.1...IIJILI.. . . . . . . . 31.14 2: ... fie... e .o

F... ... .. 3.

..v. .r. .5 .
l...v3..... . 4....

: \I. blur}.
_. . if... 1.1... Jan
abs}; 1.. «.1Lnfi-oomldr!

4|; .r: . l 4
.2 r it .95

Ln!‘

.fi.r...wumq.».nn. .

. ( h... «at-r133:
2..

. . «=2. . .
3.

..! ..».

.._.....

 

ugfimti , , . . . , . . 3. - .xfiskF.a.uno,... .nurufin.

 

l!

 

l

 

 

 

 

 

NIH"!

 

 

l!

 

 

 

 

 

 

 

00788 5860
f * -‘
LIBRARY
Michigan State
University
L‘ I
This is to certify that the

dissertation entitled

VIOLENCE IN STORIES ABOUT SCENES OF
AFFILIATION AND ACHIEVEMENT: NEW
EVIDENCE AGAINST POLLAK AND GILLIGAN'S REPORT

presented by

Loren A. Brooks

has been accepted towards fulfillment
of the requirements for

Ph.D. Psychology

degree in

 

 

cl

’

 

Major Eolessor ‘

DateJl/fl I/fi

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

 

MICHIGANSTATUE ) f“ "

MUN!!!)WWII!“MIMI”!!! 5* ~> .

5/

(.2 /

,7 I
,/

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmetive Action/Equal Opportunity Institmion
smarts-e.-

VIOLENCE IN STORIES ABOUT SCENES OF AFFILIATION AND ACHIEVEMENT:
NEW EVIDENCE AGAINST POLLAK AND GILLIGAN'S REPORT

BY

Loren A. Brooks

A DISSERTATION

Submitted to Michigan State University
in partial fulfillment of the requirements

for the degree of
DOCTORATE OF PHILOSOPHY
Department of Psychology

I 988

DIG/IBLJIDDI

L

ABSTRACT

VIOLENCE IN STORIES ABOUT SCENES OF AFFILIATION AND ACHIEVEMENT:
NEW EVIDENCE AGAINST POLLAK AND GILLIGAN'S REPORT

bY

Loren A. Brooks

Pollak and Gilligan (I982) reported that men more frequently wrote
stories containing violence to pictorial stimuli featuring intimacy while
women more frequently wrote stories containing violence to pictures
depicting achievement. These differences were attributed to distinct
sex-dif f erences in modes of self-definition and interpersonal posture. In
a subsequent study featuring methodological advances, Benton, Hernandez,
Schmidt, Schmitz, Stone, and Weiner ( I 983) were unable to replicate
Pollak and Gilligan's findings.

In an effort to clarify these divergent findings, the current work
further explored the presence of violence in similar stories but employing
an original set of pictorial stimuli balanced for both intensity level and
for the featured picture characters' sex.

Participants appeared to be a representitive sample of

undergraduates based on their scores on Spence and Helmreich's (i978)

Personal Attributes Questionnaire (DAG) and Wiggins' (I979) Interpersonal

Adjective Scales. Responses to the present stimuli also failed to support
Pollak and Gilligan‘s findings but were closer to Benton et al.'s report.
Both men and women wrote significantly more stories containing violence
to scenes of Intimate Affiliation, as compared to scenes of Moderate
Affiliation and to scenes of Achievement, at both Intense and Moderate
levels. The role of the featured picture characters' sex could not be fully
ascertained due to a flawed pictorial stimulus. The order in which stimuli
were presented significantly influenced the number of scored violent
responses given by participants. PAQ gender identity did not relate to
stories scored as containing violence in response to Intimate Opposite-Sex
Affiliation or Intense Opposite-Sex Achievement pictures for either men
or women.

Although Pollak and Gilligan succeeded in drawing research attention
to some important issues concerning gender, the failure of successive and
more methodologically advanced studies to confirm their provocative
findings raises some serious questions about the latter's validity.
Definitive answers to questions raised by Pollak and Gilligan must be
based on more sophisticated studies which explore order and intensity of

StImUII.

ACKNOWLEDGMENTS

This manuscript is dedicated to my advisor and friend, John Hurley
Ph. D., to my aunt, Marilyn Slough, and to my friend David Rosenberg. The
relationships I shared with John and my Aunt Marilyn inspired my
Interests in interpersonal relationships and in the experiences of women.
Along the way, they also taught me about feminism and about listening to
my own thoughts and feelings.

My friendship with David began when I entered graduate school and
will remain important to me for the rest of my life. The feelings,
thoughts, and experiences we have shared have strongly influenced my
development.

Additional thanks go to Linda Forrest, Ph.D. for her consistent

comments, insights, and inspiration.

IV

TABLE OF CONTENTS

List of Tables .

List of Appendices

Chapter I - INTRODUCTION
Gilligan's Work

Related Theoretical and Empirical Work

Aggressive Imagery in Thematic Apperception Test Stories .

Fear of Success
Challenges to the Work of Pollak and Gilligan .
The Current Study
Relationship of the Personal Attributes Questionnaire .
Hypotheses
CHAPTER II - METHOD .
Participants
Measures

Procedure .

vii

VIII

18

24

29

31

33

37

39

39

SI

CHAPTER III - RESULTS
Interpersonal Adjective Scales .
Picture Orientation .
Interrater Reliabiltiy
Violence Scores in Pictorial Stimuli
Content Analysis
Analysis of Variance
Hypotheses for Men
Hypotheses for Women .

Order Effects

Relationship of Violence Scores to the PAO .

CHAPTER VI - DISCUSSION .
APPENDICES

REFERENCES

VI

52

52

54

.59

65

70

71

74

76

77

BI

82

97

12I

Table I.

Table 2.

Table 3.

Table 4.

Table 5.

Table 6.

Table 7.
Table 6.

Table 9.

LIST OF TABLES

Rages
Sequential Presentation of Pictorial Stimuli . . . . 46
Interpersonal Adjective Scales and Scores . . . . . 53

Rated Affiliation and Achievement Pulls of Pictorial Stimuli 55
Rated Intensity Levels of Affiliation and Achievement Pictorial
Stimuli.................S7
Frequency of Stories Containing the Mention of Violence . 60
Percentage of Participants Who Wrote Stories Containing .
Violence Elicited by Achievement and Affiliation . . . 67
Content Analysis of Stories Scored as Containing Violence 72
Analysis of Varience: Sex x Pictorial Stimuli. . . . . 73

Comparison of Individual Hypotheses . . . . . . . 75

Table l0. Frequency of Stories Containing Violence in Five Sequences 79

Table l I. Analysis Of Varience: Order X Pictorial Stimuli. . . . 80

vii

G.

H.

LIST OF APPENDICES

Coding Scheme of Interview Data (Lyons, I981)
Pictorial Stimuli

Pictorial Stimuli Content Questionnaire .
Personal Attributes Questionnaire

Interpersonal Adjectives Questionnaire .

Group Administration Instructions for Pictorial Stimuli .

Study Explanation for Participants

Guidelines for Rating Stories

Ia. Scored Responses of Female Participants

Ib. Scored Responses of Male Participants

J.

Types of Violence in Scored Violent Responses .

viii

97

99

108

109

111

114

115

117

118

119

120

INTRODUCTION

Personality theorists have long recognized that the self develops
through interaction with others. In her work on moral development
Gilligan (I982) has explored “self in relation to others." She reported two
distinct modes of interacting which are broadly related to gender and
development across the life span. Her research has indicated that
generally men develop an identity Which is based on objective reciprocity
In relationships (ob jective/separate self) and women primarily form a
definition of self based on connection with others and an orientation of
care (the connected self). Supporting this theory Pollak and Gilligan
( I982) found that men tended to report more aggressive behavior in
picture stimuli displaying scenes of intimacy while women tended to
report aggressive behavior in pictures depicting achievement situations.
The validity of this finding has been the subject of controversy due to the
limited range of pictorial stimuli utilized. It remains unclear whether
women responded to general achievement cues or cues of competition and
whether men respond differently to varying levels of intimacy.

Attempting to further distinguish the different modes of
self-definition and interpersonal orientations Gilligan has delineated, this
study will focus on clarifying the findings of Pollak and Gilligan (1982) by

1

I)

using a different set of pictorial stimuli. These stimuli will consider
both the level of intensity and the sex of characters in scenes of
affiliation and achievement. The pictures used will be similar to those
composing the Thematic Apperception Test (Murray, l943) but have been
specifically designed for this study. The Personal Attributes
Questionnaire (PAO: Spence & Helmreich, i978) will also be administered
to clarify the linkages between Gilligan‘s modes of self-definition and
gender identity.
W

Gilligan's (i977) exploration of moral development led her to
conclude that classical developmental theories, most notably Kohlberg's
theory of moral development, inadequately represented the experiences of
women (i977) by neglecting the relational or interpersonal component of
self -def inition in the developmental process. She has primarily based her
conclusions on three studies which have employed interviews. In doing so
she has assumed, "that the way people talk about their lives is of
significance, that the language they use and the connections they make
reveal the world that they see and in which they act" (Gilligan, l982). in
her milestone work, m aotfferent Voice Gilligan ( 1982) identified three

primary studies she conducted as the college student study the aoortton

bl

decis/on stud); and the rig/its and responsibilities study.

The college student study examined "view of self," thoughts about
morality, personal moral conflict, and choices made about conflicts. The
abortion decision study explored the relationships between experienced
conflict, morality, and development in women considering abortion. The
rig/its and responsibilities study further considered identity, morality,
and moral conflict by systematically comparing responses of men and
women to real and hypothetical moral dilemmas. While Gilligan has
focused on the relationship between self-definition and the moral decision
making process, the postures which she has illuminated appear to be
modes of interpersonal interaction instead of individual perspectives
experienced in isolation. The existence of these gender linked and
divergent interpersonal postures have challenged the theories of Erikson
and Kohlberg. They can be further understood through the work of
Chodorow (i978) and more recently have been expanded upon by Lyons
(l983).

WW

Erikson (i950) postulated eight stages of psychosocial development

which included: trust versus mistrust, autonomy versus shame and doubt,

initiative versus guilt, industry versus inferiority, identity versus

identity confusion, intimacy versus isolation, generativity versus
self -absorption, and integrity versus despair and disgust.

In the trust versus mistrust period occurring during infancy the task
to be mastered involves experiencing relationships and individuating
within these relationships. Autonomy versus shame and doubt is marked
by a growing separateness evolving in the walking child. This move
towards greater autonomy continues during the initiative versus guilt
stage but is challenged as the child enters the Oedipal period, or middle
childhood years, by the realization that to join the world of adults, they
must join them and learn to do what they do. This transition signifies the
crisis encountered in the industry versus inferiority phase. Gilligan
( i982) commented:

But about whom is Erikson talking? . . . it turns out to be

the male child. For the female, Erikson (i968) says, the

sequence is a bit different. While for men, identity

precedes intimacy and generativity in the optimal cycle

of human separation and attachment, for women these tasks

seem instead to be fused. intimacy goes along with identity,

as the female comes to know herself as she is known,

through her relationships With others.

Yet despite Erikson‘s observation of sex differences,

his chart of life-cycle stages remains unchanged: identity

continues to precede intimacy as male experience continues

to define his life-cycle conception. (p. i2).

In addition to limiting development to the male life cycle, Erikson's
conceptualization allows little preparation for the first adult phase of
development, intimacy versus isolation. According to Erikson, the sole
precursor to this stage occurs during infancy with trust versus mistrust.
The remainder of their development revolves around separation.
Attachment almost appears to be "a developmental impediment" (Gilligan,
i982,p.i3)

in a similar vein, Freud (l925) earlier noted that because women
were unable to obtain Oedipal resolution they developed superego's that
were never “so inexorable, so impersonal, so independent, of its emotional
origins as we require it to be in men" . . . and that . . . "for women the level
of what is ethically normal is different from what it is in men" (1925, pp.
257-258). He concluded that the judgements of women were influenced by
their feelings and that by comparison they were less just.

Nancy Chodorow (i978) brought a feminine perspective to the

understanding of human development. She pointed out that because women

are the primary providers of child care, particularly through age five, life
begins for both males and females with a primary identification to mother.
Consequently, separation is experienced differently by the genders. "A boy
gives up his mother in order to avoid punishment, but identifies with his
father because he can then gain the benefits of being the one who gives
punishment, of being masculine and superior" (Chodorow, i978, p. i i3).
While according to psychoanalytic theory, "penis envy--the feminine form
of the castration complex--leads a girl to turn to her father exclusively,
and thenceforth to see her mother only as a sexual rival, . . . the girl
abandons the mother as a love object with far more embitterment and
finality than the boy" (Freud, i925, p. 253). Chodorow posited that due to
the similarity of mother and daughter it is more difficult for a girl to have
a sense of individuation and in ways she is an extension of her mother. In
other words girls, "retain a sense of self which has preoedipal" (Chodorow,
I978, p.i IS) characteristics. This means that girls continue to experience
themselves as involved in merging, attachment characterized by primary
identification, and fusion of Identification and object. Boys, having been
cut off from the preoedipal relationship loose their sense of empathy in an
effort to cement ego boundaries. Consequently, their personalities revolve

around autonomy and separation.

Girls emerge from this period with a basis for "empathy“

built into their primary definition of self in a way that

boys do not. Girls emerge with a stronger basis for

experiencing another's needs and feelings as one's own.

Furthermore, girls do not define themselves in terms

of the denial of preoedipal relational modes to the

same extent as do boys. Therefore, regression to these

modes tends not to feel as much a basic threat to their ego.

_ (Chodorow, I978, p. I67).

In her observations of i8i f if th-grade children Lever (I976) observed
that boys more often played out of doors, in age-heterogeneous groups,
competitively, and that their games lasted longer than those of girls.
interestingly, boys quarrelled constantly but no game was interrupted for
more than seven minutes while arguments among girls tended to terminate
games. Based on his own observations Piaget (i932) asserted that
legality was less well developed in girls.

Piaget studied childhood moral development from preschool to middle
childhood but left off at adolescence. Picking up where Piaget stopped
Kohlberg (I958, l981 ) specifically examined adolescent moral

development. Kohlberg's (l9S8, I98I) theory of moral development was

based on a study of eighty-four boys aged IO to 16 who were followed for
twenty years at three year intervals. He described six stages of moral
development progressing from an egocentric understanding of fairness
based on individual needs (stages one and two), an understanding of equity
marked by conformity to societal norms (stages three and four), and
finally an understanding of fairness which considered equality and
reciprocity (stages five and six). He noted that stage one development
carries a perceived vulnerability to powerful figures into adolescence and
individuals in this stage focus on staying out of trouble. in stage two the
orientation continued to be needs based but the omnipotence of powerful
others fades in importance. Stage three is embodied by a stereotypical
emphasis on goodness and the importance of the needs of others and in
stage four this sentiment Is extended to society at large and contains
concern for social preservation. in postconventional morality, or stages
five and six, equality and reciprocity are central and eventually respect
for the individual becomes paramount. Kohlberg‘s initial findings were
primarily based on responses to the hypothetical dilemma of a male
character named Heinz. In this dilemma Heinz must decide whether or not

to steal a drug in order to save his wife's life.

While Kohlberg generally defended these stages of development, a

\O

phenomenon observed in college students known as moral relativism
challenged this theory. Moraliy relative judgements were represented in
the notion that objective moral judgement was impossible and that,
therefore, people should do what they feel or believe is right. On the
surface this appeared to be a regression back to stage two development
but looked at differently this was not necessarily true. Gilligan noted that
it was women and relativists who generally defied the boundaries of
Kohlberg's coding schema. it appeared that the contextual situation was
salient for these individuals.

When Gilligan began examining the moral development of girls and
woman she found that the thought processes that went into their moral
decisions were qualitatively different from those of boys and men. She
initially noted these differences in her studies of college students
(Gilligan & Murphy, i979) and women considering abortion (Gilligan &
Belenky, i980).

in the college studentstudy 25 undergraduates were randomly
selected from a group of sophomores who had enrolled in a course on moral
and political choice. They were Interviewed about personal moral
conflicts and life choices as seniors and five years later. Gilligan

believed that some of the struggles involved in moral decision making for

I 0

women were reflected in the experiences of Clair (pseudonym) who had
enrolled in the course on moral choice and eventually went on to become a
physician. in the course of the study Clair described her own
transformation, from a person absorbed in pleasing others and frightened
of her own assertiveness, to a self -assured woman who identified the
importance of caring for others around her. Reflecting upon Clair's
development and the experiences of other women Gilligan asserted that:

The conflict between self and other thus constitutes

the central moral problem for women, posing a dilemma

whose resolution requires a reconciliation between

femininity and adulthood. In the absence of such a

reconciliation, the moral problem cannot be resolved.

