• TABLE OF CONTENTS
HIDE
 Title Page
 Acknowledgement
 Table of Contents
 Abstract
 Theoretical and empirical...
 Hypothesis
 Method
 Results
 Discussion and conclusions
 Appendices
 References
 Biographical sketch






Title: Sex role development and identity achievement
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00098114/00001
 Material Information
Title: Sex role development and identity achievement
Physical Description: vi, 144 leaves : ill. ; 28cm.
Language: English
Creator: Deldin, Lauren Sue, 1946-
Publication Date: 1976
Copyright Date: 1976
 Subjects
Subject: Sex role   ( lcsh )
Identity (Psychology)   ( lcsh )
Androgyny (Psychology)   ( lcsh )
Psychology thesis Ph. D   ( lcsh )
Dissertations, Academic -- Psychology -- UF   ( lcsh )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Thesis: Thesis--University of Florida.
Bibliography: Bibliography: leaves 139-143.
General Note: Typescript.
General Note: Vita.
Statement of Responsibility: by Lauren Sue Deldin.
 Record Information
Bibliographic ID: UF00098114
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000169541
oclc - 02907415
notis - AAT5949

Downloads

This item has the following downloads:

PDF ( 3 MBs ) ( PDF )


Table of Contents
    Title Page
        Page i
    Acknowledgement
        Page ii
    Table of Contents
        Page iii
    Abstract
        Page iv
        Page v
        Page vi
    Theoretical and empirical background
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
    Hypothesis
        Page 30
        Page 31
        Page 32
        Page 33
    Method
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
    Results
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
    Discussion and conclusions
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
    Appendices
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
    References
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
    Biographical sketch
        Page 144
        Page 145
        Page 146
        Page 147
Full Text











SEX ROLE DEVELOPMENT AND
IDENTITY ACHIEVEMENT














By

LAUREN SUE DELDIN
















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






UNIVERSITY OF FLORIDA


1976













ACKNOWLEDGEMENTS


I want to thank my Chairman, Dr. Audrey Schumacher,

for her enthusiasm and imagination which were so important

in the beginning and final stages of this project and for

her guidance and support throughout. For their helpful

comments and support, I would also like to thank the other

committee members: Dr. Ben Barger, Dr. Susan Hoffman,

Dr. Milan Kolarik, and Dr. Marilyn Zweig.

Special thanks are directed to Dr. Tom Kelley for being

so generous and cheerful in his assistance with statistical

design and interpretation.

I would also like to thank Terry Turner, Leif Grazi,

and Joe Seitz for help with the cumbersome task of organizing

materials and data collection.

To Roger Keroack, Pat Korb, and Suzanne Nickeson, I

give unending appreciation for their fortitude in scoring

and the love and laughter that they gave to this project and

to me.

And finally, I thank Dianne Downing for the arduous

task of typing and Breck Parker for his constant support.














TABLE OF CONTENTS


ACKNOWLEDGEMENTS

ABSTRACT

CHAPTERS:

I THEORETICAL AND EMPIRICAL BACKGROUND

II HYPOTHESES

III METHOD

IV RESULTS

V DISCUSSION AND CONCLUSIONS

APPENDICES:

I Ego Identity Incomplete Sentences Blank

II Ego Identity Incomplete Sentences Blank
Scoring Manual

III Adjective Check List for Same-Sex Best
Friend and Information Sheet for
Hollingshead Scale

IV Adjective Check List for Same-Sex Parent

V Parent-Child Interaction Rating Scales

VI Bem Sex Role Inventory

REFERENCES

BIOGRAPHICAL SKETCH


Page

ii

iv



1

30

34

54

73



93


96



124

125

126

135

139

144










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



SEX ROLE DEVELOPMENT AND
IDENTITY ACHIEVEMENT

By

Lauren Sue Deldin

June, 1976

Chairman: Dr. Audrey Schumacher
Major Department: Psychology

Utilizing the framework of Erikson's theory to con-

ceptualize the relation between sex role development and

overall identity formation, and the related.empirical evi-

dence on parental sex role identification and peer group

influence, we chose to study the contribution to identity

achievement of three factors.

The three factors were the level of nurturance of the

late adolescent's relation with his/her same-sex parent

(measured by the Parent-Child Interaction Rating Scales

developed by Hilbrun), the level of the father's masculine

sex role stereotyping/the mother's feminine sex role stereo-

typing (measured by the nine sex-typed scales of the Adjective

Check List), and the level of matching between the same-sex

parent's and peer's sex role stereotyping (measured by

comparison of the subject's separate ratings of parent and

peer on the Adjective Check List). We investigated the re-

lated issues of whether exposure to both masculine and








feminine sex role stereotypes in parent and peer models

is integrated into an androgynous sex role self-concept

(measured by Bem's Sex Role Inventory) and whether more

androgynous self-concept would be related to greater

identity consolidation. Additionally, based on the premise

that the traits of the traditional female role are antago-

nistic to identity development and previous findings of a

male advantage in identity achievement, we predicted that

our male subjects would score significantly higher than the

females on our measure of identity the Ego Identity

Incomplete Sentences Blank developed by Marcia and modified

to include sexuality items.

The five instruments were administered in two sessions

to the 56 male and 56 female subjects pooled from the in-

troductory behavioral studies classes at a Community College.

There were no significant effects on identity achieve-

ment for any of the three factors or between androgyny and

identity achievement. However, several important trends

emerged. The most important influence on male identity achieve-

ment was found to be the level of perceived sex role typing

in the adolescent's father. Male students with a highly

masculine stereotyped father scored higher on identity than

those with a less masculine father. From the patterns that

emerged in the female androgyny scores we supported the

general hypothesis that the reaction of the daughter to the

mother-daughter relationship and of the sex role preference

of the mother has a determining effect on the direction of








the daughter's sex role self-concept whether she develops

toward feminine stereotyping or androgyny. The influence

of the peer's sex role norms is merely an adjunct in this

process. As we discussed, perhaps the most striking finding

was the negligible impact for both males and females of the

peer's sex role norms. From the above trends it was con-

cluded that the early influence of the same-sex parent far

outweighs the effect of the peer sex role model the youth

is exposed to in late adolescence.

A rather unexpected and noteworthy finding was that

the females scored substantially higher on identity achieve-

ment than the males. The following explanations for this

finding were proposed and evaluated: a higher than normative

incidence of androgyny in our female subjects, conflicts

between the males self-perceptions and actualizations of

masculine sex-typed traits, and sanction by the Women's

Movement for the development of traits in females that

facilitate identity achievement.













CHAPTER I
THEORETICAL AND EMPIRICAL BACKGROUND



Erik Erikson has conceptualized the focal theme of

adolescence as a crisis in identity, precipitated by the

discontinuity of development occurring at that time. He

attributes the disruption of adolescence not only to physical

changes and increased sexual demands but also to the strain

of assuming new social roles as well as the difficulties

engendered in relating these new elements to identity frag-

ments developed in earlier stages. The outcome of this crisis,

the success or failure of re-integration, depends on the pos-

sibility of integrating significant identifications as well

as libidinal needs, favored skills, and constitutional givens

with the available social roles and involves choices and

h decisions which result in commitments for life (Erikson, 1968).

To assemble all these converging elements at the end

of childhood is a formidable task, and it follows that

adolescence is a time of "normative crisis, i.e., a normal

phase of increased conflict" (Erikson, 1968, p. 163). As

such, it is not defined by the self-perpetuating propensities

of neurotic and psychotic crises, but rather is more

traversable, being characterized by an abundance of

energy which serves both to support the ego in its search

for and playful engagement with new opportunities








as well as to revive dormant anxiety and arouse new conflict

and confusion (Erikson, 1968).

To accomplish this integration of identity elements,

Erikson believes adolescents need above all a psychosocial

moratorium, an intermediate period between childhood and

adulthood characterized by a delay of adult commitments.

It is up to the society to offer a moratorium during which

the individual ego, through role experimentation and fantasy,

may examine and attempt to integrate old and new ways of being.

The end result of this process is the selective repudiation

and mutual assimulation of all significant childhood ident-

ifications into a coherentgestalt uniquely suited to the

individual .and yet in concordance with the adult roles offered

by his society at that time (Erikson, 1968). For it is of

great importance to the adolescent's identity formation that

his more finalized self-definition be verified in experiences

of psychological "fittedness." However, this experience of

"fittedness" involves more than the adolescent filling an

appropriate slot. This is a subtle process which Erikson

calls recognition a type of response to the growing person,

at every new step in his development, that informs him that

his individual way of being is a successful variant of his

group's identity (Erikson, 1963).

In short, whether the identity crisis is successful or

unsuccessful depends on 1) the developmental history of the

individual as reflected in variations in ego abilities and

in the quality of childhood identifications as well as








2) the present situation of the adolescent, involving the

extent of recognition available in the role opportunities

of his peer group and the quality of the moratorium provided

by the larger society.

The unsuccessful resolution of the crisis has several

variations, including too early "commitment" imposed on the

youth by community recognition at a critical moment when he

has aroused displeasure, the youth's own foreclosure of his

identity development by concentration on one area to the ex-

clusion of other vital aspects, and, finally, acute identity

diffusion an extreme version of the normative negative pole

of the identity crisis. This pole, labeled identity confusion

or role confusion, consists of the conflicts, confusion, and

anxiety aroused in this normative crisis which must also enter

the process of identity formation and hopefully be integrated.

The positive resolution of this crisis is marked by the

achievement of a sense of identity defined by Erikson as "the

accrued confidence that the inner sameness and continuity pre-

pared in the past are matched by the sameness and continuity

of one's meaning for others" (Erikson, 1963, p. 261). Many

writers on adolescence agree with Erikson that the development

of sex roles plays an integral part in the adolescent's overall

identity formation (Blos, 1961; Dignan, 1965; Heilbrun, 1964a; .

Kagan, 1964; Sanford, 1967). Sex roles are a mandatory and

all encompassing structure that the growing individual uti-

lizes in interpreting his experience. These typologies

grant the youth an "inner sameness and continuity" that is









matched by his "sameness and continuity for others" over an

intensive and extensive range of his experiences of himself

and others.

The fact that many of our youth are now rejecting much

of the content of this societal recognition of their exper-

ience does not mean that they as a generation are less clear

about their own bodily experience but rather that they object

to the particular type of recognition our society offers -

that is, the traditional sex roles (Erikson, 1968). While

the relationship between the parental sex role preference

and either the adolescent's own sex role or his identity

achievement has been the subject of several studies, no one

has yet assessed the effect on identity formation of the

alternative sex role norms of the late adolescent's peer

culture. The purpose of this study is to investigate the

contribution to identity achievement of both those factors

already deemed to be crucial in parental sex role identifi-

cation as well as the interaction effect of those factors

and peer sex role prescriptions.

Identity theory postulates that the ease of re-synthesis

of identification into a more final self-definition is in-

fluenced by both past experience with the models of the

child's family and the options that are available to the

adolescent in the larger society. In order to predict the

relation between sex role development and identity formation,

we may translate this basic postulate into the specifics of

the process of sex role identification. The resulting.








prediction is that the "quality and coherence of the sex

role identifications the adolescent must attempt to integrate

will vary according to the degree of satisfactory interaction

he has experienced with a trustworthy and meaningful hier-

archy of sex roles in his family" (Erikson, 1956). The

literature on parental sex role identification provides

evidence and theory that support and clarify the specific

hypotheses that we will derive from this global prediction.

The first factor we will consider in this prediction

is the degree of satisfactory interaction the adolescent has

experienced with the primary models of sex roles in his

family his parents. A nurturant relation is one type of

satisfactory interaction with parents that has been shown

to be significantly related to the coherence of the adoles-

cent's sex role identifications. Several studies demonstrate

that the degree to which a nurturant, rewarding relationship

exists between a youth and his same-sex parent is related to

the strength of the youth's identification with this parent

and the strength of the appropriate sex role preference made

by the youth (Heilbrun, 1964b; Mussen, 1961; Mussen & Distler,

1959; Mussen & Rutherford, 1963; Payne & Mussen, 1956). The

occurrence of a nurturant relation with the opposite-sex

parent has not been shown to be related to the youth's sex

role identification (Mussen & Distler, 1959; Mussen &

Rutherford, 1963).

One study has related this one aspect of the process of

sex role development the relation between parental nurturance








and the coherence of sex role identifications to the

outcome of integration of sex role identifications overall

identity achievement (Heilbrun, 1964b). High father nurtur-

ance was found to be significantly related to high role

consistency a measure of overall identity for male

adolescents. Heilbrun helped clarify the role of same-sex

parental nurturance in the relation between sex role develop-

ment and identity achievement by demonstrating that the

interaction between nurturance and parental sex role prefer-

ence had a significant effect on identity achievement. There

was no simple relation between parental sex role preference

and identity, for it was necessary to take into account the

nurturance factor to determine if the parent's sex role was

imitated and, thus, could effect the adolescent's identity

integration. When the father presented a low masculine model

for the male adolescent, greater nurturance was associated

with lower identity achievement while greater nurturance in

a moderate or high masculine model was related to higher

identity achievement.

