrey, 1985; p. 10) as more
descriptive of this line of development and form of action and
empowerment than ”agency” or "autonomy" .

Gender differences in adolescent identity development support these

theoretical distinctions. For adolescent males, identity develOpment is

12
focused on gaining autonomy, assertiveness, and independence (Dusek &
Flaherty, 1981), with issues of occupation, activity, and achievement of
major concern (Douvan & Adelson, 1966; Jesselson, Greenberger &
McConochie, 1977a). YOung men tend to define themselves in terms of
their competence (Hodgson & Fischer, 1979), and relationships*with
others are strongly influenced by issues of separateness and autonomy.
Friendships are activityhoriented (Jehnson and Aries, 1983), serve
shared goals of achievement and heterosexual ego-building (Thorbecke &
Grostevant, 1982), and are based on expectations of mutual aid (Kon &
Losenkov, 1978).

For adolescent girls, affiliation is a primary concern (Douvan &
Adelson, 1966), with females defining themselves in terms of who they
are in relation to others (Hodgson & Fischer, 1979). Friendships are of
more salience for young women than young men (Douvan & Adelson, 1966),
and are built on expectations of understanding and emotional support
(Kon.& Iosenkov, 1978). Girls use their friendships to provide
self-differentiating experiences through exploring and clarifying their
identitites in relation to each other, for more articulated
representations of themselves (Jesselson, Greenberger, & McConochie,
1977b). Interpersonal relationships and skills may often be more

important sources of self-esteem and personal achievement than career.t¥L'

goals. The fusion of identity with attachment noted in early
development (Chodorow, 1974) continues here as a fusion of identity with
intimacy, where a young woman's sense of self continues to be developed,
defined, and sustained through making and maintaining relationships, and
her personal competency judged by standards of responsibility and care

for others.

vi.»
~.

\'l‘:

13

Gilligan's Model

 

Gilligan's work (1980, 1982) on moral development in adolescence
and adulthood links differences in the way men and women perceive,
interpret, and respond to situations of moral conflict and choice, to
underlying views of the self and its relation to others. Both images
originate in the early parentrchild relationship and express central
truths about human experience. The image of the self as separate, in a
hierarchy of relationships of power governed and controlled by systems
of logic and law, is derived from the inequality of the parent—child
relationship, and gives rise to an ethic of justice. This ethic insures
that self and others will be treated fairly, on the basis of equal
worth, despite differences in power. The image of the self as
connected, in a web or network of relationships of interdependence
activated and sustained by activities of care and communication, echoes
the emotional connectedness of parent and child, giving rise to an ethic
of care. This ethic insures that no one will be hurt or left out, that
everyone will be responded to on their own terms.

Gilligan generally sees these two orientations as related to
gender, with the separate/objective self and justice orientation
describing the developmental paths of boys and men, and the connected
self and care orientation more descriptive of women's concerns and
development. .Lyons (1983) tested Gilligan's hypotheses, and found that
in constructing, resolving, and evaluating real-life moral conflict,
individuals empdoy both justice and care considerations but use one mode
predominantly. Choice of mode is related but not confined to gender,
‘with men using justice/rights predominantly and women using

care/response.

14
In terms of self-concept, Lyons also found that men and women
define themselves in relation to others with equal frequency, but
characterize this relationship differently. Men more frequently use
characterizations of a separate/objective self in self-definitions, and

womn more frequently use characterizations of a connected self.»

¢->

 

Finally, in her sample, regardless of sex, individuals characterizing
themselves predominantly in connected terms more frequently used
considerations of care in moral conflicts. . Individuals characterizing
themselves in separate/objective terms more frequently used
considerations of justice, indicating an important linkage between nodes
of self-definition/ identity and modes of moral choice.

By articulating another moral orientation, Gilligan expands
understandings of morality and moral development which have previously
focused on the use of fairness, human rights, and moral principles to
resolve moral problems of conflicting claims of individuals and groups.
Moral choices in this earlier model have been evaluated in terms of he»
they were made and whetler values or principles were maintained
(Kolhberg, 1981) . The often noted tendency of women to infuse their
moral judgements with interpersonal and affective concerns has usually
been seen as inadequate moral development or a failure of impartial or
objective morality in this earlier model.

In Gilligan's model, the intermingling of moral judgeients with
interpersonal concerns characteristic of women becomes an equally moral
and responsive perspective . Here conflicting claims denote fractures of
relationship or failures of response, to be resolved through activities
of care and communication. The goals of moral action are to maintain

connection, promote human welfare, and prevent harm. Moral actions and

15
decisions made on this basis are evaluated in terms of their
consequences for individuals and for relationships, whether connection
was maintained, restored, enhanced. The values and visions of women's
experience find a positive and powerful voice.

Gilligan and her associates (Gilligan, 1982; Iangsdale and
Gilligan, 1980; Lyons, 1983) have begun to outline developmental
patterns of an orientation of care in women. Because identity and
intimacy are interconnected for women, the development of this
orientation also involves changes in women's understandings of
themselves. Also, women's construction of moral problems as conflicting
responsibilities, rather than competing rights, ties development of
their moral thinking to changes in their understanding of responsibility
and relationships, to the development of an increasingly complex
psychological and contextual logic of relationships, rather than to the
formal and abstract logic of the justice approach.

The sequence Gilligan describes emerged from an analysis of women's
moral language, their thinking, and the ways they reflected on their-
thought in situations of crisis and life transitions. In this sequence,
the initial focus is self-care for survival. Tl'e self is the object of er ._
concern because relationships are disappointing, and the women
experience themselves as alone. A transitional phase follows in which, ,.

' 5

this focus is judged ”selfish", due to a developing understanding of the ‘
'\.
connection between self and other which is captured in the concept of

responsibility. Growth is marked as a move from selfishness to
responsibility, from isolation to social participation, from an external /
morality of imposed social sanctions to a morality of shared norms and ;

expectations. This transition signals an enhancement in self-north, and

16
depends upon a self-concept which can acknowledge one's potential for
being good, and hence worthy of social inclusion. . __q\

The traditional conventions of femininity dictate the morality that
characterizes the second perspective: a ”maternal" morality equating
goodness with caring for others, and care with self-sacrifice. The
moral strength of this position is intimately related to the
vulnerability created by the opposition of responsiveness to self and
others. Assuming responsibility for care, attending to others' voices
and taking into account other points of view, sensitivity to the needs
of others, and a reluctance to judge are positive strengths, which can
lead, paradoxically, to diffusion and confusion of judgement when one's
gag voice is lost. When no option exists to satisfy all interests at
stake, the feminine identification of goodness with self-sacrifice
clearly defines the primacy of other-care above self-care. The ideal of
selflessness and the obligation to care for others are pitted against
the "selfishness” of attending to oneself. And when care is defined as
not hurting, assertion and initiative become potentially immoral in
treir power to hurt.

The next transition is prompted by a conflict between compassion
and autonomy, between feminine virtue and adult power, as it leads to
self-deceptive evasion of awareness of choice and responsibility for
one's choices, and confusion about the extent of control and
responsibility for others. The exclusion of the truth and reality of
women's own agency and reeds, of herself as a recipient of care, creates
problems in relationships. Being for others presupposes a passive

innocence of one's own power and needs that is constantly at risk for

confrontation with the activity of care as it encounters choice, moral

17
decisions. A confused notion of responsibility protects feminine virtue
and innocence: she is assigned responsibility for the actions of
others, while others become responsible for the choices sl'e makes. Her
assertion becomes disguised as response. Because the morality of care
is embedded in the psychology of dependence, control and responsibility
for others are exaggerated, and control and responsibility for self are
minimized, leaving women vulnerable to excessive guilt and powerless
resentment. This resentment may be particularly strong when
self-sacrifice is not returned or rewarded in ways that match
expectations, conscious or otherwise. When relationships are secured by
masking one's own desire and avoiding self-responsibility and conflict,
the resulting confusion about locus of responsibility creates a problem
of truth that leads to reconsideration of tie meaning of care and the
relationship between self and other.

This second transition is marked by a reappearance of the issue of
"selfishness" as woren question whether it is selfish or responsible,
moral or immoral, to include their own needs as worthy of concern and
care, being responsible to themselves as well as to others. The
criterion for judgement shifts from external approval/goodness, to
personal integrity/inner judgements of truth. Being responsible to and
for oneself means acknowledging the intentions, reality, and
consequences of one's actions, verifying a capacity for inner judgement
and the legitimacy of one's own point of view. It may also mean that
the needs of the self must be uncovered and explored.

The tension between selfishness and responsibility is resolved
through the discovery that responsiveness to self and others are
connected, rather than opposed, through the injunction of

18
nonviolence/not hurting. This is elevated to a principle governing all
moral judgement and action, including the conventions of feminine
self-abnegation and sacrifice. A new'moral equality between self and
other now seeks to balance claims of self and others, transferm the
definition of care, deepen the understanding of relationships, and
integrate care with justice. care becomes the self-chosen principle of
a judgement that is increasingly aware of both the differentiation of
self and other, and the interdependence of self and others in the
psychology of human relationships. .

Transition to the third perspective of self and other care chosen
freely requires a self-concept that can support and defend the worth of
the self in relation to others. Acknowledging selfdworth, claiming the
power to choose, and accepting the responsibility for choice is related
to recognizing anger and frustration as the psychological costs of the
indirect action prescribed by traditional femininity. When the
acquisition of adult power through choice and responsibility no longer
means the loss of feminine sensitivity and compassion, the restructuring
of a morality of care assumes that power, and considers it in the

context of interdependence and connection.

The Revised Relationship Self Inventory

 

Each of Gilligan's three perspectives represents a more complex
understanding of the relationship between self and other, and each
transition involves a critical reinterpretation of the conflict between
responsibility and selfishness. Initial concern with survival gives way
to a fbcus on goodness, and finally to a reflective understanding of

care in the resolution of conflicts in relationships. Gilligan's theory

19
was developed from her study of college student women making career
decisions and facing other life transitions during their college years
and in the first five years following graduation. She also studied the
role of conflict in development through interviews with women deciding
whether or not to have an abortion. Her theory ties developmental
changes in the understanding of care to crises and major life
transitions. The development of a measure of her model of development
indicates that this model is appropriate for the study of self
development in different populations of women, and can be assessed in
survey format, allowing issues such as the importance of life
transitions and the nature of deve10pmental changes to be researched
more easily (Reinhart, Pearson, Kamptner, Cornwell, Barnes, Stromen, &
Donelson, 1985; Pearson, Reinhart, Donelson, Stromen, & Barnes, 1985;
Stromen, Reinhart, Pearson, Barnes, Blank, Cebollero, Cornwell,
Donelson, & Kamptner, 1987).

The original Relationship Self Inventory (RSI) was developed and
examined in a sample of 526 women aged 21 to 85, and consists of 27
items in four internally reliable (at all ages) scales measuring the
Connected Self and the three perspectives of Self Care from Need (SCN) ,
Primacy of Other Care (POC), and Self and Other Care Chosen Freely
(SOCCF) (Reinhart et al, 1985) . Iters tapping the Separate/Objective
Self did not measure consistently any psychological construct meaningful
to this sample and were not included in the R51.

Ninety percent of the sample folnd the Connected Self scale
self-descriptive, with sixty percent indicating their care orientation
as described by the Primacy of Other (POC) scale. The FCC scale was

self-descriptive of more women in the two older age groups (42-85) than

20
it was of the women in the youngest group (21-41) . There were no
differences between homemakers and other women's ratings on the FCC
scale, but hoiemakers (48% of the sample) were less likely to be
described by the Self Care from Need (SCN) and the Self and Other Care
Chosen Freely (SOCCF) scales than other respondents.

In another study (Pearson et al., 1985) , the responses of married,
separated, divorced, single, and widowed women on the RSI were examined,
using marital status as an indicator of a past life transition (e.g.,
married to separated, separated to divorced), to explore the validity of
Gilligan's proposed developmental sequence. As expected, the marital
status groups did not differ in mean scores on the Connected Self scale.
Divorced and separated women scored higher than married women on the SCN
and SOCCF scales as predicted. However, no differences among married
women, and divorced or separated women were found on the FCC. Although
many of the patterns of care support Gilligan's theory of the centrality
of the ”comected self" and the association of life transitions with
differences in patterns of care of self and others, the data suggest
that women perceive themselves to be in more than one focus area at the
same time. This implies that the developmental sequence is not a stage
sequence. It is also not clear whether a life transition causes a
change in perception of care and/or such a change precipitates a life
transition.

TheRSI has sincebeenrevisedonasampleof 930worenand228men
(Stromen et al., 1987) . A measure of the Separate/Gajective Self was
added, to form four internally reliable and consistent scales for both
men and women: 1) Separate/Oajective Self (S/O), 2) Relational/Connected

Self (CS), 3) Primacy of Other Care (POC) , and 4) Self and Other Care

21
Chosen Freely (SOCCF) . The Self Care from Need (SCN) , which formed a
reliable scale originally, failed to do so in the revision and was
dropped from the RRSI. In the revised version, subjects' responses to
SCN items tended to fall into the Separate/Objective scale, which did
not appear in the first version, and into SOCCF. Ebrpected patterns of
interrelationship support the distinctions between Separate/abjective
and Relational/Connected selves and among meanings of care within
Relational/Connected Self (see Table c2 in Appendix C) . The data also
support Gilligan's view tint relational/connected and separate/objective
selves are different but not opposite dimensions, with moderate
interrelationships .

My research used the RRSI to identify woren who described
themselves as relational/connected selves, and then to further identify
subgroups for comparison: those women who described themselves in terms
of Primacy of Other Care (POC) were compared to women describing
themselves in terms of Self and Other Care Chosen Freely (SOICF) on

dimensions of depression, self-esteem, anger, and power.

Depression

 

Depression has been well-documented as a substantial, pervasive,
and serious problem for wolen: 20 to 30% of all men are estimated to
suffer from depression, often moderately severe, at sole point in their
lives (Weissman & Klerman, 1979) . Sex differences in depression are a
consistent finding across institutional settings, where twice as many
women as men undergo depressive episodes, and three times as many woven
as men in the age group 25-44 seek help for depression through

outpatient services (Carmen, Russo, 8 Miller, 1984). Surveys of

22
nonpatient populations produce similar statistics. Other research
indicates that these results are not artifacts reflecting gender
differences in help-seeking behavior, willingness to acknowledge
psychological symptoms, or diagnostic labeling bias. Reported
differences thus represent real gender differences in depression, for
which current understandings of biological, endocrinological, and
genetic contributions provide insufficient explanation (Carmen, Russo, &
Miller, 1984).

Recent interest in the impact of life stress for depression reveals
no consistent sex differences in reports of stress on the life stress
scales usually employed. However, a critical review of this research
suggests methodological biases of the scales themselves which make them

less relevant for women's lives, and in doing so, illuminates concerns

and dynamics which are central in understanding women's vulnerability to
depression (Makosky, 1980).

Most stress scales emphasize acute changes in life conditions
rather than chronic conditions common to women (poverty, large family
size, economic discrimination, health problems), exclude some areas of
direct stress for women (abortion, sexual abuse, sex discrimination),
and weigh life events according to the degree of distress experienced by
the average person, obscuring felt differences in events with
significant long-term‘consequences for women (marriage, childbirth,
separation, divorce). Other research supports a strong relationship
between female depression and other ferms of mental illness, and the
stress and powerlessness associated with the social, psychological, and
economic realities of women's disadvantaged status. Finally, life

stress scales focus on personal stressors, ignoring data that suggests

23
women view stresses in terms of significant others as well as
themselves. ‘WOmen report more recent stressful events than men do when
asked to list events happening to them and to important others
(Dohrenwend, 1976). Their psychological symptom scores are determined
more by events they cannot control (i.e., things that happen to others),
a pattern not true of men and indicative of women's greater sense of
responsibility for the well-being of others (Dohrenwend, 1973).

These last findings suggest the relevancy of theories which
acknowledge the role and significance of interpersonal concerns in
women's development and identity for understanding gender differences in
depression. Self-inerelation theory provides a meaningful connection
between certain aspects of women's psychological development and key
dynamics of depression described by several cognitive and affective
accounts of the etiology of depression. Issues of self-esteem, anger,
and power are involved in the interplay between central features of
relational identity and depression, and will be briefly discussed here,
'with fuller elaboration to follow.

Kaplan (1986) argues that central dynamics of depression are
essentially distorted aberrations of aspects of women's normative
development in western society. She uses selfeinrrelation theory to
illustrate how women's felt responsibility fer relationships can lead to

vulnerability to loss, inhibition.of action and assertion, inhibition of
anger, and low self-esteem» when connection is thwarted, threatened, or

devalued. Other related approaches to the study of depression explore
predisposing personality factors (Salzman, 1975; Chodoff, 1974) in
individual personality structures; Kaplan examines features of

personality structure common to women as a group, linking them to

24
existing accounts and descriptions of depression, refining these
accounts in the process.

The experience of emotional loss and separation through death,
abandonment, or emotional disconnection is a conron feature of
psychoanalytic, object relations, and social accounts of depression.
Relationship loss of mother before age 11 figures prominently in
childhood experiences of women who become depressed as compared to women
who do not (Brown & Harris, 1978), and relational loss appears to be a
colmon precipitant for women seeking therapy for depression (Weissman &
Klerman, 1979; Schwartz & JUroff, 1979). Conversely, a close
relationship with a confidante provides wolen a buffer against becoming
depressed under stressful life circumstances (Belle, 1982) . Intimacy
with a spouse serves a similar protective function (Brown & Harris,
1978), and depressed women seek help from others to counter depressive
states more frequently than depressed men do (Padesky & Hammen, 1981).

In traditional theories, loss of the other as a source of
gratification of oral needs results in lowered self-esteem (narcissistic
injury), or is perceived as rejection/abandonment with disrupted
attachment bonds. Self-inrrelation as the core self-structure of women
emphasizes the role of connection to others in the development of
identity and self-esteem. According to this model, loss of mother and
others is experienced not only as unidirectional object loss, but as .
disconfirmation of relational self-structure and as a lost opportunity
for actively nurturing the relational process as a mutually empathic and
empowering process.

Here, loss of intimacy is experienced as a failure of the self as

well as loss of love. Fears of separation and abandonment are

25
heightened when responsibility for maintaining relationships is a
central component of one's core identity. Grief, guilt, shame, and
feelings of inadequacy and worthlessness are the intermingled results of
experienced loss for the relational self. Even without actual loss,
experiences of disappointment at lack of mutuality of understanding with
others can be felt as loss and self-diminishment for a person whose
self-concept includes being skilled at facilitating reciprocity and
affective connection in relationships.

For persons with self-concepts organized around performance or
personal achievement, experiences of loss other than relationship loss
(e.g. loss of jcb, retirement, failing exams, losing a competition,
etc.) may be equally devastating in their challenge to identity, role,
values and existential purpose, ego ideal (Bibring, 1953). To the
extent.that.women's experience of self is organized around being able to
make and maintain affiliations, the threat of disruption of these ties
is perceived not just as loss of a relationship, but as a total loss of
self (Miller, 1976). Assuming responsibility for relational failure may
stimulate further attempts at connection or repair, action which
supports others, but the experience of self-doubt, of disconfirmation,
is likely to result in the inhibition of other forms of action,
specifically actions which further one's own goals. The deep fear of
relational fracture and the concomitant threat to the integrity and
authenticity of the relational self leads to constriction of a range of
activities and modes of expression to preserve relational ties,
particularly when self-assertion and personal strivings are seen as

hurting or depriving others (Kaplan, 1986).

26

Inhibition of action or assertiveness is a characteristic sign of
depression noted in psychodynamic (Bibring, 1953), cognitive (Beck,
1972), and behavioral (Seligman, 1975) accounts of depression. This
inhibition is described as a powerless ego state, as the result of a
triad of negative cognitions about one's self, the world, and one's
future, and as a consequence of "learned helplessness", respectively.
Kaplan (1986) suggests that helplessness for women is consistent with
Seligman's view of behavioral inhibition as a result of loss of control
of reinforcement modes for behavior, but the woman is additionally
burdened by self-blame and responsibility, due to the impact of such
loss of control on core self-structure. These attributions are also
consonant with cognitive accounts linking such attributions to
depression. She also differentiates degrees of helplessness, depending
on whether action is viewed as selfish or destructive in a relational
context, citing Miller's (1976) point that depressed women can be very
active if their actions are perceived as occurring in the "proper
context” of being facilitative of others.

The inhibition of anger and aggression is a central theme of early
psychoanalytic accounts of depression, and has been hypothesized as the
result of loss of an ambivalentlybloved object. Angry feelings towards
the other are turned on the self through introjection, to preserve
loving feelings for the other. This intrapsychic process can be
confirmed interpersonal ly for women, when others view their expressions
of anger as powerfully destructive, as evidenced in strong social
sex-role proscriptions against anger for women (Ierner, 1977;
Bernardez-Bonesatti, 1978). The internalized view of anger as bad and

destructive of relationShips leads to its inhibition, which is in itself

27
disempowering. Poverlessness may generate more anger, which is viewed
as further confirmation of badness. The struggle to contain anger can
also contribute to inhibited action. Women's greater socialization for
expressions of distress ratler than anger allows the transformation of
anger to distress, which protects others and relationships at the cost
of women's clarity about their own affective states.

Low self-esteem is a major corponent of most major theories of
depression, whether it is seen as a contributing factor or product of
depression in exaggerated feelings of worthlessness and inadequacy. When
women's self-worth is intimately tied to relational capacities, the felt
responsibility for failures of relationship creates self-doubt about
relational worth. This is intensified when self-worth is exclusively
tied to one role and other sources of self-esteem are restricted. And
when relational qualities are devalued or misinterpreted as dependency
(Miller, 1976; Stiver, 1984) in the larger social world, women may not
fully value these capacities as strengths, or receive much validation
for them as learned skills, adaptive values, and a source of power and
action in the world. The paradoxical dilemma of deriving self-worth
from actions which demand self-sacrifice (when care of others is opposed
to self-care) is inherent in women's normative role.