The “good woman" masks assertion in evasion, denying

responsibility by claiming only to meet the needs of others,

while the "bad women" forgoes or renounces the commitments

that bind her In self -deception and betrayal. it is precisely

this dilemma--the conflict between compassion and autonomy,

between virtue and power-~which the feminine voice

struggles to resolve in its effort to reclaim the self and to

solve the moral problem in such a way that no one is hurt.

(Gilligan, I982, p. 70-71).

When Clair considered the Heinz dilemma as a senior she focused on
the "failure of response." She believed that Heinz should have done
whatever necessary to save his wife's life, but further, that the
pharmacist also had a responsibility to life and an obligation to make an
exception in the case of Heinz's wife. in this instance Clair associated
morality with connection and care.

This relationship reflects the conflict between care of others and
self-care. This conflict readily surfaces when a woman is required to
decide whether or not to have an abortion. in a study of this conflict 29
women ranging in age from fifteen to thirty-three, and coming from
divergent ethnic and socioeconomic classifications were referred to the
investigators by pregnancy counseling services and abortion clinics, and
were interviewed In their first trimester. These women participated In
the study for a variety of reasons. Several used the interviews to clarify
their thoughts and feelings, some were encouraged to participate by their
counselors, and others were willing to offer their time to research. Some
of the women had failed to use contraception, had denied the possibility of
becoming pregnant, used the pregnancy to cement or test relationships, or

had reconsidered after deciding to have a child. Of these women, four gave

birth, two miscarried, twenty-one chose abortion, and two were undecided
and could not be reached for f oilow-up.

Describing herself and the process she used to make decisions,
another participant in this study, Illustrated the tenuous balance of self
and other care.

"It is taking a life. Even though it is not formed, it is the

potential, and to me it Is still taking a life. But i have to

think of mine, my son's, and my husband's. And at first I

thought it was for selfish reasons, but it is not. I believe

that, too, some of it is selfish. I don't want another one

right now; i am not ready for it." (Gilligan, i982, p. 83)

While examining such womens' responses, Gilligan ( 1982) noted a
sequence in the development of morality and care. The first phase of this
sequence was characterized by care for self in order to survive. This was
followed by the perception of being selfish. The initial feeling of
responsibility grew into a mentality which sought to insure care for
others. The Inequality of this arrangement stimulated "a reconsideration
of relationszs in an effort to sort out the confusion between self -
sacrifice and care inherent in the conventions of feminine goodness"

(Gilligan, i982, p. 74 ). Through her interveiws with women Gilligan

i 3

observed that ultimately, care became freely chosen and self -care was
perceived as equally important. The phases she described have been
summarized as: (a) care of self is necessary because others will not care,
(b) care of others has priority over care of self, and (c) care of self is
chosen freely since care for all, including self, is important (Pearson,
Kamptner, Cornwell, Barnes, Strommen, & Donelson, I985).

Throughout Gilligan's work women described themselves in terms of
connection with other people. Helping and giving were central to their
self -perceptions. Self-evaluation was dependent upon their ability to
maintain relationships and care for people. Reports of achievement were
largely absent from women's self -descriptions and many of them spoke of
the conflict they experienced between achievement and care. Because
their descriptions spoke of an identity formed by a mode of being-in-
relation-to-others, Lyons (I983) has characterized womens'
self-definition as the "connected" self.

Men used adjectives that conveyed separateness to describe
themselves as opposed to describing their way of being with others. They
often described themselves as intelligent, competitive, rational, creative,
intense, or sensitive. Their identities appeared more distinct and their

standard 0f self-assessment seemed more based on achievement. Their

I4

attachments to others are ancillary to their sense of identity. Their
voices describe an identity based on a mode of being-ln-relation-to-
others-as a "objective/separate" self (Lyons, I983).

In an effort to view these perspectives across the life span Gilligan
conducted the rignts andresponsloilitlos study which further considered
identity, morality, and moral conflict through hypothetical moral
dilemmas (Gilligan, Langdale, Lyons, & Murphy, 1974). This was done
systematically by including men and women matched for age, intelligence,
education, occupation, and social class at nine points across the life span
beginning with 6 and ending with 60. Within this group there were two
men and two women from the ages: 8, I2, l5, i9, 22, 27, 36, 46, and 60+.
Although this sample size was small Gilligan noted that this was
characteristic of theory-building research in the area of moral
devel0pment. The milestone work of Piaget was based on a sample of 20
boys and Kohlberg's initial sample consisted of 58 males.

The data collected by Gilligan contained of responses to a five part
interview that was conducted in an open-ended manner. The interview
included the following Inquiries.

I. Looking back over the past year/five years what stands out for

you?

i5

2. The Heinz dilemma (one of Kohlberg's hypothetical moral

dilemmas).

3. A real life dilemma generated by the individual in response to a

question about a personal experience of moral conflict and choice

4. How would you describe yourself? Is the way you describe

yourself now different from the way you described yourself in the

past? if you see change how would you account for that?

5. What does morality mean to you? What makes something a moral

problem for you? What does responsibility mean to you? When

responsibility to self and others conflict, how should one choose?

The coding schema for written responses was created by Lyons (i98l)
and distinguished the definitions of "self as connected in relation to
others and as separate/objective in relation to others“ (Gilligan,
unpublished manuscript). The four general constructs coded were:
construction of the problem, resolution of the problem, evaluation of the
resolution, and the responses to the "describe yourself “ question. This
coding schema appears in Appendix A.

The data reflected that modes of self -def inition were clearly gender
related but not limited to gender. In response to real-iii e moral decisions

75% (i2) of the women, but only l4% (2) of the men chose an orientation of

’

lo

responsiveness, wnile 79% (I i) of the men chose an orientation of rights
and responsibility but only 25% (4) women responded in the rights and
responsibility mode. Furthermore, 37% (6) of the women never expressed
any consideration of rights and 36% (6) of the men never employed the
responsiveness mode.

A comparison of predominant orientations and modes of
self -def inition demonstrated that females who defined themselves as
“connected" more frequently utilized a responsive mode when confronted
with real life moral conflicts, whereas men who considered themselves as
"separated/objective" called upon considerations of rights and
responsibility in real-life moral dilemmas.

Only a single notable difference was found when comparing data from
different age groups. Adult women constructed problems or moral
dilemmas concerning "rights" at a frequency that did not occur for females
of other ages. in spite of this, they continued to use the responsive mode
of morality for the resolution of these dilemmas but "care of self "
interestingly dropped out of consideration for women over twenty-seven
in this sample. Men more consistently selected a "rights" mode across the
life span. Their use of "responsiveness" peaked during adolescence, when

it nearly equaled their choice of the "rights" mode.

This empirical data conflicts with Gilligan's (I982) later theoretical
inferences based on all of the interview information. She concluded that
postures characterized by "rights" and "responsiveness" or connection
versus separation developed over time through maturation and that this
maturation consisted of a transcendence through infantile, adolescent, and
adult needs (Gilligan, I982). Primarily based on the experiences of
women, she suggested that while connected and separate self identities
were broadly associated with gender, maturation fostered their
integration and that adults used both postures. She believed that for
women the absolute of care changed as the importance of personal
integrity became more salient and that for men the absolute of truth and
reciprocity shifted due to experiences which demonstrated differences
between personal perceptions and the perceptions of others. She used the
experience of "Alex" to exemplify these changes. After earlier
participating in the college student study, Alex became an attorney and the
change in his perspective was evident in his comments about a
relationship which, to his surprise, ended had painfully.

People have real emotional needs to be attached to

something, and equality doesn't give you attachment.

Equality fractures society and places on every person

I 8

the burden of standing on his own two feet (Gilligan, I 982, p. I67).
Ar i im inThmiA r inT ri

Pollak and Gilligan ( l982) became interested in comparing sex
differences in violent imagery using pictorial stimuli (similar to the
Thematic Apperception Test-TAT) when they observed unexpected violence
in men's stories written about what appeared to be a peaceful scene of a
man and a woman "sitting on a bench by a river next to a low bridge." They
Initially believed the violence perceived in this scene by men might
parallel Horner's ( i968) finding that women wrote more violent stories to
sentence stems featuring achievement situations. They were interested in
clarifying whether differences in aggressive fantasies reflected
systematic perceptual differences. They did this by gathering stories in
response to four pictures similar to classic Thematic Apperception Test
cards.

Their curiosity about gender differences In interpersonal orientations
was stimulated by previous literature. It had been shown that men are
generally more aggressive than women (Terman & Tyler, i954). This was
not only true for Americans but held up cross culturally (Whiting & Pope,
l973). Maccoby and Jacklin (1974) found that women tended to express

their aggression In disguised ways, while men were generally more direct,

l9

but primarily turned their aggression towards each other. May (1980),
interpreting fantasies of men and women, hypothesized that women
inhibited aggression and men failed to perceive relationships.

Directly related to the work of Pollak and Gilligan, Bramante (I970)
reported that men responded with disaster filled stories when shown a
romantic film. Aside from Pollak and Gilligan's study, Bramante's
investigation has been one of the few considerations of the relationship
between aggression and affiliation. Pollak and Gilligan wanted to explore
whether men and women associated different amounts of aggression or
violence with affiliation.

Like Maccoby and Jacklin (i974), Pollak and Gilligan (I982)
distinguished aggression from violence in that aggression is the intent of
one person to hurt another, while violence surpasses mere intent.
Violence was defined as the infliction of "homicide, suicide, death by
accident, rape or forcible violation, physical assault, kidnapping, or fatal
disease." The authors used pictures depicting work settings and human
closeness to represent achievement and affiliative situations.

Eighty-eight men and 50 women enrolled in a course on Human
Motivation at Harvard participated in their study. Students were asked to

write a brief and imaginative story for each picture and a series 0f

20

questions served as guidelines (Appendix F). Two of the pictures selected
for this study depicted achievement and two displayed affiliation. The
first achievement picture featured a man sitting alone at a desk in a high
rise apartment building. A photograph of his wife and children sat on the
desk, and there was a large window in front of him. The second consisted
of two women dressed in white laboratory coats. The woman in the
background watched while the woman in the foreground handled the test
tubes. The first affiliation picture was of a man and a woman sitting next
to each other on a bench. There was a river in front of them, a low bridge
to the side, and university buildings appeared in the background. The
second affiliation scene showed a male and a female trapeze artist. A man
was shown hanging upside down from his knees on a trapeze bar while a
woman on another bar grasped his wrists in midair.

These stories were scored for the presence or absence of violence.
Violence appearing in at f iliative or achievement situations was
distinguished and noted. Figurative violence ("Jef f was sick to death of
the performance") was not included in violent scores. Lastly, a content
analysis was made in which violence arising from situations described as
af f iliative or achievement was scored regardless of scene category.

Of the men In this sample, SI% (45) wrote violent stories versus only

21

22 o (i I) of the women. A quarter of the men (22) wrote violent stories to
scenes of affiliation and 6% (6.8) wrote violent stories to scenes of
achievement, and l7% (19.3) constructed violent stories to both affiliation
and achievement scenes. No mention of violence occurred in the stories of
48.9% (43) of the male participants. Conversely, 6% (3) of the women
wrote violent stories to scenes of affiliation, I6% (8) perceived violence
in pictures depicting achievement, and none wrote stories with violence in
both scenes. The vast majority of women (78% or 39 of 50) wrote stories
with no violence. ‘

The content analysis was aimed at determining which situations men
and women perceived as dangerous. Three categories of content were
described: violence associated with situations of affiliation, violence
associated with situations of achievement, and violence associated with
neither affiliation nor affiliation but taking place by accident or due to
natural causes. The results for the men showed that 26. I % (23) described
violence and danger in affiliative situations, l.i% (1) described danger in
achievement situations, 3.4% (3) described danger in both situations, 20.5
( l8) perceived danger in neither, and 48.9% (43) reported no danger.

Women never described danger in situations of affiliation, i2% (6)

described danger in achievement Situations, none described danger in DOiZi'l

scenes, 10% (5) wrote about violence associated with neither situation,
and 78% (39) reported no violence.

Computed separately for men and women, statistical analyses found
that a significantly greater number of men perceived danger or violence in
stories of affiliation. Pollak and Gilligan noted that men appeared to
associate danger with entrapment or rejection and women related danger
to isolation. Accordingly, the trapeze picture, the only picture with
physical contact, elicited the most violence from men and the man sitting
at his desk, the only person alone, received the most violent stories from
women. Another provocative observation made by the authors was that
22% of the women versus 6% of the men built the presence of a safety net
into the stories about the trapeze artists. Additionally, 40% of the men
clearly specified the absence of a net in contrast to only 6% of the women
who excluded this safeguard.

The students who participated in this work appeared to perceive
danger in different situations. Men generally associated danger with
situations of interpersonal connection while women perceived danger in
situations of achievement which reflected competition and isolation. In
his analysis of male and female fantasies May (i980) distinguished what

he called the male posture of "pride" and the female stance of “caring." In

23

measuring female development against the male posture he insinuated that
the absence of aggression in woman was unusual or inadequate. This is
similar to Freud's (i931) idea that "the essence of femininity lies in the
repression of aggression" (Pollak and Gilligan, I982, p. I65). The results
of this study challenged the work of May by suggesting that woman do not
perceive relationships as dangerous and consequently they are not
experienced as aggressive. Pollak and Gilligan suggested that women may
stay in relationships which are harmful to them, not because they are
masochistic, but because it may take them longer to experience and
realize the danger in interpersonal relationships. it is possible that
women perceive relationships as safe because they prevent isolation,
thereby affirming self-identity, and men generally feel more comfortable
in competitive situations with clearly defined boundaries and rules where
their self -perceptions are not challenged.

The recent work of Helgeson and Sharpsteen (I987) supported the
findings of Pollak and Gilligan. They asked undergraduate men (50) and
women (6 I) to write stories elicited by the picture of the acrobats used
by Pollak and Gilligan. They directed the attention of participants to
achievement and affiliation by placing the following sentences beneath the

pi CtUFES.

a. Achievement

These people have worked hard for many years to reach

their present level of achievement. They strive to improve

their skills with each performance.

b. Affiliation

These people have had a close relationship for many years.

They have shared many activities and experiences, which

have created an intimate bond between them.

The authors reported that there was an interaction between
respondents“ sex and instructions (1-; = 4.99, p < .05). Men wrote more
stories containing violence to the scene with an affiliative caption
(women [:1 = .07, men :1 = .27) and women wrote more stories that
mentioned violence to the scene with an achievement caption (women L1 =
.3l, men [:1 =03).

W

The study of fear of success in women has begun to explore how
women experience achievement and how achievement interacts with
self -perception. in her study of achievement motivation and performance
in competitive and non-competitive situations Horner ( i 968) found that

women who performed higher In non-competitive situations, when

25

compared to competitive situations, were more likely to write stories
characterized by high fear of success imagery. Women who performed
higher in competitive situations were more likely to write stories
characterized by low fear of success imagery. She also found that the
achievement motivation of women was lower in mixed-sex competitive
situations compared to same-sex competitive situations and
non-competitive situations.

Some of the fear of success imagery women reported in response to
achievement situations was striking, if not bizarre. Responses were
written about situations typified by a story about "Anne," a medical
student at the top of her class. in one account Anne was beaten and
maimed for life by her envious classmates. Ninety women and
eighty-eight men participated in this study and 65% of the women versus
8% of the men reported imagery connoting fear of success. The women
who displayed fear of success and performed better in non-competitive
situations appeared to associate success with social isolation and loss of
femininity. Supporting this finding Peplau (1976) found that women with
traditional sex-role attitudes and fear of success were particularly apt to
associate competition with aggression.

Horner's findings and the fear of success construct have received

26

marked theoretical attention and empirical reconsideration. Organizing
about half of the Fear of Success (FOS) literature existing at the time,
Tresemer (i976) pooled the findings of ICC studies assessing FOS. He
narrowed that group of studies by analyzing only samples which included
males and females from the same subpopulation. Cumulatively, he
considered 56 samples from 42 studies which represented close to 6,000
men and women. He found that women told only a slightly greater number
of FOS stories but that there was "tremendous variability between
studies." Tresemer noted that the content of male and female stories was
strikingly different, and suggested that in the absence of a single scoring
method, male and female data may have been scored differently for FOS.
Male stories were consistently more bizarre, cynical, hostile, pessimistic,
and included more jokes, death, and violence. Females wrote more stories
which contained social rejection, affiliative loss, and loss of femininity.
He concluded that such thematic differences might reflect the differing
psychodynamic significance of FOS for men and women.