A second factor represented in the predicted relation

between sex role development and identity achievement is.the

"meaningful hierarchy" of sex roles in the adolescent's

family. The degree of masculinity of the father and femininity

of the mother is one aspect of the social meaningfulness of

familial sex roles that several investigators believe should

influence the quality and coherence of the child's sex role

identifications (Heilbrun, 1964b; Kagan, 1964; Mussen &








Rutherford, 1963). However, the evidence is not clear-cut.

The specific assumptions that a highly feminine mother would

foster her daughter's femininity, and a highly masculine

father would foster his son's masculinity were not supported

in studies by Mussen & Rutherford (1963) and Payne & Mussen

(1956).

Considering the male youth first, both studies concluded

that the role of the parents in masculinizing their son has

nothing to do with the masculinity of the father or the

femininity of the mother but is simply that the father provide

Initial motivation, through provision of a nurturant and

rewarding relationship, and that the son imitate his behavior.

Then by generalization the son will continue to emulate other

men, including more masculine ones. Their explanation assumes

the provision of much greater assistance from the social

environment in the development of the boy's sex role ident-

ification than the girl's. This assumption of differential

assistance is widely supported by observation and experimental

evidence (Bardwick, 1971; Brown, 1958; Heilbrun, 1964b; Lynn,

1966; McKee & Sherriffs, 1957; Rudy, 1968; West as cited in

Mussen & Rutherford, 1963). In short, the general culture

of the parental generation greatly assists the family in the

masculinization of the boy by providing well articulated cues

for the masculine role and by consistently rewarding the boy

for learning them.

However, subsequent to Payne & Mussen's and Mussen &

Rutherford's studies, Heilbrun demonstrated that the degree








of masculinity of the father's sex role preference does

significantly effect the process of sex role identification

and overall identity achievement (1964b). As we noted in

the above discussion of the factor of nurturance, Heilbrun

found that moderate or high masculinity of the father model

(when high nurturance was also present) was significantly

related to high identity achievement, while low masculinity

of the father model (with high nurturance present) was

associated with low identity achievement for male adolescents.

S In his discussion of these results Heilbrun implicates

a third factor in the relation between sex role development

and identity formation the similarity of familial roles and

the sex roles of the adolescent's society. This factor is

emphasized in our general thesis and elsewhere in Erikson's

writings on identity. One way in which familial roles form

a "trustworthy hierarchy" is the degree to which they match

the sex role expectations of the adolescent's present society.

Heilbrun explained his results in terms of the matching or

lack of matching between parental sex roles and socially

reinforced sex role behavior. Specifically, if a boy has a

-high or moderate masculine father and greater nurturance is

present, the highly imitated sex role behavior will also tend

to be reinforced by his social environment, resulting in

greater overall role consistency Heilbrun's measure of

identity achievement.

In order to trace the experimental evidence on the

effects of factors two and three on the female adolescent's








sex role development and identity achievement we will return

to Mussen & Rutherford's findings. While the mother's

femininity was not related to the daughter's sex role pre-

ference, the father's possession of a high degree of masculin-

ity of interests and attitudes and his active encouragement

of the girl's participation in appropriate sex-typed activities
Tended to foster feminization of the daughter (1963). Mussen

& Rutherford concluded that the parents must play a more

forceful and direct role in the development of the girl's

sex role than they do for the boy.

Their explanation was that in comparison to the boy, the

girl receives little assistance from the social environment

in the development of her sex role identification (Mussen &

Rutherford, 1963). Heilbrun's explanation of his findings on

females (1964b) supports this thesis of differential assistance

offered by the social environment. Heilbrun found that the

interaction between nurturance and model sex role preference

does not have a significant effect on role consistency in

female adolescents. He concluded that this relation does

not appear for the female because she is exposed, not clear-

cut and consistent sex role standards as the male is, but to

conflicting social rewards which fail to reinforce her identi-

fication with her mother, whether or not the mother presents

a highly feminine role preference.

Thus, both Mussen & Rutherford's and Heilbrun's con-

clusions attribute the consolidation of the female's sex

role identifications to the influence of her family, with






10
little or no consistent recognition offered by the social

environment. However, though there is less definitive

evidence than for the male)there are indications that our

third factor the similiarity of familial sex roles and

the sex roles of the youth's present social environment -

may have an important influence on the girl's level of

identity achievement. In one of these studies Heilbrun

tapped the outcome of parental sex role identification -

the establishment of a sex role identity and its relation

to identity achievement. Sex role identity was assessed by

conformity to masculine or feminine stereotypes, and identity

strength was indicated by degree of role consistency. For

females, Heilbrun found that either high conformity to the

feminine stereotype or high conformity to the masculine

stereotype resulted in high identity achievement whereas

an intermediate sexual identity was related to lower role

consistency (1964a). As Heilbrun's first study (1964b)

showed no relation between maternal sex role preference,

high nurturance, and identity achievement in female adolescents,

it doesn't seem likely that identification with either a very

feminine or a more masculine mother by itself is responsible

for the establishment of these two types of sex role identities

and the related successful identity achievement. Nor would

identification with a highly masculine father account for

the process of sex role consolidation and identity achieve-

ment in the feminine stereotype girls whose modal parental

identification Heilbrun has shown to be with a highly feminine








mother. We might look for some other factor in the process

of sex role identification that supported the consolidation

of the girl's developing parental sex role identification

into one of these stereotypes, which in turn produced high

role consistency.

With this in mind let us consider Dignan's study of a

group of college women in a very different social environ-

ment than the one in Heilbrun's studies. She found a signif-

icant positive relation between high maternal identification

and identity achievement (1965). We would not expect such

a simple and significant relation in light of Mussen &

Rutherford's and Heilbrun's (1964b) findings. One possible

explanation lies in the unique nature of the social environ-

ment that Dignan's women were in a small Catholic women's

college in the Midwest. This milieu may present consistent

and clear-cut standards and models for the woman's sex role

and, thus, support the development of her sex role identity

contingent only upon her having made an initial identifi-

cation with her mother that motivated her to generalize to

other female role models. In line with this interpretation

we would expect that both stereotypical groups in Heilbrun's

study had found sex role standards and models in their sur-

roundings that helped them successfully consolidate their

sex role identifications with their parents.

Such a fortunate situation does not seem to exist in

the environment of many American female adolescents. Sanford

reports that a study of females at Vassar, over the period








1952-1958, revealed that seniors showed more disturbance

with respect to identity than freshmen. Comparing these

findings to current observations of males at Stanford and

Berkeley, Sanford found that the senior year for males was

not distinguished by an identity crisis. Instead it appeared

that the most critical time for identity was earlier, in

connection with choice of a major (Sanford, 1966).

Constantinople's study in 1969 indicated that while women

seem to be further advanced in identity development when

they enter college, the men show greater gains in maturity

during this period. The lesser degree of identity consoli-

dation in the female subjects may be due to the lack of

recognition of important aspects of a woman's feminine

identity potentials.

As has often been noted, the American culture is male

dominant in orientation and recognizes almost exclusively,

activities, interests, experiences, etc., in line with the

masculine role (Brown, 1969; Heilbrun, 1964b; Lynn, 1966;

Mussen & Rutherford, 1963). In our society, the domain of

women is not clearly designated, the few predominately

feminine activities are not prestigeful, and this condition

is reflected by our educational system, in which preparation

for womanhood is not undertaken with seriousness (Brown, 1969).

As Constantinople (1969) concludes, the academic environment

does not seriously offer preparation and experimentation

with choices and activities involved in being a wife and

mother recognition of a more feminine parental sex role


I








identification but instead offers only masculine oppor-

tunities to the female student. Thus, the prolonged identity

diffusion for many females in this college environment and

in many other environments of our society may be attributed

to the lack of matching between present social sex role op-

portunities and previous identification with the feminine

sex role in the family.

The lesser degree of identity achievement in the female

adolescent may also be due to the traditional female sex

role itself. Evidence supporting this thesis involves two

aspects of the highly feminine sex role: 1) conflict over

accepting the prescribed role due to the limits and negative

valence of the role and 2) acceptance of the role reinforces

personality traits that are antagonistic to the development

of ego strength and identity achievement.

Several studies indicate that the majority of girls,

whatever the quality of the rest of their identity struggles,

experience appreciable diffusion due to conflicts over

assuming the traditional female sex role. Howard found that

whether or not a girl had expressed anxiety over the other

aspects of identity diffusion, she would show concern in the

area of sexual identity (1960). Other studies indicative

of such a pervasive conflict show that 20-30% of adult fe-

males, as compared to only 2-4% of adult males,have been

aware of their desire to be a member of the opposite sex

(Brown, 1958) and, that, in early adolescence females view

feminine traits and interests as less desirable than males

regard masculine traits and interests (Rudy, 1968).









Two inferences can be made from these studies. The

first is that the conflict arises because the available

female sex role does not affirm a wide enough range of human

experiences. Being prescribed by the role to define them-

selves very narrowly as sex objects or potential mothers

interferes with full identity attainment because of the role

obligations and self-conceptions engendered as well as the

conflict and rejection produced by such confining limits

(Sanford, 1967). Actually restriction and suppression of

individual potentials is also operative through the pre-

scribed masculine sex role. Being only "really masculine"

or only "really feminine" is less than being a fully human

man or woman.

The other inference is that the traits ascribed to the

traditional female sex role are rejected or present conflict

because they are less socially desirable. This inference is

supported by the findings of Broverman et al. across several

different samples of college men and women. "Stereotypically

masculine traits are more often perceived to be desirable

than are stereotypically feminine characteristics" (1972).

When Broverman et al. administered their questionnaire under

self-concept instructions the effect of a negatively valenced

sex role upon the identity process of young women was brought

into focus. Self-concepts of college men and women were

found to be significantly different on the sex role stereo-

typed items. Since many more of the feminine traits are

negatively valued than are the masculine traits, these women








were reporting more negative self-concepts than the men.

Thus, the lesser degree of identity consolidation in the

female adolescent may in part be due to the bind of having

to choose between aligning one's self with more socially

desirable behaviors or identifying with the sex role be-

haviors designated to her by her society. From the results

of Broverman et al. the tendency of female choice is to

denigrate herself, an effect that is evidence of the power-

ful social pressures on the adolescent to conform to the

sex role standards of the society.

Moreover, many of the personality traits of the des-

ignated role appear to be antagonistic to the development

of ego strength, identity achievement, and healthy adult

behavior. As described earlier the process of identity

achievement involves on the part of the adolescent his own

active and individuating search a questioning of roles,

ideals, and past identifications, role experimentation, and

rethinking as well as the modeling of the established guide-

lines of parents and society. It would appear that the

greater degree of passive, dependent, and submissive behavior

expected and tolerated in the female, in accordance with her

sex role, would obstruct the achievement of an independent

identity. Supporting this view Stone and Church observe

that through cultural instruction, girls become motivated

by the notion that attachment to a man is the way to identity,

but the point of this intimate connection is to belong to

the man and to utilize him as a more potent extension of the








self rather than to establish a separate identity (1968).

Sanford also questions the effects of intimacy before

identity on the girl's identity development (1966).

While a strong affiliation need not necessarily carry

with it a tendency toward excessive dependency and lack of

independent achievement, researchers in the psychology of

women have stressed the mutual exclusivity of the traits

of affiliation and independent achievement in the female

sex role stereotype (Lunneborg & Rosenwood, 1972). Horner's

well known thesis is that individual achievement and femi-

ninity are viewed as two mutually exclusive goals. If a

woman achieves, she expects social rejection. Bardwick

(1971) sees more room in our society for women to be psycho-

logically free to pursue work achievement but not until after

they have become completely secure in their affiliative

relationships. Women still largely perceive the world in

interpersonal terms and value themselves only insofar as

they are loved by others. There is some recent research

that indicates that the stereotypical one-sided development

and recognition of affiliation needs and the disregard or

rejection of achievement needs in women has a negative effect

on identity achievement and ego strength.

Erikson postulates that the resolution of the school

child's ego crisis of industry/inferiority influences his

later crisis of identity to a great degree. The positive

resolution of this crisis is theorized to support the devel-

oping sense of identity through the provision of a sense of








competency and a willingness to apply oneself to concrete

pursuits and goals that are preparatory to productive roles

in the larger society (1968). Bauer & Snyder (1972) have

labeled the outcome of the school age crisis as achievement

motivation and tested the predicted relation between identity

strength and achievement motivation. They did find that

individuals who demonstrate high achievement motivation

scored significantly higher on an ego identity measure than

subjects with low achievement motivation. Thus, the young

woman identified with the traditional sex role prescription

of renounced or delayed achievement motivation would be

expected to score lower on identity achievement.

SIn researching the relation between sex role attitudes

and psychological well-being, Gump (1972) utilized the

dimension of self- or other-orientation to differentiate

women's sex role attitudes. The other-oriented woman had

a traditional outlook. These women found personal fulfill-

ment through fostering the fulfillment of others, primarily

a husband and children. The self-oriented woman was one

who embraced the achievement orientation of our culture,

seeking fulfillment through the maximization of her own

potential. Gump found that women with high ego strength

scores characteristically had well defined educational and

vocational plans while low scorers were dependent on a man

for future plans; high scorers certainly made plans which

took into account their interest in a man, but they tended

to provide themselves with many options rather than to








pursue a single and solely affiliative alternative; and,

finally, women high on ego strength were more often self-

oriented than other-oriented. Based on these findings,

Gump questions whether it is possible to possess high ego

strength in the context of the traditional female role.