Self-in-relation theory demonstrates how extrere forms of
curtailment of women's normative developmental patterns can create the
intra-psychic conditions that characterize depression, rooting the
question of gender differences in depression in tie particular
vulnerability of women's developmental patterns. What then accounts for
those woren who are not seriously troubled by depression? Can theories

of the psychology of the connected self explain differences among women

28
related to their vulnerability to depression? Gilligan's (1982)
elaboration of different perspectives of care among relational women may
provide a way of answering these questions.

Gilligan's description of the second perspective or phase of
development, where care means caring for and not hurting others,
highlights the connection between women's relational strengths and
vulnerabilities noted earlier. Others are clearly the beneficiaries of
women's investment in relationships that empower and support human
development, and women also grow through and enjoy their relational
competence in these relationships. However, in a context where women
and their caring/communion are devalued, where their needs are denied
and unmet in nonmutual relationships, relational skills can become
intertwined with dependency and anger in ways that are clearly
problematic for women, and ultimately for relationships and an ethic of
care.

various theorists trace the roots of this confound to family
relationships, viewed in an historical context. Female personality has
been shaped by the historicallyhcreated nurturing imperative for women
to meet not only the needs of dependent children and elderly family
members, but the material and emotional needs of their partners as well,
an assumption of unilateral and unconditional nurturance (Chodorow,
1974; Miller, 1976; westkott, 1986). Women are encouraged to develop
strengths that are valid only when used for others, and are idealized
for denying their own compelling needs and desires.

This process produces undernurtured nurturers, subject to
underlying fear and anger at having to deny self, in being devalued.

Anger is turned against the self, and the appropriation of behaviors to

29

ingratiate oneself to important others forms a fundamental
characterological split, recently described by Alice Miller (1981, 1983,
1984) and others. The undernurtured nurturer feels herself to be
unworthy, but hopes for acceptance through meeting the needs of others.
These tensions then corrupt the very goodness towards others that the
nurturing imperative (or ethic of care in its second-phase
conceptualization) demands. Outcoles of this process may include
expressions of martyrdom when one's giving is not fully appreciated
(Horney , 1950) , experiences of vague depression accompanying
self-sacrifice in close relationships (Horney, 1942) , and underlying
resentment towards those for whom responsibility is felt (Horney, 1937) .
Woren may also nurture others in compensation for their om feelings of
worthlessness, in an inauthentic and "powerless responsibility” (Rich,
1976) , if they are not able to nurture themselves without fear or guilt.

This intertwining of dependency and care is noted by Gilligan as a
problem for women in her second stage, causing confusion about
responsibility and locus of control. When responsibility for
relationships becores too inclusive, when women feel responsible for the
actions and feelings of others, which are beyond their control, their
vulnerability to loss and rejection intensifies, and hence their
vulnerability to depression. When relationships are preserved at the
cost of self-development, feelings of personal helplessness and low
self-esteem are heightened. The conflictual opposition of self-care
with caring for others, and the equation of care with not hurting
others , creates a potential vulnerabi lity to depression . This research
tested the hypotheses that depression would be positively associated
with Primacy of Other Care (POC) and negatively associated with Self and

30
Other Care Chosen Freely (SOCCF) . Also, women who characterized
themselves by POC would be more likely to experience depression than

women who endorsed SOCCF as descriptive of their relational orientation.

Self-Esteem

 

Self-esteem, variously defined and measured, occurs frequently as a
variable in studies of development and adjustment for many groups,
including women. As discussed and explored here, self-esteem is defined
as a component of self-concept, as the privately experienced dimension
of self-assessment. Self-esteem is a positive or negative attitude or
evaluation of the self, with high self-esteem expressing the feeling
that one is "good enough," a view of oneself as a person of worth, a
self-respecting rather than self-aggrandizing position (Rosenberg,
1965). Neither deficiencies nor strengths are exaggerated or ignored,
and self-acceptance anticipates future growth and self-improvement. Low
self-esteem implies a negative view of the self as not worthy of esteem,
an expression of self-dissatisfaction or more extreme self-rejection and
selfecontempt.

The view that females have lower self-esteem than males was one of
the societal and professional beliefs about sex differences reported by
Maccdby and Jacklin (1974) as unfOunded. In reviewing literature
dealing primarily with preadolescents (1400 studies compared to 16 on
adolescent females), they report findings suggesting that the sexes are
similar in overall self-satisfaction and self-confidence, with
differences in the areas of functioning where greater self-confidence is
felt. These areas seem to differ according to sex roles: girls in

grades 6-12 have more positive views than boys about themselves with

31
respect to social confidence, while boys are higher than girls in
selfdviews about achievement and leadership (Monge, 1973). The school
environment seems to reward girls' performance skills, self-control, and
social skills, so that girls aged 9-13 are likely to have more favorable
views of themselves than do boys, and their self-esteem increases during
this period while that of boys decreases (Scares & Scares, 1969).

In high school, adolescents experience increased pressure fer
adherence to sex roles, and the content of their self-concepts diverges
towards increasing consistency with sex roles (carlson, 1965). The
interpersonal and communal focus of adolescent girls and the agentic
achievement concerns of adolescent boys noted earlier define the realms
and most important sources of self-esteem considered appropriate fOr
each sex. Although content of self-concept is different, favorability
of selfdview is equal (Carlson, 1965).

After high school and beyond, there is no clear pattern of
differences in self-esteem beyond the broad generalization that women as
a groupldo not have higher self-esteem than men as a group. When sex
differences occur, women in college or later years are more likely to
score lower than men than the reverse pattern. ‘Within each sex,
psychological masculinity itself is more strongly related to self-esteem
than is femininity (Spence, Helmreich & Stapp, 1975; Wetter, 1975) .
Environmental support for one's selfdview matters greatly: self-esteem
in one's role depends on the environment allowing behavioral expression
relevant to one's selfdview, and rewarding one's efforts in ways that
are personally meaningful (Donelson & Gullahorn, 1977). Thus a woman
whose self-esteem rests largely on her sense of personal competence in

creating and nurturing her family life can have high self-esteem if she

32
is free to act in that role and if her family supports and acknowledges
her efforts. Most variation in self-esteem is likely based on the
content of one's self-concept and ideals, with low self-regard
experienced when felt deficiencies concern central characteristics of
one's self-concept, in relation to the attributes rewarded, censured or
ignored in the environment (Donelson & Gullahorn, 1977).

Rosenberg (1965) found that, beginning in adolescence, the
selfdvalues of girls emphasize values of interpersonal harmony and
success (being likeable, easy to get along with, friendly, etc.), and
what he called "tender virtues" (1965, p. 254) of being kind and
considerate, sympathetic and understanding. Boys were more likely to
stress motoric values and physical courage, and value interpersonal
control or dominance in relationships with others (being good at getting
people to do what you want). Social class affected the selfdvalues of
boys but not of girls. These internalized standards and values about
how she ought to be influence a woman's evaluation of the person she is,
with interpersonal competence a central realm of achievement or
competency. In Gilligan's language, women judge themselves on their
ability to»care, to respond in relationShips, and on the quality of that
care. Selfeesteem‘would be related to relational women's general
“response/ability," and to»their specific understanding of what care
entails.

Differences in the meanings of care, in the selfdvalues of women
described by Gilligan's developmental perspectives, may be related to
variations in selfeesteem among women. ‘When the ideal or ethic of care
extends to others but does not include the self, one's caring for others

is the standard for personal evaluation and sense of accomplishment or

33
positive self-esteem. Cues from others becole important indicators of
the quality or efficacy of that care; if being for others becoles being
good enough for others, then self-esteem becores increasingly vulnerable
to external rather than internal evaluation. The view of the self about
the self may be lost if sensitivity to others' voices and views is not
balanced by attention to one's own voice.

The ideal of self-sacrifice as it infuses a maternal morality can
also lead a self-caring woman to judge herself as bad, inadequate, less
moral, etc., for that very concern with self. If capacities for
responding to others are developed at the expense of other areas of
self-development and self-expression, other sources of self-esteem are
diminished. When women are encouraged to develop self-definition
through others for whom they care, the distinction between one's own and
others' needs blurs. Needs and desires denied their own legitimate
expression may find vicarious expression through overidentification with
others. Self-esteem becomes tied to others' status, failures, and
successes, which is ultimately unsatisfying and problematic for
relationships with others. And finally, if the equation of care as not
hurting others is not subject to inner judgements of integrity, caring
for others can create self-alienation, when meeting others' expectations
and pleasing others creates distance or conflict with other values,
reeds, and desires of the self.

The inclusion of self as a recipient of care is both a result of
self-esteem (viewing the self as worthy of one's care) and reinforces
it. Caring for others and oneself creates the standard against which
care is evaluated, with self-care viewed as a responsibility to oneself.

Gilligan describes the criterion for judgement shifting from external

34
approval / goodness to personal integrity/ inner judgements of truth, in
the process of being responsible to and for oneself. Others'
expectations or requests for care are evaluated according to judgements
of truth and norexploitation which also prohibit self-exploitation.

When self-development is not contrasted with empowering others in
their development, women are more free to develop interests and skills
in other areas as well as relational competence, and to enjoy
self-esteem from a broader variety of sources. When participation in
other roles and goals is not seen as threatening one's important
connections to others, self-development is less ambivalent and
conflictual than when it is viewed as selfish or in moral opposition to
caring for others. Since society does not fully recognize or value the
competencies involved in the feminine role, and holds a narrow agentic
view of achievement, relational strengths and skills do not receive the
social support or reward that other skills receive. Until such views
change, self-esteem which draws on multiple roles and sources will be
enhanced.

This research tested the hypothesis that women who characterized
ttemselves by Primacy of Other Care (POC) would have lower self-esteem
than women who endorsed Self and Other Care Chosen Freely (SOIIZF) .
Self-esteem was also expected to be negatively associated with POC and

positively associated with SOCCF.

ESE
The psychology of women and anger has received particular attention

as the feminist movement raised awareness of the personal, social,

political, and economic oppression of women, and challenged role

35

expectations that denied women legitimate response to this oppression.
Assertiveness training and the use of social support and validation in
consciousness-raising groups were interventions with the goal of helping
women have and use their anger more effectively to change their lives.
Other accounts of the intrapyschic and interpersonal origins of problems
with anger for women attempt to untangle internal ambivalence that
undermines the usefulness of modeling, practice, and support, exploring
the relationship of anger to women's core self-concepts or identity.

Many theorists have noted the strong cultural prohibitions against
the expression of anger by women, and relate these taboos to the
unconscious beliefs of men and women in the omnipotence and destructive
power of female anger (Bernardez-Bonesatti, 1978; Lerner, 1974;
Dinnerstein, 1976; Lederer, 1968) . This belief has its roots in the
early experience of the infant with his/her mother, and persists
unconsciously in irrational fears of female aggression. The feminine
sex—role stereotype of the al l-caring, nonaggressive woman is an attempt
to ensure the binding of this fantasied destructive power through its
denial. Stereotypes of the angry woman are particularly vengeful in the
negative characterization of feminine anger . The expression of anger in
the defense of helpless or dependent others is permitted as a response
of protective care and nurturance , but the expression of anger in the
cause of women themselves is discouraged and shamed as selfish,
unfeminine, demanding, irrational, or extreme.

when identity for women is defined solely by caregiving functions,
and care is defined absolutely as not hurting, anger poses a threat to
core identity, femininity, and self-esteem. The opposition of anger

with caring remains unchallenged by the experience that anger may

36
reflect self-care and self-respect. It can also be a form of caring for
and respecting others, an expression of intimacy, rather than
dissolution or destruction of engagement.

Another way in which anger challenges relational identity and
meanings of care is through the phenomenology of anger itself. The
experience of anger has been described as involving feeling separate,
different, and alone in the assertion of differences with another person
(Bernardez-Bonesatti, 1978). The temporary loss of felt connection, and
'with it a sense of usefulness and self-value as a self-in-relation, can
trigger separation anxiety and an unconscious fear of object loss,
producing maxed expressions of anger accompanied by tears, guilt, and
sadness. The mixed nature of this communication may contaminate or
defuse the interpersonal impact of the anger.

The shift from anger to tears is a transformation of anger into
expressions of distress, sadness, or depression, which are socially more
acceptable than the direct expressions of anger by'a woman. It is also
intended to nullify the anger which distances and seems to threaten her
sense of connection; the expression of hurt and distress also elicits
concern and restores connection by emphasizing the value of the other to
the self (Ierner, 1980).

If anger is experienced as incompatible with core identity,
destructive of relationships, and in opposition to caring, it may be
denied or repressed, inhibited through the use of withdrawal, and/or
viewed negatively. It.may be expressed indirectly through somatization/
symptom formation, and tranformed to sadness/diStress/hurt/depression.

waever, if anger is viewed as a legitimate expression of

self-care, when selfecare itself is legitimized, anger can become a

37
signal to the self that something is wrong. As physical pain protects
the body, the pain of anger preserves the integrity of the self (Ierner,
1985) . Anger is the "voice of self-respect," as it is used in the
expression of protest over the violation of personal rights and needs
(Bernardez-Bonesatti, 1978) . The capacity to be aware of the intensity,
source, and purpose of one's own anger has been linked to self-esteem,
personal well-being, goal-directed activity, and creativity
(Bernardez-Bonesatti, 1978) . Anger may be viewed positively rather than
negatively, approached rationally for understanding of its signal
function and meaning, and acknowledged in communication with others.
Resolution can be more genuinely attempted if anger has been fully
acknowledged and not denied or ignored .

This research explored care orientation/ views of self-and-
other-care in relation to anger and ways of responding to one's own
anger. Women who described their care orientation on the POC scale
might be angry as often as women who described themselves using the
SOCCF scale, or possibly experienced more anger due to lack of
self-nurturance and unrewarded self-sacrifice, but were not expected to
admit to such anger on a measure of frequency and intensity of anger
experience. While it might be hypothesized that POC women would have
higher scores on an anger scale than SOCCF women, the conflict of
acknowledging anger for the self-image of POC women was likely to
influence anger scale scores, such that no differences would occur
between the two groups in terms of their responses on a measure of
anger.

Differences in the ways anger was viewed and expressed should

characterize the two groups of women, however. In responding to an

38
open-ended sentence completion stem ('When I get angry I___") , POC
women were expected to characterize themselves using categories of
denial of anger, withdrawal from anger/and/or substitute activity,
negative judgements about anger or the angry self, sometization and
transformation of anger to distress. Both POC and SOCCF women were
expected to be concerned with a focus on responsibility for hurting
others, on the impact of anger on relationships. SOCCF women might
characterize themselves as using rational means of problem-solving about
anger, holding positive views of anger, and communicating their feelings

to others, towards resolution or reconciliation.

Power

 

The ideas and model presented here are drawn largely from personal
and professional experience as a woman working with other women,
listening to common struggles and concern with power, and attempting to
develop a model of power that addresses women's experiences. As a
clinician involved with individuals and small interpersonal groups, my
orientation is towards a concept and model of power framed in the
language of individual personality and develogrent, rather than defined
in social-political context and terms as influence, status, etc.

Traditional discussions of personal power defined and experienced
as adequacy, ambition, competition, power over others, and agentic
achievement and success describe central concerns of the
separate/objective self, of men's experiences of power. However, even
as competent women in positions of institutional power and control
struggle with similar issues of competency, competition, and success,

they do so with different images and experiences of identity and

39
relationship. These images and experiences shape the particular form
and meaning "power" then takes in their lives. For a model of power to
be accurate and helpful, it must address felt experiences of power in
relation to core images of identity. For women, this means a model of
power which is consonant with the values and strengths of a relational
orientation, which explores the experienced effect of one's power on,
through, and in relationships, in terms of interdependence and
connection. A model of power for the relational self is needed to
clarify the particular definitions and perceptions of power, sources of
power, conflicts with power, and possible interventions towards
increased power, which reflect women's experiences. A basic framework
for understanding power has emerged from listening to women talk and
deal with issues of power in their lives.

When asked what power means to them, women often disclose
fragmented images of power that echo dominant cultural images of power
which are problematic for women. Power is associated with masculinity,
‘with efficacy and influence, with manipulation of others, as power over
others which can become abusive or hurtful of others: power is
confounded with violence. Wbmen are thus ambivalently attracted to
power, wanting to feel powerful and not powerless, but uncertain about
the relationship of self and others in power.

Although women can discuss power in relationship to role or status,
as authority and responsibility, in hierardhical relationships,
particularly'when they are fUnctioning in professional settings and
positions, they often express quite different concerns when asked about
their conflicts with power. When interviewed in a pOpular magazine for

professional women, women in high-ranking positions in business,

40
government, education, etc., identified elements of power in similar
terms as their male counterparts, but then revealed more personal
concerns with power off-tape, at the end of the interview (Stautberg,
1985) . They talked in terms of guilt, separation from others,
aloneness, fear of becoming arrogant, invulnerable or distant,
conflicting responsibilities, etc.

In a therapy group, competent graduate student woven reported
feeling powerless and guilty when saying “no” to others, setting limits
on responsibility for others, disappointing others, coping with
separation and differentiation, dealing with the stigma of selfishness
in self-care and self-development, asking for and taking help from
others, fearing confrontation with others and also exploitation by
others, etc. Gilligan and Lyon's language of connection, of
interdependence and network , of care and response surface in women ' 5
discussions of power. Here again women's relational identity and views
of care contain potential experiences/sources of power and challenges to
that power.

Those aspects of women's power most comfortable for women and men
often involve capacities for empowering others, relational strengths
used in the service, care, and development of others (Miller, 1976). To
feel powerful through elpowering others does not conflict with
relational identity, although it is an experience of power devalued in
the larger social world. This devaluation can lead to woven not
recognizing interpersonal skills as skills, and ignoring the power of
their capacities for care. The challenge of power for women comes in
reclaiming relational strengths and values as a source of power against

dominant cultural views, and in using that power for themselves as well

41
as others. ‘Women struggle with claiming power for themselves as selves
and understanding that power in terms of changing self-images.

Wbmen's fears in confronting and integrating power with their sense
of self have been described in terms of troublesome equations of power
and selfishness, power and destructiveness, and power and abandonment
(Miller, 1982). When acting on one's own interests and motivation is
viewed as selfish, as not enhancing others, to be powerful challenges a
self-concept based in the opposition of self-and-other-care. If
self-determined action is morally wrong or evil, feeling powerful can be
experienced as the psychic equivalent of being destructively aggressive.
If being powerful means loss of interdependence, not needing others,
core identity is threatened, and women anticipate feeling alone,
abandoned, and isolated in their experience of power.

Gilligan describes some of these conflicts for women who define
themselves as caring for others and not self. She also describes the
power of choice in the reorganization of caring as a self-chosen
principle applied to self and others, in the discovery that
responsiveness to self and others are connected rather than opposed.

This shift in the meanings of care has implications fOr women
dealing with the challenge of power to their identity. It is difficult
fer women to own and use their own power UNTESS they resolve the
conflict of selfeother care, tolerate the reality and tension of
multiple and possibly conflicting claims on their caring, and
acknowledge responsibility for their own actions of choice, WITHIN the
web of relationships that sustains their identity.

This view of power relates it to central issues of depression,

self-esteem, anger, and guilt, as they are encountered in the process of

42

reclaiming self-care, and as they contribute to experiences of
power/powerlessness for relational women. It also contains the seeds
for interventions with women, to increase their sense of personal power.
Recognizing the limits of responsibility to others and the extension of
it to oneself deals directly writh questions of guilt and assertion.
Perceptions of choice and willingness to accept responsibility for
choices contribute to a sense of efficacy and agency for one's own life,
and to experiences of mutual care as genuine and freely given and
received.

This model was developed through clinical experience with women and
was examined empirically in an initial exploration of the ways a
nonclinical population of women described their experiences of power "in
their own voices," using an open-ended sentence format. The responses
of women to the sentence stem "When I feel powerful, it is
because__" were coded for sources of power which might be
differentially descriptive of POC women and SOCCF woven. POC women were
expected to deny feeling powerful and view power as inauthentic, as a
negative experience. POC and SOCCF women might both find power from
relationships with others. SOCCF women were expected to experience
power from self-esteem and power from perceiving choices about oneself,
one's life. Their experiences of power would not be limited only to .
sources easily reconciled with traditional feminine self-concept and

role.

Wof Model and Research Hypotheses

 

This research was focused on the use of the Revised Relationship

Self Inventory (RRSI) to identify men who were similar in their

43
descriptions of themselves as relational/connected selves, but who
differed in their understanding of the role of self and other in
activities of care, as reflected in their differential endorsements of
the Primacy of Other Care (POC) and Self and Other care Chosen Freely
(SOCCF) scales. These two scales were constructed to measure different
manifestations of the relational/connected self corresponding to
Gilligan's (1982) final two developmental perspectives on the meaning of
care for self and others. This research tested the value of these
differentiating perspectives in accounting for differences among women
in areas of interest for the psychology of women: depression, self-
esteem, anger, and power.

Describing expected relationships between relational identity style
and other variables would test and expand Gilligan's model of relational
identity development for women. Predicted relationships between the two
relational perspectives on self and other care measured by POC and
SOCCF, and the variables of depression, self-esteem, anger, and power

are summarized below:

1. Depression was expected to be positively associated with Primacy of
Other Care (POC) , and negatively associated with Self and Other Care
Chosen Freely (SOCCF).

2. Self-esteem was expected to be negatively associated with POC, and
positively associated with SOCCF.

3. Wbmen whose relational identity was described by high POC/low SOCCF
were expected to be more vulnerable to experiencing depression than
women whose relational identity was described by high SOCCF/low POC.

4.

6A.

GB.