Tresemer‘s metanalysis also found that FOS has declined since about
i969 and this decline has paralleled increases in female attendance to
medical schools. It has frequently been hypothesized that women with

greater ability would experience more FOS because success was within

27

their grasp. This relationship was not eVIdent in the present studies in
which ability included experimental performance tasks, SAT scores, GPA‘s,
IO scores, career goals, etc. Many studies have sought to link gender-role
identity or identification with stereotypically traditional roles with FOS.
Gender-role Identity has been measured by career choice, employment of
mother, college major and so on. Together, studies considering
gender-identity did not find a significant relationship between this
dimension and F05. Measures of anxiety have also not been related to FOS.
Higher FOS scores have been related to high external locus of control
scores. While Horner‘s study has not been replicated in it's entirety, the
information synthesized by Tresemer does call the F05 construct into
question and illustrates the lack of clarity which surrounds it.

Olsen and Willemsen ( I978) illustrated that when the ”deviance"
aspect or isolation of Anne in her class was removed, the negative
outcomes of success lessened. They observed this by varying the classic
story about Anne finding herself at the top of her medical school class
after first term finals. In one of the descriptions all of her classmates
were men and in the other description half were men and half are women.
They gathered information from descriptions with both a female (Anne)

and a male (John) protagonist. They assessed negative feelings on the part

of the achiever (Anne or John), negative reactions of others, other negative
outcomes, positive feelings of the achiever, positive reactions from
others, and other negative outcomes.

Males and females did not differ in their mention of negative feelings
and sex of respondent did not interact with sex of the protagonist.
Females reported more negative reactions from others when Anne was in
an all male class and positive feelings were mentioned less frequently for
females in the all male class than for males or females in the mixed-sex
class.

Studying women's performance in mixed-sex situations LaNoue and
Curtis (I985) found that following a mixed-sex "led to f all situation"
women, when compared to men, had lower expectations or their
performance, they performed worse, they rewarded themselves less, and
they attributed their failure to lack of ability significantly more. These
differences were not evident in same-sex tasks nor when tasks were
performed alone. Women's negative attributions in mixed-sex situations
were altered when given instructions which encouraged effort and
success.

Paludi and Fankell-Hauser ( I986) examined FOS using interviews.

When women were asked if they had been in situations, and had feared

success, 9 l % of their sample (ranging in age from late teens to 80) replied
"no." A striking percentage of women (96%) did report that they wondered
if their achievements were worth the effort and many of them struggled
with the issue of costs and benefits. Most frequently they cited
relationships and children as being important considerations (Paludi &

F ankeIl-Hauser, I986). These authors suggested that the FOS concept be
abandon in favor of an achievement versus cost model.

While the existence of FOS as a measurable personality construct
remains unclear it appears that men and women generally perceive
achievement differently. This difference can effect performance, and
there may be costs attached to achievement for women just as there may
be costs for men in the interpersonal realm.

k f i

Pollak and Gilligan conducted a study which is one of the few
empirical examinations of Gilligan's theories of development, and has been
criticized on several accounts. Some of the most salient questions have
come from Benton, Hernandez, Schmidt, Schmitz, Stone, and Weiner (I983).
Their major challenges concern the classification of TAT cards into
motivational categories, failure to control for sex of persons depicted in

TAT pictures, the limited scoring schema, and what they perceived of as

 

3O

unusual instructions.

Benton et al. argued that one of the pictures intended to depict
achievement contained features of affiliation and that one of the pictures
meant to depict affiliation included elements of achievement.
Specifically, in the stimulus depicting the man at his desk, they noted the
man appeared to be looking at a family photograph, and they asserted that
the scene with the acrobatics could be considered a professional or an
achievement situation. To determine whether the pictures represented
their intended themes, the authors asked students to rate each picture for
its portrayal of affiliation or achievement. Additionally, the pictures
were scored for affiliation and achievement imagery with a standard TAT
schema (Atkinson, i958). The results showed that the two pictures did
not appear to display their intended themes. Although the scene of the
man at his desk was primarily rated as an achievement picture, it elicited
more affiliation imagery as scored with “standard TAT scoring"
procedures. The picture of the trapeze artists was seen as depicting
achievement and it elicited more achievement Imagery.

To make up for these difficulties Benton et al. added two cards which
they believed to portray affiliation and achievement less ambiguously.

These displayed a man and a woman on a hillside and two men in suits

exchanging documents and were rated as depicting their intended themes.
Using these two new pictures, and the cards used by Pollak and
Gilligan, Benton et al. found that while men perceived more violence in
general, there was no significant difference in the violence scores of men
and women written in response to scenes of affiliation and achievement.
They further contended that the scoring criteria used by Pollak and
Gilligan was stringent and restrictive because mild physical violence and
nonfatal accidents were not considered. To clarify the impact of scoring
criteria a more liberal scoring schema that utilized nonlethal harm was
applied in addition to the original scoring system. They found that
differences in the scoring systems did not significantly effect results.
The authors also believed that Pollak and Gilligan's instructions to
write an "interesting and imaginative" story may have pulled for an
unusually high number of violent stories. They gave one half of the sample
the instructions used by Pollak and Gilligan and one half more "neutral"

instructions but found no significant differences between the two sets of

instructions.

W511
Although Benton et al. f alled to reproduce the findings of Pollak and

Gilligan, they concluded that "a more exhaustive examination of dramatic

32

and intuitively resonant findings that might have social implications" was
warranted. While their work raised important challenges and questions
concerning Pollak and Gilligan's study and Gilligan's theoretical
observations, further research appears necessary to clarify these
provocative and contrasting findings.

Pollak and Gilligan's study used the trapeze scene and the man and
woman on a bench to represent affiliation. Benton et al. added a scene
depicting a man and woman on a hillside. Achievement was represented by
three women in a lab and a man at a desk by Pollak and Gilligan and Benton
et al. added two men in suits exchanging documents to this set. The work
of Pollak and Gilligan was explorative and it is understandable they might
not have foreseen the shortcomings of their stimuli. Although the focus of
the Benton et al. study was not solely on the type of stimuli used, they
gave no clear explanation about what guided their choice of the additional
pictures, except that they believed, and later found, that the cards
displaying the trapeze artists and the man at his desk were perceived as
both affiliation and achievement scenes. The cards they chose did elicit
imagery and ratings which were consistent with the themes they intended
to portray but these stimuli still appear to be limited.

While the previous studies were well Intentioned, the visual stimuli

 

33

used in these studies was not fully balanced for level of intensity or sex
of picture character. It has been suggested that women have the most
difficulty with achievement in mixed sex competitive situations (LaNoue
& Curtis, 1 985; Horner, i968) and that men may relate danger to
affiliation, following separation from a female primary care giver early in
life (Chodorow, i974), and because it leads to what is perceived as
entrapment (Pollak & Gilligan, I982). Based on these findings and
premises it seems important to consider issues of intensity level and sex
of depicted character.

Differences in self -def inition and Interpersonal posture as discussed
by Gilligan and as reflected in stories written to pro jectives has primarily
been linked to sex differences. It also seemed worthwhile to explore their
relationship to gender identity. The Personal Attributes Questionnaire
(PAD: Spence & Helmreich, I978) appears to be one of the more useful and
valid measures of sex-role orientation and style.

:-. HI. 0 "‘ 04A 1 ' Q.‘ .lle.

Although considerable work remains to be done in the area of gender
identity measurement, the PAO (Appendix D) has received somewhat more
empirical support as a measure of gender identity than several other

instruments including the Bem Sex Role Inventory (BSRI: Bem, i974). Both

34

the BSRI and PAO were based on the assumption that men and women
internalize sex-role behaviors based on what is defined by their families
and peers as socially desirable. Each instrument contains general scales
of masculinity and femininity, and identifies the position of androgyny, in
which masculine and feminine qualities are thought to be accessible parts
of an individual's behavior. The BSRI contains a separate scale which
measures an individual's tendency to endorse socially desirable traits as
self -descriptive, whereas, the PAO does not measure the tendency to
endorse socially desirable traits because the items were selected to be
equally desirable, although some items were assumed to be more desirable
for one sex or the other. The social desirability of test items was
determined by independent judges for the BSRI and persons were asked to
characterize "ideal" individuals of both sexes on the PAC). The PAO
contains a scale (Masculinity-Femininity) composed of items that have
been perceived as more desirable for one sex.

An additional difference between these measures is that the PAO
consists of bipolar scales of Masculinity (M), Femininity (F), and
Masculinity-Femininity (M-F), whereas the BSRI contains the unipolar
subscales Masculinity, Femininity, and Social Desirability. While each

measure identifies masculine, feminine, androgynous, and undifferentiated

 

35

gender positions using a median split scoring method, their construction
and content differ slightly. Spence and Helmreich (1978) have credited
these differences with the minimal correlations between BSRI and PAO
scores.

Some of the strongest criticisms of the BSRI which led to the
selecting the PAO as a measure of gender identity, have been the numerous
multiple factors that have appeared in the BSRI, the apparent negativity of
several femininity scale adjectives, and BSRI's relatively weaker
predictive validity. The Masculinity subscale alone has yielded the
following factors in different studies: assertiveness, interpersonal
sensitivity, and maturity (Pedhazur & Tetenbaum, I979) and dominance,
incisiveness, insensitivity, and independence (Pearson, I980). This
multiplicity of BSRI factors challenges the alleged unidimensionality of
its Masculinity and Femininity scales. Although items on the BSRI were
selected based on their desirability for one sex or the other and Bem
(i974) claimed that all items are "relatively desirable" for either sex, the
Masculinity subscale appears to contain more socially desirable traits
than the Femininity subscale (Gross, Batlis, Small, & Erdwins, i979;
Pedhasur & Tetenbaum, I979; Puglisi, i980). items shy, gullible, and

childlike of the BSRI's Femininity scale have been consistently judged as

undesirable.

The BSRI also appears to be less sensitive to gender schema theory
than the PAO (Hungerford & Sowbolew-Shubin, 1987). Bem ( l98l) ,
proposed that gender orientation was in part determined by a readiness to
"encode and to organize information--including information about the
self --in terms of the cultural definitions of maleness and femaleness that
constitute the society‘s gender schema“ (Bem, I98l, p. 363). Testing this
premise Hungerford and Sowbolew-Shubin ( I987) examined response
latencies when persons were asked to determine whether masculine and
feminine phrases (controlled for social desirability, sex-linked content,
and length) were like "me" or "not me." As predicted, individuals
classified as masculine or feminine by the PAO and the BSRI identified
same-sex typed phrases as self -descriptive faster than did those
classified as either androgynous or undifferentiated. Persons classified
as androgynous by the PAO responded faster to masculine phrases than did
feminine individuals. A similar pattern held when PAO "androgynous"
persons were compared to persons classified as masculine in reference to
feminine phrases. Supporting the PAO's stronger linkage to gender schema,
the response times of persons judged as androgynous by the BSRI did not

differ significantly.

37

if Gilligan's observations about differences in the interpersonal
orientations of men and women were accurate, it was reasonable to expect
that the reflection of these differences culled by projective measures
would be significantly related to PAO scores. Consequently, PAO
masculinity scores would be expected to correlate significantly and
positively with high ratings of violence in scenes of affiliation, while PAO
femininity scores would be expected to correlate significantly and
positively with high ratings of violence in scenes of achievement. The
strength of these correlations would be expected to vary depending upon
the nature and intensity of achievement and af f iliative stimuli.

Additionally, the Interpersonal Adjective Scale (IAS) developed by
Wiggins ( I 979) will be used assess the degree of sample distortion due to
self-selection. Participants' scores will be compared with the scores of
similar samples to assure that this group of Individuals falls within an
average range of behavior.

HYPOTHESES

W

i. Men will write significantly more stories containing violence in
response to all pictorial stimuli as compared to the number of stories

written by women.

38

2. Men will write significantly more stories containing violence in
response to scenes of Intimate annsite-Sex Affiliation as compared to
scenes of WW (M2 versus MI ).

3. Men will write significantly more stories containing violence in
response to scenes of W as compared to
scenes of W (MI versus M5).

4. Men will write significantly more stories containing violence in
response to scenes of WW as compared to
scenes of W (M2 versus M4).

5. Men who write stories containing violence in response to scenes of
WW will have significantly higher PAO

Masculinity scores than men who do not write stories containing violence

in response to scenes of intimate annsite-ng Affiliation (M2 and PAO M
scores)
W

I. Women will write significantly more stories containing violence

in response to scenes of W as compared
to scenes of WW (F3 versus F4).

2. Women will write significantly more stories containing violence

in response to scenes of W as compared

to scenes of Monetate Same-Sex Acnieyement (F 3 versus F6).

3. Women will write significantly more stories containing violence
in response to scenes of W as compared to
scenes of WW (F4 versus F2).

4. Women who write stories containing violence in response to
scenes of WW will have significantly higher
PAO Femininity scores than women who do not write stories containing
violence in response to scenes of WW (M4
and PAO F scores).

METHOD
W

Two hundred and seventy-one men and women participated in this
study. One hundred men and one hundred women composed the main body of
participants from which data were collected. All were undergraduates
enrolled in introductory psychology courses at Michigan State University
during the fall quarter of i987. The remaining 7i participants ( 38 women
and 33 men), enrolled in an introductory psychology course in the spring
quarter of 1987, rated the pictorial stimuli for affiliative and

achievement content.

All volunteers were recruited from the psychology department‘s

40

research participation pool which allows students to receive credits
toward higher course grades by participating in psychological research.
The pool is primarily composed of students enrolled in introductory
courses. Some instructors require research participation and others give
credit toward course grades for time spent participating in research. Most
introductory course instructors require that students obtain four points of
course work by serving as a research participant. Each half hour of
participation is worth one point. Students choose the experiments they
would like to participate in and attend them when they are scheduled.
W
W

The pictures or scenes used in this study were devised by the author
and were fashioned after those in the Thematic Apperception Test (Murray,
I943). The TAT was originally designed by Morgan and Murray in i935 and
may be scored and used in the assessment of personality. TAT responses
have been shown to reveal primitive motivations (Combs, i947) and
perceptions elicited from its pictures have been associated with behavior
(Mussen & Naylor, I954; Purcell, I956). The projective power of the TAT
seemed clear in Karon's ( l98i) study of patients who had received electric

shock treatment following suicide attempts. While those patients

41

generally showed little memory of what precipitated such attempts, the
events that led to their actions appeared evident in their TAT protocols.

Karon also pointed out that while the TAT is rich in information that
might have clinical applications, the establishment of a simple scoring
system for data elicited by the pictures may have minimal clinical value
because most of the complex and unique information about individual
personality might not be scorable. Contrarily, the TAT appears to be a
useful empirical tool because targeted stimuli can be selected and used to
assess specific aspects of personality if reliable scoring systems can be
developed (Karon, I98I ).

For the current study, the author created eight pictures shown in
Appendix B which attempted to portray affiliation and achievement
stimuli in situations balanced for gender and level of intensity. While the
TAT stimuli were drawn by artists, the present pictures are photographs.

Four of the pictures (two affiliation - intended to portray moderate
and intimate levels; and two achievement - intended to portray moderate
and intimate levels) were used to explore how men and women perceived
situations containing both sexes. Featuring a male and female, these
pictures were shown to all participants. Four other pictures (two male -

intended to portray moderate affiliation and achievement and two female -

42

intended to portray moderate affiliation and achievement) were used to
explore participants‘ impressions of same-sex interactions and were
shown to either men or women.

To enhance the likelihood that these pictures would be perceived as
affiliation or achievement situations, sentences stems were placed
beneath each illustration. Verbal leads have been successfully used to
gather projective material by Winterbottom (i958), Lowell (McClelland, et
at, I953), French (I955), and Atkinson and Litwin (I960). The following
list describes stimuli shown to participants.

WW

MI & F i. A man and a woman talking (Moderate Opposite-Sex

Affiliation).

Michael and Carla have known each other for about two

years.

M2 & F2. A man and women embracing (Intimate Opposite-Sex

Affiliation).

Elizabeth and Nathan have been close for two years.

M3 & F3. A man and a woman seated at a table working together on a

project (Moderate Opposite-Sex Achievement).

Eve and Mark are department store managers in their

43

company.

M4 & F4. A man and a woman seated at a table working together on a
project (Intense Opposite-Sex Achievement).

Stuart and Alexandra are high ranking executives in their

firm.

Wm:

M5. Two men talking (Moderate Same-Sex Affiliation).

Phillip and Brian have been friends for two years.

M6. Two men seated across from one another apparently working
together (Moderate Same-Sex Achievement).

Peter and David are physicians. They are discussing a new cancer
drug.

W

F5. Two women talking (Moderate Same-Sex Affiliation).

Caroline and Theresa have been friends for two years.

F6. Two women seated across from one another apparently working
together (Moderate Same-Sex Achievement).

Julia and Sharon are physicians. They are discussing a new cancer
drug.

These pictures were designed to illustrate affiliation and

44

achievement between and among sexes at moderate and strong levels of
intensity. Both affiliation and achievement scenes depict associations
between individuals of the same and opposite-sexes. They also intended to
feature two levels of intensity in affiliation (moderate and intimate) and
achievement scenes (moderate and Intense). While Moderate Same-Sex
Achievement and Intimate Opposite-Sex Affiliation were compared;
Intimate Same-Sex Affiliation was not used in comparisons because
homosexuality was not targeted as an issue in this study.

To help insure that the pictures pulled for their intended themes, each
was rated independently by a group of undergraduates for their depiction
of affiliation and achievement. Coinciding with Benton et al.'s study,
affiliation was defined as a concern with "the establishment and
maintenance of interpersonal relationships" (Atkinson, I958, p. 205) and
achievement was described as concern with "success and good
performance" (Atkinson, i958, p. i8l ). Ratings were made on a 7-point
scale anchored by "does not portray" and "certainly portrays" affiliation or
achievement (Appendix C).

The stories were scored for violence employing the straightforward
system used by Pollak and Gilligan (i982) in which violence was scored

wholly on presence or absence. Violence included: death, suicide,

45

homicide, assault, rape, or fatal disease. in the analyses performed by
Pollak and Gilligan, high intercoder reliability was apparently achieved.
The authors reported I00% agreement on the overall presence of violence,
93% agreement on the presence of violence in situations of affiliation and
achievement, and 84% agreement on the content analysis.

Pictures were presented In five sequences to allow for an assessment
of order effects. The sequences varied in their presentation of theme and
intensity (Table i). Sequence i presented same-sex pictures first and
began with an affiliation picture. Sequence lI also presented same-sex
pictures first but began with an the achievement picture. Sequences III
and iV began with the more intense opposite-sex pictures but alternated
the presentation of the affiliative and achievement pictures. Affiliation,
achievement, and level of intensity were alternated in Sequence V.
W

Spence and Helmreich originally derived the PAO (i978) from a study
by Rosenkrantz, Vogel, Bee, Broverman, and Broverman (I968) in which
students identified characteristics which differentiated men and women.
They asked students to rate over I30 items as typical of adult men versus

adult women, male college students versus female college students, and as

46
Table l
Sequential Presentation of Pictorial Stimuli

Sequence I
Moderate Same-Sex Affiliation
Moderate Same-Sex Achievement
Moderate 0pp05ite-Sex Affiliation
Moderate Opposite-Sex Achievement
Intense Opposite-Sex Affiliation

Intimate Opposite-Sex Achievement

Sequence 2
Moderate Same-Sex Achievement
Moderate Same-Sex Affiliation
Moderate Opposite-Sex Achievement
Moderate Opposite-Sex Affiliation
intense OppOSIte-Sex Achievement
Intimate Opposite-Sex Affiliation

Sequence 3
Intimate Opposite-Sex Affiliation
intense Opposite-Sex Achievement
Moderate Opposite-Sex Affiliation
Moderate Opposite-Sex Achievement
Moderate Same-Sex Affiliation

Moderate Same-Sex Achievement

Sequence 4
intense 0pposute-Sex Achievement
intimate Opposite-Sex Affiliation
Moderate Opposite-Sex Achievement
Moderate Opposite-Sex Affiliation
Moderate Same-Sex Achievement
Moderate Same-Sex Affiliation

Sequence 5
Moderate Opposite-Sex Affiliation
Intense Opposite-Sex Achievement
Intimate Opposite-Sex Affiliation
Moderate Same-Sex Affiliation
Moderate Same-Sex Achievement
Moderate Same-Sex Affiliation

47

typical of the ideal individual of either sex for both college students and
adults. ideal ratings were more closely linked with either extreme
masculinity or femininity, leading the researchers to conclude that PAO
items represented socially desirable attributes.

The PAO consists of three scales labeled Masculinity (M), F emininity
(F), and Masculinity-Femininity (M-F). The M and F scales include items
considered equally socially desirable for each sex, but are more often
associated with one sex (e.g., independence). The M-F scale contains items
whose social desirability varies between men and women (e.g.,
aggressiveness was judged as desirable for men and nonaggressive
desirable for women). The authors noted that their M scale generally
reflected instrumental, agentic characteristics and the F scale generally
referred to expressive, communal attributes, while the M-F contains
elements of both postures. The PAQ scalesiwere empirically supported by
correlations in which each item was more highly correlated with the scale
to which it belonged than either of the two other scales.

The original form of the PAC) was a 55 item semantic differential
with five possible responses ranging from "Not at all . . . (e.g.,
aggressive)" to "Very . . . (e.g., aggressive)" This initial version was

subsequently shortened to a 24-item inventory (see Appendix D) based on

48

correlational data. This briefer versions' M, F, and M-F scales of the short
version correlated .93, .93, and .9I with their longer counterparts. Items
on the PAO are scored from 0 to 4. The scores on individual items of each
scale are summed to obtain scale scores. The possible scores for each
scale range from 0 to 32.

Using the median-split method (Spence & Helmreich, I978)
individuals' are categorized as Masculine, Feminine, Androgynous, or
Undifferentiated depending upon the position of their responses above or
below the medians on the M and F scales. M-F scale scores are also
determined by median splits. Due to the loss of information involved in
categorization, the median split method was only be used for descriptive
purposes in the present study and raw scores were be used in statistical
analyses.

l tiv

After reviewing much of the theoretical and empirical literature on
Interpersonal behavior, Wiggins (I979) constructed a taxonomy of trait
descriptive terms which appears to reflect the predominant features of
interpersonal behavior. He Introduced eight scales labeled
Ambitious--Dominant, Gregarious--Extraverted, Warm--Agreeable,

Unassuming--Ingenuous, Lazy--Submissive, Aloof--Introverted,

49

Cold--Ouarrelsome, and Arrogant--Calculating. The intercorrelations of
these bipolar scales formed a circumplex undergirded by the primary
dimensions of dominance and warmth. These two dimensions have
repeatedly been recognized as the principal dimensions of interpersonal
behavior (Hurley, i980; Kiesler, i983; LaForge & Suczek, I955; Wiggins,
I982).

The items composing the eight interpersonal scales introduced by
Wiggins are shown in Appendix E. Each scale contains I6 items, adding to
i28 total items. Participants were asked to assess how accurately each
item described them on an 8-point Likert scale ranging from “extremely
inaccurate" to "extremely accurate." Scores were derived by summing the
responses for each item in each scale, resulting in scores ranging from i6
to i28.

Circumplex models of interpersonal behavior have been viewed as
"reflections of the cognitive ordering of interpersonal behavior by
observers, rather than an objective ordering of these behaviors by actors
in the real world" (Gifford & O'Conner, I987, p. iOI9) (Shweder &
D'Andrade, l979; Conte & Plutchik, I98l ). Wiggins' scales have been
shown to possess predictive validity (Gifford & O'Connor, I987) and

adequate internal consistency (Wiggins, i979).

50

Moving beyond the quantification of reported acts as behavioral
criteria with which to compare interpersonal measures, Gifford ( l982)
related personal space or preferred interpersonal distance (iPD) to
Wiggins' Interpersonal Adjective Scales. in this study participants
specified the amount of iPD they were most comfortable within 18
vignettes. As expected, the strongest positive correlations occurred
between iPD preference and Cold--Ouarrelsome (.50) and the strongest
negative correlation occurred between iPD preference and
Extraverted--Gregarious (-.35). Thus, Cold-~Ouarrelsome accounted for
from 25% to 50% (Ozer, i985) of the variances associated with personal
space preferences.

Extending this self -report finding, Gifford and O‘Conner ( I 987)
correlated acts initiated and gestures made during group conversations
with responses on the iAS. While few of the correlations were strong,
behaviors and questionnaire responses were generally related in the
expected directions.

in addition to their apparent validity, Wiggins' interpersonal scales
appear to be a particularly appropriate measure to use in the assessment

Of this StUdYS participants because they measure inter personal posture.

Si

BLQQQQLILQ

To assure that pictorial stimuli presented affiliative or achievement
situations, the pictures were separately rated by 38 undergraduate women
and 33 undergraduate men. The women received the series of female
pictures and the men received the series intended for males.

The primary group of participants for this study was obtained from
the Michigan State University research participant pool. The title used to
inform participants about the study was “Gender Differences in
Personality." This title appeared on the participant sign-up sheets for the
purpose of data collection. Both a female and a male monitor were present
at each of several data collections to minimize experimenter effects.
They both looked like they were in their mid-twenties and dressed
casually but neatly.

in same-sex groups ranging from l0 to 25, data were separately
collected from men and women. Each was given a questionnaire packet
containing: instructions (Appendix F), an informed consent sheet, the six
pictorial stimuli, paper on which to write these responses, the PAC, and
the iAS. They were asked to write these stories before completing the
PAO and IAS. After receiving these questionnaire packets, they were also

given a verbal explanation or the procedures and the instructions were

52

reviewed.

The pictures were given with an adaptation of the Atkinson (l958)
group administration format which appears in Appendix F. The
instructions were altered only slightly to fit the current study. The most
significant change is due to the fact that these pictures were presented in
Xeroxed form instead of with a slide projector. Using procedures similar
to those of Pollak and Gilligan, students were asked to write brief,
imaginative, dramatic, and interesting stories for each picture. The PAO
and the lAS were given with standard instructions.

Following the collection of data each participant received a written
debriefing sheet (Appendix G).

RESULTS
i r A 'v I

The lAS (Wiggins, i979) was used to determine whether this sample
represented similar undergraduate populations on the two principal
dimensions personality generally denoted as Dominance and Warmth. The
means and standard deviations on these scales and six less central
dimensions of personality appear in Table 2. Scores of the individuals in

this study appear with data collected by Wiggins from 523 research

53

Table 2
Interpersonal Adjective Scales and Scores

Mann: W
Lemates £19.11: Eemates £191.95
ted 4532 1.6.6. use use use LAB. use

Ambitiousl 5.32 5.72 5.35 5.87 .92 1.01 .87 .95
Dominant

Arrogantl 3.59 3.60 4.19 4.30 1.12 1.07 1.07 1.02
Calculating

(2010/ 2.29 2.48 2.82 2.95 .91 .80 1.01 .89
Guerrelsome

Aloofi 3.11 3.87 3.59 4.28 .97 1.17 1.11 1.19
Introverted

Lazg/ 3.44 4.13 3.42 3.95 1.04 1.01 1.05 .94
Submissive

Unassuming/ 4.56 5.14 4.06 4.64 .96 1.00 .89 .93
Ingenuous

Warm! 6.39 7.08 6.00 6.64 .90 .76 -95 -80
Agreeabie

Gregorious/ 6.10 6.54 5.70 6.17 .67 .96 .ee .96
Extravarted

Grand ”gong 4.22 4.82 4.39 4.85 .96 .97 .96 .96

'LAB (present) sample, n‘s = 90 women and 90 men

2WIggins, n's .-. 328 women and 195 men. See note in text (0. 54) concerning the atypical
method that Wiggins used to score all scales in this sample.

54

participants in four separate studies (Wiggins, 1979). The mean scores
for females in the current study and Wiggins‘ sample correlated .98 (n <
.001) and the mean scores for the males in the current study and Wiggins'
sample correlated .99 (p < .001 ). The scores of this studies participants
averaged .46 points lower than those of Wiggins' sample. This difference
is due to the use of different scoring systems by Wiggins and by the
current author. While Wiggins‘ scale normally has an eight-point range, he
used a nine-point scale In the sample cited. The fifth position
was used as a score for items which had been skipped or not marked boldly
enough on a computer sheet. The nine-point scale shifts scores between 5
and 8 up one point. The present work used the eight-point scale which
accounts for the difference in scores.

W

The ratings for the affiliative and achievement qualities contained in
the pictures were obtained from the Card Rating Form shown in Appendix
C. Mean ratings of each picture‘s pull for af f iliativeness and achievement
were compared with two tailed t-tests. The results appear in Table 3 and

Indicated that all pictures were significantly more likely to elicit the

intended themes (:2 < .01 ).

55

Table 3

Rated Affiliation and Achievement Pulls of Pictorial Stimuli (n -.- 71)

Stimuli

Moderate
Opposite-Sex
Affiliation

Intense

Opposite-Sex
Achievement

intimate
Opposite-Sex
Affiliation

Moderate
Opposite-Sex
Achievement

Moderate
Same-Sex
Achievement

Moderate
Same-Sex
Affiliation

11291151212612:

mum mm
£12m SD.

3.96 1.44 2.55 1.44
3.55 1.65 5.66 1.51
6.54 1.00 4.41 2.12
3.66 1.45 4.76 1.53
3.60 1.74 5.54 1.47
6.10 .93 3.69 1.93

6.01

-8.21

7.88

-5.00

-7.89

8.58

70

7O

70

70

70

70

2-tail

00‘

.01

00‘

.01

.01

.Ol

56

Several of the pictures were perceived as depicting situations of
affiliation and achievement which placed them at intended levels of
intensity. Intimate Opposite-Sex Affiliation was rated as containing
significantly more affiliation than either Moderate Opposite-Sex
Affiliation (t = 14.58, n < .0001) or Moderate Same-Sex Affiliation (t =
3.18, n < .05). Moderate Same-Sex Affiliation was seen as significantly
more at f iliative than Moderate Opposite-Sex Affiliation (t = 12.55, n <
.0001 ) (Table 4). This was likely due to the caption accompanying the
Moderate Same-Sex Affiliation picture which read, "Phillip and Brian/
Caroline and Theresa have been friends for two years." Although the
picture parallels the Moderate Opposite-Sex Affiliation picture, the
caption replicates the one accompanying the intimate Opposite-Sex
Affiliation picture because it mentions "two years" instead of the intended
two weeks. This error illustrates the impact written material can have on
projective stimuli. The impact of this oversight is discussed in the
Analysis of Variance section which reviews individual hypotheses.

The Intense Opposite-Sex Achievement picture was rated as depicting
significantly more achievement than the Moderate Opposite-Sex
Achievement picture (t = 4.5, n < .01) but not more than the Moderate

Same-Sex Achievement picture. The Moderate Same-Sex Achievement

57

Table 4

Rated Intensity Levels of Affiliation and Achievement Pictorial Stimuli*

M Kim titanic MW

intimate Opposite-
Sex Affiliation x
Mabrate ODDOSIte-

Sex Affiliation 2.56 l 4.58 70 .0001

intimate Opposite-

Sex Affiliation x

Metbrate Same-

Sex Affiliation .44 3.18 70 .002

Moderate Opposite-

Sex Affiliation x

Moderate Same-

Sex Affiliation 2. 14 12.55 70 .0001

Intense Opposite-

Sex Achievement x

Moderate Opposite-

SexAchievement .89 4.60 70 .0001

intense Opposite-

Sex Achievement x

Moderate Same-

SeiiAchievement .13 .63 70 .53

Methrate Opposite-
Sex Achievement x
Momrate Same-

Sex Achievement -.76 -3.69 70 .0004

”fl-7i

58

picture was also given a higher achievement content rating than the
Moderate Opposite-Sex Achievement picture (t = -3.69, n < .05).

The significance levels of these t-tests were adjusted because the
number of tests performed exceeded the treatment variable degrees of
freedom present (gt = 5). Six pictures were assessed and 12 t-tests were
performed. The adjustments were made based on the calculations of Sidak
(1967). The actual adjusted levels of significance were taken from a table
produced by Games (1977).

The Moderate Same-Sex Achievement picture pulled more strongly for
violence than intended, apparently due to the presence of physicians in the
picture and the accompanying caption, "Peter and David/Julia and Sharon
are physicians. They are discussing a new cancer drug." The mention of
cancer drew a large number of responses related to death and f atai illness.
Consequently, stories written in response to this picture were not
included In the current analysis.

This exclusion only interfered with female Hypothesis 2 directed at
gender differences in achievement situations. This hypothesis contrasted
Moderate Same-Sex Achievement with Moderate Opposite-Sex
Achievement. It was predicted that women would write more stories

containing violence to scenes of Moderate Opposite-Sex Achievement.

59

Table 5 shows that women wrote only four stories in response to pictures
of Moderate Opposite-Sex Achievement which were scored as containing
violence. This frequency, makes it unlikely that stories written by women
in response to Moderate Opposite-Sex Achievement would contain
significantly more scores for violence than those written about scenes of
Moderate Same-Sex Achievement.

The Moderate Opposite-Sex Achievement picture's sequential
placement apparently also influenced the frequency of the mention of
violence in subsequent pictures. This will be discussed in a review of
order effects produced by the pictures.
inteccateLBeliabim

Two-hundred participants (100 females and 100 males) each
contributed sets of six stories to the current study. Twenty sets of
stories (10 male and 10 female) were randomly selected and used to train
raters. The raters were two undergraduate students (a male and a female)
who were blind to this studies hypotheses. They received independent
Study (PSY 490) course credits for their assistance with this project.

They were trained to rate stories for the presence or absence of

violence according to Pollak and Gilligan‘s scoring schema. The guidelines

60

Table 5

Frequency of Stories Containing the Mention of Violence (all n's = 90)

Moderate Moderate Moderate intimate intense
Same-Sex OppositecSex Opposite-Sex Opposite-Sex Opposite-Sex Total
Affiliation Affiliation Achievement Affiliation Achievement Violence

m

Total 6 6 4 16 2
Violence

11913.5

'TL_

 

Total 12 5 6 16 s

44
Violence

19.1.31 20 1 1 10 32 7 60

61

they used to rate these stories are in Appendix H. Violence included the
mention of homicide, suicide, death by accident, rape or forcible violation,
physical assault, kidnapping, or fatal disease. if violence as defined by
the scoring schema was mentioned once or more, stories were given a
violence rating of 1.

After an Initial rating of the training set of 20 stories, it was also
decided to include the mention of abortion as violence because it can be
considered a human death and is thought by some to be a homicidal act.
Although this addition deviates from Pollak and Gilligan‘s scoring system,
there were only five mentions of abortion, leaving data sets generally
comparable. Interestingly, four mentions of abortion appeared in stories
written by men. Three of these were written in response to Intimate
Opposite-Sex Affiliation scenes.

Similar to the methodology of Pollak and Gilligan, each story which
obtained a violence rating was also rated for it's content. Content ratings
assessed whether stories scored for mentions of violence were generally
situations involving affiliation or achievement. In addition to the content
classifications of affiliation and achievement, the options of both
affiliative and achievement, and neither af f iiiative nor achievement, were

available to raters. An example of a story containing elements of both

62

affiliation and achievement might be one that described a husband and
wife working together in the same business. Stories containing neither
affiliation nor achievement might have been written about only one of the
picture characters or might have had little to do with the content of
pictures.

The initial rater training session included a general orientation to the
study, an explanation of scoring procedures, and a discussion of some
possible rating dilemmas. Following this training, the set of 20 randomly
selected stories were rated for violence and content by each of the raters.
Both raters agreed on the presence of violence in 83% (5 out of 6) of these
stories. Of these five stories that received violence ratings from both
raters, 60% (3) were given the same content ratings.

A second phase of training included a review of ratings that had been
given by each rater and a discussion of rating discrepancies. Agreement
was reached on several stories which had been difficult to score. It was
during this discussion that it was decided that abortion would be given a
score for violence.

After this retraining raters scored the stories of the larger group of
180 (90 male and 90 female) participants for the mention of violence and

story content. After each rater had scored 90 stories, raters were asked

 

63

to rescore the original group of 20 training stories to see if agreement on
their content ratings had changed. In the second rating of the set of 20
stories, raters agreed on the presence of violence in 7 of the 8 stories
scored for violence. They agreed on the content ratings of the six stories
which they had both rated as violent.

Although the raters obtained high interrater reliability for the small
group of 20 stories, this agreement declined in the larger set of stories.
Of the 900 total stories written by participants (5 each, excluding stories
written In response to Moderate Same-Sex Achievement), raters
cumulatively reported 88 mentions of violence. They agreed about the
presence of violence in 63% (55) of these stories, 95% (52) of which they
also gave the same content rating.

The low interrater agreement on the large set of stories appeared to
be due to difficulties in rating assaultive situations, confusion about the
scoring of abortion, and general rater disagreement. There were several
instances when the occurrence of death was overlooked or unclear when
terms such as "late" ("late husband“) or "loss" ("the loss of a friend“) were
used to describe people. In some stories it was not clear whether
alcoholism could be considered a fatal disease and vague phrases like

"serious accidents“ had to be scored in the context or the entire Story.

64

Comparatively minor assaults were sometime difficult for raters to agree
upon. In one story, a character vomited on another and it was not clear
whether this act was an intentional assault. Ultimately,this story did not
receive a score for violence. On several occasions one rater gave a
violence score to the mention of abortion only when it was carried
through. The decision had previously been to score the mention of abortion
as violent in accordance with mentions of other forms of violence. Death
of animals or violence directed at animals was also counted as violent in
this system. One normally very conscientious rater, seemed to miss
several blatant mentions of violence in the larger set of data. These
oversights may be related to preoccupations with the death of her brother
which occurred midway through the coding process.

in determining the final scores for violence and content, stories with
discrepant ratings were appraised by a third rater (Rater C) who was also
blind to the study's hypotheses and the story writer's sex. The use of a
third rater to resolve rating discrepancies was also used by Pollak and
Gilligan. Rater C was a 24 year-old male graduate of Michigan State
University working in the local area. After being trained to use Pollak and
Gilligan's scoring system for violence, he rated the set of 20 stories

which had been used to train the other raters. After their second training,

65

the first two raters (Rater A and Rater B) had agreed on 7 out of 8
mentions of violence and on 6 of the content ratings for these 7 stories.
The third rater agreed with Rater A on all violence scores and with and
Rater B on all but one story. There was also agreement between Raters A
and C on 7 content scores and between Raters B and C on 6 content scores.

After demonstrating reasonable interrater agreement with the two
original raters, Rater C was asked to score the 33 stories which had
received discrepant scores of violence by the first two raters. Rater C‘s
data were used to force a consensus between raters. Consensus was
reached for all of the violence scores and all of the content scores except
for one given a "neither" content rating by Rater A, no rating for violence
by Rater B, and an achievement content score by Rater C. This content
rating was excluded from subsequent analyses.
WW

Based on the initial ratings of Raters A and B and those later added
by Rater C, the total number of stories containing violence for each
stimulus condition appears in Table 5. Men collectively wrote 44 stories
which mentioned violence at least once and women wrote 36 stories which

contained violence. Responses to the Intense Opposite-Sex Affiliation

66

stimulus constituted 40% (32 of 80) of all stories containing violence.

in addition to these data, Table 6 presents the percentages of men and
women who wrote stories containing violence in response to either
affiliation pictures or achievement pictures, or to both affiliation and
achievement pictures. The affiliation and achievement categories in this
table are composed of separately pooled ratings of responses to the
affiliation and achievement pictures.

The frequency of scored violence for all participants appears in
Appendix 1. Ten of 30 and six of 24 women wrote more than one story
scored as containing violence. Three men and two women wrote more than
two stories scored to contain violence.

in Table 6, the current data are also compared to the findings of
Pollak and Gilligan and those of Benton et al. This comparison includes
both 4 and 5 picture sets. The four picture set excludes the Moderate
Same-Sex Affiliation picture, making it more comparable to the data of
the other authors. The inclusion of the Moderate Same-Sex Affiliation
picture Increased the potential number of violent ratings in the affiliation

category because it is not balanced by an achievement picture.

67

Table 6

Percentage of Participants Who Wrote Stories Containing Violence Elicited
by Achievement and Affiliation:
A Comparison of Three Studies

min
Total No
Author: Sex Affiliation Annimmcm Bath Minions: Violence
Pollakt. M-88 25% 6.8% 19.3% 51% 49%
Gilligan
Benton, et al. M-90 20% 7% 45% 55%
Brooks
Four Pictures M-90 18.8% 5.6% 5.6% 22.2% 77.8%
Brooks M-90 22.2% 5.6% 7.8% 35.5% 64.4%
Five Pictures
Emma
19.1.31 tie

Author: 5.8! WWMWW

Pollak 8. F-SO 6% 16% 0% 22% 78%
Gilligan

Benton et al. F-i 14 12% 5% 24% 76%
Brooks F-90 1 7.8% 2.2% 3.3% 23.3% 76.6%
Fou‘ Pictures

Brooks F -90 21.1% 2.2% 3.3% 26.7% 73.3%

Five Pictures

68

Pollak and Gilligan reported that 25% of their men, as compared to 6%
of their women, wrote stories containing violence to pictures of
affiliation. In contrast, 6.8% of the men and 16% of the women wrote
stories containing violence to scenes of achievement. While no women
wrote stories containing violence to scenes of both affiliation and
achievement, 19.3% of the men wrote such stories. Altogether, 51% of
Pollak and Gilligan‘s men and 22% of their women wrote at least one story
scored as containing violence.

Of the present study men, 22.2% responded to the four picture series
with at least one story containing violence, as did 23.3% of the women.
The percentage of men who wrote stories containing violence in the
present study is much smaller (22.2% vs 45%) than that of Pollak and
Gilligans‘ investigation. Another notable difference between these two
studies is that 15.5% of the present women as compared to 6% of Pollak
and Gilligans' women wrote stories containing violence in response to
scenes of affiliation. Additionally, only 2.2% of the women wrote stories
containing violence in response to achievement stimuli, in comparison to
Pollak and Gilligans' 16%.

The present findings more closely approximated those of Benton et al.

who found that 20% of the men and 12% of the women wrote stories scored

6‘9

:3" containing Violence to affiliation stimuli and that 7% of the men and 5%
of the women wrote stories containing violence to achievement stimuli.
These figures must be confirmed through correspondence with Benton et
al. because they do not coincide with the total violence scores reported by
these authors. Benton et al. stated that, "Concerning the amount of
violence imagery, 45% of the male and 24% of the female respondents
expressed hostility to at least one of the TAT cards" (p. 1 170). The
percentages of violence reported for men (20% and 7%) and women (12%
and 5%) in af f iliative and achievement situations do not add to the total
frequencies for mentions of violence. Perhaps Benton et al. did not equate
violence with hostility.

Appendix J gives a detailed breakdown of the types of violence in the
present responses scored as violent. Some responses included more than
one type of violence, accounting for the differences in total mentions of
violence in this table and the frequency of stories scored as containing
violence in Table 5. Death by accident was the most frequently occurring
type of violence in stories written by both men and women. Fatal illness
and homicide were the second and third most frequently mentioned types

of violence.

 

70

Content Analysis

In addition to the independent content ratings (by 71 undergraduates)
of affiliative and achievement pictures, a content analysis was made of
all stories scored for violence. The purpose of this analysis was to
examine whether the content of stories differed from the intended theme
of the stimuli. In other words, were respondents writing stories about
affiliation to scenes of affiliation?

Each story scored for violence received, a content rating of
affiliation, achievement, both, or neither. This differed from Pollak and
Gilligan‘s content analysis in one significant respect. Pollak and Gilligan
rated the specific context in which violence occurred as either af f iliative,
achievement, both, or neither; whereas, the current study rated the
stories which contained violence as affiliative, achievement, both, or
neither. This was an unintended departure from Pollak and Gilligans'
methodology. Helgeson and Sharpsteen ( 1987) also performed a content
analysis that was similar to that of the present study. Pollak and
Gilligans' limited description of their content analysis may have
contributed to these deviations.

Pollak and Gilligan's content analysis results did not differ

Significantly from their findings analyzed “by situations" (Table 6).

71

Although the current content analysis was not fully comparable to theirs,
it has prov1ded information about whether the content of violent stories
differed from the intended theme of the stimuli.

The results of the content analysis, given in Table 7, show the
content ratings of pictures in each category. Based on the three rater
consensus, 91% of the stories scored as containing violence were in
categories congruent with their stimulus type (affiliation vs.
achievement). This means that stories scores scored for violence in
response to pictures of affiliation usually focused on relationships, while
responses to pictures of achievement which contained violence, generally
centered around issues of competition and accomplishment. Because story
content linked so strongly to category of stimuli, a separate analysis
based on story content seemed prof itless.

Analxsisofiamnce

A repeated measures ANOVA was performed on the violence scores
agreed upon by raters (80 ratings of violence, Table 5) for men and women
across pictorial stimuli. There was a main effect for mention of violence
across pictures (_F_ = 8.6, n < .0001), but no main effect for respondent's sex

(E = .56, n < .46) (Table 8). The main effect for the mention of violence

72

Table 7

Content Analysis of Stories Scored as Containing Violence (all n's = 90)

Moderate Moderate Moderate intimate intense

Same-Sex Opposne-Sex Opposite-Sex Opposite-Sex Opposne-Sex
Affiliation Affiliation Achievement Affiliation Achievement Totals

1.2mm

Lemons
Affiliation B 6 1 16 o 29
Achievement 0 o 3 o 2 5
Both 0 o 0 o 0 0
Neither 0 0 0 0 O 0
TotalVioience B 6 4 15 2 36
No Violence 82 B4 86 74 an 414

119.133.
Affiliation 11 5 1 16 1 34
Achievement 0 o 4 0 I 5
Both 0 0 0 3 3
Neither 1 0 0 0 o 1
Total Violence 12 5 5' 16 s 43
No Violence 78 85 84 74 85 407

’1 content rating excluded due to rating disagreement

73

Table 8

Analysis of Variance

Sex x Pictorial Stimuli

m Qt §§ 11$. E P.
Ortbr (A) l .071 .071 .555 .457
Pictorial Stimuli (8) 4 2.03 .575 8.6 .0001
(AB) 4 .096 .024 .357 .839

Subjects 1 78 22.82 . 128
within Groups
8 x suiects within 712 46.60 .067

Groups

was probably largely due to the greater number of stories scored as
containing violence by both men and women in response to the Intimate
Opposite-Sex Affiliation picture. There was no interaction between sex
and the mention of violence (E = .36, n <84).

l-tests were used to assess individual hypotheses and results are
given in Table 9.
MW

1. To all pictorial stimuli, men will write significantly more stories
scored to contain violence than will women. Rejecting this hypothesis,
men did not write significantly more stories scored as containing violence
than did women (E - .56, n < .46).

2. Men will write significantly more stories scored to contain

violence in response to scenes of intimata annaita-Sex Affiiiatinn as
compared to scenes of Mngnrata annaitg-Sax Aftiliatinn (M2 versus M1).

As expected, the stories men wrote in response to scenes of Intimate
Opposite-Sex Affiliation were more frequently scored as containing
violence than those stories written about scenes of Moderate Opposite-Sex
Affiliation (t = 2.60, n <01).

3. Men will write significantly more stories scored to contain

violence in response to scenes of W as

Commune

Intimate Opposite-Sex
Affiliation x Moderate
Opposite-Sex Affiliation

Moderate Opposue-Sex
Affiliation x Moderate
Same-Sex Affiliation

intimate Opposite Sex
Affiliation x Intense
Opposite-Sex Achievement

Intimate Opposite Sex
Affiliation x Intense
0pp051te-Sex Achievement

intense Opposite-Sex
Achievement x Moderate
Opposite-Sex Achievement

n = 90 males 6. 90 females

75

Table 9

Comparisons of Individual ngotheses

5221 mm
11 .12
n -.06
11 .12
r -.16
r 4022

2.60

-i.97

2.96

-3.75

- 1.00

m

89

89

89

89

89

2"! 'I I!

.012

.052

.004

.0003

76

compared to scenes of Modfiate Same-Sex Affiliation (M1 versus M5).
Rejecting this hypothesis, men wrote significantly more stories
containing violence in response to the scene of Moderate Same-Sex
Affiliation than to the picture of Moderate Opposite-Sex Affiliation (t =
-1.97, n < .05). This may in part be due to the more intimate caption
accompanying the Moderate Same-Sex Affiliation picture ("Phillip and
Brianl Caroline and Theresa have been friends for two years"). When
comparing Moderate Same-Sex Affiliation and Intimate Opposite-Sex
Affiliation responses which shared similar captions, the discrepancy
between scores was slight (12 vs. 16), implicating the importance of
situational intensity versus sex of story character.

4. Men will write significantly more stories scored to contain

violence in response to the scene of intimate anneite-fiex Af f iliatinn as
compared to the scene of intense anneitefieg nehieyement (M2 versus

M4). This hypothesis was sustained (t = 2.96, n < .01).

W

1. Women will write significantly more stories scored to contain
violence In response to scenes of W as
compared to scenes of moderategmsitegfiexmentent (F3 versus F 4).

This hypothesis was rejected it = 1.0, n < .32).

2. Women will write significantly more stories scored to contain

violence in response to scenes of Mnnenate anneite-Sex Aenievement as
compared to scenes of Monerate fiame-Sex Aehievement (F3 versus F 6).

Although this comparison was not performed because of the earlier
noted problems with the Moderate Same-Sex Achievement, it seems
unlikely that this hypothesis would have been supported due to the very
low number of violent responses in this category (4 stories of 90).

3. Women will write significantly more stories scored to contain

violence in response to scenes of W as
compared to scenes of Intimate annsite-fiex Attiliatinn (F4 versus F2).
Sharply rejecting this hypothesis, women wrote significantly more stories
scored as containing violence in response to scenes of intimate Af f iliatinn
witn tne anneite-iex (t = -3.75, n < .0003).

it was not necessary to adjust the levels of significance for these
t-tests because the number of comparisons performed (5) was less than
the number of treatment (or pictorial stimuli) levels.
W

The pictures used in this study to gain information about
self -def inition and interpersonal posture were administered in the five

different orders noted Table 1. Violence scores derived from these

78

different sequences are shown in Table 10. A repeated measures ANOVA
contrasting pictorial stimuli and order of pictorial stimuli was performed
(Table 11). There was no main effect for order (E = 1.70, n < .15) but there
was a main effect for the number of violent responses across pictorial
stimuli (E = 8.80, n < .0001). There was also an interaction between
picture order and violence across pictorial stimuli (E = 1.87, n <02).

Although data collected from the Moderate Same-Sex Achievement
picture was excluded from the present analyses, it appears that the early
viewing of this card may have had a significant impact on the number
stories written by men which contained violence. Of the male stories
which were scored as containing violence, 45% (20) were evoked by the
sequence of stories in which the Moderate Same-Sex Achievement
(Sequence 11) picture appeared first. Sequence 1, featuring Moderate
Same-Sex Affiliation first, generated 22.7% (10) of the males‘ stories
scored as containing violence.

The largest number of stories containing violence and written by
women occurred in the sequence which presented the Intimate
Opposite-Sex Affiliation picture at the beginning (Sequence lli). Eleven or

32.4% of the female stories with violence appeared In this group. The

sequence of pictures that pulled the second largest number of stories

79

Table 10
Frequency of Stories Containing Violence in Five Picture Sequences
Moderate Moderate Moderate intimate intense

Picture Orders Same-Sex Opposite-Sex Opposite-Sex Opposite-SexOpposite-Sex
Affiliation Affiliation Achievement Affiliation Achievement Totals

Males
Sequence 1 5 2 2 1 0 10
Sequence 2 5 1 3 7 4‘ 21
Sequence 3 O 2 1 2 i 6
Sequence 4 o 0 0 2 0 2
Sequence 5 2 0 0 4 0 6
sum 12 5 6 16 5 44

Emotes

Sequence 1 2 2 I 3 O 8
Sequence 2 0 0 1 2 1 4
Seqdence 3 3 2 2 4 ~ 0 ll
Sequence 4 I 2 0 1 1 5
Sequence 5 2 0 0 6 0 8
Sum 8 6 4 16 2 36

n=90 males 8. 90 females

Solute

Orthr (A)

Pictorial Stimuli (81
(AB)

Subjects

within Groups

8 x sujects within
Groups

1 75
700

80

Table 1 1

AnaIySis of Variance

Order x Pictorial Stimuli

Si

.86
2.3
1.96

22.03
45.74

Elf:

.21
.58
.12

.13
.07

E E

1.7 .15
8.8 .0001
1.87 .02

 

81

scored as containing violence from women was the set which began with
Moderate Same-Sex Achievement (Sequence 5, 8 stories or 23.5%).
Relatinnshin nf Vinlenee fienree tn the PAO

The Androgynous, Feminine, Masculine, and Undifferentiated
categorizations are based on the median-split method of scoring. This
method uses the rounded-of f average of male and female median scores as
an intercategory cutoff. The median scores for females on the Masculinity
and Femininity Scales were 21 and 25. The males parallel medians were
22.5 and 23. Thus, the averaged median would be 22 ([21 + 22.5] / 2) for
Masculinity and 24 ([25 + 231/ 2) for Femininity. Very similar medians of
21 on Masculinity and 23 on Femininity were reported from data based on
715 college students by Spence & Helmreich (1978).

Of the 90 females in this study, 16 (18%) described themselves as
Androgynous, 38 (42%) as Feminine, 18 (20%) as Masculine, and 18 (20%)
as Undifferentiated. Twenty-three (25.5%) of the males rated themselves
as Androgynous, 17 or 18.8% as Feminine, 26 or 28.8% as Masculine, and 24
or 26.6% as Undifferentiated.

WM

4. Women who write stories stored as containing violence in

response to scenes of WW will have

82

significantly higher PAO Femininity scores than women who do not write
stories scored as containing violence in response to scenes of intense
anneite-fiex Achievement (M2 and PAO F scores). This hypothesis was
rejected because the mean PAO Femininity score of women who wrote
stories containing violence to scenes of Intense anneite-Sex
Achievement (m = 26) did not differ Significantly from those of the other
women (131 = 24.40). Although not statistically, this difference was in the
expected direction.

5. Men who write stories containing violence in response to scenes of
intimate anneite-Sex Affiliation will have significantly higher PAO
Masculinity scores than men who do not write stories scored as containing

violence in response to scenes 0f Intimate QQQQSite‘fiezs AI filiatinn (M2

and PAO M scores). As with women, men who wrote stories scored as

containing violence to scenes of intimate anneite-Qex Affiliatinn ([31 =
22.25) did not differ significantly on PAO Masculinity from the other men
([1 = 22.19).
DISCUSSION
This study attempted to clarify the debate between Pollak and
Gilligan, and Benton et al., regarding differences in the mention of

violence, in the written responses of men and women to scenes of

83

achievement and affiliation. in addition to being contradictory, the
findings of these previous authors appeared limited due to their use of
stimuli unbalanced for the sex of depicted characters and inattentive to
the intensity level of the situations presented. The present work
addressed these issues by constructing equivalent sets of pictorial
stimuli which presented achievement and affiliation themes using same-
and opp051te-sex pictorial characters at both moderate and high levels of
intensity.

It was hoped that stories written in response to the projective
materials created for this study would yield further insights into how men
and women experience affiliation and achievement situations. Although
the roles of men and women in our society are rapidly changing and
merging in some important respects, numerous theoretical and empirical
works (Horner, 1968; Chodorow, 1974; Gilligan, 1977; Eccles, 1987)
continue to indicate that men and women perceive and respond differently
to interactions featuring intimacy, competition, and success. These
differences have been attributed to issues ranging from self -def inition to
variations in personality originating in separation from
a female primary care-giver.

in file Reproduction oi/‘iotner/ng, Chodorow provided a penetrating

84

look at how men and women develop different personality structures in
response to separation from a female primary care-giver. She believes
that in their effort and ability to separate clearly and distinctly from an
opposite-sex care provider, males develop more autonomous personalities
and move away from some of the affiliative bonds associated with
preoedipal relationships. In addition to an evolutionary path which has
propelled men away from the role of nurturer and into the role of provider,
Chodorow posits that the separation process encourages the formation
of clearer ego boundaries in men than in women and causes men to
deemphasize the importance of relationships.

Paralleling men's historic exclusion from care-giving processes,
women have been excluded from socially recognized professions and denied
the opportunity to acquire resources. The struggle faced by women in
adapting to achievement environments were illustrated in Horner's (1968)
doctoral dissertation. She alleged that some women experience what she
referred to as, ”fear of success" in competitive situations. In her study
the anxiety surrounding accomplishment appeared to weaken women‘s
motivation to compete against men. Some of the experiences women have
in achievement situations may arise because women tend to place a higher

priority on relationships than do men. Success is often accomplished

85

through assertive striving wherein one person gains competitive Victory
over another. Women have typically been discouraged from competitive
and aggressive behavior because these behaviors can be perceived as
threatening to the needs of others (Bernardez, 1983) which conflicts with
the maternal expectations placed on women. Consequently, it may be
especially anxiety-provoking for women to seek success in competitive
occupations.

Gilligans' book, ln aflifierent Voice, has resonated with the
perceptions of many women. She illustrated that men and women seem to
define themselves and their purposes differently and consequently make
moral decisions based upon different sets of priorities. interviewing men
and women about how they made personal moral decisions, Gilligan
observed that men appeared more concerned with rights and responsibility
and that women seemed to place a greater emphasis on the maintenance of
relationships and care of others.

Gilligan linked her observations about these divergent perspectives
with the findings she shared with Pollak 0n the mention of violence in
af f iliative and achievement scenes. These authors believed that men more
frequently wrote stories containing violence in response to scenes of

affiliation and women more frequently wrote stories containing violence

86

in resoonse to scenes of achievement due to their differing self-
definitions and resulting sets of priorities.

The present findings did not support the empirical work of Pollak and
Gilligan. The frequency at which men and women wrote stories containing
violence did not differ across the present pictorial stimuli. They
separately wrote similar numbers of stories scored as containing violence
in response to both af f iliative and achievement stimuli and the scenes
featuring affiliation elicited more stories containing violence from both
men and women. These findings are based on both the number of men and
women who wrote stories containing violence in response to affiliative
and achievement situations (Table 6) and also on the number of stories
scored as containing violence for each category of pictorial stimuli (Table
5).

The findings of this study are more similar to those of Benton et al.,
who were unable to reproduce the results of Pollak and Gilligan in either a
replication or in an extension which used two additional pictures. Benton
et al. reported that men and women more frequently wrote stories
containing violence (20% and 12%) to scenes of affiliation than to scenes
of achievement (7% and 5%). In the present study, using the set of two

oppOSIte-sex affiliation (Intimate and moderate) pictures and two

0pp051te-sex achievement pictures (Intense and moderate), 18.8% of the
men and 17.8% of the women wrote at least one story scored to contain
violence to a scene of affiliation (Table 6), but only 5.6% of the men and
2.2% of the women wrote at least one story which was scored as
containing violence in response to scenes featuring achievement (Table 6).
These findings seem likely to be generally representative of other
undergraduate populations based on the nearly identical IAS and PAO
scores of this study's participants and other undergraduate populations.
Helgeson and Sharpsteen (1987) recently conducted a study in which
they used only the Acrobatic picture to elicit stories from 50 men and 61
women. This picture was accompanied by either an affiliative or an
achievement caption to focus the respondent's attention. Although their
findings supported Pollak and Gilligan's, this may have been due to the
smaller number of subjects in each experimental condition (25) and to this
picture's implicit potential for violence. Additionally, some subjects
continued to perceive this picture as an achievement scene even when the
affiliative caption was presented. There was no significant difference in
the number of stories classified as affiliative (49.1%) or achievement
(42.1%) wnich had been written in response to the affiliative stimuli.

A11 stimuli created for the present work were independently judged

88

as depicting either the intended themes of achievement or affiliation.
Potentially more reflective of real-life affiliation and achievement
situations than previous sets of pictorial stimuli, both the intimate
Opposite-Sex Affiliation and Intense Opposite-Sex Achievement pictures
were also judged to contain a higher intensity of their intended content
category than their moderate counterparts. This allowed for the
assessment of perceptions at two different levels of affiliation and
achievement.

The assessment of differences due to sex of featured characters was
complicated because the Moderate Same-Sex Achievement stimuli pulled
for violent imagery through its caption's mention of cancer, and the
caption of the Moderate Same-Sex Affiliation stimuli pulled for intimacy,
reading "two years" instead of "two weeks." it is noteworthy that the
Moderate Same-Sex Achievement scene also received the highest ratings
for contained achievement imagery. This is probably due to the presence
of physicians in the picture.

The men who participated in this study responded differently to high
levels of affiliation and achievement stimuli and t0 Intimate and moderate
levels of affiliation. They produced more stories containing violence in

intimate Affiliation as opposed to intense Achievement situations (t =

89

2.96, n < .01) and they more frequently wrote stories containing violence
in response to scenes of Intimate Opposite-Sex Affiliation as compared to
Moderate Opposite-Sex Affiliation (t = 2.60, n < .01). The differences
expected to emerge from sex of story character in af f iliative situations
were not found. Men wrote more stories scored as containing violence in
response to scenes of Moderate Same-Sex Affiliation than to Moderate
Opposite-Sex Affiliation (t = -1.97, n < .05). It is not clear whether this is
due to the presence of same-sex intimacy or the higher level intimacy
elicited by the Moderate Same-Sex Affiliation stimuli and it's "two year"
relationship caption.

Women also wrote more stories scored as containing violence in
response to scenes of intimate affiliation as compared to scenes of
intense achievement and to scenes of intimate affiliation as compared to
scenes of moderate affiliation stimuli. Although the comparison of
opposite-sex and same-sex picture characters in achievement situations
containing women were not performed, it would have been unlikely for the
Moderate Opposite-Sex Achievement picture to have stimulated a
significantly greater number of stories containing violence in any
comparison due to its infrequent evocation of responses scored for

violence.

90

The mention of violence in stories seemed strongly influenced by the
order of picture presentation. Men wrote more stories containing violence
in the sequence Which first featured Moderate Same-Sex Achievement,
followed by Moderate Same-Sex Affiliation (45.5%). The sequence which
received the second greatest number of violent stories by men began with
Moderate Same-Sex Affiliation and presented Moderate Same-Sex
Achievement second (23.5%). As discussed previously, the early
presentation of the Moderate Same-Sex Achievement picture containing
the "cancer" reference, may have set the tone for other violent stories.
The Moderate Same-Sex Affiliation picture may have caused men to
increase their violent responses in defense against homosexual feelings.

Women wrote more stories containing violence in the sequence which
featured the Intimate Opposite-Sex Affiliation picture first and the
intense Opposite-Sex Achievement picture second (12.2%). While men
responded most strongly to same-sex pictures, level of intensity elicited
the most violence from women.

Although interrater reliability on the large set of 180 stories was
markedly lower than on the training sample of 20 stories, a third rater
was used to achieve consensus ratings between these discrepant ratings.

The third rater obtained a high level of agreement with the first two

raters on the initial set of 20 training stories. This raters‘ scoring ability
likely benefited from the author's greater knowledge of scoring
difficulties during training.

In addition to the lack of a clear relationship between sex and the
mention violence, no linkage was found between participants gender
identity and their mentions of violence. Men who wrote stories containing
violence in response to Intimate Opposite-Sex Affiliation did not have
significantly higher mean PAO Masculinity scores than other men.
Similarly, the mean PAO Femininity scores for women who had written
stories containing violence in response to scenes of intense Opposite-Sex
Achievement did not differ from those women who had not written violent
stories in this category. These results are in agreement with similar
findings by Helgeson and Sharpsteen ( 1987) who also reported a lack of a
relationship between respondents' gender identities and their scores for
violence based on written to af f iliative and achievement situations.
Violent responses may be more related to broader cross-sex personality
variables than to gender-linked traits.

in another recent study which examined Gilligan's theories of moral
reasoning, PAO scores failed to relate to the moral decision making

process. Friedman, Robinson, and Friedman (1987) asked undergraduates to

92

score moral decisions for the presence of 12 criteria. These criteria were
composed of elements containing both Kohlberg's and Gilligan's systems
moral reasoning. The authors found no link between sex and approaches to
moral judgement. Participants' PAO scores also failed to yield a
relationship between moral reasoning of gender Identity. A group of
individuals who had made moral decisions based on care was evident in the
correlation between items based on Gilligan‘s theory, but this group was
composed of roughly equal proportions of males and females. The lack of a
relationship between PAO scores and moral reasoning strategies may be
due to the PAO‘s insensitivity to the aspects of gender emphasized by
Gilligan.

If sex differences are associated with self -def initions, interpersonal
orientations, and moral reasoning, the mention of violence in response to
scenes of affiliation and achievement may not be the most accurate
method of observing these differences. in response to this present set of
stimuli's generally more realistic and balanced (for interaction intensity
level and featured character's sex) structure, both men's and women's
responses to scenes of intimate Opposite-Sex interaction were scored as
containing significantly more violence. The Moderate Same-Sex

Affiliation picture, featuring a caption similar to the intimate

\O
Qt

OpposNe-Sex picture, received the second greatest number of violent
stories from both men and women. At two levels of intensity, both men
and women appear to be perceiving intimate affiliation as more dangerous
than achievement.

This may be because the need for Intimacy is more primitive and more
pervasive than the need for achievement (Maslow, 1970). Harlow's (1959)
work with monkeys has shown that the need for nurturance overrides even
the need for food in the face of deprivation. McKinney (1986) has found
that young monkeys display neurological and physiological changes within
24 hours of separation from their mothers. Their heart rates and body
temperatures dropped during sleep and their behaviors looked similar to
depressive human responses. intimacy is important to human development
throughout Iif e and strongly effects the intellectual and personal
development of young children (Skeels, 1966). in the current study the
presence of intimate affiliative stimuli elicited more violence than other
stimuli. Perhaps the presence of intimacy evokes concern about it's
stability and inherently harbors danger and risk. If this is so, the
salience of affiliation and it's accompanying danger transcended sex and
gender differences in the present study. While men and women may

generally define themselves differently according to Gilligan's theory,

both seemed more apt to associate intimate affiliation with violence.

This general salience of affiliation must be viewed in light of the
population which participated in this study. The findings may be specific
to the age of the research participants who were largely college
sophomores (I9 year-olds). The negotiation of intimacy may be more
pressing at this age than issues of achievement that may surface more
clearly In a student's senior year or after graduation. Attendance at a
university tends to lengthen the Oedipal period and delay adulthood.

Beyond the need to replicate these findings because of the limited
degree of interrater agreement, additional issues raised by the present
results call for further exploration. A thorough examination of the levels
of Intensity with which each stimuli is perceived seems required. it
would seem desirable to distinguish between, low, moderate, and high
levels of stimulus intensityperhaps while separately considering the
perceptions of men and women. More widely varying the age of
participants also seems desirable to obtain a broader perspective on
related developmental periods.

The effect of picture presentation sequence on results strongly
suggests that variables like intensity level and featured characters‘ sex

play a major role in the amount of violence present in responses to

95

pictorial stimuli. A more rigorous examination of order effects in future
studies of this nature is recommended. While optimally three levels of
stimuli with multiple character sex configurations might be examined,
this study has shown that Intense Same-Sex Affiliation stimuli need to be
compared to Moderate Same-Sex Affiliation stimuli for response
differences and differences due to order of presentation. An exploration
of the impact of physical contact in pictorial stimuli might also be
informative. The rating of Moderate Same-Sex Affiliation as more
affiliative than Moderate Opposite-Sex Affiliation, in light of it's more
intimate "two-year" relationship caption, and the rating of the Moderate
Same-Sex Achievement scene as the most achievement oriented picture,
indicate that much more attention needs to be given to how pictorial
stimuli are perceived prior to their use.

This present work shows that intimate affiliation stimuli elicited a
greater number of violent responses from both men and women than did
either moderately affiliative stimuli or any level of achievement stimuli.
The frequency of scored violence in responses was strongly influenced by
the order in which stimuli were viewed. Although it is clear that gender
differences exist, the extent and nuances of these variations were not

reflected in the current data and appear to require further innovative

96

empirical examination. The violence aroused by intimate affiliation
stimuli for both men and women raises new questions about how both

sexes experience intimacy.

. 9 7
Appendix A

HORALITY AS CARE AND MORALITY AS JUSTICE: A SCHEME FOR
CODIRG COflSIDERATIONS OF RESPONSE AND CONSIDERATIONS OF RIGHTS

A. The Construction of the Problem
Considerations of Response (t3?.1..

1.
2.

3.

4.
5

General effects to others (unelaborated);

Maintenance or restoration of relationships; or
response to another considering interdependence;
Welfare/well-being of another or the avoidance

of conflict; or, the alleviation of another's
burden/hurtlsuffering (physical or psychological);
Considers the ”situation vs./over the principle";
considers care of self; care of self vs. care of others.

Considerations of Rights (Justice)

1.
2.
3.

4.
5.

General effects to the self (unelaborated
including “trouble“ ”how decide");

Obligations. duty or commitments;

Standards. rules or principles for self or society;
or. considers fairness, that is, how one would like
to be treated if in other's place;

Considers the "principle vs./over the situation“;
Considers that others have their own contexts;

The Resolution of the Problem/Conflict
Considerations of Response (Cars)

1.
2.

3.

4.
5

General effects to others (unelaborated);

Maintenance or restoration of relationships; or

response to another considering interdependence;
welfare/well-being of another or the avoidance of
conflict; or, the alleviation of another's
burden/hurt/suffering (physical or psychological);
Considers the ”situation vs. the principle“;

Considers care of self; care of self vs./care of others;

Considerations of Rights (Justice)

1.
2.
3.

4.
5.

General effects to the self (unelaborated
including 'trouble" 'decision”);

Obligations. duty or commitments;

Standards. rules or principles for self or society;
or. considers fairness. that is. how one would like
to be treated if in other's place;

Considers the "principle vs./over the situation“;
Considers that others have their own contexts.

The Evaluation of_the Resolution
Considerations of Response (Care)

1.
2.

what happened/how worked out;
Uhether relationships maintained/restored;

Considerations of Rights (Justice)

1.
2.

How decided/thought about/Justified;
whether values/standardslprinciples maintained.

©1931 by None Lyons

98
A.SCHEME FOR CODIHG RESPONSES TO THE “DESCRIBE YOURSELF” QUESTION

I. General and Factual
I.‘ Ceneral factual
2. Physical characteristics
3. Identifying activities
4. identifying possessions
5. Social status

‘2. Abilities and Agency
1. Ceneral ability
2. Agency
3. Physical abilities ‘
4. Intellectual abilities

3. Psychological
I: Interests (likes/dislikes)
2. Traits/dispositions
3. Beliefs. values
4. Pre-occupations

4. Relational Component
. nnected in relation to others:
1. Have relationships: (relationships are there)
2.’ Abilities in relationships: (make. sustain; to care. to
do things for others)
3. Traits/dispositions in relationships; (help others)
4. Concern: for the good of another in their terms
5. Pre-occupations: with doing good for another; with _h_o_w
to do good
3. Separate/objective in relation to others
I. Have relationships: (relationships part of obligations]
commitments; instrumental)
2. Abilities in relationships: (skill in interacting with
others)
3. Traits/dispositions in relationships; (act in reciprocity;
live up to duty/obligation; commitment; fairness)
4. Concern: for others in light of principles, values.
beliefs or general good of society)
5. Pre-occupations: with doing good for society; with whether
to do good for others)

. Summary statements
Self-evaluating Commentary
1. In self's terms
2. In self in relation to others
a. Connected self
b. Separate self

@1931 by Nona Lyons

99

Appendix B

lOO

.ehhm phogo ch

mm>flusomxm mcflxcmu swag mum muocmxma< paw ummzum

 

ucwEw>ofi£U< xwmlwuflmommo mmcwucH

lOl

.xchEou.uHm£u.cH

A
2...

5..

'~:>.v.- 7:33 at

£1;

s

q.

mummmsz uQmEpummmp mum gum: Cam m>m

 

ucwEw>0w£U< xwmuwuwmoamo wumuwpoz

Intimate Opposite-Sex Affiliation

102

 

 

 

 

Elizabeth and Nathan have been close for two years.

103

.WL;.:93

 

codumaaammd xmmuwufimoamo mucnopoz

104

HQUCMO

3mm m mcflmmzomflp

mum hone

.msuv
.mcmaowmmnm mum Ufi>mo paw nmumm

 

Dcwfiwxrmflgnud Kmmlwfimw wUMHGUOZ

105

.mumwm osu Mom mncmfiuu coon m>mn coaum 6cm mflaflnm

 

COwumewwm< xmmIOEmm mumumpoz

m

@cflmmsomflp mum xmne

 

.msnp mmucmo

.mcmfloflmxnm mum conmnm ppm

ucwEm>mߣU< XmmnwEmm mucumpo:

amaze

107

.mnmmm can now mpcmfluu sown m>mn mmmnmne paw mzaaoumo

 

 

:ofiumaaflmm4 Xmmlmfimm wumumpoz

108

Appendix C

Female Male

 

 

On the pages that follow you will find six pictures. Use the scale below to
rate each picture for the amount of affiliation (defined below) and the
amount of achievement (defined below) displayed. Please read the
definitions and scale before completing your ratings.

Affiliation: The establishment and maintenance of interpersonal
relationships.

Achievement: The attainment and interest in success and 9000
performance.

 

l 2 3 4 S 6 7
does not portrays somewhat moderately portrays portrays certainly
portray a little bit portrays portrays quite a bit a great deal portrays

l. a. Affiliation
b. Achievement

ii. a. Affiliation
b. Achievement

Ill. a. Affiliation
b. Achievement

Iv. a. Affiliation
b. Achievement

v. a. Affiliation
b. Achievement

 

VI. a. Affiliation __
b. Achievement

 

109

Appendix D

Instructions:
The items below inquire about what kind of a person you think you are.
Each item consists of a pair of characteristics. with the letters A-E in

between.

For example:

.Not at all artistic

AOOIOBOOOOCOOOODOOCOE

Very artistic

Each pair describes contradictory characteristics--that is. you

cannot be both at the same time. such as very artistic and not at all
artistic.

The letters form a scale between the two extremes.
a letter which describes where 123 fall on the scale.

think you have no artistic ability. you would choose A.
are pretty good. you might choose D.

choose C. and so forth.

1.

2.

10.

11.

Not at all aggressive

Not at all independent

Not at all emotional

Very submissive

hot at all excitable
in a major crisis

Very passive

Not at all able to

devote self completely
to others

Very rough

Not at all helpful
to others

Not at all competitive

Very home oriented

A....B....C....D....E
A....B....C....D....E
A....B....C....D....E
A....B....C....D....E
A....B....C....D....E

AOOOOBOOOOCOOOODOOOOE

AOOOOBOOOOCOCOODOOOOE

AOOOOBOOOOCOOCODOOOOE

AOOOOBOOOOCCOOODOOOOE

AOOOOBOOOOCOOOODOOOOE

AOOOOBOOOCCOOOODOOOOE

You are to choose

For example. if you

If you think you

If you are only medium. you might

Very aggressive
Very independent
Very emotional
Very dominant
Very excitable in
a major crisis

Very active

Able to devote self
completely to others

Very gentle

Very helpful to
others

Very competitive

Very wordly

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

23.

2k.

110

Not at all kind
AOOOOBOOOOCOOOODOOOOE

Indifferent to others'
approval

AOOOOBOOOOCOOOODOOOOE

.Feelings not easily

hurt
AOOOOBOOOOCOOOODOOOOE

Not at all aware of
feelings of others

AOOCOBOOOOCOCOODOOOOE

Can make decisions
easily

AIOOOBOOOOCOOOODOOOOE

Gives up very easily
AOOOOBOOOOCOOOODOOOOE

Never cries
AOOOOBOOOOCOOOODOOOOE

Not at all self-confident
AOOOOBOOOOCOOOODOOOOE

Feels very inferior
AOOOOBOOOOCOOOODOOOOE

Not at all understanding
of others

AOOOOBOOOOCOOOODOOOOE

Very cold in relations
with others

AOOOOBOOOOCOOOODOOOOE

Very little need for
security

AOOOOBOOOOCOOOODOOOOE

Goes to pieces under
pressure

AOOOOBOOOOCOOOODOOOOE

Very kind

Highly needful of
o thers ' approval

Feelings easily hurt

Very aware of
feelings of others

Has difficulty making
decisions

Never gives up easily

Cries very easily

Very self-confident

Feels very superior

Very understanding
of others

Very warm in relations
with others

Very strong need for
security

Stands up well under
pressure

111

Appendix E
Hale Pamela

DIRECTIONS - SECTION A

On the two pagee that follow, you will find a liat of worde that are uaed

to deecribe people'e personal characteriatice. For each word in the liet.
indicate how accurately the word deacribea on. The accuracy with which a
word deecribea you ie to be judged on the fol owing ecalet

l 3. 3 i 2 9. 1 2
Extremely Very Quite Slightly Slightly Quite Very threaely
inaccurate inaccurate inaccurate inaccurate accurate arcurete accurate accurate

Coneider the world IOLD. flow accurately does that word deecribe you an a
peraonl If you think that thin word ie a quite accurate deacription of you,
write the number "6" to the left of the item:

6 BOLD

 

If you think that this word in a ali htly inaccurate description of you, write
the number "4" next to it. it it e very;ihaccurete, write the number "2”. etc.

If you are uncertain of the weaning of a word, raiee your hand and we will
define it for you.

I‘ 2
DO NOT WHITE

3 '2" ‘3'

1" THIS! SPACES

l 2 - 3
Extr;hely. . V;ry . Quite .Sli
inaccurate inaccurate inaccurate anac
___(006) BASHFUL ___(029)
___(007) unausrwcsstxxc ___(030)
_(oon) uncrvu. _(031)
_(009) uwmcuurmc _(032)
_(010) aoasrrux. _(033)
___(011) unvaxu ___(03A)
___(012) KIND ___(035)
___(013) NONEGOIISTICAL ___(036)
_(014) uncucxous _(037)
‘___(015) StLr-DISCIPLINED ___(038)
_(016) roacarux. _(039)
___(017) uwotcarrlvz .___(oe0)
___(018) coucsxrtp ___(051)
'___(019) cooeanarlvr ___(062)
___(020) tweensowat. ___5043)
_(021)- swr _(044)
‘___(022) unsctrcous¢tous ___fi065)
___(023) UNSLY ___(oael
___(oza) etzasanr ___(oa7)
___(nzs) coucrxrtzss ___(048)
___(026) coneawtonantr ___(oe9)
.___(021) DISTANT ___(050)
___(028) unaurwoaxrarxvz ___(051)

Please

Ehtly

112

FOR" A

i. 2
Slightly

CUKIIC ICCUIIC.

IMPRACTICAL
smearwruc
coop-unwrap
sucu'r
ANTI-SOCIAL
powtncratuc
pansxsrcwr
cunnrnc

FIRM
unaccatss 1v:
EHOTIONAL
SHELL-HZADED
unconntar
HELL-HANNERED
UNCRAFTY
unxwpusrntous
DISRESPECTPUL
thr
otsoncanxch
GENIAL
CORDIAL
unnato
ACCOMMODATING

go on to Form B

 

.2
Quite .
accurate
(052)
(053)
(056)
(055)
(056)
_(057)
_(058)
(059)
(060)
(061)
(062)
(063)
_;_}064)
_(065)
_(oas)
____(oo7)
_(068)
_(069)
_(070)
_(071)
___(072)
___(073)
___(074)

Very
accurate

7 2
Extremely
accurate
COLD-HEARTED
canrrr
SELF-ASSURED
SBLP-CONFIDBNT
rtwtb

sraatt
RurRLtss
twousrntous
uwcnacar
CALCULATING
rework-wcaarto
unpemnsrurxva
SLY

cassarun
unsutttwc
wnmwczss
BIG-HEADED
uneaooucrlve
uwrwoaoucw
twtwusxasrxc
ILL-HANNERED
unnavaactuc
uwtht

2

113

DO NOT WRITE IN THESE SPACES

1

Extremely

inaccurat

_(ooa)
_(oo7)
___}003)
___(000)
___(010)
__(on)
____(012)
__(013)
_(ou.)
_(015)
_(016)
__(on)
___(oxa)
__(me)
_(020)
___(021)
__(022)
_(023)
_(024)
___(025l
_m2m
_« 027)
_(028)

.2.

Very

e inaccurate

DELIBERATIVE
CRUEL
APPRECIATIVE
FLAUNTY
FORCELESS
OVERFORHARD
UNDEVIOUS
UNCUNNING
VIVACIOUS
EXPLOITATIVE
UNCOOPERATIVE
PERKY .
CHARITABLE
ASSERTIVE
UNARGUHENTATIVE
UNUILD
NEIGHBOURLY
UNSYHPATHETIC
UNSOCIABLE
SELF-DOUBTINC
APPROACHABLE
STEADY
BOASTLESS

FORM 3

2 i 2.

Quite Slightly Slightly
inaccurate inaccurate accurate
CUILELESS
FORGIVING
UNSHY
IHPOLITE
COURTEOUS
UNNEIGHBOURLY
EXTROVERTED
HARD-HEARTED
CONGENIAL
UNDEHANDING
JOVIAL

_(029)
___(030)
_(031)
__(032)
___(033)
_(034)
_(035)
__(o36)
__(037)
___(038)
____(039)
_(01.0) otssocuu.
_( 01.1) unseuxtmc
_(01.2) ovcacuwwmc
__(oz.3) PERSEVERING
_(oat.) Pancuscttss
_( 045) 'pxscouaraous
_(01.6) TENDER
___(o:.7) oncamzco
_(o.'.a) rumour
__(oz.9i cocxr

(050) IRON-HEARTED

6

Quit:

ICCUECCC

___(051)
___(052)
___(053)
___(osa>
.___(055)
___(oss)
.___(057)
_____(os9)
___(060)

Very

.CCUTIEI

(058)

___}061)

(062)

___(063)

(060)

Extremely

ICCUEIC.

TRICKY
INCONSISTENT
GENTLE-HEARTED
INTROVERTED
RESPECTFUL
BOISTEROUS
INHARD
SOFT-HEARTED
LAZY
DOMINANT
OUTCOING
SELF-EFFACING
UNCHARITABLE
MEEK

l l4
Appendix F
Atkinson Group Administrationi"r

Please fill out the information blank on the top of your answer sheet
[name and student number]. This is a test of your creative imagination.
Eight pictures are included in the first part of your questionnaire packet.
The page preceding your answer sheets is a series of questions which can
be used to guide your stories. Notice that you have eight answer sheets
which correspond to the eight pictures which you have been given.

Obviously there are no right answers, so you may feel free to make up
any kind of a story you choose for each picture. The more vivid and
dramatic the better. I would suggest writing as fast but as legibly as you
can in order to make your story about the picture as vivid, detailed, and
imaginative as you can. Remember this is a test of creative imagination.
Do not merely describe the picture. Anyone can do that. Make up a creative
story about each picture.

What is happening? Who are the people? What has led up tothis
situation? What has happened in the past? What is being thought? What
is wanted? By whom?

*l‘iodif led for the current study.

P

l is
Appendix G
Experimenter: Loren Brooks, MA.

Title: Gender Differences in interpersonal Orientation

Study Explanation

Personality theorists have long recognized that the self develops
through interaction With others. in her work on moral development
Gilligan ( l982) has explored "self in relation to others". She has observed
two distinct modes of interacting Which are broadly related to gender and
development across the life span. Her research has indicated that men
generally develop an identity which is based on objective reciprocity in
relationships (objective/separate self) and women primarily form a
definition of self based on connection with others and an orientation of
care (the connected self). Strongly supporting this theory Pollak and
Gilligan (1982) found that men tended to report more aggressive behavior
in picture stimuli displaying scenes of intimacy while women tended to
report aggressive behavior in pictures depicting achievement situations.
The validity of this study has been the subject of controversy due to the
limited range of pictorial stimuli utilized. It remains unclear whether

women responded to general achievement cues or cues of competition and

ll6

whether men respond differently to varying levels of intimacy.

Attempting to further distinguish the different modes of
self-definition and interpersonal orientations Gilligan delineated, this
study will focuses on clarifying the findings of Pollak and Gilligan (i982)
by using new pictorial stimuli. The pictures you wrote stories about were
designed specifically to assess achievement and affiliation at different
levels for both men and women. These pictures are similar to the
projective test developed by Murray (1943) called the Thematic
Apperception Test. You also filled out The Personal Attributes
Questionnaire (PAO: Spence & Helmreich, i978) which is a measure of
gender identity. it is expected that women who report more violence in
scenes of achievement will have higher femininity scores and men who
have report more violence in scenes of affiliation will have higher
masculinity scores.

Lastly, you filled out a standard measure of personality. This was
included to insure that this subject sample falls into the average range of

behavior.

H7

ADDende l-l
Guidelines for Rating Stories

Violence Score

Give a score of one if: DEATH, HOMICIDE, SUICIDE, RAPE, PHYSICAL
ASSAULT, OR FATAL DISEASE are mentigneg.

Fatal Diseases include diseases that are deadly, not non-fatal illness like:
blindness, impotence or sterility.

Pregnancy does not receive a score for violence.
Abortion-The mention of abortion is given a score for violence.

Death or violence in association with an animal can be given a score for
violence.

Content Analysis
Af f iliatign as a gamma QI gangar

Violence was described as occurring in the context of situations of
affiliation

A'm r

VIOIGOCB was described as occurring in the COOtGXt Of situations Of
achievement

Bath Affiliation and Aghiavamant as a source pf dangar

Violence was described as occurring in the context of situations of both
affiliation and achievment

Nemml‘ Al I IHQHQH ”Q! AghIayamant

Violence was described as occurring in the context of neither affiliation
nor achievment

118

Appendix la

Scored Responses of Female Paricipants

Moderate Moderate Moderate Intimate W008.
Same-Sex Opposite-Sex Opposite-Sex Opposite-Sex OPPOSIN'SOX

Affiliation Affilation Achievement Affiliation ACMOVOMMI

 

'Participants who wrote more than one story scored as
containing violende.

   

119

Amnmmtb
Scored Responses of Male Paricipants

tioderate rdoderate rioderate Intwnate Intense
Same-Sex Opposite-Sex Opposite-Sex Opposite-Sex OPPOSNO'SOX
Affiliation Affilation Achievement Affiliation Achievement

  

 

*Participants who wrote more than one story scored as
containing violence.

   

t

#4...

120

ADDENdiXJ
V“! 'n r i
511mm §92£

Moderate M 3 6 i i i i 0 i 1 4

Same-Sex

Affiliation F 3 3 i 0 0 I 0 0 8

Moderate M 2 i 0 i 0 i 2 0 7

Opposite-Sex

Affiiation F I 2 0 I i I 0 0 6

Moderate M 1 i 0 0 0 4 0 0 6

Opposite-Sex

Achievement F 0 0 l 0 0 2 0 0 3

intimate M i 0 3 i 0 3 0 0 0 I 7

Opposite-Sex

Affiliation F l 0 4 i i 0 0 0 0 l 6

Intense M 0 l l 0 0 3 0 0 S

Opposite-Sex

Achievement F 0 O 0 i 0 I 0 0 2

mm M i 6 I 2 3 2 4 9 2 i 49
F i 4 9 3 3 i S 0 0 35

121

Bibliography

Atkinson, J. w. (Ed). (1958). Motives in fantasy, aotioo, ano sooiotv. New
York: Van Nostrand.

Atkinson, J. W. & Litwin, G. H. (1960). Achievement motive and test
anxiety conceived as motive to approach success and motive to avoid
failure. Journal of Aonormal Sooial Psychology, t_'>_Q, 52-63.

Benton, C. J, Hernandez, A. C. R., Schmidt, A., Schmitz, M. 0., Stone, A. J, &
Weiner, B. (1983). Is hostility linked with affiliation among males
and with achievement among females? A critique of Pollak and

Gilligan. Journal of Eocsooality and Social Esyonology, Vol. 55, No. 5,
1 167-1 171.

Bem, S. (1974). The measurement of psychological androgyny. ooomal of
Consulting aoo glioioal Psyooology. Vol. 42, No. 2, 155-162.

Bem, S. (1981). Gender schema theory: A cognitive account of sex typing.
Psychological Boviow, BB, 354-364.

Bernardez, T. (l983). Women's groups. In M. Rosenbaum (Ed.),fiaoopook_gLf
Short-loom Inocaoy orouos (pp. 1 19-338). New York: McGraw-Hill.

Bramante, M. (1970). Sex differences in fantasy pattern: A replication
and elaboration. (Doctoral Dissertation, City University of New York)

Dissertation ADSLEBQLS Iotoonational, 31, 2273B-2274B. (University
Microfilms No. 70-195460.

Chodorow, N. (1974). WW9, Berkeley: University
of California Press.

Combs, A. w. (1947). A comparison study of motivation as revealed in

thematic apperception stories and autobiography. ooumal of gljoioa]
W991 31 267-277.

1 "‘2
II
.—

Conte, H. R. & Plutchik, R. ( 1981). A circumplex mooel for interpersonal

personality traits. ooocnal of Personality and Social Psychology, 40,
701-71 1.

Erikson, E. (1950). leoooogjoojocjcly. New York: Norton.

Eccles, J. S. (1987). Gender roles and women‘s achievement-related
decisions. Psychology of womon Qoartccly, 1_l_,135-172.

French, E. G. (1955). Some characteristics of achievement motivation.

Joumal of Exoocimontal Psychology, 5_Q, 232-236.

Freud S (1961) W
W trans. and ed. James Strachey. London: The

Hograth Press.

Freud, S. (1925). Some psychical consequences of the anatomical

distinction between the sexes. Vol, Xlx. Tho staooaro ool'tioo of tho

comolotc Esychologlcal works of Sigmond Froog, trans. and ed. James
Strachey. London: The Hograth Press.

Freud, S. (1931). Witt __0.L__LV XX- W0 n

complete Psychological works of Sigmund F reog, trans. and ed. James
Strachey. London: The Hograth Press.

Friedman, w., Robinson, A., & Friedman, B. (1987). Sex differences in

moral judgment? A test of Gilligan's theory, W
Mariachi. LL. '57-‘16

Games, P A. (1977). An improved 1 table for simultaneous control on g

COMFaStS W12.
53] 34.

Gifford, R. (l982). Projected interpersonal distance and orientation
choices: Personality, sex, and social situation. W
Quactcclx. 45. 145-152.

123

Gifford, R. & O'Conner, B. (1987). The interpersonal circumplex as a

behavior map. Journal of Eorsonality and Social Psychology, S2,
1019-1026.

Gilligan, C. (1977). In a different voice: Women's conceptions of the self
and of morality. Harv r i n 1 view fl, 481-517.

Gilligan, C. (1982). In a difforont voice. Cambridge: Harvard University
Press.

Gilligan, C., & Belenky, M. F. (1980). A naturalistic study of abortion
decisions. In R. Selman & R. Yando (Eds), Clinical-ooyoloomontal

psychology, Now diroctions f oc chilo oovcloomcht, No, 7,, (pp. 69-90).
San Francisco: Jossey-Bass.

Gilligan, C., Langdale, 5., Lyons, N., & Murphy, M. (1974). The contribution
of women's thought to developmental theory: The elimination of sex
bias in moral development research and education. Submitted to the
Natlohal lhstltotc of Eoocatlon, Harvard University.

Gilligan, C., & Murphy, J. M. (1979). Development from adolescence to
adulthood: The philosopher and the "Dilemma of the F act.“ In D. Kuhn
(Ed.), I ‘ ‘ -oIO‘VI‘Hl ‘1 0‘ 0 o n I m n 'I‘ ' LQL
child ocyoloomont, No, S, (pp. 85-99). San Francisco: Jossey-Bass.

Gilligan,C. (1984). W
h Iimin in xbi inmrl ev n

W. Manuscript submitted to the National Institute of
Education.

Gross, R., Batlis, N., Small, A. & Erdwins, C. (1979). Factor structure of
the Bem Sex Role Inventory and the personal attributes questionnaire,
Journal of Consulting and Clinical Psychology, £11, I I22-I I24.

124

Harlow, H. F. & Zimmerman, R. R. (1959). Affectional responses in the
infant monkey, Sclepce, 110, 421-432.

Helgeson, V. & Sharpsteen, D. (1987). Perceptions of danger in
achievement and af f iliation situations: An extension of the Pollak

and Gilligan versus Benton et al. debate, Joorhal of Personality and
Wm 53. 727-733.

Horner, M. S. (1968). Sex differences in achievement motivation and
performance in competitive and noncompetitive situations. (Doctoral

Dissertation, University of Michigan) W
W 407B. University Microfilms it6912135.

 

Hungerford, J. K., 8Sobolew-Shubin, A. P. (1987). Sex-role identity,

gender identity, and self -schemata, Psychology of Women Ooarterly,
14, I-10.

Hurley, J. R. (1980). Two interpersonal dimensions relevant to group and
family therapy. In L. R. Wolberg & M. L. Aronson (Eds), Sroop ano
tamlly therapy, ISSQ (pp. 65-78). New York: Brunner-Mazel.

Karon, B. (1981). The Thematic Apperception Test. In A. I. Rabin (Ed),

Assessment with projective techhlooes, New York: Springer
Publishing.

Kiesler, D. L. (1983). The 1982 interpersonal circle: Ataxonomy for
complementarity in human transaction. Psychological Review, 9_Q,
185-214.

Kohlberg, L. (1958). The development of modes of thinking and choices in-
years 10 and 16. lrotoral Dissectatloh, Universlty of Chicago).

Kohlberg, L. (1981). T 'l I I n San Francisco:
Harper and Row.

LaForge, R., & Suczek, R. F. (1955). The interpersonal dimension of

1
I
I _..I

 

125

personality: 111. An interpersonal checklist. optimal of Personality.
24, 94-1 12.

Lanoue, J. & Curtis, R. (1985). Improving women's performance in
mixed-sex situations by effort attributions. Psychology of Women
Quactcclx. 9. 337-356.

Lever, J. (1976). Sex differences in the games children play. Social
pnoplemsjl: 478-487.

Lyons, N. (1981). Manual for coding responses to the question: How would
you describe yourself to yourself? Unpublished manuscript. Hamlarg
Unlyerslty, 1891b.

Lyons, N. (1983). Two perspective: on self, relationships, and morality.
WW ii. 125'145.

Maslow, A. H. (1970). Motlyatlomandpecsonahty, (2nd ed). New York:
Harper and Row.

May, R. (1974). Sexandialnasy. New York: Norton.

Maccoby, E., & Jacklin, M. (1974). The psychology of sex gltfetences.
Stanford: Stanford University Press.

McClelland, D. C., Atkinson, J. w., Clark, R. A., &Lowell, E. L. (1953) Ihe
achjeyementmotjye. New York: Appleton- Century- Crofts.

McKinney, w. T. (1986). Primate separation studies. Psychiatric
Annals, to, 281-287.

Murray, H. (1943). WW1. Cambridge:
Harvard University.

Mussen, P., & Naylor, H. (1954). Relationship between overt and fantasy
aggression. MW 52. 235-240.

126

Olsen, N. 8 Willemsen, E. (1978). Fear of success-Fact or fiction? Th

Jogmal of psychology, es, 65-70. _

02er, D. J. (1985). Correlation and the coefficient of determination,

Wm 91. 307-315.

Paludi, M. 8 F ankeIl-Hauser, J. (1986). An Idiographic approach to the

study of women's achievement striving. W
Quactcclx. .10. 89-100.

Pearson, C. (1980). A factor analytic study of the items in three selected
sex role instruments, Bsychotpgmalfiponts, 36, I I I9-1 126.

Pearson, J. L. Kamptner, N. L, Cornwell, K., Barnes, C. L, Strommen, E A.,
8Donelson, E. (1985). Michigan State University, Presented at the
‘n “no I‘O‘ 0' ‘4" iii ‘Nlllo

:QLQDLQ, Canada.

Pedhazur, E. J., 8 Tetenbaum, T. L. (1979). Bem Sex Role Inventory: A

theoretical and methodological critique. ooucnalnLEecsonanmLand
Widow. 11 996-1016.

Peplau, L. A. (I976). Impact or fear Of SUCCESS and SGX" role attitudes on

women s competitive achievement. WM
Esxchologx. 3.4. 561 -.568

Piaget, J. (1932). IhemocaLtoggementothechho. New York: The Free
Press.

Pollak, 5., 8 Gilligan, C. (I982). Images of violence In Thematic

Apperception test stories. W
Psychologyglz, No. I: 159-167.

Puglisi, J. T. (1980). Equating the social desireability of Bem's sex role
Inventory's Masculinity and Femininity subscales, JonenaLpt
WW 5% 272'276.

127

Purcell, K. (1956). The Thematic Apperception Test and antisocial
behavior. ooocnal of Consulting Psychology, 2_Q, 449-456.

Rosenkrantz, P. S., Vogel, S. R., Bee, H., Broverman, I. K., 8 Broverman, D. M.
(1968). Sex role stereotypes and self concepts in college student.
tlopnnal of Consoltlng ang Clinical Psychology, 12, 287-295.

Sidak, Z. (1967). Rectangular confidence regions for the means of

multivariate normal distribution, WW
ASSQCJEUQD. 62. 626‘33.

Shweder, R. A., 8 D'Andrade, R. G. (1979). Accurate reflection or
systematic distortion? A reply to Block, Weiss, and Thorn. ooocnal of
Pecsonality and Social Psychology, 31, 1075-1084.

Skeels, H. (1966). Adult status of children with contrasting early life

experiences WWW
Dexelooment. 31 (N0 3)

Spence, J., 8 Helmreich, R. (I978). mascnltmtyangfimtnlnlty. Austin:
University of Texas Press.

Terman, L., 8Tyler, L. (1954). Psychological sex differences. In

Carmichael, L. (Ed), W New
York: Wiley.

Tresemer, D. (1976). The cumulative record of research on "Fear of
success". SexPoles, 2, 217-236.

Wiggins, J. S. (1979). Apsychological taxonomy of trait- descriptive

terms: the interpersonal domain, Wat
mm 31. 395- 412

Wiggins, J. S. (1982). Circumplex Models of interpersonal behavior in

clinical psychology. In P. C. Dendal 8 J. N. W
W (pp. 183-222). New York:

I28
Wiley.

Winterbottom, M R. (1958). The relation of need for achievement to
learning experiences in independence and mastery. In Atkinson, J. W.

(EdM) WW3 4781 Princeton

Van Nostrand.

"IIIIIIIIIIIIIIIIIIIIIIIIII“