Her findings suggest that,"purposiveness, resourcefulness,

and self-direction may be inconsistent with adoption of a

role limited to the traditional, other-oriented goals and

satisfactions" (Gump, 1972).

Nonetheless, other observers of adolescents believe it

is more adaptive for the girl's overall personality develop-

ment not to actively and independently define herself during

the adolescent years, that the high dependency and high

affiliative/low independence and low achievement prescrip-

tions of the traditional role are more adaptive for women.

Douvan found that ego integration in girls was linked to

interpersonal skill and sensitivity (1960). To a greater

extent than in the male their identity remains diffuse and

incomplete because it ultimately depends not so much on what

they are but on who their husbands are (Douvan, 1960; Douvan

& Adelson, 1966). Carlson (1965) found that over a six-year

period of adolescent development the female increased in an

orientation which emphasizes interpersonal experience in her

concept of herself, while the male increased in the orienta-

tion which produces conceptions of self which are independent

of concern with social experiences. Although a socially

oriented self-concept implies an over-reliance upon social








feedback and potential instability of self-esteem, self-

esteem was not lower in the socially oriented adolescent

girl. Carlson explains this paradox by pointing out that

social orientation is considered to be the appropriate style

for women according to the traditional sex role, and, thus,

does not lead to devalued self-esteem. Indeed in their

application of the identity status measures to 49 college

senior women, Marcia & Friedman (1970) found that lowest

self-esteem and the next to highest anxiety level existed

among the identity achievement group the group that actively

and independently questions, experiments, and then formulates

its own firm commitments. Highest self-esteem and lowest

anxiety levels were found among foreclosure women. The

latter have not experienced a period of "crisis" but have

followed parentally determined commitments. Their inter-

pretation, in line with Douvan & Adelson's and Carlson's

conclusions, is that possibly the foreclosure identity is

a particularly adaptive one for women. As compared to men,

for whom there is a social demand that they go through an

identity crisis, women are expected to remain somewhat

dependent on their families or future husband for direction.

Thus, high social support, due to their fulfillment of sex

role expectations leads to less anxiety and higher self-

esteem in the foreclosure women, while the identity achieve-

ment group's "male-like" process of achieving an identity

may alienate them from their peers and may be quite threat-

ening to the girl herself because it involves breaking with


_ I








her family, resulting in lower self-esteem and higher

anxiety (1970).

However, Marcia & Friedman failed to administer their

overall measure of identity strength to the women, and they

aren't on safe ground empirically generalizing from the

statuses differential performance on the self-esteem measure

as the self-esteem and identity measures appear to be tapping

two related but distinct aspects of personality (Marcia,

1964) or from the anxiety measure as foreclosure men, shown

to have lower overall identity scores than the identity

achievement group, also scored lowest on the anxiety measure

(Marcia, 1967). Nonetheless, they go on to sum up their

findings on the identity process in women by saying that

achieving an identity seems more difficult for women than

for men so women may actually be rewarded in terms of psycho-

logical health for not doing it for remaining foreclosures.

Thus, the implication of Marcia & Friedman as well as

Carlson, Douvan, and Adelson's conclusions is that a healthy

young woman's development is less healthy than a healthy

young man's. Indeed, the findings of Broverman et al. (1972)

demonstrate that clinicians utilize a double standard of

health for the product of the adolescent struggle adult

men and adult women. They found that "the general standard

of health (adult, sex-unspecified) is actually applied to

men only, while healthy women are perceived as significantly

less healthy by adult standards."








Thus, because of our sex role biased conceptions of what

is healthy in the individual, there is less socially sanc-

tioned opportunity for an adolescent who is a woman to

participate in an identity quest or to enjoy the results -

healthy adulthood. If women adopt the behaviors specified

as desirable for adolescents and then for adults, they risk

censure for their failure to be appropriately feminine, but

if they adopt the behaviors that are designated as feminine,

-.they are necessarily deficient with respect to the general

standard of adolescent and then adult behavior.

There is hope, however. Several recent studies on

adolescents report a seeming breakdown of sex role stereo-

types a move towards an integration of healthful masculine

and feminine traits in each individual. Lunnenborg &

Rosenwood (1972) sought to replicate Bardwick's earlier

study of the relative strength of affiliative and achieve-

ment needs in college women using both men and women this

time. While Bardwick had found all affiliative responses

and not one achievement response in her female subjects, of

the four tests of differences between the sexes on the two

needs in the present study only one reached significance in

line with the traditional stereotypes. They conclude that

"it would be more accurate for psychologists to describe

college men and women as currently both possessing these

needs, with men becoming more concerned with loving and

close interpersonal relationships (than in the past) and

women more concerned with pride in school and work achieve-

ment."








SUsing a model of sex role identification/personal

adjustment based on the traditional stereotypes, Williams'

(1972) hypothesis that adolescent girls identified with

retiring-passive mothers would have the highest level of

adjustment was not supported. Not only did-she find that

girls identified.with an ascendant-dominant father were the

healthiestbut the distribution of identifications with

parental types was markedly different from that obtained in

an earlier study in 1967. In that study the most frequent

category for girls was retiring-passive mother while the most

frequent group in the present study was ascendant-dominant

father, and 2/3 of the girls ascribed to themselves mostly

ascendant-dominant behavioral traits.

Finally, a later study of the identity statuses in women

revealed very different results than those reported by Marcia

& Friedman. In Schenkel & Marcia's (1972) study, the identity

achievement group scored higher on self-esteem than all other

statuses while in the earlier study that status scored the

lowest. In addition, the identity achievement group did

not score unusually high on the anxiety measure as they had

in the Marcia & Friedman study. In explaining the difference

the authors suggest that the current study, executed three

years later, may have been dealing with a new generation of

college women influenced by the current women's movement.

Subjects may have been experiencing a sanction for making

independent decisions and felt more support from their peers

while going through the process.








Thus, in a time of changing sex role prescriptions,

to fully understand and predict the effect of parental sex

role preferences on the male as well as the female adoles-

cent's sex role and overall identity development, it may be

necessary to also look at the parental sex roles in terms

of the sex role prescriptions of the late adolescent's

present environment. More explicitly, we can now state

factor three as follows: the degree to which parental sex

role preferences match the sex role expectations of the

adolescent's present society significantly influence the

ease of identity consolidation by effecting the quality and

coherence of the sex role identifications which the adoles-

cent must integrate. Our third factor is central to Erikson's

theory of identity formation, for identity integration at

adolescence involves the selective repudiation and assim-

iliation of all significant childhood identifications into

a coherent gestalt uniquely suited to the individual adoles-

cent and yet in concordance with the roles offered by his

society at that time (Erikson, 1968).

The source of sex role norms in the adolescent's present

society is not that obvious. The established view is that

the peer group is the most available temporary anchor to the

adolescent in the midst of his confusion and doubts and,

thus, is the prime source of roles and opinions to test out

and confirm his many and fluctuating selves (Douvan &

Adelson, 1966; Erikson, 1963; Horrocks, 1969; Marin & Cohen,

1971; Potvin, 1964; Rogers, 1969; Sherif & Sherif, 1965;








Stone & Church, 1968). However, just what the lasting gains

are for identity formation is an issue that Erikson has not

definitively elaborated and on which the relevant literature

is vague or conflicting. On the one hand, experimental

study and observation have led to the conclusion that the

peer group does not help to differentiate the adolescent's

budding identity but is merely a temporary support against

identity diffusion and often hinders growth by encouraging

a denial of the self (Douvan & Adelson, 1966; Potvin, 1964).

On the other hand, there is considerable experimental sup-

port for a more basic and pervasive influence of the peer

group on identity formation. To begin with, sociological

and social-psychological studies have demonstrated that the

peer group does develop its own norms and values which

direct members' behaviors (Epperson, 1969; Sherif & Sherif,

1964; Smith, 1969). In addition, other social-psychological

studies demonstrate that these peer norms, identifications,

and appraisals not only regulate the immediate behavior of

the adolescent, but are utilized by him as external references

against which he fashions his inner sameness and continuity,

in other words, they become part of his emerging identity

(Douvan & Adelson, 1966; Kipnis, 1961; Manis, 1955; Sherif

& Sherif, 1965).

If on the basis of the above evidence we conclude that

the peer group is not only the most available source of roles

and opinions for the adolescent but that these references

have a basic and lasting effect on identity formation, then








the degree of matching between the parental sex role

preference and the sex role norms of the adolescent's peer

group is a suitable measure of factor three.

However, the more general culture of the parental

generation is also an available source of sex role norms.

Earlier we noted Mussen & Rutherford's and Heilbrun's

assumption and the supporting data that indicated that the

norms of this culture are available from early development

on to assist the parents in the masculinization of the boy

by providing well articulated models and by consistently

rewarding the boy for learning them. In this way long

before adolescence and through adolescence the youth has

been exposed to the traditional sex role stereotype of the

general culture which will have aided in the consolidation

of his sex role identity to the extent that it recognized

and confirmed the sex role identification he has made with

his father. We have already included this effect of the

traditional sex role stereotypes upon sex role development

and identity achievement in our second factor: the degree

to which the father's sex role preference is confirmed by

the traditional sex role stereotype is, in other words, the

degree of masculinity of the father's sex role, and the

degree to which the mother's sex role preference matches

the traditional sex role stereotype is, in other words, the

degree of femininity of the mother's sex role.

Utilizing identity theory to explore the complex in-

fluences of two sets of sex role norms we can conceptualize








both the traditional stereotypes of the parental culture

and the peer sex role norms in terms of the extent of

recognition they provide of the youth's initial identifi-

cation with his/her parent's sex role preference. Identi-

fication with the parental preference would seem to provide

the highest degree of recognition if that preference both

matched the parental culture's stereotype and the peer norms.

In this case the adolescent's sex role self-definition built

upon identification with his parent has been consistently

"recognized" verified in experiences of psychosocial fit-

tedness from childhood through adolescence.

However, the recognition provided by these social roles

will facilitate identity development only to the extent that

the roles encompass and legitimize the individual's inner

experience. Several studies have discussed the limitations

of both the feminine and masculine traditional sex roles

(Brown, 1958; Douvan & Adelson, 1966; Howard, 1960; Rudy,

1968; Sanford, 1967; Smith, as cited in Lynn, 1966). The

conclusion is that the traditional sex role stereotypes do

not affirm a wide enough range of human experiences and,

thus, total and consistent identification with just the

masculine stereotype or just the feminine stereotype would

be expected to result in a considerable amount of identity

diffusion. Thus, in the case considered above in which

parental sex role identification both matched the general

culture's stereotype and the peer group norms, the individ-

ual's opportunities for recognition of his inner experiences








are all alike and similarly limited, resulting in some

remaining identity diffusion. This would be the case of

a boy with a high masculine father (matching between parental

sex role identification and the general culture's stereo-

type) whose peer group norms do match the familial identi-

fication. This category may be considered those who most

need "men's liberation." They receive no external recogni-

tion or opportunity to express more feminine sides of the

self and, thus, identity integration is less complete.

Non-matching between parental sex role identification

and both the traditional sex role of the general culture

and the peer group norms is a situation of very low recogni-

tion. Thus, the individual does not have the societal con-

firmation of his childhood identification necessary to

consolidate an adult identity. This would be the case of a

boy with a low masculine father (non-matching between parental

sex role identification and the general culture's stereo-

type) whose peer group's sex role prescriptions do not match

the familial identification.

Matching between parental identification and either the

general culture's stereotype or the peer group norms, but

not both, is a situation of moderate matching, and, thus,

the individual has the societal confirmation of his child-

hood identification necessary to anchor his self-image but

also the opportunity through another sex role stereotype to

express another side of his human experience. This would be

the following two cases: 1) A boy with a low masculine








father (non-matching between parental identification and

the general culture's stereotype) whose peer group's sex

role norms do match the familial identification. 2) A boy

with a high masculine father (matching between parental

identification and the general culture's stereotype) whose

peer group norms do not match the familial identification.

Projecting from the above research on the effect of the

traditional feminine role itself on female identity achieve-

ment, the operation of this same underlying principle of an

inverted U curve relation between levels of matching and

identity may not appear in the interaction between factors

two and three in female identity achievement. The negative

social valence of feminine traits and the impediment to

identity achievement of feminine traits such as high depen-

dency, submission, lack of independent decision making, and

low achievement may complicate the effects of matching on

identity achievement. For example, identification with a

high feminine mother and the early learning of traits of

high dependency, lack of autonomy, etc., may interfere with

the girl's utilization of the positive identity opportunity

of a low feminine peer group. Rather than use this group

as an opportunity to identify with and express another side

of her self and, thus, better integrate her identity, she

may just depend on them.

1> Having also considered the possibility that the sex-

typed individual may be less "mentally healthy" than one

who combines both masculine and feminine personality traits,








Bem (1974) has developed a sex role inventory that makes it

possible to assess mixed as well as stereotypical sex role

self-concepts. While in the past a person has had to score

as either masculine or feminine on sex role measures because

these characteristics have been conceptualized as bipolar

ends of a single continuum, Bem has constructed a sex role

inventory that treats masculinity and femininity as two

independent dimensions. On the Bem Sex Role Inventory (BSRI)

it is possible to characterize a person as masculine sex-

typed, feminine sex-typed, or androgynous a mixed sex role

that endorses equally both masculine and feminine attributes.