44
High POC/ low SOCCF woren were also expected to have lower self-
esteem than high SOCCF] low POC women.
The two groups of women were not expected to differ on a measure of
the frequency and intensity of their experience of anger. It might
be hypothesized that high POC/ low SOCCF women would have higher
scores on an anger scale than high SOCCF/ low POC women due to lack
of self-nurturance and unrewarded self-sacrifice. However, the
conflict of high POC/ low SOCCF women about acknowledging anger was
expected to influence self-report data. Thus, no differences were
expected to occur between the two groups on a measure of frequency
and intensity of anger.

Information about differences in the ways anger was viewed and
expressed by these groups of women was sought using the open-ended
sentence completion "When I get angry I __". Codes were
developed to categorize responses elicited by the open-ended format,
and hypotheses linked particular categories to the relational
perspectives of POC and SOCCF.

The fol lowing types of responses to anger were expected to be more
characteristic of high POC/ low SOCCF women than high SOCCF/ low POC
women: denying anger, erotional and/or physical withdrawal
(including use of substitute activity to displace angry tension),
negative judgetents about anger or the angry self, physical
reactions or somatization, and the transformation of anger to
expressions of tearfulness, sadness, or depression.

The fol lowing types of responses were expected to be more
characteristic of high SOCCF/ low POC women than high POC/ low SOCCF

woren: rational means of problem-solving about anger (including

6C.

7A.

7C.

45
introspective means and talking to others not involved in the
conflict), positive views of anger or the angry self, and attempts
at communication of feelings towards resolution and reconciliation.
Both groups of women were expected to produce responses to anger
expressing concern about the impact of anger on relationships,
indicating a shared sense of responsibility for others
characteristic of relational women.

In a similar way, the responses ofwomen to the sentence stem
"When I feel powerful, it is because " were coded for sources
of power.

Denial of being powerful or viewing being powerful as a negative
experience was expected to be more characteristic of high POC/ low
SOCCF men than high SOCCF/low Pm women.

Power from self-esteem or other positive feelings about the self,
and power from perceiving choices about oneself/one's life were
expected to be more characteristic of high SOCCF/ low POC women than
high POC/ low SOCCF women.

Both groups of women were expected to have similar experiences of
power from relationships with others, as relational ly-oriented

women.

METHODODOGY

The data used in this research were drawn from a large data base
collected in the period of April, 1985 through JUne, 1985, by a research
group to which this researcher belongs. The group began as an infonmal
study group on adult development, composed of interested graduate
students and faculty in developmental and clinical psychology. After
studying Gilligan's work on moral development, the group began work on
an instrument to measure her model, which became the original RSI. This
researcher joined the group in fall, 1984, and participated in the
revision work on the RSI, choosing instruments and designing validity
research, preparing packets and collecting data, cleaning data and
constructing codebooks, and in training and supervision of coders for

sentence completion data.

Participants

 

In the RRSI research project, data were collected from three
populations: high school students, college students, and adult women not
in college. The high school students were seniors at an Eastern Los
Angeles high school and were not included in the subsample used in the
this research, which was drawn from the combined samples of
undergraduate and adult women.

The undergraduate sample was drawn from the population of Michigan
State university students enrolled in introductory psychology classes in
spring term (April-anne) 1985. Only the female undergraduate students

46

47
who had completed all relevant instruments were used as subjects in this
research.

The adult.women sample was drawn from the population of women.who
attended college Week, an onecampus adult enrichment program sponsored
by the Home.Extension Service of Michigan State university; These women
typically are married with children, and have a high school education,
some with college experience. mly women who completed all relevant
instruments were used as participants in this research.

In order to identify women who were similar in their descriptions
of themselves as connected and relationally oriented, female
undergraduate and College week subjects were selected if they had mean
scale scores greater than or equal to 3.5 on the 5-point Relational Self
Scale, indicating that this scale was self—descriptive of them. The
resulting sample of 323 subjects included 61 female undergraduates and
262 College week women.

To further identify women on their endorsement of the Primacy of
Other (POC) scale and the Self and Other Care Chosen Freely (SOCCF)
scale, groups were formed using the median split method. The initial
plan for creating ”types" of relationship styles based on POC and SOCCF
scores used scale anchor points and meanings, with scores greater than
or equal to 3.5 considered ”high" and scores less than or equal to 2.5
considered 'low' on the two scales. Because of the limited range of
scores Obtained, particularly on SOCCF where there were no ”low“ scores
using this criterion, the median split procedure was chosen.

Wbmen with POC scores greater than 3.2 and SOCCF scores less than
3.9 formed one group (high POC/low SOCCF, n = 74). Women with POC

scores less than 3.2 and SOCCF scores greater than 3.9 formed another

48
group (higm SOCCF/ low POC, n = 86) . These two groups represented the
two ”types" of relationship style which were of primary interest for
explorations of similarities and differences in depression, self-esteem,
anger, and power. The other groups (high POC/high SOCCF, n = 81, and

low POC/ low SOCCF, n = 82) represented the remaining classifications.

Procedure

Undergraduate participants were recruited with sign-up sheets
posted in their introductory psychology courses. They were tested in
small groups of 10-25 students by female research group members.
Research packets were administered and checked for completion, with
credit given only for completed packets. College Week women were
recruited with the approval of the College Week administration.
Research packets were included with introductory materials on College
Week given to women upon arriving on campus . Completed packets were
returned to collection boxes at several different sites convenient to
the men's classes and dorms.

Ethical Review

 

This research was approved by the University Committee on Research
Involving Human Subjects (UCRIHS) . A cover letter explaining the

research for informed consent appears in Appendix A.

Instruments
The instruments used for this study were part of a research packet
which contained the RRSI and other instruments. The RRSI appeared first
in all packets, with the remaining instruments counterbalanced across

six orders. The following instruments were used in this study and appear

49
in Appendix B: Relational Self Inventory (RSI), Center for
Epidemiological Studies Depression Scale (CBS-D), Rosenberg Self-Esteem
Inventory (SE), the anger subscale of the EASI-III Temperament Survey
(items 6-9), and sentence completion items. As described below, the
Relational Self Inventory was revised to become the Revised Relationship

Self Inventory (RRSI), included in Appendix C.

Revised Relationship Self Inventory

 

The Revised Relationship Self Inventory (RRSI) is the product of a
research group's efforts to measure constructs described by Gilligan
(1982) and Lyons (1983), through the creation and use of easily-
administered scales (Reinhart et al, 1985: Pearson et al, 1985: Strommen
et al, 1987). In revising the original RSI, additional items were
generated, reviewed, and modified. Ninetybseven items (original RSI
items and new items) were randomly ordered to form an inventory.

The revised inventory was administered to a total of 930 women and
228 men. It was administered as part of a research packet, described
above, completed by male and female undergraduate college students and

adult College Week women (described under Participants). The inventory

 

was also administered by itself, as a single instrument, to a small
group of high school students, to additional undergraduates, and to a
small group of separated and divorced people in another research
project; none of these groups were included in the present study.
Participants rated the selfedescriptive value of items on a five-point
scale: 1= "Not like me at all”: 5= "Very much like me".

Oonfirmatory cluster analyses yielded fOur internally consistent,

reliable scales: Separate/Objective Self: Relational/Connected Self;

50
Primacy of Other Care: and Self and Other Care Chosen Freely. Patterns
of response on all scales were similar for men and women, and scale
intercorrelations formed expected patterns in both women's and men's
data (Stromen et al., 1987) . The Revised Relationship Self Inventory
(RRSI) consists of the 60 items from these four scales randomized, with
five-point self-descriptive response options. The RRSI, scale
reliabilities, scale intercorrelations, and itemrscale total
correlations are included in Appendix C. validation studies on the RRSI

are currently underway.

center fOr Epidemiological Studies Depression Scale

The center for Epidemiological Studies Depression Scale (CBS-D) was
used to measure depression. The CBS—D (Radloff, 1977) is a 20-item
self-report scale designed to measure depressive symptomatology in the
general population. Respondents are asked to indicate how often‘within
the past week they have experienced the items as applicable to
themselves, using a 4-point rating scale of frequency per week, where 1:
Rarely or ane of the Time (less than one day), 2= Some or a Little of
the Time (1-2 days), 3= Occasionally or a Moderate Amount of Time (3-4
days), and 4= Most or All of the Time (5-7 days).

Radloff (1977) reports very high internal consistency (.85) and
adequate testvretest repeatability (.54). validity was established by
patterns of correlations with other self-report measures, by corre-
lations with clinical ratings of depression, and by relationships with
other variables which support its construct.validity (significant life
events, perceived need for treatment, improvement after treatment).

Reliability, validity, and factor structure were similar across a wide

51
variety of demographic characteristics in the general population samples
tested.

Rosenberg Self-Esteem Inventory

 

The Rosenberg Self-Esteem Inventory (Rosenberg, 1965) was used to
measure self-esteem. This is a 10—item self-report scale which asks
respondents to strongly disagree, disagree, agree, or strongly agree
with items reflecting positive or negative attitudes towards the self.
Rosenberg (1965) reports test-retest reliability of .92 and internal
consistency of .72. The scale has been shown to be empirically related
to depressive affect, anxiety, and peer-group reputation (Rosenberg,
1979). Convergent validity has been shown with measures of the same
concept based on different methods: Kelly Repertory Test (a self-ideal
discrepancy test), a self-image questionnaire, and psychiatrist's

ratings (Silber & Tippett, 1965) .

Anger Scale

 

The EASI-III Temperament Survey (Buss and Plomin, 1975) is a
50—item selfereport survey of aspects of temperament, using a S-point
Iikert-type rating scale from 1= Net at all like me, to 5= Very much
like me. Only the subscales measuring Sociability, Anger, and Fear were
included in the research packe . The Anger subscale was edited by the
research group: one item was deleted ("men displeased, I let people
know it right away"), and two items were rewritten to remove specific
behavioral descriptions of anger thought to be more typical of a pattern
of expressing anger that.might characterize a separate/objective

perspective. "I yell and scream more than most people my age" became “I

52
get angry more often than most people my age,“ and “I am known as
hot-blooded and quick-tempered” became "I frequently get angry." Only

the revised Anger subscale was used in this research.

Sentence Completions

 

A sentence completion test was developed by the research group and
included in the research packets because Gil ligan and others have
questioned whether similar terms and language, particularly moral
language, are interpreted differently according to one' s moral
orientation or relationship style. For example, "responsibility" might
have quite different meanings for someone using a justice orientation
than for someone operating from considerations of care. An open-ended
sentence completion format might elicit the variety of meanings such
language could hold.

Using the stem "To me [term] means __," participants were asked
to define the following terms: responsibility, selfish, reciprocity and
dependency. Also included were the following statements: I feel in
control when _: I feel out of control when _: I hurt others when
_: men I get angry I _: men I feel powerful it is because
_. ally the last two sentence corpletions regarding responses to
anger and power were used in this research. The sentence completion item
about power was written by this researcher, and included in the research
packet to elicit and explore the varieties of experiences and soirces of
power, following an earlier interest in this issue developed through
clinical work (described under M) .

A codebook for coding sentence completion responses was developed
by the research group. Coding categories for responses to each sentence

53

completion stem were developed separately, using theoretical and
empirical considerations. Group members worked with samples of responses
from different subjects, shaping coding categories to address the
variety of responses received. Attempts were made to describe the coding
categories using language from our theoretical framework, linking coding
categories to larger theoretical constructs wherever possible.
Categories were refined and clarified through group discussion and
consensus. Descriptive labels, coding criteria, and appropriate
examples were collected for each code, and were compiled in a codebook.

The codes fOr sentence completions about anger and power were
refined by a subgroup of the research group. This researcher devised
the first version of codes for the sentence completion about power,
basing them largely on theoretical oonceptualizations about the sources
of power derivative from a separate/objective orientation, and the
experiences of power consonant with a relational/connected orientation.
Further articulation, clarification, and development of these codes,
including the addition of categories to address the range of responses
received, was accomplished through group participation and discussion.
Final codes for power were completed during the introduction and
training of students serving as coders fOr the sentence completion data.

The codes for the sentence completion about anger were developed
through a similar process of working with samples of responses and
creating coding categories. categories were created to reflect the
phenomenology of responses to one's own anger, and incorporated in
category descriptions theoretical and clinical perspectives on the
dynamics of anger. codes were developed by the research subgroup, and

54
completed during the training of student coders. The complete codebook

for anger and powerful appears in Appendix D.

Sentence Completion Codes for Poerful

 

Responses to the stem, "When I feel PCMERFUL it is because __,"
were coded using nonexclusive categories: responses received multiple
codes if appropriate. The codes were categorical variables, i.e. , noted
as present or absent. Coding categories included codes that were not
specifically related to the research hypotheses of this study, as well
as codes for which hypotheses were made. The codes that were directly
relevant to research questions on the experiences of power for women
included: DENIAL of being or feeling powerful, PCMER from/in
RELATIONSHIPS with others, POWER from SELF-ESTER! and/or other positive
feelings about the self, and PCMER from awareness of CHOICE about one's
self/own life.

Direct statements denying the personal experience of poer ("I
never feel poerful") , statelents describing poer as inauthentic, as
derivative of a negative experience ("I 'm overcompensating for a feeling
of insecurity") , and staterents describing power as a negative
experience ("I feel I am, but don't really like to") were coded as
DENIAL of being or feeling powerful. The first two types of statements
were differentiated from staterents of poer as negative in another
subcategory. Because of the small number of responses in either
subcategory, the subcategories were col lapsed for analysis. Responses
receiving this code seeted to reflect mixed feelings about power,

including helplessness, defeatist views about self-efficacy,

55
ambivalence, and fears about having power. Power was not seen as a
positive, autonomous experience.

Responses coded as experiences of POVEIR from/in RETATIONSHIPS with
others reflected feelings of power derived from connections with others,
interpersonally grounded. Responses receiving this code were not
necessarily reflective of a connected/relational self orientation, but
related perceived poer to aspects of relationships characterized by
five subcodes. The first three subcodes included poer from erotional
acceptance by others, being liked or loved ‘("I know I am loved by
important people in my life”) , poer from unidirectional dependency,
with the self as source of support for others (”someone depends on me") ,
and power from others' respect for one's skills or competency ("Others
respect my opinions and actions") . Poer from self-and-other-care,
including self as a recipient of care ("I have done sorething worthwhile
for others and myself") and power from relationships, nonspecified ("I
have a good relationship with those I am working with") were the last
two subcodes. Because of the small number of responses in any
subcategory, the five were collapsed for analysis.

ENTER from SHE-ESTHER and/or other positive feelings about the
self included statements of self-confidence ("I feel good about myself
and my potential") , self-liking ("I like myself") , and other positive
feelings ("At peace with myself": "My inner being is calm”) . PWER from
awareness of CHOICE about oneself/own life reflected the capacity to
experience the self as able to choose, with freedom and responsibility
for one's choices. Statements coded here euphasized seeing oneself as
having options and the agency and freedom for self-definition ("I am in

charge of myself," ”I'm doing what I want,” "I control my life”). While

56
this perception of the self might lead to increased efficacy, ownership,
and control of one's life, choice was the central there of this
category, not ”control” per se. Statetents of "I am in control” with no
further elaboration, and staterents of situational control, not
including control of self or others ("I have control of a situation") ,
were coded separately in a subcategory that was not used for analysis
because the meaning of these responses was unclear.

The retaining categories were not specifically related to the
hypotheses of this study. Positional power, or POVER from one's
POSITION or role in a hierarchy relative to others' positions,
emphasized poer over others, and experiences where power for one person
might be at the expense of another's power or status. This type of
power might be role-conferred, and was often expressed in statements of
comparison or in competitive imagery. Examples included ”I am in charge
of a situation and my decision is the final one,” "I have something
other people want," "I have knowledge or some other advantage on my
side," and "I am with someone younger than me."

POWER from moral rightness or moral certainty captured those
responses indicating that respondents derived power from feeling morally
or conventionally in the "right.” Bcamples included "I'm absolutely
sure of my convictions“ and "I have stood up for what's right.” MEIR
from ACHIEVEMENT/WW included satisfaction at task
completion, accomplishment, success, achievelent, hard work, etc. Here
poer was experienced through the process of achieving and/or the
product of one's achievetent and work. Goals might not be self-chosen:
the elphasis was on the activity of achieving as the source of poer.
Examples were "I have cote up the hard way and have worked hard to get

57
what I have now," "I have succeeded in doing something well," and "I
feel like I accotplished something."

PWER with BALANCE involved dealing with multiple dimensions and
juggling them successfully ("I have been true to myself and still gotten
others to understand without hurting them; I can cope with my many
responsibilities and still smile and be happy") . POWER from
PREDICI‘ABILITY included responses where predictability, usually through
knowledge or planning, contributed to a sense of comfort and limited
control ("I am prepared for a situation," "I know what I'm doing").
PASSIVE or IMPERSONAL PGNEIR reflected experiences of general contentment
and passive well-being (”Everything is going great," ”Things are going
OK") , with a subcategory for religious references ("Because of my
reliance on God,” ”I'm letting Divine Power express throigh me") .
Responses that were not scorable by any of the codes described here

received a separate code, nor SCORABLE (e.g.,"I have energy to spare”).

Sentence Cowletion Codes for Anger

 

Responses to the sentence completion stem, "When I GET ANGRY I
_, " were coded using nonexclusive categories: responses received
multiple codes if appropriate. Codes were categorical variables, i.e.,
noted as present or absent. Categories related to this research
included DENIAL of anger, focus on RESPQJSIBILITY for others
HURT/concern for impact of anger on others, WITHDRAWAL, JUDGEMENTS about
anger, SJMATI ZAT‘ICN, RATICNAL approaches, transformation of ABEER to
SADNESS/DISTRESS, and coMUNICA'now/Rnsowrlom attempts.

Direct statements denying the experience of being angry were coded

as DENIAL of anger (”I rarely get angry," "I never get angry”).

58
Responses reflecting concern for the impact of one's anger on others,
with the view that anger might hurt others or damage relationships, were
coded for their focus on RESPONSIBILITY for others HUM ("Wish to avoid
hurting others,“ “Fear the results of the anger in my words and actions,
upon others and myself") .

Attempts at holding anger in, at withdrawal from feeling angry,
expressing anger, or the conflictual situation itself, were coded as
WITHDRAWAL. This category reflected respondents ' attempts to retreat
from their anger and/or the conflictual situation. WITHDRAWAL might be
physical, as in going off alone, or exotional, as in becoming quiet and
pulling back inside. There might be an active wish not to talk or
communicate, and to hide feelings. Angry feelings and tension might be
channeled into physical activity and discharged indirectly in a
sub] imated fashion, distinguished from cathartic physical acting-out.
Subcategories differentiated withdrawal and/or substitute activity from
attempts at regulating/ impulse control against losing control. Ebramples
of withdrawal and/or substitute activity included "Get quiet or do
house-work fast-like,” ”Want to be alone,” ”Bottle the motions as much
as possible so as not to let others know I'm upset,” and ”Take a walk".
Ekamples of attempts at regulating/impilse control included "Usually
walk away, cool my terper" and "Try to keep a lid on it for fear of
losing control.“ Subcategories were collapsed for analysis because they
reflected aspects of one type of response to anger.

JUDGENHTTS about anger had subcategories for negative and positive
judgements about anger or the aneg self, which were analyzed
separately. Negative judgements included ”Don't like myself," "Get

upset because I do not like the feeling,” and "Lose control of myself

59
and feel degraded.“ Positive judgements included ”Feel I have a right
to vent anger when it is hurting me,” and ”I feel good after, because
that has been a prdblem for me (keeping anger to myself).'

SOMATIZATION included physical reactions or symptoms experienced as
a result of getting angry (”Get a headache, muscle spasm, or other
physical symptom,” ”My stomach gets in a knot and I can't breathe
easily”). RATIONAL approaches to anger included three subcategories.
Introspective problem-solving/analysis through rational means reflected
attempts, usually by oneself alone, to think through the problem or
examine the angry feelings Objectively ("Become very analytical,“ "Write
in my journal,” and “Think it out"). Other subcategories were
discussion involving another person who was not the recipient of anger
("Go talk to someone who is not involved with the fight"), and religious
solutions ("Pray, then act,” "Use prayer").

The first subcategory of introspective problem-solving was analysed
separately, then combined with the second subcategory of discussion with
others for analysis.

Transformation of ANGER.to SADNESS/DISTRESS described responses
where anger had been tranformed into expressions of sadness,
tearfulness, distress, or depression. This category was specifically
for tranformations of the affect of anger to expressions of the affect
of distress or sadness, not all possible tranformations of anger to
other phenomena. All expressions of tearfulness or crying were included
in one subcategory ('Cry,” Usually get upset and end up in tears,”
I'Start crying”). Expressions of depression and sad or hurt feelings
were coded in a second subcategory ("Cry and feel sorry for myself,"

”Become depressed,” ”Feel hurt"). The two subcategories were combined

60
fer analysis because they reflected similar dynamic processes of
transfbrmation.

WICATIQT/ RESOLUTION attempts were divided into those attempts
at communication of feelings and reconciliation with the partner to the
conflict, and impersonal statements about correcting or resolving the
problem which did not specify others. Dcamples of communication of
feelings/reconciliation attempts with others were ”Talk‘with the other
person," and “Tell the person I'm angry with.” Impersonal statements
about correcting the problem included "Try to resolve the problem,” and
”Try to correct the situation which made me angry.” The first
subcategory was used separately in analysis, then in combination with
the second subcategory.

Other codes were developed which were not specifically linked to
hypotheses of this research. These included BEHAVIORAL expression of
anger and REACTIVE AGGRESSION towards others,‘with a final category of
ZNOT SOORAELE, for responses falling outside the categories. Most of the
responses deemed NOT SOORABLE by these categories seemed to be
misreadings of the stem: the most cormon misreading was "I get angry
when... people try to hurt me or someone I love, I get a lower grade
than I expect, I don't live up to my own expectations or I am tired,"
etc.

BEHAVIORALmexpression of anger was intended to capture various
forms of emotional or physical expressions of anger in an acting-out
fashion, intended fer venting affect rather than for communication. One
.subcategory included physical responses, with possible aggression
towards Objects but not others as targets (“Throw old, Chipped coffee

mugs against the garage wall," “Throw things, slam doors”). verbal

61

expressions comprised another subcategory ("Scream and yell," "Sometimes
shout but don't hold grudges”). Responses including both physical and
verbal experiences (”Throw things or say things I don't mean") and
expressions of anger not specified ("I blow up,” ”Tend to overreact and
show a lot of emotion") were the third and fourth subcategories.

REACTIVE AGGRESSION towards others, including retaliation, was
created to diffentiate responses of behavioral expression from responses
of aggression towards others as a result of anger. Subcategories
included physical aggression ("Hit the other person," ”Throw something
at the other person, lose control"), verbal aggression ("Deliberately
say things I know will hurt or will make the other person angry or I
cvercritize," ”Say mean things to get back at the other person”), and
mode of aggression unspecified ("1 fight back any way I can,” ”Take it

out on my family and friends").

Data coding

 

Sentence Completion coding

 

Four undergraduates (two men and two women) and one female graduate
student served as coders for the sentence completion data, receiving
academic credit for their involvement in this research project. The
coders were trained by this researcher and two other members of the
research group.

Initial training involved detailed presentation and discussion of
the codes for sentence completion responses about poer. This was
fol lowed by trial coding of samples of sentence completions from student
respondents from two undergraduate courses not included in the sample

for this researCh. All student coders and research group members

62
independently coded sample sentence completions for power and then
discussed the coding used, checking for consensus on the application of
the codes, clarifying meaning and intent of coding categories, and
highlighting problems. Trial coding was repeated until students were
familar with the codes and able to use them in agreement with research
group members' usuage.

coders then coded sentence completion data for experiences of
power. During coding, an additional training session was held to
clarify the subcategories for Power from RELATIONSHIPS with others,
because inspection of completed codes for this category showed
reliability drift. coders then reevaluated all previous uses of this
code, either changing or confirming their initial coding. Training for
coding sentence completion responses to anger was initiated after coding
was completed for the data on power, and was conducted as described
above.

All sentence completion responses were coded by two coders working
separately. coders recorded their individual coding, and resolved
differences for a final code through discussion with each other. When
they were unable to reach consensus, this researcher resolved the coding
problem. Final codes for each item‘were entered into the computer by
pairs of coders, with one coder reading and the other entering data.
Random checks of codes revealed no errors.

Analyses

Descriptive statistics were calculated providing frequencies, means
and standard deviations of sample demographics and scales. Using

cronbach's alpha, reliabilities for the CES—D, Rosenberg Self-Esteem

63
Inventory, and EASI-III Anger scale were computed for the subsample used
in this research. Inter-rater reliability was calculated on random
samples of the final codes from each pair of coders, since final codes
were reached by consensus. For each of the 4 pairs of coders, ten
powerful and ten anger responses (if both were coded) were selected and
coded by a research group member other than this researcher. The
research group member was an active participant in code construction and
coder training, and served as the standard against.which coder pair
final ratings were evaluated. Inter-rater reliability was calculated
using Cohen's kappa, comparing her coding of the selected responses with
the final codes from each coding pair.

A correlational matrix was computed to explore hypotheses regarding
expected patterns of correlations for the two relational perspectives of
POC and SOCCF on the variables of depression, self-esteem, and anger.
The total subsample of 323 women was used in the creation of the
correlational matrix. Regression analyses were calculated to further
explore the contributions of POC and SOCCF in predicting these external
variables.

The analyses just described included all undergraduate and college
week women identifying themselves as connected and relationally oriented
(with mean scores greater than or equal to 3.5 on the 5-point Relational
Self Scale) regardless of the relative levels of their scores on POC and
SOCCF. Because women could score high on both scales, it was necessary
to identify groups of women scoring high on only one of these scales,
thus forming ”refined types" for further analysis. Four groups were
created, using the median scores on POC and SOCCF to split the sample
along dimensions of high/low POC and high/low SOCCF. women in the high

64
POC/low SOCCF group and men in the high SOCCF/low POC group were
considered ”refined types" of primary interest for comparisons on the
variables of depression, self-esteem, and anger. Analyses of variance
with Scheffe post hoc corparisons were used to compare the groups for
expected differences on depression and self-esteem, and expected
similarity on the measurerent of anger.

The responses of the two groups of high POC/ low SOCCF and high
SOCCF/ low Pa? woten on the sentence cmpletions concerning anger and
power were examined using all possible classification categories.
Frequencies and percents were reported in a descriptive account of how
women with different care orientations responded to their anger and how
they experienced their power.

To compare the responses of high POC/ low SOCCF and high SOCCF/ low
Pm woten on specific anger and power variables (see Sentence Completion

Codes for Poerful and Anger), the Chi square statistic was selected

 

because it is the appropriate statistical test for use with categorical
variables. Because it required a mutually exclusive category system,
and the classification scheme used for each variable had codes which
were not mutually exclusive, separate Chi squares were calculated for
each variable of interest. These compared the expected distribution
with the observed distribution for significant differences.

Independent Chi squares were calculated for the following anger
variables hypothesized to be more characteristic of high POC/ low SOCCF
than high SCXZCF/low POC woren: DENIAL of anger, WITHDRAWAL and/or
substitute activity, SGIATIZATICN, NEEATIVE judgerents aboit anger or
the angry self, and transformation of ANGER to SADNESS/DIST'RESS.

Separate Chi squares were calculated for the following variables

65
hypothesized to be more characteristic of high SOCCF/ low POC than high
POC/ low smCF women: POSITIVE judgements about anger, RATICNAL
approaches including introspective problem—solving and discussion with
others not part of the conflict, and MUNICATION/RESOIDTIGNI attempts.
A Gui square was calculated to look at responses categorized as focusing
on RESPGTSIBILITY for others HURT, where no differences were expected
between high POC/low SOCCF and high SCECF/low POC woven in their use of
this category.

Independent Chi squares were also calculated to explore differences
between high POC/low SOCCF and high SOCCF/low Poc women in their
experiences of power. A separate G11 square was calculated using the
power variable hypothesized to be more characteristic of high POC/ low
SOCCF than high SOCCF/ low POC woven: DENIAL of poer. Separate Gui
squares were calculated using power variables hypothesized to be more
characteristic of SOCCF than POC woven: poer from SELF-ESIEEM and other
positive feelings about the self, and power from CHOICE about one's
life. A Gui square was calculated using the variable of power from
RELATIONSHIPS, which was expected to characterize high POC/ low SOCCF and

high SOOCF/ low POC women similarly.

RESULTS
Results are presented fol lowing the order described in the previous
section on Analyses. Results will be discussed as they relate to the

research hypotheses, and summarized at the end of this chapter.

Descriptive Statistics

 

Two hundred seventy undergraduate men and woven completed research
packets for credit for their psychology classes. The students ranged in
age from 16-55 and were primarily Caucasian. 516 adult women attending
College Week voluntarily completed research packets during College Week.
These women range in age from 18-78 years and were primarily Caucasian.
The sample selected for this research consisted of those undergraduate
and College Week women who completed all relevant instruments and who
has mean scores greater than or equal to 3.5 on the 5-point Relational
Self Scale. Table 1 presents the means and standard deviations for the
resulting sample of 61 female undergraduates and 262 College Week woven
for the variables Relational/Connected Self (CS), Primacy of Other Care
(POC) , Self and Other Care Chosen Freely (SGICF) , depression, self-
esteem and anger. This sample did not report being very depressed,
similar to levels of depression measured in general population samples:
when tranformed to correspond with the rating scale used in this
research, the mean and standard deviation reported by Radloff (1977) for

one such hoisehold sample were 1.46 and .43, respectively. This sample

66

67
did not report being very angry, and had generally high self-esteem,
with little variability.

Multiple groups confirmatory cluster analysis (Hunter, 1975, 1977;
NUnnally, 1978) used on male and female data separately and combined
produced four internally reliable and consistent subscales for both men
and women in the revision of the RRSI on a total sample of 930 women and
228 men. These subscales were examined across age groups, using
confirmatory cluster analyses and scale intercorrelations. Internal
consistencies of the scales were similar across age groups, indicating
that the scales were reliable for both younger and older respondents.
The general patterns of scale intercorrelations were also similar. Thus
the subsample for this research was drawn from the two samples combined.
To confirm the appropriateness of combining the 61 undergraduate women
and the 262 college week.women for this subsample, tftests were computed
for the RRSI variables CS, POC, and SOCCF. There were no significant
differences between the 61 undergraduates and the 262 college week
women, who composed the total subsample of 323 women, on the RRSI
variables. comparisons of the two groups on CS, POC, and SOCCF are
presented in Table 2.

Using cronbach's alpha, reliabilities for the CES-D, Rosenberg
Self-Esteem Inventory, and the revised EASI—III Anger scale were
computed for the subsample used in this research. Internal consistency
for the CBS—D was .91. Scale reliability fer the Rosenberg Self-Esteem
Inventory was .83, and scale reliability fOr the revised Anger scale was
'.69. Inter-rater reliability for the sentence completion coding was

calculated using cohen's kappa. Kappas for the four pairs doing the

68

.mmm ... m .302

 

 

om. 84TH: mam m 5785: Eton $94 273%
3. 845.: RA v ism. fichEH ammummvfimw 885mg
Sumac. 2QO
am. immense . 34 a teammate $th 808235de sou H356
S. 85-8.8 3.,“ m E809 38m 3me 586 $8 550 ea £8
me. 85.4.2.2 Ra m 88. 28m 83 unto mo sonata
am. 86:98 84 m Go. 28m name Buooccoohaoflmamm
om mmwoom no waste :8: 2.8m mo mmcmm Saunas

 

 

mafia mom 2033\er oucocoum ocm mama:

H 23..

Table 2

69

Comparisons of Undergraduate8 and Collgle Weekb Women in Research Sample

 

 

 

Mean SD t-va lue DF Two-tail
‘ Probability
CS
Undergraduates 4 . 22 . 32 - . 36 108 . 84 . 72
College Week 4.24 .40
PCXI
Undergraduates 3.24 .46 -.51 95.65 .61
College Week 3.27 .50
SOCCF
Undergraduates 3.97 .46 1.16 92.46 .25
College Week 3.90 .47
33 = 61. bn = 262.

7O
coding of poer responses were .57, .76, .89, and .89. Kappas for the

pairs who coded anger responses were .75, .78, and .86.

Relationships amom Relational Perspectives and Other Variables

 

Pearson correlations were computed to explore hypothesized
relationships among the relational perspectives of POC and SOCCF, and
depression, self-esteem, and anger. The total subsample of 323 woven
was used in the creation of the correlational matrix, presented in Table
3. Correlations corrected for attenuation are presented below the
diagonal in Table 3.

As expected of different manifestations of the relational/connected
self, both POC and SOCCF were significantly positively correlated with
the Relational/Connected Self Scale. The absence of a significant
correlation between POC and SOCCF indicated the relative independence of
these two perspectives. These correlations among tre RRSI variables
were similar to the interscale correlations produced by the total sample
in the revision of the RRSI, presented in Table 2c in Appendix c.

As described in hypothesis 1, depression was expected to be
positively associated with POC, and negatively associated with SOCCF.
The correlation between depression and POC indicated a positive
relationship that did not reach significance; depression and SOCCF were
basically independent. The level of depression acknowledged by this
sample was generally quite low, with little variability, which may have
masked relationships between depression and the two variables of
interest. The lack of a stronger positive relationship between
depression and POC was particularly noteworthy, because tie correlation

between depression and POC in the total female sample (N = 930) was .19.

71
Table 3

Intercorrelations amonLAll Measures

 

 

 

CS PCXI SOCCF CBS-D SE EASI-III
CS (.76) .33** .24** -.06 .11* -.11*
Pa? .45** (.68) .07 .08 -.19** .02
m .31** .09 (.78) .01 .08 -.11*
CBS-D -.O7 .10 .01 (.90) -.41** .32**
SE .13* -.25** .09 . -.47** (.84) -.29**
EASI-III .15* .02 -.14* .40** -.38** (.69)

 

Note. N = 323. Alphas are on the diagonal.

below the diagonal .

Corrected for attenuation

* p < .001, two-tailed test. ** p < .05, two-tailed test.

72
A regression analysis using POC and SOCCF with depression was not
significant, indicating that neither relational perspective was
important in predicting depression, _F(2, 320) = 1.07, p = .34.

Because women in this subsample had scores of 3.5 or above on the
Relational/Connected Self scale, perhaps this selection factor
influenced the relationship between POC and depression. Perhaps being
strongly relational protected against the vulnerability to depression
associated with Pm in our total sample. To test this, partial
correlations were computed, with CS held constant. The relationship
between POC and depression changed only slightly when the
Relational/Connected Self scale scores were held constant (from .08 to
.10) . This was also true for the relationship between SOCCF and
depression, when CS was held constant (from .01 to .02) .

As predicted in hypothesis 2, self-esteem was expected to be
negatively associated with POC, and positively associated with SOCCF.
Self-esteem was significantly negatively correlated with POC, as
expected. The correlation between self-esteem and SOCCF indicated a
weak positive relationship. The relationship between self-esteem and
SOCCF was in the predicted direction, but did not attain significance.
However, using the E to _z transformation to test differences in
correlations, the correlations (corrected for attenuation) of self-
esteem with POC (_r; = -.25) and self-esteem with SCXCF (5 =.09) differed
significantly and in the predicted direction, _z_ (320) = 4.692, p < .001.
The two relational variables produced the hypothesized pattern of
(relationships to self-esteem, and related to self-esteem in
significantly different ways. A stepwise regression analysis was

significant for POC as a predictor of self-esteem, with SOCCF not

73
contributing significantly to the regression equation, {(1, 321) =
12.34, p = .001.

‘No predictions were made regarding the relationships between POC,
SOCCF, and anger as measured by the EASI-III Anger Scale. Hypothesis 5
predicted no differences between the two "pure types” on this measure,
because it was assumed that possible real differences in intensity and
frequency of anger would not be revealed on a direct, self-report
measure. Concerns with social desirability factors particularly salient
‘with respect to women's anger were expected to mask differences related
to POC and SOCCF. correlations indicated no relationship between anger
and POC. However, SOCCF was significantly negatively related to anger.
Thus, greater endorsement of balanced self and other care was associated
‘with less frequent and intense reported experiences of anger. When the
correlations (corrected for attenuation) of anger with POC (£.= .02) and
anger with SOCCF (E = -.14) were compared using the _r_ to z
transformation, they differed significantly, _z_(320) = 2.14, p < .05,
indicating that POC and SOCCF related quite differently to anger.

A regression analysis was performed using POC and SOCCF to predict
anger, which was not significant, F(2, 320) = 1.97, p = .14. Since
self-esteem also correlated significantly with anger, regression was
also performed using POC, SOCCF, and self-esteem to determine their
relative strengths in predicting anger. iNeither POC or SOCCF were
important in predicting anger: self-esteem*was significant, suggesting
that the obtained correlations may reflect the mediating effect of self-
‘esteem, £(1, 322) = 30.69, p = .001.

Other significant relationships were found among depression, self-

esteem, and anger, which were not themselves the focus of this study.

74
Depression was negatively associated with self-esteem, and positively
with anger. Self-esteem was significantly negatively related to anger.
These three variables form a ”depressive triad” , such that high
depression is associated with low self-esteem and high anger. The
relationship between depression and self-esteem has been noted in
clinical and theoretical accounts of depression (Kaplan, 1986) .
Unconscious or introjected anger has been described as a dynamic aspect
of depression in early psychoanalytic accounts of depression. For this
sample, acknowledged dissatisfaction and anger was related to

depression.

Comparisons of High POC/Low SOCCF and High SOCCF/Low POC Women on

Depression, Self-esteem, and Anger

 

The preceding analyses included all women regardless of the
relative levels of their scores on POC and SOCCF. To the extent to
which woven score high on both these scales, results may be unclear.
Therefore, 91‘pr of women who represented "refined types" by scoring
high on only one of these scales were identified for further analysis.
Four groups were created through a median split procedure using the
median scores on POC and SJOCF scales. Woven with POC scores greater
than 3.2 and SOCCF scores less than 3.9 formed the high POC/low SOCCF
group (2 = '74). Women with POC scores less than 3.2 and SOCCF scores
greater than 3.9 formed the high SOCCF/low POC groip (n = 86). Woven
with scores above the medians for POC and SOCCF were considered high
Pal/high SOCCF (_n = 81): women with scores below both medians were low

POC/low SOCCF (n = 82) . Means and standard deviations for the four

75
groups on the variables of depression, self-esteem, and anger appear in
Table 4.

The high POC/low SOCCF and high SOCCF/low POC groups represented
relational "types“ of primary theoretical interest for group comparisons
on depression, self-esteem, and anger. Analyses of variance with
Scheffe post hoc comparisons were used to explore differences among the
groups described in hypotheses 3, 4, and 5. The conservative Scheffe
post hoc procedure was chosen instead of planned comparisons because of
interest in all four groups, with interest in the high SOCCF/high POC
and low SOCCF/low POC groups being exploratory in nature.

Hypothesis 3 stated that women whose relational identity was
described by high POC/low SOCCF were expected to be more vulnerable to
experiencing depression than women whose relational identity was
described by high SOCCF/low POC. Table 4 shows CBS-D score means for
these two groups to be virtually identical (1.52 and 1.54), and indeed,
none of the four groups differed significantly on depression, §(3, 319)
= .28, p_= .84. These findings corroborate the correlational results
indicating lack of a strong relationship»between depression and either
relationship style.

As described in hypothesis 4, high POC/low SOCCF women were also
expected to have significantly lower self-esteem than high SOCCF/lcw'POC
woven. Looking at the overall patterns, high S(I:CF/low POC women had
higher self-esteem (M = 3.40) than high POC/low SOCCF women (M = 3.19).
Both low POC groips, regardless of SOCCF status, had higher mean self-
esteem (3.40 and 3.29) than the twwo high poc groups (3.19 and 3.19).
Regression results showed that low POC scores were more important than

high SOCCF scores for high self-esteem. An analysis of variance was

Table 4

76

Means and Standard Deviations for Groups

 

 

Groups CES-D SE EAST-III

High POC/low SOCCF
Mean 1.52 3.19 2.09
Standard Deviation .48 .51 .77
n_= 74

High SOCCF/low POC
Mean 1.54 3.40 1.98
Standard Deviation .55 .45 .78
n’= 86

High SOCCF/high POC
Mean 1.58 3.19 2.22
Standard Deviation .53 .50 .88
2.: 81

Low SOCCF/low POC
Mean 1.52 3.29 2.24
Standard Deviation .44 .47 .77

n_= 82

 

77
significant, indicating that the four groups differed in terms of self-
esteem, MB, 319) = 3.65, p = .01. However, the only two groups that
significantly differed at the .05 level were the high SOCCF/ low PG:
women (M = 3.4) and the high SJCCF/high Pm women (M = 3.19). while a
similar difference existed between the two groups of primary interest,
high SOCCF/low POC women (M = 3.4) and high POC/low SOCCF women (M =
3.19) , this did not reach statistical significance, possibly because of
differences in standard deviations.

For reasons discussed earlier, hypothesis 5 stated that the two
groups were not expected to differ on a measure of the frequency and
intensity of anger. Although these two groups had lower mean scores
than high SOCCF/high POC and low SOCCF/low POC women, there was no
significant difference among the 91‘pr on anger, _F_‘ (3, 322) = 1.85, p =
.14. Although correlational analysis showed a significant negative
relationship between SOCCF and anger, regression analyses did not
indicate that either SOCCF or POC was significantly important in

predicting anger .

Comparisons of High POC/Low SOCCF and High SOCCF/Low POC Woven on Anger
and Poor Sentence Covpletion Data

Theresponsesofthetwogroipsofwovenontheangerandpower
sentence covpletion itevs were examined using all possible coding
categories. The 74 high POC/ low SOCCF women produced 101 coded anger
responses and 86 coded responses for poer, indicating multiple coding
‘ for the sentence covpletions for anger and for power. The 86 high
soccrl low POC woven produced 117 coded anger responses and 99 coded
power responses, indicating multiple coding for both sentence

completions .

78
m

Wbmen with different care orientations responded to "When I get
ANGRY I _____f in similar ways. Frequencies and percents of responses
to the sentence completion item regarding anger are reported in Table 5.
Table 6 contains anger categories organized from most to least
frequently used by groups. The three most frequent categories of
response for both POC and SOCCF women were behavioral expressions of
anger, withdrawal and/or substitute activity, and tranformation of anger
to sadness/distress. For both groups, behavioral expression included
physical, verbal, both physical and verbal together, and mode of
expression unspecified, with verbal expression of anger, such as yelling
or screaming, most commonly used. Withdrawal and/or substitute activity
included attempts at using withdrawal for impulse control (”Get quiet,"
”Bottle the emotions," "Take a walk,” and ”Usually walk away, cool my
temper”). Anger transformed into expressions of sadness or depression
("cry,” "Usually get upset and end up in tears,” "Become depressed”) was
the third most common response for both groips.

Least frequent for both groups were responses of reactive
aggression, most often verbal, denial of feeling or being angry, and
sovatization. Other categories were used with slightly different
frequencies by the two groups of women, between the most and least
commonly used categories just discussed. The only category in which POC
and SOCCF women really differed was negative judgements about anger or
the angry self,‘with POC women making more negative judgements (16) than
‘SOOCF‘women (9).