In our discussion of level of matching we were exploring

the possibility that the adolescent who has had exposure to

both masculine and feminine sex role expectations in parent

and peer cultures would score high on identity achievement

because of the greater extent of recognition of the individ-

ual's inner experience provided by a wider range of role

opportunities. We can look at this question by analysis of

the interaction of factors two and three. If we assessed

the adolescent on Bem's scale we could explore two related

questions. 1) We could follow the process of sex role

identification "inside the individual" and see whether mixed

role opportunities are indeed integrated into the individual's

identity resulting in an androgynous sex role self-concept

and then 2) we could test the relation between androgyny in

the adolescent's sex role self-concept, as well as in his

sex role opportunities, and his level of identity achieve-

ment.













CHAPTER II
HYPOTHESES


Utilizing the framework of Erikson's identity theory

to conceptualize the relation between sex role development

and overall identity formation, we formulated the following

general thesis: The quality and coherence of the sex role

identifications that the adolescent must attempt to inte-

grate will vary according to the degree of satisfactory

interaction he has experienced with a trustworthy and mean-

ingful hierarchy of sex roles in his family. In the previous

section, based on the related empirical evidence on parental

sex role identification and peer group influence, we devel-

oped, and in some cases supported, the operation of three

specific factors from this general thesis. The following

are the hypotheses we used to test the contributions of each

of these separate factors and of their interactions:

1) A rewarding and nurturant relationship with his

same-sex parent will be positively related to the level of

overall identity formation achieved in late adolescence.

(Factor 1)

2) a. For the male adolescent, the level of his father's

masculine sex role stereotyping will be related to the level

of overall identity formation he achieves in late adolescence.

(Factor 2)








b. For the female adolescent, the level of her

mother's feminine sex role stereotyping will be related to

the level of overall identity formation she achieves in late

adolescence.

3) a. For the female adolescent, the level of matching

between her mother's feminine sex role stereotyping and the

feminine stereotyping of her peer will be related to the

level of overall identity formation achieved. (Factor 3)

b. For the male adolescent, the level of matching

between his father's masculine sex role stereotyping and the

masculine stereotyping of his peer will be related to the

level of overall identity formation achieved.

4) a. For the male adolescent, the level of overall

identity formation achieved as a function of his father's

level of nurturance will differ according to the level of

his father's masculine sex role stereotyping. (Interaction

of Factors 1 & 2)

b. For the female adolescent, the level of overall

identity formation achieved as a function of her mother's

level of nurturance will differ according to the level of

her mother's feminine sex role stereotyping.

5) a. For the female adolescent, the level of overall

identity formation achieved as a function of the level of

matching between her mother's sex role and the feminine sex

role stereotype of the general culture will differ according

to the level of matching between her mother's feminine sex

role stereotyping and the feminine stereotyping of her peer.

(Interaction of Factors 2 & 3)








b. For the male adolescent, the level of overall

identity formation achieved as a function of the level of

matching between his father's sex role and the masculine

sex role stereotype of the general culture will differ

according to the level of matching between his father's

masculine sex role stereotyping and the masculine stereo-

typing of his peer.

6) a. For the male adolescent, the level of overall

identity formation achieved as a function of the interaction

of his father's level of nurturance and the level of his

father's masculine sex role stereotyping will differ according

to the level of matching between his father's masculine sex

role stereotyping and the masculine stereotyping of his

peer. (Interaction of Factors 1, 2, & 3)

b. For the female adolescent, the level of overall

identity formation achieved as a function of the interaction

of her mother's level of nurturance and the level of her

mother's feminine sex role stereotyping will differ according

to the level of matching between her mother's feminine sex

role stereotyping and the feminine stereotyping of her peer.

7) The level of overall identity achievement for male

adolescents will be significantly higher than the level for

female adolescents.

8) a. For the female adolescent, the degree of androg-

yny in her sex role self-concept as a function of the inter-

action of her mother's level of nurturance and the level of

matching between her mother's sex role and the feminine sex








role stereotype of the general culture will differ according

to the level of matching between her mother's feminine sex

role stereotyping and the feminine stereotyping of her peer.

(Interaction of Factors 1, 2, & 3)

b. For the male adolescent, the degree of androgyny

in his sex role self-concept as a function of the interaction

of his father's level of nurturance and the level of matching

between his father's sex role and the masculine sex role

stereotype of the general culture will differ according to

the level of matching between his father's masculine sex

role stereotyping and the masculine stereotyping of his peer.

9) a. For the male adolescent, the degree of androgyny

in his sex role self-concept will be related to the level of

overall identity formation achieved.

b. For the female adolescent, the degree of androg-

yny in her sex role self-concept will be related to the level

of overall identity formation achieved.













CHAPTER III
METHOD

Subjects



Subjects were 56 male and 56 female students attending

the introductory behavioral studies classes at Santa Fe

Community College. Participation in the study was made a

course requirement. This college and course were used with

the hope of tapping a wider range of adolescent sex role

styles i.e. very traditional as well as ultra-liberated -

than may be found in university introductory psychology

courses.

Of the subjects pooled in this manner, data obtained

only from those in the age bracket of 18-25 were used in the

main analyses of this study. This selection was based on

the following reasons: 1) We are studying factors affecting

the ease or difficulty of identity crisis and consolidation,

and theoretically late adolescence has been indicated as the

time of this crisis. Thus, variations between subjects on

the accomplishment of this developmental task should be most

extreme at this time, and, taking into consideration wide

individual and cultural variations, most theories define late

adolescence as covering this particular time span (Muus,

1962). 2) Previous researchers have typically used subjects

either in this discrete age range or in the category of








freshman senior in college, and, thus, comparison of our

results would seem more valid if we matched their data on

this dimension.

Data was collected from additional BE classes until we

had 56 male and 56 female subjects in this age range.



Measurement of Variables


Nurturance

The Parent-Child Interaction Rating Scales (P-CIRS)

developed by Heilbrun for his study of the relation between

parental sex roles, nurturance, and consistency of adolescent

behavior was selected as our measure of the factor of nur-

turance. One advantage of this omnibus measure is that it

brings together the various facets of nurturant behavior

that previous studies have shown to correlate with high

parental identification and/or high identification with the

appropriate sex role (Heilbrun, 1964b; Mussen, 1961; Mussen

& Rutherford, 1963; Payne & Mussen, 1956). Another ad-

vantage of the P-CIRS is that it was specifically designed

to evaluate the late adolescent's impression of his rela-

tionships with parents while growing up and, thus, precisely

fits the developmental time focus of our study. Although

scores on the P-CIRS have not yet been related to theoreti-

cally consistent behavioral measures, Heilbrun reports psycho-

metric analyses of mother/father differences on the scale

that are evidence of its validity as a measure of nurturance








(1964b). In his administration of the P-CIRS to 61 male

and 63 female college students, he found: 1) The mother

was attributed higher total nurturance significantly more

often than the father which is in accord with several studies

indicating that mothers provide greater nurturance to their

children than fathers. 2) More masculine fathers were

perceived as significantly less nurturant than more feminine

fathers by the combined male and female subjects, as would

be expected from the research evidence that nurturance is

a feminine sex-typed trait.

The eight nurturant modes tapped by the scale are

1) affection I (degree of affection felt for subject);

2) affection II (degree of affection physically expressed

toward subject); 3) approval of subject and his behavior;

4) sharing of personal feelings and experiences; 5) concrete

giving (gifts, money, etc.) to subject; 6) encouragement

of subject in meeting responsibilities and pursuing personal

interests; 7) trust placed in subject; 8) sense of security

felt by subject in relations with parents.

Each mode is presented with a five-point rating scale

with each point anchored by a descriptive phrase. Five

always represents the highest degree of perceived nurturance,

although the direction of scoring is alternated from one

page to the next to counteract position rating sets. Female

subjects rated their mother for each mode, and male subjects

rated their father for each mode. The mother/father


~








nurturance score for each subject is the cumulative total

of the subject's eight ratings for his same-sex parent.

The median father nurturance score for the males was used

as the cutoff to establish high and low father nurturance

groups. Likewise, the median mother nurturance score for

the females was chosen as the cutoff for high and low nur-

turance groups.



Parental Sex Role Stereotyping

The level of masculine stereotyping of the father's

sex role and the level of feminine stereotyping of the

mother's sex role were assessed by the subject's ratings of

his mother/father on the Adjective Check List (ACL) (Gough

& Heilbrun, 1965). From the 300 behavioral adjectives of

the ACL the subject was asked to check those which were

descriptive of his same-sex parent. The adjectives were

scored on 15 personality variables taken from Murray's

catalogue of manifest needs. Nine of these 15 variables

have been found to be sex-typed based upon the ratings of

400 male and female college students.1 Achievement, auton-

omy, dominance, and endurance were rated as more character-

istic of fathers while deference, affiliation, succorance,




iSubjects in Heilbrun's study (1964b) rated their
parents on the same 15 descriptive paragraphs used by judges
in the initial selection of adjectives for the 15 ACL scales.
Thus, there is reason to presume that the behavioral var-
iables being assessed in the ACL parental description are
the same as those rated by Heilbrun's students. On this
basis we can apply the stereotyping of those nine variables
to their representatives in the ACL scales.








abasement, and nurturance were selected as more character-

istic of mothers.

In our scoring we first established high and low groups

for mothers on each of the nine sex-typed scales by utilizing

the median score on each scale. The same procedure was used

to establish high and low groups for fathers on the nine

scales. Then the number of times on these nine scales a

male subject's ratings of his father produced a high or low

score in line with the masculine stereotype was calculated

for each subject. Likewise, the number of times on these

nine scales a female subject's ratings of her mother pro-

duced a high or low score in line with the feminine stereo-

type was calculated for each subject. Finally, the median

score for fathers on this last calculation of frequency of

conformity to the masculine stereotype, in terms of these

nine scales, was chosen as the cut-off to establish high

and low groups of masculine sex role stereotyping. Similarly,

the median score for frequency of conformity to the feminine

stereotype, in terms of the nine scales, was chosen as the

cut-off to establish high and low groups of feminine sex

role stereotyping for mothers.



Level of Matching between Parental and Peer Sex Roles

To assess the factor of matching between parent and

peer sex roles we needed instruments that met the following

two demands. 1) The assessment of the parent's and peer's

sex roles would both measure the same set of specific








behaviors. Otherwise, the matching of the two sex role

models would be impossible to determine objectively. 2) It

is not obvious to the subject that he is being asked to

make comparisons between his parent's sex role and his peer's

sex role.

We employed a comparison of a subject's ratings of his

same-sex parent and his same-sex peer on the ACL because it

met both of these demands as well as providing 1) a measure

of the masculinity or femininity of parental sex roles based

on 400 college students' perceptions of these stereotypes

and 2) a precise scoring method for determining the degree

of similarity of same-sex peer and parental sex roles for

each subject.

To reduce the subject's tendency to compare parent and

peer ratings on the ACL, other measures and time were inter-

polated between the two ratings. Administration and scoring

of the ACL to assess parental sex role preference has been

described above. At a different testing session the subject

was asked to check those adjectives on the ACL which were

descriptive of his same-sex best friend.

The level of matching between the subject's perceptions

of parental and peer sex roles was determined as follows:

The subject's ratings of his same-sex best friend were cate-

gorized as high (H) or low (L) on each of the nine scales

on the basis of the median cutoff established for the same-

sex parental scores. Then, following the same procedure

described to determine parental sex role stereotyping, the








number of times on these nine scales a subject's ratings of

his best friend produced a H or L in line with the masculine

stereotype was calculated for each male subject. For

females, we computed the number of times on these nine

scales a subject's rating of her best friend produced a H

or L in line with the feminine stereotype. Next, the median

score within the male peer distribution for frequency of

conformity to the masculine stereotype was chosen as the

cut-off to establish H and L groups of masculine sex role

stereotyping for male peers. Likewise, the median score

within the female peer distribution for frequency of con-

formity to the female stereotype was chosen as the cut-off

to establish H and L groups of feminine sex role stereo-

typing for female peers. Having classified both parent and

peer groups into H and L groups on their sex appropriate

stereotype, a subject's parent and peer were categorized as

matching if both scored H or both scored L, and as non-

matching if they differed.



Socioeconomic Status

As the sex role prescriptions of the subject's parents

and peers may be strongly influenced by their particular

social class membership, any relations found between identity

achievement and parent and peer sex roles may be confounded

by the influence of the factor of SES. The Hollingshead

scale (1965), a two factor index of social class, was in-

cluded in the design in order to provide a means of assessing

the presence of such possible contamination.








Identity Achievement

The Ego Identity Incomplete Sentences Blank (EI-ISB)

was chosen as the measure of overall identity achievement.