According to hypothesis 6A, the following types of responses were

expected to be more characteristic of high POC/low SOCCF than high

79
Table 5

Frequencies and Percents* for Anger Categories by Groups

 

 

 

Anger high POC/low SOCCF high SOCCF/low POC

Denial of Anger 1, .01 O, 0

Focus of impact of anger/others hurt 3, .03 4, .03

Withdrawl and/or 25, .24 29, .25
substitute activity (23) (25)
Attempts at impulse control (2) (4)

Jodgments about anger

Negative 16, .16 9, .08
Positive 0 0
Somatization O, 0 l, .009
Behavioral expression 27, .27 32, .27
Physical (l) (1)
verbal (19) (21)
Both physical and verbal (1) (1)
Mode unspecified (6) (9)
Rational approaches 5, .05 7, .06
Introspective problem-solving (3) (5)
Tell others not part of conflict (2) (1)
(Religious references-not used) (0) (1)

(table continues)

80

 

 

Anger high POC/low SOCCF high SOCCF/IOW'POC
Transformation of anger to sadness 16, .16 16, .14
Cry (13) (13)
Other feelings of hurt,
depression (3) (3)

communication/resolution

attempts with partner 3, .03 5, .04
(Impersonal statements of
resolution - not used) 0 l
Reactive aggression 2, .02 4, .03
Physical (0) (0)
verbal (2) (1)
Mode unspecified (0) (3)
NOt scorable 3, .03 9, .08
TOTAL 101 117

 

* Percents based on numbers of responses, not number of subjects.

81

Table 6

Anger Categories Organized Most to Least Frequently Used by Groug

 

 

High POC/ low SOCCF High SOCCF/ low POC
Behavioral expression Behavioral expression
mainly verbal (27) mainly verbal (32)
Withdrawal (25) Withdrawal (29)
Transformation to sadness (16)/ Transformation to sadness (16)
Negative view (16) Not scorable (9)/
Rational means (5) Negative View (9)
Not scorable (3)/ Rational means (7)
Communication attempts (3)/ Communication attempts (5)
Focus on others hurt (3) Focus on others hurt (4)/
Reactive aggression (2) Reactive aggression (4)
Denial (1)/ Somatization (1)

Somatization (1) Denial (0)

 

82
SOCCF/ low POC women: denial of anger, withdrawal and/or substitute
activity, negative judgements about anger or the angry self,
somatization, and the transformation of anger to sadness/distress.
Independent Gui squares were covputed for withdrawal and/ or substitute
activity with subcategories combined, negative judgments, and
tranforvvation of anger to sadness with subcategories combined. Two
variables (denial of anger and sovatization) had Ms of 1 or 0 and could
not be used. The groups did not differ in terms of their use of
responses of withdrawal, x2 (1, M = 160) = 0, p = .99, or responses of
transformation of anger to sadness, x2 (1, M = 160) = 0, p = 1.0. The
two groups of women had equal numbers of responses of transformation,
with POC women having only slightly higher percentage of responses in
this category. As expected, POC women were more represented in the
category of negative judgevents about anger than SOCCF women, X2 (1, M =
160) = 3.75, p = .05.

As described in hypothesis 6B, the fol lowing types of responses
were expected to be more characteristic of high SOCCF/ low POC women than
high POC/ low SOCCF women: rational means of problem-solving aboit
anger, positive views of anger or the angry self, and attempts at
covmunication of feelings towards resolution and reconciliation.
Independent G'ii squares were computed for rational means of problem-
solving and attevpts at covmunication: there were no responses of
positive views of anger . The subcategory of introspective problem-
solving throigh rational means (alone, usually) was analysed separately,
‘ x2 (1, _13 = 160) = .02, p = .88, and then combined with the subcategory
involving discussion with others not part of the conflict, x2 (1, g =

160) = 0, p = 1.0. For comunication/resolution attempts, only those

83

‘with the partner in conflict were used, excluding the subcategory of
abstract or impersonal attempts, x2 (1, M =160) = .26, p = .61. Chi-
squares were not significant for the hypothesized variables: POC and
SOCCF women were not differentiated by these categories of responses to
anger.

Both groups were expected to produce responses expressing concern
about the impact of anger on relationships, as expressed in hypothesis
6c. The groups were represented equally in this category, x2 (1, M =

160) = o, p = 1.0.

Power

 

High POC/low SOCCF and high SOCCF/low POC women responded similarly
to "When I feel POWERFUL it is because _____7. Frequencies and percents
of responses to the sentence completion item regarding power are
reported in Table 7. Table 8 contains the power categories organized
from most to least frequently used by groups. For the groups, the two
most frequently used categories were power from achievement and power
from being ”in control”. Power from achievement indicated satisfaction
‘with one's performance, accomplishments, efforts, hard work, etc. The
other category was created as a subcategory under power from choice
about oneself, to separate responses of "I am in control" with no
further elaboration, and responses of situational control (”I have
things under control") from statements of control over oneself or over
choices, decisions, in one's life. Statements of explicit control over
(others were coded under power from position. Thus the subcategory had
no theoretical definition and contained responses which were not coded

under control of self or others. It was not clear what responses of ”in

84

Table 7

Frequencies and Percents* for Poor Categories by Groups

 

 

Poer High POC/Low SOCCF High SOCCF/ low POC
Denial of poer 3, .03 7, .07
Denial (2) (6)
Poer as negative (1) (1)
Power from relationships 12, .14. 8, .08
Evvotional acceptance (3) (0)
Dependency (3) (0)
Respect, competence (1) (5)
Caring for self and others (0) (0)
Mode nonspecified/other (5) (3)
Poer from position 3, .03 6, .06
Poor from right 2, .02 1, .01
Power from achievement 14, .16 21, .21
Power from self-esteem 11, .13 9, .09
Power from choice 3, .03 7, .07

Statements of "in control" 14, .16 19, .19
Poer with balance 1, .01 1, .01
' Power from predictability 8, .09 5, .05

(table continues)

85

 

Power

High POC/ Low SOCCF

High SOCCF/ low poc

 

Impersonal or passive poer
Passive

Religious reference

Not scorable

TOTAL

10,
(9)
(1)

5!

86

.12 10, .10
(6)
(4)

.06 5, .05
99

 

* Percents based on number of responses, not number of subjects.

Table 8

Poer Categories Organized from Most to Least Frequently Used by Groups

 

 

High POC/low SOCCF

High SOCCF/ low POC

 

Achievement (14)
"In control" (14)
Relationships (12)
Self-esteem (11)
Impersonal poor (10)
Predictability (8)
Not scorable (5)
Position (3)]
Goice (3)/
Denial (3)

Right (2)

Balance (1)

Achievevent (21)
"In control" (19)
Impersonal power (10)
Self-esteem (9)
Relationships (8)
Choice (7)/

Denial (7)
Position (6)
Predictability (5)/
Not scorable (5)
Right (1)/

Balance (1)

 

86
control" meant to respondents, but the frequency of such responses
seeved to neccessitate a separate category for them.

Poer from moral certainty or right, and power with balance were
the least frequently used categories for both groups of women. Other
categories were used with slightly different frequencies by POC and
SOCCF woven, between the most and least commonly used categories.

Denial of being powerful or viewing being powerful as a negative
experience was expected to be more characteristic of high POC/ low SOCCF
women than high SOCCF/low POC woven, as described in hypothesis 7A. A
Gui square was computed for this category, which was not significant, x2
(l, M = 160) = .54, p = 46. In contrast to what was expected, SOCCF
women were more represented here than Pm women, although the difference
was not significant.

Hypothesis 73 described the categories expected to be more
characteristic of high SOCCF/ low POC women than high POC/ low SOCCF
women: power from self-esteem or other positive feelings aboit the self,
and poer from perceiving choices about oneself/one's life. Independent
Chi squares computed for these variables were not significant: for power
from self-esteem, x2 (1, M = 160) = .36, p = .55, for power from
choice, x2 (1, M = 160) = .54, p = .45. As expected, soccr woven were
more represented than poc woven in power from choice, but the difference
was not significant. In contrast, POC woven were more represented in
poer from self-esteem than were SOCCF women, although the difference
was also not significant. Thus, these categories did not differentiate
' the two groups.

Both groips were expected to have similar experiences of poer from
relationships with others (subcategories combined), as relational ly-

 

87
oriented women, according to hypothesis 7c. POC women were more
represented in this category than SOCCF women, but a Chi square was not
significant, x2 (1, _1g = 160) = 1.16, p = .28. Both groups did
experience power from relationships about equally, though it is
interesting to note that the majority of responses from SOCCF women
mentioned power from respect or acknowledgement of covpetence in
relationship, while evotional acceptance from others and dependency were

mentioned only by members of the POC groip.

Wof Results Related to Hypotheses

 

The following hypotheses were formally stated at the onset. of this
study:

1. Depression was expected to be positively associated with
Primacy of Other Care (POC) , and negatively associated with Self and
Other Care Chosen Freely (SOCCF) .

Depression was positively related to POC, but that relationship was
not significant. Depression and SOCCF were independent, nonrelated.
Neither POC nor SOCCF were useful in predicting depression. The absence
of a stronger relationship between POC and depression was unexpected,
given that such a relationship had been found for POC and depression
using the entire sample in validity studies. Because this sample was
selected for moderate to high scores on the Relational/Connected Self,
this variable was held constant to see whether it influenced the
relationships between poc, EJCCF, and depression. The relationship
between pee and depression was not basically altered: this was also true
for the relationship between SOCCF and depression.

2. Self-esteem was expected to be negatively associated with POC,

and positively associated with SOGZF.

88

Self-esteem was significantly negatively related to POC, as
expected. The relationship between self-esteem and SOCCF was in the
expected direction, but was not significant. HOwever, the correlations
between self-esteem and POC and SOCCF were significantly different and
in the predicted direction, indicating that these two variables related
quite differently to self-esteem. The strength of the relationship
between self-esteem and POC made POC a predictor of self-esteem, with
SOCCF not contributing significantly to its prediction.

3. women whose relational identity was described by high POC/low
SOCCF were expected to be more vulnerable to experiencing depression
than women whose relational identity was described by high SOCCF/low
POC.

Comparison of the subgroups on depression was not significant,
indicating that the two groups'did not differ in terms of experiencing
depression. These results support the correlational findings showing no
relationship between depression and either relationship style.

4. High POC/low SOCCF women were also expected to have lower self-
esteem than high SOCCF/low POC women.

High POC/low SOCCF women did have lower self-esteem than high
SOCCF/low'POC women, but this difference did not reach significance at
the .05 level. A.similar difference between the higher selfeesteem of
high SOCCF/low’POC women and the lower self-esteem of high SOCCF/ high
Pm woven was significant, possibly because of differences in standard
deviations. Examination of the means for all four groups indicated that
I both IOW'POC groups, regardless of SOCCF status, had higher self-esteem
scores than the two high POC groups. The negative relationship between

POC and self-esteem noted in the correlational and regression analyses

89
may have been more important for self-esteem than the contribution of
high SOCCF scores.

5. The two 91‘pr of women were not expected to differ on a
measure of the frequency and intensity of their experience of anger.

As expected, the two groups were not significantly different on
this measure of anger. However, correlational analysis showed a
significant negative relationship between StDCF and anger; POC and anger
were basically independent. The correlations of anger with POC and
SOCCF were significantly different, indicating that these two variables
related differently to anger.

6A. The fol lowing types of responses to sentence completions
eliciting women's responses to their anger were expected to be more
characteristic of high POC/ low SOCCF women than high SOCCF/ low POC
woven: denial of anger, withdrawal and/or substitute activity, negative
judgements about anger or the angry self, sovatization, and the
transformation of anger to sadness/distress.

For both groups, withdrawal and transformation of anger to
sadness/distress were the second and third most frequently used
categorizations of anger responses, after behavioral expression of
anger. Overall, both grons of woven characterized their anger in quite
similar ways. Denial of anger and sovatization were used so
infrequently as to preclude statistical analysis. POC and SOCCF women
were not significantly differentiated by their use of withdrawal or
tranformation, althoigh PG: woven had a slightly higher percentage of
7 total responses in this category. POC women did make significantly more

negative judgevents about anger than did SOCCF woven.

90

6B. The fol lowing types of responses to anger were expected to be
more characteristic of high SOCCF/ low POC woven than high POC/ low M
women: rational means of problem-solving aboit anger, positive views of
anger or the angry self, and attempts at covmunication of feelings
towards resolution and reconciliation.

Neither group produced responses characterized as positive views of
anger. Comparisons of the other two categories indicated that POC and
SOCCF women were not significantly differentiated by their use of these
categories of responses to anger, although there were sore slight
differences. In terms of rational approaches, POC women used
introspective and interpersonal (tell others who are not part of the
conflict) means of problem—solving almost equally, while SOCCF woven had
more responses for introspective means (5) than interpersonal (l) .

SOCCF woven also had more covmunication/resolution attevpts with
partners (5) than did PCX: women (3).

6c . Both groips were expected to produce responses to anger
expressing concern about the impact of anger on relationships,
indicating a shared sense of responsibility for others characteristic of
relational women.

Both groups were represented equal ly in this category, as expected .
However, this was not a frequently used category of response for either
group, indicating that other concerns and more behavioral
characterizations of anger experience were predominant.

7A. Responses to a sentence covpletion aboit experiences of poer
that were coded as denial of being or feeling powerful or viewing being
poerful as a negative experience were expected to be more character-

istic of high Pat/low SOCCF women than high smCF/low POC women.

91

Overall , both groups of women characterized their experiences of
poer in similar ways. In contrast to what was expected for the
category of denial of power , SOCCF were more represented here (7) than
POC woven (3) , although the difference was not significant.

78. Poer from self-esteem or other positive feelings about the
self, and power from perceiving choices about oneself/one's life were
expected to be more characteristic of high SOCCF/ low POC women than high
POC/ low SOCCF women.

These categories did not significantly differentiate the groups as
hoped . SOCCF women were more represented than POC women in power from
choice, but POC women were more represented in poer from self-esteem,
although the differences were not significant.

7c. Both groips were expected to have similar experiences of power
from relationships with others as relationally-oriented woven.

The two groups were not significantly different in their
experiencing power from relationships. POC women were slightly more
represented in this category than SOCCF woven, and this category was
more frequently used for POC experiences of power than for SOCCF
experiences of power. It was interesting to note that POC women
experienced power in relationships primarily from evotional acceptance
and unidirectional dependency, while SOCCF woven experienced power
primarily from respect and acknowledgement of competence by others.

DISCUSSIQI

This research attempted to explore the psychological correlates of
different perspectives on the role of self and others in activities of
care, among woven who defined relational concerns as central to their
identity. As different manifestations of the relational/connected self,
these two perspectives on the meaning of care for self and others may be
useful in accounting for differences among women in areas of interest
for the psychology of woven: depression, self-esteem, anger, and poer.
The results of this study described the relationships between relational
identity style and depression, self-esteem, anger, and poer, and raised
interesting questions about how women perceive and respond to issues of
self- and other-care.

Hypothesized relationships between Primacy of Other Care (POC) and
Self and Other Care Giosen Freely (SOCCF) and the variables of
depression, self-esteem, and anger were only partially supported by the
results. These associations were reflected in later tests of the
performance of special subgroups of relational woven with respect to

trese same variables .

Depression

 

As measured here, neither relational perspective was significantly
related to depression. A significant positive relationship had been
‘ posited for POC and depression: a nonsignificant positive relationship
was found. A significant negative relationship between SOCCF and
92

93
depression was hypothesized: SOCCF and depression were basically
independent. Neither were important as predictors of depression. These
results were echoed in a comparison of women identified as special
subgroups of high POC/low SOCCF and high SOCCF/low POC fbr differences
on depression. Nb«differences were found between the groups of women,
and group means indicated that both groups acknowledged very infrequent
experiences of depressive symptoms.

The discovery of a significant negative relationship for POC and
depression in the total sample from which my study's subsample was drawn
raised a concern that the subsample selection criterion of scores
greater than or equal to 3.5 on the Relational/connected Self scale
might be influencing results. Perhaps being strongly relational served
a protective function against the vulnerability to depression noted for
POC in the total sample. This concern was explored and did not alter
the results significantly. Thus, the lack of predicted relationships
between relationship style and depression did not seem to be related to
endorsement of relational/connected self. However, it is possible that
ceiling effects due to the restricted range of scores on the CS scale
obscure existing relationships.

The lack of support for hypothesized relationships between
depression and relationship style suggests that these relational
perspectives may not be helpful in accounting fbr depressive
symptomatology in woven, at least women such as those in this sample.
It is possible that the very normalcy of this sample masked significant
’relationships between depression and relationship style because of
reduced variance in depression scores. This was a normal, nonclinical

population of female undergraduates and women functioning in the

94
community, similar to the general population samples on which the CES-D
scale was validated and with which it is intended to be used for
epidemiologic studies of depression. The mean, standard deviation and
distribution of scores of my sample were very similar to those noted for
the general population samples (Radloff, 1977). The distributions of
CES-D scores in these samples are typically very skewed, with small
standard deviations and large proportions of low scores. Psychiatric
patient samples, in contrast, produce symmetrical distributions with
large standard deviations, a pattern consistent with the interpretation
of the scale as related to a pathological condition more typical of a
patient population than a household sample (Radloff, 1977). Perhaps
hypothesized relationships between relationship style and depression as
measured by the CES-D scale would be better tested with a more depressed
sample.

Both relationship perspectives are described as normative
relational styles, different manifestations of the relational/connected
self, itself a nonpathological fOrm of identity. As such, perhaps the
distinctions about the role of self and others that differentiate the
two perspectives are not meaningful in accounting for depression as a
clinical problem for women. When Kaplan (1986) describes how extreme
forms of curtailment of women's normative developmental patterns may
create the intrapsychic conditions characterizing depression, she does
not suggest that these patterns themselves result in depression.

Perhaps there are pathological aspects of identity development
relevant fOr depression that are not captured by the descriptions of
relational style used here. There may be factors of developvvental loss,

genetic contributions, biochemical interactions, etc., that interact

95
with relational identity style to produce depression. Context or
environment may interact with relationship style as well. Perhaps my
sample were woven for whom POC status worked reasonably well and/or was
rewarded in their life context, leaving little basis for dissatisfaction
with it. Perhaps in a nontraditional population such as professional
women or single parents, where women struggle with multiple roles and
extensive devands on time and energy, relationships between depression

and relational style would be more evident.-

Self-esteem

 

The results provided somewhat more support for hypothesized
relationships between relational perspective and self-esteem. As
expected, a significant negative relationship between POC and self-
esteem was found. Putting others before self in activities of care was
related to low self-esteem. Self-esteem was expected to be positively
related to SOCCF; results indicated a nonsignificant relationship in the
predicted direction. Correlations of self-esteem with POC and with
SOCCF were significantly different and in the predicted direction,
indicating that POC and SOCCF related quite differently to self-esteem.

POC was a significant predictor of self-esteem, and SOCCF was not.
Comparing the groups of high POC/low SOCCF and high SCIECF/low PG: women»
on self-esteem indicated that group differences on self-esteem were in
the predicted direction but were not significant, with high PCE/ low
SGICF woven having lower self-esteem. Results concerning the underlying
' relationships between self-esteem and relationship style suggested that
high POC contributed more to the difference between groups than did high
SOCCF.

96

Perhaps relationship style had more meaning for self-esteem than
depression because both involved direct attitudes, views, and beliefs
about the self, sometimes in relation to others. Two of the ten items
on the self-esteem inventory involve contrasting the self with others
for evaluation. Items on the POC contrast attending to others' desires,
requests, reactions, and feelings*with giving up one's own desires,
neglecting personal values, and giving false impressions to preserve
relationships. What is more difficult to explain is the lack of
significant relationship between SOCCF and self-esteem. On the SOCCF
scale, hurting others, deciding to say no, and exercising choice about
activities of care is balanced with the need for self-care, and several
items stress responsibility for oneself. Self-care may result from
perceived and desired responsibility for self, rather than from self-

respect or self-esteem per se.

Agger
No specific hypotheses were made for relationships between POC and

SOCCF and anger measured by the EASI-III Anger scale, revised for this
research. INo differences were expected to occur between high POC/low
SOCCF and high SOCCF/low POC women on this measure of the frequency and
intensity of anger, because it was assumed that possible real
differences would not be revealed on a direct selfereport measure.
concerns with social desirability factors particularly salient with
respect tO‘women's anger were expected to mask differences related to
{POC and SOCCF. It was hoped that responses to the sentence completion
about anger could be examined for differences in the ways POC and SOCCF

women experienced and expressed their anger.

97

Results shoed no significant differences between the groups on the
self-report measure, with both group means below the midpoint of the
scale, indicating that neither groip endorsed frequent or intense anger
experiences. However, correlational results indicated a significant
negative relationship between SOCCF and anger; POC and anger were
independent. Thus, greater endorsevent of balanced self and other care
experienced as choice was associated with less frequent and intense
reported experiences of anger. Correlations of anger with POC and with
SOCCF were significantly different, indicating that these variables
related differently to anger.

The sentence completion "men I get ANGRY I _" allowed
respondents to describe their reactions to anger in their own words
throigh the open-ended format. Their responses were then coded using
categories created by the research groip. These categories were created
to reflect the phenorenology of responses to anger and incorporated
theoretical and clinical perspectives on the dynamics of anger.
Hypotheses identified the categories expected to be more characteristic
of high POC/low SOCCF women, those expected to characterize the
responses of high soccvv low POC women, and the category in which both
groups might be equal ly represented . Categories were analyzed
separately because responses could received multiple coding.

Most of the hypotheses positing differential anger responses
according to relational perspective were not confirmed or supported by
the data, except the hypothesis predicting that negative judgements of
anger would be more characteristic of high POC/ low SOCCF women than high
SOCCF/ low POC woven. The hypothesis positing no differences in the use

of the category focusing on the impact of anger on relationships was

98
also confirmed. Relational style did not differentiate types of anger
responses. Woven with different care orientations responded to anger
quite similarly. The overall frequencies can be discussed as a
descriptive account of the ways relational woven express and experience
anger.

The most frequently given responses for both groips fell in the
category of behavioral expressions of anger , and included physical ,
verbal and nonspecified modes of expressions, with verbal expressions
clearly most common. This category was conceptualized as a cathartic,
nonrational venting of angry feelings, and was not hypothesized to
relate to either relational perspective particularly. Next most common
were responses categorized as withdrawal, which included attempts to
hold anger in, attevpts to displace angry tension into substitutive
activity, and attempts at impulse control. The third most covmon means
of expressing anger was throigh its transformation into expressions of
sadness/distress, including crying and feeling hurt or depressed.
(Negative views of anger shared third most commonly used category with
transformation of anger for high POC/ low SOCCF woven.)