This choice was made on the basis of the instrument's

theoretical development and construct validity. The 23 items

of this projective test were chosen from a pool of 50 by

five clinically trained psychologists on the basis of the

relevance of the items to Erikson's conception of ego iden-

tity (Marcia, 1964). Although every measure of identity in

the literature has purported to be based on Erikson's theory,

all others have arbitrarily chosen to assess only some part

of the criteria Erikson has defined as indicative of ego

identity while the present measure may provide a more valid

measure by including a fuller range of the criteria. These \

criteria are 1) the presence or absence of some period of

re-thinking, trying out, and actively choosing among alter-

native life plans, labeled by Erikson and Marcia as crisis;

2) the degree of personal investment expressed in a course

of action or belief, labeled by Erikson and Marcia as com-

mitment; 3) degree of continuity of the self-definition over

time; 4) degree of differences between real and ideal selves;

5) degree of ego strength the ability to orient to, plan,

and initiate activity that integrates the timetable of the

organism with the structure of social institutions; 6) the

endorsement and attainment of finding a niche in society that

really fits the individual, concretely represented in the

two adolescent phase tasks of selection of an occupation








and formation of an ideology.

Unlike most other measures of ego identity, the EI-ISB

has been used in more than one study and has been shown to

be significantly and/or consistently related to performance

on a whole network of behaviors in the direction predicted

by identity theory. The EI-ISB was found to be signifi-

cantly related to an interview measure of the four statuses

of identity development derived from the two psycho-social

criteria of crisis and commitment (Marcia, 1964; 1966).

The hypothesized linear relation of the four statuses on the

EI-ISB was found to be statistically significant with the

order as follows: Diffusion lowest, then Foreclosure,

Moratorium, and Achievement highest. A significant cor-

relation was found between the EI-ISB and a self-esteem

questionnaire as hypothesized in both the 1964 and 1966

studies. In both studies the correlation between the EI-ISB

and a conceptual task used as a measure of vulnerability to

evaluative stress were non-significant but in the expected

direction of higher identity related to better performance.

In the pilot study for the 1964 project scores on the EI-ISB

were significantly higher for junior and senior subjects

than for freshmen and sophomores. Class differences were

in the same direction for the main study but non-significant.

However, findings on female identity development in-

dicated the need to modify the EI-ISB before it could serve

as a valid measure of identity achievement in that sex.

Schenkel & Marcia (1972) found that ego identity statuses in








women based on an interview dealing with premarital inter-

course differed more on the dependent variables of self-

esteem, anxiety, and authoritarianism than those based on

the ORP interview the standard occupation, religion,

politics interview that the men's statuses have been based

on in Marcia's series. They concluded that the higher

predictive power of the sex interview supports Erikson's

idea that a woman's potential for reproduction, with its

biological and social implications,must be taken into account

in her identity development. Identity as a process is

probably not different for men and womenbut the issues

around which the process occurs may be, and these differences

in issues also may be influenced by changes in the social

scene. For example, with more college men and women living

as couples, interpersonal issues such as intimacy and pre-

marital intercourse, may have become important in the iden-

tity formation of college men as well. Hence any measure

for determining identity must remain flexible in content in

order to tap validly the underlying process (Schenkel &

Marcia, 1972).

To take into account the sexual area for both men and

women we included the following five stems which were each

adaptations of a parallel question concerning vocational or

ideological choice in the original questionnaire:

To change my mind about my feelings toward pre-
marital intercourse
(like stem 18)

When I consider my sexual standards and behavior
in the light of my family's
(like stem 3)








Sticking to my standards on pre-marital inter-
course (like stem 5)

Whether I sleep with someone depends on
_(like stem 15)
As compared with my views on sex in high school,
I (like stem 16)

Additionally, on the basis of their finding that the

female statuses based on the occupational interview had the

poorest predictive power, Marcia & Schenkel suggested that

occupation has to be more broadly conceived for women. This

assumption was based on Horner's finding that professional

commitment was tempered with marriage and family planning

for many women and is supported by Gump's findings that

women high on ego strength had well defined vocational plans

that took into account their interest in a man. Several

questions in the original EI-ISB are specifically directed

to the occupational area and, thus, to accurately tap the

degree of crisis and commitment in the female adolescent

on those questions, scoring was modified to focus not only

on crisis and commitment to a specific job title but also

on the degree of crisis and commitment to a total life plan

including vocation, marriage, and family planning. On the

basis of studies and speculation showing an increase in the

importance of the affiliative need in college men this modi-

fied scoring was done for males also.

Each of the stems in the EI-ISB was given a score of

3, 2, or 1 according to criteria and examples provided in

the EI-ISB Scoring Manual (Appendix II). The Manual is based








on the above six theoretical criteria and response examples

taken from pilot studies by Marcia and the present author.

Reliability of measurement. Marcia (1964) reports

inter-scorer reliability on the EI-ISB but gives no infor-

mation on the test-retest reliability of this instrument.

A pilot study of our modified EI-ISB was conducted prior to

the main study to evaluate this dimension of our measure as

well as to provide practice protocols for training the three

psychology graduate students who would rate the EI-ISB.

The EI-ISB was administered twice, with a two week

interval intervening, to 16 subjects from two BE 100 classes

at Santa Fe Junior College.1 The first 16 protocols were

each scored by the three raters during practice scoring

sessions, and then each protocol was assigned a final score

agreed upon by all three raters and the experimenter. The

second administration was scored after the raters had reached

their criterion of inter-rater agreement, and, thus, the

protocols were merely divided among the three for scoring.

The test-retest method for estimating reliability based

on a one-way analysis of variance was applied to these two

sets of scores (Lord & Novick, 1968). The estimated reli-

ability coefficient was .77. Interpreting this statistic,



A third class of five subjects was included in the
first administration of the pilot study but could not be
re-tested during the same time interval and, thus, was not
included in this analysis.








we can say that 3/4 of the variance in these scores is

due to individual differences between the subjects, and only

1/4 of the variance is attributable to error arising from

differences in the two testing situations.or changes in the

subject's ego identity over the two week interval. Con-

sidering the present stage of development of this measure,

this reliability factor appeared satisfactory.

Training of raters and inter-rater agreement. As

briefly described above, protocols from the first admini-

stration of the EI-ISB in the pilot study were used to train

the three raters. However, prior to use of this pilot data,

the raters were trained in scoring of the EI-ISB through study

of Erikson's theory of identity formation, Marcia's concepts

of crisis and commitment, and the EI-ISB scoring manual.

Following individual study, raters and experimenter met three

times to iron out any misunderstandings concerning the theo-

retical construct we were attempting to measure and the speci-

fic scoring criteria for each item in the EI-ISB. During two

of these sessions the raters practiced scoring protocols from

another pilot study conducted by the experimenter.1 All dis-

crepancies in scoring were discussed, item by item.




1At the time of this study the author was employed at
a residential treatment program for drug dependent adolescents.
All five instruments of the main study were administered to
five male and five female volunteers from this program in
order to evaluate the comprehensibility of both written
instructions and questionnaire items as well as to estimate
time allotment per questionnaire.









The 23 protocols from the Santa Fe pilot study were

divided into three sets, to be scored prior to three more

training sessions. All training was aimed at establishing

a high level of inter-rater agreement on the same protocol,

item by item, not at high inter-rater reliability. For the

purposes of our study we needed to establish not just a

predictable relation between raters' scores but as near

identity of scores as possible. This necessity was based

on two methodological givens. It would be very uneconomical

in terms of human labor to rate 112 protocols three times.

We wanted to establish that the raters were rating identi-

cally enough that we could divide the protocols among the

three of them for scoring. Secondly, discrepancies of five

points or more on a given protocol, a situation that can

exist even though inter-rater reliability is high, would

influence what level of identity achievement was assigned

to that subject and, thus, possibly confound any numerical

relations between identity and the three factors being in-

vestigated.

Each training session consisted of discussion of dis-

crepant scores, clarification of relevant scoring criteria,

and assignment of a final score agreed upon by all three

raters and the experimenter.

Before discussion and adjustment of scores, inter-rater

agreement for the current set was assessed by an F test

based on a one-way analysis of variance with repeated measures.

Summaries of these three analyses are presented in Tables









1-3. Only the F test of the second practice rating was

significant at the .05 level, indicating that only in the

second set was there a significant difference between raters'

scores on the same subjects. By the third sample the raters

were within two points of each other (with a possible score

range of 54 points) for six of the nine subjects and differed

by only three points on the other three subjects of that

sample. At this point we felt all three raters were operating

with identical criteria in mind, and further practice would

not erase differences due to human error. The obtained F

for this sample was .06, far below the 3.63 required for

significance at the .05 level. A test of raters' agreement

over the combined three sets produced a non-significant F

at the .05 level (Table 4). We concluded that there was no

overall significant difference between raters now, and we

could proceed with scoring the main study, following one more

special training session.

This seventh session was devoted to a review of scoring

procedures for the five new stems, added to Marcia's EI-ISB

in order to take into account identity development in the

area of sexual values and behavior. Through our experience

scoring these five stems in the pilot study by using the

scoring criteria of the five parallel stems in the original

EI-ISB, we had agreed upon several revisions in these criteria

that make them more applicable to the new items. These re-

visions were based on the translation of the parallel criteria

into the specifics of sexual issues. Additionally, answers









Summary Tables of F Tests of Raters
Scoring for Three Samples of Pilot Study


TABLE 1


SAMPLE 1


Source SS df MS F

Raters 25.6 2 12.8 1.22

Error 83.7 8 10.5



TABLE 2 SAMPLE 2


Source SS df MS .F

Raters 56.07 2 28.04 5.14*

Error 87.26 16 5.45



TABLE 3 SAMPLE 3


Source SS df MS F

Raters .3 2 .15 .06

Error 41. 16 2.6



TABLE 4 3 SAMPLES TAKEN TOGETHER


Source SS df MS F.

Raters 29.83 2 14.92 2.48

Error 264.17 44 6.


*p_.05








typifying each of the three criteria levels for each new

stem were compiled during this session from the three

samples of the pilot study. These five stems were numbered

5, 11, 18, 24, and 28, and scoring criteria and examples

for each are presented in the scoring manual (Appendix II).

In order to insure continued rater agreement through-

out the scoring of the main study, the experimenter met with

the raters for practice scoring and comparison after the

first and second thirds of the 112 protocols had been scored.



Androgyny

The degree of androgyny of the adolescent's sex role

self-concept was assessed by the Bem Sex Role Inventory (BSRI).

The BSRI was chosen on the basis of its theoretical develop-

ment, the substantiation of several of these theoretical

assumptions through psychometric analyses of normative data,

and the beginnings of construct validation for the scale.

As Bem's goal was to be able to assess the relative degree

of both masculinity and femininity in the person's self-

concept, the test was developed so that, unlike existing sex

role inventories, it did not construe masculinity and feminin-

ity as bi-polar ends on a single continuum. On this basis

the BSRI was designed to include a Masculinity Scale to

reflect the level of endorsement of masculine sex-typed,

socially desirable personality characteristics and a

Femininity Scale to reflect the level of endorsement

of feminine sex-typed, socially desirable personality








characteristics. The test is constructed so that these two

scores are free to vary independently. A person is charac-

terized as masculine sex-typed, feminine sex-typed, or

androgynous as a function of the difference between his

scores on these two scales. A person is, thus, sex-typed

to the extent this difference score is high, and androgynous

to the extent the difference is low. Finally, a Social

Desirability Scale was included to assess whether the in-

ventory was simply tapping a general tendency to endorse

socially desirable traits.

Results from the normative samples (M = 561, F = 356)

demonstrate: 1) the Masculinity and Femininity scores are

empirically independent, 2) the Androgyny score is reliable

over a four week interval, and 3) the Social Desirability

score is uncorrelated with the Androgyny score and, thus,

the Androgyny score is not simply a measure of a general

tendency to respond in a socially desirable direction (Bem,

1974).

The first evidence of construct validity for the BSRI

comes from a recent study. First, the BSRI was administered

to classify subjects as masculine and feminine sex-typed or

as androgynous. As predicted, the so classified androgynous

subjects of both sexes exhibited masculine or feminine

characteristics depending on the situational appropriateness

of these behaviors while sex-typed subjects displayed be-

havioral deficits largely related to predicted inhibitions

of sex-typed inappropriate behavior (Bem, 1975).








In the administration of the BSRI the subject indicates

on a seven-point scale how well each of 60 masculine,

feminine, and neutral personality characteristics describes

himself/herself. The scale ranges from one ("Never or almost

never true") to seven ("Always or almost always true") and

is labeled at each point. The subject's masculinity score

equals the mean self-rating for all endorsed masculine items,

and the femininity score equals the mean self-rating for all

endorsed feminine items. The index of androgyny is the dif-

ference between a subject's masculinity and femininity scores.

The closer the androgyny score is to zero, the higher the

degree of androgyny present in that subject's sex role self

concept, while the higher the absolute value of the androgyny

score the more the person is sex-typed or sex reversed.



Procedure



The following instructions and sequence of events were

presented in each BE 100 class used in the study: In the

first testing session, the ACL for the same-sex best friend

and the PC-IRS were administered, in that order. Directions

for both tests were printed on the individual booklets.

(Refer to Appendices III & V) Information for the Hollings-

head Scale was requested on a cover sheet to the ACL.

One week later, the BSRI, the EI-ISB, and the ACL for

the same-sex parent were passed out and completed in class.

Instructions for all three were printed on the individual





53


booklets (Refer to Appendices VI, I, & IV).

Following the collection of all data, the class was

informed of the major hypotheses of the study and the rela-

tion of each instrument to these factors.














CHAPTER IV
RESULTS

Identity Achievement and the Factors of Nurturance,
Parental Sex Role Stereotyping, and Matching of
Parent and Peer Sex Role Stereotyping



Our first six hypotheses dealt with the contribution

to identity achievement of the three factors of nurturance,

parental sex role stereotyping, and matching of parental and

peer sex role stereotyping. Factor effects and interaction

effects were assessed, for males and females separately, by

a three-way analysis of variance of identity scores, using

a fixed effect, 2 x 2 x 2 design. The two levels of each

factor were as follows: High and Low Nurturance (HN, LN);

High and Low Masculine Stereotyping/Feminine Stereotyping

(HM, LM/HF, LF); Matching and Non-Matching (Match, NMatch).

As there were unequal cell frequencies, a least squares

analysis was performed. A summary of these analyses are

presented in Tables 5 and 6.

Imposing a .05 significance level, there were no signif-

icant effects for males or females. From the results of our

study we are unable to reject the null hypothesis of no

relation between level of identity achievement in late

adolescence and these three factors. However, inspection

of the analysis of male identity scores reveals one trend

worth noting: the obtained F for the factor level of father's










TABLE 5

Summary Table of Analysis of
Variance of Male Subjects' Scores on the EI-ISB



Source Partial SS df F Prob>F

Nurturance 5.6746 1 .1941 .6615

Parental Sex
Role 82.4810 1 2.8213 .0994

Matching .2098 1 .0072 .9328

Nurturance x
Parent 22.0483 1 .7542 .3895

Nurturance x
Match .5888 1 .0201 .8877

Parent x Match 25.1626 1 .8607 .3582

Nurturance x
Parent x
Match .0033 1 .0001 .9916









TABLE 6


Summary Table of Analysis of
Variance of Female Subjects' Scores on EI-ISB


Source Partial SS df F Prob) F

Nurturance 26.9055 1 1.0800 .3039

Parental Sex
Role .0506 1 .0020 .9642

Matching 10.7049 1 .4297 .5153

Nurturance x
Parent 4.9345 1 .1981 .6583

Nurturance x
Match 44.5649 1 1.7889 .1874

Parent x Match 23.1901 1 .9309 .3395

Nurturance x
Parent x
Match 58.1345 1 2.3336 .1332








masculine sex role stereotyping is .0995. Although the

unmet pre-requisite of an overall significant F prevents

statistical judgement of how significant the differences

between the identity scores of the two levels of this factor

are, we did calculate the means of these two groups. As

Table 7 indicates the mean identity score of male students

that perceived their father to be highly masculine stereo-

typed was greater than the mean score of those males with

low masculine stereotyped fathers.

As the presence of a father with a sex role preference

that is recognized by the general culture's stereotype

appears to facilitate identity achievement we wondered what

the impact was, if any, of peer recognition of this paternal

model. As there was no significant effect found for the

Matching factor alone or in interaction with the other two

factors, what we are reporting is, again, only a trend which

further research may support or reject. In Figure 1 we have

graphed the means of the four combinations of level of

father's sex role and level of peer matching. These are

presented separately for high and low nurturance groups.

The four categories of father's sex role stereotype and

peer matching line up in these positions from highest to

lowest identity achievement:

1) High Masculine Father/Non Matching of Peer Sex Role
(Both HN and LN groups)

2) High Masculine Father/Matching of Peer Sex Role
(HN group)

3) Low Masculine Father/Matching of Peer Sex Role
(LN group)








TABLE 7

Mean Scores on the EI-ISB of the
Two Levels of Father's Sex Role Preference (Male Subjects)



Levels N Mean

High Masculine 24 54.372

Low Masculine 32 51.539


Low Nurturance


HM



LM

H

LM


Match


NMatch


High Nurturance

HM



HM






LM LM


Match


NMatch


FIGURE 1

Graph of the Mean EI-ISB
Four Combinations of Level of
Stereotyping and Levell of Peer
Low Nurturance


Scores of the
Father's Masculine
Matching for High and
Groups








High Masculine Father/Matching of Peer Sex Role
(LN group)

Low Masculine Father/Matching of Peer Sex Role
(HN group)

4) Low Masculine Father/Non Matching of Peer Sex Role
(Both HN and LN groups)

Let us look at this:line-up in terms of the impact of

combined levels of general cultural and peer recognition.

Scoring highest in identity achievement, as expected from

our discussion of recognition on pages 26-28, and substan-

tially higher than all other groups, is one of the two cate-

gories exhibiting moderate recognition HM/NMatch. Moderate

recognition is the result of recognition of a high masculine

stereotyped paternal identification by the general culture's

stereotype but a lack of recognition of the paternal identi-

fication by the peer's sex role. Next in line is the HM/

Match (HN) group. This category represents high combined

recognition, the result of matching of the high masculine

stereotyped paternal identification by both the general cul-

ture's stereotype and the peer's sex role. We would have

expected them to score low in identity achievement (refer to

pp. 26-28). The third position is held by three groups

representing two recognition categories: 1) LM/Match (HN &

LN groups) representing moderate recognition produced by the

combination of a lack of matching between the low masculine

stereotyped paternal identification and the general culture's

stereotype but confirmation of the paternal identification

by the peer's sex role. 2) HM/M (LN group) representing

high combined recognition as analyzed for the HM/Match








(HN group) above. Thus, within this position we have a

high recognition group scoring low in identity as expected

and both groups of a moderate recognition category scoring

unexpectedly low. The lowest position, in line with our

expectations, is held by the category exhibiting low recog-

nition LM/NMatch (Both HN & LN groups). Low recognition

is the result of a lack of recognition of the low masculine

stereotyped paternal identification by either the general

culture's stereotype or the peer's sex role.



Difference in Identity Achievement
between Male and Female Subjects



Our seventh hypothesis was that the level of identity

achievement for male adolescents would be significantly

higher than the level achieved by female adolescents. Mean

identity scores for male and female subjects are presented

in Table 8. Using a one tailed t-test of the difference

between the two means, the obtained t of -3.443 did not reach

significance (at the .05 level with 120 df, t)1.658). It

is certainly worth noting that if we had predicted that fe-

males would score higher than males, the sign of the obtained

t would have been reversed, and the obtained t of +3.443

would be highly significant (at the .001 level with 120 df,

t03.160).

One methodological explanation for the unexpected

direction of this difference is that the inclusion of the










TABLE 8

Mean EI-ISB Scores for
Male and Female Subjects



Males Females

N 56 56

M 52.446 55.821

Variance 29.015 24.804





TABLE 9

"Adjusted" EI-ISB Scores for
Male and Female Subjects



Males Females

N 56 56

M 42.589 45.054

Variance 24.865 18.270








five stems dealing with sexual values had tipped the scale

in favor of the females. Traditional role definitions as

well as recent observation of college students support the

premise that adolescent females may be more advanced than

their male counterparts in the area of sexual ideals. To

test this possibility by taking into account the contribution

to the overall identity score of the five stems evaluating

sexual values, we subtracted the scores on these items from

each subject's total score and performed a t-test of the

adjusted male and female means. Mean identity scores minus

the five sex questions are presented in Table 9. Again using

the reversed hypothesis of female identity achievement being

greater than male achievement, the obtained t of 2.811 is still

highly significant (at the .005 level with 120 df, t)2.617).

Thus, the inclusion of sex related questions on the identity

measure did not bias the results in favor of the women.

With or without these questions, the late adolescent females

in our sample are substantially advanced in identity for-

mation compared to the males.



Androgyny and the Interaction of the
Factors of Nurturance, Parental Sex
Role Stereotypes, and Matching



Hypothesis 8 explored the process of sex role identifi-

cation "inside the individual" by predicting that the presence

or absence of mixed stereotypic sex role opportunities in

parent and peer models influence the degree of androgyny in








the youth's own sex role self-concept. Before looking at

the results for hypothesis 8 we want to present the BSRI

scores for our sample as they differed in important ways

from Bem's norms. 'We classified our sample as feminine,

near feminine, androgynous, near masculine, and masculine

according to the same cut-off points Bem used in reporting

her normative data (Bem, 1974). We remind the reader that

our index of androgyny is the difference between a subject's

masculine and feminine scores on the BSRI. The closer the

difference score is to zero, the higher the degree of androg-

yny, while the higher the absolute value of the androgyny

score, the more the person is sex-typed or sex-reversed. As

Bem's cut-offs were in terms of the Androgyny t ratio, we had

merely to divide these cut-off numbers by 2.323 in order to

establish corresponding cut-off points on our "difference

score" scale of Androgyny. Table 10 presents this classifi-

cation schema applied to both our male and female subjects.

As Bem herself notes her cut-off points are somewhat

arbitrary and the percentages she presents are based on norm

samples too small to generalize to all college students across

the country. With this caution in mind we'll just note that

two consistent differences exist between our junior college

sample and both Bem's college and junior college samples.

Santa Fe women are much less feminine sex-typed and much more

androgynous while Santa Fe men are somewhat less androgynous

and much more near masculine or masculine sex-typed.


__































H t-I
-1 M
0 Mi
u u



1H 0
r-)d 0
0 0


rl0 J)
(3 )B 0i








o to -r
0 *UO



H Z *-ri


444


i -- E
Ho ad
m *Ir






Og






C
W1 a


a






0H
n3 II








O II









Sr-




Od









43 Ln
Id i1
0 II


H
6-1

+J.i0

O II


fOi



0 r

II II


0' ff1 N
VD in m e m

















N N N1 r-


0 c0 CO
.T m


N, CO


Ln oo o
cN H- tn








In terms of our three specific factors, hypothesis 8

dealt only with the interaction effect of all three on

androgyny scores. A three-way analysis of variance, for

each sex separately, using the same design as the analysis

of identity scores, was performed to evaluate the signifi-

cance of this interaction effect. Summaries of these

complete analyses are presented in Tables 11 and 12.

As can be seen from the tables the three-way inter-

action effect for males and females did not reach signifi-

cance. Thus, from the results of our study we are unable

to reject the null hypothesis of no relation between the

degree of androgyny in late adolescence and the interaction

of the three factors of same-sex parental nurturance, same-

sex parental sex role stereotype, and matching of parental

and peer sex roles.

None of the F tests of the other possible effects of

those three factors on androgyny reach significance and,

thus, do not allow even post hoc support for any significant

relation between our conception of important variables in

sex role development and the concept of androgyny. However,

the F test of the hypothesized three-way interaction effect

does approach significance in the analysis of female scores.

The obtained F is .0926. Likewise, the obtained F for the

Matching effect on female androgyny scores is not far below

significance level, having a probability of .0964. Although

we can not say anything conclusive concerning the different

groups' androgyny scores, we decided to calculate the means










TABLE 11

Summary Table of Analysis of
Variance of Female Subjects' Scores on BSRI


Source Partial SS df F Prob >F

Nurturance .0156 1 .0589 .8092

Parental Sex Role .0692 1 .2623 .6109

Matching .7592 1 2.8751 .0964

Nurturance x Parent .3267 1 1.2373 .2715

Nurturance x Matching .0450 1 .1703 .6817

Parent x Matching .5444 1 2.0618 .1575

Nurturance x Parent x
Matching .7779 1 2.9458 .0926










TABLE 12

Summary Table of Analysis of
Variance of Male Subjects' Scores on BSRI


Source Partial SS df F Prob > F


Nurturance .1803 1 .8962 .3485

Parental Sex Role .0172 1 .0855 .7712

Matching .2031 1 1.0095 .3201

Nurturance x Parent .0038 1 .0187 .8918

Nurturance x Matching .0230 1 .1145 .7365

Parent x Matching .3007 1 1.4943 .2275

Nurturance x Parent x
Matching .0695 1 .3455 .5595








of the female Match and NMatch groups and to graph the

androgyny scores of the eight different categories of the

three factor interaction to see if we could discern any

pattern related to our theoretical formulations.

As Table 13 indicates the mean androgyny score of the

female subjects with mother and peer sex role stereotypes

that match scored closer to androgyny than those whose

preferences did not match. In terms of Bem's cut-offs,

though, both groups fall within the near feminine category.

Figure 2 presents the androgyny scores of the eight

possible combinations of the three factors, with the HN and

LN groups graphed separately. The following is the sequence

from most androgynous to most feminine stereotyped of the

four HN groups:

1) Low Feminine Mother/Matching of Peer Sex Role
(This group is the only one of the four HN
classifications to score within Bem's androgyny
cut-off.)

2) High Feminine Mother/Matching of Peer Sex Role

3) Low Feminine Mother/Non-Matching of Peer Sex Role
(Positions two and three have almost identical
scores. Both are substantially lower on androgyny
than the first group, scoring within Bem's near
feminine classification.)

4) High Feminine Mother/Non-Matching of Peer Sex Role
(This group scores just .03 points below Bem's
feminine stereotype cut-off.)

The following is the continuum from most androgynous

to most feminine stereotyped of the four LN groups:

1) High Feminine Mother/Matching of Peer Sex Role
(This is the only one of the four LN groups to
score within Bem's androgyny cut-off.)