It is interesting to note that these findings are quite similar to
the results of another study of the frequency of anger in woven in a
contevporary, ”normal" population (Dickstein, Bell, & Lin, 1980). In
this study of 2078 women, responses to an interview question ("What do
yoidoorhowdoyoishowitwhenyoiareangryr) weregrouped into
categories similar to those used in this research. The most frequently
- given responses involved verbal-facial expressions, including
nonrational verbal expressions and nonverbal acting oit, such as

scowling or pouting. The second most covmon response was withdrawal .

99
(The study by Dickstein et a1. had a separate category for nondirected
physical activity corresponding to the substitute activity included with
other withdrawal attempts in my research.) Crying was the third most
frequent response. Least frequent responses were physical syvvptoms and
aggressive behavior towards persons, which also correspond to the least
frequently given responses in my research. This research also suggested
that the most frequent forms of anger expression were distinct modes of
expression, such that woven seeved to use one or the other of these
modes rather than a combination.

Holding anger inside and crying have been considered
stereotypical ly "appropriate" ways of dealing with angry feelings for
women, and were hypothesized to be more characteristic of POC than SOCCF
women. However, both groups produced similar responses of withdrawal
and transformation.

The next most frequent response by both groups combined was
negative views of anger or the angry self. Responses coded here rarely
described how anger was expressed, but instead communicated the
respondents' judgements of dislike and disapproval of being angry or of
themselves when angry. This category was expected to be used more
frequently by POC woven, who did in fact produce significantly more
negative views (16, 16%) than SOCCF woven (9, 8%).

Rational means of problem-solving about anger, nonscorable
responses, covmunication/ resolution attevpts, and focus on the impact of
anger/others hurt were categories used with decreasing frequency by the
women. Rational means of problem-solving, including talking with others
not part of the conflict, and covmunication of feelings to the partner

in conflict towards resolution were expected to characterize SOCCF women

100
more than POC, but did not significantly. SOCCF women did make slightly
more communication attempts (5, 4%) compared to POC (3, 3%), and made
more introspective problem-solving responses than interpersonal. POC
women made almost equal use of both introspective and interpersonal
means.

Both groups produced very few responses of denial of anger,
reactive aggression, and somatization. These could be considered the
most pathological categories: overt denial_of ever being angry seemed to
be a very defensive response, and reactive aggression specified intended
aggression towards persons. Acknowledgement of symptoms as a response
to anger required psychological sophistication and conscious awareness
of a psychological defense that is usually unconscious. Denial of anger
and somatization.were expected to characterize POC women more than
SOCCF, but both groups produced only a very few responses in these

categories.

Power

 

The sentence completion "When I feel POWERFUL it is because _____"
was included in this research as an initial exploration of the ways
women describe their experiences of power in their own words.

Hypotheses were made about how POC and SOCCF women might differentially
characterize their sources and experiences of power. Again, these
hypotheses were not supported by the sentence completion data. Only the
hypothesis of no differences in experiencing power from relationships
was supported. women with different care orientation claimed similar
sources of personal power. These sources can be described for a

phenomenological account of power for relational women.

101

Power from achievement was the most common source for both groups.
This category, along with power from right and power from position, had
been conceptualized as experiences of power reflective of a
separate/objective orientation, not a relational orientation. Thus,
none of these three were expected to be important to relational women
and were not included in the hypotheses. (The other two categories were
in fact infrequently used by both groups of women.) Achievement as a
source of power may reflect the dominant cultural emphasis on and
valuing of agenetic achievement. It.may'also reflect a shifting
emphasis by women towards experiences that have been denied them, made
newly important as potential experiences of competency apart from
relational strengths.

Also unexpected was the frequency of responses of power from being
"in control". These responses indicate that "control" is somehow
importantly related to power, but it is not clear what this meant to
respondents. Responses dealing with control over oneself and one's
decisions and choices were coded under choice, and responses dealing
‘with control over others were coded under power from position, leaving
unclear, nonelaborated statements of control in a separate category that
was theoretically ambiguous.

Power from relationships, power from self-esteem, and impersonal or
passive power were the next most common sources for both.groups
combined. Although the differences were not significant, POC women
produced slightly more responses of deriving power from relationships
(12, 14%) than SOCCF women (8, 8%), giving responses based primarily on
emotional acceptance and dependency. SOCCF responses, in contrast, were

based primarily on respect and acknowledgement of competence.

102

Impersonal or passive power, power from self-esteem, power from
predictability, power from choice, nonscorable responses, and power from
position were used in descending order for the groups combined, with
slightly varying order by relational perspective. Power from self-
esteem and power from choice were expected to characterize SOCCF women
more than POC women. SOCCF women did use choice slightly more (7, 7%)
than POC women (3, 3%), but POC women gave more responses involving
self-esteem or other feelings about the self (11, .13%) than did SOCCF
women (9, .09%).

Least frequently used categories were denial of power, power from
moral certainty or right, and power from balancing multiple dimensions.
Denial of power was expected to characterize POC women (3, 3%) but
instead was more typical of SOCCF women (7, 7%), although not

significantly so.

Relationshipgiamong Depression, Self-esteem, and Anger

 

Significant relationships were found among depression, self-esteem,
and anger as measured by the EASI-III Anger scale. Depression.was
negatively related with self-esteem and positively related to anger.
Self-esteem was also negatively related to anger. These findings are
similar to clinical descriptions of some types of depressions where
anger and frustration are acknowledged and experienced rather than
internalized. These relationships were not the fbcus of this study,
which explored the relationships of these variables to two relational
perspectives on self and others, but do provide validation fOr these

measures of depression, self-esteem and anger.

103

Measurement Considerations

 

The focus of this research was on the use of the Revised
Relationship Self Inventory (RRSI) to identify woven who were similar in
their descriptions of themselves as relational/connected selves, but who
differed in their understanding of the role of self and other in
activities of care, as reflected in their differential endorsevents of
the Primacy of Other Care (POC) and Self and Other Care Giosen Freely
(SOCCF) scales. These scales measured different manifestations of the
relational/connected self corresponding to Gilligan's (1982) final two
developmental perspectives on the meaning of care for self and others.
The results of this study were sovewhat disappointing in that these
relational perspectives did not relate as strongly to other variables as
predicted, although there were different patterns of relationship that
distinguished the two perspectives from each other, particularly in
terms of self-esteem and anger. This raises questions about the RRSI's
validity. A discussion of the three RRSI scales used in this research
may identify some important measurevent concerns .

The RRSI scales use a five-point scale ranging from 1 = "Not like
me at all” to 5 = "Very much like me", with an unlabeled midpoint of 3.
In order to identify woven who described thevselves as relational , the
selection criterion for my subsample was scale-defined, using the anchor
points of the appropriate scale: mean scores greater than or equal to
3.5 on the Relational/Connected Self scale, excluding mean scores
falling around the midpoint of the scales. However, the mean CS score
for fevales in the total sample (M = 930) was 4.1, with standard
deviation of .51 . This indicates that most women were describing

themselves as strongly relational. The mean score on the SOCCF scale

104
was also quite high: 3.9, with standard deviation of .50. For my
subsample (_r_v_ = 323), CS mean score was 4.23 and SOCCF 3.91. Such high
mean scores indicated restricted ranges and low variability on these
scales, which likely influenced the results due to ceiling effects.

The initial plan for creating "refined types" of relationship
styles using scores on POC and SOCCF was to consider scores greater than
or equal 3.5 as ”high” and scores less than or equal to 2.5 as "low" on
the two scales, using scale anchor points and meanings. Because of the
limited range of scores, particularly on SOCCF, where the minimum SOCCF
score was 2.6, this plan was revised. There were no ”low" SOCCF scores
using this criterion. The median split procedure was chosen, although
the groups forvved in this way becove sample-dependent. Definitions of
the groups as high POC/low SOCCF and high SCECF/low POC becove somewhat
arbitrary, because individuals in the high POC/ low SOCCF group included
those endorsing the SOCCF scale as self-descriptive, using the anchor
points of the scale itself. The high means and small standard
deviations of these scales indicate that women were not differentiating
themselves in their use of the scales.

Anotl'er significant problem concerns the ways respondents endorsed
POC and SOCCF scales simultaneously, which challenges Gilligan's
developmental sequencing of these two perspectives on self and others .
Both scales were significantly positively correlated with CS, as is
appropriate for scales measuring different manifestations of the
Relational/Connected Self. The scales were weakly positively correlated
with each other. The median split method used to create “pure types”
for analyses in this study produced for groups: high POC/low SOCCF,
high SOCCF/low POC, low POC/low SOCCF, and high POC/high SOCCF.

105
(Because of the limited range of scores on SOCCF, designations of "low
SOCCF" remain sovewhat arbitrary.)

The relative independence of the scales would be expected to
produce the foirdway distribution foind. Theoretically, hoever, the
manifestations of relational identity that POC and SOCCF were created to
measure involve conflicting views about the self and others. Based on
these differences, POC and SOCCF might be expected to be negatively
correlated. Given our underlying conceptualizations of these two
perspectives, it is difficult to know how to interpret the endorsevents
defining the last two groups of low POC/ low SOCCF and high POC/high
SOCCF woven.

While it was not the goal of this research to explain or describe
these last two groups theoretically, it is important to note the
problems they posed for the theory of the relational self and these
measures of it. For the low SOCCF/ low POC group, if neither scale was
particularly descriptive of their relational style, but they were
describing themselves as relational Iconnected, perhaps POC and SOCCF are
not the only manifestations of relational orientation. The groip
scoring high on both POC and SOCCF were particularly problematic because
they endorsed two different positions on the role of self and other in
activities of care .

E:xamining how these two groips compared with the subgroips of
primary interest (high POC/ low SOCCF and high SOCCF/ low PCXI) on
depression, self-esteem, and anger does not seem to offer greater
theoretical clarity. The low POC/ low SOCCF group was very similar to
the two ”refined types” on depression, had a mean score on self-esteem

between the means for the POC and SOCCF groups (with higher self-esteem

106
than either of the two groips high on POC) , arid had a mean score for
anger that was similar to that of the high POC/high SOCCF group, with
both higher than the means for the POC and SOCCF groups. The high
POC/high SOCCF groip was the most depressed, shared loer self-esteem
with the high POC/low SCDCF group, and had similar anger as the low
POC/ low SOCCF group, perhaps related to the conflict between both views
of self and other care. However, differences among the means were not
very large, with only differences in self-esteem being significant.

The group of women scoring high on both scales was surprising large
and points to underlying assumptions about the relationship between POC
and SOCCF. An examination of items from the scales reveals items that
seem to clearly reflect conflicting perspectives that would be difficult
to endorse simultaneously. For example, contrast ”The best way to help
soveone is to do what they ask even if you don't really want to do it"
(POC) , with ”If soveone asks me for a favor I have a responsibility to
think about whether or not I want to do the favor" (SOCCF) . Other items
contrast neglecting one's own values in decision-making to keep a
relationship (POC) , with the importance of using one's own values to
make choices (SOCCF) . Theoretically, these manifestations of relational
identity are distinguished by a significant shift in the understanding
of the role of self and others in relationships, from a position that
self and other care are opposed, with caring for others taking
precedence over self care (POC) , to a position where self and other care
are seen as covplementary and freely chosen (SOCCF) . Perhaps woven
' endorsing both positions simultaneously are expressing conflict over
making such a shift in understanding or are in developmental process of
making that shift.

107

Because these perspectives represent conflicting views of the
relationship between self and others, measures of these relational
perspectives might be expected to be negatively correlated. Hypotheses
linking these perspectives to other variables such as depression and
self-esteem posited relationships in opposite directions and described
expected differences according to differentiating perspectives.
chever, POC and SOCCF were not negatively correlated, and were both
endorsed simultaneously by some women. ‘What might explain how*women
perceived and responded to these scales?

One possibility is that respondents may have seen the central issue
of the role of self and others in relationships along a continuum, with
concerns for self care at one endpoint, balanced with concerns for
caring for others at the other. POC items, with their stress on others,
may have represented one pole, with SOCCF items representing the
alternative focus on self. Individuals may have endorsed items from
both scales, without making the self-other distinctions intended to
differentiate the scales, as they moved back and forth between both
concerns, attempting to respond to both. Another possibility is that
scale items about how relationships are handled were considered and
responded to from such varied relational contexts that distinctions were
lost. It may be that.women considered some items in the context of
family relationships, others in terms of friendships, work settings,
‘with peers, with nonpeers, etc., and that imagined context influenced
responses. women may also be in transition, endorsing conflicting
positions as they sort out issues of self and others.

As a self-report instrument, the RRSI scales may also be unable to

accurately tap consistent patterns of responses of care. For more

108
traditional woven, the POC position may be viewed as socially desirable;
women with feminist views may aspire to the SOCCF perspective. It may
be difficult for woven to accurately report how they handle the claims
of the self and of others for care, not because it is a trivial concern
or dimension, but because it is such a central concern. If these issues
are central in woven's identity, as I think they are, self-report itevs
intended to measure them tap only the conscious dimension of self-
awareness. If these views of self and other are fundamental aspects of
identity on both consciois and unconscious levels, it may be difficult
to isolate and identify the perspectives or "screens” through which
experiences are filtered when responding to a short set of items. Women
may not be readily able to accurately describe their behavior and
articulate their basic beliefs aboit responses of care, particularly if
they experience conscious or unconscious conflict about their beliefs or
actions . They may respond by describing thevselves in terms they
consider socially desirable or that support idealized images of
themselves acting in response to self and others.

These results suggest that more work is needed to understand how
these scales worked and how they might be improved as measures of
relational perspectives. Other analyses using the RRSI and other
measures for convergent and divergent validity may offer helpful

information.

Other Measurevent Issues

 

Relational perspective did not differentiate types of anger and
poer responses on sentence covpletions . This may have been due to the

measurement problems just discussed for the RRSI scales. Sentence

109
completions were also single items, and as such produced limited
information.

For the sentence completion data with anger and power, it is also
not clear how much coding error existed. Interrater reliabilities as
calculated were lower than expected or desired. The reported
reliabilities for coder pairs with a research group member as an
external standard were calculated several months after coding was
completed. Although the researdh group member's initial agreement with
my coding was quite high during the training of the coders, the
intervention of time spent on other projects and lack of continued
practice with the codes may have influenced her coding of items for

reliability calculations, lowering agreement.

§EEEEEXL
Adthough this study did not confirm some of the relationships

described as a test and expansion of the model of the relational self,
it did illuminate problems of measurement that remain to be resolved.
The relationships between perspectives on the meaning of care for self
and others and basic issues of depression, self-esteem, anger and power
remain intriguing and vital concerns fer clinical work with women, and
hopefully for continued empirical articulation and investigation as

well.

APPENDICES

APPENDIX A

RESEARCH-{WW

110

June 1985
Dear College Week Participant:

Daring College Week 1984, a research project took place that asked many
of the participants how relationships were a part of their identity. The
results were very clear. Last year's College Week women were consistent-
ly a "caring” group, who highly valued being interconnected with others.
We found these results so important that Jare Pearson is currently
teaching a course for this year's College Week that describes, discusses
and expands upon our research. It is titled, "Women's Development: The
Importance of Empathy, Care and Relationships." We also presented the
results of the research at a psychology conference, and have found a
high level of interest in this topic from other social scientists.

We are continuing this line of study, and hope that you will decide to
participate. This year, we are asking that all College week participants
take an hour or so to sit down and complete the enclosed questionnaire
packet. Please note that your participation is purely voluntary. If you
participated in last year's study and are interested in participating
this year, we would greatly appreciate having your input again. You'll
notice that your answers are confidential-—do not put your name or other
identifying information on the questionnaire itself. While we hope that
you will complete the entire packet, you are free to stop at any point
if you choose to do 30.

After you have completed the packet, return it, and the pencil, in the
envelope to your main dorm desk, where there will be a box labeled
”Relationship Identity Project," and simply leave it in the box. If you
choose to complete only part of the packet, or none at all, please
return your packet in the envelope as well.

If you would like to receive a brief summary of this year's project
outcomes, please fill out your name and address on the last page of the
packet, and return it to the same box seflate from your envelope
containing your packet.

To be sure that you understand your rights as a research participant,
read through the following statements:

I freely consent to take part in the study of relationship identity
being conducted under the supervision of Dr. Ellen Strommen, Professor,
Department of Psychology, Michigan State University. I understand that
the study deals with relationships in people's lives; I have been given
a clear explanation of my part in this work, which is to complete a
questionnaire.

I understand that I am free to discontinue my participation in the study
at any time without penalty.

I understand that tl'e results of the study will be treated in strict
confidentiality and that I will remain anonymous. Within these

111

restrictions, group results of the study will be made available to me at
my request. '

I understand that my participation in the study does not guarantee any
beneficial results to me.

I understand that, at my request, I can receive additional explanation
of the study after my participation is completed.

I understand that my compliance in completing the questionnaire
constitutes my informed consent for participation in the study.

If you agree with the statements, then go ahead and begin work on the
questionnaire. Thank you for your assistance in completing the
questionnaire.

Sincerely,

Ellen A. Strommen
Professor

APPENDIX B

IIBTRUMENI‘S

PLEASE NOTE:

Copyrighted materials in this document have
not been filmed at the request of the author.
They are available for consultation. however.
in the author's university library.

These consist of pages:

112-121

122-132

U-M-I

112

Relational Self Inventory (RSI)

 

Instructions: Reach each statement below and decide how much it
descfibes you. Using the following rating scale, select the most
appropriate response and blacken the corresponding circle on the blue
answer sheet.

 

Not like Very much
me at all like me
1 2 3 4 5

1. I believe I must care for myself because others are not concerned
with my needs.

2. When I help someone I feel good because I've done my duty.
3. Mat is right is right.

4. It's worse for me to be a failure in my chosen vocation than to have
no one with whom to share my life.

5. Activities of care that I perform expand both me and others.
6. Caring about other people is important to me.

7. True responsibility involves making sure my needs are cared for as
well as the needs of others.

8. I enjoy taking care of my own health.
9. Love is an activity, not something you have.

10. I believe that in order to survive I must concentrate more on taking
care of myself than on taking care of others.

11. I try not to think about the feelings of otters when there is a
principle at stake.

12. It's hard for me to tell others how much I care about them.
13. Doing things for others makes me happy.

14. Sometimes Ihavetoaccepthurtingsomeoneelse ifIamtodothe
things that are important in my own life.

15. If other people are going to sacrifice something they want for my
sake I want them to understand what they are doing.

16.Iwanttolearntostandonmycwntwofeet.

17. All you really need to do to help someone is to love them.

113

Not like very much
me at all like me
1 2 3 4 5

18. I can feel confident in myself even when I do not have the approval
of those who are close to me.

19. If someone does something for me, I reciprocate by doing something
for them.

20. I like to acquire many acquaintances and friends.

21. In choosing a vocation, helping others is more important to me than

22.

23.

24.
25.

26.
27.
28.
29.

30.

31.

32.
33.

34.

35.

36.

money, prestige, or personal challenge.

I cannot always do what my loved ones want, if it causes me to make
a sacrifice.

When I am feeling "needy," I am comfortable asking others to help
out rather than doing it all myself.

In a close relationship you nearly always give up more than you get.

When I make a decision it's important to use my own values to make
the right choice.

I expect others to treat me as I treat them.
I like competing with others.
Relationships are a central part of my identity.

If someone offers to do something for me, I should accept the offer
even if I really want something else.

Even though I am sensitive to others' feelings, I make decisions
based upon what I feel is best for me.

I feel better when I choose to do a favor than when I think I am
expected to.

The feelings of others are not relevant when deciding what is right.

I try to approach relationshipS‘with the same organization and
efficiency as I approach my work.

Those about.whom I care deeply are part of who I am.

I cannot choose to help someone else if it will hinder my self-
development.

The worst thing that could happen in a friendship*would be to have
my friend reject me.

114

th like very much
me at all like me

1 2 3 4 5
37. I do not want others to be responsible for me.

38.

39.

40.

41.

42.
43.

44.

45.
46.
47.

48.

49.

50.

51.

52.

53.

54.

55.

I no longer think this way, but I used to believe that the greatest
good is selfesacrifice.

It is necessary for me to take responsibility for the effect my
actions have on others.

I cannot afford to give attention to the opinions of others when I
am certain I am correct.

Loving is like a contract: If its provisions aren't met, you
wouldn't love the person any more.

Being unselfish with others is a way I make myself happy.
I deserve the love of others as much as they deserve my love.

If someone asks me fOr a favor I have a responsibility to think
about whether or not I want to do the favor.

Sometimes others do for me what I want to be able to do for myself.
I feel empty if I'm not closely involved with someone else.

When a friend traps me with demands and negotiation has not worked,
I am likely to end the friendship.

I find it hard to sympathize with people whose misfortunes I believe
are due mainly to their own shortcomings.

I like to see myself as interconnected with a network of friends.

I often try to act on the belief that self-interest is one of the
worst problems facing society.

Sometimes I think I do too much for others and not enough for
myself.

Those who are strong and happy deserve my care as much as those who
are needy.

I believe that I must care for myself because others are not
responsible for me.

The people whom I admire are those who seem to be in close personal
relationships.

I make decisions based upon what I believe is best for me and mine.

115

Not like Very much
me at all like me
1 2 3 4 S
56. In my every day life I am guided by the notion of "an eye for an eye

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.
68.
69.

70.

71.

72.

and a tooth for a tooth.”

I believe that one of the most important things that parents can
teach their children is how to cooperate and live in harmony with
others.

EVen though it's difficult, I have learred to say no to others when
I need to take care of myself.

Sometimes a good way to support others is to tell them of your own
faults and problems.

In order to continue a relationship it has to let both of us grow.

The best way to help someone is to do what they ask even if you
don't really want to do it.

I often tell people what to do when they are having trouble making a
decision.

Being unselfish with others is more important than making myself
happy.

If I am to help another person it is important to me to understand
my own motives.

I want to be responsible for myself.

I feel that my development has been shaped more by the persons I
care about than by what I do and accomplish.

I accept my obligations and expect others to do the same.
I try to curb my anger for fear of hurting otters.

I no longer think this way, but I used to believe that true
responsibility is the same as caring for others, even if it means
less care for myself.

If I knewIweretodiewithintheyear, Iwouldbemoreconcerred
for my loved ones than for my unfinished occupational goals.

Sometimes a good way to give to others is to tell them what you need
for yourself.

What it all boils down to is that the only person I can rely on is
myself.

116

Not like Very much
me at all like me
1 2 3 4 5

73. I would never compromise something I truly believe in.

74. To sustain a relationship I play many roles.

75. I don't feel very pleased with myself if I help someone
”automatically” without thinking of what I'm doing.

76. (hoe I've worked out my position on some issue I stick to it.

77. In making decisions, I can neglect my own values in order to keep a
relationship.

78. When I am feeling “needy,” I think I have the right to ask others to
help out rather than doing it all myself.

79. You've got to look out for yourself or the demands of circumstances
and other people will eat you up.

80. men dealing with a tough situation, my first concern is to be fair.

81. To keep relationships going, I often tell others I care more about
them than I really do.

82. Men important changes are going on in my life, I like to retreat
into myself for awhile.

83. I believe that I have to look out for myself and mine, and let
others shift for themselves.

84. If what I want to do upsets other people, I try to think again to
see if I really want to do it.

85. Before I can be sure I really care for someone I have to know my
true feelings and reasons.

86. Being your own person is doing whatever you want, as long as you do
not step on other people's rights or wants.

87. I will not let otl'ers help me unless I can do tl'e same thing for
them.

88. My own personal achievements are rarely important enough to justify
causing hurt and pain to others.

89. If I am really sure that what I want to do is right, I do it even if
it upsets other people.

90. To really help someone, it is as important to know them and their

desires as it is to love them.

117

Not like Very much
me at all like me
1 2 3 4 5

91. I am guided by the principle of treating others as I want to be
treated.

92. I often keep quiet rather than hurt soreone's feeling, even if it
means giving a false impression.

93. A close friend is soreone who will help you whenever you need help
and knows you will help if they need it.

94. I don't often do much for others unless they can do sore good for me
later on.

95. People who don't work hard to accomplish respectable goals can't
expect me to help when they're in trouble.

118

(CESPD) Center for Epidemiological Studies Depression Scale

 

Below is a list of some of the ways you may have felt or behaved. By
using the same small red answer sheet, indicate how often you have felt
this way during the past week by darkening the appropriately numbered
circle. USe the following scale to make your responses.

Rarely or ane Some or a Little Occasionally or Most or All

of the Time of the Time a Moderate Amount of the Time

(less than 1 day) (1-2 days) of time (3-4 days) (5—7 days)
1 2 3 4

l. I was bothered by things that don't usually bother me.
2. I did not feel like eating: my appetite was poor.

3. I felt that I could not shake off the blues even with help from my
family and friends.

4. I felt that I was just as good as other people.
5. I had trouble keeping my mind on.what I was doing.
6. I felt depressed.

7. I felt that everything I did was an effort.

8. I felt hopeful about the future.

9. I thought my life had been a failure.

10. I felt fearful.

11. My sleep was restless.

12. I was happy.

13. I talked less than usual.

14. I felt lonely.

15. People were unfriendly.

16. I enjoyed life.

17. I had crying spells.

18. I felt sad.

19. I felt that people disliked me.

20. I could not get "going."

119
Rosenberg Self-Esteem Inventory

Instructions: Read each statement below and decide how much it
describes you. Using the following rating scales, select the most
appropriate response and blacken the corresponding circle on the same
blue answer sheet.

 

Strongly Agree Disagree Strongly
Agree Disagree
1 2 3 4

1. I feel I'm a person of worth, at least on an equal plane with
others.

2. I feel that I have a number of good qualities.

3. All in all, I am inclined to feel I am a failure.

4. I am able to do things as well as most other people.
5. I feel I do not have much to be proud of.

6. I take a positive attitude toward myself.

7. On the whole, I am satisfied with myself.

8. I wish I could have more respect for myself.

9. I certainly feel useless at times.

10. At times I think I am no good at all.

120
EASI-III Temperament Survey

Instructions: Rate each of the following statements by deciding whether
it is not at all like you see yourself, very much like you see yourself,
or somewhere in between. Using the same small answer sheet, blacken the
circle corresponding to your response. USe the following scale to make

your responses.

 

‘Not at All Somewhat very Much
Like Me like Me like Me
1 2 3 4 5

1. I make friends very quickly.

2. I am very sociable.

3. I tend to be shy.

4. I usually prefer to do things alone.

5. I have many friends.

6. There are many things that annoy me.
7. I frequently get angry.
8. It takes a lot to get me mad.

9. I get angry more often than most people my age.

10. I am easily frightened.

11. I often feel insecure.

12. I tend to be nervous in new situations.

13. I have fewer fears than most people my age.
14. When I get scared I panic.

121

Sentence Completion

 

The following are sentence completion items. Please complete each
statement in the space provided .

1. To me RESPONSIBILITY means

 

 

2. To me SELFISH means

 

 

3. To me REEIPROCITY means

 

4. I feel IN C(NI‘ROL when

 

 

5. IfeelOUTOFCCNTROLWhen

 

 

6. IHURTOTHERSwhen

 

 

7. WhenIGETANGRYI

 

 

8. MienIFEELPmERFULitisbecause

 

 

9. TomeDEPENDENCYmeans

 

 

APPH‘IDIX C
REVISED RELATICNSHIP SELF INVENTORY
RRSI RELIABILITIES AND SCALE WTICNS
RRSI ITEM-SCALE 'IUI‘AL (IRRELATICNS

122

Revised Relationship Self Inventory (RSI)

 

Instructions: Read each statement below and decided how much it
déscribes you. Using the following rating scale, select the most
appropriate response and blacken the corresponding circle on your answer
sheet.

 

Not Like Very Much
Me at All Like Me
1 2 3 4 5

1. I often try to act on tle belief that self—interest is one of the
worst problems facing society.

2. A close friend is soreone who will help you whenever you need help
and knows that you will help if they need it.

3. I cannot choose to help soreone else if it will hinder my self-
development.

4. I want to be responsible for myself.

5. In making decisions, I can neglect my own values in order to keep a
relationship.

6. I find it hard to sympathize with people whose misfortunes I believe
are due mainly to their shortcomings.

7. I try to curb my anger for fear of hurting others.

8. Being unselfish with otl'ers is more important than making myself
happy.

9. loving is like a contract: If its provisions aren't met, you
wouldn't love He person anymore.

10. In my everyday life I am guided by the notion of "an eye for an eye
and a tooth for a tooth.”

11.Iwanttolearntostandonmyowntwofeet.

12. I believe that one of the most important things that parents can '
teach their children is how to cooperate and live in harmony with
others. .

13. Itrynottothinkabouttlefeelingsofotrerswhentlere isa
principle at stake.

14. I don't often do much for others unless they can do sore good for me
later on.

15. Activities of care that I perform expand both me and others.

123

Not Like Very Much
Me at All Like Me

1 2 3 4 S
16. If what I want to do upsets other people, I try to think again to

17.

18.

19.

20.

21.

22.
23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

see if I really want to do it.
I do not want others to be responsible for me.

I am guided by the principle of treating others as I want to be
treated.

I believe that I have to look out for myself and mine, and let
others shift for themselves.

Being unselfish with others is a way I make myself happy.

When a friend traps me with demands and negotiation has not worked,
I am likely to end the friendship.

I feel empty if I'm not loosely involved with soreone else.

Sometimes I have to accept hurting someone else if I am to do the
things that are important in my own life.

In order to continue a relationship it has to let both of us grow.

I feel that my developrrent has been shaped more by the persons I can
about than by what I do and accomplish.

People who don't work hard to accomplish respectable goals can't
expect me to help when they're in trouble.

Relationships are a central part of my identity.

I often keep quiet rather than hurt soreone's feelings, even if it
means giving a false impression.

If someone offers to do something for me, I should accept the offer
even if I really want sorething else.

The worst thing that could happen in a friendship would be to have
my friend reject me.

If I am really sure tlat what I want to do is right, I do it even if
it upsets other people.

Before I can be sure I really care for sorecne I have to know my
true feelings.

What it all boils down to is that the only person I can rely on is
myself.

124

th Like very Much
Me at All Like Me

1 2 3 4 5
34. Even though I am sensitive to others' feelings, I make decisions

35.

36.
37.
38.
39.

40.

41.

42.

43.
44.

45.
46.
47.

48.

49.
50.
51.

52.

based upon.what I feel is best for me.

Even though it's difficult, I have learned to say no to others when
I need to take care of myself.

I like to see myself as interconnected with a network of friends.
Those about whom I care deeply are part of who I am.
I accept my obligations and expect others to do the same.

I believe that I must care for myself because others are not
responsible.

The people whom I admire are those who seem to be in close personal
relationships.

It is necessary for me to take responsibility for the effect my
actions have on others.

True responsibility involves making sure my needs are cared for as
well as the needs of others.

The feelings of others are not relevant when deciding what is right.

If someone asks me for a favor I have a responsibility to think
about whether or not I want to do the favor.

I make decisions based upon what I believe is best for me and mine.
Once I've worked out my position on some issue I stick to it.

I believe that in order to survive I must concentrate more on taking
care of myself than on taking care of others.

The best way to helprsomeone is to do what they ask even if you
don't really want to do it.

Doing things for others makes me happy.
Allyoureallyreedtodotol'elpsoreoreistolovethem.
I deserve the love of others as much as they deserve my love.

Ybu've got to look out for yourself or the demands of circumstances
and of other people will eat you up.

125

Not Like Very Much

Me at All Like Me
1 2 3 4 5

53. I cannot afford to give attention to the opinions of others when I
am certain I am correct.

54. If someone does sorething for me, I reciprocate by doing something
for them.

55. Caring about other people is important to me.

56. If other people are going to sacrifice something they want for my
sake I want them to understand what they are doing.

57. When I make a decision it's important to use my own values to make
tre right education.

58. I try to approach relationships with the same organization and
efficiency as I approach my work.

59. If I am to help another person it is important to me to understand
my own motives.

60. I like to acquire many acquaintances and friends.

Table C1.

Wbmen
(N = 930)

Men
(N = 228)

Table C2.

Separate/
Objective
Self

Connected/
Relational
Self

Primacy of
Other Care

Self and
Other Care
Chosen Freely

126

Revised Relationship Self Inventory

Reliabilities and Scale Intercorrelations

Separate/
Objective

Self

.77

.85

Separate/
Objective
Self

1.00":1

-.33

-001

.26

Reliabilities (alpha)

connected/
Relational
Self

.76

.76

Scale Intercorrelations*

Connected/
Relational
Self

-023

1.00

.58

 

Primacy of
Other Care

.68

.67

Primacy of
Other Care

.09

.56

1.00

.19

Self and
Other Care
Chosen Freely

Self and
Other Care
Chosen Freely

.52

.10

1.00

* Intercorrelations for women above the diagonal: intercorrelations for
men below the diagonal.

aCorrected for attenuation.

127
Table C3. Item-Scale Total Cbrrelations and Scale Reliabilities of the

Revised Relationship Self Inventgry

 

Wbmen Men
Separate/Objective Self

Item-total Scale Item—total Scale
Item Correlationa Alpha Correlationa Alpha

I believe that in order to

survive I must concentrate

more on taking care of myself

than on taking care of others. .50 .77 .49 .85

I try not to think about the
feelings of others when there
is a principle at stake. .36 .37

Even though I am sensitive to

others' feelings, I make

decisions based upon what I

feel is best for me. .31 .41

The feelings of others are not
relevant when deciding what is
right. .39 .47

I try to approach relationships
‘with the same organization and
efficiency as I approach my work. .21 .36

I cannot chose to helprsomeone
else if it will hinder my self-
develcpment. .50 .58

I cannot afferd to give attention
to the opinions of others when I
am certain I am correct. .45 .59

loving is like a contract: If

its provisions aren't.met4 you

wouldn't love the person any

more. .41 .36

When a friend traps me with

demands and negotiation has not

worked, I am likely to end the

relationship. .32 .36

(table continues)

128

Item-total Scale Itemrtotal Scale
Item Correlationa Alpha Correlationa Alpha

I find it hard to sympathize with
people whose misfortunes I believe
are due mainly to their own

shortcomings. .43 .53

I make decisions based upon what
I believe is best for me and
mine. .32 .46

In my everyday life I am guided
by the notion of "an eye fOr an
eye and a tooth for a tooth.“ .43 _ .62

What it all boils down to is
that the only person I can rely
on is myself. .40 .48

Once I've worked out my position
on some issue I stick to it. .23 .37

You've got to look out for

yourself or the demands of

circumstances and other peOple

will eat you up. .46 .54

I believe that I have to look
out for myself and mine, and let
others shift for themselves. .57 .71

I don't often dermuCh for others
unless they can do some good for
me later on. .41 .49

People who don't.work hard to

accomplish respectable goals

can't expect me to help‘when

they're in trouble. .47 .56

 

Relational/Connected Self

Activities of care that I perform
expand both me and others. .50 .76 .60 .76

Caring about other people is
important to me. .59 .67

Doing things for others makes
me happy. .51 .60

(table continues)

129

Item—total Scale Item-total Scale
Item Correlationa Alpha Correlationa Alpha

If someone does something for me,
I reciprocate by doing something
for them. .42 .52

I like to acquire many
acquaintances and friends. .43 .30

Relationships are a central part
of my identity. .48 .39

Those about whom I care deeply
are part of whom I am. .51 .45

It is necessary for me to take
responsibility for the effect my
actions have on others. .40 .46

Being unselfish with others is
a way I make myself happy. .38 .35

I like to see myself as
interconnected with a network of
friends. .42 .30

I believe that one of the most

important things that parents can

teach their children is how to

cooperate and live in harmony*with

others. .41 .44

I am guided by the principle of
treating others as I want to be
treated. .39 .45

 

Primacy of Other Care

All you really need to do to help
someone is to love them. .29 .68 .35 .67

If someone offers to do something

for me, I should accept the offer

even if I really want something

else. .41 .44

The worst thing that can happen
in a friendship would be to have
my friend reject me. .35 .47

(table continues)

130

Item-total Scale Item—total Scale
Item Correlationa Alpha Correlationa Alpha

I feel empty if I'm not closely
involved with soreone else. .32 .36

I often try to act on the belief
that self-interest is one of the
worst problems facing society. .30 .33

The people whom I admire are
those who seem to be in close
personal relationships. .33 .26

The best way to help soteone is
to do what they ask even if you
don't really want to do it. .43 .30

Being unselfish with otters is
more important than making
myself happy. .48 .52

I feel that my development has
been shaped more by the persons
I care aboutthanbywhat Ido
and accomplish. .37 .19

ltrytocurbmyangerfor fear
of hurting others. .41 .46

In making decision, I can neglect
my own values in order to keep
a relationship. .28 .23

If what I want to do upsets other
people, I try to think again to
see if I really want to do it. .36 .30

I often keep quiet rather than
hurt soreone's feelings, even if it
means giving a false impression. .43 .45

A close friend is someone who will

help you whenever you need help

and laws that you will help if

tley need it. .34 .27

 

(table continues)

131

Item—total Scale Item-total Scale
Item Correlationa Alpha Correlationa Alpha

Self and Other care Chosen Freely

True responsibility involves
making sure my needs are cared for
as well as the needs of others. .38 .78 .40 .77

Sometimes I have to accept hurting

someone else if I am to do the

things that are important in my

own life. .30 .17

If other people are going to

sacrifice something they want for

my sake I want them to understand

what they are doing. .40 .44

I want to learn to stand on my
own two feet. .53 .59

I do not want others to be
responsible for me. .35 .46

I deserve the love of others as
much as they deserve my love. .31 .32

If someone asks me for a favor I

have a responsibility to think

about whether or not I want to

do the favor. .43 .40

I believe that I must care for
myself because others are not
responsible for me. .45 .46

EVen though it's difficult, I

have learned to say no to others

when I need to take care of

myself. .31 .42

In order to continue a
relationship it has to let both
of us grow. .52 .40

If I am to help another person
it is important to me to understand
my own motives. .47 .38

(table continueS)

132

Item-total
Item Correlationa Alpha
I want to be responsible for
myself. .63

I accept my obligations and expect
others to do the same. .47

Before I can be sure I really
care for someone I have to know
my true feelings. .37

When I make a decision it's
important to use my own values to
make the right choice. .43

If I am really sure that what
I want to do is right, I do it
even if it upsets others. .42

Item—total Scale
Correlationa Alpha

.50

.51

.32

.45

.46

 

aCorrected for item overlap.

APPENDIX D

cwmooxsmmmmmcouss

133

When I FEEL PWERFUL it is because

1.

 

ITEM COMPIEI‘ICN
0 = not completed
1 = corpleted

DENIAL of being or feeling powerful

This code is for fairly direct staterents of denial or staterents of
power as negative or bad. Do not use this code when you infer from
the subject's response that s/he doesn't really feel very powerful
with the type of power s/he describes.

 

0 absent

1 denial of, or experience of power as unauthentic, as derivative
of another negative experience; power is not felt as an autonomous
experience. Ex. I can't remember feeling powerful. I ate spinach
(tee hee), I don't really feel powerful. I never feel powerful.
I'm overcompensating for a feeling of insecurity.

2 = power as ative experience. Ex. I fbrget temporarily who is
in charge, a ee ing I don't take pride in. I feel I am, but don't
really like to.

Power from/in RELATIONSHIPS with otters
This is NOT necessarily reflective of a connected-self orientation.

0 absent

1 emotional acceptance by others, liking, being loved by others.
Br. Others are accepting of the way I act or feel. I have received
a lot of strokes-I can then do anything. I lmcw I am loved by
important people in my life. I feel confident and loved (also 7.1) .
2 = dependency, others depending on self, for one-way experiences of
dependency with the self as giver/source of support for others,
being needed. Ex. Soreone depends on me. I can take care of
others, help them. I have helped soreore. I can help otters.

3 = respect, competence. Ex. PeOple listen when I'm imparting
knowledge on subjects I'm well versed in. I have accomplished a
goal and been recognized by soreone whose opinion I value (also
coded 6.1). Others respect my opinions and actions. I've dore
solething that is considered good by otters. People look up to me.
4 = Self- and other-care, including self as recipient of care. Dr.
I have done sorething worthwhile for otters and myself. '
5 = from relationships, others, but can't distinguish. Ex. People
will respond to me when I ask tlem to do a favor for me.

 

 

 

Positional power: power from one's POSITIQI or role in a hierarchy,
power over otl'ers, not necessarily coercive or against others, may
be role-conferred, via status, possessions, seen in staterents of
comparison, imbalance, “more than." Co'rpetitive imagery may be
used, where power for one person is at the expense of another's
power or status. ”Imagine a hierarchy based on some measure of a
standard, whether it is dominance, skill, age, possessions, parental
authority, job ladder, etc., where power comes from one's position
in that hierarchy relative to others' positions. Please be careful
not to let negative stereotypes of power as abusive exploitation or

5.

134

manipulation color your use of this code. There is nothing
inherently abusive about this type of power, e.g., parents are
legitimately powerful in their roles as parents of children younger,
less competent, etc., without having to express it though violence
or abuse. If you respond to a staterent of power as hurtful or
negative, look carefully at the source of the power: if it is from
sadism or manipulation or a statement of ability to get one's own
way, it may best be coded 12.1.

0 absent

1 present. Ex. Gained a position of authority and/or my
technical and comunicable (sic) skills. I have control (being a
leader, controlling a number of people in any given situation). I
am with soreone younger than me. I feel superior to the people
around me. I have sorething other people want. I am in charge of a
situation and my decision is the final one. My reliance on others
is not as much as their reliance on me. I have He upper hand. I
set short term goals and can control otter people. I have knowledge
or some other advantage on my side. I can do something that others
can't.

Power from moral sense of RIGHTness, moral certainty. This code is
for the experience of power coming from being certain one is morally
or conventionally in the ”right."

0 absent

1 - present. Ex. I think I'm doing the right thing whether others
agree or not. I'm absolutely sure of my convictions. I'm doing the
right thing and the other people lmow it. I have proved my point.

I have stood up for what's right.

Power from AGTIEVDIENT/ACCINPLISI-mr (task inferred), including
satisfaction at corpletion of task, satisfaction at accomplishment,
success, achievement, hard work, etc. Here power is experienced
through the process of achieving and/or the product of one's
achievement and work. It may/may not be a self-chosen goal: the
erphasis is on the doing, tl'e activity of achieving as the source of
power.

0 absent

1 present. Ex. I have succeeded. I've learned sorething I have
succeeded in doing sorething well. I've achieved a goal. I'm
satisfied with a job well done. I have accorplished a goal. I have
achieved something I've worked hard for. I feel like I accorplished
something. Ihavecomeupthehardwayandhaveworkedhardtoget
what I have now. I have controlled a particular situation or
accomplished sore goal (also coded 8.2) . I have accorplished or
achieved what I want (also coded 8.1). I feel good about myself
because of something that I've accomplished (also coded 7.1). I am
doing well and achieving what I set out to do (also coded 7.1).

7.

8.

135
Power from SEIF-EST‘EEIVI and/or other positive feelings about the self

0 absent

1 present. Ex. I feel confident and loved (also 3.1). I like
myself. I feel good abort myself and my potential. I feel
fulfilled. I feel good. Increased self-confidence. My inner being
is calm. At peace with myself. I feel confident and trust in God
(also coded 11.2) . I am sure of my capabilities. I am not having
feelings of depression and I feel in control over anything that
might happen (also coded 8.2-note this as example of a mtive
expression of 7.1). I feel good abort myself or what I'm doing
(also coded 8.1). Note other multiply-coded example above in 6.1.

Power from awareness of CHOICE about self/own life, experiencing the
self as able to choose, with freedom and reSponsibility for own
choices, feeling identification with one's own choices. Here the
emphasis is on seeing oneself as having options and the agency and
freedom for self-definition. This leads to increased efficacy,
omership and sense of control of one's life, but "control" per se
is not the central theme of this category (choice is).

0 absent

1 present

Ex. I have control of myself in a situation. I control my life.
I'm doing what I want; I have accomplished a goal I wanted to reach.
(These examples would be coded both here and under 6, power from
achievement.) I feel comfortable in what I'm doing. I am in charge
of myself. I am in control of myself. I'm doing what I want. I
feel that I can support myself and deal with my problem without
somebody elses intervention. (Note other examples doubled-coded
under 6.1 and 7.1 above.) I can change my life in ways that will
bring me happiness.

2 = for statements of "I am in control," with no further
elaboration, or statements ofTI am in control of the
situation/situation at hand" (impersonal or abstract Eneral
itatements of control, not over people—4.1 or over self—8.1). Ex.
I feel in contrcfl over anything that might lappen. I have taken
control of a situation. I have controlled a particular situation.

I have controlled a particular situation. I have control over the
things around me.

 

Power with BAIAMIE: dealing with multiple dimensions, juggling them
successfully.

0 = absent

1 = present

Dr. I have been true to myself and still gotten others to
understand without hurting them. I can cope with my many
responsibilities and still smile and be happy.

10.

11.

12.

136

Power from PREDICTABILITY, usually through knowledge or planning,
wtere predictability contributes to a sense of comfort and limited
control.

0 absent

1 present

Dr. I know what's going on. I am prepared for a situation. I feel
I have answers, I know where I'm headed. I am in control of a
situation and know what to do. I understand the reason and
direction of what I'm doing. I know what I'm doing. (Note—if you
are confused between coding 8.1 or 10.1 for a response, use 10.1 if
the response resembles examples here and if using 10.1 means less
inference than 8.1.)

PASSIVE or IMPERSCNAL power

0 absent

1 present

Ex. Things are going OK. All's going well in my life. Everything
is going great.

2 = God. Ex. I'm letting Divine Power express throrgh me. Because
of my reliance on God. I know it's God's will that I am following
His teachings according to the Bible .

NOI' SCORABLE by these categories. Always consider that a strange
response might be a misreading of the stem. Also, a response can be
meaningful and interesting, but just not fit any of orr codes and
fall outside our areas of interest: in this sense, it is not
”garbage" but simply not useful for tle coding scheme.

absent
not scorable. Ex. I have energy to spare. I feel that nothing
do can go wrong.

0
1
I
NOTE: 12.1 should never appear as part of a double code. Use any

other appropriate code instead of 12.1 or 12.1 if no other codes are
appropriate.

137

MienIGE'T‘ANGRYI

1.

 

Item corpletion

0=completed
l=not completed

DENIAL of anger

0=absent
1=present. Ex. I rarely get angry. I never get angry. I
don 't get angry except at myself.

Focus on RESPCNSIBILITY for others HURT/concern for impact of anger
on others, with view that anger hurts others, damages relationships.

0=absent

1=present. Ex. Wish to avoid hurting others. Hurt those I

don't want to hurt. Fear the results of the anger in my words and
action-upon others and myself. Don't believe in hurting anyone so
I try to keep my cool (also 4.2). Try not to let it affect others.

Holding anger in; WITHDRAWAL from feeling angry, expressing
anger or from conflictual situation; not communicating,

and/or channeling feelings into physical activity (not
acting-out) . "This category reflects an Anger-In dimension
where the subject retreats from his/her anger and/or the
conflictual situation. The withdrawal may be physical, as

in going off alone, or emotional, as in becoming quiet and
pulling back inside. There may be an active wish not to talk or
comunicate, and to hide feelings. The feelings and tension may be
channeled into physical activity and discharged indirectly in a
sublimated fashion rather than through cathartic physical
acting-out, as in doing horse-work, yardwork or cleaning
"furiously."

0=absent

1=withdrawal and/or substitute activity. Ex. Do nothing. Refuse to
express it. Don't cormrnicate. Get quiet or do housework
fast-like. Becore depressed and withdrawn (also 9.2) . Want to be
alone. Sit and stew over it. Brood. Fume about it inside of me
but don't let it show. Clam up. Retreat into myself. Bottle the
erotions as much as possible so as not to let others know I'm upset.
Try to be alone. Try to hold it in. Becore quiet and withdraw into
my own thoughts. Take a walk.

2=attempts at regulating/imgrlse control against losing control.
Count to 10 to try to keep my terper uerer control. Try to control
my anger. Take time to cool off. Have to find an ortlet for the
anger or I might do sotething rash. Usually walk away, cool my
terper. Try to keep a lid on it for fear of losing control.

 

 

5.

7.

8.

138
JUDGEMENTS abort anger

0=absent

1=n§ative judgements about anger or the angry self, no mention of
impact on others as in 3.1. Ex. Feel guilty. Say the wrong thing.
Don't like myself. I get nasty. Say things I wish I hadn't said.
Feel sorry. lose control of myself and feel degraded. Do not like
myself. Say things I regret/shouldn't. Get upset because I do not
like the feeling. get mad at myself for being angry.

2=g§itive judgements about anger or the self as angry. Ex. I have a
right to get angry. Feel I have a right to vent anger when it is
hurting me. I feel good after, because that has been a problem for
me (keeping anger to myself). I used to feel guilty... now I
recognize anger...accept it as an emotion I have a right to
experience... not just a right but...a duty when situations warrant
it.

SQGATIZATION

0=absent

1=present, expressed as physical more. Ex. Get a head-
ache, muscle spasm, or other physical synptom. My stomach
gets in a knot and I can't breathe easily. I get migraines.

BEHAVIORAL expression of anger (as cathartic acting-out)

0=absent

1=physical expression, possible aggression towards objects,

(does not include people as targets). Ex. Throw things,

slam doors. Throw old, chipped coffee mugs against the

garage wall. Throw whatever is handy, especially pillows.
2=verbal expression, nonrational venting. Ex. Scream. Yell.
Scream and yell at the closest (family) people to me. Talk loud,
make rash statements. Explode and sport ort things I regret later
(also 5.1). Soretimes short but don't hold grudges.
3=bothjrhysical and verbal acting-out. Ex. Bang drawers,

shut doors and sometimes yell at the person who annoys me. Throw
things or say things I don't mean (also 5.1).

4=expression of anger not specified (but IUI' cry). Ex. I

blow up! Tend to overreact and show a lot of erotion (also 5.1) .

 

 

RATIQIAL approach (not venting)

0=absent

1=introspective problem-solving/analysis throrgh rational

means (usually implicit: by oneself). Ex. Write in jounal. Think
it out. Need time out to assess where the anger is really coming
from. Become very analytical. Am now learning to look for the
reason why. mite. Come back to the sorrce of anger and view it in
a more objective perspective.

Zwith another person (not the recipient of anger). Ex.

Tell my problems to a friend. Go talk to sorecne who is

not involved with the fight.

3=with G_og. Dr. Pray, then act. Use prayer.

 

 

9.

10.

11.

12.

139

Transformation of ANGER.TO SADNESS/DISTRESS. This category
of transformation of affect is ONLY for anger changed to
sadness/tearfulness/distress/depression/hurt feelings,

not anger to some other phenomena.

0=absent

leggy. Ex. cry. Start crying and then I can't voice my feelings.
Usually get upset and end up in tears. Cry.

2=depressed, sad or hurt feelings. Ex. Cry and feel sorry for
myself. Am feeling hurt. Become depressed. Feel hurt.

COMMUNICATION/RESOLUTION'attempts

0=absent

1=communication of feelings, reconciliation attempts including with
partner with whom in conflict. Ex. voice my feelings. Express it
verbally; verbalize it. Talk with the other person. Tell the
person I'm angry with.

2=impersonal statements about correcting , resolving the problem
(others not specified). Ex. Try to resolve the problem. Try to
correct the situation which made me angry.

 

REACTIVE AGGRESSION towards others, retaliation (aggression,
not expression).

0=absent

l=physica1 aggression towards people. Ex. Hit the other person.
Throw something at the other person, lose control.

2=verbal aggressigg towards people. Ex. Deliberately say

things I know will hurt or will make the other person angry or I
cvercriticize. Say mean things to provoke further argument. Say mean
things to get back at the other person.

3=mode unspecified. Ex. I fight back any way I can. I tend

to take it out on others. Have a tendency to take my anger out on my
family and friends. Tend to provoke others into more argument.

 

NOI‘ SCORABLE by these categories. Please always consider
that a response may be a misreading of the stem: the most
common misreading here seems to be I GET ANGRY‘WHEN .
I have to stay at home all the time, people try to hurt me
or someone I love, I get a lower grade than I expect, I
don't live up the my own expectations or I am tired, etc.

 

0=absent

1=not scorable

***NOTE: 12.1 should NEVER.appear as part of a double
code. If any other code fits, use that first.
12.1 is for responses that do not fit anywhere
in our coding scheme.

REFERENCES

REFERENCES

Bakan, D. (1966) . fie duality of human existence: An essay on psychology
and religion. Chicago: Rand McNally.

 

 

Beck, A. T. (1972). Depression: causes and treatment. Philadelphia:
University of Pennsylvania Press.

 

Bernardez-Bonesatti, T. (1978) . Women and anger: Conflicts with
aggression in contemporary women. Journal of tle American Medical
Woren's Association, 33(5), 215-219.

 

 

Bibring, E. (1953). The mechanism of depression. In P. Greenacre
(Ed.), Affective Disorders. NéW'Ybrk: International University
Press.

 

Blos, P. (1980) . Modifications in the traditional psychoanalytic theory
of female adolescent development. In S. Feinstein (Ed.) , Adolescent
Psychiatry VIII. Chicago: University of Chicago Press.

 

 

Broverman, I., Vogel, S., Broverman, D., Clarkson, F., s. Rosenkrantz, P.
(1972) . Sex-role stereotypes: a current appraisal. Jourral of
Social Issues, _2_8_, 59-78.

 

 

Brown, G., & Harris, T. (1978). Social origins of depression: A study
of psychiatric disorders in women. New York: The Free Press.

Buss, A. H., & Plomin, R. (1975). A temperament theory of personality
development. New York: Wiley & Sons.

 

 

Carlson, R. (1965) . Stability and change in the adolescent's
self-image. Child DevelOpment, g, 659-666.

 

Carmen, E., Russo, N., 8 Miller, J. B. (1984). Inequality and women's
mental health: an overview. In Rieker, P., 8. Carmen, E. (Eds.), go;
gender gap in psychotherapy. New York: Plenum Press.

 

Chodoff, P. (1974). Depressive personality. In R. J. Friedman and M.
M. Katz (Eds.) , The psycholggy of depression: contemporary theory
and research. Washington, D.C.: V. H. Winston.

 

 

 

Chodororr, N. (1974) . Family structure and feminine personality. In M.
Z. Rosaldo & L. lamphere (Eds.), Women, culture and society.
Stanford: Stanford University Press.

 

140

141

Chodorow, N. (1976) . Oedipal asymmetries and heterosexual knots. Social
Problems, _28, 454-468.

Cooley, C. (1902) . Human nature and the social order. New York:
Scribner.

Dinnerstein, D. (1976) . The mermaid and the minotaur: Sexual
arrangements and human malaise. New York: Harper & Row.

Dohrenwend, B. S. (1973). Social status and stressful life events.
Journal of Personality and Social Psychology, 38(2) , 225-235.

Dohrenwend, B. S. (1976) . Anticipation and control of stressful life
events: an exploratory analysis. Paper presented at the Eastern
Psychological Association Convention, New York.

Donelson, E., & Gullahorn, J. (Eds.). (1977). Women: A psychological
perspective. New York: Wiley & Sons.

 

Douvan, E., & Adelson, J. (1966). The adolescent experience. New York:
Wiley & Sons.

 

Drsek, J., & Flaherty, J. (1981). The development of the self-concept
during the adolescent years. Monographs of the Society for
Reasearch in Child Development, 46, Serial No. 191.

 

Erikson, E. H. (1950). Childhood and society. New York: W. W. Norton.

 

Erikson, E. H. (1963). Childhood and society (2nd ed.). New York: W.
W. Norton.

Erikson, E. H. (1968) . Identity: Youth and crisis. New York: W. W.
Norton.

 

Gilligan, C. (1982). In afdifferent voice: Psychological theory and
worren's development. Cambridge, Ma.: Harvard University Press.

 

Gilligan, C. & Iangsdale, S. (1980). The contribution of woman's
thought to developmental theory: The elimination of sex bias in
moral development research and education. Interim report: National
Institution of Education .

Gleason, N. (1985) . Daughters and mothers: college women look at their
relationships. Works in Progress. Wel lesley, MA: Stone Center
Working Papers Series .

 

Hodgson, J. W., & Fischer, J. L. (1979) . Sex differences in identity
and intimacy development in college youth. Jorrnal of Youth and
Adolescence, _8_, 37-80.

 

 

Hoffman, M. L. (1977) . Sex differences in apathy and related
behaviors. Psychological Bulletin, 84(4), 712-722.

142

Horney, K. (1937) . The neurotic firsonality of our time. New York: W.
W. Norton.

Horney, K. (1942) . Self analysis. New York: W. W. Norton.

Horney, K. (1950) . Neurosis and human growth. New York: W. W. Norton.

Hunter, J. E. (1977, March). Cluster analysis: Reliability, construct
validity and the multiple indicators approach to measurement. Paper

presented at the workshop titled 1iAdvanced Statistics“ at the U. S.
Civil Service Commission.

 

 

Hunter, J. E., Cohen, S. H., & Nicol, T. S. (1975). PACKAGE: A set of
correlations and factor analysis programs. Unpublished manuscript,
Michigan State University.

 

Johnson, F. L., & Aries, E. J. (1983). Conversational patterns among
same-sex pairs of late-adolescent close friends. Journal of Genetic
Psychology, 142, 225-238.

 

 

Jordan, J. (1984) . Empathy and self boundaries. Work in Progress.
Wellesley, Mass.: Stone Center Working Papers Series.

 

Jordan, J., Surrey, J., & Kaplan, A. (1983). Women and empathy. Work
in Progress. Wellesley, Mass.: Stone Center Working Papers Series.

 

Josselson, R., Greenberger, E., & McConochie, D. (1977a).
Phenomenological aspects of psychosocial maturity in adolescence.
Part 1. Boys. Journal of Youth and Adolescence, _6_, 25-55.

 

Josselson, R., Greenberger, E., & McConochie, D. (1977b).
Phenomenological aspects of psychosocial maturity in adolescence.
Part 2. Girls. Journal of Youth and Adolescence, 6, 145-167.

 

Kaplan, A. (1986) . The self-in-relation: implications for depression in
woren. Psychotherapy, 23(2), 234-242.

Kaplan, A., & Klein, R. (1985). The relational self in late adolescent
women. Work in Progress. Wellesley, MA: Stone Center Working
Papers SeT‘res.

Kohlberg, L. (1981) . The philosophy of moral development. San
Francisco: Harper & Row.

 

Kohut, H. (1971). Analysis of the self. New York: International
University Press .

Kon, I. S., & losenkov, V. A. (1978). Friendship in adolescence:
Values and behavior. Jorrnal of Marriage and the Family, 40,
143-155.

lederer, W. (1968). The fear of women. New York: Grune & Stratton.

143

lerner, H. (1974) . Early origins of envy and devaluation of women:
Implications for sex-role stereotypes. Bulletin of the Menninger
Clinic, fl, 538-553.

 

Lerner, H. (1977) . The taboos against female anger. Mennipger
Perspective, 8(4), 5-11.

lerner, H. (1980) . Internal prohibitions against female anger.
American Journal of Psychoanalysis, flu) , 137-148.

lerner, H. (1985) . The dance of anger. New York: Harper & Row.

levinson, D. J. (1978). The seasons of a man's life. New York: Alfred
A. Knopf.

Lyons, N. (1983). Two perspectives: on self, relationships, and
morality. Harvard Educational Review, _53, 125-245.

Maccoby, E., & Jacklin, C. (1974) . The psychology of sex differences.
Palo Alto: Stanford University Press.

 

Mahler, M., Pine, F., & Barman, A. (1975). The psychological birth of
the human infant: Symbiosis and individuation. New York: Basic
Books.

 

Makosky, V. (1980) . Stress and the mental health of women: A discussion
of research and issues. In M. Guttentag, S. Salasin, & D. Belle
(Eds.) , The mental health of women. New York: Academic Press.

Mead, G. (1934). Mind, self and society. Chicago: University of
Chicago Press.

Miller, A. (1981). The drama of the gifted child. New York: Basic
Books.

Miller, A. (1983) . For your own good: Hidden cruelty in child-rearing
and the roots of violence. New York: Farrar, Strauss & Giroux.

Miller, A. (1984) . Thou shalt not be aware: Society's betrayal of the
child. New York: Farrar, Strauss & Girorx.

Miller, J. B. (1976). Toward a newpsychology of woren. Boston: Beacon
Press.

Miller, J. B. (1982). Women and power. Work in Progress. Wellesley,
MA: Stone Center Working Papers Series.

Miller, J. B. (1983). The construction of anger in women and men. Work
in Progress. Wellesley, MA: Stone Center Working Papers Series.

Miller, J. B. (1984). The development of women's sense of self. Work
in Prcyress. Wellesley, MA: Stone Center Working Papers Series.

144

Monge, R. H. (1973) . Developmental trends in factors of adolescent
self—concept. Developmental Psychology, 8, 382-392.

 

Nunnally, J. C. (1978). Psychometric theory. NewYork: McGraw—Hill.

 

Padesky, C. A., & Hammen, C. L. (1981). Sex differences in the
expression of depressive responses on the Beck Depression Inventory.
Journal of Abnormal Psychology, _8_6, 609-614.

Pearson, J. L., Reinhart, M. A., Donelson, F. E., Stromen, E. A., &
Barnes, C. L. (1985, April). Marital status and Gilligan's
developmental theory of care of self and other. Paper presented at
the Biennial meeting of the Society for Research in Guild
Development , Toronto , Ontario , Canada .

 

Radloff, L. S. (1977) . The CES-D scale: A self-report depression scale
for research in the general population. Applied Psychological
Measurement, 1, 385-401.

 

 

Reinhart, M. A., Pearson, J. L., Kamptnar, N. L., Cornwell, K., Barnes,
C. L., Stromen, E. A. & Donelson, E. (1985, April). Assessment of
Gilligan's model: Development of the relationship self inventory.
Paper presented at the Biennial meeting of the Society for
Research in Guild Development, Toronto, Ontario, Canada.

 

 

Rich, A. (1976). Of woman born. New York: W. W. Norton.

 

Rosenberg, M. (1965) . Society and the adolescent self-image. Princeton,
NJ: Princeton University Press.

 

Rosenberg, M. (1979). Conceiving the self. Princeton, NJ.: Princeton
University Press.

 

Salzman, L. (1974) . Interpersonal factors in depression. In F. F.
Flack & S. C. Droghi (Eds.) , The nature and treatment of depression.
New York: John Wiley.

 

Schwartz, J. C., & Juroff, D. C. (1979). Family structure and
depression in female college students: effects of parental conflict,
decision-making power and inconsistency of love. Jorrnal of
Abnormal Psychology, _8_8, 398-406.

 

 

Seligman, M. E. (1975). Helplessness. San Francisco: W. H. Freeman.

 

Silber,E., & Tippett, J. (1965). Self-esteem: clinical assessment and
measurement validation. Psychological Reports, _1_6(4) , 1017-1071.

 

Scares, A., & Scares, L. (1969). Self-perceptions of culturally
disadvantaged children. American Educational Research Journal, 6,
31-45.

 

145

Spence, J., Helmreich, R., & Stapp, J. (1975). Ratings of self and
peers on sex-role attributes and their relation to self-esteem and
conceptions of masculinity and femininity. JOurnal of Personality
and Social Psychology, 32, 29-39.

 

Stautberg, S. S. (1985, December). The elements of power. Saggy,
36-41.

Stiver, I. (1984). The meanings of "dependency“ in female-male
relationships. 'Work in Progress. wellesley, MA: Stone Center
working Papers Series.

 

Strommen, E. A., Reinhart, M. A., Pearson, J. l», Barnes, C, Lu, Blank,
L., Cebollero, A. M., cornwell, K., Donelson, E., & Kamptner, N; L.
(1987, April). Assessment of Gilligan's model: The revised
relationship self inventory. Paper presented at the Biennial
meeting of the Society for Research in Child Development,
Baltimore, MD.

 

Sullivan, H. (1953). The interpersonal theory of psychiatry. New York:
Wt Wk Nbrton & Co.

 

Surrey, J. (1985). Self-in-relation: A theory of women's development.
Wbrk in Progress. Wellesley, MA: Stone center Wbrking Papers
Series.

Thorbecke,‘W. L., and Grostevant, H. D. (1982). Sex differences in
styles of occupational identity formation in adolescence.
Developmental Psychology, 18, 396-405.

 

vaillant, G. E. (1977). Adaptation to life. Boston: Little, Brown.

 

weissman, M. N., & Klerman, G. L. (1979). Sex differences and the
etiology of depression. In E. S. Gomberg &‘V. Franks (Eds.), Gender
and disordered behavior. New Ybrk: Brunner/Mazel.

 

westkott, N. (1986). Historical-developmental roots of female
dependency. Psychotherapy, 23(2), 213-220.

wetter, R. (1975). Levels of self-esteem association with four sex-role

categories. Paper presented at the meeting of the American
Psycho ogical Association, Chicago.

 

Winnecott, D. (1971). Playing and reality. New Ybrk: Basic Books.

 

I.
IV Ul‘! a.