TABLE 13

Mean Scores on the BSRI of the
Two Levels of Matching for Female Subjects



Level N Means

Matching 28 .460

Non-Matching 28 .699


Sex .9
Typed
.8

.7

.6

.5

.4

.3

.2
SAndrog-
yny .1


Low Nurturance


.9
HF
.8

.7

.6

.5
LF
.4

.3


I I
Match NMatch


High Nurturance


J I
Match NMatch


FIGURE 2


Graph of the Mean Androgyny Scores
of the Four Combinations of Level of
Mother's Feminine Stereotyping and Level
of Peer Matching for High and Low Nurturance Groups








2) Low Feminine Mother/Non-Matching of Peer Sex Role
(Scores substantially lower on androgyny than the
first LN group but is still just .04 points higher
than Bem's androgyny cut-off.)

3) Low Feminine Mother/Matching of Peer Sex Role
(Scores well within the near feminine stereotype
classification)

4) High Feminine Mother/Non-Matching of Peer Sex Role
(This group scores just .04 points below Bem's
cut-off for the feminine stereotype classification,
receiving almost exactly the same score as its
parallel category in the HN groups.)



Identity Achievement and Androgyny



Our last hypothesis predicted that the degree of androg-

yny in the adolescent's sex role self-concept would be related

to his level of identity achievement. To evaluate this

relation, Pearson product moment correlations were computed

between EI-ISB and BSRI scores, for males and females

separately. We remind the reader that the relation we are

interested in is between identity and the bi-polar character-

istic of an androgynous/stereotyped sex role self-concept.

(Refer to pages 28-29) Thus, in this hypothesis the direction

of stereotyping masculine or feminine is not important.

For this reason the absolute values of the BSRI difference

scores were used in the computations. The obtained r did

not approach significance for either males or females. The

values of r and probabilities are given in Table 14.

As both Constantinople (1969) and Marcia (1964) found

that identity increased with year in college, and Erikson's























1)
01




4)
.a -
- H

r.to

44
-,1 0













0 )
H U)
P4












UH
rf










0a,









U
A o
o u












u 0
U









OI
OH
UO





r40
IU



U0
02


02
r-I H
rd 9











H
I
H






0,








C,

H










H
I
H
M


o ',





O* *















o0
o o

o 0




m H

H LO







C0 H

C CI










o o
0 0
* *
H 0
















H
I
H
V m


o o
0 o
o*
H o











I I






0 0!
o mn








o o
o o
o o

H 0









I I
I I


















H

M a


I I
I I











I I
I I





o o

o o
0 0


I I
I I











I I
I I


~









construct is a developmental one, we expected that scores

on the EI-ISB would be related to age of our subjects.l

Pearson product moment correlations were computed between

the two variables, for males and females separately. The

obtained r did not approach significance in either case.

(Refer to Table 14.)

Finally, we calculated r for the variables of age and

BSRI scores, in order to obtain more information about this

new construct of androgyny. Again, no significant relation-

ship was found between age and our measure of androgyny

(Table 14).


























1We used age rather than year in college as there would
probably not be much differentiation between the two possible
levels of the latter while the age levels of our subjects span
seven years.














CHAPTER V
DISCUSSION AND CONCLUSIONS



Our predictions concerning the relation between sex

role development and identity achievement in late adolescence

were based on a basic postulate of Erikson's theory: The

ease of re-synthesis of sex role identifications into a more

final self-definition is influenced both by past experience

with the models of the youth's family and the role opportun-

ities that are presently available to the adolescent in the

larger society. We expanded upon previous work in this area

by including among these role opportunities the alternative

sex role norms of the late adolescent's peer culture. Al-

though none of our results were significant at the .05 level,

several important trends emerged. Perhaps the most striking

finding is the negligible impact of the peer's sex role norms.

From trends in both the male and female data the early in-

fluence of the same-sex parent far outweighs the effect of

the peer sex role model the youth is exposed to in late adoles-

cence. As the details of the same-sex parental influence

differ for males and females, we will discuss them separately.

The most important influence on male identity achieve-

ment is the level of perceived traditional sex role typing

in the adolescent's father. Male students with a highly

masculine stereotyped father scored higher on identity than









those with a less masculine father. This trend is in

opposition to Mussen and Rutherford's and Payne and Mussen's

findings. However, it does support Heilbrun's findings

(1964b) and conclusion that the presence of a father with

a sex role preference that is reinforced by the traditional

sex role norms facilitates identity consolidation.

As this aspect of societal recognition of the paternal

identification appeared to facilitate identity consolidation

we explored what effect on identity achievement, if any, the

addition of peer recognition of the paternal model produced.

The results show that the four possible combinations of

general culture and peer culture recognition of the parental

identification are related to identity achievement in an

"almost" inverted U function (Figure 3) with low and high

levels of combined recognition related to lower identity

achievement and moderate levels related to higher identity

achievement. The obtained relation isn't perfect, but it

does generally support our conceptualization of the effects

of different levels of recognition on identity consolidation.

We'll review this conceptualization briefly and then discuss

the implications of its strengths and weaknesses empirically.

We had conceptualized the complex influence of the two sets

of sex role norms, from the general culture and the peer

culture, in terms of the extent of combined recognition they

provide of the youth's initial identification with his same-

sex parent's sex role preference (See pp. 26-28). Based on

theory and research we concluded that the stereotypical







75














-C
'a




0 ,
4 "




-,


Smr




0 *t
-,O





S0 0.
N a a g s -

U u-1-











,4 *-
C O






U)d 1 C C1





a 0








h i H






_1 -P
0 r.


> Cl)

hi '1C
N^& a








masculine role (or the stereotypical feminine role) is too

limited to encompass and legitimize the whole range of a

human's inner experience. Consequently, we assumed that

total and consistent recognition by the general culture's

norms and peer norms of an identification with a high

masculine stereotyped father, a case of high combined

recognition, would be expected to result in a considerable

amount of identity diffusion as the youth's opportunities

for recognition of his inner experience are all alike and

similarly limited. We expected that recognition of the

parental identification by either the general culture's

norms or peer norms, but not both, a case of moderate

recognition, would provide optimum conditions for identity

consolidation. Finally, a combination of no recognition

of the parental identification by either the general culture's

or peer norms, a case of low recognition, was seen as dis-

ruptive as the youth doesn't have any source of societal

confirmation of his childhood identification.

This schema breaks down empirically because of the low

impact of peer recognition on identity consolidation when

the father is low masculine stereotyped and, thus, general

culture recognition is lacking. As can be seen in Figure 3

our conceptualization that peer recognition of an identifi-

cation with a low masculine father would facilitate higher

identity achievement is only partially upheld. The low

masculine father groups with matching peer sex role score

higher than the groups with non-matching peer sex role.


~








However, peer support is not enough to counter-balance the

impact of the lack of a high masculine stereotyped father

(the lack of general culture support): Three of the low

masculine groups score lower than all four of the high

masculine father groups.

Likewise, the introduction of a new role possibility

by a non-matching peer group when the father is high masculine

does appear to heighten identity consolidation, as expected,

but not to a significant degree (Compare positions of

HM/Match and HM/NMatch groups in Figure 3). In conclusion,

for males, the peer sex role, whether it matches the father's

preference or introduces a new role possibility, can only tone

up or down the influence of the level of masculine stereo-

typing perceived in the paternal sex role.

The near significant trends in the analysis of female

androgyny scores also point to the importance of the early

influence of the same-sex parent. Let us take a look at the

line-up of female androgyny scores for the eight combinations

of nurturance, feminine stereotyping in the mother, and peer

matching. (Refer to pp. 68-70 in the Results) Two interesting

patterns emerge. The first concerns the near significant

effect of our third factor matching of mother and peer sex

roles. By looking at the graph one can see that the increase

in androgyny due to the matching factor can be accounted for

by the two groups HN/LF/Match and LN/HF/Match which not

only score substantially more androgynous than both the four

NMatch groups but the other two Match groups as well. Let









us take a look at the meaning of the three factor levels and

speculate on how they combine in these two groups to facili-

tate androgyny. For the HN/LF/Match group, high identifi-

cation with a low feminine stereotyped mother and the sup-

portive recognition for more masculine behavior by a

similarly low feminine stereotyped peer, amidst conditioning

by the general culture for high feminine stereotyping, may

enable a woman to develop a more androgynous sex role. In

interpreting the factors for the LN/HF/Match group, we pro-

pose a rebellion hypothesis to explain the outcome of a very

androgynous self-concept in conjunction with a maternal

model of high feminine stereotypy.. A low nurturant re-

lationship with a high feminine stereotyped mother for some

girls triggers rebellion against the mother and an attempt

to become her opposite. In terms of sex roles, this means

developing the masculine side of her personality. In line

with this hypothesis, the choice of a high feminine stereo-

typed peer as a best friend may be an attempt to gain the

long sought after "good mother." This peer choice balances

the daughter's masculine side by identification with feminine

traits in the best friend.

A second pattern that can be seen in the androgyny

scores of our eight groups is that the two scoring most

stereotypical have in common a high feminine stereotyped

mother and a peer with a non-matching sex role. Whether or

not a good relationship exists with the mother doesn't

appear to effect the outcome of high feminine stereotypy in









the daughter. These are the HN/HF/NMatch and LN/HF/NMatch

groups. These two start off with a maternal model of high

feminine stereotypy for sex role identification, and the

addition of a new role opportunity in adolescence doesn't

appear to be used in integrating a more androgynous self-

concept. As we speculated on page 28, the early learning

of feminine traits of dependency, submissiveness, and lack

of autonomy may have interfered with these particular girls'

utilization of the balancing identity opportunity of a low

feminine stereotyped peer model. Rather than use this group

to identify with and, thus, express the masculine side of

her self, she may just depend on them. Here we are proposing

a different reaction to the low nurturant mother of the

LN/HF/NMatch group than to the one in the LN/HF/Match group

discussed above. While the above group may have reacted to

maternal indifference with rebellion and an attempt to replace

the mother by a similar peer, the present group would appear

already too feminine stereotyped to rebel. In line with their

high feminine stereotyping they chose a high masculine stereo-

typed peer to depend on.

More support for the hypothesis of rebellion against a

low nurturant mother in certain girls is evidenced by the

reversal of positions on the androgyny scale of the HF/Match

and LF/Match groups depending on whether or not a good rela-

tionship with the mother exists. (Refer to Figure 2J When

a good relationship with the mother exists, the mother's sex

role preference is mirrored by the daughter in her own sex


__








role self-concept. If the mother is low feminine stereotyped,

the daughter becomes androgynous, if the mother is high

feminine stereotyped, the daughter becomes near stereotyped.

When a low nurturant relation with the mother exists, the

mother's preference is tied to development of its opposite

in the daughter's sex role self-concept. When the mother

is low feminine, the daughter now becomes near feminine

stereotyped, and when she is high feminine, the daughter

becomes androgynous.

A related hypothesis for these last two groups with a

low nurturant mother and matching peer sex roles is that they

score opposite to their mother's sex role preference due to

identification with a highly nurturant father. This specu-

lation was prompted by data by Heilbrun (1968) which indi-

cates that the manifestation of both masculine goal orienta-

tion and feminine interpersonal sensitivity in college females'

behavior, behavior which we would classify as androgynous,

is tied to identification with the father rather than the

mother. Clinical experience of the identification patterns

of women with uninvolved mothers also promotes this hypothesis.

In order to investigate this possibility, we related the level

of father nurturance to mother nurturance in our female

subjects. We found that out of the 15 LN/Match subjects

there were only two subjects in which the level of father

nurturance wasn't also low. Thus, there is no support in

our data for the assumption that the daughter's preference

of a sex role opposite to her mother's when the mother is









low nurturant is accounted for by a reaction of increased

closeness and identification with the father. Indeed, out

of all 29 subjects with LN mothers there were only five in

which this situation was balanced by HN fathers. What appears

to us to be the most striking finding in relating levels of

mother and father nurturance is that, in most cases, levels

of mother and father nurturance were the same. Out of the

56 subjects, there were only 12 differences in levels of

nurturance. A girl perceived her parents very much the

same in regards to how nurturant they were with her. In

terms of family dynamics, it is worth noting that this same

finding is true of our male sample. Out of all 29 male sub-

jects with LN fathers there were only seven in which low

father nurturance was balanced by high mother nurturance.

Out of the 56 male subjects there were only 14 differences

in levels of paternal and maternal nurturance. A boy also

perceives his parents very much the same in regards to how

nurturant they were with him.

The patterns we have presented highlight the determining

effect of the mother on her daughter's sex role self-concept

through the interactive influence of the level of nurturance

of the mother-daughter relation and the sex role preference

of the mother. Our speculation about these influences have

either ignored or explained the contribution of the peer's

sex role preference as an adjunct in working out of the

primary influence of the mother. We decided to check out

these assumptions by starting with the sex role preference








of the peer and tracing its relation to the direction of the

subject's own sex role self-concept. Determining the sex

role preference of the peer directly from the Match factor,

we looked at the four groups having low feminine stereotyped

peers. These were the following groups: HN/HF/NMatch;

HN/LF/Match; LN/HF/NMatch; LN/LF/Match. As can be seen from

the graph in Figure 2, three of the four groups having low

feminine stereotyped peers occupy the three highest positions

for feminine sex role stereotyping. The fourth group,

HN/LF/Match, discussed on page 78, scores within the androg-.

yny cut-off. Of the four groups having high feminine stereo-

typed peers HN/LF/NMatch; HN/HF/Match; LN/LF/NMatch;

LN/HF/Match the direction of the subjects' BSRI scores is

not highest feminine stereotyping, as might be expected, but

middle range scores for three of the four groups. Moreover,

the most androgynous score of all eight groups is received

by the fourth group, LN/HF/Match, discussed on page 78. From

this pattern we see that a low feminine peer a good model

for more masculine traits can only facilitate androgyny

in the female adolescent when certain conditions exist in

the mother-daughter relationship, namely, low femininity in

the mother and a highly nurturant relationship. Likewise,

the effect of high femininity in the peer is controlled by

the interaction of the level of the sex role preference of

the mother and the degree of nurturance in the mother-daughter

relation. In no way can it facilitate the highest levels

of feminine stereotyping, and in one case it actually coexists









with highest androgyny scores.

From the above discussion of female androgyny scores

we can tease out the general hypothesis that the reaction

of the daughter to the interactive influence of the level

of nurturance in the mother-daughter relationship and of

the sex role preference of the mother has a determining

effect on the direction of the daughter's sex role self-

concept whether she develops towards feminine stereotyping

or androgyny. Moreover, the data substantiate the related

hypothesis that the direction of influence of the role

norms of the peer is determined not by whether these norms

are high feminine stereotyped or low feminine stereotyped

but by how they fit into the early channelization of sex

role development produced by the girl's reaction to the

maternal model.

Specifically, we posited two main variables in the

daughter's reaction to the maternal model:

1) Imitation of or rebellion against the maternal
model

2) Choice of a best friend to: a. provide sup-
portive recognition of the maternal model; b. be
someone to depend on; c. be a role model sub-
stitute for the low nurturant mother.

We did not hypothesize about the specific workings of these

variables for three of the groups (LN/LF/NMatch; HN/LF/NMatch;

HN/HF/Match) because their intermediate scores on the androg-

yny continuum do not allow much room for speculation about

the directional impact of these variables.








It would be necessary for further research to check out

our hypotheses by demonstrating significant differences

between the eight groups in the predicted directions on

measures of: 1) Attitude toward mother, i.e., presence of

rebellion or identification 2) Orientation toward peer sex

role behavior, i.e., imitation of like behavior or perfor-

mance of role opposite behavior 3) Relation of choice of

best friend to unresolved needs in the mother-daughter rela-

tionship. At present, there is evidence in the literature

to support our conceptualizations of the female adolescent's

reaction to a low nurturant maternal model and of the in-

fluence of the development of high feminine stereotypy on

behavioral alternatives. A rebellion hypothesis in con-

junction with the choice of a peer similar to the mother, to

explain the final sex role self-concept of the adolescent

girl with a low nurturant mother,is grounded in a review of

research on individual styles of adolescent identity formation.

Many researchers posit and/or demonstrate a rebellion style

of handling the adolescent identity crisis that is related

to low acceptance and lack of encouragement of individuality

by parents (Karr & Dent, 1970; Keniston, 1965, 1968, 1970;

Murphy et al., 1963; Nixon, 1966). This seemingly.contra-

dictory style of a need to become independent of the authority

figure by endorsing his opposite behavior combined with a

need for guidance and identity, manifested by imitating

similar behavior, has been demonstrated to be true of Marcia's

"rebel" group the moratorium status (Podd et al., 1970).








Just these behavioral contradictions are often evidenced in

clinical experience with adolescents who are struggling to

establish an autonomous identity amidst a background of

judgemental and cold relations with parents.

One group that definitely does not fit this rebellion

hypothesis is the LN/HF/NMatch group which develops in a

highly feminine stereotyped direction. In a low nurturant

relation with the mother, why don't they rebel like the other

two LN groups (LF/Match and HF/Match), who score markedly

opposite to their mother's sex role preference, appear to

have done? From our finding of no significant effect of the

nurturance x parental sex role factor on androgyny but a

trend for the interaction effect of nurturance x parental

sex role x matching, it seems that knowledge of just the

level of the mother's nurturance and her sex role preference

is not enough to predict the daughter's own sex role self-

concept. There appear to be individual differences in re-

actions to the same levels on these two factors which we

believe are related to choice of a matching or non-matching

peer. Again, further research could investigate the dimen-

sions of these individual differences.

For three of the groups (LN/HF/NMatch; HN/HF/NMatch;

LN/LF/Match) we proposed that the early development of high

feminine stereotypy interfered with their utilization of the

balancing identity opportunity of a low feminine stereotyped

peer model. The girl's early identification with traits of

dependency, submission, etc., in the context of the high








feminine stereotype, interfered with imitation of and

expression of the more masculine behavior of her best friend.

She is trapped in her habitual style and just depends on her

more dominant, autonomous friend. Such a proposition is in

line with Bem's general hypothesis on the consequences of

sex role typing: "a non-androgynous sex role can seriously

restrict the range of behaviors available to an individual

as he or she moves from situation to situation" (1972). The

highly sex-typed person is seen as becoming motivated during

the course of sex role socialization to keep his behavior

consistent with an internalized sex role standard, a goal

which he accomplishes by suppressing any behavior that might

be considered undesirable or inappropriate for his sex

(Kagan, 1964). Thus, the HF female would be expected to

inhibit the expression of more masculine behaviors in order

to maintain her self-image as feminine. A recent study of

Bem's supports this hypothesis: Feminine stereotyped females

unlike androgynous or masculine stereotyped females failed

to show masculine independence even in a situation in which

it would have been the most appropriate style (1975).

A rather unexpected finding was the direction of the

substantial difference in identity achievement between the

male and female subjects of our sample. Based on previous

findings of male/female differences in identity achievement

during the college years (Constantinople, 1969; Sanford,

1966), as well as evidence concerning aspects of the tradi-

tional feminine role that have been shown to be antagonistic








to identity integration (refer to pp. 15-18), we predicted

that male subjects would score significantly higher on

identity achievement than females. On the contrary, our

data shows that the females scored substantially higher than

the males. As we noted in the Results section, if we had

made the opposite prediction and, thus, reversed the sign

of the t-test, the obtained difference would have been

highly significant.

There are two plausible explanations for this reversal.

First of all, our prediction of a current difference in

identity achievement between male and female students in

the first two years of their college career (Santa Fe is a

junior college) was based on evidence of a male/female dif-

ference in longitudinal changes in identity integration over

the four years of college. Specifically, Constantinople

found that women scored higher than men on identity achieve-

ment and significantly lower on identity diffusion when they

entered college, but the men made greater gains in identity

consolidation during their college experience. By the senior

year females scored higher on identity diffusion than the

males, though not significantly. Likewise, Sanford's findings

were that there was a greater degree of identity problems in

senior women than in freshmen women, while problems in iden-

tity for men were greater earlier in college. It may be

possible that our male and female subjects will undergo the

same reversal in positions on identity achievement by the

time they reach the end of the educational/training phase of








their lives.

A second explanation is that there are differences

between Santa Fe students and students attending other

colleges in terms of their sex role identities and consequent

overall identity integration. There is evidence for such

differences in our data. Although their scores on the BSRI

indicate that Santa Fe (SF) males see themselves as more

masculine sex-typed than the normative males, we would guess

that male students enrolled in a junior college are generally

lower than those accepted at four year colleges on one index

of how society evaluates manhood: success at attaining

academic/vocational goals. This possible discrepancy between

perceptions of themselves and their actualization of important

masculine ideals may interfere with and result in lower

identity integration for these males as compared to other

college males. At the same time, the females in our sample

score substantially higher on androgyny than the normative

college population, indicating a perception of themselves

as possessing both stereotypical masculine and feminine traits.

These females combine identification with society's approved

way of being female feminine traits and identification

with masculine standards of independent achievement, autonomy,

etc. Our literature review showed that these latter traits,

lacking in the traditional feminine role, are vital to the

development of high ego strength and identity achievement.

Thus, these joint identifications in the androgynous females

may have facilitated their identity achievement and raised








it above the level attained by the more feminine sex-typed,

normative college students. Indeed, Heilbrun found that

female college students who combined masculine goal orienta-

tion and feminine interpersonal sensitivity score higher than

all other females on behavioral effectiveness in college

(1965, 1968, 1970).

With this last explanation we would have expected a

high correlation between our measures of androgyny and

identity: There was virtually none. A final explanation

is that Schenkel & Marcia (1972) were right in their inter-

pretation of a decrease in anxiety and an increase in self-

esteem in the identity achievement group over the three year

interval between two related studies on women. They posited

that women in college right now, influenced by the women's

movement, may be experiencing a sanction for making inde-

pendent decisions, finding and asserting themselves, in

short, a push towards identity integration. On the other

hand we would add, college males are falling behind in iden-

tity achievement due to the confusion they are experiencing

in their own role definition stemming from dramatic changes

in the espoused standards of the complementary female role.

At present the embryonic Men's Movement lacks the recognition

necessary to serve as a vehicle for clarifying and resolving

sex role diffusion.

Our last two explanations for the female advantage in

identity achievement have in common the assumption that a

higher than normative frequency of an androgynous sex role








among our women, or at least sanction by the Women's Move-

ment for actualization of masculine traits, is responsible.

This premise is contradicted by our data on two counts. As

mentioned above, our measures of androgyny and identity have

practically a zero correlation. Furthermore, if the Libera-

tion Movement is facilitating the occurrence of masculine

or androgynous traits in our female subject's self-percep-

tions, we would expect that this change in sex role norms

would be reflected in the girls' perceptions of their best

friends. On the contrary, a casual perusal by the author of

the ACL ratings of parents and peers revealed that both male

and female subjects in our study did not perceive their

peers as any less stereotyped than their parents. Moreover,

the percentage of both parents and peers rated as possessing

a mixture of masculine and feminine traits on the ACL was far

below what we'd expect by chance, indicating that sex role

stereotyping is still determining our subjects' perception

of peers as well as parents.1



1These statements are based on a comparison of the
distributions of subjects' ratings of their same-sex parent
and peer, in line with the traditional masculine or feminine
stereotypes on the nine sex-typed scales of the ACL. As the
ACL measure is set up so that higher scores for male parent
or peer represent possession of many masculine traits and
few feminine traits, and low scores represent possession of
few masculine traits and many feminine traits (vice versa
for female parents and peers), the middle-most scores of
four or five represent equal representation of both masculine
and feminine traits, or androgynous sex roles. For both male
and female subjects, we found that the number of peers scored
in the middle-most range of the continuum was exactly the
same as the number of same-sex parents scored in that range.
Most importantly, as these distributions meet the criteria








Unfortunately, we can't use this important evidence

of differences in the frequency of androgynous versus

stereotypical roles between the female subjects group and

both parent and peer groups to help us unravel the relation

between sex role development and identity. The distribution

of subjects' sex role self-concepts was arrived at by a very

different measure (BSRI) than the distributions of both

parent and peer sex roles (ACL), and there is no correlational

data on the two. As seen with hindsight, we wish we had

given the BSRI to all three groups so we could validly com-

pare differences in the distributions. However, at the time

of formulating the study we chose the ACL measure for parent

and peer sex roles in order to correspond to measures used

by Heilbrun in his studies of this area. Future research in

this area using the BSRI for all three groups would not only

more validly test our predicted relations between parent and

peer sex role norms and the subject's own sex role self-

concept but could scientifically evaluate the very popular

question: To what degree and in what directions are male

and female sex role norms changing? An answer to this last




for a bi-nomial distribution, we decided to compare the
percentages of male and female parents and peers scored
within the androgynous range on the ACL with the expected
probability of the middle two classes of the bi-nomial.
While only 25% of each of the four sample distributions were
scored within the androgynous classes, if chance alone were
operating the expected percentage would be 49.









question might facilitate identity formation for many

adolescents. It would provide the growing youth with an

ideal of how to be a man or woman that grounds him in the

realities of his present society and might simultaneously

lighten his struggle to find himself by clearing out rigid

stereotypical expectancies.













APPENDIX I
INCOMPLETE SENTENCES BLANK

EGO IDENTITY

Name Age Class

Marital Status Date

Complete these sentences to express your real feelings.

Try to do every one. Be sure to make a complete sentence.



1. For me success would be



2. The difference between me as I am and as I'd like to be



3. When I consider my goals in the light of my family's

goals

4. I'm at my best when



5. To change my mind about my feelings toward pre-marital

intercourse

6. Sticking to one occupational choice



7. When I let myself go I



8. I chose to come to this college after



9. If someone were to ask me who I am, I would say








I am really convinced that



Whether I sleep with someone depends on



When I was a child, I

whereas, now I

I know that I can always depend on



(Choose only one) I am

I am not

It seems as though I always



I wish I could make up my mind about



What happens to me depends on



When I consider my sexual standards and behavior in the

light of my family's

As compared with four years ago, I


I belong to


To change my mind about feelings toward religion



If one commits oneself




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs