Citation
The characteristics of masculinity, femininity, androgyny, age, and academic ability

Material Information

Title:
The characteristics of masculinity, femininity, androgyny, age, and academic ability as factors in the retention of women students in college programs in business administration
Creator:
Wilson, Carol Popejoy, 1941-
Publication Date:
Language:
English
Physical Description:
viii, 95 leaves : ; 28 cm.

Subjects

Subjects / Keywords:
Business administration ( jstor )
College students ( jstor )
College transfer students ( jstor )
Colleges ( jstor )
Femininity ( jstor )
Inventories ( jstor )
Masculinity ( jstor )
Social discrimination ( jstor )
Women ( jstor )
Womens studies ( jstor )
Business education -- Florida ( lcsh )
Characters and characteristics ( lcsh )
Curriculum and Instruction thesis Ph. D
Dissertations, Academic -- Curriculum and Instruction -- UF
Women college students -- Florida ( lcsh )
Greater Orlando ( local )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1981.
Bibliography:
Bibliography: leaves 92-94.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Carol Popejoy Wilson.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
028134248 ( ALEPH )
07916452 ( OCLC )

Downloads

This item has the following downloads:


Full Text












THE CHARACTERISTICS OF MASCULINITY,
FEMININITY, ANDROGYNY, AGE, AND ACADEMIC ABILITY AS FACTORS IN THE RETENTION OF WOMEN STUDENTS IN COLLEGE PROGRAMS IN BUSINESS ADMINISTRATION







BY

CAROL POPEJOY WILSON


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


1981

























Copyright 1981



by



Carol Popejoy Wilson













ACKNOWLEDGMENTS

This study Is dedicated to my husband, Skip, and my son, Jay, for their patience, encouragement, and understanding; and to my parents, Ira and Dorothy Popejoy, for their constant support and concern.

The writer expresses her appreciation to Dr. Glenna Carr, chairman

of the supervisory committee, for her guidance, support, and encouragement. The writer also expresses her appreciation to the other committee members: Dr. Eugene Todd, Dr. Fred Goddard, Dr. Arthur Lewis, and Dr. William Drummond.

The writer is especially grateful to her colleagues at the University of Central Florida: Dr. Charles Dzuiban, Mr. Tom Peeples, Dr. Gordon McAleer, Dr. Bill Brown, Dr. LeVester Tubbs, Dr. Sandra Guest, Dr. Ralph Gunter, and Ms. Phyllis Smith, who assisted the writer in the study.








TABLE OF CONTENTS


PAGE

ACKNOWLEDGEMENTS ................. I11

LIST OF TABLES ............................................ v

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

CHAPTER

ONE INTRODUCTION .................................. 1

Need for the Study 9 Statement of the Problem ...................... 10
Definition of Terms ........................... 10
De IIm Ita tlIons . . . . ...0 00.........................e 0 12
Description of the Instruments ................ 12
Demographic Profile ........................... 15
Statement of Hypotheses ....................... 16

TWO REVIEW OF THE RESEARCH ........................... 17

Conflicts from Sex-Role Stereotyping .......... 17
Evolution of Sex Roles ........................ 19
Sex Roles of Women In Business ................ 20
Sex Roles of Women in Education ............... 24

THREE PROCEDURES USED IN CONDUCTING THE STUDY .......... 27 FOUR ANALYSIS OF DATA ................................. 31

Analysis of Descriptive Data .................. 33
Analysis of Open-ended Question ............... 42
Descriptive Analysis of Bem Sex-Role Inventory. 46 Discriminant Analysis Findings ................ 53
Classification of Cases ....................... 64
Test of Hypotheses and Results ................ 65

FIVE CONCLUSIONS AND RECOMMENDATIONS ................... 70

APPENDICES

A DEMOGRAPHIC PROFILE DATA SHEET ........................ 76

B LETTER TO GROUPS I AND II ............................ 79

C INSTRUCTIONS FOR THE BEM SEX-ROLE INVENTORY AND
THE BEM SEX-ROLE INVENTORY ......................... 81
D ADMINISTRATION AND SCORING GUIDE FOR THE BEM
SEX-ROLE INVENTORY ................. ......... 84













LIST OF TABLES


Table Page

1. Sex of Students in Groups I, II, III and IV ............. 35

2. Ages of Students in Groups I, II, III and IV ............ 36

3. Harital Status of Students In Groups I, II,
III, and IV ............................................. 37

4. Employment Status of Students In Groups I,
II, III, and IV ........................................ 38

5. Grade Point Averages for Students in
Groups I, II, III, and IV ............................... 40

6. Students Classifications in Groups I, II, III and IV ................................ .......... 41

7. Scholastic Aptitude Test Scores for Students in Groups I and II ..................................... 43

8. Number of Students in Groups I, II, III, and IV with Children ........................................ 44

9. Year of High School Graduation for Students In Groups 1, II, III, and IV ............................... 45

10. Responses by Group !1 Students to the Four Categories Derived from the Open-ended
Question on the Demographic Profile Data
Sheet . 47

11. Interpretation of Categories Derived from
the Bem Sex-Role Inventory .............................. 50

12. Comparison of Undifferentiated, Masculine, Androgynous
and Feminine Bem Sex-Role Inventory Classifications
by Group ................................................ 51









List of Tables-continued


Table

13. Standardized DiscrImInant Function Coefficients At the .05 Level of Significance Presented In Order of
Importance to the Discriminant Function and the
Levels of Measurement Used In the Assignment of Values
to the Discriminating Variables ......................... 56

14. Means and Standard Deviations for Discriminating
Variables at .05 Level of Probability for Groups
I and II ................................................ 58

15. Means and Standard Deviations for the Bem SexRole Inventory Masculine, Feminine and
Difference Standard Scores for Groups I and II .......... 59

16. Classification of Subjects In Groups I and II Utilizing Nine Discriminating Variables ................. 61














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

THE CHARACTERISTICS OF MASCULINITY,
FEMININITY, ANDROGYNY, AGE, AND ACADEMIC ABILITY
AS FACTORS IN THE RETENTION OF WOMEN STUDENTS
IN COLLEGE PROGRAMS IN BUSINESS ADMINISTRATION

By

Carol Popejoy Wilson

August 1981

Chairman: Dr. Glenna Carr Major Department: Curriculum and Instruction

The purpose of this study was to analyze the effects of age, academic ability, masculinity, femininity, and androgyny on the retention of women students in a program of business administration. One hundred and fifty-nine students enrolled at the University of Central Florida during the spring quarter of 1979 were Included in the study. There were four groups; the two major groups were Group I women students enrolled In the College of Business Administration, and Group II, women students who had transferred out of the College of Business Administration. Group III, women education majors, and Group IV, men business acninistration majors, were used as comparison groups to see If their masculine, feminine and androgyny scores on the Bem Sex-Role Inventory would differ greatly from those In Groups I and II.









The Bem Sex-Role Inventory was used to measure students' masculine, feminine, and androgyny scores. A demographic profile data sheet and university records were used to gather additional data. All data for the four groups were compared through cross tabulation, and data on subjects In Group I and II were analyzed and classified through a discriminant analysis.

A discriminant function composed of nine discriminating variables at the .05 level of significance resulted from the analysis. The variables are, In the order of their Importance, (1) student classification, (2) marital status, (3) feminine standard score, (4) Scholastic Aptitude Test score, (5) difference standard score, (6) employment status, (7) age,

(8) grade point average, and (9) masculine standard score.

The major conclusion was that given data on the nine variables listed above for a new woman student who has selected business administration as a major, a prediction concerning her staying in the business administration curriculum can be made with a 69.7 percent accuracy rate, and a prediction of her transferring out of business administration can be made with a 62.5 percent chance of accuracy.


viii














CHAPTER I
INTRODUCTION



The governing clause of Title IX of the 1972 Education Amendments to the Civil Rights Act of 1964 sets the stage for eliminating sex discrimination. The law, which was originally introduced In 1971 as an amendment to the Civil Rights Act of 1964, presents its purpose in Its opening statement as follows:

No person In the United States shall, on the basis of sex, be excluded from participation In, be denied the
benefits of, or be subjected to discrimination under
any education program or activity receiving Federal
financial assistance... (Taylor and SchavIlk, 1975,
p. 5.)
No one can be excluded from participating In any academic, occupational training, or other educational program within institutions using Federal funds. "The practices that are specifically prohibited by Title IX as well as other practices that maintain stereotypic sex roles should be examined" (Taylor and Schavlik, 1975, p. 15). Further, "Any special services that are provided to help students overcome the effects of previous sex-role socialization should be described and evaluated by those responsible for them" (Taylor and Schavilk, 1975, p. 28). Examples are counselling services offered by professional counselors, clinical and consulting psychologists, psychiatrists, academic counselors or advisers, residence hall counselors, and admissions counselors.









"All who assist students in making educational and career choices are likely to influence, consciously or unconsciously, the decisions of some students" (Taylor and Schavlik, 1975, p. 27).

Without a doubt, the thrust of Title IX should provide an Incentive and push institutions of higher education in that direction of eliminating Inequality which Is based on sex-role stereotyping. What took the American society so long to recognize the stunting effects sex stereotyping has had on women? According to Kate Mueller, women had to wait until the 1970's for the women's liberation movement to awaken the public to women's problems and needs. This wait for women to begin to ask for equal rights, in Mueller's opinion, can be traced to several factors.

1. In the 1950's many of the women indicated they were not

too unhappy with their own situations.

2. A women's movement would mean full utilization of birth

control, including the birth control pill, so that maternal

responsibilities could be carefully planned. It took

awhile for the birth control pill to be accepted.

3. It wasn't until the later 1960's that women's groups

learned how to organize, and how to apply verbal protests.

4. New ways of looking at love and new attitudes and behavior

about sex relations have slowly developed during the last forty years. Along with these new attitudes and behaviors

came planned parenthood and lower birthrates.









5. As late as the 1950's and 1960's, women were still

suffering under Freud's description of women as being

Inferior.

6. During the 1950's and 1960's there was the "utter

perfidy" of the mass media toward homemakers.

7. Passage of the Equal Rights Amendment, which was written

and introduced in Congress in the early 1920's, was not

strongly supported by any of the organized women's groups

until the late 1960's and early 1970's (Mueller, 1977,

pp. 43-45).

The focus of this study is to determine If the characteristics of masculinity, femininity, androgyny, age and academic ability of women students are significant influences on the retention of women students In business administration programs at the post secondary level. Is there a correlation between these factors and the retention of women students In business administration, which is considered to be a stereotypically nontraditional field of study for women?

A stereotype is defined as an "unvarying form or pattern, having no individuality, as though cast from a mold" (Webster, 1971, p. 860). When stereotype is applied to sex roles, the definition would indicate all women are alike In certain respects, Just as all men are alike and they both possess a certain degree of fixed or conventional expressions, notions, character and mental patterns.










The roots of sex-role stereotyping are deep In our culture.

Cross points out how effectively de Tocqueville briefly and concisely described male/female stereotypes of nearly 150 years ago.

In no country has such constant care been taken as in America to trace two clearly distinct lines of action for the two sexes and to make them keep pace one with
the other, but in two pathways that are always different.
Women took no part in business or in politics; they never
managed the outward concerns of the family. With rare
exceptions they were confined to the quiet circle of
domestic employment. (Cross, 1965, p. vii)

Cross offered a possible explanation of sex roles when she

said,

Americans had simply applied to the sexes the great
principle of political economy which governs the
manufacturers of our age by carefully dividing the duties of man from those of women in order that the
great work of society may be better carried on.
(Cross, 1965, p. vili)

Some Improvements have been made In the betterment of women's

role In society since 1835; however, in the evolution and maturation

of the Individual, sex-role stereotyping continues to be reinforced.

For example, studies cited by Tibbets (1976) show problems connected

with sex-role stereotyping:

Stereotypically masculine traits, such as Independence, dominance, competition, achievement, drive, leadership,
decisiveness and logic are considered more socially
desirable than are stereotypically feminine attributes, such as dependence, emotionalism, submissiveness, passivity, Indecisiveness, and lack of logic . . . both men
and women agree that for an adult, regardless of sex,
stereotyped male characteristics were "healthier" than
were stereotyped female characteristics. (Tibbets, 1976,
p. 179)

The control that sex-role stereotyping Injects into the work

environment and its effects on the degree of success of women In









work was Investigated by the National Commission on the Observance of International Women's Year In 1976. The Commission specifically addressed four areas. "The Commission examined women entering the skilled crafts and blue collar Jobs; secretaries; women who own their own businesses; and new ways In which pay rates are being evaluated" (National Commission on the Observance of International Women's Year, 1976, p. 58). The Commission undertook the following:

They looked into practices In the construction industry
to determine how nondiscrimination laws are applied.
Construction work has been a white male preserve for so
long that a "macho" folklore has grown up around it. But
modern technology has made nearly all construction jobs
easier and has brought many of them within the capability of many women. (National Commission on the Observance of
International Women's Year, 1976, p. 58)

The Commission further Investigated problems that women who want to go Into business on their own encounter. The investigation revealed enormous problems for these women. Arthur S. Flemming, Chairman of the U. S. Commission on Civil Rights, referred to these

problems as "staggering." In a 1975 report he wrote:

Our Investigations reveal that minority and female-owned
firms encounter problems of staggering proportions In
obtaining Information on federal, state and local government contracting opportunities In time to submit bids,
and In obtaining working capital necessary for effective
marketing and bidding. . .
. Minority and female entrepreneurs also encounter a
great deal of skepticism regarding their ability to perform adequately on government contracts. (Flemming,
1975, p. 1)

More and more women are entering the labor force. According to the National Commission on the Observance of International Women's Year (1976), there Is a need for opportunities for women to enter nontraditional fields of work . . . that is, those fields which have









traditionally been open to males only. The steady Increase In the number of women entering the labor force Is reflected in the percentage Increase during the last four years. Women represented 40.5 percent of the civilian labor force in 1976. This percentage increased to 41.7 percent In 1978, and was up to 41.9 percent through Hay, 1979. (U. S. Bureau of the Census, 1979, P. 392).

Within the one million divorces which occurred in 1975, only 14 percent of the women Involved were awarded alimony. Generally, child support payments are less than half enough to support the children. It is apparent that more women needed to work In 1979 than did In 1976. Limited availability of employment for women, because of their being confined to the few "traditional" fields of work for women such as teaching, nursing and secretarial work, has been exacerbated by sex-role stereotyping In all phases of American life (National Commission on the Observance of International Women's Year, 1976, p. 57).

Students of all ages are not shielded from the effects of sexrole stereotyping. In fact they are often overtly subjected to such stereotyping all through school and during their formative stages In life. Ms. Verheyden-Hilliard painted the following picture at recess time on the playing fields at an elementary school.

"John says: 'Get away, we don't want any girls playing.'

Mary says: 'They won't let me play with them.'

Educator 'Well . . . that's the way boys are. They
says: are too rough, anyway. Play with the
girls." (Verheyden-Hilliard, 1975, P. 151)









If boys are supported and Indeed encouraged in this behavior for all of their formative years (boys will be
boys), there Is no way to expect them to suddenly turn about as adults and see female peers as equals. Women
have remained, and will remain, the little girls that
society allowed and encouraged little boys to shove aside
so that they could play with the boys without Interference
i . . and they're still doing it when they're 45.
(Verheyden-Hilliard, 1975, p. 151)

According to Ms. Verheyden-Hilliard, "The message conveyed to

girls Is that boys have a right to exclude them and that they are too weak to compete anyway. The girls retreat to the girls games and the female careers where they have been forcefully told they belong" (Verheyden-Hilliard, 1975, p. 152).

Girls and women apparently lack strong success models to follow. Universities are more lacking In such role models than are elementary schools. The closer to the top of the academic ladder, the greater the preponderance of males. Statistics for secondary schools shbw this trend.

in the eleven-year periods between 1950-51 and
1960-61 the number of women serving as junior high and senior high school principals dropped from 18% to 3.8%
for all secondary schools . . . for the same eleven year
periods the percentages of women elementary principals
decreased by nearly 20%. (Van Heir, 1975, p. 163)

More recent statistics cited by the National Center for Education Statistics revealed the State of Florida as having 23 percent women in principal and assistant principalships (National Center for Education Statistics, 1978, p. 19).

It is recognized that women who enter Institutions of higher education today are more likely to choose careers traditionally dominated by men than were women ten years ago. However, differences in choice of career between men and women have not been








substantially reduced. The percentage of women who entered the fields of physical sciences, biological sciences, business, and engineering increased at the same rate as that of men from 1961 to 1971. The percentage of women choosing these professions Increased from 21 percent to 26 percent. The percentage of men choosing these professlons Increased from 56 to 61 during the same period of time (Peng and Jaffee, 1979, p. 286). The career fields generally selected by women continue to be those traditionally dominated by women, such as education and nursing.

Why does the problem of sex differences In such areas as career choice persist? There is general agreement among writers and educators that women continue to limit their own choices and substantially contribute to their own second-rate status in society. According to Sylvia-Lee Tibbets,

Women choose to be Inferior--not because they are, nor
necessarily because they want to be--but because: (a) they have been taught to believe they are or should be Inferior;
(b) they are afraid to appear unfeminine; or (c) they
are not fully aware of their situation and do not realize
that they are being treated as second class citizens or
that they do have a legitimate complaint. (Tibbetts,
1975, p. 178)

Rather than holding women entirely responsible for their position in society, It would be more accurate to describe women as "victims" of a society steeped In sex-role stereotyping. When individuals do not realize that a situation exists, there is nothing they can do to Improve it; however, once a woman becomes aware of the sexist nature of our society, she becomes responsible for the Improvement of her own position In society. For example, her recognition of the possibility for people, and especially for her, to be both masculine and feminine









would enable her to express anger, assert her preferences, trust her own Judgment, and take control of her life situations, whatever they might be. According to Dr. Sandra Bem,

For fully effective and healthy human functioning, both masculinity and femininity must each be tempered by the
other, and the two must be Integrated into a more balanced,
a more fully human, a truly androgynous personality. An
androgynous personality would thus represent the very
best of what masculinity and femininity have each come to
represent, and the more negative exaggerations of masculinity and femininity would tend to be cancelled out.
(Bem, 1975, p. 4)

Need for the Study

If sex-role stereotyping is detrimental to the successful completion of academic programs and the attainment of educational goals established by women students, an awareness of the problem by educators is the first step in helping to eliminate the problem. Today's women need to be made aware of the expanding choices of academic programs and Job opportunities available to them. A major responsibility of the educator Is to assist students in their efforts to expand their professional horizons, exercise their full potential In their chosen fields of study and to take advantage of their freedom to make choices without being hindered by the detrimental effects of sex-role stereotyping by the Individual and society.

It is Important for educators and society at large to challenge

the myths that have kept women In their "place." Daughters, as well as sons, have to believe that they can achieve their dreams and perform to the level of their abilities; consequently, "We must encourage








the growing recognition by women and men that equal rights Is a matter of simple Justice" (National Commission on the Observance of International Women's Year, 1976, p. v).

Women students who chose business as their profession, which Is still a nontraditional field of study for women today, should be free to strive for top level positions within the profession. Women students should be freed from the direct and subtle effects of sex-role stereotyping In their efforts to complete their academic programs In such nontraditional fields of study as business administration.


Statement of the Problem

It was the purpose of this study to determine whether the characteristics of masculinity, femininity, androgyny, age, and academic ability are factors which correlate with the retention of women students In programs of study in the College of Business Administration at the University of Central Florida.


Definition of Terms

The following definitions apply to the terms as they are used in this study.

Stereotype: An unvarying form or pattern, having no Individuality, as though cast from a mold.

Sex-role stereotypes: Socially designated behaviors that differentiate between men and women; women commonly possess a common pattern of conventional expression, notion, character, or mental attitude, Just as men do, but the attitudes and patterns differ greatly.






I1

Bern Sex-Role Inventory: An Inventory consisting of sixty personality characteristics, twenty of which are Judged to be socially desirable traits for men, twenty of which are Judged to be socially desirable traits for women and twenty characteristics which serve as filler items.

Masculine characteristics: A set of characteristics on the Ben SexRole Inventory judged to be more desirable in American society for a man than for a woman.

Feminine characteristics: A set of characteristics on the Bem Sex-Role Inventory Judged to be more desirable In American society for a woman than for a man.

Androgyny: A score on the Bern Sex-Role Inventory which indicates that the Individual Is not strongly self-identified with either masculine or feminine characteristics. Group 1: Those women students enrolled in the College of Business Administration at the University of Central Florida during the Spring

Quarter, 1980, who have completed ninety (90) quarter hours of college study including at least two (2) courses In business administration. Group I1: Those women students enrolled at the University of Central Florida during the Spring Quarter, 1980, who have taken at least two

(2) courses In business administration, and who have transferred out of the College of Business Administration at some point between January, 1979, and June, 1980.

Age: The actual age of women students In Groups I and II as stated on their demographic data sheets.








Academic Ability: Scores on the Scholastic Apptltude Test, American College Test or the Florida Twelfth Grade Test which have been converted to a common score utilizing the 1977 State University System of Florida Equivalency Document.


Delimitations

All of the subjects In this study were enrolled In the College of Business Administration at the University of Central Florida between January, 1979, and June, 1980.

There was no attempt to make inferences about the entire female student population In the State University System of Florida or the entire student population at the University of Central Florida. Inferences were confined to the stated population Included In the study.

Description of the Instruments

The Bem Sex-Role Inventory is described by Dr. Sandra Lipsitz Bem as follows:

Both In psychology and In society at large, femininity
and masculinity have long been conceptualized as opposite
ends of a single bipolar dimension. More recently, however, scholars In a number of disciplines have begun to
concern themselves with the concept of psychological
androgyny, a term that denotes the Integration of femininity and masculinity within a single Individual. The
concept of psychological androgyny Implies that it Is
possible for an Individual to be both compassionate and
assertive, both expressive and instrumental, both feminine and masculine, depending upon the situational appropriateness of these various modalities; and It further Implies
that an Individual may even blend these complimentary
modalities In a single act, being able, for example, to
fire an employee If the circumstances warrant it but with
sensitivity for the human emotion that such an act Inevitably produces.









Before empirical research on the concept of psychological androgyny could be Initiated, however, It was necessary to develop a new type of sex-role Inventory, one that would not automatically build In an Inverse relationship between femininity and masculinity as previous Inventories had done. On most Inventories, Items are empirlcally defined as feminine or masculine on the basis of differential endorsement by females and males, and a person filling out the Inventory is said to be either feminine or masculine as a function of which sex she or he most closely resembles. Although It is possible for a person to earn a score that falls halfway between the two extremes and thereby reveal that she or he does not closely resemble either sex, a person cannot make the strong statement on such an inventory that she or he is androgynous.

The Bem Sex-Role Inventory (BSRI) was designed to Implement empirical research on psychological androgyny. It contains 60 personality characteristics, 20 of which are stereotypically feminine (e.g., affectionate, gentle, understanding, sensitive to the needs of others), and 20 of which are stereotypically masculine (e.g., ambitious, selfreliant, Independent, assertive). The BSRI also contains 20 characteristics that serve as filler Items (e.g., truthful, happy, conceited). When taking the BSRI, a person is asked to Indicate on a 7-point scale how well each of the 60 characteristics describes herself or himself. The scale
ranges from I ("Never or almost never true") to 7 ("Always or almost always true") and is labeled at each point.

The BSRi has two features distinguishing It from most masculinity-femininity scales. The BSRI treats femininity and masculinity as two independent dimensions rather than as two ends of a single dimension, thereby enabling a person to indicate whether she or he is high on both dimensions ("androgynous"), low on both dimensions ("undifferentiated") or, high on one dimension but low on the other (either
"feminine" or "masculine").

In addition, the BSRI is based on a conception of the traditionally sex-typed person as someone who is highly attuned to cultural definitions of self-appropriate behavior and who uses such definitions as the Ideal standard against which her or his own behavior Is to be evaluated. In this view, the traditionally sex-typed person Is motivated to keep her or his behavior consistent with an Idealized image of femininity or masculinity, a goal she or he presumably accomplishes both by selecting behaviors and attributes that enhance the Image and by avoiding behaviors and attributes that violate the Image. Accordingly, Items








were selected as feminine or masculine on the basis of
cultural definitions of sex-typed social desirability and
not on the basis of differential endorsement by females and
males, i.e., a characteristic qualified as feminine if it was judged to be traditionally more desirable In American
society for a woman than for a man, and it qualified as
masculine if It was judged to be more desirable In American
society for a man than for a woman.

The BSRI is essentially self-administering and may be
given to large groups as well as to Individuals. It has also been administered by mail in several studies. The inventory consists of 60 adjectives and phrases printed
on a single sheet with Instructions and space for personal information about the subject on the reverse side. (Bem,
1979, pp. 1-3)

The Bem Sex-Role Inventory Initially was acministered to 444 male and 279 female students in introductory psychology at Stanford University during the Winter and Spring of 1973. It was also administered to an additional 117 male and 77 female students at Foothill Junior College. The data that these students provided represent the normative data for the Bem Sex-Role Inventory.

Subjects were classified as sex typed, whether masculine or

femininine, if the androgyny t ratio reached statistical significance (/t/ > 2.025, df - 38, p < .05), and they are classified as androgynous If the absolute value of the t ratio is less than or equal to one (Bem, 1974, p. 161).

Dr. Bem estimated the internal consistency of the Bem Sex-Role Inventory by computing the coefficient alpha separately for the masculinity, femininity, and social desirability scores of the subjects in each of the two normative samples. The results Indicated all three scores to be highly reliable, both In the Stanford sample '1masculinity a - .86; femininity a - .80; social desirability a.70" (Bem, 1974, p. 159).









Because the reliability of the Androgyny t ratio could not be calculated directly, coefficient alpha was computed for
the highly correlated Androgyny difference score, femininity
masculinity, using the formula provided by Nunnally (1967) for linear combinations. The reliability of the androgyny
difference score was .85 for the Stanford sample and .86
for the Foothill sample. (Bem, 1974, p. 159)

In an effort to determine test-retest reliability, the Bem

Sex-Role Inventory was administered for a second time to 28 males and 28 females from the Stanford normative sample. The second administration of the Inventory took place approximately four weeks after the first administration.

During this second administration, subjects were told that
we were interested In how their responses on the test might vary over time, and they were explicitly Instructed not to
try to remember how they had responded previously. Productmoment correlations were computed between the first and second administrations for the Masculinity, Femininity,
Androgyny, and Social Desirability scores. All four scores
proved to be highly reliable over the four-week Interval
(Masculinity r - 90; Femininity r - .90; Androgyny r - 93;
Social Desirability r - .89). (Bem, 1974, p. 160) Demographic Profile

The demographic profile data sheet was designed to elicit pertinent background information from the subjects. Most of the Information requested on the demographic profile data sheet also was collected by the Investigator from the student's permanent file and University data tapes. Collecting the same Information from two sources was Intended to provide a cross check on the data.

The following Information was available only from the demographic profile data sheet: Address for Summer Quarter; marital status, employment outside the home; occupation; and number of children.









The final question, "If you were enrolled in the College of Business Administration after January 1, 1979, and subsequently have transferred to another college at U.C.F., please Indicate in the space provided below your reasons for the transfer," provided the investigator the opportunity to analyze the various reasons for the transfer.

Statement of Hypotheses

The major hypothesis of this study was that an individual's membership In either Group 1 or Group II can be predicted on the basis of age, academic ability, and the Bem Sex-Role Inventory characteristics. The following sub-hypotheses In the null form were tested.

1. The masculinity score derived from the Bern Sex-Role Inventory

is a variable which does not discriminate when used in the

prediction of group membership in either Groups I or II.

2. The femininity score derived from the Bern Sex-Role Inventory

Is a variable which does not discriminate when used In the

prediction of group membership in either Groups I or II.

3. The androgyny score derived from the Bem Sex-Role Inventory

is a variable which does not discriminate when used In the

prediction of group membership in either Groups I or II.

4. The age of the Individual is a variable which does not discriminate when used in the prediction of group membership In

either Groups I or II.

5. The individual's academic ability is a variable which does

not discriminate when used In the prediction of group membership in either Groups I or II.













CHAPTER Ii
REVIEW OF THE RESEARCH



The purpose of this review was to study sex-role stereotyping as it related to women In business and education. Research reviewed concerned the questions, what Is sex-role stereotyping, how has It evolved, does It exist today, and If so, how does It affect student performance.


Conflicts from Sex-Role Stereotyping

The conflicts which result from some sex-role stereotyping

have been studied by many researchers. One of the most noted studies Is Matina Horner's "Toward an Understanding of Achievement Related Conflicts In Women" (Homer, 1972). Horner tested freshman and sophomore students, male and female, at a large mid-western university. These students responded to a cue added to the end of the Standard Thematic Apperceptive Test (TAT). The cue Is "After first term finals, Anne finds herself at the top of the medical school class." Horner concluded that approximately "65% of the girls were disconcerted, troubled, or confused by the cue" (Horner, 1972, p. 162). She further concluded that a fear of success Imagery dominates responses from women students and is relatively absent in male responses, and that conflicts emerge between role expectations and aspirations due to









previously Imposed stereotypes (Homer, 1972, p. 162). The expectancy that success In achievement related situations will be followed by negative consequences arouses fear of success In otherwise achievement oriented women which then Inhibits their performance and levels of aspiration.

In an earlier study, Homer hypothesized that the motive to avoid success is significantly greater In women than It is in men (Horner, 1969). She further hypothesized that this tendency Is even greater in high achievement oriented, high ability women who are capable of success than It is in women at the other extreme; that Is, lower achievement oriented or lower ability women who cannot and will not seek to achieve. Homer's research was based on 178 students who were administered the Standard Thematic Apperceptive Test. Her findings suggest that most women will exercise their full potential only In a non-competitive situation, especially if they are competing with men. This tendency Is stronger among women with an Intense anxiety about success. One of the most significant findings of this research Is the complexity of achievement motivation In women. Her results Indicated that fear of success was found to be aroused In situations in which there was concern over or anxiety about competitiveness and Its aggressive overtones. When these anxieties occurred, women experienced conflict and adjusted their behaviors to their Internalized sex-role stereotypes.

The effects of the various traditional roles performed by women, such as wife and mother, have been studied. One researcher examined









the question of whether women who have experienced these traditional roles experienced less conflict about achievement (Tomlison-Keasey, 1974). Subjects In the Tomllson-Keasey study were women students from Douglass College and University College at Rutgers University whose enrolled women students differed In terms of age, marital status and number of children. A young unmarried sample was drawn from Douglass College and an older married sample was drawn from the University College. The procedure followed Homer's study by using a cue response, and the results indicated a significantly different level of fear of success. In the study it was hypothesized that women who were performing as mothers and wives, which were perceived as appropriate achievements, experienced less fear of success and reduced anxiety than did those women who were not performing these "appropriate" roles. The results were as follows:

� . women who had satisfied the mother and the wife roles
are not as likely to evidence fear of success. unmarried
subjects had significantly more fear of success . . . then did
married subjects . . . and women with no children had significantly more fear of success . . . than did women with children.
(Tomlinson-Keasey, 1974, p. 235)

Evolution of Sex Roles

The changing of sex roles has been slow moving. Myth and fact are intertwined. Some very significant "ole wives tales," however, are laid to rest (Tangri, 1972). For example, data Indicate that women who seek to enter predominantly male-dominated professions do not show evidence of having Identified with their fathers over their mothers. Daughters who have more educated working mothers in maledominated occupations will view their mothers as role models. Tangri










also found that women in nontraditional fields or occupations do not reject the traditional female roles of wife and mother. They do plan, however, to postpone marriage and have fewer children than do the more traditional women. Also they do not view themselves as masculine women (Tangri, 1972, p. 196).

Studies reveal findings that vividly point out the existence In our contemporary society of "clearly defined sex-role stereotypes for men and women contrary to the phenomenon of unisex roles currently touted In the media" (Broverman, 1972, p. 75). In his examination of the relationship between mothers' employment status and sex-role perceptions of college students, daughters of employed mothers perceived fewer differences between men and women than did daughters of homemaker mothers. Also, "since more feminine traits are negatively valued than are masculine traits, women tend to have more negative self-concepts than do men" (Broverman, 1972, p. 750). The double standard emerges. According to Broverman:

If women adopt the behaviors specified as desirable 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 standards of adult behavior.
(Broverman, 1972, p. 75)


Sex Roles of Women In Business

Few significant studies in the area of women in business have

been found. There is an abundance of opinion and testimony, however, on the topic. For example, an article appeared in Business Week Magazine which stressed the barriers are still up for women at the









top executive levels. There was noted, however, a new trend for women graduates who are getting job offers that were closed to those women who graduated just a few years before them. In the opinion of Ms. Grim, the first woman vice-president of Carson, Price, Scott and Co., "A woman who Is determined to play In a game so fixed, had better be prepared to look like a girl, act like a lady, think like a man and work like a dog" (For Women, a Difficult Climb to the Top, 1969, p. 45).

Fretz and Hayman conducted a survey with twenty business organizations, ten of which are industrial companies from among the first 100 companies on the Fortune "500" list. The organizations that participated In the survey employed nearly two million people. The sample organizations Included women at all levels of employment. Thirty six percent of the persons employed In the twenty business organizations that participated In the survey were women. However, a women at the managerial and professional levels accounted for less than I percent of the total number of employees. Fretz and Hayman concluded from their survey that women are hindred by their own preconceived Ideas of what their roles In business should be (Fretz and Hayman, 1973, p. 134).

One of only a few available studies with data regarding women In the work force as early as 1920 is presented by Elizabeth Waldman. Most of the data are from the Current Population Survey conducted by the Bureau of the Census for the Bureau of Labor Statistics. They relate to the population 16 years of age and over. The typical working woman In 1920 was single, about 28 years of age and from the









working class. The typical working woman of today is married and half of them are over 39 years of age. Chances are that the working women of today are supplementing the family Income (Waldman, 1970, pp. 10-18).

In an attempt to eliminate the myths about women managers,

research has been conducted to provide Insight Into the similarities and dissimilarities of men and women managers. As opposed to a predominantly psychological approach to women's attitudes, values, and beliefs and how these factors cause them to behave In a business situation, Reif and Newstrom utilized an approach based on women managers' perceptions of their work environment. The purpose was to determine If their views of the formal and informal aspects of organizations are different from the views of men. If this was true, a possible conclusion would be that women should be treated differently from men. The methodology centered on the semantic differential technique, which measures perceptions along two independent dimensions, evaluative and potency. A questionnaire was administered to 286 men and 55 women In management development programs In business and government organizations. It was concluded that women and men managers are more similar than dissimilar in their feelings about their work climate. Based on this, Reif and Newstrom concluded "that decisions made about women on the basis of their sex, without considering such individual factors as background, education, experience, personality and potential, are likely to be wrong" (Reif and Newstrom, 1975, p. 78).

Recent studies have yielded contradictory results concerning

male and female attitudes and behavior as they work in organizations.









Some studies found persons following traditional sex-role stereotypes; in others, they have not. Powell and Butterfield explored the possibility that sex-role Identification is a more Important variable than is gender in an individual's perceptions of traits desirable for management personnel. One hundred and ten graduate students with Jobs in the business community and 575 undergraduate business students participated in the study conducted by Powell and Butterfield. The students completed the Bem Sex-Role Inventory for both themselves and for a person whom they believed to be a "good manager."

As expected, individuals' sex-role Identifications significantly affected their perceptions of traits desirable
for management personnel, while sex had virtually no effect.
The study concludes that sex-role identification Is a
variable deserving of further attention, particularly when
sex-related differences are examined. Also, a graduate
woman revealed more masculine traits than feminine In their
self-descriptions, suggesting that a masculine standard
for management may nullify the femininity of women in or
aspiring to management positions. (Powell and Butterfield,
1978, p. 1)

Virginia Schein cited that limited attention has been paid to the Impact of sex-role stereotypical thinking on the perceived and actual performance of women in management. Her Investigation of prior research in the areas of sex-role stereotypes and performance perceptions revealed an over-riding theme that sex role stereotyping has a definite and negative impact on the selection of women into managerial positions. She further stressed the need for research on the relationship between power and political behavior and effective managerial performance.









Limited opportunities to acquire work-related power acquisition behaviors and exclusion from political/
Influence networks within organizations can, then, limit
the performance effectiveness of the women with the
potential to do well, as well as, in the long run diminish
her motivation to perform. As such, a prime research target
should be the role of power acquisition behaviors in effective managerial functioning and the effect of sex and sexrole stereotypical thinking on acquisition of and opportunity to use these behaviors. (Schein, 1978, p. 266)

Jeanne Herman and Karen Gyllstrom discovered through their recent study of 500 employees stratified by sex and Job classification in a major midwestern university that myths are dispelled concerning Inter-role conflict for women employed full-time. "A potential employer could not justify hiring a male over an equally competent female on the grounds that the woman will experience greater Interrole conflict." The results show that married women with children who work full time report no more conflict between job and family responsibilities than do men who hold a comparable number of social roles (Herman and Gyllstrom, 1978, p. 330).

Sex Roles of Women In Education

Taylor concluded in her doctoral dissertation that, all things being equal, superintendents of schools are not likely to hire women as school administrators. The researcher concluded that the only factor having any real and constant significance of who will be hired and who will not be hired is the sex of the applicant (Taylor, 1973, p. 128).

The percentage of women elementary school administrators during the past forty years has steadily declined. Information on the high school administrative level showed the same trend. Data provided by the National Council of Administrative Women In Education Indicated the following:








in the eleven-year periods between 1950-51 and 1960-61,
the number of women serving as junior high and senior high
school principals dropped from 18% to 3.8% for all secondary
schools. (Van Heir, 1975, p. 163)

More recent statistics cited by the National Center for Education Statistics revealed that In the State of Florida in 1977, 23 percent of public school principals and assistant principals were women (National Center for Education Statistics, 1978, p. 19).

Young (1976) "researched popular myths In public education concerning women in the field of educational leadership at the elementary and secondary level. For each myth cited, contrary evidence evolved from the study. Several of the most quoted myths that were disproved are (l) women rarely possess the credentials for administrative positions, (2) women seldom aspire or apply for administrative positions, (3) women will not make the required commitment to an administrative position, and (4) women do not have to work because they are merely after a second income (Young, 1976, p. 83).

Borgstrom, Whiteley, and Rudolph studied the Impact of recent

societal changes on women's educational aspirations. The researchers utilized data from a student survey administered by the American Council on Education. They compared the data with survey responses of freshman students at the University of California, Santa Barbara (UCSB), from 1968 to 1978. They were attempting to characterize the University of California, Santa Barbara, campus culture today and show how and when UCSB students have changed over the past ten years.









Women are currently much more undecided about their majors
than are their male counterparts and even substantially
more than women In past years, perhaps reflecting some
confusion about the Increased options available to them.
While showing significant decreases In areas traditionally
associated as being "women's majors" (English, fine arts, social sciences) as well as the less traditional math and
statistics, "undecided" women students have shown Increased Interest in professional fields such as business, engineering and the health professions. (Borgstrom, Whiteley and
Rudolph, 1980, pp. 6-7)

In a recent study by Pulg-Casauranc, women in three fields of

academic study were Investigated. The Edwards Personal Preference

Schedule (EPPS) and the Bem Sex-Role Inventory (BSRI) were used to

assess the relationship between personality characteristics and psychological androgyny for females In traditional (female-dominated), nontraditional (male-dominated) and neutral (dominated by neither men

nor women) fields of academic study. A random sample of 185 women from

junior and senior classes of traditional, nontraditional and neutral

academic departments participated In the study.

The data Indicated that the three groups could also be differentiated by their psychological androgyny scores.
Significant differences for the sample group found the
nontraditional group demonstrated the highest degree of psychological androgyny, although all three groups fell within the range defined as psychologically androgynous.
(Pulg-Casauranc, 1977, p. 5)

The most significant finding of the study was that the females In

the three groups thought to be qualitatively different, tended to be

more similar than disparate when It came to displaying and Internalizing

behaviors they perceived as appropriate (Puig-Casauranc, 1977, p. 7).














CHAPTER I II
PROCEDURES USED IN CONDUCTING THE STUDY



The following procedures were used in the study:

1. The sample population included women students who were

enrolled In the University of Central Florida College of Business Administration for at least one academic quarter

between January, 1979, and June, 1980, and who were enrolled

at the University of Central Florida during the Spring

Quarter, 1980. This sample population was found by using

University records and official University data tapes.

2. The sample population was stratified Into two groups.

Group I: Those women students enrolled in the College of

Business Administration at the University of

Central Florida during the Spring Quarter, 1980,

who had completed ninety quarter hours of college

study Including at least two courses taught in the

College of Business Administration.

Group I1: Those women students enrolled at the University

of Central Florida during the Spring Quarter,

1980, who took at least two courses in business

administration, and who had transferred out of the

College of Business Administration at some point

between January, 1979, and June, 1980.









3. The ages of women students In Groups I and I! were obtained

from the student's permanent record and official University

data tapes.

4. Scholastic Aptitude Test scores, American College Test

scores or Florida Twelfth Grade Test scores were obtained

from the student's permanent record and official University

data tapes. The scores were converted to a common base

utilizing the State University System Equivalency Document

(Jackson, 1977).

5. In the Spring Quarter of 1980, following the assignment of

students to classes, a list of all women students in Groups

I and II was compiled from University registration Information.

6. Demographic data for Groups I and II were obtained from a

demographic profile data sheet developed and administered

by the Investigator (Appendix A).

7. A letter (Appendix B) was mailed to each person selected,

requesting her cooperation in completing the Bem Sex-Role

Inventory which was administered In a conveniently located

room on campus.

8. When a large enough sample was not produced by the procedure

in 17, follow-up telephone calls, extending a second Invitation to participate in the study, were made with careful attention being given to avoid pressuring students during

the telephone conversation. A minimum of fifty percent

participation by women students contacted in Groups I and II

was sought.









9. The investigator administered the Bern Sex-Role Inventory

(Appendix C).

10. Responses to the final question on the demographic profile

data sheet, "If you were enrolled in the College of Business

Administration after January, 1979, and subsequently have

transferred to another college at the University of Central Florida, please indicate in the space provided your reasons for the transfers," were reviewed and categorized by a panel

of knowledgeable professionals with expertise In the counselling of students. The purpose of categorizing responses

was to enable the Investigator to cite the main reasons

indicated for transferring out of the College of Business

Administration and to compare these responses with the

findings of the study. It was recognized that those students

in Group II may perceive their reasons for transferring out of the College of Business Administration in many different

ways. This question was Included only to give additional

Information concerning reasons for transferring out of the

College of Business Administration. A check was made to

determine if the reasons did or did not tend to support

the major questions posed in this study.

11. As a means of providing comparative data on the masculinity,

femininity and androgyny scores, a sample of junior and senior

male students enrolled in selected classes In the College of Business Administration at the University of Central Florida

(Group IV), and a sample of women students enrolled in selected








classes In the College of Education at the University of Central Florida who had never been enrolled in the College of Business Administration (Group III) completed the Bern Sex-Role Inventory In their classrooms. These data were Included In the study to address questions surrounding the concept of the traditionally sex-typed person as follows:

(1) Did women students majoring In education, a traditionally feminine field of study, produce a more feminine score on the Bern Sex-Role Inventory than did the other groups In the study?

(2) Did men students majoring In business administration, a traditionally masculine field of study, produce a more masculine score on the Bem Sex-Role Inventory than did the other groups in the study?














CHAPTER IV
ANALYSIS OF DATA



A total of 159 subjects were included in the study. These subjects were divided Into four groups as follows: Group I (thirty-three subjects) was composed of a sample of women enrolled In the College of Business Administration at the University of Central Florida during the spring quarter, 1980, all of whom had completed ninety quarter hours of college study, Including at least two courses taught In the College of Business Administration. Group II (forty subjects) was made up of a sample of women students enrolled at the University of Central Florida during the spring quarter, 1980, all of whom had taken at least two courses in business administration, and who had transferred out of the College of Business Administration at some point between January, 1979, and June, 1980. Group III (twenty-three subjects) was a sample of women students enrolled in selected classes In the College of Education who had never been enrolled In the College of Business Administration. Group IV (sixty-three subjects) was a sample of male students enrolled In classes In the College of Business Administration during the spring quarter, 1980.

Groups I and II were the groups upon which this study was focused.

Groups III and IV were included in the study as a means of providing comparative data; consequently, Groups III and IV were not used in the discriminant analysis portion of the study, but only in the descriptive analysis. Groups III and IV were Included to address questions surrounding






32

the concept of the traditionally sex-typed person, such as: (1) Did women students majoring In education, a traditionally feminine field of study, produce a more feminine score on the Bem Sex-Role Inventory than did the other groups in the study? (2) Did men students majoring In business administration, a traditionally masculine field of study, produce a more masculine score on the Bem Sex-Role Inventory than did the other groups in the study?

The Statistical Package for the Social Sciences (SPSS) computer program was used to analyze the data. A discriminant analysis with significance at .05 was used to test the data. The discriminant analysis enabled the researcher to statistically distinguish between groups by selecting the Independent variables on which the groups differed.

The discriminant analysis produced a discriminant function, a regression equation, with group membership as the dependent variable and demographic data along with scores on the Bem Sex-Role Inventory as the Independent variables. The results statistically Indicated to which group each subject probably belonged. The analysis discriminated between the two groups (Groups I and II) and predicted which subjects were or should be in either Groups I or II.

The discriminant analysis was used to predict whether a subject belonged to Group I or Group II on the basis of the selected independent variables, and to determine the importance of these Independent variables in making optimal assignments of the subjects to the two groups. A more in-depth discussion of the discriminant analysis appears In the "Discriminant Analysis Findings" portion of this chapter.










A cross tabulation of variables by groups was performed to describe these variables as they related to the four groups In the study. These frequency distributions compared variables for one group to the same variables for another group. The exact differences were more clearly revealed when the percentages were examined because of the difference In total numbers of subjects in each of the four groups. An analysis of these four groups Is reported in the following narrative discussion and tables.


Analysis of Descriptive Data

Twelve variables were used In the analysis of descriptive data.

Seven variables from the demographic profile data sheet were pertinent to the descriptive analysis. The questions on the demographic profile data sheet illiciting current college of enrollment and previous colleges of enrollment were used only for verification of assignment of subjects to groups. The name of the high school attended and occupation of the student were recorded but were not used In the description of the groups.

The seven variables from the demographic profile data sheet which were considered pertinent to the descriptive analysis were (1) age, (2) marital status, (3) employment status, (4) number of children, (5) grade point average, (6) student's enrollment classification (freshman, sophomore, junior or senior), and (7) year of high school graduation. Two variables, the sex of the student and Scholastic Aptitude Test score or its equivalent, were taken from the student's permanent university record.








Sex

Included In the four groups cited in the study were sixty-three male business students, all of whom were Included In Group IV, and ninety-six women students In Groups I, II, and III (Table 1).


The 159 participants ranged in age from 18 to 55 years. Approximately 75 percent of the participants were under 25 years of age. Fifteen percent of Group I, 10 percent of Group II, 25.9 percent of Group III and 17.5 percent of Group IV ranged in age from 25 to 34 years. Six percent of Group I, 10 percent of Group 1i, 8.6 percent of Group III, and 8 percent of Group IV ranged in age from 35 to 55 years (Table 2). Marital Status

The information on marital status was collected in categories of

single or divorced, married, or widowed. An analysis of the 159 participants revealed that 73 percent were single or divorced, 24.5 percent were married and 2.5 percent were widowed. An analysis by group showed Group I, women business students, as having the highest percentage of single or divorced members. Women education majors (Group III) had the smallest percent of single or divorced members (Table 3). EmIployment

An analysis of the employment status, which was collected in categories of full time, part time, or unemployed, Indicated a high percentage (78.3) of unemployed members in Group III, women education majors. Group I had 30.3 percent unemployed, Group II had 40 percent unemployed and Group IV had 39.7 percent unemployed (Table 4).














Table I

Sex of Students In
Groups I, 11, III and IV


Number of Students Percentage of Total Students Male Female Used In the Study


Major Groups


Group Ia Group IIb




Group IIIc Group IVd


20.8 25.2


Comparison Groups
14.5


39.5


aWomen students enrolled in the College of Business Administration. bwomen students who had transferred out of the College of Business
Administration Into another college within the University. CWomen students enrolled in the College of Education. dMen students enrolled In the College of Business Administration.













Table 2

Ages of Students
In Groups I, II, III and IV


Percentages of
Ages of Group Ia Group IIb Group IIIC Group IVd Students (N-33) (N-40) (N-23) (N-63)

18 to 24 78.8 80.0 65.2 74.6 25 to 34 15.1 10.0 25.9 17.5 35 to 55 6.0 10.0 8.6 8.0



aWomen students enrolled in the College of Business Administration. bWomen students who had transferred out of the College of Business
Administration Into another college within the University. CWomen students enrolled In the College of Education. dMen students enrolled in the College of Business Administration.













Table 3

Marital Status of Students In Groups I, 1i, III, and IV


Number of Students and Percentages of Groups Group Ia Group Ib Group IIIc Group IVd


Single/Divorced
Number 27 29 12 48 Percent 81.8 72.5 52.2 76.2

Married
Number 6 9 9 15
Percent 18.2 22.5 39.1 23.8 Widowed
Number 0 2 2 0 Percent 0 5.0 8.7 0



awomen students enrolled In the College of Business Administration. bwomen students who had transferred out of the College of Business
Administration into another college within the University. CWomen students enrolled In the College of Education. dMen students enrolled In the College of Business Administration.














Table 4

Employment Status of Students in Groups I, 1i, 1I1, and IV


Number of Cases
and Percentages of Groups
Status Group Ia Group 11b Group IIIc Group IVd


Full Time
Number 2 7 1 9
Percent 6.1 17.5 4.3 14.3

Part Time
Number 21 17 4 29
Percent 63.6 42.5 17.4 46.0

Unemployed
Number 10 16 18 25 Percent 30.3 40.0 78.3 39.7



auomen students enrolled in the College of Business Administration. buomen students who had transferred out of the College of Business
Administration Into another college within the University. CWomen students enrolled in the College of Education. dMen students enrolled in the College of Business Administration.









Grade Point Average

The range of grade point averages In college work for all groups was from 1.87 to 3.99 on a 4.00 scale. Grade point averages were determined from the students' permanent university records. The women students in Group II had the lowest grade point averages; twenty-three percent of this group had grade point averages of 2.49 or below. In all of the groups 44.6 percent to 56.3 percent had a grade point average of 3.0 or better in their college work (Table 5). No special minimum grade point average was required for admission to the College of Business Administration or the College of Education; however, a minimum of 2.0 overall grade point average was required for the admission of all degree seeking students (Table 5).

Student Classification

The Information on student classification was collected in categories of freshmen, sophomore, junior, and senior. Seventy-seven and five tenths percent of the students In Group II, women students who transferred out of the College of Business Administration, were classified at the junior level or above. All the participants in Groups I, III, and IV were classified at the junior level or above (Table 6). At the University of Central Florida students are permitted to enter the college of their choice at the beginning of their freshman year.

Scholastic Aptitude Test Scores

Scholastic Aptitude Test scores or their equivalent, as measured on the American College Test or the Florida Twelfth Grade test, for Groups I and II were taken from the permanent university records. The scores ranged from 590 to 1350. The highest score possible Is 1600 and the













Table 5

Grade Point Averages for Students
In Groups I, 1i, III, and IV


Percentage of Each Group Within Grade Point Rangesa

1.87 to 2.49 2.50 to 2.99 3.00 and Above Group I b 15.1 30.2 54.6 Group Ic 22.5 32.5 45.0 Group I1Id 8.6 34.6 56.3 Group IVe 14.4 41.4 44.6


aGrade point averages are computed on a four point scale 4.0 - A grade. bwomen students enrolled In the College of Business Administration. CWomen students who had transferred out of the College of Business
Administration into another college within the University. dWomen students enrolled in the College of Education. eMen students enrolled in the College of Business Administration.














Table 6

Student Classifications In Groups I, II, II1 and IV


Percentages of Groups
Student
Classifications Group I Group 11b Group IIIc Group IV


Freshman 0 10.0 0 0 Sophomore 0 12.5 0 0 Junior 24.2 45.0 69.6 9.5 Senior 75.8 32.5 30.4 90.5



aWomen students enrolled In the College of Business Administration. bwomen students who had transferred out of the College of Business
Administration into another college within the University. CWomen students enrolled In the College of Education. dMen students enrolled in the College of Business Administration.











lowest Is 400. Forty-five and two tenths percent of Group I scored over 1,000, as compared to 32.5 percent of Group II (Table 7). Scholastic Aptitude Test scores or their equivalents were not compiled for the subjects In Groups III and IV. These groups were Included In the study for the purpose of comparing results on the Bern Sex-Role Inventory (Table 7). Number of Children

The women students In the College of Education (Group III) reported

having the grestest number of children; 34.7 percent had from one to three children. The greatest number of children reported by any one participant was six. A majority of each group had no children (Table 8). Year of High School Graduation

Participants In the study reported the year of their graduation from high school on the demographic Information sheet. The range was from 1943 to 1979 with 77 percent of the participants having graduated since 1973 (Table 9).


Analysis of Open-ended Question

Three professionals with expertise In the counselling of university

students were selected to review and categorize the open-ended question on the demographic profile data sheet. Each reviewer had a minimum of ten years of academic and vocational counselling experience In Institutions of higher education.

The open-ended question was, "If enrolled in the College of Business

Administration after January, 1979, and subsequently transferred to another college at the University of Central Florida, please Indicate in the space provided your reasons for the transfer."













Table 7

Scholastic Aptitude Test Scores for Students In Groups I and II


Percentages of Groups
Ranges for
SAT Scoresa Group Ib Group lIc


590 - 800 03 18 801 - 1000 51 50 1001 - 1200 36 24 1201 - 1350 09 08


aSAT Is the abbreviation for Scholastic Aptitude Test. The highest
possible score is 1600.
bWomen students enrolled In the College of Business Administration. CWomen students who had transferred out of the College of Business
Administration into another college within the University.














Table 8

Number of Students in
Groups I, II, 11, and IV with
Children


Number of Group Ia Group I1b Group IIIc Group IVd Children (N-33) (N,,40) (N-23) (N-63)


One 1 3 3 1 Two 1 1 5 6 Three 1 2 0 3 Four 0 0 0 1 Five 0 1 0 0 Six 0 1 0 0 None 30 32 15 52



aWomen students enrolled In the College of Business Administration. b/omen students who had transferred out of the College of Business
Administration Into another college within the University. CWomen students enrolled In the College of Education. dMen students enrolled in the College of Business Administration.














Table 9

Year of High School Graduation
for Students In Groups I, II, i11, and IV


Year
1943 to 1959
Number Per t


of Graduation from High School
1960 to 1969 1970 to 1979
Numer Percent Number Percent


Group Ia 2 6.1 2 6.0 29 87.9 Group 3b 7.5 3 7.5 34 85.0 Group IIIc 2 8.6 3 12.9 18 78.0 Group IVd 3 4.8 7 11.2 53 84.2



a Women students enrolled In the College of Business Administration. bwomen students who had transferred out of the College of Business
Administration into another college within the University. CWomen students enrolled in the College of Education. dMen students enrolled In the College of Business Administration.










Approximately 43 percent of the forty students in Group II, those women students who transferred out of the College of Business Administration, responded to the open-ended question. Four broad categories of answers emerged in the process of grouping the participants' responses, and all responses fell Into these four categories. They were (1) indecision as to field of study, (2) change of career Interests, (3) expediency of completion of a degree in another field of study, and (4) boredom or lack of enthusiasm for the business curriculum.

The category receiving the greatest response was "boredom or lack of enthusiasm for the business curriculum." Forty-one percent, seven out of the seventeen responding to the question, were assigned to this category (Table 10). None of the respondents stated that the academic work in business administration was too difficult or that they transferred out of business administration because they were getting low grades In the business courses,

Descriptive Analysis of
Bem Sex-Role Inventory

Sixty personality characteristics are listed on the Bern Sex-Role Inventory. Twenty of the characteristics are stereotypically feminine, twenty are stereotypically masculine and twenty serve as filler items. Participants in the study indicated how well each of the sixty personality characteristics described herself or himself on a scale from one to seven. The scale ranged from one, "Never or almost never true," to seven, "Always or almost always true." The sixty personality characteristics are placed on the Bem Sex-Role Inventory in the following manner: (I) The first














Table 10

Responses by Group II Students to the
Four Categories Derived from the Open-ended Question on the
Demographic Profile Date Sheet


Number of Responses
Categories (N-17)


Indecision as to Field of Study 3 Change in Career Interests 3 Expediency of Completion of Degree in
Another Field of Study 4 Boredom or Lack of Enthusiasm for the
Business Curriculum 7


Seventeen of the 40 students in Group 1i responded


to the question.


Note: Open-ended question was, "If enrolled In the College of Business
Administration after January, 1979, and subsequently transferred to
another college at the University of Central Florida, please Indicate
In the space provided your reasons for the transfer."









adjective and every third one thereafter Is masculine; (2) the second adjective and every third one thereafter is feminine; (3) the third adJective and every third one thereafter is filler.

Scoring of the Bem Sex-Role Inventory Is conducted as follows:

The first step is the calculation of each subject's femininity ("a") and masculinity ("b") scores, which are the averages of
each subjects' ratings of the feminine and masculine adjectives
on the BSRI. That Is, a given subject's femininity score Is the
mean of that subject's ratings on the feminine adjectives, and
that same subject's masculinity score is the mean of her or his
ratings on the masculine adjectives. (Bern, 1979, p. 1)

The total score for each subject's ratings on the feminine items

was divided by twenty, If all characteristics were rated by the subject, and the same procedure was used for calculating the ratings on the masculine characteristics. All subjects rated all Items on the Bem Sex-Role Inventory. A feminine raw score (a) and a masculine raw score (b) were the result of this process.

The masculine and feminine standard scores were determined from the

masculine and feminine mean raw scores, using Table 1 in the Administration and Scoring Guide for the Bern Sex-Role Inventory (Appendix D). A difference score was computed by subtracting the masculine standard score from the feminine standard score. The difference score was converted to a difference standard score by using Table 2 in the Administration and Scoring Guide for the Bern Sex-Role Inventory (Appendix D). The median for the feminine raw score, 4.90, and for the masculine raw score, 4.95, is based on normative data. The median for the difference standard score, also based on normative data, Is 0 which was converted to a T-score of 50 by using Table 2 in the Administration and Scoring Guide for the Bem SexRole Inventory (Appendix D).









Each subject was classified into one of four categories based on the

median raw scores of the normative sample and the subject's mean raw scores. All participants In the study were placed In one of the following four categories: (1) masculine, which denotes a low feminine score and high masculine score, (2) feminine, which denotes a low masculine score and a high feminine score, (3) androgynous, which denotes a high masculine score and a high feminine score, or (4) undifferentiated, which denotes a low masculine score and a low feminine score (Table 11). "High difference scores In either direction Indicate a tendency to be strongly sex-typed, positive scores Indicating femininity and negative scores Indicating masculinity" (Bem, 1979, p. 3). Positive scores indicate a generally more feminine orientation, negative scores a more masculine orientation, and scores near zero an androgynous orientation.

Thirteen and three tenths percent of the subjects In Group I (women students enrolled in the College of Business Administration), 15.0 percent of the subjects In Group II (women students who had transferred out of the College of Business Administration), and 13.6 percent of the subjects In Group III (women students enrolled in the College of Education) were placed in the masculine category as a result of their scores on the Bem Sex-Role Inventory. Sixty-two and one tenth percent of the subjects in Group IV (men students enrolled in the College of Business Administration) were placed In the masculine category (Table 12).

Thirty-six and seven tenths percent of women students enrolled in the College of Business Administration (Group I) were classified as feminine as a result of their scores on the Bern Sex-Role Inventory. Thirty-seven










Table 11

Interpretation of Categories
Derived from the
Bern Sex-Role Inventory


Below Median


Masculine Score


Above Median


Below Median
Fern in i ne
Score Above
Median


Undifferentiated Masculine (low feminine - low masculine) (low feminine - high masculine)


Feminine Androgynous (high feminine - low masculine) (high feminine - high masculine)


Medians Based on Normative Data


Feminine Raw Score

Masculine Raw Score


4.90

4.95














Table 12

Comparison of Undifferentiated, Masculine,
Androgynous, and Feminine Bem Sex-Role Inventory Classifications by Group



Bem Sex-Role Inventory Percentages of Groups Classifications Group Ia Group jib Group I11c Group IVd


Undifferentiated 6.7 7.5 13.6 17.2 (low masculine and
low feminine)

Masculine 13.3 15.0 13.6 62.1
(low feminine and
high masculine)

Androgynous 43.3 40.0 45.5 19.0 (high feminine and
high masculine)

FemInine 36.7 37.5 27.3 1.7 (low masculine and
high feminine)


aWomen students enrolled In the College of Business Administration. bWomen students who had transferred out of the College of Business
Administration into another college within the University. CWomen students enrolled in the College of Education. dMen students enrolled In the College of Business Administration.








and one half percent of the women students who had transferred out of the College of Business Administration (Group II) and 27.3 percent of women students enrolled In the College of Education (Group III) were placed in the feminine category as a result of their scores on the Bem Sex-Role Inventory. The percent of male business students (Group IV) who were placed in the feminine category as a result of their scores on the Bem Sex-Role Inventory was 1.7.

Forty-three and three tenths percent of the subjects In Group I (woMen students enrolled in the College of Business Administration), 40.0 percent of the subjects In Group II (women students who had transferred out of the College of Business Administration), and 45.5 percent of the subjects in Group III (women students enrolled In the College of Education) were placed in the androgynous category as a result of their scores on the Bem Sex-Role Inventory. The percent of subjects In Group IV (men students enrolled in the College of Business Administration) who were placed In the androgynous category as a result of their scores on the Bem Sex-Role Inventory was 19.0.

Groups III and IV were Included in the analysis of the four categories of masculine, feminine, androgynous, and undifferentiated in order to address the following questions. (1) Did women students majoring in education, a traditionally feminine field of study, produce a more feminine score on the Bern Sex-Role Inventory than did the other groups In the study? (2) Did men students majoring in business administration, a traditionally masculine field of study, produce a more masculine score on the Bem Sex-Role Inventory than did the other groups In the study? Women majoring In education (Group III) did not produce a more feminine score than did the other groups.









Twenty-seven and three tenths percent of the women majoring In education

were categorized as feminine, as compared to 36.7 percent of women business

majors, 37.5 percent of women who transferred out of the College of Business Administration, and 1.7 percent of the men business majors. Men

students majoring in business administration did produce a more masculine

score than did the other groups. Sixty-two and one tenth percent of men

majoring in business administration were categorized as masculine, as compared to 13.3 percent of women business majors, 15.0 percent of women who

transferred out of the College of Business Administration, and 13.6 percent of the women education majors.


Discriminant Analysis Findings

The data were analyzed through a discriminant analysis.

The mathematical objective of discriminant analysis is to weigh and linearly combine the discriminating variables In
some fashion so that the groups are forced to be as statistically distinct as possible.

Discriminant analysis attempts to do this by forming one or
more linear combinations of the discriminating variables.
These 'discriminant functions" are of the form

D - d 11ZI + d12Z2 + . .. + dpZp

where D, Is the score on discriminant function I, the d's are weighting coefficients and the Z's are the standardized values
of the p discriminating variables used in the analysis. The
maximum number of functions which can be derived Is either one
less than the number of groups or equal to the number of discriminating variables, if there are more groups than variables.
Ideally, the discriminant scores (D's) for the cases within
a particular group will be fairly similar. At any rate, the functions are formed In such a way as to maximize the separation of the groups. Once the discriminant functions have
been derived, we are able to pursue the two research objectives
of this technique; analysis and classification.









The analysis aspects of this technique provide several tools for the interpretation of data. Among these are statistical
tests for measuring the success with which the discriminating
variables actually discriminate when combined Into the discriminant functions....The result is often of major theoretical
significance, and statistical tests are Included for this purpose. Since the discriminant functions can be thought of as the axes of a geometric space, they can be used to study the spatial relationships among the groups. . . More importantly,
the weighting coefficients can be interpreted much as In
multiple regression or factor analysis. In this respect, they
serve to Identify the variables which contribute most to differentiation along the respective dimension (function). (Nie,
1975, pp. 435-436)

The discriminant analysis was used to statistically distinguish between Group I, women students enrolled in the College of Business Administration, and Group ii, women students who transferred out of the College of Business Administration.

The analysis resulted in a linear combination of independent variables (discriminant function) that tended to maximize the separation of the two groups. The rule for determining the number of equations in a discriminant analysis Is to utilize one function less than the number of groups in the study, or to use the number of equations (discriminant functions) equal to the number of discriminating variables, whichever is smaller. Consequently, one equation (discriminant function) was derived In this study because two groups were being studied.

When testing for statistical significance of the discriminating variables, a significance level of .05 was found.

The direct method was used in order to select the Independent variables which would contribute to the discriminant analysis. All eleven of the possible Independent variables were entered into the analysis concurrently, and the discriminant function was created directly from the entire set of










variables, with two variables being rejected or not contributing to the discriminant function. The direct method puts all possible variables In at one time, as opposed to a stepwise procedure wherein possible variables

are entered one at a time.

The direct method of entry was selected to produce a discriminant function that would, to the best of its ability, statistically distinguish between Groups I and II.

Information concerning the methods used for assigning values to the eleven variables Is discussed below, and the methods used for assigning values to each of the nine discriminating variables which surfaced are Indicated by footnotes In Table 13.

When data are being collected, a process of measurement is used in order to assign a value or score to the observed phenomenon or variable. In this study an ordinal level of measurement was used for some of the variables. When it was possible to rank order a category according to some criteria, an ordinal level of measurement was used. For instance the order of freshman, sophomore, junior, and senior In the student classification variable was ranked 1, 2, 3, 4, with freshman being I and senior being 4. The ordinal level of measurement assumes a reasoned order for value or score assignment, but does not require a known distance between ranks or scores (Nie, 1975, p. 5).

"The interval level of measurement has an additional property in that the distances between the categories are defined in terms of fixed and equal

units" (Nie, 1975, p. 5). Scholastic Aptitude Test scores, for Instance, Indicate the same or fixed distance between 800 and 900 or 1000 and 1100.






Table 13


Standardized Discriminant Function Coefficients
at the .05 Level of Significance Presented In Order
of Importance to the Discriminant Function and the Levels
of Measurement Used In the Assignment of Values
to the Discriminating Variables


Discriminating Variables Standardized Discriminant Levels of (Assigned Values) Function Coefficients* Measurement

Student Classification 0.854 Ordinal**
(freshman - 1, sophomore - 2,
junior - 3, senior - 4)

Marital Status -0.632 Ordinal
(single/divorced = 1,
married - 2, widowed - 3)

Feminine Standard Score from the -0.368 Interval
Bem Sex-Role Inventory
(actual scores)

Scholastic Aptitude Test Score 0.287 Interval
(actual scores)

Difference Standard Score from the 0.158 Interval
Bem Sex-Role Inventory
(actual scores)

Employment Status 0.117 Ordinal
(full time - 1, part time - 2
unemployed - 3)

Age 0.041 Ratio
(actual ages)

Grade Point Average -0.027 Ratio
(actual averages)

Masculine Standard Score from the -0.023 Interval
Bem Sex-Role Inventory
(actual scores)
*The Influence of one variable to the function In relation to the
Influence of another variable may be made by Ignoring the sign
and comparing coefficients. Zero to +1 drives the case to Group I,
and a 0 to -1 drives the case to Group II.
**Ordinal level measurements rank order categories according to some
criteria.
***Interval level measurements have fixed and equal units of distance
between categorical ranks.
****Ratio level measurements have known zero points and have fixed and
equal distances between categorical values.









"The ratio level of measurement has all of the properties of an interval level with the additional property that the zero point is inherently defined by the measurement scheme" (Nie, 1975, p. 5). Age Indicates the same or fixed distance between age 18 and 19 or age 25 and 26, and has the additional property of a zero point. Table 13 shows the values or scores assigned to each variable and the level of measurement used in each assignment.

The eleven variables subjected to the direct method included seven

Items from the demographic profile data sheet (1) age, (2) marital status,

(3) employment status, (4) grade point average, (5) student classification,

(6) number of children, and (7) year of high school graduation, one Item from the permanent University file (Scholastic Aptitude Test score or Its equivalent), and the three standard scores derived from the personality characteristics on the Bern Sex-Role Inventory (masculine standard score, feminine standard score, and difference standard score). Items six and seven from the demographic profile data sheet were rejected, leaving nine variables contributing to the discriminant function.

The nine discriminating variables which surfaced In the analysis as contributing to the discriminant function were (1) age of the participant,

(2) marital status, (3) employment status, (4) grade point average, (5) student classification, (6) Scholastic Aptitude Test score, (7) Bem SexRole Inventory masculine standard score, (8) Bem Sex-Role Inventory feminine standard score, and (9) Bem Sex-Role Inventory difference standard score.

The means and standard deviations for the nine discriminating variables are presented in Tables 14 and 15.














Table 14

Means and Standard Deviations
for Discriminating Variables at .05 Level
of Probability for Groups I and II


Discriminating Group Ia Group I1b
Variables Mean S.D. Mean S.D.


Age, (range - 18 to 55) 23.21 6.10 23.70 7.71 Marital Statusd 1.18 0.39 1.33 0.57 Employment Statuse 2.24 0.56 2.23 0.73 Grade Point Averagef 2.98 0.44 2.91 0.56 Student Classification 3.73 0.52 3.03 1.07 SAT Scoresg 987.12 141.42 930.50 183.35


aWomen students enrolled in the College of Business Administration. i/omen students who had transferred out of the College of Business
Administration Into another college with the University. cS.D. Is the abbreviation for standard deviation. dMarital status was recorded as single or divorced (Value - 1),
married (Value - 2), or widowed (Value - 3).
eEmployment status was measured as full time (Value - 1), part time
Value - 2), or unemployed (Value - 3).
fGrade point averages were computed on a four point scale (4.0 - A grade).


gscholastic Aptitude Test scores are normally
range from 500 to 1500 points or higher.
score for admission to the University of


computed on a maximum The current minimum Central Florida is 800.













Table 15

Means and Standard Deviations for the Bern Sex-Role inventory
Masculine, Feminine and Difference Standard Scores
for Groups I and II


Discriminating Group I Group 11C
Variablesa Mean S.D. d Mean S.D.


Masculine Standard Scoree 52.67 10.69 54.30 9.26
(range - 39 to 69)

Feminine Standard Scoref 55.50 11.81 58.43 7.57
(range - 44 to 73)

Difference Standard Scoreg 62.77 23.55 65.28 11.25
(range - 51 to 82)


aThree of the nine variables which surfaced at
cance during the discriminant analysis.


a .05 level of signifi-


bWomen students enrolled In the College of Business Administration. CWomen students who had transferred out of the College of Business
Administration into another college within the University. dS.D. is the abbreviation for standard deviation. eStandard score computed from responses to twenty masculine personality
characteristics on the Bem Sex-Role Inventory.
fStandard score computed from responses to twenty feminine personality
characteristics on the Bem Sex-Role Inventory.
gstandard score computed from the difference between the masculine and
feminine scores on the Bem Sex-Role Inventory.










The discriminant function coefficients were computed for the nine discriminating variables. The coefficients represent the relative contributions of the nine discriminating variables to the equation (discriminant function). They serve to Identify the variables which contributed most to the discriminant function. The coefficients for the discriminating variables are shown on Table 13.

In determining the relative weight of a particular variable, the sign of the coefficient Is ignored, and the value of the coefficient for any one variable may be compared to another. For instance, the student classification coefficient projects 20.8 times more Influence on the discriminant function than does the age coefficient. The relative weight In this example Is determined by dividing the coefficient for the age variable (.041) into the coefficient for the student classification variable (.854). (.854 divided by .041 - 20.8).

Discriminant scores were computed for each one of the participants

in Group I and II. The computation process was as follows: The standard value for the first discriminating variable was multiplied by its corresponding coefficient; this product was added to the product of the second discriminating variable which likewise was calculated by multiplying the standard value of the variable by its corresponding coefficient. This process took place for each of the nine discriminating variables, resulting in the computation of a "discriminant score" for each of the subjects In Groups I and II (Table 16).








Table 16

Classification of Subjects In
Groups I and II Utilizing Nine Discriminating Variables


Actual Group Predicted Group Discriminant Subjects Membership Membership Scores***
i i iii


1
1
2****
1 2****



1


1
1
I 2****


2****

1 2****
1
I
1
1

I

1
1


0.8478 1.7815
-0.2951 1.0657
-0.0610
1.0075
0.6711 0.5212 1.3583
-0.6170
-0.1134 1.6095
-0.3881 0.3993 0.8615 0.5013 0.1766
-0.2285
0.9171
1.5142
-0.9658 1.3706 0.7354 1.4973
0.8941
-0.0366
-0.4384 1.1789 1.5886


*Women students enrolled In the College of Business Administration.
**Women students who had transferred out of the College of Business Administration.
***Scores obtained when coefficents are applied to the nine discriminating variables. Zero to +1 drives the case to Group 1, and
a 0 to -1 drives the case to Group II.
****These subjects were not properly classified in their original group. NOTE: Age, marital status, employment status, grade point average,
student classification, Scholastic Aptitude Test scores, and
Bem Sex-Role Inventory masculine, feminine and difference scores
were the nine discriminating variables.











Table 16-continued


Actual Group Predicted Group Discriminant Subjects Membership Membership Scores***


2**
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2


1
2****

1
2
2
1 ****
2
2
2
2
2 1"***


2
1 ****
2
2
2
1 ****
2
2 I**** 1*** 1"***
2


0.9984 0.2017
-0.1695 0.7602
-2.5037
-0.1481 0.1248
-0.5548
-1.9723
-3.3995
-1.2594 0.0087
0.5658 0.6596
1.0142
-1.1935
1.1894
-2.5562
-1.8649
-0.9820
1.6084
0.3359
-0.1308
-0.5588 0.6834
0.2010 0.2789
0.4473
-0.5730


*Women students enrolled in the College of Business Administration.
**Women students who had transferred out of the College of Business Administration.
***Scores obtained when coefficients are applied to the nine discriminating variables. Zero to +1 drives the case to Group I, and
a 0 to -1 drives the case to Group 11.
****These subjects were not properly classified In their original group. NOTE: Age, marital status, employment status, grade point average,
student classification, Scholastic Aptitude Test scores, and
Bem Sex-Role Inventory masculine, feminine and difference scores
were the nine discriminating variables.








Table 16-continued


Actual Group Predicted Group Discriminant Subjects Membership Membership Scores***

59 2** I**** 0.4533 60 2 2 -1.0115 61 2 l**** 0.1311 62 2 2 -0.7055 63 2 2 -1.4451 64 2 2 -0.2870 65 2 2 -0.3365 66 2 2 -1.2436 67 2 2 -0.7151 68 2 2 -2.1282 69 2 2 -2.5279 70 2 1**** 1.0655 71 2 2 -0.5948 72 2 2 -0.2146 73 2 1**** 0.9951

*Women students enrolled in the College of Business Administration.
**Women students who had transferred out of the College of Business
Administration.
***Scores obtained when coefficients are applied to the nine discriminating variables. Zero to +1 drives the case to Group 1, and
a 0 to -l drives the case to Group II.
****These subjects were not properly classified In their original group. NOTE: Age, marital status, employment status, grade point average,
students classification, Scholastic Aptitude Test scores, and
Bem Sex-Role Inventory masculine, feminine and difference scores
were the nine discriminating variables.









Classification of Cases

The use of discriminant analysis as a classification technique
comes after the initial computation. Once a set of variables
is found which provides satisfactory discrimination for the cases with known group memberships, a set of classification
functions can be derived which will permit the classification
of new cases with unknown memberships. (Nie, Hull, Jenkins,
Steinbrenner and Brent, p. 436)

If variables are found which do well in predicting to which group the subjects belong, then these variables can be used to predict the likelihood of group membership for other students.

As a check on how well the equation (discriminant function) worked, the subjects were placed in either Group I or Group II according to the discriminant scores on the nine discriminating variables. Group membership was predicted, based on the nine discriminating variables. The goal of the classification process as it relates to a discriminant analysis was to measure the success of the equation (discriminant function) by observing how many subjects were properly classified Into their groups. Twenty-three subjects out of thirty-three subjects in Group I, women business majors, were properly classified. Twenty-flve subjects out of forty subjects in Group II, women students who transferred out of the College of Business Into another college within the University, were properly classified Into their group (Table 16).

The data for Groups I and II were subjected to the classification process. When the coefficients were applied to the nine discriminating variables, and a discriminant score of less than zero to minus one resulted, the subject was placed in Group I. A discriminant score of more than zero to plus one resulted in the subject's being placed In Group I for classification purposes.









When all the subjects In Group I were subjected to the classification process, 69.7 percent of them were properly placed In Group I; consequently, it was possible to predict group membership for Group I, women business majors, 69.7 percent of the time.

When all the subjects In Group I were subjected to the classification process, 62.5 percent of them were properly placed In Group II; consequently, it was possible to predict group membership for Group II, women students who transferred from the College of Business Administration to another college in the University, 62.5 percent of the time.

For any new woman student wishing to enter as a freshman In the College of Business Administration from whom data on the nine discriminating variables could be gathered, and for whom a discriminant score could be computed, the classification process could be performed. If she were placed In Group I as a result of her discriminant score, It could be predicted at a 69.7 percent level of probability that she would remain enrolled in the College of Business Administration. If as a result of the classification process, the same new woman student was placed In Group II, it could be predicted at a 62.5 percent level of probability that she would transfer out of the College of Business Administration.


Test of Hypotheses and Results

The hypotheses of this study were that an individual's membership

in either Group I or Group II is predictable on the basis of age, academic ability, and the Bem Sex-Role Inventory characteristics. The following sub-hypotheses in the null form served as the basis for making predictions of group membership.









1. The masculine score derived from the Bem Sex-Role
Inventory is a variable which does not discriminate when used In the prediction of group membership in
either Groups I or II.

This null hypothesis was rejected. The masculine standard score

was one of the nine variables which was shown to be discriminating in the discriminant analysis. The corresponding coefficient for the masculine standard score was -.023 at the .05 level of significance and was ninth out of nine discriminating variables. The coefficient for the next highest discriminating variable, grade point average (-.027), provides 1.17 times the influence of the coefficient for the masculine variable (-.023). The variable with the greatest influence, student classification (.854), has 37.13 times more Influence than does the coefficient for the masculine variable (-.023). The masculine variable, combined with the remaining eight variables In the discriminant function, resulted in the successful classification of 69.7 percent of the subjects In Group I and 62.5 percent of the subjects in Group II (Table 13).

2. The feminine score derived from the Bem Sex-Role Inventory Is a variable which does not discriminate when
used In the prediction of group membership In either
Groups I or II.
This null hypothesis was rejected. The feminine standard score was one of the nine variables that was shown to be discriminating in the discriminant analysis. The coefficient for the feminine standard score was

-.368 at the .05 level of significance and was third out of nine discriminating variables. The coefficient for the next highest discriminating variable, marital status (-.632), provides 1.71 times the influence of the coefficient for the feminine variable (-.368). The variable with

the greatest influence, student classification (.854), has 2.32 times more










Influence than does the coefficient for the feminine variable (-.368). The feminine variable, combined with the remaining eight variables In the discriminant function, resulted in the successful classification of 69.7 percent of the subjects In Group I and 62.5 percent of the subjects in Group I (Table 13).

3. The androgyny classification derived from the Bern
Sex-Role Inventory is a variable which does not
discriminate when used In the prediction of group
membership In either Group I or II.

This null hypothesis was rejected. The androgynous classification was determined through the difference standard score. The difference standard score was one of the nine variables that was shown to be discriminating In the discriminant analysis. The corresponding coefficient for the difference standard score was .158 at the .05 level of significance and was fifth out of nine discriminating variables. The coefficient for the next highest discriminating variable, Scholastic Aptitude Test score (.287), provides 1.82 times the Influence of the coefficient for the difference standard score (.158). The variable with the greatest influence, student classification (.854), has 5.41 times more Influence than does the coefficient for the difference score (androgyny) variable (.158). This difference variable, combined with the remaining eight variables In the discriminant function, resulted in the successful classification of 69.7 percent of the subjects In Group I and 62.5 percent of the subjects In Group II (Table 13).

4. The age of the Individual Is a variable which does not
discriminate when used in the prediction of group membership in either Groups I or II.









This null hypothesis was rejected. The age of the participants was one of the nine variables that surfaced as being discriminating. The corresponding coefficient for age was .041 at the .05 level of significance and was seventh out of nine discriminating variables. The coefficient for the next highest discriminating variable, employment status (.117), provides 2.85 times the influence of the coefficient for age (.041). The variable with the greatest Influence, student classification (.854), has 20.83 times more Influence than does the coefficient for the age variable (.041). The inclusion of the age variable, along with the remaining eight variables which were discriminating, resulted in the ability of the function to successfully predict group membership 69.7 percent of the time for Group I and 62.5 percent of the time for Group II (Table 13).

5. The Individual's academic ability Is a variable which
does not discriminate when used in the prediction of
group membership In either Groups I or I1.
This null hypothesis was rejected. Academic ability, as measured by the Scholastic Aptitude Test score, or Its equivalent, was one of the nine variables which resulted from the discriminant analysis. The correspondIng coefficient for academic ability was .287 at the .05 level of significance and was fourth out of nine discriminating variables. The coefficient for the next highest discriminating variable, feminine standard score (-.368), provides 1.28 times the Influence of the coefficient for academic ability (.287). The variable with the greatest Influence, student classification (.854), has 2.98 times more Influence than does the coefficient for the academic ability variable (.287). The inclusion of the academic






69


ability variable, along with the remaining eight discriminating variables, resulted In the successful classification of 69.7 percent of cases In Group I and 62.5 percent of the cases in Group II (Table 13).















CHAPTER V
CONCLUSIONS AND RECOMMENDATIONS



The purpose of this study was to determine If the characteristics of masculinity, femininity, androgyny, academic ability and age could be used to predict whether women students In the College of Business Administration at the University of Central Florida would remain In that college or transfer to another college within the University.

Data was collected from four groups of students at the University of Central Florida. The two major groups were Group I, women students enrolled in the College of Business Administration, and Group II, women students who had transferred out of the College of Business Administration Into another college within the University. Group III, women students enrolled In the College of Education, and Group IV, men students enrolled in the College of Business Administration, were not included for the purpose of predicting retention; these two groups were used to address questions surrounding the concept of the traditionally sex-typed person, such as: (1) Did women students majoring In education, a traditionally feminine field of study, produce a more feminine score on the Bem Sex-Role Inventory than did the other groups in the study? (2) Did men students majoring In business administration, a traditionally masculine field of study, produce a more masculine score on the Bem Sex-Role Inventory than did the other groups In the study?









Eleven variables for students in Groups I and II were subjected to a discriminant analysis technique and nine of these variables surfaced as discriminating at the .05 level of significance. Listed according to their relative Influence In the prediciton of group membership (either Group I or Group II), the nine discriminating variables were, (I) student classification, (2) marital status, (3) Bem Sex-Role Inventory feminine standard score, (4) Scholastic Aptitude Test score (5) Bern Sex-Role Inventory difference standard score, (6) employment status, (7) age, (8) grade point average and (9) Bem Sex-Role Inventory masculine standard score. Although age, academic ability, masculinity, femininity, and androgyny (as determined through the difference score on the Bem Sex-Role Inventory) were variables that surfaced as discriminating In the analysis, only femininity (feminine standard score) realized a coefficient with relatively strong Influence on the prediction of group membership.

Analyzing of data on the nine discriminating variables for a new woman student, who as a freshman selected business administration as

her major field of study, will enable a prediction to be made regarding the probability of her remaining in the College of Business Administration. A classification made as a result of the discriminant analysis of her data that would put her in the group which would indicate she would tend to remain In the College of Business Administration could be said to be correct 69.7 percent of the time. A classification made as a result of the discriminant analysis of her data which put her in the group that would Indicate she would tend to transfer out of the College of Business Administration could be said to be correct 62.5 percent of the time. This











conclusion is drawn because group membership was predictable in 69.7 percent of the cases classified in Group I, women business majors, and 62.5 percent of the cases in Group II, women who had transferred out of business administration. The combination of the nine discriminating variables in a discriminant analysis made these predictions possible.

Students in Group I, women students enrolled in the College of Business Administration, and Group II, women students who had transferred out of the College of Business Administration, were similar in age, marital status, and employment status. Groups I and II also were similar in academic ability, as measured by grade point averages and Scholastic Aptitude Test scores or their equivalents.

Students in Group I!, women students who had transferred out of the College of Business Administration, apparently had not made the transfer for reasons of poor academic ability. Thirty-nine out of forty students In Group II maintained an over-all grade point average of 2.0 or better. Those responding to the open-ended question (seventeen out of forty) did not Indicate that their transfer had anything to do with the business admintstration courses being too difficult for them; however, it is likely that students would not tend to state that course difficulty was a factor in their transferring out of the College of Business Administration. They might have tended to omit an answer to the question rather than state that the work was too difficult. The poor response to the open-ended question leaves real doubt as to the students' reasons for transferring out of Bustness Administration.










From Dr. Sandra Bem's research on sex role characteristics, the

assumption has been made that those people with androgynous traits can be said to have a more balanced personality than do those with strong masculine or strong feminine traits. According to this study, women business majors were classified as androgynous at a slightly higher percentage rate (43.3 percent as opposed to 40.0 percent) than were women who transferred out of the College of Business Administration. In the

comparison groups, Groups III and IV, it was interesting to note that women education majors in Group III had a higher percentage of those classified as androgynous than did any other group. It was also Interesting to note that Group III had the largest percentage of older members (over 25), the largest percentage of married subjects, and the largest percentage of members with children than did any other group. It seems reasonable to conclude that maturity Is a definite factor in women students' self perceptions, which tend to put them into an androgynous classification on the Bern Sex-Role Inventory. Group IV, men business majors, tended to be strongly sex-typed with 62.1 percent classified as masculine, and only 19.0 percent classified as androgynous.

Reconmendat ions

I. It is recommended that similar studies, using the same measurement

Instruments In combination with additional measurement instruments, be

conducted for the purpose of strengthening or expanding the discriminant function and increasing its ability to predict the retention

or transfer of women students in business administration.










2. It is recommended that similar studies, using the same measurement

Instruments In combination with additional measurement Instruments,

be conducted for the purpose of strengthening or expanding a discriminant function and Increasing its ability to predict the retention or

transfer of men students in business administration.

3. Due to the limitations of this study to one Institution of higher

education, it is recommended this study be replicated In other

Institutions.

4. In counseling women students who designate business administration

as a major field of study, counselors should gather and analyze data

on the nine variables cited in this study for the purpose of counseling

these students.

5. Additional studies should be done to determine the reasons students

transfer out of the College of Business Administration.

6. Studies should be done to find answers to the following questions:

Does the business administration curriculum and the teaching of that curriculum tend to attract and retain students with certain sex-role

characteristics? If this Is true, should efforts be made to change

the curriculum and/or the methods of teaching? Would such changes produce students who would be more effective In their jobs in the

business community? Is the business administration curriculum

tailored to be more attractive and palatable to those students who

tend to have more masculine characteristics than androgynous or

feminine characteristics?















APPENDIX A

DEMOGRAPHIC PROFILE DATA SHEET














PERSONAL INFORMATION FORM


Social Security # Phone #


(Last)


(First)


(Middle/Maiden)


Address at which you can be reached:


Spring Quarter, 1980


Summer Quarter, 1980


Bi rthdate:


(Street) (City) (State) (ZIp) (Street) (City) (State) (Zip)


Age:


Marital Status: S Employment outside the home:


ingle/Divorced


Marrled


Widowed


'Full Time Part Time Unemployed


Occupation:


Number of Children


Grade Point Average (Overall College Work, Including Junior College, if Applicable)


Student Classification: Freshman


Sophomore


Junior


Senior


(Circle one)

College at U C F in which you are currently enrolled If you have been enrolled In a college at U C F other than the one in which you are now enrolled, please list those college In order of Enrollment


High School from which you graduated:


Year


OVER


Name:


IIIIIIIIIII


I I


III II II






77


Personal Information Form: (continued)

If you are currently enrolled in the College of Business Administration, do not answer the following question on this Personal Information Form.

If you were enrolled in the College of Business Administration
after January 1, 1979, and subsequently have transferred to
another college at UCF, please Indicate below your reasons
for the transfer.















APPENDIX B

LETTER TO GROUPS I AND II














May 6, 1980


Dear

One of our primary goals is to lend assistance to students enrolled at the UnLversity of Central Florida as they work toward the successful completion of their academic programs. As part of our effort to assist you, we need to gather information that will enable us to develop and offer programs that will help you reach your educational and professional goals.

You have been selected as part of a random sample of women students; consequently, your participation will be critical to the accuracy of this study. Please be assured that your Involvement will be treated confidentially and will be used only In combination with others to
compare responses.

Information will be gathered during the end of Spring Quarter. We would appreciate your taking about 10 to 15 minutes of your time while on campus to come by Room 282 in the Administration Building any time during the hours posted below. In return we would be happy to share with you a
summary of the research results and our plans to provide programs to enhance your professional development.

Thursday, May 22 10 a.m. to 3 p.m. Room 280, Conference Room Friday, May 23 10 a.m. to 3 p.m. Room 280, Conference Room Tuesday, May 27 10 a.m. to 3 p.m. Room 280, Conference Room Wednesday, May 28 10 a.m. to 3 p.m. Room 280, Conference Room Thursday, May 29 10 a.m. to 3 p.m. Room 280, Conference Room

Sincerely,


Carol P. Wilson
Dean of Women

CPW:o















APPENDIX C

INSTRUCTIONS FOR THE BEM SEX-ROLE INVENTORY AND
THE BEM SEX-ROLE INVENTORY









BEM INVENTORY


Developed by Sandra L. Bem,Ph.D.


Age Sex


Phone No. or Address Date 19


If a student: School


Yr. in School


If not a student: Occupation


DIRECTIONS

On the opposite side of this sheet, you will find listed a number of personality characteristics. We would like you to use those characteristics to describe yourself, that Is, we would like you to indicate, on a scale from I to 7, how true of you each of these characteristics Is. Please do not leave any characteristic unmarked.


Example: sly
Write a Write a Write a Write a Write a Write a Write a


never or almost never true that you are sly.
usually not true that you are sly. sometimes but Infrequently true that you are sly. occasionally true that you are sly. often true that you are sly. usually true that you are sly. always or almost always true that you are sly.


Thus, If you feel It is sometimes but infrequently true that you are "sly," never or almost never true that you are "malicious," always or almost always true that you are "irresponsible," and often true that you are "carefree," then you would rate these characteristics as follows:


Sly 3i
Ma IcI o IUs I


CONSULTING PSYCHOLOGISTS PRESS, INC.
577 College Avenue Palo Alto, California 94306

Copyright, 1978, by Consulting Psychologists Press, Inc. All rights reserved. Duplication of this form by any process is a violation of the copyright laws of the United States except when authorized in writing by the Publisher.


Name












Never or almost never true


I
Usually
not true


Sometimes but Infrequently
true


| I
Occasionally Often
true true


Usual ly true


Always or almost always true


Defend my own beliefs Affectionate Conscientious Independent Sympathetic

Moody
Assertive Sensitive to needs
of others Reliable Strong personality Understanding Jealous Forceful Compassionate Truthful Have Leadership
abilities
Eager to soothe
hurt feelings


p - ~*


Adaptable


Flatterable


Domi nant Theatrical Tender Self-Sufficient Conceited Loyal Willing to take a Happy
stand
Love Children Individualistic Tactful Soft-spoken Aggressive Unpredictable

Gentle Masculine Conventional Gullible Self-reliant Solemn Yielding Competitive
Helpful Childlike Athletic Likeable Cheerful Ambitious Unsystematic Do not use harsh language


Analytical


Sincere


Secretive Shy Act as a leader Willing to take risks Inefficient Feminine Warm Make decisions easily Friendly


R.S.


S.S.















APPENDIX D

ADMINISTRATION AND SCORING GUIDE FOR
THE BEM SEX-ROLE INVENTORY








ADMINISTRATION AND SCORING GUIDE FOR THE BEM SEX-ROLE INVENTORY*

Sandra Llpsltz Bem, Ph.D.
Cornell University

INTRODUCTION

Both In psychology and In society at large, femininity and masculinity have long been comceptualized as opposite ends of a single bipolar dimension. More recently, however, scholars in a number of disciplines have begun to concern themselves with the concept of psychological androgyny, a term that denotes the Integration of femininity and masculinity within a single Individual. The concept of psychological androgyny implies that it is possible for an Individual to be both compassionate and assertive, both expressive and instrumental, both feminine and masculine, depending upon the situational appropriateness of these various modalities; and It further Implies that an Individual may even blend these complementary modalities In a single act, being able, for example, to fire an employee if the circumstances warrant It but with sensitivity for the human emotion that such an act inevitably produces.

Before empirical research on the concept of psychological androgyny could be initiated, however, It was necessary to develop a new type of sex-role Inventory, one that would not automatically build In an Inverse relationship between femininity and masculinity as previous Inventories had done. On most inventories, Items are empirically defined as feminine or masculine on the basis of differential endorsement by females and males, and a person filling out the inventory Is said to be either feminine or masculine as a function of which sex she or he most closely resembles. Although It Is possible for a person to earn a score that falls halfway between the two extremes and thereby reveal that she or he does not closely resemble either sex, a person cannot make the strong statement on such an Inventory that she or he Is androgynous.

The Bem Sex-Role Inventory (BSRI) was designed to Implement empirical research on psychological androgyny. It contains 60 personality characteristics, 20 of which are stereotypically feminine (e.g., affectionate, gentle, understanding, sensitive to the needs of others), and 20 of which are stereotypically masculine (e.g., ambitious, self-reliant, Independent, assertive). The BSRI also contains 20 characteristics that serve as filler Items (e.g. truthful, happy, conceited). When taking the BSRI, a person Is asked to indicate on a 7-point scale how well each of the 60 characteristics describes herself or himself. The scale ranges from 1 ("never or almost never true") to 7 ("Always or almost always true") and Is labeled at each point.



*This material Is distributed free with orders for the Bem Sex-Role Inventory pending publication of a complete Manual for the test. Copyright 1979, by Consulting Psychologist Press, Inc., 577 College Avenue, Palo Alto, CA 94306.










The BSRI has two features distinguishing It from most masculinityfemininity scales. The BSRI treats femininity and masculinity as two independent dimensions rather than as two ends of a single dimension, thereby enabling a person to indicate whether she or he is high on both dimensions ("androgynous"), low on both dimensions ("undifferentiated") or, high on one dimension bult-Tow on the other (either "feminine" or "mascul ine").

In addition, the BSRI Is based on a conception of the traditionally sex-typed person as someone who Is highly attuned to cultural definitions of sex-appropriate behavior and who uses such definitions as the Ideal standard against which her or his own behavior Is to be evaluated. In this view, the traditionally sex-typed person Is motivated to keep her or his behavior consistent with an idealized Image of femininity or masculinity, a goal she or he presumably accomplishes both by selecting behaviors and attributes that enhance the Image and by avoiding behaviors and attributes that violate the image. Accordingly, Items were selected as feminine or masculine on the basis of cultural definitions of sextyped social desirability and not on the basis of differential endorsement by females and males, I.e., a characteristic qualified as feminine If it was judged to be traditionally more desirable In American society for a woman than for a man, and It qualified as masculine If It was judged to be more desirable in American society for a man than for a woman.

For a full discussion of the development of the BSRI, the reader should consult The Measurement of Psychological Androgyny by Sandra L. Bem, Journal of Consulting and Clinical Psychology, 1974, Vol. 42, pp. 155-162.

ADMINISTRATION AND SCORING

The BSRI is essentially self-administering and may be given to large groups as well as to Individuals. It has also been administered by mail in several studies. The inventory consists of 60 adjectives and phrases printed on a single sheet with Instruction and space for personal Information about the subject on the reverse side. Although It has been used primarily with college students and adults, the items in the BSRI should be comprehensible to most upper elementary school children.

The test is labeled simply "Bem Inventory" to reduce the possibility that responses might be influenced by a knowledge of the purpose of the scales. Most subjects complete the original inventory In 15 minutes or less and the short form in 10 minutes. The test Is arranged so that the 30 short form Items appear first and, where time is limited, subjects may be Instructed to stop after the item "conventional" or at the heavy black line in the middle column.

It may be helpful to remind subjects orally that it Is Important that they not skip any Items. Questions about the meaning of any item should be answered In as neutral a fashion as possible. It Is recommended that lead pencils with erasers be used so that any changes In responses will be clearly made to facilitate accurate scoring.









Handscoring the BSRI Is a relatively simple clerical task which may be facilitated by use of a pocket calculator. The first step is the calculation of each subject's Femininity ("a") and Masculinity ("b") scores, which are the averages of subject's ratings of the feminine and masculine adjectives on the BSRI. That is a given subject's Femininity Score Is the mean of that subject's ratings on the feminine adjectives, and that same subject's Masculinity Score Is the mean of her or his ratings on the masculine adjectives. The placement of adjectives on the BSRI is as follows:
1) The first adjective and every third one thereafter is masculine.
2) The second adjective and every third one thereafter is feminine.
3) The third adjective and every third one thereafter is filler.

Users may wish to make templates for the two scales by punching holes In blank answer sheets so that only the feminine or masculine ratings are
visible at any one time.

In any event, total the sum of the ratings for either scale and divide by the number of items rated. Unless the subject has omitted Items, the divisor will be 20 for the original form or 10 for the short form, but it Is important to correct the divisor when Items have been omitted.

The average of the ratings for the Femininity scale Is entered in box for the "a" raw score (RS) at the bottom of the answer sheet. The average of tl7-hasculinity ratings is entered In the box labeled "b".

Now look up the standard scores (SS) for each of the raw scores using Table 1, page 14,* and enter them in the SS boxes under the raw score boxes. The T-scores in Table I are adjusted so that females and males are equally represented. Be careful to use the correct column depending upon whether you have administered the original or the short form.

Next obtain the Difference Score by subtracting the "b" SS from the "a" SS. Be sure to retain the correct sign, plus (+) for positive and minus (-) for negative. High scores In either direction Indicate a tendency to be strongly sex-typed (or sex-reversed), positive scores Indicating femininity and negative scores Indicating masculinity.

After obtaining the a-b score, consult Table 2 to find the Standard Score for the difference. Note that one column in Table 2 gives the Short
Form SS's and one gives the Original Form SS's.

Now you may return to the upper (raw score) boxes and classify each subject in one of four categories based on the relationship between the median raw scores of the normative sample and the subject's scores on each scale.


*Instructions for scoring by computer using punched cards are available without charge from the Publisher.










Masculinity Score
Below Median Above Median


Femininity
Score


Below Median


Above Med I an


Based on the normative data on Stanford students (sexes combined), the medians are as follows:


Original Form


Short Form


Femininity RS Masculinity RS


4.90 4.95


5.50

4.80


For a discussion of the significance of the classification, see Bem, S. L. On the Utility of Alternative Procedures for Assessing Psychological Androgyny. Journal of Consulting and Clinical Psychology, 1977, Vol. 45, pp. 196-205, as well as the article referred to on Page 2.



*The normative data In Tables I and 2 Is based on samples of Stanford University undergraduates (476 males and 340 females) weighted to equalize the sex distribution.


Und ifferent I ated Masculine TTv-Iow) 7T&. fem.-high nasc.


Feminine Androgynous "Figh fem.-low masc.) "'hIgh fem.-high masc.)








Table 1

T-Scores for Femininity and Masculinity
Based on the 1978 Stanford Sample (Sexes Combined) N-816


Femininity(a) Mascullnlty(b) Femininity(a) Masculinlty(b) Stand. Score Stand. Score Stand. Score tand. Score
Raw R:aw
Orig Short Score Short Orig Orig Short core Short Orig
-15 -5 1.00 2 -8 36 33 4.00 40 36
-11 1.05 -7 37 4.05 37
-13 -4 1.10 3 -7 38 34 4.10 41 38
-12 1.15 -6 39 4.15 38
-11 -3 1.20 5 -5 39 35 4.20 42 39
-i1 1.25 -4 40 4.25 40
-10 -2 1.30 6 -4 41 36 4.30 43 40
-9 1.35 -3 42 4.35 41
-9 0 1.40 7 -2 43 38 4.40 45 42
-7 1.45 -1 44 4.45 43
-6 8 1.50 8 -1 45 39 4.50 46 43
-5 1.55 0 45 4.55 44
-5 2 1.60 10 1 46 40 4.60 47 45
-4 1.65 1 47 4.65 46
-3 3 1.70 11 2 48 41 4.70 48 46
-2 175 3 49 4.75 47
-1 5 1.80 12 4 50 43 4.80 50 48 0 1.85 4 51 4.85 49 1 6 1.90 13 5 51 44 4.90 51 49 1 1.95 6 1 52 4.951 50 2 7 20 15 7 53 45 5.001 52 51 3 2.05 7 54 5.05 51 4 8 2.10 16 8 55 46 5.10 53 52 5 2.15 9 56 5.15 53 6 10 2.20 17 10 56 48 5.20 55 54 6 2.25 10 57 5.2 54 7 11 2.30 18 11 58 49 5.30 56 55 8 2.35 12 59 5.35 56 9 12 2.40 20 13 60 50 5.40 57 57 10 2.45 13 61 5.45 57 S 14 2.50 21 14 61 52 5-W0 58 58 12 2.55 15 62 5.55 59 13 15 2.60 22 15 63 53 5.60 60 60 13 2.65 16 64 5.65 61 14 16 2.70 23 17 65 54 5.701 61 61











Table 1-continued


Femlnlnlty(a) Mascullnlty(b) Feminlnlty(a) Mascullnlty(b) Stand. Score tand. Score Stand. Score Stand. Score
Raw Raw
Orlh Short Scor Short Or IShg T5 2.75 U 6So 5.75 b2 16 17 2.80 25 18 67 55 5.80 62 63 17 2.85 19 67 5.85 63 17 19 2.90 26 20 68 57 5.90 63 64 18 2.95 21 69 5.95 65 19 20 3.00 27 21 70 5 6.00 65 65 20 3.05 22 71 6.05 66 21 21 3.10 28 23 72 59 6.10 66 67 22 3.15 24 73 6.15 68 23 22 3.20 30 24 73 60 6.20 67 68 23 3.25 25 74 6.25 69 24 24 3.30 31 26 75 6 30 68 70 25 3.35 26 76 6.35 71 26 25 3.40 32 27 77 63 6.40 70 71 27 3.45 28 78 6.45 72 27 26 3.50 33 29 78 64 6.50 71 73 28 3.55 29 79 6.55 74 29 27 3.60 35 30 80 65 6.60 72 74 30 3.65 31 81 6.65 75 31 28 3.70 36 32 82 67 6.70 73 76 32 3.75 32 83 6.75 76 33 30 3.80 37 33 84 68 6.80 75 77 34 3.85 34 84 6.85 78 34 31 3.90 38 35 85 69 6.90 76 79 35 3.95 35 86 6.95 79 87 71 7.00 77 80








Table 2

T-Scores for the Femininity minus Masculinity Difference


Short Diff. Orig. Short Diff. Orig. Form SS Form Form SS Form T-Score (a-b) T-Score T-Score (a-b) T-Score

12 -50 17 51 +1 51 13 -49 17 52 +2 51 13 -48 18 52 +3 52 14 -47 19 53 +4 53 15 -46 19 54 +5 53 16 -45 20 55 +6 54 17 -44 21 55 +7 55 17 -43 21 56 +8 55 18 -42 22 57 +9 56 19 -41 23 58 +10 57 20 -40 23 58 +11 57 20 -39 24 59 +12 58 21 -38 25 60 +13 59 22 -37 25 61 +14 59 23 -36 26 61 +15 60 23 -35 27 62 +lb 61 24 -34 27 63 +17 61 25 -33 28 64 +18 62 26 -32 29 64 +19 63 26 -31 29 65 +20 63 27 -30 30 66 +21 64 28 -29 31 67 +22 65 29 -28 31 67 +23 65 29 -27 32 68 +24 66 30 -26 33 69 +25 67 31 -25 33 70 +26 67 32 -24 34 71 +27 68 32 -23 35 71 +28 69 33 -22 35 72 +29 69 34 -21 36 73 +30 70 35 -20 37 74 +31 71 36 -19 37 74 +32 71 36 -18 38 75 +33 72 37 -17 39 76 +34 73 38 -16 39 77 +35 73 39 -15 40 77 +36 74 39 -14 41 78 +37 75 40 -13 41 79 +38 75 41 -12 42 80 +39 76 42 -11 43 80 +40 77










Table 2-continued


Short Diff. Orig. Short Diff. Orig. Form SS Form Form SS Form T-Score (a-b) T-Score T-Score (a-b) T-Score


42 -10 43 81 +41 77 43 -9 44 82 +42 78 44 -8 45 83 +43 79 45 -7 45 83 +44 79 45 -6 46 84 +45 80 46 -5 47 85 +46 81 47 -4 47 86 +47 82 48 -3 48 87 +48 82 48 -2 49 87 +49 83 49 -1 49 88 +50 84 50 0 50













REFERENCES CITED


Bem, S. L. The Measurement of Psychological Androgyny. Journal of
Consulting and Clinical Psychology, 1974, 42(2), 155-162.

Bern, S. L. Beyond Androgyny: Some Presumptuous Prescriptions for a
Liberated Sexual Identity. American Psychological Association-National Institute of Mental Health Conference on the Research
Needs for Women, Madison, Wisconsin, May 31, 1975.

Bem, S. L. Administration and Scoring Guide for the Bem Sex-Role Inventory.
Palo Alto, Calif.: Consulting Psychologist Press, Inc., 1979.

Borgstrom, K. F., Whiteley, S., & Rudolph, P. Today's Undergraduates:
Are They Really Any Different? Journal of National Association of
Women Deans, Administrators and Counselors, Spring, 1980, 3(3), 3-9.

Brovermen, I. Sex-Role Stereotypes: A Current Appraisal. Journal of
Social Issues, 1972, 28, 74-77.

Cross, B. M. The Educated Woman In America. New York: Teachers College
Press, 1965.

Flemming, A. Minorities and Women as Government Contractors. Washington,
D. C.: U. S. Printing Office, May, 1975.

For Women, A Difficult Climb to the Top. Business Week, August 2, 1969,
42-46.

Fretz, C. F., & Hayman, J. Progress for Women--Men are Still More Equal.
Harvard Business Review, September--October, 1973, 51, 133-142.

Herman, J. B., 6 Gyllstrom, K. K. Working Men and Women: Inter- and
Intra-Role Conflict. Psychology of Women Quarterly, Summer, 1978,
1(4), 319-333.

Horner, M. S. Fail: Bright Women. Psychology Today, 1969, 42, 36-38.

Horner, M. S. Achievement-Related Conflicts In Women. Journal of Social
Issues, 1972, 28, 157-175.




Full Text
Tibbetts, S. Sex-Role Stereotyping: Why Women Discriminate Against Them
selves. Journal of National Association of Women Deans, Administrators
and Counselors, Summer, 1975, 38(*0, 177~^83
Tibbetts, S. Elementary Schools: Do They Stereotype or Feminize? Journal
of National Association of Women Deans, Administrators and Counselors,
Pa 11, 1976, 40(1), 27-33-
TomI1Inson-Keasey, C. Role Variables: Their Influence on Female Moti
vational Constructs. Journal of Counseling Psychology, 197**, 21(3),
332-337.
U. S. Bureau of the Census. Statistical Abstract of the United States.
Washington, D. C.: U. S. Government Printing Office, 1979.
Van Heir, E. J. Sexual Discrimination In School Administration Opportunities.
Journal of National Association of Women Deans, Administrators and
Counselors, Summer, 1975, 38(4), 163-167.
Verheyden-Hl11 lard, M. E. Kindergarten: The Training Ground for Women In
Administration. Journal of National Association of Women Deans,
Administrators and Counselors, Summer, 1975, 38(A), 151~155.
Waldman, E. Changes In the Labor Force Activity of Women. Monthly Labor
Review, June, 1970, 10-18.
Webster's Seventh New Collegiate Dictionary. Springfield, Mass., G. £ C.
Merrlam Company, 1971.
Young, C. L. Women In School Administration and Supervision: A New
Leadership Dimension. NASSP Bulletin, May, 1976, 83-85.


66
1. The masculine score derived from the Bern Sex-Role
Inventory Is a variable which does not discriminate
when used In the prediction of group membership In
either Groups I or 11.
This null hypothesis was rejected. The masculine standard score
was one of the nine variables which was shown to be discriminating in the
discriminant analysis. The corresponding coefficient for the masculine
standard score was -.023 at the .05 level of significance and was ninth
out of nine discriminating variables. The coefficient for the next
highest discriminating variable, grade point average (-.027), provides
1.17 times the Influence of the coefficient for the masculine variable
(-.023). The variable with the greatest Influence, student classifi
cation (.85*0, has 37.13 times more influence than does the coefficient
for the masculine variable (-.023). The masculine variable, combined
with the remaining eight variables In the discriminant function, resulted
in the successful classification of 69.7 percent of the subjects In
Group I and 62.5 percent of the subjects In Group II (Table 13).
2. The feminine score derived from the Bern Sex-Role Inven
tory Is a variable which does not discriminate when
used In the prediction of group membership In either
Groups I or II.
This null hypothesis was rejected. The feminine standard score was
one of the nine variables that was shown to be discriminating In the dis
criminant analysis. The coefficient for the feminine standard score was
-.368 at the .05 level of significance and was third out of nine discri
minating variables. The coefficient for the next highest discriminating
variable, marital status (-.632), provides 1.71 times the influence of
the coefficient for the feminine variable (-.368). The variable with
the greatest Influence, student classification (.85*0, has 2.32 times more


61
Table 16
Classification of Subjects In
Groups I and II Utilizing
Nine Discriminating Variables
Subjects
Actual Group
Membershlp
Predicted Group
Membership
Discriminant
Scores***
1
1 *
1
0.8A78
2
1
1
1.7815
3
1
2****
-0.2951
A
1
1
1.0657
5
1
2****
-0.0610
6
1
1
1.0075
7
l
1
0.6711
8
1
1
0.5212
9
1
1
1.3583
10
1
2****
-0.6170
11
1
2****
-0.113A
12
1
1
1.6095
13
1
2****
-0.3881
1A
1
1
0.3993
15
1
1
0.8615
16
l
1
0.5013
17
1
1
0.1766
18
1
2****
-0.2285
19
1
1
0.9171
20
1
1
1.51A2
21
1
2****
-0.9658
22
1
1
1.3706
23
1
1
0.735A
2A
1
1
1.A973
25
1
1
O.89AI
26
1
2****
-0.0366
27
1
2****
-O.A38A
28
1
1
1.1789
29
1
1
1.5886
*Women students enrolled In the College of Business Administration.
**Women students who had transferred out of the College of Business
Administration.
***Scores obtained when coefflcents are applied to the nine discri
minating variables. Zero to +1 drives the case to Group I, and
a 0 to -1 drives the case to Group II.
****These subjects were not properly classified In their original group.
NOTE: Age, marital status, employment status, grade point average,
student classification, Scholastic Aptitude Test scores, and
Bern Sex-Role Inventory masculine, feminine and difference scores
were the nine discriminating variables.


26
Women are currently much more undecided about their majors
than are their male counterparts and even substantially
more than women In past years, perhaps reflecting some
confusion about the increased options available to them.
While showing significant decreases in areas traditionally
associated as being "women's majors" (English, fine arts,
social sciences) as well as the less traditional math and
statistics, "undecided" women students have shown increased
Interest in professional fields such as business, engineer
ing and the health professions. (Borgstrom, Whiteley and
Rudolph, 1980, pp. 6-7)
In a recent study by Pulg-Casauranc, women In three fields of
academic study were Investigated. The Edwards Personal Preference
Schedule (EPPS) and the Bern Sex-Role Inventory (BSRl) were used to
assess the relationship between personality characteristics and psycho
logical androgyny for females in traditional (female-dominated), non-
traditional (male-dominated) and neutral (dominated by neither men
nor women) fields of academic study. A random sample of 185 women from
Junior and senior classes of traditional, nontradttional and neutral
academic departments participated in the study.
The data Indicated that the three groups could also be
differentiated by their psychological androgyny scores.
Significant differences for the sample group found the
nontradltional group demonstrated the highest degree of
psychological androgyny, although all three groups fell
within the range defined as psychologically androgynous.
(Pulg-Casauranc, 1977, p. 5)
The most significant finding of the study was that the females in
the three groups thought to be qualitatively different, tended to be
more similar than disparate when It came to displaying and internalizing
behaviors they perceived as appropriate (Pulg-Casauranc, 1977, p. 7).


22
working class. The typical working woman of today Is married and
half of them are over 39 years of age. Chances are that the working
women of today are supplementing the family Income (Waldnan, 1970,
pp. 10-18).
In an attempt to eliminate the myths about women managers,
research has been conducted to provide Insight Into the similarities
and dissimilarities of men and women managers. As opposed to a pre
dominantly psychological approach to women's attitudes, values, and
beliefs and how these factors cause them to behave In a business
situation, Relf and Newstrom utilized an approach based on women
managers' perceptions of their work environment. The purpose was to
determine If their views of the formal and Informal aspects of organi
zations are different from the views of men. If this was true, a
possible conclusion would be that women should be treated differently
from men. The methodology centered on the semantic differential
technique, which measures perceptions along two independent dimensions,
evaluative and potency. A questionnaire was administered to 286 men
and 55 women In management development programs In business and govern
ment organizations. It was concluded that women and men managers are
more similar than dissimilar In their feelings about their work climate.
Based on this, Relf and Newstrom concluded "that decisions made about
women on the basis of their sex, without considering such individual
factors as background, education, experience, personality and potential,
are likely to be wrong" (Relf and Newstrom, 1975, p. 78).
Recent studies have yielded contradictory results concerning
male and female attitudes and behavior as they work In organizations.


68
This null hypothesis was rejected. The age of the participants was
one of the nine variables that surfaced as being discriminating. The
corresponding coefficient for age was .0**1 at the .05 level of signifi
cance and was seventh out of nine discriminating variables. The coeffi
cient for the next highest discriminating variable, employment status
(.117), provides 2.85 times the Influence of the coefficient for age
(.041). The variable with the greatest Influence, student classification
(.85*0, has 20.83 times more Influence than does the coefficient for the
age variable (.0**1). The Inclusion of the age variable, along with the
remaining eight variables which were discriminating, resulted in the
ability of the function to successfully predict group membership 69.7
percent of the time for Group I and 62.5 percent of the time for Group II
(Table 13).
5. The Individual's academic ability Is a variable which
does not discriminate when used In the prediction of
group membership In either Groups I or II.
This null hypothesis was rejected. Academic ability, as measured by
the Scholastic Aptitude Test score, or Its equivalent, was one of the nine
variables which resulted from the discriminant analysis. The correspond
ing coefficient for academic ability was .287 at the .05 level of signifi
cance and was fourth out of nine discriminating variables. The coefficient
for the next highest discriminating variable, feminine standard score
(-.368), provides 1.28 times the Influence of the coefficient for academic
ability (.287). The variable with the greatest Influence, student classi
fication (.85**), has 2.98 times more Influence than does the coefficient
for the academic ability variable (.287). The Inclusion of the academic


Student Classifications
In Groups I, II, III and IV
Student
Classifications
Group Ia
Percentages of Groups
Group II*3 Group lllC
Group IVd
Freshman
0
10.0
0
0
Sophomore
0
12.5
0
0
Junior
24.2
45.0
69.6
9.5
Senior
75.8
32.5
30.4
90.5
aWomen students enrolled In the College of Business Administration.
^Women students who had transferred out of the College of Business
Administration Into another college within the University.
S/omen students enrolled In the College of Education.
aMen students enrolled In the College of Business Administration.


5^
The analysts aspects of this technique provide several tools
for the interpretation of data. Among these are statistical
tests for measuring the success with which the discriminating
variables actually discriminate when combined Into the discri
minant functions....The result Is often of major theoretical
significance, and statistical tests are Included for this pur
pose. Since the discriminant functions can be thought of as
the axes of a geometric space, they can be used to study the
spatial relationships among the groups. . More importantly,
the weighting coefficients can be interpreted much as In
multiple regression or factor analysis. In this respect, they
serve to Identify the variables which contribute most to dif
ferentiation along the respective dimension (function). (Me,
1975, PP. A35-436)
The discriminant analysis was used to statistically distinguish be
tween Group I, women students enrolled in the College of Business Adminis
tration, and Group II, women students who transferred out of the College
of Business Administration.
The analysis resulted In a linear combination of independent vari
ables (discriminant function) that tended to maximize the separation of
the two groups. The rule for determining the number of equations in a
discriminant analysis is to utilize one function less than the ninber of
groups In the study, or to use the number of equations (discriminant
functions) equal to the number of discriminating variables, whichever Is
smaller. Consequently, one equation (discriminant function) was derived
In this study because two groups were being studied.
When testing for statistical significance of the discriminating vari
ables, a significance level of .05 was found.
The direct method was used In order to select the independent variables
which would contribute to the discriminant analysis. All eleven of the pos
sible independent variables were entered Into the analysts concurrently,
and the discriminant function was created directly from the entire set of


CHAPTER 1
INTRODUCTION
The governing clause of Title IX of the 1972 Education Amendments
to the Civil Rights Act of 1964 sets the stage for eliminating sex
discrimination. The law, which was originally introduced in 1971 as
an amendment to the Civil Rights Act of 1964, presents Its purpose in
Its opening statement as follows:
No person in the United States shall, on the basis of
sex, be excluded from participation in, be denied the
benefits of, or be subjected to discrimination under
any education program or activity receiving Federal
financial assistance. . (Taylor and Schavlik, 1975
p. 5.)
No one can be excluded from participating in any academic, occu
pational training, or other educational program within institutions
using Federal funds. "The practices that are specifically prohibited
by Title IX as well as other practices that maintain stereotypic sex
roles should be examined" (Taylor and Schavlik, 1975 p. 15). Further,
"Any special services that are provided to help students overcome the
effects of previous sex-role socialization should be described and eval
uated by those responsible for them" (Taylor and Schavlik, 1975, p. 28).
Examples are counselling services offered by professional counselors,
clinical and consulting psychologists, psychiatrists, academic counselors
or advisers, residence hall counselors, and admissions counselors.
1


APPENDIX C
INSTRUCTIONS FOR THE BEM
AND
THE BEM SEX-ROLE
SEX-ROLE INVENTORY
INVENTORY


73
From Dr. Sandra Bern's research on sex role characteristics, the
assumption has been made that those people with androgynous traits can
be said to have a more balanced personality than do those with strong
masculine or strong feminine traits. According to this study, women
business majors were classified as androgynous at a slightly higher per
centage rate (43.3 percent as opposed to 40.0 percent) than were women
who transferred out of the College of Business Administration. In the
comparison groups, Groups III and IV, It was Interesting to note that
women education majors In Group III had a higher percentage of those
classified as androgynous than did any other group. It was also interesting
to note that Group III had the largest percentage of older members (over
25), the largest percentage of married subjects, and the largest percen
tage of members with children than did any other group. It seems reasonable
to conclude that maturity Is a definite factor In women students' self
perceptions, which tend to put them Into an androgynous classification
on the Bern Sex-Role Inventory. Group IV, men business majors, tended to
be strongly sex-typed with 62.1 percent classified as masculine, and only
19.0 percent classified as androgynous.
RecommendatIons
I. It Is recommended that similar studies, using the same measurement
Instruments In combination with additional measurement Instruments, be
conducted for the purpose of strengthening or expanding the discri
minant function and Increasing Its ability to predict the retention
or transfer of women students In business administration.


65
When all the subjects In Group I were subjected to the classifica
tion process, 69.7 percent of them were properly placed In Group I;
consequently, It was possible to predict group membership for Group I,
women business majors, 69.7 percent of the time.
When all the subjects In Group II were subjected to the classifi
cation process, 62.5 percent of them were properly placed In Group II;
consequently, It was possible to predict group membership for Group II,
women students who transferred from the College of Business Administra
tion to another college In the University, 62.5 percent of the time.
For any new woman student wishing to enter as a freshman In the
College of Business Administration from whom data on the nine discri
minating variables could be gathered, and for whom a discriminant score
could be computed, the classification process could be performed. If
she were placed In Group I as a result of her discriminant score, It
could be predicted at a 69.7 percent level of probability that she would
remain enrolled In the College of Business Administration. If as a
result of the classification process, the same new woman student was
placed In Group II, It could be predicted at a 62.5 percent level of
probability that she would transfer out of the College of Business
Administration.
Test of Hypotheses and Results
The hypotheses of this study were that an Individual's membership
In either Group I or Group II Is predictable on the basis of age, academic
ability, and the Bern Sex-Role Inventory characteristics. The following
sub-hypotheses In the null form served as the basis for making predictions
of group membership.


Table 10
k7
Responses by Group 11 Students to the
Four Categories Derived from the
Open-ended Question on the
Demographic Profile Date Sheet
Categories
Number of Responses
(N=17)
Indecision as to Field of Study 3
Change In Career Interests 3
Expediency of Completion of Degree In
Another Field of Study U
Boredom or Lack of Enthusiasm for the
Business Curriculum 7
Seventeen of the kO students In Group II responded to the question.
Note: Open-ended question was, "If enrolled in the College of Business
Administration after January, 1979, and subsequently transferred to
another college at the University of Central Florida, please Indicate
in the space provided your reasons for the transfer."


n
Bern Sex-Role Inventory: An Inventory consisting of sixty personality
characteristics, twenty of which are judged to be socially desirable
traits for men, twenty of which are Judged to be socially desirable
traits for women and twenty characteristics which serve as filler
Items.
Masculine characteristics: A set of characteristics on the Bern Sex-
Role Inventory judged to be more desirable In American society for a
man than for a woman.
Feminine characteristics: A set of characteristics on the Bern
Sex-Role Inventory Judged to be more desirable In American society
for a woman than for a man.
Androgyny: A score on the Bern Sex-Role Inventory which Indicates
that the Individual Is not strongly self-identified with either mas
culine or feminine characteristics.
Group 1: Those women students enrolled in the College of Business
Administration at the University of Central Florida during the Spring
Quarter, 1980, who have completed ninety (90) quarter hours of college
study Including at least two (2) courses In business administration.
Group II: Those women students enrolled at the University of Central
Florida during the Spring Quarter, 1980, who have taken at least two
(2) courses In business administration, and who have transferred out of
the College of Business Administration at some point between January,
1979 and June, 1980.
Age: The actual age of women students In Groups ( and II as stated
on their demographic data sheets.


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
THE CHARACTERISTICS OF MASCULINITY,
FEMININITY, ANDROGYNY, AGE, AND ACADEMIC ABILITY
AS FACTORS IN THE RETENTION OF WOMEN STUDENTS
IN COLLEGE PROGRAMS IN BUSINESS ADMINISTRATION
By
Carol Popejoy Wilson
August 1981
Chairman: Dr. Glenna Carr
Major Department: Curriculum and Instruction
The purpose of this study was to analyze the effects of age, aca
demic ability, masculinity, femininity, and androgyny on the retention
of women students in a program of business administration. One hundred
and fifty-nine students enrolled at the University of Central Florida
during the spring quarter of 1979 were Included in the study. There
were four groups; the two major groups were Group I, women students
enrolled in the College of Business Administration, and Group II, women
students who had transferred out of the College of Business Administration.
Group lit, women education majors, and Group IV, men business adminis
tration majors, were used as comparison groups to see if their masculine,
feminine and androgyny scores on the Bern Sex-Role Inventory would differ
greatly from those in Groups I and I!.
vil


77
Personal Information Form: (continued)
If you are currently enrolled In the College of Business Acknlnistration,
do not answer the following question on this Personal Information Form.
If you were enrolled in the College of Business Administration
after January 1, 1979, and subsequently have transferred to
another college at UCF, please Indicate below your reasons
for the transfer.


3
5. As late as the 1950's and 1960's, women were still
suffering under Freud's description of women as being
Inferior.
6. During the 1950's and 1960's there was the "utter
perfidy" of the mass media toward homemakers.
7. Passage of the Equal Rights Amendment, which was written
and Introduced in Congress in the early 1920's, was not
strongly supported by any of the organized women's groups
until the late 1960's and early 1970*s (Mueller, 1977,
pp. A3-A5).
The focus of this study is to determine if the characteristics
of masculinity, femininity, androgyny, age and academic ability of
women students are significant influences on the retention of women
students In business administration programs at the post secondary
level. Is there a correlation between these factors and the re
tention of women students in business administration, which is con
sidered to be a stereotypically nontraditional field of study for
women?
A stereotype is defined as an "unvarying form or pattern, having
no Individuality, as though cast from a mold" (Webster, 1971, p. 860).
When stereotype Is applied to sex roles, the definition would indicate
all women are alike In certain respects, just as all men are alike
and they both possess a certain degree of fixed or conventional ex
pressions, notions, character and mental patterns.


13
Before empirical research on the concept of psycho
logical androgyny could be Initiated, however, It was
necessary to develop a new type of sex-role Inventory, one
that would not automatically build In an Inverse relation
ship between femininity and masculinity as previous In
ventories had done. On most Inventories, Items are empiri
cally defined as feminine or masculine on the basis of
differential endorsement by females and males, and a person
filling out the Inventory Is said to be either feminine or
masculine as a function of which sex she or he most closely
resembles. Although It Is possible for a person to earn a
score that falls halfway between the two extremes and there
by reveal that she or he does not closely resemble either
sex, a person cannot make the strong statement on such an
Inventory that she or he Is androgynous.
The Bern Sex-Role Inventory (BSRI) was designed to
Implement empirical research on psychological androgyny.
It contains 60 personality characteristics, 20 of which are
stereotypically feminine (e.g., affectionate, gentle, under
standing, sensitive to the needs of others), and 20 of
which are stereotypically masculine (e.g., ambitious, self-
reliant, Independent, assertive). The BSRI also contains
20 characteristics that serve as filler Items (e.g., truth
ful, happy, conceited). When taking the BSRI, a person Is
asked to Indicate on a 7-polnt scale how well each of the
60 characteristics describes herself or himself. The' scale
ranges from 1 ("Never or almost never true") to 7 ("Always
or almost always true") and Is labeled at each point.
The BSRI has two features distinguishing It from most
masculinity-femininity scales. The BSRI treats femininity
and masculinity as two Independent dimensions rather than
as two ends of a single dimension, thereby enabling a per
son to Indicate whether she or he Is high on both dimensions
("androgynous"), low on both dimensions ("undifferentiated")
or, high on one dimension but low on the other (either
"femlntne" or "masculine").
In addition, the BSRI is based on a conception of the
traditionally sex-typed person as someone who Is highly
attuned to cultural definitions of self-appropriate be
havior and who uses such definitions as the Ideal standard
against which her or his own behavior Is to be evaluated.
In this view, the traditionally sex-typed person Is moti
vated to keep her or his behavior consistent with an
Idealized Image of femininity or masculinity, a goal she or
he presumably accomplishes both by selecting behaviors and
attributes that enhance the Image and by avoiding behaviors
and attributes that violate the Image. Accordingly, Items


REFERENCES CITED
Bern, S. L. The Measurement of Psychological Androgyny. Journal of
Consulting and Clinical Psychology, 1974, 42(2), 155162.
Bern, S. L. Beyond Androgyny: Some Presumptuous Prescriptions for a
Liberated Sexual Identity. American Psychological Association
National Institute of Mental Health Conference on the ResearcfT"
Needs for Women, Madison, Wisconsin, May 31 1975
Bern, S. L. Administration and Scoring Guide for the Bern Sex-Role Inventory.
Palo Alto, Calif.: Consulting Psychologist Press, Inc., 1979-
Borgstrom, K. F., Whlteley, S., 6 Rudolph, P. Today's Undergraduates:
Are They Really Any Different? Journal of National Association of
Women Deans, Administrators and Counse lors, Spring, 19&0, 43(3),"T-9.
Brovermen, I. Sex-Role Stereotypes: A Current Appraisal. Journal of
Social Issues, 1972, 28, 74-77.
Cross, B. M. The Educated Woman In America. New York: Teachers College
Press, 19^51
Flemming, A. Minorities and Women as Government Contractors. Washington,
D. C.: U. t. Printing Office, May, )9?5.
For Women, A Difficult Climb to the Top. Business Week, August 2, 1969,
42-1*6.
Fretz, C. F., 6 Hayman, J. Progress for WomenMen are Still More Equal.
Harvard Business Review, SeptemberOctober, 1973 5J_* 133" 142.
Herman, J. B., £ Gyllstrom, K. K. Working Men and Women: Inter- and
Intra-Role Conflict. Psychology of Women Quarterly, Summer, 1978,
U4), 319-333.
Horner, M. S. Fall: Bright Women. Psychology Today, 1969, 42, 36-38.
Horner, M. S. Achievement-Related Conflicts in Women. Journal of Social
Issues, 1972, 28, 157-175.
92


72
conclusion is drawn because group membership was predictable in 69.7 per
cent of the cases classified in Group I, women business majors, and 62.5
percent of the cases in Group II, women who had transferred out of business
administration. The combination of the nine discriminating variables in
a discriminant analysis made these predictions possible.
Students in Group I, women students enrolled in the College of Busi
ness Administration, and Group II, women students who had transferred out
of the College of Business Administration, were similar in age, marital
status, and employment status. Groups I and II also were similar in aca
demic ability, as measured by grade point averages and Scholastic Aptitude
Test scores or their equivalents.
Students in Group II, women students who had transferred out of the
College of Business Administration, apparently had not made the transfer
for reasons of poor academic ability. Thirty-nine out of forty students
in Group II maintained an over-all grade point average of 2.0 or better.
Those responding to the open-ended question (seventeen out of forty) did
not indicate that their transfer had anything to do with the business
administration courses being too difficult for them; however, it is likely
that students would not tend to state that course difficulty was a factor
in their transferring out of the College of Business Administration. They
might have tended to omit an answer to the question rather than state that
the work was too difficult. The poor response to the open-ended question
leaves real doubt as to the students' reasons for transferring out of
Business Administration.


lowest Is I00. Forty-five and two tenths percent of Group I scored over
1,000, as compared to 32.5 percent of Group II (Table 7). Scholastic
Aptitude Test scores or their equivalents were not compiled for the sub
jects In Groups III and IV. These groups were Included In the study for
the purpose of comparing results on the Bern Sex-Role Inventory (Table 7)*
Number of Children
The women students In the College of Education (Group III) reported
having the grestest number of children; 3^.7 percent had from one to three
children. The greatest number of children reported by any one participant
was six. A majority of each group had no children (Table 8).
Year of High School Graduation
Participants In the study reported the year of their graduation from
high school on the demographic Information sheet. The range was from 19^3
to 1979 with 77 percent of the participants having graduated since 1973
(Table 9).
Analysis of Open-ended Question
Three professionals with expertise In the counselling of university
students were selected to review and categorize the open-ended question on
the demographic profile data sheet. Each reviewer had a minimum of ten
years of academic and vocational counselling experience In Institutions of
higher education.
The open-ended question was, "If enrolled In the College of Business
Administration after January, 1979, and subsequently transferred to another
college at the University of Central Florida, please Indicate In the space
provided your reasons for the transfer."


91
Table 2-contlnued
Short
Diff.
Orlg.
Short
Diff.
Orlg.
Form
SS
Form
Form
SS
Form
T-Score
(a-b)
T-Score
T-Score
(a-b)
T-Score
42
-10
1*3
81
+41
77
43
-9
44
82
+42
78
44
-8
45
83
+43
79
45
-7
45
83
+44
79
45
-6
46
84
+45
80
A6
-5
47
85
+46
81
~TT
-4
~FT
5T~
+47
5T"
48
-3
48
87
+48
82
48
-2
49
87
+49
83
49
-1
49
88
+50
84
50
0
50


ADMINISTRATION AND SCORING GUIDE FOR THE BEM SEX-ROLE INVENTORY*
Sandra Lipsltz Bern, Ph.D.
Cornell University
INTRODUCTION
Both In psychology and In society at large, femininity and masculinity
have long been conceptualIzed as opposite ends of a single bipolar dimension.
More recently, however, scholars In a number of disciplines have begun to
concern themselves with the concept of psychological androgyny, a term that
denotes the Integration of femininity and masculinity within a single indi
vidual. The concept of psychological androgyny Implies that it is possible
for an Individual to be both compassionate and assertive, both expressive
and Instrumental, both feminine and masculine, depending upon the situa
tional appropriateness of these various modalities; and it further implies
that an individual may even blend these complementary modalities in a
single act, being able, for example, to fire an employee if the circum
stances warrant it but with sensitivity for the human emotion that such an
act inevitably produces.
Before empirical research on the concept of psychological androgyny
could be Initiated, however, it was necessary to develop a new type of
sex-role Inventory, one that would not automatically build in an Inverse
relationship between femininity and masculinity as previous Inventories
had done. On most Inventories, lterns are empirically defined as feminine
or masculine on the basis of differential endorsement by females and males,
and a person filling out the inventory Is said to be either feminine or
masculine as a function of which sex she or he most closely resembles.
Although It is possible for a person to earn a score that falls halfway
between the two extremes and thereby reveal that she or he does not closely
resemble either sex, a person cannot make the strong statement on such an
inventory that she or he is androgynous.
The Bern Sex-Role Inventory (BSRl) was designed to Implement empirical
research on psychological androgyny. It contains 60 personality character
istics, 20 of which are stereotypically feminine (e.g., affectionate,
gentle, understanding, sensitive to the needs of others), and 20 of which
are stereotypically masculine (e.g., ambitious, self-reliant, Independent,
assertive). The BSRl also contains 20 characteristics that serve as filler
items (e.g. truthful, happy, conceited). When taking the BSRl, a person
is asked to indicate on a 7Point scale how well each of the 60 characteris
tics describes herself or himself. The scale ranges from 1 ("never or
almost never true") to 7 ("Always or almost always true") and is labeled
at each point.
*This material is distributed free with orders for the Bern Sex-Role In
ventory pending publication of a complete Manual for the test. Copyright
1979, by Consulting Psychologist Press, Inc., 577 College Avenue, Palo
Alto, CA 94306.
84


36
Table 2
Ages of Students
In Groups I, II, III and IV
Percentages
of
Ages of
Group Ia
Group II*5
Group lllc
Group IV^
Students
(N33)
(N-40)
(N-23)
(N-63)
18 to 2k
78.8
80.0
65.2
7k.S
25 to 3k
15.1
10.0
25.9
17.5
35 to 55
6.0
10.0
8.6
8.0
tornen students enrolled In the College of Business Administration.
^Women students who had transferred out of the College of Business
Administration Into another college within the University.
S/omen students enrolled In the College of Education.
^Men students enrolled In the College of Business Administration.


86
Handscoring the BSRI Is a relatively simple clerical task which may
be facilitated by use of a pocket calculator. The first step Is the
calculation of each subject's Femininity ("a") and Masculinity ("b")
scores, which are the averages of subject's ratings of the feminine and
masculine adjectives on the BSRI. That Is a given subject's Femininity
Score Is the mean of that subject's ratings on the feminine adjectives,
and that same subject's Masculinity Score Is the mean of her or his
ratings on the masculine adjectives. The placement of adjectives on the
BSRI Is as follows:
1) The first adjective and every third one thereafter Is masculine.
2) The second adjective and every third one thereafter Is feminine.
3) The third adjective and every third one thereafter Is filler.
Users may wish to make templates for the two scales by punching holes
In blank answer sheets so that only the feminine or masculine ratings are
visible at any one time.
In any event, total the sum of the ratings for either scale and divide
by the number of Items rated. Unless the subject has omitted Items, the
divisor will be 20 for the original form or 10 for the short form, but it
Is important to correct the divisor when Items have been omitted.
The average of the ratings for the Feminlnlty scale Is entered In
box for the "a11 raw score (RS) at the bottom of the answer sheet. The
average of the Masculinity ratings Is entered In the box labeled "b".
Now look up the standard scores (SS) for each of the raw scores using
Table 1, page l,* and enter them In the SS boxes under the raw score boxes.
The T-scores In Table 1 are adjusted so that females and males are equally
represented. Be careful to use the correct column depending upon whether
you have administered the original or the short form.
Next obtain the Difference Score by subtracting the "b" SS from the
"a" SS. Be sure to retain the correct sign, plus (+) for positive and
minus (-) for negative. High scores In either direction Indicate a ten
dency to be strongly sex-typed (or sex-reversed), positive scores Indicating
femininity and negative scores Indicating masculinity.
After obtaining the a-b score, consult Table 2 to find the Standard
Score for the difference. Note that one column In Table 2 gives the Short
Form SS's and one gives the Original Form SS's.
Now you may return to the upper (raw score) boxes and classify each
subject In one of four categories based on the relationship between the
median raw scores of the normative sample and the subject's scores on
each scale.
Instructions for scoring by computer using punched cards are available
without charge from the Publisher.


The Bern Sex-Role Inventory was used to measure students1 masculine,
feminine, and androgyny scores. A demographic profile data sheet and
university records were used to gather additional data. All data for the
four groups were compared through cross tabulation, and data on subjects
In Group I and II were analyzed and classified through a discriminant
analysis.
A discriminant function composed of nine discriminating variables at
the .05 level of significance resulted from the analysis. The variables
are, In the order of their importance, (1) student classification, (2)
marital status, (3) feminine standard score, (4) Scholastic Aptitude Test
score, (5) difference standard score, (6) employment status, (7) age,
(8) grade point average, and (9) masculine standard score.
The major conclusion was that given data on the nine variables listed
above for a new woman student who has selected business administration as
a major, a prediction concerning her staying in the business administration
curriculum can be made with a 69.7 percent accuracy rate, and a prediction
of her transferring out of business administration can be made with a 62.5
percent chance of accuracy.
viii


7
If boys are supported and Indeed encouraged In this be
havior for all of their formative years (boys will be
boys), there Is no way to expect them to suddenly turn
about as adults and see female peers as equals. Women
have remained, and will remain, the little girls that
society allowed and encouraged little boys to shove aside
so that they could play with the boys without Interference
. . and they're still doing it when they're 45.
(Verheyden-Hi11 lard, 1975, p. 151)
According to Hs. Verheyden-Hi11 lard, "The message conveyed to
girls Is that boys have a right to exclude them and that they are too
weak to compete anyway. The girls retreat to the girls games and the
female careers where they have been forcefully told they belong"
(Verheyden-Hi11iard, 1975, p. 152).
Girls and women apparently lack strong success models to follow.
Universities are more lacking In such role models than are elementary
schools. The closer to the top of the academic ladder, the greater
the preponderance of males. Statistics for secondary schools shbw
this trend.
... In the eleven-year periods between 1950-51 and
1960-61 the number of women serving as junior high and
senior high school principals dropped from 18% to 3.8%
for all secondary schools . for the same eleven year
periods the percentages of women elementary principals
decreased by nearly 20%. (Van Heir, 1975 p. 163)
More recent statistics cited by the National Center for Education
Statistics revealed the State of Florida as having 23 percent women
In principal and assistant principalshlps (National Center for Education
Statistics, 1978, p. 19).
It Is recognized that women who enter institutions of higher
education today are more likely to choose careers tradition
ally dominated by men than were women ten years ago. However, dif
ferences in choice of career between men and women have not been


APPENDIX B
LETTER TO GROUPS I AND II


Year of High School Graduation
for Students In Groups I, II, III, and IV
Year of Graduation from High School
1943 to 1959 i960 to 1969 1970 to 1979
Number Percent Number Percent Number Percent
Group Ia
2
6.1
2
6.0
29
87.9
b
Group 11
3
7.5
3
7.5
34
85.0
Group 11 lC
2
8.6
3
12.9
18
78.0
Group IV^
3
4. 8
7
11.2
53
84.2
aWomen students
enrol led
in the
Col lege of
Business
Administration,
^Women students who had transferred out of the College of Business
Administration into another college within the University.
S/omen students enrolled in the College of Education.
Men students enrolled in the College of Business Administration.


39
Grade Point Average
The range of grade point averages In college work for all groups was
from 1.87 to 3.99 on a 4.00 scale. Grade point averages were determined
from the students' permanent university records. The women students In
Group II had the lowest grade point averages; twenty-three percent of this
group had grade point averages of 2.49 or below. In all of the groups
44.6 percent to 56.3 percent had a grade point average of 3.0 or better In
their college work (Table 5). No special minimum grade point average was
required for admission to the College of Business Administration or the
College of Education; however, a minimum of 2.0 overall grade point aver
age was required for the admission of all degree seeking students (Table
5).
Student Classification
The information on student classification was collected In categories
of freshmen, sophomore, junior, and senior. Seventy-seven and five tenths
percent of the students In Group II, women students who transferred out of
the College of Business Administration, were classified at the junior level
or above. All the participants In Groups I, III, and IV were classified
at the junior level or above (Table 6). At the University of Central
Florida students are permitted to enter the college of their choice at the
beginning of their freshman year.
Scholastic Aptitude Test Scores
Scholastic Aptitude Test scores or their equivalent, as measured on
the American College Test or the Florida Twelfth Grade test, for Groups I
and II were taken from the permanent university records. The scores
ranged from 590 to 1350. The highest score possible Is 1600 and the


CHAPTER II
REVIEW OF THE RESEARCH
The purpose of this review was to study sex-role stereotyping as
It related to women In business and education. Research reviewed
concerned the questions, what Is sex-role stereotyping, how has it
evolved, does it exist today, and if so, how does it affect student
performance.
Conflicts from Sex-Role Stereotyping
The conflicts which result from some sex-role stereotyping
have been studied by many researchers. One of the most noted studies
is Matina Horner's "Toward an Understanding of Achievement Related
Conflicts in Women" (Horner, 1972). Horner tested freshman and sopho
more students, male and female, at a large mid-western university.
These students responded to a cue added to the end of the Standard
Thematic Apperceptive Test (TAT). The cue is "After first term finals,
Anne finds herself at the top of the medical school class." Horner
concluded that approximately "65$ of the girls were disconcerted,
troubled, or confused by the cue" (Horner, 1972, p. 162). She further
concluded that a fear of success imagery dominates responses from
women students and is relatively absent in male responses, and that
conflicts emerge between role expectations and aspirations due to
17


5
work was Investigated by the National Commission on the Observance of
International Women's Year In 1976. The Commission specifically ad
dressed four areas. "The Commission examined women entering the
skilled crafts and blue collar Jobs; secretaries; women who own their
own businesses; and new ways In which pay rates are being evaluated"
(National Commission on the Observance of International Women's Year,
1976, p. 58). The Commission undertook the following:
They looked Into practices In the construction Industry
to determine how nondiscrimination laws are applied.
Construction work has been a white male preserve for so
long that a "macho" folklore has grown up around It. But
modern technology has made nearly all construction jobs
easier and has brought many of them within the capability
of many women. (National Commission on the Observance of
International Women's Year, 1976, p. 58)
The Commission further Investigated problems that women who
want to go Into business on their own encounter. The Investiga
tion revealed enormous problems for these women. Arthur S. Flemming,
Chairman of the U. S. Commission on Civil Rights, referred to these
problems as "staggering." In a 1975 report he wrote:
Our Investigations reveal that minority and female-owned
firms encounter problems of staggering proportions In
obtaining Information on federal, state and local govern
ment contracting opportunities In time to submit bids,
and In obtaining working capital necessary for effective
marketing and bidding. . .
. . Minority and female entrepreneurs also encounter a
great deal of skepticism regarding their ability to per
form adequately on government contracts. (Flemming,
1975, p. 0
More and more women are entering the labor force. According to
the National Commission on the Observance of International Women's
Year (1976), there Is a need for opportunities for women to enter
nontradltlonal fields of work . that Is, those fields which have


k
The roots of sex-role stereotyping are deep In our culture.
Cross points out how effectively de Tocqueville briefly and con
cisely described male/female stereotypes of nearly 150 years ago.
In no country has such constant care been taken as In
America to trace two clearly distinct lines of action
for the two sexes and to make them keep pace one with
the other, but In two pathways that are always different.
Women took no part In business or In politics; they never
managed the outward concerns of the family. With rare
exceptions they were confined to the quiet circle of
domestic employment. (Cross, 1965, p. vii)
Cross offered a possible explanation of sex roles when she
said,
Americans had simply applied to the sexes the great
principle of political economy which governs the
manufacturers of our age by carefully dividing the
duties of man from those of women In order that the
great work of society may be better carried on.
(Cross, 1965, p. vii1)
Some Improvements have been made In the betterment of women's
role in society since 1835; however, In the evolution and maturation
of the Individual, sex-role stereotyping continues to be reinforced.
For example, studies cited by Tibbets (1976) show problems connected
with sex-role stereotyping:
Stereotypically mascultne traits, such as Independence,
dominance, competition, achievement, drive, leadership,
decisiveness and logic are considered more socially
desirable than are stereotypically feminine attributes,
such as dependence, emotionalism, submissiveness, pas
sivity, Indecisiveness, and lack of logic ... both men
and women agree that for an adult, regardless of sex,
stereotyped male characteristics were "healthier" than
were stereotyped female characteristics. (Tibbets, 1976,
P. 179)
The control that sex-role stereotyping injects Into the work
environment and its effects on the degree of success of women in


85
The BSRI has two features distinguishing it from most masculinity-
femininity scales. The BSRI treats femininity and masculinity as two
independent dimensions rather than as two ends of a single dimension,
thereby enabling a person to Indicate whether she or he Is high on both
dimensions ("androgynous") low on both dimensions ("undifferentiated")
or, high on one dimension but low on the other (either "feminine" or
"masculine").
In addition, the BSRI is based on a conception of the traditionally
sex-typed person as someone who is highly attuned to cultural definitions
of sex-appropriate behavior and who uses such definitions as the ideal
standard against which her or his own behavior is to be evaluated. In
this view, the traditionally sex-typed person Is motivated to keep her or
his behavior consistent with an Idealized Image of femininity or mascu
linity, a goal she or he presumably accomplishes both by selecting be
haviors and attributes that enhance the Image and by avoiding behaviors
and attributes that violate the Image. Accordingly, items were selected
as feminine or masculine on the basis of cultural definitions of sex-
typed social desirability and not on the basis of differential endorse
ment by females and males, i.e., a characteristic qualified as feminine
If it was judged to be traditionally more desirable in American society
for a woman than for a man, and it qualified as masculine If it was judged
to be more desirable In American society for a man than for a woman.
For a full discussion of the development of the BSRI, the reader should
consult The Measurement of Psychological Androgyny by Sandra L. Bern, Journal
of Consulting and Clinical Psychology, 197^* Vol. *2, pp. 155" 162.
ADMINISTRATION AND SCORING
The BSRI is essentially self-administering and may be given to large
groups as well as to Individuals. It has also been administered by mail
in several studies. The inventory consists of 60 adjectives and phrases
printed on a single sheet with instruction and space for personal infor
mation about the subject on the reverse side. Although it has been used
primarily with college students and adults, the Items in the BSRI should
be comprehensible to most upper elementary school children.
The test is labeled simply "Bern Inventory" to reduce the possibility
that responses might be Influenced by a knowledge of the purpose of the
scales. Most subjects complete the original Inventory in 15 minutes or
less and the short form in 10 minutes. The test is arranged so that the
30 short form Items appear first and, where time is limited, subjects may
be Instructed to stop after the Item "conventional" or at the heavy black
line in the middle column.
It may be helpful to remind subjects orally that it is Important that
they not skip any items. Questions about the meaning of any Item should be
answered in as neutral a fashion as possible. It is recommended that lead
pencils with erasers be used so that any changes In responses will be clearly
made to facilitate accurate scoring.


! certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and Is fully
adequate, In scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, In scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
'jUioUJA
William H. Drummond
Professor, Instructional Leadership
and Support
This dissertation was submitted to the Graduate Faculty of the Division
of Curriculum and Instruction in the College of Education and to the
Graduate Council, and was accepted as partial fulfillment of the require
ments for the degree of Doctor of Philosophy.
Dean for Graduate Studies and Research


67
influence than does the coefficient for the feminine variable (-.368).
The feminine variable, combined with the remaining eight variables in the
discriminant function, resulted in the successful classification of 69.7
percent of the subjects in Group I and 62.5 percent of the subjects in
Group II (Table 13).
3. The androgyny classification derived from the Bern
Sex-Role Inventory is a variable which does not
discriminate when used in the prediction of group
membership in either Group I or II.
This null hypothesis was rejected. The androgynous classification
was determined through the difference standard score. The difference
standard score was one of the nine variables that was shown to be discri
minating In the discriminant analysis. The corresponding coefficient for
the difference standard score was .158 at the .05 level of significance
and was fifth out of nine discriminating variables. The coefficient for
the next highest discriminating variable, Scholastic Aptitude Test score
(.287), provides 1.82 times the Influence of the coefficient for the dif
ference standard score (.158). The variable with the greatest influence,
student classification (.85*0, has 5.^1 times more Influence than does
the coefficient for the difference score (androgyny) variable (.158).
This difference variable, combined with the remaining eight variables in
the discriminant function, resulted in the successful classification of
69.7 percent of the subjects in Group I and 62.5 percent of the subjects
In Group 11 (Table 13).
k. The age of the individual Is a variable which does not
discriminate when used In the prediction of group mem
bership In either Groups 1 or II.


4o
Table 5
Grade Point Averages for Students
In Groups I, II, III, and IV
0
Percentage of Each Group Within Grade Point Ranges
1.87 to 2.49 2.50 to 2.99 3.00 and Above
Group 1^
15.1
30.2
54.6
Group llc
22.5
32.5
45.0
Group IIId
8.6
34.6
56.3
Group IV
14.4
41.4
44.6
aGrade point averages are computed on a four point scale 4.0 A grade.
^Women students enrolled In the College of Business Administration.
Sfomen students who had transferred out of the College of Business
Administration Into another college within the University.
tornen students enrolled In the College of Education.
eMen students enrolled In the College of Business Administration.


16
The final question, "If you were enrolled In the College of Business
Administration after January 1, 1979* and subsequently have transferred
to another college at U.C.F., please indicate In the space provided
below your reasons for the transfer," provided the investigator the
opportunity to analyze the various reasons for the transfer.
Statement of Hypotheses
The major hypothesis of this study was that an individual's member
ship in either Group 1 or Group II can be predicted on the basis of age,
academic ability, and the Bern Sex-Role Inventory characteristics. The
following sub-hypotheses in the null form were tested.
1. The masculinity score derived from the Bern Sex-Role Inventory
is a variable which does not discriminate when used in the
prediction of group membership in either Groups I or II.
2. The femininity score derived from the Bern Sex-Role Inventory
Is a variable which does not discriminate when used in the
prediction of group membership In either Groups I or II.
3. The androgyny score derived from the Bern Sex-Role Inventory
is a variable which does not discriminate when used In the
prediction of group membership in either Groups I or II.
4. The age of the Individual Is a variable which does not dis
criminate when used in the prediction of group membership In
either Groups I or II.
5. The Individual's academic ability Is a variable which does
not discriminate when used in the prediction of group member
ship In either Groups I or II.


8
substantially reduced. The percentage of women who entered the fields
of physical sciences, biological sciences, business, and engineering
Increased at the same rate as that of men from 1961 to 1971. The
percentage of women choosing these professions Increased from 21
percent to 26 percent. The percentage of men choosing these pro
fessions Increased from 56 to 61 during the same period of time (Peng
and Jaffee, 1979 p. 286). The career fields generally selected by
women continue to be those traditionally dominated by women, such as
education and nursing.
Why does the problem of sex differences In such areas as career
choice persist? There Is general agreement among writers and educa
tors that women continue to limit their own choices and substantially
contribute to their own second-rate status In society. According to
Sylvia-Lee TIbbets,
Women choose to be Inferiornot because they are, nor
necessarily because they want to bebut because: (a) they
have been taught to believe they are or should be Inferior;
(b) they are afraid to appear unfemlnlne; or (c) they
are not fully aware of their situation and do not realize
that they are being treated as second class citizens or
that they do have a legitimate complaint. (Tibbetts,
1975, p. 178)
Rather than holding women entirely responsible for their position
In society, It would be more accurate to describe women as "victims"
of a society steeped In sex-role stereotyping. When Individuals do
not realize that a situation exists, there Is nothing they can do to
Improve It; however, once a woman becomes aware of the sexist nature
of our society, she becomes responsible for the Improvement of her
own position In society. For example, her recognition of the possibility
for people, and especially for her, to be both masculine and feminine


23
Some studies found persons following traditional sex-role stereotypes;
In others, they have not. Powell and Butterfield explored the possi
bility that sex-role Identification Is a more Important variable than
Is gender In an Individual's perceptions of traits desirable for
management personnel. One hundred and ten graduate students with
jobs In the business community and 575 undergraduate business students
participated In the study conducted by Powell and Butterfield. The
students completed the Bern Sex-Role Inventory for both themselves and
for a person whom they believed to be a "good manager."
As expected, Individuals' sex-role Identifications sig
nificantly affected their perceptions of traits desirable
for management personnel, while sex had virtually no effect.
The study concludes that sex-role Identification Is a
variable deserving of further attention, particularly when
sex-related differences are examined. Also, a graduate
woman revealed more masculine traits than feminine In their
self-descriptions, suggesting that a masculine standard
for management may nullify the femininity of women in or
aspiring to management positions. (Powell and Butterfield,
1978, p. 1)
Virginia Schein cited that limited attention has been paid to the
Impact of sex-role stereotypical thinking on the perceived and actual
performance of women In management. Her investigation of prior research
In the areas of sex-role stereotypes and performance perceptions revealed
an over-ridtng theme that sex role stereotyping has a definite and nega
tive impact on the selection of women Into managerial positions. She
further stressed the need for research on the relationship between power
and political behavior and effective managerial performance.


TABLE OF CONTENTS
PAGE
ACKNOWLEDGEMENTS HI
LIST OF TABLES v
ABSTRACT vi I
CHAPTER
ONE INTRODUCTION 1
Need for the Study 9
Statement of the Problem 10
Definition of Terms 10
Delimitations 12
Description of the Instruments 12
Demographic Profile 15
Statement of Hypotheses 16
TWO REVIEW OF THE RESEARCH 17
Conflicts from Sex-Role Stereotyping 17
Evolution of Sex Roles 19
Sex Roles of Women In Business 20
Sex Roles of Women in Education 24
THREE PROCEDURES USED IN CONDUCTING THE STUDY 27
FOUR ANALYSIS OF DATA 31
Analysis of Descriptive Data 33
Analysis of Open-ended Question 42
Descriptive Analysis of Bern Sex-Role Inventory. 46
Discriminant Analysis Findings 53
Classification of Cases 64
Test of Hypotheses and Results 65
FIVE CONCLUSIONS AND RECOMMENDATIONS 70
APPENDICES
A DEMOGRAPHIC PROFILE DATA SHEET 76
B LETTER TO GROUPS I AND II 79
C INSTRUCTIONS FOR THE BEM SEX-ROLE INVENTORY AND
THE BEM SEX-ROLE INVENTORY 8l
D ADMINISTRATION AND SCORING GUIDE FOR THE BEM
SEX-ROLE INVENTORY 84


55
variables, with two variables being rejected or not contributing to the
discriminant function. The direct method puts all possible variables In
at one time, as opposed to a stepwise procedure wherein possible variables
are entered one at a time.
The direct method of entry was selected to produce a discriminant
function that would, to the best of its ability, statistically distin
guish between Groups I and il.
Information concerning the methods used for assigning values to the
eleven variables Is discussed below, and the methods used for assigning
values to each of the nine discriminating variables which surfaced are
indicated by footnotes in Table 13.
When data are being collected, a process of measurement Is used in
order to assign a value or score to the observed phenomenon or variable.
In this study an ordinal level of measurement was used for some of the
variables. When it was possible to rank order a category according to
some criteria, an ordinal level of measurement was used. For Instance
the order of freshman, sophomore, junior, and senior in the student clas
sification variable was ranked 1, 2, 3 with freshman being 1 and senior
being 4. The ordinal level of measurement assumes a reasoned order for
value or score assignment, but does not require a known distance between
ranks or scores (Nle, 1975, p. 5).
"The Interval level of measurement has an additional property in that
the distances between the categories are defined In terms of fixed and equal
units" (Nie, 1975, p. 5). Scholastic Aptitude Test scores, for instance,
indicate the same or fixed distance between 800 and 900 or 1000 and 1100.


CHAPTER V
CONCLUSIONS AND RECOMMENDATIONS
The purpose of this study was to determine If the characteristics
of masculinity, femininity, androgyny, academic ability and age could
be used to predict whether women students In the College of Business
Administration at the University of Central Florida would remain In that
college or transfer to another college within the University.
Data was collected from four groups of students at the University of
Central Florida. The two major groups were Group I, women students en
rolled In the College of Business Administration, and Group II, women
students who had transferred out of the College of Business Administration
Into another college within the University. Group III, women students
enrolled In the College of Education, and Group IV, men students enrolled
In the College of Business Administration, were not Included for the pur
pose of predicting retention; these two groups were used to address ques
tions surrounding the concept of the traditionally sex-typed person, such
as: (1) Did women students majoring in education, a traditionally feminine
field of study, produce a more feminine score on the Bern Sex-Role Inventory
than did the other groups In the study? (2) Did men students majoring In
business administration, a traditionally masculine field of study, produce
a more masculine score on the Bern Sex-Role Inventory than did the other
groups In the study?
70


29
9.The Investigator administered the Bern Sex-Role Inventory
(Appendix C).
10. Responses to the final question on the demographic profile
data sheet, "If you were enrolled In the College of Business
Administration after January, 1979, and subsequently have
transferred to another college at the University of Central
Florida, please Indicate In the space provided your reasons
for the transfers," were reviewed and categorized by a panel
of knowledgeable professionals with expertise In the coun
selling of students. The purpose of categorizing responses
was to enable the Investigator to cite the main reasons
Indicated for transferring out of the College of Business
Administration and to compare these responses with the
findings of the study. It was recognized that those students
In Group II may perceive their reasons for transferring out
of the College of Business Administration In many different
ways. This question was Included only to give additional
Information concerning reasons for transferring out of the
College of Business Administration. A check was made to
determine If the reasons did or did not tend to support
the major questions posed In this study.
11. As a means of providing comparative data on the masculinity,
femininity and androgyny scores, a sample of junior and senior
male students enrolled In selected classes In the College of
Business Administration at the University of Central Florida
(Group IV), and a sample of women students enrolled In selected


18
previously Imposed stereotypes (Horner, 1972, p. 162). The expec
tancy that success In achievement related situations will be followed by
negative consequences arouses fear of success In otherwise achievement
oriented women which then Inhibits their performance and levels of
aspiration.
In an earlier study, Horner hypothesized that the motive to avoid
success Is significantly greater In women than It Is In men (Horner,
1969). She further hypothesized that this tendency Is even greater In
high achievement oriented, high ability women who are capable of success
than It Is In women at the other extreme; that Is, lower achievement
oriented or lower ability women who cannot and will not seek to achieve.
Horner's research was based on 178 students who were administered the
Standard Thematic Apperceptive Test. Her findings suggest that most
women will exercise their full potential only In a non-competltlve situa
tion, especially If they are competing with men. This tendency Is stronger
among women with an Intense anxiety about success. One of the most
significant findings of this research Is the complexity of achievement
motivation In women. Her results Indicated that fear of success was
found to be aroused In situations In which there was concern over or
anxiety about competitiveness and Its aggressive overtones. When these
anxieties occurred, women experienced conflict and adjusted their be
haviors to their Internalized sex-role stereotypes.
The effects of the various traditional roles performed by women,
such as wife and mother, have been studied. One researcher examined


28
3. The ages of women students In Groups 1 and II were obtained
from the student's permanent record and official University
data tapes.
k. Scholastic Aptitude Test scores, American College Test
scores or Florida Twelfth Grade Test scores were obtained
from the student's permanent record and official University
data tapes. The scores were converted to a common base
utilizing the State University System Equivalency Docunent
(Jackson, 1977).
5. In the Spring Quarter of 1980, following the assignment of
students to classes, a list of all women students In Groups
I and II was compiled from University registration Information.
6. Demographic data for Groups I and II were obtained from a
demographic profile data sheet developed and administered
by the Investigator (Appendix A).
7. A letter (Appendix B) was mailed to each person selected,
requesting her cooperation In completing the Bern Sex-Role
Inventory which was administered In a conveniently located
room on carpus.
8. When a large enough sanple was not produced by the procedure
In #7, follow-up telephone calls, extending a second Invl-
tatton to participate In the study, were made with careful
attention being given to avoid pressuring students during
the telephone conversation. A minimum of fifty percent
participation by women students contacted In Groups I and II
was sought.


64
Classification of Cases
The use of discriminant analysis as a classification technique
comes after the initial computation. Once a set of variables
Is found which provides satisfactory discrimination for the
cases with known group memberships, a set of classification
functions can be derived which will permit the classification
of new cases with unknown memberships. (Nte, Hull, Jenkins,
Steinbrenner and Brent, p. 436)
If variables are found which do well in predicting to which group the
subjects belong, then these variables can be used to predict the likeli
hood of group membership for other students.
As a check on how well the equation (discriminant function) worked,
the subjects were placed In either Group I or Group II according to the
discriminant scores on the nine discriminating variables. Group member
ship was predicted, based on the nine discriminating variables. The goal
of the classification process as It relates to a discriminant analysis
was to measure the success of the equation (discriminant function) by ob
serving how many subjects were properly classified into their groups.
Twenty-three subjects out of thirty-three subjects in Group I, women
business majors, were properly classified. Twenty-five subjects out of
forty subjects in Group II, women students who transferred out of the
College of Business Into another college within the University, were
properly classified into their group (Table 16).
The data for Groups I and II were subjected to the classification
process. When the coefficients were applied to the nine discriminating
variables, and a discriminant score of less than zero to minus one re
sulted, the subject was placed in Group II. A discriminant score of more
than zero to plus one resulted in the subject's being placed In Group I
for classification purposes.


List of Tables-contlnued
Table
13. Standardized Discriminant Function Coefficients At
the .05 Level of Significance Presented In Order of
Importance to the Discriminant Function and the
Levels of Measurement Used In the Assignment of Values
to the Discriminating Variables 56
]k. Means and Standard Deviations for Discriminating
Variables at .05 Level of Probability for Groups
I and II 58
15. Means and Standard Deviations for the Bern Sex-
Role Inventory Masculine, Feminine and
Difference Standard Scores for Groups I and II 59
16. Classification of Subjects in Groups I and II
Utilizing Nine Discriminating Variables 61
vl


12
Academic Ability: Scores on the Scholastic Apptltude Test, American
College Test or the Florida Twelfth Grade Test which have been con
verted to a corrmon score utilizing the 1977 State University System
of Florida Equivalency Document.
Delimitations
All of the subjects In this study were enrolled In the College
of Business Administration at the University of Central Florida
between January, 1979, and June, 1980.
There was no attempt to make inferences about the entire female
student population In the State University System of Florida or the
entire student population at the University of Central Florida.
Inferences were confined to the stated population Included In the
study.
Description of the Instruments
The Bern Sex-Role Inventory Is described by Dr. Sandra Llpsltz
Bern as follows:
Both In psychology and In society at large, femininity
and masculinity have long been conceptualized as opposite
ends of a single bipolar dimension. More recently, how
ever, scholars In a number of disciplines have begun to
concern themselves with the concept of psychological
androgyny, a term that denotes the Integration of femi
ninity and masculinity within a single Individual. The
concept of psychological androgyny Irrplles that It is
possible for an Individual to be both compassionate and
assertive, both expressive and Instrumental, both feminine
and masculine, depending upon the situational appropriate
ness of these various modalities; and it further Implies
that an Individual may even blend these complimentary
modalities in a single act, being able, for example, to
fire an employee If the circumstances warrant It but with
sensitivity for the human emotion that such an act Inevit
ably produces.


21
top executive levels. There was noted, however, a new trend for
women graduates who are getting job offers that were closed to those
women who graduated Just a few years before them. In the opinion of Ms.
Grimm, the first woman vice-president of Carson, Price, Scott and Co.,
"A woman who Is determined to play In a game so fixed, had better be
prepared to look like a girl, act like a lady, think like a man and work
like a dog" (For Women, a Difficult Climb to the Top, 1969, p. **5).
Fretz and Hayman conducted a survey with twenty business organi
zations, ten of which are Industrial companies from among the first
100 companies on the Fortune "500" list. The organizations that
participated In the survey employed nearly two million people. The
sample organizations Included women at all levels of employment.
Thirty six percent of the persons employed In the twenty business
organizations that participated In the survey were women. However, a
women at the managerial and professional levels accounted for less than
1 percent of the total number of employees. Fretz and Hayman con
cluded from their survey that women are hlndred by their own precon
ceived Ideas of what their roles In business should be (Fretz and
Hayman, 1973, p. 13*0.
One of only a few available studies with data regarding women in
the work force as early as 1920 Is presented by Elizabeth Waldman.
Most of the data are from the Current Population Survey conducted by
the Bureau of the Census for the Bureau of Labor Statistics. They
relate to the population 16 years of age and over. The typical
v*>rklng woman In 1920 was single, about 28 years of age and from the


Table 13
56
Standardized Discriminant Function Coefficients
at the .05 Level of Significance Presented in Order
of Importance to the Discriminant Function and the Levels
of Measurement Used in the Assignment of Values
to the Discriminating Variables
Discriminating Variables Standardized Discriminant Levels of
(Assigned Values) Function Coefficients* Measurement
Student Classification
(freshman 13 1, sophomore 2,
junior 3, senior 4)
0.85^
Ordinal**
Marital Status
(single/divorced 1,
married 2, widowed = 3)
-0.632
Ordinal
Feminine Standard Score from the
Bern Sex-Role Inventory
(actual scores)
-0.368
K-4
Interval
Scholastic Aptitude Test Score
(actual scores)
0.287
Interval
Difference Standard Score from the
Bern Sex-Role Inventory
(actual scores)
0.158
Interval
Employment Status
(full time 1, part time 2
unemployed * 3)
0.117
Ordinal
Age
(actual ages)
0.041
. ****
Ratio
Grade Point Average
(actual averages)
-0.027
Ratio
Masculine Standard Score from the
Bern Sex-Role Inventory
(actual scores)
-0.023
Interval
*The influence of one variable to
the function
In relation to the
Influence of another variable may be made by Ignoring the sign
and comparing coefficients. Zero to +1 drives the case to Group I,
and a 0 to -1 drives the case to Group II.
**0rd!nal level measurements rank order categories according to some
criteria.
***lnterva1 level measurements have fixed and equal units of distance
between categorical ranks.
****Rat!o level measurements have known zero points and have fixed and
equal distances between categorical values.


33
A cross tabulation of variables by groups was performed to describe
these variables as they related to the four groups in the study. These
frequency distributions compared variables for one group to the same
variables for another group. The exact differences were more clearly re
vealed when the percentages were examined because of the difference in
total numbers of subjects in each of the four groups. An analysis of
these four groups is reported in the following narrative discussion and
tables.
Analysis of Descriptive Data
Twelve variables were used in the analysis of descriptive data.
Seven variables from the demographic profile data sheet were pertinent to
the descriptive analysis. The questions on the demographic profile data
sheet Illtcltlng current college of enrollment and previous colleges of
enrollment were used only for verification of assignment of subjects to
groups. The name of the high school attended and occupation of the stu
dent were recorded but were not used In the description of the groups.
The seven variables from the demographic profile data sheet which
were considered pertinent to the descriptive analysis were (1) age, (2)
marital status, (3) employment status, (A) number of children, (5) grade
point average, (6) student's enrollment classification (freshman, sopho
more, junior or senior), and (7) year of high school graduation. Two
variables, the sex of the student and Scholastic Aptitude Test score or
its equivalent, were taken from the student's permanent university record.


32
the concept of the traditionally sex-typed person, such as: (1) Did women
students majoring In education, a traditionally feminine field of study,
produce a more feminine score on the Bern Sex-Role Inventory than did the
other groups In the study? (2) Did men students majoring in business ad
ministration, a traditionally masculine field of study, produce a more
masculine score on the Bern Sex-Role Inventory than did the other groups
In the study?
The Statistical Package for the Social Sciences (SPSS) computer pro
gram was used to analyze the data. A discriminant analysis with signifi
cance at .05 was used to test the data. The discriminant analysis enabled
the researcher to statistically distinguish between groups by selecting
the independent variables on which the groups differed.
The discriminant analysis produced a discriminant function, a re
gression equation, with group membership as the dependent variable and
demographic data along with scores on the Bern Sex-Role Inventory as the
Independent variables. The results statistically indicated to which group
each subject probably belonged. The analysis discriminated between the
two groups (Groups I and II) and predicted which subjects were or should
be In either Groups I or II.
The discriminant analysis was used to predict whether a subject be
longed to Group I or Group II on the basis of the selected independent
variables, and to determine the Importance of these Independent variables
In making optimal assignments of the subjects to the two groups. A more
In-depth discussion of the discriminant analysis appears In the "Discrl-
mlnant Analysis Findings" portion of this chapter.


59
Table 15
Means and Standard Deviations
for the Bern Sex-Role Inventory
Masculine, Feminine and Difference Standard Scores
for Groups I and 11
Discriminating
Variables3
Group I*3
Mean S.D.^
Q
Group 11
Mean S.D.
0
Masculine Standard Score
(range 39 to 69)
52.67
10.69
54.30
9.26
Feminine Standard Score^
(range * 44 to 73)
55.50
11.81
58.43
7.57
Difference Standard Score9
(range 51 to 82)
62.77
23.55
65.28
11.25
Three of the nine variables which surfaced at a .05 level of signifi
cance during the discriminant analysis.
^Women students enrolled in the College of Business Administration.
cWomen students who had transferred out of the College of Business
Administration into another college within the University.
^S.D. is the abbreviation for standard deviation.
0
Standard score computed from responses to twenty masculine personality
characteristics on the Bern Sex-Role Inventory.
Standard score computed from responses to twenty feminine personality
characteristics on the Bern Sex-Role Inventory.
^Standard score computed from the difference between the masculine and
feminine scores on the Bern Sex-Role Inventory.


THE CHARACTERISTICS OF MASCULINITY,
FEMININITY, ANDROGYNY, AGE, AND ACADEMIC ABILITY
AS FACTORS IN THE RETENTION OF WOMEN STUDENTS
IN COLLEGE PROGRAMS IN BUSINESS ADMINISTRATION
BY
CAROL POPEJOY WILSON
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
1981

Copyright 1981
by
Carol Popejoy Wilson

ACKNOWLEDGMENTS
This study Is dedicated to my husband, Skip, and my son, Jay, for
their patience, encouragement, and understanding; and to my parents, Ira
and Dorothy Popejoy, for their constant support and concern.
The writer expresses her appreciation to Dr. Glenna Carr, chairman
of the supervisory committee, for her guidance, support, and encouragement.
The writer also expresses her appreciation to the other committee members:
Dr. Eugene Todd, Dr. Fred Goddard, Dr. Arthur Lewis, and Dr. William
Drummond.
The writer Is especially grateful to her colleagues at the Univer
sity of Central Florida: Dr. Charles Dzuiban, Mr. Tom Peeples, Dr. Gordon
McAleer, Dr. Bill Brown, Dr. LeVester Tubbs, Dr. Sandra Guest, Dr. Ralph
Gunter, and Ms. Phyllis Smith, who assisted the writer in the study.
Hi

TABLE OF CONTENTS
PAGE
ACKNOWLEDGEMENTS HI
LIST OF TABLES v
ABSTRACT vi I
CHAPTER
ONE INTRODUCTION 1
Need for the Study 9
Statement of the Problem 10
Definition of Terms 10
Delimitations 12
Description of the Instruments 12
Demographic Profile 15
Statement of Hypotheses 16
TWO REVIEW OF THE RESEARCH 17
Conflicts from Sex-Role Stereotyping 17
Evolution of Sex Roles 19
Sex Roles of Women In Business 20
Sex Roles of Women in Education 24
THREE PROCEDURES USED IN CONDUCTING THE STUDY 27
FOUR ANALYSIS OF DATA 31
Analysis of Descriptive Data 33
Analysis of Open-ended Question 42
Descriptive Analysis of Bern Sex-Role Inventory. 46
Discriminant Analysis Findings 53
Classification of Cases 64
Test of Hypotheses and Results 65
FIVE CONCLUSIONS AND RECOMMENDATIONS 70
APPENDICES
A DEMOGRAPHIC PROFILE DATA SHEET 76
B LETTER TO GROUPS I AND II 79
C INSTRUCTIONS FOR THE BEM SEX-ROLE INVENTORY AND
THE BEM SEX-ROLE INVENTORY 8l
D ADMINISTRATION AND SCORING GUIDE FOR THE BEM
SEX-ROLE INVENTORY 84

LIST OF TABLES
Table Page
1. Sex of Students In Groups I, II, III and IV 35
2. Ages of Students In Groups I, II, III and IV 36
3. Marital Status of Students In Groups I, II,
III, and IV 37
4. Employment Status of Students In Groups I,
II, III, and IV 38
5. Grade Point Averages for Students In
Groups I, II, III, and IV 40
6. Students Classifications In Groups I, II,
III and IV Al
7. Scholastic Aptitude Test Scores for Students
In Groups I and II 43
8. Number of Students In Groups I, II, III, and
IV with Children 44
9. Year of High School Graduation for Students In
Groups I, II, III, and IV 45
10. Responses by Group II Students to the Four
Categories Derived from the Open-ended
Question on the Demographic Profile Data
Sheet 47
11. Interpretation of Categories Derived from
the Bern Sex-Role Inventory 50
12. Comparison of Undifferentiated, Masculine, Androgynous
and Feminine Bern Sex-Role Inventory Classifications
by Group 51
v

List of Tables-contlnued
Table
13. Standardized Discriminant Function Coefficients At
the .05 Level of Significance Presented In Order of
Importance to the Discriminant Function and the
Levels of Measurement Used In the Assignment of Values
to the Discriminating Variables 56
]k. Means and Standard Deviations for Discriminating
Variables at .05 Level of Probability for Groups
I and II 58
15. Means and Standard Deviations for the Bern Sex-
Role Inventory Masculine, Feminine and
Difference Standard Scores for Groups I and II 59
16. Classification of Subjects in Groups I and II
Utilizing Nine Discriminating Variables 61
vl

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
THE CHARACTERISTICS OF MASCULINITY,
FEMININITY, ANDROGYNY, AGE, AND ACADEMIC ABILITY
AS FACTORS IN THE RETENTION OF WOMEN STUDENTS
IN COLLEGE PROGRAMS IN BUSINESS ADMINISTRATION
By
Carol Popejoy Wilson
August 1981
Chairman: Dr. Glenna Carr
Major Department: Curriculum and Instruction
The purpose of this study was to analyze the effects of age, aca
demic ability, masculinity, femininity, and androgyny on the retention
of women students in a program of business administration. One hundred
and fifty-nine students enrolled at the University of Central Florida
during the spring quarter of 1979 were Included in the study. There
were four groups; the two major groups were Group I, women students
enrolled in the College of Business Administration, and Group II, women
students who had transferred out of the College of Business Administration.
Group lit, women education majors, and Group IV, men business adminis
tration majors, were used as comparison groups to see if their masculine,
feminine and androgyny scores on the Bern Sex-Role Inventory would differ
greatly from those in Groups I and I!.
vil

The Bern Sex-Role Inventory was used to measure students1 masculine,
feminine, and androgyny scores. A demographic profile data sheet and
university records were used to gather additional data. All data for the
four groups were compared through cross tabulation, and data on subjects
In Group I and II were analyzed and classified through a discriminant
analysis.
A discriminant function composed of nine discriminating variables at
the .05 level of significance resulted from the analysis. The variables
are, In the order of their importance, (1) student classification, (2)
marital status, (3) feminine standard score, (4) Scholastic Aptitude Test
score, (5) difference standard score, (6) employment status, (7) age,
(8) grade point average, and (9) masculine standard score.
The major conclusion was that given data on the nine variables listed
above for a new woman student who has selected business administration as
a major, a prediction concerning her staying in the business administration
curriculum can be made with a 69.7 percent accuracy rate, and a prediction
of her transferring out of business administration can be made with a 62.5
percent chance of accuracy.
viii

CHAPTER 1
INTRODUCTION
The governing clause of Title IX of the 1972 Education Amendments
to the Civil Rights Act of 1964 sets the stage for eliminating sex
discrimination. The law, which was originally introduced in 1971 as
an amendment to the Civil Rights Act of 1964, presents Its purpose in
Its opening statement as follows:
No person in the United States shall, on the basis of
sex, be excluded from participation in, be denied the
benefits of, or be subjected to discrimination under
any education program or activity receiving Federal
financial assistance. . (Taylor and Schavlik, 1975
p. 5.)
No one can be excluded from participating in any academic, occu
pational training, or other educational program within institutions
using Federal funds. "The practices that are specifically prohibited
by Title IX as well as other practices that maintain stereotypic sex
roles should be examined" (Taylor and Schavlik, 1975 p. 15). Further,
"Any special services that are provided to help students overcome the
effects of previous sex-role socialization should be described and eval
uated by those responsible for them" (Taylor and Schavlik, 1975, p. 28).
Examples are counselling services offered by professional counselors,
clinical and consulting psychologists, psychiatrists, academic counselors
or advisers, residence hall counselors, and admissions counselors.
1

2
"All who assist students in making educational and career choices
are likely to Influence, consciously or unconsciously, the decisions
of some students" (Taylor and Schavlik, 1975, p. 27).
Without a doubt, the thrust of Title IX should provide an incen
tive and push institutions of higher education in that direction of
eliminating Inequality which Is based on sex-role stereotyping. What
took the American society so long to recognize the stunting effects
sex stereotyping has had on women? According to Kate Mueller, women
had to wait until the 1970's for the women's liberation movement to
awaken the public to women's problems and needs. This wait for women
to begin to ask for equal rights, in Mueller's opinion, can be traced
to several factors.
1. In the 1950's many of the women indicated they were not
too unhappy with their own situations.
2. A women's movement would mean full utilization of birth
control, Including the birth control pill, so that maternal
responsibilities could be carefully planned. It took
awhile for the birth control pill to be accepted.
3. It wasn't until the later 1960's that women's groups
learned how to organize, and how to apply verbal protests.
A. New ways of looking at love and new attitudes and behavior
about sex relations have slowly developed during the last
forty years. Along with these new attitudes and behaviors
came planned parenthood and lower birthrates.

3
5. As late as the 1950's and 1960's, women were still
suffering under Freud's description of women as being
Inferior.
6. During the 1950's and 1960's there was the "utter
perfidy" of the mass media toward homemakers.
7. Passage of the Equal Rights Amendment, which was written
and Introduced in Congress in the early 1920's, was not
strongly supported by any of the organized women's groups
until the late 1960's and early 1970*s (Mueller, 1977,
pp. A3-A5).
The focus of this study is to determine if the characteristics
of masculinity, femininity, androgyny, age and academic ability of
women students are significant influences on the retention of women
students In business administration programs at the post secondary
level. Is there a correlation between these factors and the re
tention of women students in business administration, which is con
sidered to be a stereotypically nontraditional field of study for
women?
A stereotype is defined as an "unvarying form or pattern, having
no Individuality, as though cast from a mold" (Webster, 1971, p. 860).
When stereotype Is applied to sex roles, the definition would indicate
all women are alike In certain respects, just as all men are alike
and they both possess a certain degree of fixed or conventional ex
pressions, notions, character and mental patterns.

k
The roots of sex-role stereotyping are deep In our culture.
Cross points out how effectively de Tocqueville briefly and con
cisely described male/female stereotypes of nearly 150 years ago.
In no country has such constant care been taken as In
America to trace two clearly distinct lines of action
for the two sexes and to make them keep pace one with
the other, but In two pathways that are always different.
Women took no part In business or In politics; they never
managed the outward concerns of the family. With rare
exceptions they were confined to the quiet circle of
domestic employment. (Cross, 1965, p. vii)
Cross offered a possible explanation of sex roles when she
said,
Americans had simply applied to the sexes the great
principle of political economy which governs the
manufacturers of our age by carefully dividing the
duties of man from those of women In order that the
great work of society may be better carried on.
(Cross, 1965, p. vii1)
Some Improvements have been made In the betterment of women's
role in society since 1835; however, In the evolution and maturation
of the Individual, sex-role stereotyping continues to be reinforced.
For example, studies cited by Tibbets (1976) show problems connected
with sex-role stereotyping:
Stereotypically mascultne traits, such as Independence,
dominance, competition, achievement, drive, leadership,
decisiveness and logic are considered more socially
desirable than are stereotypically feminine attributes,
such as dependence, emotionalism, submissiveness, pas
sivity, Indecisiveness, and lack of logic ... both men
and women agree that for an adult, regardless of sex,
stereotyped male characteristics were "healthier" than
were stereotyped female characteristics. (Tibbets, 1976,
P. 179)
The control that sex-role stereotyping injects Into the work
environment and its effects on the degree of success of women in

5
work was Investigated by the National Commission on the Observance of
International Women's Year In 1976. The Commission specifically ad
dressed four areas. "The Commission examined women entering the
skilled crafts and blue collar Jobs; secretaries; women who own their
own businesses; and new ways In which pay rates are being evaluated"
(National Commission on the Observance of International Women's Year,
1976, p. 58). The Commission undertook the following:
They looked Into practices In the construction Industry
to determine how nondiscrimination laws are applied.
Construction work has been a white male preserve for so
long that a "macho" folklore has grown up around It. But
modern technology has made nearly all construction jobs
easier and has brought many of them within the capability
of many women. (National Commission on the Observance of
International Women's Year, 1976, p. 58)
The Commission further Investigated problems that women who
want to go Into business on their own encounter. The Investiga
tion revealed enormous problems for these women. Arthur S. Flemming,
Chairman of the U. S. Commission on Civil Rights, referred to these
problems as "staggering." In a 1975 report he wrote:
Our Investigations reveal that minority and female-owned
firms encounter problems of staggering proportions In
obtaining Information on federal, state and local govern
ment contracting opportunities In time to submit bids,
and In obtaining working capital necessary for effective
marketing and bidding. . .
. . Minority and female entrepreneurs also encounter a
great deal of skepticism regarding their ability to per
form adequately on government contracts. (Flemming,
1975, p. 0
More and more women are entering the labor force. According to
the National Commission on the Observance of International Women's
Year (1976), there Is a need for opportunities for women to enter
nontradltlonal fields of work . that Is, those fields which have

6
traditionally been open to males only. The steady Increase In the
number of women entering the labor force Is reflected In the per
centage Increase during the last four years. Women represented
40.5 percent of the civilian labor force tn 1976. This percentage
Increased to 41.7 percent In 1978, and was up to 41.9 percent through
Hay, 1979. (U. S. Bureau of the Census, 1979, p. 392).
Within the one million divorces which occurred In 1975, only
14 percent of the women Involved were awarded alimony. Generally,
child support payments are less than half enough to support the
children. It Is apparent that more women needed to work in 1979
than did In 1976. Limited availability of employment for women,
because of their being confined to the few "traditional" fields of
work for women such as teaching, nursing and secretarial work, has
been exacerbated by sex-role stereotyping In all phases of American
life (National Commission on the Observance of International Women's
Year, 1976, p. 57).
Students of a 11 ages are not shielded from the effects of sex-
role stereotyping. In fact they are often overtly subjected to such
stereotyping all through school and during their formative stages in
life. Ms. Verheyden-Hl11 lard painted the following picture at recess
time on the playing fields at an elementary school.
"John says: 'Get away, we don't want any girls playing.'
Mary says: 'They won't let me play with them.'
Educator 'Well . that's the way boys are. They
says: are too rough, anyway. Play with the
girls." (Verheyden-Hi11 lard, 1975, p. 151)

7
If boys are supported and Indeed encouraged In this be
havior for all of their formative years (boys will be
boys), there Is no way to expect them to suddenly turn
about as adults and see female peers as equals. Women
have remained, and will remain, the little girls that
society allowed and encouraged little boys to shove aside
so that they could play with the boys without Interference
. . and they're still doing it when they're 45.
(Verheyden-Hi11 lard, 1975, p. 151)
According to Hs. Verheyden-Hi11 lard, "The message conveyed to
girls Is that boys have a right to exclude them and that they are too
weak to compete anyway. The girls retreat to the girls games and the
female careers where they have been forcefully told they belong"
(Verheyden-Hi11iard, 1975, p. 152).
Girls and women apparently lack strong success models to follow.
Universities are more lacking In such role models than are elementary
schools. The closer to the top of the academic ladder, the greater
the preponderance of males. Statistics for secondary schools shbw
this trend.
... In the eleven-year periods between 1950-51 and
1960-61 the number of women serving as junior high and
senior high school principals dropped from 18% to 3.8%
for all secondary schools . for the same eleven year
periods the percentages of women elementary principals
decreased by nearly 20%. (Van Heir, 1975 p. 163)
More recent statistics cited by the National Center for Education
Statistics revealed the State of Florida as having 23 percent women
In principal and assistant principalshlps (National Center for Education
Statistics, 1978, p. 19).
It Is recognized that women who enter institutions of higher
education today are more likely to choose careers tradition
ally dominated by men than were women ten years ago. However, dif
ferences in choice of career between men and women have not been

8
substantially reduced. The percentage of women who entered the fields
of physical sciences, biological sciences, business, and engineering
Increased at the same rate as that of men from 1961 to 1971. The
percentage of women choosing these professions Increased from 21
percent to 26 percent. The percentage of men choosing these pro
fessions Increased from 56 to 61 during the same period of time (Peng
and Jaffee, 1979 p. 286). The career fields generally selected by
women continue to be those traditionally dominated by women, such as
education and nursing.
Why does the problem of sex differences In such areas as career
choice persist? There Is general agreement among writers and educa
tors that women continue to limit their own choices and substantially
contribute to their own second-rate status In society. According to
Sylvia-Lee TIbbets,
Women choose to be Inferiornot because they are, nor
necessarily because they want to bebut because: (a) they
have been taught to believe they are or should be Inferior;
(b) they are afraid to appear unfemlnlne; or (c) they
are not fully aware of their situation and do not realize
that they are being treated as second class citizens or
that they do have a legitimate complaint. (Tibbetts,
1975, p. 178)
Rather than holding women entirely responsible for their position
In society, It would be more accurate to describe women as "victims"
of a society steeped In sex-role stereotyping. When Individuals do
not realize that a situation exists, there Is nothing they can do to
Improve It; however, once a woman becomes aware of the sexist nature
of our society, she becomes responsible for the Improvement of her
own position In society. For example, her recognition of the possibility
for people, and especially for her, to be both masculine and feminine

9
would enable her to express anger, assert her preferences, trust her
own Judgment, and take control of her life situations, whatever they
might be. According to Dr. Sandra Bern,
For fully effective and healthy himan functioning, both
masculinity and femininity must each be tempered by the
other, and the two must be Integrated Into a more balanced,
a more fully human, a truly androgynous personality. An
androgynous personality would thus represent the very
best of what masculinity and femininity have each come to
represent, and the more negative exaggerations of mascu
linity and femininity would tend to be cancelled out.
(Bern, 1975, p. k)
Need for the Study
If sex-role stereotyping Is detrimental to the successful com
pletion of academic programs and the attainment of educational goals
established by women students, an awareness of the problem by educa
tors Is the first step In helping to eliminate the problem. Today's
women need to be made aware of the expanding choices of academic
programs and Job opportunities available to them. A major responsi
bility of the educator Is to assist students In their efforts to
expand their professional horizons, exercise their full potential in
their chosen fields of study and to take advantage of their freedom
to make choices without being hindered by the detrimental effects of
sex-role stereotyping by the Individual and society.
It Is Important for educators and society at large to challenge
the myths that have kept women In their "place." Daughters, as well
as sons, have to believe that they can achieve their dreams and per
form to the level of their abilities; consequently, "We must encourage

10
the growing recognition by women and men that equal rights Is a matter
of simple Justice" (National Commission on the Observance of Inter
national Women's Year, 1976, p. v).
Women students who chose business as their profession, which is
still a nontradltJonal field of study for women today, should be free
to strive for top level positions within the profession. Women stu
dents should be freed from the direct and subtle effects of sex-role
stereotyping In their efforts to complete their academic programs In
such nontradltional fields of study as business administration.
Statement of the Problem
It was the purpose of this study to determine whether the charac
teristics of masculinity, femininity, androgyny, age, and academic
ability are factors which correlate with the retention of women stu
dents In programs of study In the College of Business Administration
at the University of Central Florida.
Definition of Terms
The following definitions apply to the terms as they are used
In this study.
Stereotype: An unvarying form or pattern, having no Individuality,
as though cast from a mold.
Sex-role stereotypes: Socially designated behaviors that differen
tiate between men and women; women commonly possess a common pattern
of conventional expression, notion, character, or mental attitude,
Just as men do, but the attitudes and patterns differ greatly.

n
Bern Sex-Role Inventory: An Inventory consisting of sixty personality
characteristics, twenty of which are judged to be socially desirable
traits for men, twenty of which are Judged to be socially desirable
traits for women and twenty characteristics which serve as filler
Items.
Masculine characteristics: A set of characteristics on the Bern Sex-
Role Inventory judged to be more desirable In American society for a
man than for a woman.
Feminine characteristics: A set of characteristics on the Bern
Sex-Role Inventory Judged to be more desirable In American society
for a woman than for a man.
Androgyny: A score on the Bern Sex-Role Inventory which Indicates
that the Individual Is not strongly self-identified with either mas
culine or feminine characteristics.
Group 1: Those women students enrolled in the College of Business
Administration at the University of Central Florida during the Spring
Quarter, 1980, who have completed ninety (90) quarter hours of college
study Including at least two (2) courses In business administration.
Group II: Those women students enrolled at the University of Central
Florida during the Spring Quarter, 1980, who have taken at least two
(2) courses In business administration, and who have transferred out of
the College of Business Administration at some point between January,
1979 and June, 1980.
Age: The actual age of women students In Groups ( and II as stated
on their demographic data sheets.

12
Academic Ability: Scores on the Scholastic Apptltude Test, American
College Test or the Florida Twelfth Grade Test which have been con
verted to a corrmon score utilizing the 1977 State University System
of Florida Equivalency Document.
Delimitations
All of the subjects In this study were enrolled In the College
of Business Administration at the University of Central Florida
between January, 1979, and June, 1980.
There was no attempt to make inferences about the entire female
student population In the State University System of Florida or the
entire student population at the University of Central Florida.
Inferences were confined to the stated population Included In the
study.
Description of the Instruments
The Bern Sex-Role Inventory Is described by Dr. Sandra Llpsltz
Bern as follows:
Both In psychology and In society at large, femininity
and masculinity have long been conceptualized as opposite
ends of a single bipolar dimension. More recently, how
ever, scholars In a number of disciplines have begun to
concern themselves with the concept of psychological
androgyny, a term that denotes the Integration of femi
ninity and masculinity within a single Individual. The
concept of psychological androgyny Irrplles that It is
possible for an Individual to be both compassionate and
assertive, both expressive and Instrumental, both feminine
and masculine, depending upon the situational appropriate
ness of these various modalities; and it further Implies
that an Individual may even blend these complimentary
modalities in a single act, being able, for example, to
fire an employee If the circumstances warrant It but with
sensitivity for the human emotion that such an act Inevit
ably produces.

13
Before empirical research on the concept of psycho
logical androgyny could be Initiated, however, It was
necessary to develop a new type of sex-role Inventory, one
that would not automatically build In an Inverse relation
ship between femininity and masculinity as previous In
ventories had done. On most Inventories, Items are empiri
cally defined as feminine or masculine on the basis of
differential endorsement by females and males, and a person
filling out the Inventory Is said to be either feminine or
masculine as a function of which sex she or he most closely
resembles. Although It Is possible for a person to earn a
score that falls halfway between the two extremes and there
by reveal that she or he does not closely resemble either
sex, a person cannot make the strong statement on such an
Inventory that she or he Is androgynous.
The Bern Sex-Role Inventory (BSRI) was designed to
Implement empirical research on psychological androgyny.
It contains 60 personality characteristics, 20 of which are
stereotypically feminine (e.g., affectionate, gentle, under
standing, sensitive to the needs of others), and 20 of
which are stereotypically masculine (e.g., ambitious, self-
reliant, Independent, assertive). The BSRI also contains
20 characteristics that serve as filler Items (e.g., truth
ful, happy, conceited). When taking the BSRI, a person Is
asked to Indicate on a 7-polnt scale how well each of the
60 characteristics describes herself or himself. The' scale
ranges from 1 ("Never or almost never true") to 7 ("Always
or almost always true") and Is labeled at each point.
The BSRI has two features distinguishing It from most
masculinity-femininity scales. The BSRI treats femininity
and masculinity as two Independent dimensions rather than
as two ends of a single dimension, thereby enabling a per
son to Indicate whether she or he Is high on both dimensions
("androgynous"), low on both dimensions ("undifferentiated")
or, high on one dimension but low on the other (either
"femlntne" or "masculine").
In addition, the BSRI is based on a conception of the
traditionally sex-typed person as someone who Is highly
attuned to cultural definitions of self-appropriate be
havior and who uses such definitions as the Ideal standard
against which her or his own behavior Is to be evaluated.
In this view, the traditionally sex-typed person Is moti
vated to keep her or his behavior consistent with an
Idealized Image of femininity or masculinity, a goal she or
he presumably accomplishes both by selecting behaviors and
attributes that enhance the Image and by avoiding behaviors
and attributes that violate the Image. Accordingly, Items

14
were selected as feminine or masculine on the basis of
cultural definitions of sex-typed social desirability and
not on the basis of differential endorsement by females and
males, i.e., a characteristic qualified as feminine If it
was Judged to be traditionally more desirable In American
society for a woman than for a man, and It qualified as
masculine if It was Judged to be more desirable in American
society for a man than for a woman.
The BSRI is essentially self-administering and may be
given to large groups as well as to Individuals. It has
also been administered by mail in several studies. The
inventory consists of 60 adjectives and phrases printed
on a single sheet with Instructions and space for personal
Information about the subject on the reverse side. (Bern,
1979, PP. 1-3)
The Bern Sex-Role Inventory initially was administered to 444
male and 279 female students In Introductory psychology at Stanford
University during the Winter and Spring of 1973. It was also adminis
tered to an additional 117 male and 77 female students at Foothill
Junior College. The data that these students provided represent the
normative data for the Bern Sex-Role Inventory.
Subjects were classified as sex typed, whether masculine or
feminlnine, If the androgyny t ratio reached statistical significance
(/t/ _> 2.025, df 38, p < .05), and they are classified as androgynous
If the absolute value of the t ratio Is less than or equal to one
(Bern, 1974, p. 161).
Dr. Bern estimated the Internal consistency of the Bern Sex-Role
Inventory by computing the coefficient alpha separately for the mas
culinity, femininity, and social desirability scores of the siijects
In each of the two normative samples. The results Indicated all
three scores to be highly reliable, both In the Stanford sample
"masculinity a .86; femininity a .80; social desirability a
.70" (Bern, 1974, p. 159).

15
Because the reliability of the Androgyny t ratio could not
be calculated directly, coefficient alpha was computed for
the highly correlated Androgyny difference score, femininity
masculinity, using the formula provided by Nunnally (1967)
for linear combinations. The reliability of the androgyny
difference score was .85 for the Stanford sample and .86
for the Foothill sample. (Bern, 197**, p. 159)
In an effort to determine test-retest reliability, the Bern
Sex-Role Inventory was administered for a second time to 28 males and
28 females from the Stanford normative sample. The second adminis
tration of the Inventory took place approximately four weeks after
the first administration.
During this second administration, subjects were told that
we were Interested In how their responses on the test might
vary over time, and they were explicitly Instructed not to
try to remember how they had responded previously. Product-
moment correlations were computed between the first and
second administrations for the Masculinity, Femininity,
Androgyny, and Social Desirability scores. All four scores
proved to be highly reliable over the four-week Interval
(Masculinity r 90; Feminintty r .90; Androgyny r 93;
Social Desirability r .89). (Bern, 197**, p. 160)
Demographic Profile
The demographic profile data sheet was designed to elicit perti
nent background Information from the subjects. Most of the information
requested on the demographic profile data sheet also was collected by
the Investigator from the student's permanent file and University
data tapes. Collecting the same Information from two sources was
Intended to provide a cross check on the data.
The following Information was available only from the demographic
profile data sheet: Address for Sumer Quarter; marital status,
employment outside the home; occupation; and number of children.

16
The final question, "If you were enrolled In the College of Business
Administration after January 1, 1979* and subsequently have transferred
to another college at U.C.F., please indicate In the space provided
below your reasons for the transfer," provided the investigator the
opportunity to analyze the various reasons for the transfer.
Statement of Hypotheses
The major hypothesis of this study was that an individual's member
ship in either Group 1 or Group II can be predicted on the basis of age,
academic ability, and the Bern Sex-Role Inventory characteristics. The
following sub-hypotheses in the null form were tested.
1. The masculinity score derived from the Bern Sex-Role Inventory
is a variable which does not discriminate when used in the
prediction of group membership in either Groups I or II.
2. The femininity score derived from the Bern Sex-Role Inventory
Is a variable which does not discriminate when used in the
prediction of group membership In either Groups I or II.
3. The androgyny score derived from the Bern Sex-Role Inventory
is a variable which does not discriminate when used In the
prediction of group membership in either Groups I or II.
4. The age of the Individual Is a variable which does not dis
criminate when used in the prediction of group membership In
either Groups I or II.
5. The Individual's academic ability Is a variable which does
not discriminate when used in the prediction of group member
ship In either Groups I or II.

CHAPTER II
REVIEW OF THE RESEARCH
The purpose of this review was to study sex-role stereotyping as
It related to women In business and education. Research reviewed
concerned the questions, what Is sex-role stereotyping, how has it
evolved, does it exist today, and if so, how does it affect student
performance.
Conflicts from Sex-Role Stereotyping
The conflicts which result from some sex-role stereotyping
have been studied by many researchers. One of the most noted studies
is Matina Horner's "Toward an Understanding of Achievement Related
Conflicts in Women" (Horner, 1972). Horner tested freshman and sopho
more students, male and female, at a large mid-western university.
These students responded to a cue added to the end of the Standard
Thematic Apperceptive Test (TAT). The cue is "After first term finals,
Anne finds herself at the top of the medical school class." Horner
concluded that approximately "65$ of the girls were disconcerted,
troubled, or confused by the cue" (Horner, 1972, p. 162). She further
concluded that a fear of success imagery dominates responses from
women students and is relatively absent in male responses, and that
conflicts emerge between role expectations and aspirations due to
17

18
previously Imposed stereotypes (Horner, 1972, p. 162). The expec
tancy that success In achievement related situations will be followed by
negative consequences arouses fear of success In otherwise achievement
oriented women which then Inhibits their performance and levels of
aspiration.
In an earlier study, Horner hypothesized that the motive to avoid
success Is significantly greater In women than It Is In men (Horner,
1969). She further hypothesized that this tendency Is even greater In
high achievement oriented, high ability women who are capable of success
than It Is In women at the other extreme; that Is, lower achievement
oriented or lower ability women who cannot and will not seek to achieve.
Horner's research was based on 178 students who were administered the
Standard Thematic Apperceptive Test. Her findings suggest that most
women will exercise their full potential only In a non-competltlve situa
tion, especially If they are competing with men. This tendency Is stronger
among women with an Intense anxiety about success. One of the most
significant findings of this research Is the complexity of achievement
motivation In women. Her results Indicated that fear of success was
found to be aroused In situations In which there was concern over or
anxiety about competitiveness and Its aggressive overtones. When these
anxieties occurred, women experienced conflict and adjusted their be
haviors to their Internalized sex-role stereotypes.
The effects of the various traditional roles performed by women,
such as wife and mother, have been studied. One researcher examined

19
the question of whether women who have experienced these traditional
roles experienced less conflict about achievement (TomlIson-Keasey,
197*0. Subjects In the TomlIson-Keasey study were women students from
Douglass College and University College at Rutgers Untverslty whose
enrolled women students differed In terms of age, marital status and
number of children. A young unmarried sample was drawn from Douglass
College and an older married sample was drawn from the University College.
The procedure followed Homer's study by using a cue response, and the
results Indicated a significantly different level of fear of success.
In the study It was hypothesized that women who were performing as
mothers and wives, which were perceived as appropriate achievements,
experienced less fear of success and reduced anxiety than did those
women who were not performing these "appropriate" roles. The results
were as follows:
... women who had satisfied the mother and the wife roles
are not as likely to evidence fear of success. . unmarried
subjects had significantly more fear of success . than did
married subjects . and women with no children had signi
ficantly more fear of success ... than did women with children.
(Toml Inson-Keasey, 197**, p. 235)
Evolution of Sex Roles
The changing of sex roles has been slow moving. Myth and fact
are Intertwined. Some very significant "ole wives tales," however,
are laid to rest (Tangrl, 1972). For example, data Indicate that
women who seek to enter predominantly male-dominated professions do not
show evidence of having Identified with their fathers over their
mothers. Daughters who have more educated working mothers In male-
dominated occupations will view their mothers as role models. Tangrl

20
also found that women In nontrad!tlonal fields or occupations do not
reject the traditional female roles of wife and mother. They do plan,
however, to postpone marriage and have fewer children than do the more
traditional women. Also they do not view themselves as masculine women
(Tangri, 1972, p. 196).
Studies reveal findings that vividly point out the existence in
our contemporary society of "clearly defined sex-role stereotypes for
men and women contrary to the phenomenon of unisex roles currently
touted In the media" (Broverman, 1972, p. 75). In his examination of
the relationship between mothers' employment status and sex-role per
ceptions of college students, daughters of employed mothers perceived
fewer differences between men and women than did daughters of home
maker mothers. Also, "since more feminine traits are negatively
valued than are masculine traits, women tend to have more negative
self-concepts than do men" (Broverman, 1972, p. 750). The double
standard emerges. According to Broverman:
If women adopt the behaviors specified as desirable for
adults, they risk censure for their failure to be appro
priately feminine, but if they adopt the behaviors that
are designated as feminine, they are necessarily deficient
with respect to the general standards of adult behavior.
(Broverman, 1972, p. 75)
Sex Roles of Women in Business
Few significant studies In the area of women In business have
been found. There is an abundance of opinion and testimony, however,
on the topic. For example, an article appeared In Business Week
Magazine which stressed the barriers are still up for women at the

21
top executive levels. There was noted, however, a new trend for
women graduates who are getting job offers that were closed to those
women who graduated Just a few years before them. In the opinion of Ms.
Grimm, the first woman vice-president of Carson, Price, Scott and Co.,
"A woman who Is determined to play In a game so fixed, had better be
prepared to look like a girl, act like a lady, think like a man and work
like a dog" (For Women, a Difficult Climb to the Top, 1969, p. **5).
Fretz and Hayman conducted a survey with twenty business organi
zations, ten of which are Industrial companies from among the first
100 companies on the Fortune "500" list. The organizations that
participated In the survey employed nearly two million people. The
sample organizations Included women at all levels of employment.
Thirty six percent of the persons employed In the twenty business
organizations that participated In the survey were women. However, a
women at the managerial and professional levels accounted for less than
1 percent of the total number of employees. Fretz and Hayman con
cluded from their survey that women are hlndred by their own precon
ceived Ideas of what their roles In business should be (Fretz and
Hayman, 1973, p. 13*0.
One of only a few available studies with data regarding women in
the work force as early as 1920 Is presented by Elizabeth Waldman.
Most of the data are from the Current Population Survey conducted by
the Bureau of the Census for the Bureau of Labor Statistics. They
relate to the population 16 years of age and over. The typical
v*>rklng woman In 1920 was single, about 28 years of age and from the

22
working class. The typical working woman of today Is married and
half of them are over 39 years of age. Chances are that the working
women of today are supplementing the family Income (Waldnan, 1970,
pp. 10-18).
In an attempt to eliminate the myths about women managers,
research has been conducted to provide Insight Into the similarities
and dissimilarities of men and women managers. As opposed to a pre
dominantly psychological approach to women's attitudes, values, and
beliefs and how these factors cause them to behave In a business
situation, Relf and Newstrom utilized an approach based on women
managers' perceptions of their work environment. The purpose was to
determine If their views of the formal and Informal aspects of organi
zations are different from the views of men. If this was true, a
possible conclusion would be that women should be treated differently
from men. The methodology centered on the semantic differential
technique, which measures perceptions along two independent dimensions,
evaluative and potency. A questionnaire was administered to 286 men
and 55 women In management development programs In business and govern
ment organizations. It was concluded that women and men managers are
more similar than dissimilar In their feelings about their work climate.
Based on this, Relf and Newstrom concluded "that decisions made about
women on the basis of their sex, without considering such individual
factors as background, education, experience, personality and potential,
are likely to be wrong" (Relf and Newstrom, 1975, p. 78).
Recent studies have yielded contradictory results concerning
male and female attitudes and behavior as they work In organizations.

23
Some studies found persons following traditional sex-role stereotypes;
In others, they have not. Powell and Butterfield explored the possi
bility that sex-role Identification Is a more Important variable than
Is gender In an Individual's perceptions of traits desirable for
management personnel. One hundred and ten graduate students with
jobs In the business community and 575 undergraduate business students
participated In the study conducted by Powell and Butterfield. The
students completed the Bern Sex-Role Inventory for both themselves and
for a person whom they believed to be a "good manager."
As expected, Individuals' sex-role Identifications sig
nificantly affected their perceptions of traits desirable
for management personnel, while sex had virtually no effect.
The study concludes that sex-role Identification Is a
variable deserving of further attention, particularly when
sex-related differences are examined. Also, a graduate
woman revealed more masculine traits than feminine In their
self-descriptions, suggesting that a masculine standard
for management may nullify the femininity of women in or
aspiring to management positions. (Powell and Butterfield,
1978, p. 1)
Virginia Schein cited that limited attention has been paid to the
Impact of sex-role stereotypical thinking on the perceived and actual
performance of women In management. Her investigation of prior research
In the areas of sex-role stereotypes and performance perceptions revealed
an over-ridtng theme that sex role stereotyping has a definite and nega
tive impact on the selection of women Into managerial positions. She
further stressed the need for research on the relationship between power
and political behavior and effective managerial performance.

2k
Limited opportunities to acquire work-related power
acquisition behaviors and exclusion from political/
Influence networks within organizations can, then, limit
the performance effectiveness of the women with the
potential to do well, as well as, In the long run diminish
her motivation to perform. As such, a prime research target
should be the role of power acquisition behaviors In effec
tive managerial functioning and the effect of sex and sex-
role stereotypical thinking on acquisition of and oppor
tunity to use these behaviors. (Scheln, 1978, p. 266)
Jeanne Herman and Karen Gyllstrom discovered through their recent
study of 500 employees stratified by sex and Job classification In a
major mldwestern university that myths are dispelled concerning Inter-role
conflict for women employed full-time. "A potential employer could not
justify hiring a male over an equally competent female on the grounds
that the woman will experience greater Interrole conflict." The results
show that married women with children who work full time report no more
conflict between job and family responsibilities than do men who hold a
comparable number of social roles (Herman and Gyllstrom, 1978, p. 330).
Sex Roles of Women In Education
Taylor concluded In her doctoral dissertation that, all things
being equal, superintendents of schools are not likely to hire women
as school administrators. The researcher concluded that the only
factor having any real and constant significance of who will be hired
and who will not be hired Is the sex of the applicant (Taylor, 1973
p. 128).
The percentage of women elementary school administrators during
the past forty years has steadily declined. Information on the high
school administrative level showed the same trend. Data provided by
the National Council of Administrative Women In Education Indicated
the following:

25
... in the eleven-year periods between 1950-51 and 196061,
the number of women serving as junior high and senior high
school principals dropped from 18$ to 3*8$ for all secondary
schools. (Van Meir, 1975, p. 163)
More recent statistics cited by the National Center for Education
Statistics revealed that In the State of Florida In 1977 23 percent
of public school principals and assistant principals were women
(National Center for Education Statistics, 1978, p. 19).
Young (1976) 'researched popular myths in public education con
cerning women in the field of educational leadership at the elementary
and secondary level. For each myth cited, contrary evidence evolved
from the study. Several of the most quoted myths that were disproved
are (1) women rarely possess the credentials for administrative
positions, (2) women seldom aspire or apply for administrative posi
tions, (3) women will not make the required commitment to an adminis
trative position, and (4) women do not have to work because they are
merely after a second income (Young, 1976, p. 83).
Borgstrom, Whiteley, and Rudolph studied the Impact of recent
societal changes on women's educational aspirations. The researchers
utilized data from a student survey administered by the American
Council on Education. They compared the data with survey responses
of freshman students at the University of California, Santa Barbara
(UCSB), from 1968 to 1978. They were attempting to characterize the
University of California, Santa Barbara, campus culture today and show
how and when UCSB students have changed over the past ten years.

26
Women are currently much more undecided about their majors
than are their male counterparts and even substantially
more than women In past years, perhaps reflecting some
confusion about the increased options available to them.
While showing significant decreases in areas traditionally
associated as being "women's majors" (English, fine arts,
social sciences) as well as the less traditional math and
statistics, "undecided" women students have shown increased
Interest in professional fields such as business, engineer
ing and the health professions. (Borgstrom, Whiteley and
Rudolph, 1980, pp. 6-7)
In a recent study by Pulg-Casauranc, women In three fields of
academic study were Investigated. The Edwards Personal Preference
Schedule (EPPS) and the Bern Sex-Role Inventory (BSRl) were used to
assess the relationship between personality characteristics and psycho
logical androgyny for females in traditional (female-dominated), non-
traditional (male-dominated) and neutral (dominated by neither men
nor women) fields of academic study. A random sample of 185 women from
Junior and senior classes of traditional, nontradttional and neutral
academic departments participated in the study.
The data Indicated that the three groups could also be
differentiated by their psychological androgyny scores.
Significant differences for the sample group found the
nontradltional group demonstrated the highest degree of
psychological androgyny, although all three groups fell
within the range defined as psychologically androgynous.
(Pulg-Casauranc, 1977, p. 5)
The most significant finding of the study was that the females in
the three groups thought to be qualitatively different, tended to be
more similar than disparate when It came to displaying and internalizing
behaviors they perceived as appropriate (Pulg-Casauranc, 1977, p. 7).

CHAPTER 11 I
PROCEDURES USED IN CONDUCTING THE STUDY
The following procedures were used in the study:
1. The sampl-e population Included women students who were
enrolled In the University of Central Florida College of
Business Administration for at least one academic quarter
between January, 1979 and June, 1980, and who were enrolled
at the University of Central Florida during the Spring
Quarter, 1980. This sample population was found by using
University records and official University data tapes.
2. The sample population was stratified Into two groups.
Group I: Those women students enrolled in the College of
Business Administration at the University of
Central Florida during the Spring Quarter, 1980,
who had completed ninety quarter hours of college
study Including at least two courses taught In the
College of Business Administration.
Group II: Those women students enrolled at the University
of Central Florida during the Spring Quarter,
1980, who took at least two courses In business
administration, and who had transferred out of the
College of Business Administration at same point
between January, 1979, and June, 1980.
27

28
3. The ages of women students In Groups 1 and II were obtained
from the student's permanent record and official University
data tapes.
k. Scholastic Aptitude Test scores, American College Test
scores or Florida Twelfth Grade Test scores were obtained
from the student's permanent record and official University
data tapes. The scores were converted to a common base
utilizing the State University System Equivalency Docunent
(Jackson, 1977).
5. In the Spring Quarter of 1980, following the assignment of
students to classes, a list of all women students In Groups
I and II was compiled from University registration Information.
6. Demographic data for Groups I and II were obtained from a
demographic profile data sheet developed and administered
by the Investigator (Appendix A).
7. A letter (Appendix B) was mailed to each person selected,
requesting her cooperation In completing the Bern Sex-Role
Inventory which was administered In a conveniently located
room on carpus.
8. When a large enough sanple was not produced by the procedure
In #7, follow-up telephone calls, extending a second Invl-
tatton to participate In the study, were made with careful
attention being given to avoid pressuring students during
the telephone conversation. A minimum of fifty percent
participation by women students contacted In Groups I and II
was sought.

29
9.The Investigator administered the Bern Sex-Role Inventory
(Appendix C).
10. Responses to the final question on the demographic profile
data sheet, "If you were enrolled In the College of Business
Administration after January, 1979, and subsequently have
transferred to another college at the University of Central
Florida, please Indicate In the space provided your reasons
for the transfers," were reviewed and categorized by a panel
of knowledgeable professionals with expertise In the coun
selling of students. The purpose of categorizing responses
was to enable the Investigator to cite the main reasons
Indicated for transferring out of the College of Business
Administration and to compare these responses with the
findings of the study. It was recognized that those students
In Group II may perceive their reasons for transferring out
of the College of Business Administration In many different
ways. This question was Included only to give additional
Information concerning reasons for transferring out of the
College of Business Administration. A check was made to
determine If the reasons did or did not tend to support
the major questions posed In this study.
11. As a means of providing comparative data on the masculinity,
femininity and androgyny scores, a sample of junior and senior
male students enrolled In selected classes In the College of
Business Administration at the University of Central Florida
(Group IV), and a sample of women students enrolled In selected

30
classes In the College of Education at the University of
Central Florida who had never been enrolled In the College
of Business Administration (Group III) completed the Bern
Sex-Role Inventory In their classrooms. These data were
Included In the study to address questions surrounding the
concept of the traditionally sex-typed person as follows:
(1) Did women students majoring In education, a traditionally
feminine field of study, produce a more feminine score on the
Bern Sex-Role Inventory than did the other groups In the study?
(2) Did men students majoring In business administration, a
traditionally masculine field of study, produce a more mascu
line score on the Bern Sex-Role Inventory than did the other
groups In the study?

CHAPTER IV
ANALYSIS OF DATA
A total of 159 subjects were Included In the study. These subjects
were divided Into four groups as follows: Group I (thirty-three subjects)
was composed of a sample of women enrolled In the College of Business
Administration at the University of Central Florida during the spring
quarter, 1980, all of whom had completed ninety quarter hours of college
study, Including at least two courses taught In the College of Business
Administration. Group II (forty subjects) was made up of a sample of
women students enrolled at the University of Central Florida during the
spring quarter, 1980, all of whom had taken at least two courses In busi
ness administration, and who had transferred out of the College of Business
Administration at some point between January, 1979 and June, 1980. Group
III (twenty-three subjects) was a sample of women students enrolled In
selected classes In the College of Education who had never been enrolled
In the College of Business Administration. Group IV (sixty-three subjects)
was a sample of male students enrolled In classes in the College of Busi
ness Administration during the spring quarter, 1980.
Groups I and II were the groups upon which this study was focused.
Groups III and IV were Included In the study as a means of providing com
parative data; consequently, Groups III and IV were not used In the dis
criminant analysis portion of the study, but only In the descriptive
analysis. Groups III and IV were Included to address questions surrounding
31

32
the concept of the traditionally sex-typed person, such as: (1) Did women
students majoring In education, a traditionally feminine field of study,
produce a more feminine score on the Bern Sex-Role Inventory than did the
other groups In the study? (2) Did men students majoring in business ad
ministration, a traditionally masculine field of study, produce a more
masculine score on the Bern Sex-Role Inventory than did the other groups
In the study?
The Statistical Package for the Social Sciences (SPSS) computer pro
gram was used to analyze the data. A discriminant analysis with signifi
cance at .05 was used to test the data. The discriminant analysis enabled
the researcher to statistically distinguish between groups by selecting
the independent variables on which the groups differed.
The discriminant analysis produced a discriminant function, a re
gression equation, with group membership as the dependent variable and
demographic data along with scores on the Bern Sex-Role Inventory as the
Independent variables. The results statistically indicated to which group
each subject probably belonged. The analysis discriminated between the
two groups (Groups I and II) and predicted which subjects were or should
be In either Groups I or II.
The discriminant analysis was used to predict whether a subject be
longed to Group I or Group II on the basis of the selected independent
variables, and to determine the Importance of these Independent variables
In making optimal assignments of the subjects to the two groups. A more
In-depth discussion of the discriminant analysis appears In the "Discrl-
mlnant Analysis Findings" portion of this chapter.

33
A cross tabulation of variables by groups was performed to describe
these variables as they related to the four groups in the study. These
frequency distributions compared variables for one group to the same
variables for another group. The exact differences were more clearly re
vealed when the percentages were examined because of the difference in
total numbers of subjects in each of the four groups. An analysis of
these four groups is reported in the following narrative discussion and
tables.
Analysis of Descriptive Data
Twelve variables were used in the analysis of descriptive data.
Seven variables from the demographic profile data sheet were pertinent to
the descriptive analysis. The questions on the demographic profile data
sheet Illtcltlng current college of enrollment and previous colleges of
enrollment were used only for verification of assignment of subjects to
groups. The name of the high school attended and occupation of the stu
dent were recorded but were not used In the description of the groups.
The seven variables from the demographic profile data sheet which
were considered pertinent to the descriptive analysis were (1) age, (2)
marital status, (3) employment status, (A) number of children, (5) grade
point average, (6) student's enrollment classification (freshman, sopho
more, junior or senior), and (7) year of high school graduation. Two
variables, the sex of the student and Scholastic Aptitude Test score or
its equivalent, were taken from the student's permanent university record.

34
Sex
Included In the four groups cited In the study were sixty-three male
business students, all of whom were Included In Group IV, and nlnety-slx
women students In Groups I, II, and III (Table 1).
Age
The 159 participants ranged In age from 18 to 55 years. Approximately
75 percent of the participants were under 25 years of age. Fifteen per
cent of Group I, 10 percent of Group II, 25.9 percent of Group III and
17.5 percent of Group IV ranged In age from 25 to 34 years. Six percent
of Group I, 10 percent of Group II, 8.6 percent of Group III, and 8 per
cent of Group IV ranged In age from 35 to 55 years (Table 2).
Marital Status
The Information on marital status was collected In categories of
single or divorced, married, or widowed. An analysis of the 159 partici
pants revealed that 73 percent were single or divorced, 24.5 percent were
married and 2.5 percent were widowed. An analysis by group showed Group
I, women business students, as having the highest percentage of single or
divorced members. Women education majors (Group III) had the smallest
percent of single or divorced members (Table 3).
Employment
An analysis of the employment status, which was collected In cate
gories of full time, part time, or unemployed, Indicated a high percentage
(78.3) of unemployed members In Group III, women education majors. Group
I had 30.3 percent unemployed, Group II had 40 percent unemployed and
Group IV had 39.7 percent unemployed (Table 4).

35
Table 1
Sex of Students In
Groups I, II, III and IV
Number of Students Percentage of Total Students
Male Female Used In the Study
Major Groups
Group Ia
0
33
20.8
Group II*5
0
AO
25.2
Group III0
0
23
Comparison
Groups
1A.5
Group IV
d
63 0
39.5
aWomen students enrolled In the College of Business Administration.
^Women students who had transferred out of the College of Business
Administration into another college within the University.
Women students enrolled in the College of Education.
^Men students enrolled In the College of Business Administration.

36
Table 2
Ages of Students
In Groups I, II, III and IV
Percentages
of
Ages of
Group Ia
Group II*5
Group lllc
Group IV^
Students
(N33)
(N-40)
(N-23)
(N-63)
18 to 2k
78.8
80.0
65.2
7k.S
25 to 3k
15.1
10.0
25.9
17.5
35 to 55
6.0
10.0
8.6
8.0
tornen students enrolled In the College of Business Administration.
^Women students who had transferred out of the College of Business
Administration Into another college within the University.
S/omen students enrolled In the College of Education.
^Men students enrolled In the College of Business Administration.

37
Table 3
Marital Status of Students
In Groups I, II, III, and IV
Number of Students and Percentages of Groups
Group Ia Group II*3 Group 111C Group IV^
Single/Divorced
Number
27
29
12
48
Percent
81.8
72.5
52.2
76.2
Married
Number
6
9
9
15
Percent
18.2
22.5
39.1
23.8
Widowed
Number
0
2
2
0
Percent
0
5.0
8.7
0
tornen students enrolled In the College of Business Administration.
^Women students who had transferred out of the College of Business
Administration Into another college within the University.
S/omen students enrolled In the College of Education.
^Men students enrolled In the College of Business Administration.

38
Table 4
Employment Status of Students
In Groups I, II, III, and IV
Status
Group 1
Number of
and Percentages
Ia Group llb
Cases
of Groups
Group 111c
Groi^j IV^
Full Time
Number
2
7
1
9
Percent
6.1
17.5
4.3
14.3
Part Time
Number
21
17
4
29
Percent
63.6
42.5
17.4
46.0
Unemployed
Number
10
16
18
25
Percent
30.3
AO.O
78.3
39.7
aWomen students enrolled In the College of Business Administration.
^Women students who had transferred out of the College of Business
Administration Into another college within the University.
S/omen students enrolled In the College of Education.
^Men students enrolled In the College of Business Administration.

39
Grade Point Average
The range of grade point averages In college work for all groups was
from 1.87 to 3.99 on a 4.00 scale. Grade point averages were determined
from the students' permanent university records. The women students In
Group II had the lowest grade point averages; twenty-three percent of this
group had grade point averages of 2.49 or below. In all of the groups
44.6 percent to 56.3 percent had a grade point average of 3.0 or better In
their college work (Table 5). No special minimum grade point average was
required for admission to the College of Business Administration or the
College of Education; however, a minimum of 2.0 overall grade point aver
age was required for the admission of all degree seeking students (Table
5).
Student Classification
The information on student classification was collected In categories
of freshmen, sophomore, junior, and senior. Seventy-seven and five tenths
percent of the students In Group II, women students who transferred out of
the College of Business Administration, were classified at the junior level
or above. All the participants In Groups I, III, and IV were classified
at the junior level or above (Table 6). At the University of Central
Florida students are permitted to enter the college of their choice at the
beginning of their freshman year.
Scholastic Aptitude Test Scores
Scholastic Aptitude Test scores or their equivalent, as measured on
the American College Test or the Florida Twelfth Grade test, for Groups I
and II were taken from the permanent university records. The scores
ranged from 590 to 1350. The highest score possible Is 1600 and the

4o
Table 5
Grade Point Averages for Students
In Groups I, II, III, and IV
0
Percentage of Each Group Within Grade Point Ranges
1.87 to 2.49 2.50 to 2.99 3.00 and Above
Group 1^
15.1
30.2
54.6
Group llc
22.5
32.5
45.0
Group IIId
8.6
34.6
56.3
Group IV
14.4
41.4
44.6
aGrade point averages are computed on a four point scale 4.0 A grade.
^Women students enrolled In the College of Business Administration.
Sfomen students who had transferred out of the College of Business
Administration Into another college within the University.
tornen students enrolled In the College of Education.
eMen students enrolled In the College of Business Administration.

Student Classifications
In Groups I, II, III and IV
Student
Classifications
Group Ia
Percentages of Groups
Group II*3 Group lllC
Group IVd
Freshman
0
10.0
0
0
Sophomore
0
12.5
0
0
Junior
24.2
45.0
69.6
9.5
Senior
75.8
32.5
30.4
90.5
aWomen students enrolled In the College of Business Administration.
^Women students who had transferred out of the College of Business
Administration Into another college within the University.
S/omen students enrolled In the College of Education.
aMen students enrolled In the College of Business Administration.

lowest Is I00. Forty-five and two tenths percent of Group I scored over
1,000, as compared to 32.5 percent of Group II (Table 7). Scholastic
Aptitude Test scores or their equivalents were not compiled for the sub
jects In Groups III and IV. These groups were Included In the study for
the purpose of comparing results on the Bern Sex-Role Inventory (Table 7)*
Number of Children
The women students In the College of Education (Group III) reported
having the grestest number of children; 3^.7 percent had from one to three
children. The greatest number of children reported by any one participant
was six. A majority of each group had no children (Table 8).
Year of High School Graduation
Participants In the study reported the year of their graduation from
high school on the demographic Information sheet. The range was from 19^3
to 1979 with 77 percent of the participants having graduated since 1973
(Table 9).
Analysis of Open-ended Question
Three professionals with expertise In the counselling of university
students were selected to review and categorize the open-ended question on
the demographic profile data sheet. Each reviewer had a minimum of ten
years of academic and vocational counselling experience In Institutions of
higher education.
The open-ended question was, "If enrolled In the College of Business
Administration after January, 1979, and subsequently transferred to another
college at the University of Central Florida, please Indicate In the space
provided your reasons for the transfer."

43
Table 7
Scholastic Aptitude Test Scores
for Students in Groups I and II
Percentages of Groups
Ranges for .
SAT Scores3 Group lD Group llc
590
- 800
03
18
801
- 1000
51
50
1001
- 1200
36
24
1201
- 1350
09
08
aSAT Is the abbreviation for Scholastic Aptitude Test. The highest
possible score is 1600.
^Women students enrolled in the College of Business Administration.
S/omen students who had transferred out of the College of Business
Administration Into another college within the University.

Number of Students in
Groups I, II, III, and IV with
Children
Number of
Children
Group Ia
(N-33)
Group II*3
(N-40)
Group lllC
(N-23)
Group IV^
(N-63)
One
1
3
3
1
Two
1
1
5
6
Three
1
2
0
3
Four
0
0
0
1
Five
0
1
0
0
Six
0
1
0
0
None
30
32
15
52
aWomen students enrolled in the College of Business Administration.
^Women students who had transferred out of the College of Business
Administration Into another college within the University.
S/omen students enrolled In the College of Education.
^Men students enrolled in the College of Business Administration.

Year of High School Graduation
for Students In Groups I, II, III, and IV
Year of Graduation from High School
1943 to 1959 i960 to 1969 1970 to 1979
Number Percent Number Percent Number Percent
Group Ia
2
6.1
2
6.0
29
87.9
b
Group 11
3
7.5
3
7.5
34
85.0
Group 11 lC
2
8.6
3
12.9
18
78.0
Group IV^
3
4. 8
7
11.2
53
84.2
aWomen students
enrol led
in the
Col lege of
Business
Administration,
^Women students who had transferred out of the College of Business
Administration into another college within the University.
S/omen students enrolled in the College of Education.
Men students enrolled in the College of Business Administration.

Approximately 43 percent of the forty students In Group II, those
women students who transferred out of the College of Business Adminis
tration, responded to the open-ended question. Four broad categories of
answers emerged In the process of grouping the participants' responses,
and all responses fell Into these four categories. They were (1) Inde
cision as to field of study, (2) change of career Interests, (3) expediency
of completion of a degree In another field of study, and (4) boredom or
lack of enthusiasm for the business curriculum.
The category receiving the greatest response was "boredom or lack of
enthusiasm for the business curriculum." Forty-one percent, seven out of
the seventeen responding to the question, were assigned to this category
(Table 10). None of the respondents stated that the academic work in
business administration was too difficult or that they transferred out of
business administration because they were getting low grades In the busi
ness courses.
Descriptive Analysis of
Bern Sex-Role Inventory
Sixty personality characteristics are listed on the Bern Sex-Role
Inventory. Twenty of the characteristics are stereotypically feminine,
twenty are stereotypica1ly masculine and twenty serve as filler Items.
Participants in the study indicated how well each of the sixty personality
characteristics described herself or himself on a scale from one to seven.
The scale ranged from one, "Never or almost never true," to seven, "Always
or almost always true." The sixty personality characteristics are placed
on the Bern Sex-Role Inventory in the following manner: (1) The first

Table 10
k7
Responses by Group 11 Students to the
Four Categories Derived from the
Open-ended Question on the
Demographic Profile Date Sheet
Categories
Number of Responses
(N=17)
Indecision as to Field of Study 3
Change In Career Interests 3
Expediency of Completion of Degree In
Another Field of Study U
Boredom or Lack of Enthusiasm for the
Business Curriculum 7
Seventeen of the kO students In Group II responded to the question.
Note: Open-ended question was, "If enrolled in the College of Business
Administration after January, 1979, and subsequently transferred to
another college at the University of Central Florida, please Indicate
in the space provided your reasons for the transfer."

adjective and every third one thereafter Is masculine; (2) the second
adjective and every third one thereafter Is feminine; (3) the third ad
jective and every third one thereafter Is filler.
Scoring of the Bern Sex-Role Inventory Is conducted as follows:
The first step Is the calculation of each subject's femininity
("a") and masculinity ("b") scores, which are the averages of
each subjects' ratings of the feminine and masculine adjectives
on the BSRI. That Is, a given subject's femininity score is the
mean of that subject's ratings on the feminine adjectives, and
that same subject's masculinity score is the mean of her or his
ratings on the masculine adjectives. (Bern, 1979, p. 1)
The total score for each subject's ratings on the feminine Items
was divided by twenty, If all characteristics were rated by the subject,
and the same procedure was used for calculating the ratings on the mascu
line characteristics. All subjects rated all Items on the Bern Sex-Role
Inventory. A feminine raw score (a) and a masculine raw score (b) were
the result of this process.
The masculine and feminine standard scores were determined from the
masculine and feminine mean raw scores, using Table 1 in the Administration
and Scoring Guide for the Bern Sex-Role Inventory (Appendix D). A difference
score was computed by subtracting the masculine standard score from the
feminine standard score. The difference score was converted to a differ
ence standard score by using Table 2 in the Administration and Scoring
Guide for the Bern Sex-Role Inventory (Appendix D). The median for the
feminine raw score, A.90, and for the masculine raw score, 4.95, is based
on normative data. The median for the difference standard score, also
based on normative data, Is 0 which was converted to a T-score of 50 by
using Table 2 In the Administration and Scoring Guide for the Bern Sex-
Role Inventory (Appendix D).

49
Each subject was classified into one of four categories based on the
median raw scores of the normative sample and the subject's mean raw scores.
All participants In the study were placed In one of the following four cate
gories: (1) masculine, which denotes a low feminine score and high mas
culine score, (2) feminine, which denotes a low masculine score and a high
feminine score, (3) androgynous, which denotes a high masculine score and
a high feminine score, or (4) undifferentiated, which denotes a low masculine
score and a low feminine score (Table 11). "High difference scores In
either direction Indicate a tendency to be strongly sex-typed, positive
scores Indicating femininity and negative scores indicating masculinity"
(Bern, 1979 p. 3). Positive scores indicate a generally more feminine
orientation, negative scores a more masculine orientation, and scores near
zero an androgynous orientation.
Thirteen and three tenths percent of the subjects In Group I (women
students enrolled In the College of Business Administration), 15.0 percent
of the subjects In Group II (women students who had transferred out of the
College of Business Administration), and 13.6 percent of the subjects In
Group III (women students enrolled in the College of Education) were placed
In the masculine category as a result of their scores on the Bern Sex-Role
Inventory. Sixty-two and one tenth percent of the subjects In Group IV
(men students enrolled in the College of Business Administration) were
placed In the masculine category (Table 12).
Thirty-six and seven tenths percent of women students enrolled in the
College of Business Administration (Group I) were classified as feminine
as a result of their scores on the Bern Sex-Role Inventory. Thirty-seven

Table 11
Interpretation of Categories
Derived from the
Bern Sex-Role Inventory
Masculine Score
Below Median
Above Median
Below
Median
Undifferentiated
(low feminine low masculine)
Masculine
(low; feminine -
high masculine)
Feminine
Score
Above
Femlnlne
Androgynous
Median
(high feminine low masculine)
(high feminine
- high masculIne)
Medians Based on Normative Data
Feminine Raw Score 4.90
Masculine Raw Score A.95
vn
o

51
Table 12
Comparison of Undifferentiated, Masculine,
Androgynous, and Feminine Bern Sex-Role
Inventory Classifications by Group
Bern Sex-Role Inventory
Classifications
Group Ia
Percentages of Groups
Group l|k Group lllc Group 1
Undifferentiated
(low masculine and
low feminine)
6.7
7.5
13.6
17.2
Masculine
(low feminine and
high masculine)
13.3
15.0
13.6
62.1
Androgynous
(high feminine and
high masculIne)
^3.3
40.0
45.5
19.0
Feminine
(low masculine and
high feminine)
36.7
37.5
27.3
1.7
aWomen students enrolled In the College of Business Administration.
^Women students who had transferred out of the College of Business
Administration into another college within the University.
S/omen students enrolled In the College of Education.
^Men students enrolled In the College of Business Administration.

52
and one half percent of the women students who had transferred out of the
College of Business Administration (Group II) and 27.3 percent of women
students enrolled In the College of Education (Group III) were placed In
the feminine category as a result of their scores on the Bern Sex-Role
Inventory. The percent of male business students (Group IV) who were
placed In the feminine category as a result of their scores on the Bern
Sex-Role Inventory was 1.7.
Forty-three and three tenths percent of the subjects In Group I (women
students enrolled In the College of Business Administration), 40.0 percent
of the subjects In Group II (women students who had transferred out of the
College of Business Administration), and ^5.5 percent of the subjects In
Group III (women students enrolled In the College of Education) were placed
In the androgynous category as a result of their scores on the Bern Sex-Role
Inventory. The percent of subjects In Group IV (men students enrolled In
the College of Business Administration) who were placed In the androgynous
category as a result of their scores on the Bern Sex-Role Inventory was 19.0.
Groups III and IV were Included In the analysis of the four categories
of masculine, feminine, androgynous, and undifferentiated in order to address
the following questions. (1) Did women students majoring In education, a
traditionally feminine field of study, produce a more feminine score on the
Bern Sex-Role Inventory than did the other groups in the study? (2) Did
men students majoring In business administration, a traditionally masculine
field of study, produce a more masculine score on the Bern Sex-Role Inven
tory than did the other groups In the study? Women majoring In education
(Group III) did not produce a more feminine score than did the other groups.

53
Twenty-seven and three tenths percent of the women majoring in education
were categorized as feminine, as compared to 36.7 percent of women business
majors, 37.5 percent of women who transferred out of the College of Busi
ness Administration, and 1.7 percent of the men business majors. Men
students majoring In business administration did produce a more masculine
score than did the other groups. Sixty-two and one tenth percent of men
majoring in business administration were categorized as masculine, as com
pared to 13.3 percent of women business majors, 15.0 percent of women who
transferred out of the College of Business Administration, and 13.6 per
cent of the women education majors.
Discriminant Analysis Findings
The data were analyzed through a discriminant analysis.
The mathematical objective of discriminant analysis is to
weigh and linearly combine the discriminating variables In
some fashion so that the groups are forced to be as statisti
cally distinct as possible.
Discriminant analysis attempts to do this by forming one or
more linear combinations of the discriminating variables.
These "discriminant functions" are of the form
d,lZ,
di2Z2
+ d. Z
Ip p
where Dj Is the score on discriminant function I, the d's are
weighting coefficients and the Z's are the standardized values
of the p discriminating variables used in the analysis. The
maximum number of functions which can be derived is either one
less than the number of groups or equal to the number of dis
criminating variables, if there are more groups than variables.
Ideally, the discriminant scores (D's) for the cases within
a particular group will be fairly similar. At any rate, the
functions are formed in such a way as to maximize the separ
ation of the groups. Once the discriminant functions have
been derived, we are able to pursue the two research objectives
of this technique; analysis and class!fication.

5^
The analysts aspects of this technique provide several tools
for the interpretation of data. Among these are statistical
tests for measuring the success with which the discriminating
variables actually discriminate when combined Into the discri
minant functions....The result Is often of major theoretical
significance, and statistical tests are Included for this pur
pose. Since the discriminant functions can be thought of as
the axes of a geometric space, they can be used to study the
spatial relationships among the groups. . More importantly,
the weighting coefficients can be interpreted much as In
multiple regression or factor analysis. In this respect, they
serve to Identify the variables which contribute most to dif
ferentiation along the respective dimension (function). (Me,
1975, PP. A35-436)
The discriminant analysis was used to statistically distinguish be
tween Group I, women students enrolled in the College of Business Adminis
tration, and Group II, women students who transferred out of the College
of Business Administration.
The analysis resulted In a linear combination of independent vari
ables (discriminant function) that tended to maximize the separation of
the two groups. The rule for determining the number of equations in a
discriminant analysis is to utilize one function less than the ninber of
groups In the study, or to use the number of equations (discriminant
functions) equal to the number of discriminating variables, whichever Is
smaller. Consequently, one equation (discriminant function) was derived
In this study because two groups were being studied.
When testing for statistical significance of the discriminating vari
ables, a significance level of .05 was found.
The direct method was used In order to select the independent variables
which would contribute to the discriminant analysis. All eleven of the pos
sible independent variables were entered Into the analysts concurrently,
and the discriminant function was created directly from the entire set of

55
variables, with two variables being rejected or not contributing to the
discriminant function. The direct method puts all possible variables In
at one time, as opposed to a stepwise procedure wherein possible variables
are entered one at a time.
The direct method of entry was selected to produce a discriminant
function that would, to the best of its ability, statistically distin
guish between Groups I and il.
Information concerning the methods used for assigning values to the
eleven variables Is discussed below, and the methods used for assigning
values to each of the nine discriminating variables which surfaced are
indicated by footnotes in Table 13.
When data are being collected, a process of measurement Is used in
order to assign a value or score to the observed phenomenon or variable.
In this study an ordinal level of measurement was used for some of the
variables. When it was possible to rank order a category according to
some criteria, an ordinal level of measurement was used. For Instance
the order of freshman, sophomore, junior, and senior in the student clas
sification variable was ranked 1, 2, 3 with freshman being 1 and senior
being 4. The ordinal level of measurement assumes a reasoned order for
value or score assignment, but does not require a known distance between
ranks or scores (Nle, 1975, p. 5).
"The Interval level of measurement has an additional property in that
the distances between the categories are defined In terms of fixed and equal
units" (Nie, 1975, p. 5). Scholastic Aptitude Test scores, for instance,
indicate the same or fixed distance between 800 and 900 or 1000 and 1100.

Table 13
56
Standardized Discriminant Function Coefficients
at the .05 Level of Significance Presented in Order
of Importance to the Discriminant Function and the Levels
of Measurement Used in the Assignment of Values
to the Discriminating Variables
Discriminating Variables Standardized Discriminant Levels of
(Assigned Values) Function Coefficients* Measurement
Student Classification
(freshman 13 1, sophomore 2,
junior 3, senior 4)
0.85^
Ordinal**
Marital Status
(single/divorced 1,
married 2, widowed = 3)
-0.632
Ordinal
Feminine Standard Score from the
Bern Sex-Role Inventory
(actual scores)
-0.368
K-4
Interval
Scholastic Aptitude Test Score
(actual scores)
0.287
Interval
Difference Standard Score from the
Bern Sex-Role Inventory
(actual scores)
0.158
Interval
Employment Status
(full time 1, part time 2
unemployed * 3)
0.117
Ordinal
Age
(actual ages)
0.041
. ****
Ratio
Grade Point Average
(actual averages)
-0.027
Ratio
Masculine Standard Score from the
Bern Sex-Role Inventory
(actual scores)
-0.023
Interval
*The influence of one variable to
the function
In relation to the
Influence of another variable may be made by Ignoring the sign
and comparing coefficients. Zero to +1 drives the case to Group I,
and a 0 to -1 drives the case to Group II.
**0rd!nal level measurements rank order categories according to some
criteria.
***lnterva1 level measurements have fixed and equal units of distance
between categorical ranks.
****Rat!o level measurements have known zero points and have fixed and
equal distances between categorical values.

57
"The ratio level of measurement has all of the properties of an
Interval level with the additional property that the zero point Is In
herently defined by the measurement scheme" (Nie, 1975, p. 5). Age In
dicates the same or fixed distance between age 18 and 19 or age 25 and
26, and has the additional property of a zero point. Table 13 shows the
values or scores assigned to each variable and the level of measurement
used in each assignment.
The eleven variables subjected to the direct method Included seven
Items from the demographic profile data sheet (1) age, (2) marital status,
(3) employment status, (A) grade point average, (5) student classification,
(6) number of children, and (7) year of high school graduation, one item
from the permanent University file (Scholastic Aptitude Test score or Its
equivalent), and the three standard scores derived from the personality
characteristics on the Bern Sex-Role Inventory (masculine standard score,
feminine standard score, and difference standard score). Items six and
seven from the demographic profile data sheet were rejected, leaving nine
variables contributing to the discriminant function.
The nine discriminating variables which surfaced in the analysis as
contributing to the discriminant function were (1) age of the participant,
(2) marital status, (3) employment status, (4) grade point average, (5)
student classification, (6) Scholastic Aptitude Test score, (7) Bern Sex-
Role Inventory masculine standard score, (8) Bern Sex-Role Inventory feminine
standard score, and (9) Bern Sex-Role Inventory difference standard score.
The means and standard deviations for the nine discriminating vari
ables are presented in Tables 1k and 15.

58
Table 14
Means and Standard Deviations
for Discriminating Variables at .05 Level
of Probability for Groups I and II
Di scrtminating
Variables
Group
Mean
Ia
S.D.C
Group
Mean
h
S.D.
Age, (range 18 to 55)
23.21
6.10
23.70
7.71
Marital Status^
1.18
0.39
1.33
0.57
Employment Statusc
2.24
0.56
2.23
0.73
Grade Point Average^
2.98
0.44
2.91
0.56
Student Classification
3.73
0.52
3.03
1.07
SAT Scores9
987.12
141.42
930.50
183.35
0
Women students enrolled in the College of Business Administration.
^Women students who had transferred out of the College of Business
Administration into another college with the University.
CS.D. Is the abbreviation for standard deviation.
^Marital status was recorded as single or divorced (Value 1),
married (Value 2), or widowed (Value 3).
0
Employment status was measured as full time (Value 1), part time
Value 2), or unemployed (Value 3).
^Grade point averages were computed on a four point scale (4.0 A grade).
^Scholastic Aptitude Test scores are normally computed on a maximum
range from 500 to 1500 points or higher. The current minimum
score for admission to the University of Central Florida is 800.

59
Table 15
Means and Standard Deviations
for the Bern Sex-Role Inventory
Masculine, Feminine and Difference Standard Scores
for Groups I and 11
Discriminating
Variables3
Group I*3
Mean S.D.^
Q
Group 11
Mean S.D.
0
Masculine Standard Score
(range 39 to 69)
52.67
10.69
54.30
9.26
Feminine Standard Score^
(range * 44 to 73)
55.50
11.81
58.43
7.57
Difference Standard Score9
(range 51 to 82)
62.77
23.55
65.28
11.25
Three of the nine variables which surfaced at a .05 level of signifi
cance during the discriminant analysis.
^Women students enrolled in the College of Business Administration.
cWomen students who had transferred out of the College of Business
Administration into another college within the University.
^S.D. is the abbreviation for standard deviation.
0
Standard score computed from responses to twenty masculine personality
characteristics on the Bern Sex-Role Inventory.
Standard score computed from responses to twenty feminine personality
characteristics on the Bern Sex-Role Inventory.
^Standard score computed from the difference between the masculine and
feminine scores on the Bern Sex-Role Inventory.

60
The discriminant function coefficients were computed for the nine
discriminating variables. The coefficients represent the relative con
tributions of the nine discriminating variables to the equation (discri
minant function). They serve to identify the variables which contributed
most to the discriminant function. The coefficients for the discriminating
variables are shown on Table 13.
In determining the relative weight of a particular variable, the sign
of the coefficient Is ignored, and the value of the coefficient for any one
variable may be compared to another. For Instance, the student classifi
cation coefficient projects 20.8 times more Influence on the discriminant
function than does the age coefficient. The relative weight in this ex
ample Is determined by dividing the coefficient for the age variable
(.0*1) into the coefficient for the student classification variable (.85*0.
(.85* divided by .0*1 20.8).
Discriminant scores were computed for each one of the participants
in Group I and II. The computation process was as follows: The standard
value for the first discriminating variable was multiplied by its corres
ponding coefficient; this product was added to the product of the second
discriminating variable which likewise was calculated by multiplying the
standard value of the variable by Its corresponding coefficient. This
process took place for each of the nine discriminating variables, re
sulting In the computation of a "discriminant score" for each of the
subjects in Groups I and II (Table 16).

61
Table 16
Classification of Subjects In
Groups I and II Utilizing
Nine Discriminating Variables
Subjects
Actual Group
Membershlp
Predicted Group
Membership
Discriminant
Scores***
1
1 *
1
0.8A78
2
1
1
1.7815
3
1
2****
-0.2951
A
1
1
1.0657
5
1
2****
-0.0610
6
1
1
1.0075
7
l
1
0.6711
8
1
1
0.5212
9
1
1
1.3583
10
1
2****
-0.6170
11
1
2****
-0.113A
12
1
1
1.6095
13
1
2****
-0.3881
1A
1
1
0.3993
15
1
1
0.8615
16
l
1
0.5013
17
1
1
0.1766
18
1
2****
-0.2285
19
1
1
0.9171
20
1
1
1.51A2
21
1
2****
-0.9658
22
1
1
1.3706
23
1
1
0.735A
2A
1
1
1.A973
25
1
1
O.89AI
26
1
2****
-0.0366
27
1
2****
-O.A38A
28
1
1
1.1789
29
1
1
1.5886
*Women students enrolled In the College of Business Administration.
**Women students who had transferred out of the College of Business
Administration.
***Scores obtained when coefflcents are applied to the nine discri
minating variables. Zero to +1 drives the case to Group I, and
a 0 to -1 drives the case to Group II.
****These subjects were not properly classified In their original group.
NOTE: Age, marital status, employment status, grade point average,
student classification, Scholastic Aptitude Test scores, and
Bern Sex-Role Inventory masculine, feminine and difference scores
were the nine discriminating variables.

62
Table 16-contlnued
Actual Group
Predicted Group
Dtscriminant
Subjects
Membership
Membership
Scores***
30
1*
1
0.9984
31
1
1
0.2017
32
1
-0.1695
33
1
1
0.7602
34
2**
2
-2.5037
35
2
2
-0.1481
36
2
] ****
0.1248
37
2
2
-0.5548
38
2
2
-1.9723
39
2
2
-3.3995
40
2
2
-1.2594
41
2
2
0.0087
42
2
] ****
0.5658
43
2
| ****
0.6596
44
2
1 ****
1.0142
45
2
2
-1.1935
46
2
] ****
1.1894
47
2
2
-2.5562
48
2
2
-1.8649
49
2
2
-0.9820
50
2
1****
1.6084
51
2
] ****
0.3359
52
2
2
-0.1308
53
2
2
-0.5588
54
2
] ****
0.6834
55
2
]****
0.2010
56
2
] ****
0.2789
57
2
1****
0.4473
58
2
2
-0.5730
*Women
students
enrolled In the
Col lege of Business
Administration.
**Women
students
who had transferred out of the College of Business
Administration.
***Scores obtained when coefficients are applied to the nine discri
minating variables. Zero to +1 drives the case to Group I, and
a 0 to -1 drives the case to Group II.
****These subjects were not properly classified In their original group.
NOTE: Age, marital status, employment status, grade point average,
student classification, Scholastic Aptitude Test scores, and
Bern Sex-Role Inventory masculine, feminine and difference scores
were the nine discriminating variables.

63
Table 16-continued
Subjects
Actual Group
Membership
Predicted Group
Membership
Discriminant
Scores***
59
2**
1****
O.A533
60
2
2
-1.0115
61
2
]****
0.1311
62
2
2
-0.7055
63
2
2
-1 .A451
61*
2
2
-0.2870
65
2
2
-0.3365
66
2
2
-1.21*36
67
2
2
-0.7151
68
2
2
-2.1282
69
2
2
-2.5279
70
2
] ****
1.0655
71
2
2
-0.591*8
72
2
2
0.2116
73
2
1 ****
0.9951
*Women students enrolled In the College of Business Administration.
**Women students who had transferred out of the College of Business
Administration.
***Scores obtained when coefficients are applied to the nine discri
minating variables. Zero to +1 drives the case to Group I, and
a 0 to -1 drives the case to Group II.
****These subjects were not properly classified In their original group.
NOTE: Age, marital status, employment status, grade point average,
students classification, Scholastic Aptitude Test scores, and
Bern Sex-Role Inventory masculine, feminine and difference scores
were the nine discriminating variables.

64
Classification of Cases
The use of discriminant analysis as a classification technique
comes after the initial computation. Once a set of variables
Is found which provides satisfactory discrimination for the
cases with known group memberships, a set of classification
functions can be derived which will permit the classification
of new cases with unknown memberships. (Nte, Hull, Jenkins,
Steinbrenner and Brent, p. 436)
If variables are found which do well in predicting to which group the
subjects belong, then these variables can be used to predict the likeli
hood of group membership for other students.
As a check on how well the equation (discriminant function) worked,
the subjects were placed In either Group I or Group II according to the
discriminant scores on the nine discriminating variables. Group member
ship was predicted, based on the nine discriminating variables. The goal
of the classification process as It relates to a discriminant analysis
was to measure the success of the equation (discriminant function) by ob
serving how many subjects were properly classified into their groups.
Twenty-three subjects out of thirty-three subjects in Group I, women
business majors, were properly classified. Twenty-five subjects out of
forty subjects in Group II, women students who transferred out of the
College of Business Into another college within the University, were
properly classified into their group (Table 16).
The data for Groups I and II were subjected to the classification
process. When the coefficients were applied to the nine discriminating
variables, and a discriminant score of less than zero to minus one re
sulted, the subject was placed in Group II. A discriminant score of more
than zero to plus one resulted in the subject's being placed In Group I
for classification purposes.

65
When all the subjects In Group I were subjected to the classifica
tion process, 69.7 percent of them were properly placed In Group I;
consequently, It was possible to predict group membership for Group I,
women business majors, 69.7 percent of the time.
When all the subjects In Group II were subjected to the classifi
cation process, 62.5 percent of them were properly placed In Group II;
consequently, It was possible to predict group membership for Group II,
women students who transferred from the College of Business Administra
tion to another college In the University, 62.5 percent of the time.
For any new woman student wishing to enter as a freshman In the
College of Business Administration from whom data on the nine discri
minating variables could be gathered, and for whom a discriminant score
could be computed, the classification process could be performed. If
she were placed In Group I as a result of her discriminant score, It
could be predicted at a 69.7 percent level of probability that she would
remain enrolled In the College of Business Administration. If as a
result of the classification process, the same new woman student was
placed In Group II, It could be predicted at a 62.5 percent level of
probability that she would transfer out of the College of Business
Administration.
Test of Hypotheses and Results
The hypotheses of this study were that an Individual's membership
In either Group I or Group II Is predictable on the basis of age, academic
ability, and the Bern Sex-Role Inventory characteristics. The following
sub-hypotheses In the null form served as the basis for making predictions
of group membership.

66
1. The masculine score derived from the Bern Sex-Role
Inventory Is a variable which does not discriminate
when used In the prediction of group membership In
either Groups I or 11.
This null hypothesis was rejected. The masculine standard score
was one of the nine variables which was shown to be discriminating in the
discriminant analysis. The corresponding coefficient for the masculine
standard score was -.023 at the .05 level of significance and was ninth
out of nine discriminating variables. The coefficient for the next
highest discriminating variable, grade point average (-.027), provides
1.17 times the Influence of the coefficient for the masculine variable
(-.023). The variable with the greatest Influence, student classifi
cation (.85*0, has 37.13 times more influence than does the coefficient
for the masculine variable (-.023). The masculine variable, combined
with the remaining eight variables In the discriminant function, resulted
in the successful classification of 69.7 percent of the subjects In
Group I and 62.5 percent of the subjects In Group II (Table 13).
2. The feminine score derived from the Bern Sex-Role Inven
tory Is a variable which does not discriminate when
used In the prediction of group membership In either
Groups I or II.
This null hypothesis was rejected. The feminine standard score was
one of the nine variables that was shown to be discriminating In the dis
criminant analysis. The coefficient for the feminine standard score was
-.368 at the .05 level of significance and was third out of nine discri
minating variables. The coefficient for the next highest discriminating
variable, marital status (-.632), provides 1.71 times the influence of
the coefficient for the feminine variable (-.368). The variable with
the greatest Influence, student classification (.85*0, has 2.32 times more

67
influence than does the coefficient for the feminine variable (-.368).
The feminine variable, combined with the remaining eight variables in the
discriminant function, resulted in the successful classification of 69.7
percent of the subjects in Group I and 62.5 percent of the subjects in
Group II (Table 13).
3. The androgyny classification derived from the Bern
Sex-Role Inventory is a variable which does not
discriminate when used in the prediction of group
membership in either Group I or II.
This null hypothesis was rejected. The androgynous classification
was determined through the difference standard score. The difference
standard score was one of the nine variables that was shown to be discri
minating In the discriminant analysis. The corresponding coefficient for
the difference standard score was .158 at the .05 level of significance
and was fifth out of nine discriminating variables. The coefficient for
the next highest discriminating variable, Scholastic Aptitude Test score
(.287), provides 1.82 times the Influence of the coefficient for the dif
ference standard score (.158). The variable with the greatest influence,
student classification (.85*0, has 5.^1 times more Influence than does
the coefficient for the difference score (androgyny) variable (.158).
This difference variable, combined with the remaining eight variables in
the discriminant function, resulted in the successful classification of
69.7 percent of the subjects in Group I and 62.5 percent of the subjects
In Group 11 (Table 13).
k. The age of the individual Is a variable which does not
discriminate when used In the prediction of group mem
bership In either Groups 1 or II.

68
This null hypothesis was rejected. The age of the participants was
one of the nine variables that surfaced as being discriminating. The
corresponding coefficient for age was .0**1 at the .05 level of signifi
cance and was seventh out of nine discriminating variables. The coeffi
cient for the next highest discriminating variable, employment status
(.117), provides 2.85 times the Influence of the coefficient for age
(.041). The variable with the greatest Influence, student classification
(.85*0, has 20.83 times more Influence than does the coefficient for the
age variable (.0**1). The Inclusion of the age variable, along with the
remaining eight variables which were discriminating, resulted in the
ability of the function to successfully predict group membership 69.7
percent of the time for Group I and 62.5 percent of the time for Group II
(Table 13).
5. The Individual's academic ability Is a variable which
does not discriminate when used In the prediction of
group membership In either Groups I or II.
This null hypothesis was rejected. Academic ability, as measured by
the Scholastic Aptitude Test score, or Its equivalent, was one of the nine
variables which resulted from the discriminant analysis. The correspond
ing coefficient for academic ability was .287 at the .05 level of signifi
cance and was fourth out of nine discriminating variables. The coefficient
for the next highest discriminating variable, feminine standard score
(-.368), provides 1.28 times the Influence of the coefficient for academic
ability (.287). The variable with the greatest Influence, student classi
fication (.85**), has 2.98 times more Influence than does the coefficient
for the academic ability variable (.287). The Inclusion of the academic

69
ability variable, along with the remaining eight discriminating vari
ables, resulted In the successful classification of 69.7 percent of
cases In Group I and 62.5 percent of the cases In Group II (Table 13).

CHAPTER V
CONCLUSIONS AND RECOMMENDATIONS
The purpose of this study was to determine If the characteristics
of masculinity, femininity, androgyny, academic ability and age could
be used to predict whether women students In the College of Business
Administration at the University of Central Florida would remain In that
college or transfer to another college within the University.
Data was collected from four groups of students at the University of
Central Florida. The two major groups were Group I, women students en
rolled In the College of Business Administration, and Group II, women
students who had transferred out of the College of Business Administration
Into another college within the University. Group III, women students
enrolled In the College of Education, and Group IV, men students enrolled
In the College of Business Administration, were not Included for the pur
pose of predicting retention; these two groups were used to address ques
tions surrounding the concept of the traditionally sex-typed person, such
as: (1) Did women students majoring in education, a traditionally feminine
field of study, produce a more feminine score on the Bern Sex-Role Inventory
than did the other groups In the study? (2) Did men students majoring In
business administration, a traditionally masculine field of study, produce
a more masculine score on the Bern Sex-Role Inventory than did the other
groups In the study?
70

71
Eleven variables for students In Groups I and II were subjected to a
discriminant analysis technique and nine of these variables surfaced as
discriminating at the .05 level of significance. Listed according to their
relative Influence In the predid ton of group membership (either Group I
or Group II), the nine discriminating variables were, (1) student classi
fication, (2) marital status, (3) Bern Sex-Role Inventory feminine standard
score, (4) Scholastic Aptitude Test score (5) Bern Sex-Role Inventory dif
ference standard score, (6) employment status, (7) age, (8) grade point
average and (9) Bern Sex-Role Inventory masculine standard score. Although
age, academic ability, masculinity, femininity, and androgyny (as deter
mined through the difference score on the Bern Sex-Role Inventory) were
variables that surfaced as discriminating In the analysis, only femininity
(feminine standard score) realized a coefficient with relatively strong
Influence on the prediction of group membership.
Analyzing of data on the nine discriminating variables for a new
woman student, who as a freshman selected business administration as
her major field of study, will enable a prediction to be made regarding the
probability of her remaining In the College of Business Administration.
A classification made as a result of the discriminant analysis of her data
that would put her In the group which would Indicate she would tend to
remain In the College of Business Administration could be said to be
correct 69.7 percent of the time. A classification made as a result of the
discriminant analysis of her data which put her In the group that would
Indicate she would tend to transfer out of the College of Business Adminis
tration could be said to be correct 62.5 percent of the time. This

72
conclusion is drawn because group membership was predictable in 69.7 per
cent of the cases classified in Group I, women business majors, and 62.5
percent of the cases in Group II, women who had transferred out of business
administration. The combination of the nine discriminating variables in
a discriminant analysis made these predictions possible.
Students in Group I, women students enrolled in the College of Busi
ness Administration, and Group II, women students who had transferred out
of the College of Business Administration, were similar in age, marital
status, and employment status. Groups I and II also were similar in aca
demic ability, as measured by grade point averages and Scholastic Aptitude
Test scores or their equivalents.
Students in Group II, women students who had transferred out of the
College of Business Administration, apparently had not made the transfer
for reasons of poor academic ability. Thirty-nine out of forty students
in Group II maintained an over-all grade point average of 2.0 or better.
Those responding to the open-ended question (seventeen out of forty) did
not indicate that their transfer had anything to do with the business
administration courses being too difficult for them; however, it is likely
that students would not tend to state that course difficulty was a factor
in their transferring out of the College of Business Administration. They
might have tended to omit an answer to the question rather than state that
the work was too difficult. The poor response to the open-ended question
leaves real doubt as to the students' reasons for transferring out of
Business Administration.

73
From Dr. Sandra Bern's research on sex role characteristics, the
assumption has been made that those people with androgynous traits can
be said to have a more balanced personality than do those with strong
masculine or strong feminine traits. According to this study, women
business majors were classified as androgynous at a slightly higher per
centage rate (43.3 percent as opposed to 40.0 percent) than were women
who transferred out of the College of Business Administration. In the
comparison groups, Groups III and IV, It was Interesting to note that
women education majors In Group III had a higher percentage of those
classified as androgynous than did any other group. It was also interesting
to note that Group III had the largest percentage of older members (over
25), the largest percentage of married subjects, and the largest percen
tage of members with children than did any other group. It seems reasonable
to conclude that maturity Is a definite factor In women students' self
perceptions, which tend to put them Into an androgynous classification
on the Bern Sex-Role Inventory. Group IV, men business majors, tended to
be strongly sex-typed with 62.1 percent classified as masculine, and only
19.0 percent classified as androgynous.
RecommendatIons
I. It Is recommended that similar studies, using the same measurement
Instruments In combination with additional measurement Instruments, be
conducted for the purpose of strengthening or expanding the discri
minant function and Increasing Its ability to predict the retention
or transfer of women students In business administration.

74
2. It Is recommended that similar studies, using the same measurement
Instruments in combination wlth additional measurement Instruments,
be conducted for the purpose of strengthening or expanding a discrimi
nant function and Increasing Its ability to predict the retention or
transfer of men students In business administration.
3. Due to the limitations of this study to one Institution of higher
education, it is recommended this study be replicated In other
institutions.
4. In counseling women students who designate business administration
as a major field of study, counselors should gather and analyze data
on the nine variables cited in this study for the purpose of counseling
these students.
5. Additional studies should be done to determine the reasons students
transfer out of the College of Business Administration.
6. Studies should be done to find answers to the following questions:
Does the business administration curriculum and the teaching of that
curriculum tend to attract and retain students with certain sex-role
characteristics? If this is true, should efforts be made to change
the curriculum and/or the methods of teaching? Would such changes
produce students who would be more effective in their jobs in the
business community? Is the business administration curriculum
tailored to be more attractive and palatable to those students who
tend to have more masculine characteristics than androgynous or
feminine characteristics?

APPENDIX A
DEMOGRAPHIC PROFILE DATA SHEET

PERSONAL INFORMATION FORM
Social Security #
Phone #
Name:
(Last) (First) (Middle/Maiden)
Address at which you can be reached:
Spring Quarter, 1980
(Street) (City) (State) (Zip)
Summer Quarter, I98O
(Street) (Ctty) (State) (ZI p')
Blrthdate: Age: ______________________
Marital Status: __________ Single/Divorced Married Widowed
Employment outside the home: Ful 1 Time Part Time Unemployed
Occupation:
Number of Children Grade Point Average
(Overall College Work, Including
Junior College, If Applicable)
Student Classification: Freshman Sophomore Junior Senior
(Circle one)
College at U C F In which you are currently enrolled
If you have been enrolled In a college at U C F other than the one In
which you are now enrolled, please list those college In order of Enrollment
High School from which you graduated:
OVER
Year
76

77
Personal Information Form: (continued)
If you are currently enrolled In the College of Business Acknlnistration,
do not answer the following question on this Personal Information Form.
If you were enrolled in the College of Business Administration
after January 1, 1979, and subsequently have transferred to
another college at UCF, please Indicate below your reasons
for the transfer.

APPENDIX B
LETTER TO GROUPS I AND II

May 6, 1980
Dear
One of our primary goals Is to lend assistance to students enrolled at
the University of Central Florida as they work toward the successful
completion of their academic programs. As part of our effort to assist
you, we need to gather Information that will enable us to develop and
offer programs that will help you reach your educational and professional
goals.
You have been selected as part of a random sample of women students;
consequently, your participation will be critical to the accuracy of
this study. Please be assured that your Involvement will be treated
confidentially and will be used only In combination with others to
compare responses.
Information will be gathered during the end of Spring Quarter. We
would appreciate your taking about 10 to 15 minutes of your time v/iile on
campus to come by Room 282 In the Administration Building any time during
the hours posted below. In return we would be happy to share with you a
summary of the research results and our plans to provide programs to en
hance your professional development.
Thursday,
May 22 10 a.m.
to 3
p.m.
Room
280,
Conference
Room
Friday,
May 23 10 a.m.
to 3
p.m.
Room
280,
Conference
Room
Tuesday,
May 27 10 a.m.
to 3
p.m.
Room
280,
Conference
Room
Wednesday,
May 28 10 a.m.
to 3
p.m.
Room
280,
Conference
Room
Thursday,
May 29 10 a.m.
to 3
p.m.
Room
280,
Conference
Room
Sincerely,
Carol P. Wilson
Dean of Women
CPW:o
79

APPENDIX C
INSTRUCTIONS FOR THE BEM
AND
THE BEM SEX-ROLE
SEX-ROLE INVENTORY
INVENTORY

BEM INVENTORY
Developed by Sandra L. Bem.Ph.D.
Name Age Sex
Phone No. or Address
Date 19
If a student: School ______________________________ Yr. 1 School
If not a student: Occupation
DIRECTIONS
On the opposite side of this sheet, you will find listed a number of per
sonality characteristics. We would like you to use those characteristics
to describe yourself, that Is, we would like you to Indicate, on a scale
from 1 to 7* how true of you each of these characteristics Is. Please do
not leave any characteristic unmarked.
Example: sly
Write a 1 If It
Write a 2 If It
Write a 3 If ft
Write a A If It
Write a 5 If It
Write a 6 If It
Write a 7 If It
Is never or almost never true that you are sly.
Is usually not true that you are sly.
Is sometimes but Infrequently true that you are sly.
Is occasionally true that you are sly.
Is often true that you are sly.
Is usually true that you are sly.
Is always or almost always true that you are sly.
Thus, If you feel It Is sometimes but Infrequently true that you are "sly,"
never or almost never true that you are "malicious," always or almost al
ways true that you are "Irresponsible," and often true that you are "care
free," then you would rate these characteristics as follows:
Sly
3
1rrespons ible
7
Malicious
1
Carefree
5
CONSULTING PSYCHOLOGISTS PRESS, INC.
577 College Avenue Palo Alto, California 9^306
Copyright, 1978, by Consulting Psychologists Press, Inc. All rights re
served. Duplication of this form by any process Is a violation of the
copyright laws of the United States except when authorized In writing by
the PublIsher.
81

82
1
I
2
I
3
I
k
1
5
|
6
7
1
1
Never or
almost
never true
!
Usual 1y
not
true
1
Sometimes
but In
frequently
true
1
Occasionally
true
Often
true
Usually
true
Always or
almost
always
Defend my own beliefs
Adaptable
Flatterable
Affectionate
Dominant
Theatrical
Conscientious
Tender
Self-Sufficient
Independent
Conceited
Loyal
Sympathetic
!
Willing to take a
stand
Happy
Moody
Love Children
Individualistlc
i -
Assertive
Tactful
Soft-spoken
Sensitive to needs
of others
Aggressive
Unpredictable
Rel Iable
Gentle
Masculine
Strong personality
Conventional
Gullible
Understanding
Self-reliant
Solemn
Jealous
Yielding
Competitive
Forceful
Helpful
Childlike
Compassionate
Athletic
Likeable
Truthful
Cheerful
Ambitious
Have Leadership
abl1Itles
Unsystematic
Do not use harsh
language
Eager to soothe
hurt feelings
Analytical
Sincere
Secretive
Shy
Act as a leader
Willing to take risks
Inefficient
Feminine
Warm
Make decisions easily
Friendly
a b Class
R.S.
S.S.
a-b
SS diff

APPENDIX D
ADMINISTRATION AND SCORING GUIDE FOR
THE BEM SEX-ROLE INVENTORY

ADMINISTRATION AND SCORING GUIDE FOR THE BEM SEX-ROLE INVENTORY*
Sandra Lipsltz Bern, Ph.D.
Cornell University
INTRODUCTION
Both In psychology and In society at large, femininity and masculinity
have long been conceptualIzed as opposite ends of a single bipolar dimension.
More recently, however, scholars In a number of disciplines have begun to
concern themselves with the concept of psychological androgyny, a term that
denotes the Integration of femininity and masculinity within a single indi
vidual. The concept of psychological androgyny Implies that it is possible
for an Individual to be both compassionate and assertive, both expressive
and Instrumental, both feminine and masculine, depending upon the situa
tional appropriateness of these various modalities; and it further implies
that an individual may even blend these complementary modalities in a
single act, being able, for example, to fire an employee if the circum
stances warrant it but with sensitivity for the human emotion that such an
act inevitably produces.
Before empirical research on the concept of psychological androgyny
could be Initiated, however, it was necessary to develop a new type of
sex-role Inventory, one that would not automatically build in an Inverse
relationship between femininity and masculinity as previous Inventories
had done. On most Inventories, lterns are empirically defined as feminine
or masculine on the basis of differential endorsement by females and males,
and a person filling out the inventory Is said to be either feminine or
masculine as a function of which sex she or he most closely resembles.
Although It is possible for a person to earn a score that falls halfway
between the two extremes and thereby reveal that she or he does not closely
resemble either sex, a person cannot make the strong statement on such an
inventory that she or he is androgynous.
The Bern Sex-Role Inventory (BSRl) was designed to Implement empirical
research on psychological androgyny. It contains 60 personality character
istics, 20 of which are stereotypically feminine (e.g., affectionate,
gentle, understanding, sensitive to the needs of others), and 20 of which
are stereotypically masculine (e.g., ambitious, self-reliant, Independent,
assertive). The BSRl also contains 20 characteristics that serve as filler
items (e.g. truthful, happy, conceited). When taking the BSRl, a person
is asked to indicate on a 7Point scale how well each of the 60 characteris
tics describes herself or himself. The scale ranges from 1 ("never or
almost never true") to 7 ("Always or almost always true") and is labeled
at each point.
*This material is distributed free with orders for the Bern Sex-Role In
ventory pending publication of a complete Manual for the test. Copyright
1979, by Consulting Psychologist Press, Inc., 577 College Avenue, Palo
Alto, CA 94306.
84

85
The BSRI has two features distinguishing it from most masculinity-
femininity scales. The BSRI treats femininity and masculinity as two
independent dimensions rather than as two ends of a single dimension,
thereby enabling a person to Indicate whether she or he Is high on both
dimensions ("androgynous") low on both dimensions ("undifferentiated")
or, high on one dimension but low on the other (either "feminine" or
"masculine").
In addition, the BSRI is based on a conception of the traditionally
sex-typed person as someone who is highly attuned to cultural definitions
of sex-appropriate behavior and who uses such definitions as the ideal
standard against which her or his own behavior is to be evaluated. In
this view, the traditionally sex-typed person Is motivated to keep her or
his behavior consistent with an Idealized Image of femininity or mascu
linity, a goal she or he presumably accomplishes both by selecting be
haviors and attributes that enhance the Image and by avoiding behaviors
and attributes that violate the Image. Accordingly, items were selected
as feminine or masculine on the basis of cultural definitions of sex-
typed social desirability and not on the basis of differential endorse
ment by females and males, i.e., a characteristic qualified as feminine
If it was judged to be traditionally more desirable in American society
for a woman than for a man, and it qualified as masculine If it was judged
to be more desirable In American society for a man than for a woman.
For a full discussion of the development of the BSRI, the reader should
consult The Measurement of Psychological Androgyny by Sandra L. Bern, Journal
of Consulting and Clinical Psychology, 197^* Vol. *2, pp. 155" 162.
ADMINISTRATION AND SCORING
The BSRI is essentially self-administering and may be given to large
groups as well as to Individuals. It has also been administered by mail
in several studies. The inventory consists of 60 adjectives and phrases
printed on a single sheet with instruction and space for personal infor
mation about the subject on the reverse side. Although it has been used
primarily with college students and adults, the Items in the BSRI should
be comprehensible to most upper elementary school children.
The test is labeled simply "Bern Inventory" to reduce the possibility
that responses might be Influenced by a knowledge of the purpose of the
scales. Most subjects complete the original Inventory in 15 minutes or
less and the short form in 10 minutes. The test is arranged so that the
30 short form Items appear first and, where time is limited, subjects may
be Instructed to stop after the Item "conventional" or at the heavy black
line in the middle column.
It may be helpful to remind subjects orally that it is Important that
they not skip any items. Questions about the meaning of any Item should be
answered in as neutral a fashion as possible. It is recommended that lead
pencils with erasers be used so that any changes In responses will be clearly
made to facilitate accurate scoring.

86
Handscoring the BSRI Is a relatively simple clerical task which may
be facilitated by use of a pocket calculator. The first step Is the
calculation of each subject's Femininity ("a") and Masculinity ("b")
scores, which are the averages of subject's ratings of the feminine and
masculine adjectives on the BSRI. That Is a given subject's Femininity
Score Is the mean of that subject's ratings on the feminine adjectives,
and that same subject's Masculinity Score Is the mean of her or his
ratings on the masculine adjectives. The placement of adjectives on the
BSRI Is as follows:
1) The first adjective and every third one thereafter Is masculine.
2) The second adjective and every third one thereafter Is feminine.
3) The third adjective and every third one thereafter Is filler.
Users may wish to make templates for the two scales by punching holes
In blank answer sheets so that only the feminine or masculine ratings are
visible at any one time.
In any event, total the sum of the ratings for either scale and divide
by the number of Items rated. Unless the subject has omitted Items, the
divisor will be 20 for the original form or 10 for the short form, but it
Is important to correct the divisor when Items have been omitted.
The average of the ratings for the Feminlnlty scale Is entered In
box for the "a11 raw score (RS) at the bottom of the answer sheet. The
average of the Masculinity ratings Is entered In the box labeled "b".
Now look up the standard scores (SS) for each of the raw scores using
Table 1, page l,* and enter them In the SS boxes under the raw score boxes.
The T-scores In Table 1 are adjusted so that females and males are equally
represented. Be careful to use the correct column depending upon whether
you have administered the original or the short form.
Next obtain the Difference Score by subtracting the "b" SS from the
"a" SS. Be sure to retain the correct sign, plus (+) for positive and
minus (-) for negative. High scores In either direction Indicate a ten
dency to be strongly sex-typed (or sex-reversed), positive scores Indicating
femininity and negative scores Indicating masculinity.
After obtaining the a-b score, consult Table 2 to find the Standard
Score for the difference. Note that one column In Table 2 gives the Short
Form SS's and one gives the Original Form SS's.
Now you may return to the upper (raw score) boxes and classify each
subject In one of four categories based on the relationship between the
median raw scores of the normative sample and the subject's scores on
each scale.
Instructions for scoring by computer using punched cards are available
without charge from the Publisher.

87
Femininity
Score
Masculinity Score
Below Median Above Median
Below
Undifferentiated
Mascullne
Median
(low-low)
(low fern.-high mase.
Above
Feminine
Androgynous
Median
(High fern.-low mase.)
(high fern.-high mase.)
Based on the normative data
the medians are as follows:
on Stanford students
(sexes combined),
Original Form
Short Form
Femininity RS
4.90
5.50
Masculinity RS
4.95
4.80
For a discussion of the significance of the classification, see Bern,
S. L. On the Utility of Alternative Procedures for Assessing Psychological
Androgyny. Journal of Consulting and Clinical Psychology, 1977 Vol. 45,
pp. 1§6-205 as well as the article referred to on Page 2.
*The normative data in Tables 1 and 2 is based on samples of Stanford
University undergraduates (476 males and 340 females) weighted to equalize
the sex distribution.

88
Table 1
T-Scores for Femininity and Masculinity
Based on the 1978 Stanford Sample (Sexes Combined) N"8l6
Femlnlnlty(a)
Masculinity(b)
Femininity(a)
MasculInity(b)
Orlg
Short
Raw
Score
Short
Orlg
Orlg
Short
Raw
Score
Short
Orlg
-15
-5
1.00
2
-8
36
33
A.00
A0
36
-11
1.05
-7
37
A.05
37
-13
-A
1.10
3
-7
38
3A
A.10
Al
38
-12
1.15
-6
39
A. 15
38
-11
-3
1.20
5
-5
39
35
A. 20
A2
39
-II
1.25
-A
AO
X25
~T0~
-10
-2
1.30
6
-A
A1
36
A. 30
A3
A0
-9
1.35
-3
A2
A.35
Al
-9
0
1 .AO
7
-2
A3
38
A.A0
A5
A2
-7
1.A5
-1
AA
A.A5
A3
-6
1
1.50
8
-1
45
39
X50
-5
1.55
0
A5
A.55
AA
-5
2
1.60
10
1
A6
A0
A.60
A7
A5
-A
1.65
1
A7
A.65
A6
-3
3
1.70
11
2
A8
Al
A. 70
A8
A6
-2
1.75
3
~~W~
4.75
~TT
-1
5
1.80
12
A
50
A3
A.80
50
A8
0
1.85
A
51
A.85
A9
1
6
1.90
13
5
51
AA
A.90
51
A9
1
1.95
6
52
A.95
50
2
7
2.00
15
7
53
A5
5.00
52
51
3
2.05
7
5A
5.05
51
A
8
2.10
16
8
55
A6
5.10
53
52
5
2.15
9
56
5.15
53
6
10
2.20
17
10
56
A8
5.20
55
5A
6
2.25
HRT~
57
5.25
5T"
7
11
2.30
18
11
58
A9
5.30
56
55
8
2.35
12
59
5.35
56
9
12
2.A0
20
13
60
50
5.A0
57
57
10
2.A5
13
61
5.A5
57
II
1A
2.50
~TT
nr-
TT
52
5.50
58
55~
12
2.55
15
62
5.55
59
13
15
2.60
22
15
63
53
5.60
60
60
13
2.65
16
6A
5.65
61
1A
16
2.70
23
17
65
5A
5.70
6l
61

89
Table l-continued
Femlnlnfty(a)
Mascullnlty(b)
Femlnlnlty(a)
HasculInlty(b)
Orlg
Short
^ Raw
Score
Short
Orlg
Orlg
Short
Raw
Score
Short
P-fo
lb
T7TT
18
66
5.75
62
16
17
2.80
25
18
67
55
5.80
62
63
17
2.85
19
67
5.85
63
17
19
2.90
26
20
68
57
5.90
63
64
18
2.95
21
69
5.95
65
"T9
20
3.00
27
TT
~Jo~
58
6.00
65
65
20
3.05
22
71
6.05
66
21
21
3.10
28
23
72
59
6.10
66
67
22
3.15
24
73
6.15
68
23
22
3.20
30
24
73
60
6.20
67
68
23
3.25
25
74
6.25
69
24
24
3.30
31
IT-
75
2
6.30
68
70
25
3.35
26
76
6.35
71
26
25
3.AO
32
27
77
63
6.40
70
71
27
3.45
28
78
6.45
72
27
26
3.50
33
29
78
64
6.50
71
73
28
3.55
29
79
6.55
~W
29
27
3.60
35
30
80
65
6.60
72
74
30
3.65
31
81
6.65
75
31
28
3.70
36
32
82
67
6.70
73
76
32
3.75
32
83
6.75
76
33
30
3.8o
37
33
Si-
68
6.80
75
77
34
3.85
34
84
6.85
78
34
31
3.90
38
35
85
69
6.90
76
79
35
3.95
35
86
6.95
79
87
71
7.00
77
80

90
Table 2
T-Scores for the Femininity
minus Masculinity Difference
Short
Dlff.
Or Ig.
Short
Dlff.
Orlg.
Form
SS
Form
Form
SS
Form
T-Score
(a-b)
T-Score
T-Score
(a-b)
T-Score
12
-50
17
51
+ 1
51
13
-49
17
52
+2
51
13
-48
18
52
+3
52
14
-47
19
53
+4
53
15
-46
19
54
+5
53
"75
" -^5
20
55
+6
~W~
17
-44
21
55
+7
55
17
-43
21
56
+8
55
18
-42
22
57
+9
56
19
-41
23
58
+10
57
20
-40
23
IF"
+ 11
57
20
-39
24
59
+12
58
21
-38
25
60
+13
59
22
-37
25
61
+14
59
23
-36
26
61
+15
60
23
-35
~rr
62
+ 16
61
24
-34
27
63
+ 17
61
25
-33
28
64
+ 18
62
26
-32
29
64
+19
63
26
-31
29
65
+20
63
27
-30
30
66
+21
64
28
-29
31
67
+22
65
29
-28
31
67
+23
65
29
-27
32
68
+24
66
30
-26
33
69
+25
67
31
-25
33
70
+26
~TT
32
-24
34
71
+27
68
32
-23
35
71
+28
69
33
-22
35
72
+29
69
34
-21
36
73
+30
70
35
-20
37
~TT
+31
7l
36
-19
37
74
+32
71
36
-18
38
75
+33
72
37
-17
39
76
+34
73
38
-16
39
77
+35
73
39
-15
40
77
" +36
39
-14
41
78
+37
75
40
-13
41
79
+38
75
41
-12
42
80
+39
76
42
-11
43
80
+40
77

91
Table 2-contlnued
Short
Diff.
Orlg.
Short
Diff.
Orlg.
Form
SS
Form
Form
SS
Form
T-Score
(a-b)
T-Score
T-Score
(a-b)
T-Score
42
-10
1*3
81
+41
77
43
-9
44
82
+42
78
44
-8
45
83
+43
79
45
-7
45
83
+44
79
45
-6
46
84
+45
80
A6
-5
47
85
+46
81
~TT
-4
~FT
5T~
+47
5T"
48
-3
48
87
+48
82
48
-2
49
87
+49
83
49
-1
49
88
+50
84
50
0
50

REFERENCES CITED
Bern, S. L. The Measurement of Psychological Androgyny. Journal of
Consulting and Clinical Psychology, 1974, 42(2), 155162.
Bern, S. L. Beyond Androgyny: Some Presumptuous Prescriptions for a
Liberated Sexual Identity. American Psychological Association
National Institute of Mental Health Conference on the ResearcfT"
Needs for Women, Madison, Wisconsin, May 31 1975
Bern, S. L. Administration and Scoring Guide for the Bern Sex-Role Inventory.
Palo Alto, Calif.: Consulting Psychologist Press, Inc., 1979-
Borgstrom, K. F., Whlteley, S., 6 Rudolph, P. Today's Undergraduates:
Are They Really Any Different? Journal of National Association of
Women Deans, Administrators and Counse lors, Spring, 19&0, 43(3),"T-9.
Brovermen, I. Sex-Role Stereotypes: A Current Appraisal. Journal of
Social Issues, 1972, 28, 74-77.
Cross, B. M. The Educated Woman In America. New York: Teachers College
Press, 19^51
Flemming, A. Minorities and Women as Government Contractors. Washington,
D. C.: U. t. Printing Office, May, )9?5.
For Women, A Difficult Climb to the Top. Business Week, August 2, 1969,
42-1*6.
Fretz, C. F., 6 Hayman, J. Progress for WomenMen are Still More Equal.
Harvard Business Review, SeptemberOctober, 1973 5J_* 133" 142.
Herman, J. B., £ Gyllstrom, K. K. Working Men and Women: Inter- and
Intra-Role Conflict. Psychology of Women Quarterly, Summer, 1978,
U4), 319-333.
Horner, M. S. Fall: Bright Women. Psychology Today, 1969, 42, 36-38.
Horner, M. S. Achievement-Related Conflicts in Women. Journal of Social
Issues, 1972, 28, 157-175.
92

93
Jackson, R. A Study of 1975~76 Florida High School Seniors Taking Both
the Florida Twelfth Grade Test and the Scholastic Aptitude Test.
Atlanta, Ga., College Entrance Examination Board, Hay, 1977.
Mueller, K. The Women's Liberation Movement: What Delayed Its Impact.
Journal of National Association of Women Deans, Administrators and
Counselors, Winter, 1977, A0(2), ^3-45.
National Center for Education Statistics. Statistics of Public Elementary
and Secondary Day Schools. Washington, D. C.: U. S. Government
Printing OffIce,"1$78, 18-19.
National Commission on the Observance of International Women's Year.
To Form a More Perfect Union. Washington, D. C.: U. S. Government
Printing Office, June, 1976.
NIe, N. H., Hull, C. H., Jenkins, J. G., Stelnbrenner, K., £ Bent, D. H.
Statistical Package for the Social Sciences. New York: McGraw-Hill
Book Co., 1975.
Peng, S. S. £ Jaffe, J. Women Who Enter Male-Dominated Fields of Study In
Higher Education. American Educational Research Journal, Summer,
1979, .16(3), 285-293.
Powell, G. N., £ Butterfield, D. A. Sex and Sex-Role Identification: An
Important Distinction for Organizational Research. American Psycho
logical Association National Convention, San Francisco, California,
August 26, 1978.
Puig-Casauranc, M. Personality Characteristics and Psychological Androgyny
of Academic Women. Rocky Mountain Psychological Association Annual
Meeting, Albuquerque, New Mexico, May, 1977.
Reil, W. E., Newstrom, M. 6 Monezka, R. Exploding Some Myths About Women
Managers. California Management Review, Summer, 1975 2Z* 72-79.
Schein, V. Sex-Role Stereotyping, Ability and Performance: Prior Research
and New Directions. Personnel Psychology, Summer, 1978, 21 259"268.
Tangri, S. S. Occupational Role Innovation Among Women. Journal of
Social Issues, 1972, 28^ 190-197.
Taylor, E., £ Schavlik, D. Education Programs and Activities. Institutional
Self-Evaluation: The Title IX Requirement. American Counc1 on
Education, Office of Women in Higher Education, Washington, D. C.,
October, 1975.
Taylor, S. S. Educational Leadership: A Male Domain. Phi-Delta Kappan,
October, 1973, 55, 121-128.

Tibbetts, S. Sex-Role Stereotyping: Why Women Discriminate Against Them
selves. Journal of National Association of Women Deans, Administrators
and Counselors, Summer, 1975, 38(*0, 177~^83
Tibbetts, S. Elementary Schools: Do They Stereotype or Feminize? Journal
of National Association of Women Deans, Administrators and Counselors,
Pa 11, 1976, 40(1), 27-33-
TomI1Inson-Keasey, C. Role Variables: Their Influence on Female Moti
vational Constructs. Journal of Counseling Psychology, 197**, 21(3),
332-337.
U. S. Bureau of the Census. Statistical Abstract of the United States.
Washington, D. C.: U. S. Government Printing Office, 1979.
Van Heir, E. J. Sexual Discrimination In School Administration Opportunities.
Journal of National Association of Women Deans, Administrators and
Counselors, Summer, 1975, 38(4), 163-167.
Verheyden-Hl11 lard, M. E. Kindergarten: The Training Ground for Women In
Administration. Journal of National Association of Women Deans,
Administrators and Counselors, Summer, 1975, 38(A), 151~155.
Waldman, E. Changes In the Labor Force Activity of Women. Monthly Labor
Review, June, 1970, 10-18.
Webster's Seventh New Collegiate Dictionary. Springfield, Mass., G. £ C.
Merrlam Company, 1971.
Young, C. L. Women In School Administration and Supervision: A New
Leadership Dimension. NASSP Bulletin, May, 1976, 83-85.

BIOGRAPHICAL SKETCH
Carol Popejoy Wilson was born In Bloomington, Illinois, July 2,
19^1. She moved to Orlando, Florida, In 1950, where she attended grade
school, junior high school and high school. She was granted a Bachelor
of Science degree In business administration from the University of
Florida In 1963 and a Master of Business Administration degree from
Rollins College In 1966.
Carol was employed at the Coca-Cola Company, Foods Division, from
1963 to 1968. She became an Instructor of business administration at the
University of South Florida In 1968. Returning to Orlando, Florida, In
1969, she became an Instructor of business administration at the Univer
sity of Central Florida. Carol was appointed to the position of Dean
of Women In 1971, and Is currently serving as Associate Dean of Students
at the University of Central Florida.
95

I certify that I have read this study and that In tny opinion It
conforms to acceptable standards of scholarly presentation and is fully
adequate, In scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Glenna D. Carr, Chairman
Professor, Subject Specialization
In Teacher Education
I certify that I have read this study and that In my opinion It
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Ei^ene A. Todd
Professor, Subject Specialization
In Teacher Education
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, In scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Arthur J. Lewis
Professor, Instru
and Support
Ij
tional Leadership

! certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and Is fully
adequate, In scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, In scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
'jUioUJA
William H. Drummond
Professor, Instructional Leadership
and Support
This dissertation was submitted to the Graduate Faculty of the Division
of Curriculum and Instruction in the College of Education and to the
Graduate Council, and was accepted as partial fulfillment of the require
ments for the degree of Doctor of Philosophy.
Dean for Graduate Studies and Research



May 6, 1980
Dear
One of our primary goals Is to lend assistance to students enrolled at
the University of Central Florida as they work toward the successful
completion of their academic programs. As part of our effort to assist
you, we need to gather Information that will enable us to develop and
offer programs that will help you reach your educational and professional
goals.
You have been selected as part of a random sample of women students;
consequently, your participation will be critical to the accuracy of
this study. Please be assured that your Involvement will be treated
confidentially and will be used only In combination with others to
compare responses.
Information will be gathered during the end of Spring Quarter. We
would appreciate your taking about 10 to 15 minutes of your time v/iile on
campus to come by Room 282 In the Administration Building any time during
the hours posted below. In return we would be happy to share with you a
summary of the research results and our plans to provide programs to en
hance your professional development.
Thursday,
May 22 10 a.m.
to 3
p.m.
Room
280,
Conference
Room
Friday,
May 23 10 a.m.
to 3
p.m.
Room
280,
Conference
Room
Tuesday,
May 27 10 a.m.
to 3
p.m.
Room
280,
Conference
Room
Wednesday,
May 28 10 a.m.
to 3
p.m.
Room
280,
Conference
Room
Thursday,
May 29 10 a.m.
to 3
p.m.
Room
280,
Conference
Room
Sincerely,
Carol P. Wilson
Dean of Women
CPW:o
79


62
Table 16-contlnued
Actual Group
Predicted Group
Dtscriminant
Subjects
Membership
Membership
Scores***
30
1*
1
0.9984
31
1
1
0.2017
32
1
-0.1695
33
1
1
0.7602
34
2**
2
-2.5037
35
2
2
-0.1481
36
2
] ****
0.1248
37
2
2
-0.5548
38
2
2
-1.9723
39
2
2
-3.3995
40
2
2
-1.2594
41
2
2
0.0087
42
2
] ****
0.5658
43
2
| ****
0.6596
44
2
1 ****
1.0142
45
2
2
-1.1935
46
2
] ****
1.1894
47
2
2
-2.5562
48
2
2
-1.8649
49
2
2
-0.9820
50
2
1****
1.6084
51
2
] ****
0.3359
52
2
2
-0.1308
53
2
2
-0.5588
54
2
] ****
0.6834
55
2
]****
0.2010
56
2
] ****
0.2789
57
2
1****
0.4473
58
2
2
-0.5730
*Women
students
enrolled In the
Col lege of Business
Administration.
**Women
students
who had transferred out of the College of Business
Administration.
***Scores obtained when coefficients are applied to the nine discri
minating variables. Zero to +1 drives the case to Group I, and
a 0 to -1 drives the case to Group II.
****These subjects were not properly classified In their original group.
NOTE: Age, marital status, employment status, grade point average,
student classification, Scholastic Aptitude Test scores, and
Bern Sex-Role Inventory masculine, feminine and difference scores
were the nine discriminating variables.


82
1
I
2
I
3
I
k
1
5
|
6
7
1
1
Never or
almost
never true
!
Usual 1y
not
true
1
Sometimes
but In
frequently
true
1
Occasionally
true
Often
true
Usually
true
Always or
almost
always
Defend my own beliefs
Adaptable
Flatterable
Affectionate
Dominant
Theatrical
Conscientious
Tender
Self-Sufficient
Independent
Conceited
Loyal
Sympathetic
!
Willing to take a
stand
Happy
Moody
Love Children
Individualistlc
i -
Assertive
Tactful
Soft-spoken
Sensitive to needs
of others
Aggressive
Unpredictable
Rel Iable
Gentle
Masculine
Strong personality
Conventional
Gullible
Understanding
Self-reliant
Solemn
Jealous
Yielding
Competitive
Forceful
Helpful
Childlike
Compassionate
Athletic
Likeable
Truthful
Cheerful
Ambitious
Have Leadership
abl1Itles
Unsystematic
Do not use harsh
language
Eager to soothe
hurt feelings
Analytical
Sincere
Secretive
Shy
Act as a leader
Willing to take risks
Inefficient
Feminine
Warm
Make decisions easily
Friendly
a b Class
R.S.
S.S.
a-b
SS diff


ACKNOWLEDGMENTS
This study Is dedicated to my husband, Skip, and my son, Jay, for
their patience, encouragement, and understanding; and to my parents, Ira
and Dorothy Popejoy, for their constant support and concern.
The writer expresses her appreciation to Dr. Glenna Carr, chairman
of the supervisory committee, for her guidance, support, and encouragement.
The writer also expresses her appreciation to the other committee members:
Dr. Eugene Todd, Dr. Fred Goddard, Dr. Arthur Lewis, and Dr. William
Drummond.
The writer Is especially grateful to her colleagues at the Univer
sity of Central Florida: Dr. Charles Dzuiban, Mr. Tom Peeples, Dr. Gordon
McAleer, Dr. Bill Brown, Dr. LeVester Tubbs, Dr. Sandra Guest, Dr. Ralph
Gunter, and Ms. Phyllis Smith, who assisted the writer in the study.
Hi


51
Table 12
Comparison of Undifferentiated, Masculine,
Androgynous, and Feminine Bern Sex-Role
Inventory Classifications by Group
Bern Sex-Role Inventory
Classifications
Group Ia
Percentages of Groups
Group l|k Group lllc Group 1
Undifferentiated
(low masculine and
low feminine)
6.7
7.5
13.6
17.2
Masculine
(low feminine and
high masculine)
13.3
15.0
13.6
62.1
Androgynous
(high feminine and
high masculIne)
^3.3
40.0
45.5
19.0
Feminine
(low masculine and
high feminine)
36.7
37.5
27.3
1.7
aWomen students enrolled In the College of Business Administration.
^Women students who had transferred out of the College of Business
Administration into another college within the University.
S/omen students enrolled In the College of Education.
^Men students enrolled In the College of Business Administration.


53
Twenty-seven and three tenths percent of the women majoring in education
were categorized as feminine, as compared to 36.7 percent of women business
majors, 37.5 percent of women who transferred out of the College of Busi
ness Administration, and 1.7 percent of the men business majors. Men
students majoring In business administration did produce a more masculine
score than did the other groups. Sixty-two and one tenth percent of men
majoring in business administration were categorized as masculine, as com
pared to 13.3 percent of women business majors, 15.0 percent of women who
transferred out of the College of Business Administration, and 13.6 per
cent of the women education majors.
Discriminant Analysis Findings
The data were analyzed through a discriminant analysis.
The mathematical objective of discriminant analysis is to
weigh and linearly combine the discriminating variables In
some fashion so that the groups are forced to be as statisti
cally distinct as possible.
Discriminant analysis attempts to do this by forming one or
more linear combinations of the discriminating variables.
These "discriminant functions" are of the form
d,lZ,
di2Z2
+ d. Z
Ip p
where Dj Is the score on discriminant function I, the d's are
weighting coefficients and the Z's are the standardized values
of the p discriminating variables used in the analysis. The
maximum number of functions which can be derived is either one
less than the number of groups or equal to the number of dis
criminating variables, if there are more groups than variables.
Ideally, the discriminant scores (D's) for the cases within
a particular group will be fairly similar. At any rate, the
functions are formed in such a way as to maximize the separ
ation of the groups. Once the discriminant functions have
been derived, we are able to pursue the two research objectives
of this technique; analysis and class!fication.


9
would enable her to express anger, assert her preferences, trust her
own Judgment, and take control of her life situations, whatever they
might be. According to Dr. Sandra Bern,
For fully effective and healthy himan functioning, both
masculinity and femininity must each be tempered by the
other, and the two must be Integrated Into a more balanced,
a more fully human, a truly androgynous personality. An
androgynous personality would thus represent the very
best of what masculinity and femininity have each come to
represent, and the more negative exaggerations of mascu
linity and femininity would tend to be cancelled out.
(Bern, 1975, p. k)
Need for the Study
If sex-role stereotyping Is detrimental to the successful com
pletion of academic programs and the attainment of educational goals
established by women students, an awareness of the problem by educa
tors Is the first step In helping to eliminate the problem. Today's
women need to be made aware of the expanding choices of academic
programs and Job opportunities available to them. A major responsi
bility of the educator Is to assist students In their efforts to
expand their professional horizons, exercise their full potential in
their chosen fields of study and to take advantage of their freedom
to make choices without being hindered by the detrimental effects of
sex-role stereotyping by the Individual and society.
It Is Important for educators and society at large to challenge
the myths that have kept women In their "place." Daughters, as well
as sons, have to believe that they can achieve their dreams and per
form to the level of their abilities; consequently, "We must encourage


93
Jackson, R. A Study of 1975~76 Florida High School Seniors Taking Both
the Florida Twelfth Grade Test and the Scholastic Aptitude Test.
Atlanta, Ga., College Entrance Examination Board, Hay, 1977.
Mueller, K. The Women's Liberation Movement: What Delayed Its Impact.
Journal of National Association of Women Deans, Administrators and
Counselors, Winter, 1977, A0(2), ^3-45.
National Center for Education Statistics. Statistics of Public Elementary
and Secondary Day Schools. Washington, D. C.: U. S. Government
Printing OffIce,"1$78, 18-19.
National Commission on the Observance of International Women's Year.
To Form a More Perfect Union. Washington, D. C.: U. S. Government
Printing Office, June, 1976.
NIe, N. H., Hull, C. H., Jenkins, J. G., Stelnbrenner, K., £ Bent, D. H.
Statistical Package for the Social Sciences. New York: McGraw-Hill
Book Co., 1975.
Peng, S. S. £ Jaffe, J. Women Who Enter Male-Dominated Fields of Study In
Higher Education. American Educational Research Journal, Summer,
1979, .16(3), 285-293.
Powell, G. N., £ Butterfield, D. A. Sex and Sex-Role Identification: An
Important Distinction for Organizational Research. American Psycho
logical Association National Convention, San Francisco, California,
August 26, 1978.
Puig-Casauranc, M. Personality Characteristics and Psychological Androgyny
of Academic Women. Rocky Mountain Psychological Association Annual
Meeting, Albuquerque, New Mexico, May, 1977.
Reil, W. E., Newstrom, M. 6 Monezka, R. Exploding Some Myths About Women
Managers. California Management Review, Summer, 1975 2Z* 72-79.
Schein, V. Sex-Role Stereotyping, Ability and Performance: Prior Research
and New Directions. Personnel Psychology, Summer, 1978, 21 259"268.
Tangri, S. S. Occupational Role Innovation Among Women. Journal of
Social Issues, 1972, 28^ 190-197.
Taylor, E., £ Schavlik, D. Education Programs and Activities. Institutional
Self-Evaluation: The Title IX Requirement. American Counc1 on
Education, Office of Women in Higher Education, Washington, D. C.,
October, 1975.
Taylor, S. S. Educational Leadership: A Male Domain. Phi-Delta Kappan,
October, 1973, 55, 121-128.


34
Sex
Included In the four groups cited In the study were sixty-three male
business students, all of whom were Included In Group IV, and nlnety-slx
women students In Groups I, II, and III (Table 1).
Age
The 159 participants ranged In age from 18 to 55 years. Approximately
75 percent of the participants were under 25 years of age. Fifteen per
cent of Group I, 10 percent of Group II, 25.9 percent of Group III and
17.5 percent of Group IV ranged In age from 25 to 34 years. Six percent
of Group I, 10 percent of Group II, 8.6 percent of Group III, and 8 per
cent of Group IV ranged In age from 35 to 55 years (Table 2).
Marital Status
The Information on marital status was collected In categories of
single or divorced, married, or widowed. An analysis of the 159 partici
pants revealed that 73 percent were single or divorced, 24.5 percent were
married and 2.5 percent were widowed. An analysis by group showed Group
I, women business students, as having the highest percentage of single or
divorced members. Women education majors (Group III) had the smallest
percent of single or divorced members (Table 3).
Employment
An analysis of the employment status, which was collected In cate
gories of full time, part time, or unemployed, Indicated a high percentage
(78.3) of unemployed members In Group III, women education majors. Group
I had 30.3 percent unemployed, Group II had 40 percent unemployed and
Group IV had 39.7 percent unemployed (Table 4).


CHAPTER 11 I
PROCEDURES USED IN CONDUCTING THE STUDY
The following procedures were used in the study:
1. The sampl-e population Included women students who were
enrolled In the University of Central Florida College of
Business Administration for at least one academic quarter
between January, 1979 and June, 1980, and who were enrolled
at the University of Central Florida during the Spring
Quarter, 1980. This sample population was found by using
University records and official University data tapes.
2. The sample population was stratified Into two groups.
Group I: Those women students enrolled in the College of
Business Administration at the University of
Central Florida during the Spring Quarter, 1980,
who had completed ninety quarter hours of college
study Including at least two courses taught In the
College of Business Administration.
Group II: Those women students enrolled at the University
of Central Florida during the Spring Quarter,
1980, who took at least two courses In business
administration, and who had transferred out of the
College of Business Administration at same point
between January, 1979, and June, 1980.
27


Copyright 1981
by
Carol Popejoy Wilson


adjective and every third one thereafter Is masculine; (2) the second
adjective and every third one thereafter Is feminine; (3) the third ad
jective and every third one thereafter Is filler.
Scoring of the Bern Sex-Role Inventory Is conducted as follows:
The first step Is the calculation of each subject's femininity
("a") and masculinity ("b") scores, which are the averages of
each subjects' ratings of the feminine and masculine adjectives
on the BSRI. That Is, a given subject's femininity score is the
mean of that subject's ratings on the feminine adjectives, and
that same subject's masculinity score is the mean of her or his
ratings on the masculine adjectives. (Bern, 1979, p. 1)
The total score for each subject's ratings on the feminine Items
was divided by twenty, If all characteristics were rated by the subject,
and the same procedure was used for calculating the ratings on the mascu
line characteristics. All subjects rated all Items on the Bern Sex-Role
Inventory. A feminine raw score (a) and a masculine raw score (b) were
the result of this process.
The masculine and feminine standard scores were determined from the
masculine and feminine mean raw scores, using Table 1 in the Administration
and Scoring Guide for the Bern Sex-Role Inventory (Appendix D). A difference
score was computed by subtracting the masculine standard score from the
feminine standard score. The difference score was converted to a differ
ence standard score by using Table 2 in the Administration and Scoring
Guide for the Bern Sex-Role Inventory (Appendix D). The median for the
feminine raw score, A.90, and for the masculine raw score, 4.95, is based
on normative data. The median for the difference standard score, also
based on normative data, Is 0 which was converted to a T-score of 50 by
using Table 2 In the Administration and Scoring Guide for the Bern Sex-
Role Inventory (Appendix D).


6
traditionally been open to males only. The steady Increase In the
number of women entering the labor force Is reflected In the per
centage Increase during the last four years. Women represented
40.5 percent of the civilian labor force tn 1976. This percentage
Increased to 41.7 percent In 1978, and was up to 41.9 percent through
Hay, 1979. (U. S. Bureau of the Census, 1979, p. 392).
Within the one million divorces which occurred In 1975, only
14 percent of the women Involved were awarded alimony. Generally,
child support payments are less than half enough to support the
children. It Is apparent that more women needed to work in 1979
than did In 1976. Limited availability of employment for women,
because of their being confined to the few "traditional" fields of
work for women such as teaching, nursing and secretarial work, has
been exacerbated by sex-role stereotyping In all phases of American
life (National Commission on the Observance of International Women's
Year, 1976, p. 57).
Students of a 11 ages are not shielded from the effects of sex-
role stereotyping. In fact they are often overtly subjected to such
stereotyping all through school and during their formative stages in
life. Ms. Verheyden-Hl11 lard painted the following picture at recess
time on the playing fields at an elementary school.
"John says: 'Get away, we don't want any girls playing.'
Mary says: 'They won't let me play with them.'
Educator 'Well . that's the way boys are. They
says: are too rough, anyway. Play with the
girls." (Verheyden-Hi11 lard, 1975, p. 151)


74
2. It Is recommended that similar studies, using the same measurement
Instruments in combination wlth additional measurement Instruments,
be conducted for the purpose of strengthening or expanding a discrimi
nant function and Increasing Its ability to predict the retention or
transfer of men students In business administration.
3. Due to the limitations of this study to one Institution of higher
education, it is recommended this study be replicated In other
institutions.
4. In counseling women students who designate business administration
as a major field of study, counselors should gather and analyze data
on the nine variables cited in this study for the purpose of counseling
these students.
5. Additional studies should be done to determine the reasons students
transfer out of the College of Business Administration.
6. Studies should be done to find answers to the following questions:
Does the business administration curriculum and the teaching of that
curriculum tend to attract and retain students with certain sex-role
characteristics? If this is true, should efforts be made to change
the curriculum and/or the methods of teaching? Would such changes
produce students who would be more effective in their jobs in the
business community? Is the business administration curriculum
tailored to be more attractive and palatable to those students who
tend to have more masculine characteristics than androgynous or
feminine characteristics?


PERSONAL INFORMATION FORM
Social Security #
Phone #
Name:
(Last) (First) (Middle/Maiden)
Address at which you can be reached:
Spring Quarter, 1980
(Street) (City) (State) (Zip)
Summer Quarter, I98O
(Street) (Ctty) (State) (ZI p')
Blrthdate: Age: ______________________
Marital Status: __________ Single/Divorced Married Widowed
Employment outside the home: Ful 1 Time Part Time Unemployed
Occupation:
Number of Children Grade Point Average
(Overall College Work, Including
Junior College, If Applicable)
Student Classification: Freshman Sophomore Junior Senior
(Circle one)
College at U C F In which you are currently enrolled
If you have been enrolled In a college at U C F other than the one In
which you are now enrolled, please list those college In order of Enrollment
High School from which you graduated:
OVER
Year
76


Number of Students in
Groups I, II, III, and IV with
Children
Number of
Children
Group Ia
(N-33)
Group II*3
(N-40)
Group lllC
(N-23)
Group IV^
(N-63)
One
1
3
3
1
Two
1
1
5
6
Three
1
2
0
3
Four
0
0
0
1
Five
0
1
0
0
Six
0
1
0
0
None
30
32
15
52
aWomen students enrolled in the College of Business Administration.
^Women students who had transferred out of the College of Business
Administration Into another college within the University.
S/omen students enrolled In the College of Education.
^Men students enrolled in the College of Business Administration.


35
Table 1
Sex of Students In
Groups I, II, III and IV
Number of Students Percentage of Total Students
Male Female Used In the Study
Major Groups
Group Ia
0
33
20.8
Group II*5
0
AO
25.2
Group III0
0
23
Comparison
Groups
1A.5
Group IV
d
63 0
39.5
aWomen students enrolled In the College of Business Administration.
^Women students who had transferred out of the College of Business
Administration into another college within the University.
Women students enrolled in the College of Education.
^Men students enrolled In the College of Business Administration.


BEM INVENTORY
Developed by Sandra L. Bem.Ph.D.
Name Age Sex
Phone No. or Address
Date 19
If a student: School ______________________________ Yr. 1 School
If not a student: Occupation
DIRECTIONS
On the opposite side of this sheet, you will find listed a number of per
sonality characteristics. We would like you to use those characteristics
to describe yourself, that Is, we would like you to Indicate, on a scale
from 1 to 7* how true of you each of these characteristics Is. Please do
not leave any characteristic unmarked.
Example: sly
Write a 1 If It
Write a 2 If It
Write a 3 If ft
Write a A If It
Write a 5 If It
Write a 6 If It
Write a 7 If It
Is never or almost never true that you are sly.
Is usually not true that you are sly.
Is sometimes but Infrequently true that you are sly.
Is occasionally true that you are sly.
Is often true that you are sly.
Is usually true that you are sly.
Is always or almost always true that you are sly.
Thus, If you feel It Is sometimes but Infrequently true that you are "sly,"
never or almost never true that you are "malicious," always or almost al
ways true that you are "Irresponsible," and often true that you are "care
free," then you would rate these characteristics as follows:
Sly
3
1rrespons ible
7
Malicious
1
Carefree
5
CONSULTING PSYCHOLOGISTS PRESS, INC.
577 College Avenue Palo Alto, California 9^306
Copyright, 1978, by Consulting Psychologists Press, Inc. All rights re
served. Duplication of this form by any process Is a violation of the
copyright laws of the United States except when authorized In writing by
the PublIsher.
81


APPENDIX D
ADMINISTRATION AND SCORING GUIDE FOR
THE BEM SEX-ROLE INVENTORY


19
the question of whether women who have experienced these traditional
roles experienced less conflict about achievement (TomlIson-Keasey,
197*0. Subjects In the TomlIson-Keasey study were women students from
Douglass College and University College at Rutgers Untverslty whose
enrolled women students differed In terms of age, marital status and
number of children. A young unmarried sample was drawn from Douglass
College and an older married sample was drawn from the University College.
The procedure followed Homer's study by using a cue response, and the
results Indicated a significantly different level of fear of success.
In the study It was hypothesized that women who were performing as
mothers and wives, which were perceived as appropriate achievements,
experienced less fear of success and reduced anxiety than did those
women who were not performing these "appropriate" roles. The results
were as follows:
... women who had satisfied the mother and the wife roles
are not as likely to evidence fear of success. . unmarried
subjects had significantly more fear of success . than did
married subjects . and women with no children had signi
ficantly more fear of success ... than did women with children.
(Toml Inson-Keasey, 197**, p. 235)
Evolution of Sex Roles
The changing of sex roles has been slow moving. Myth and fact
are Intertwined. Some very significant "ole wives tales," however,
are laid to rest (Tangrl, 1972). For example, data Indicate that
women who seek to enter predominantly male-dominated professions do not
show evidence of having Identified with their fathers over their
mothers. Daughters who have more educated working mothers In male-
dominated occupations will view their mothers as role models. Tangrl


87
Femininity
Score
Masculinity Score
Below Median Above Median
Below
Undifferentiated
Mascullne
Median
(low-low)
(low fern.-high mase.
Above
Feminine
Androgynous
Median
(High fern.-low mase.)
(high fern.-high mase.)
Based on the normative data
the medians are as follows:
on Stanford students
(sexes combined),
Original Form
Short Form
Femininity RS
4.90
5.50
Masculinity RS
4.95
4.80
For a discussion of the significance of the classification, see Bern,
S. L. On the Utility of Alternative Procedures for Assessing Psychological
Androgyny. Journal of Consulting and Clinical Psychology, 1977 Vol. 45,
pp. 1§6-205 as well as the article referred to on Page 2.
*The normative data in Tables 1 and 2 is based on samples of Stanford
University undergraduates (476 males and 340 females) weighted to equalize
the sex distribution.


30
classes In the College of Education at the University of
Central Florida who had never been enrolled In the College
of Business Administration (Group III) completed the Bern
Sex-Role Inventory In their classrooms. These data were
Included In the study to address questions surrounding the
concept of the traditionally sex-typed person as follows:
(1) Did women students majoring In education, a traditionally
feminine field of study, produce a more feminine score on the
Bern Sex-Role Inventory than did the other groups In the study?
(2) Did men students majoring In business administration, a
traditionally masculine field of study, produce a more mascu
line score on the Bern Sex-Role Inventory than did the other
groups In the study?


APPENDIX A
DEMOGRAPHIC PROFILE DATA SHEET


38
Table 4
Employment Status of Students
In Groups I, II, III, and IV
Status
Group 1
Number of
and Percentages
Ia Group llb
Cases
of Groups
Group 111c
Groi^j IV^
Full Time
Number
2
7
1
9
Percent
6.1
17.5
4.3
14.3
Part Time
Number
21
17
4
29
Percent
63.6
42.5
17.4
46.0
Unemployed
Number
10
16
18
25
Percent
30.3
AO.O
78.3
39.7
aWomen students enrolled In the College of Business Administration.
^Women students who had transferred out of the College of Business
Administration Into another college within the University.
S/omen students enrolled In the College of Education.
^Men students enrolled In the College of Business Administration.


2k
Limited opportunities to acquire work-related power
acquisition behaviors and exclusion from political/
Influence networks within organizations can, then, limit
the performance effectiveness of the women with the
potential to do well, as well as, In the long run diminish
her motivation to perform. As such, a prime research target
should be the role of power acquisition behaviors In effec
tive managerial functioning and the effect of sex and sex-
role stereotypical thinking on acquisition of and oppor
tunity to use these behaviors. (Scheln, 1978, p. 266)
Jeanne Herman and Karen Gyllstrom discovered through their recent
study of 500 employees stratified by sex and Job classification In a
major mldwestern university that myths are dispelled concerning Inter-role
conflict for women employed full-time. "A potential employer could not
justify hiring a male over an equally competent female on the grounds
that the woman will experience greater Interrole conflict." The results
show that married women with children who work full time report no more
conflict between job and family responsibilities than do men who hold a
comparable number of social roles (Herman and Gyllstrom, 1978, p. 330).
Sex Roles of Women In Education
Taylor concluded In her doctoral dissertation that, all things
being equal, superintendents of schools are not likely to hire women
as school administrators. The researcher concluded that the only
factor having any real and constant significance of who will be hired
and who will not be hired Is the sex of the applicant (Taylor, 1973
p. 128).
The percentage of women elementary school administrators during
the past forty years has steadily declined. Information on the high
school administrative level showed the same trend. Data provided by
the National Council of Administrative Women In Education Indicated
the following:


BIOGRAPHICAL SKETCH
Carol Popejoy Wilson was born In Bloomington, Illinois, July 2,
19^1. She moved to Orlando, Florida, In 1950, where she attended grade
school, junior high school and high school. She was granted a Bachelor
of Science degree In business administration from the University of
Florida In 1963 and a Master of Business Administration degree from
Rollins College In 1966.
Carol was employed at the Coca-Cola Company, Foods Division, from
1963 to 1968. She became an Instructor of business administration at the
University of South Florida In 1968. Returning to Orlando, Florida, In
1969, she became an Instructor of business administration at the Univer
sity of Central Florida. Carol was appointed to the position of Dean
of Women In 1971, and Is currently serving as Associate Dean of Students
at the University of Central Florida.
95


20
also found that women In nontrad!tlonal fields or occupations do not
reject the traditional female roles of wife and mother. They do plan,
however, to postpone marriage and have fewer children than do the more
traditional women. Also they do not view themselves as masculine women
(Tangri, 1972, p. 196).
Studies reveal findings that vividly point out the existence in
our contemporary society of "clearly defined sex-role stereotypes for
men and women contrary to the phenomenon of unisex roles currently
touted In the media" (Broverman, 1972, p. 75). In his examination of
the relationship between mothers' employment status and sex-role per
ceptions of college students, daughters of employed mothers perceived
fewer differences between men and women than did daughters of home
maker mothers. Also, "since more feminine traits are negatively
valued than are masculine traits, women tend to have more negative
self-concepts than do men" (Broverman, 1972, p. 750). The double
standard emerges. According to Broverman:
If women adopt the behaviors specified as desirable for
adults, they risk censure for their failure to be appro
priately feminine, but if they adopt the behaviors that
are designated as feminine, they are necessarily deficient
with respect to the general standards of adult behavior.
(Broverman, 1972, p. 75)
Sex Roles of Women in Business
Few significant studies In the area of women In business have
been found. There is an abundance of opinion and testimony, however,
on the topic. For example, an article appeared In Business Week
Magazine which stressed the barriers are still up for women at the


71
Eleven variables for students In Groups I and II were subjected to a
discriminant analysis technique and nine of these variables surfaced as
discriminating at the .05 level of significance. Listed according to their
relative Influence In the predid ton of group membership (either Group I
or Group II), the nine discriminating variables were, (1) student classi
fication, (2) marital status, (3) Bern Sex-Role Inventory feminine standard
score, (4) Scholastic Aptitude Test score (5) Bern Sex-Role Inventory dif
ference standard score, (6) employment status, (7) age, (8) grade point
average and (9) Bern Sex-Role Inventory masculine standard score. Although
age, academic ability, masculinity, femininity, and androgyny (as deter
mined through the difference score on the Bern Sex-Role Inventory) were
variables that surfaced as discriminating In the analysis, only femininity
(feminine standard score) realized a coefficient with relatively strong
Influence on the prediction of group membership.
Analyzing of data on the nine discriminating variables for a new
woman student, who as a freshman selected business administration as
her major field of study, will enable a prediction to be made regarding the
probability of her remaining In the College of Business Administration.
A classification made as a result of the discriminant analysis of her data
that would put her In the group which would Indicate she would tend to
remain In the College of Business Administration could be said to be
correct 69.7 percent of the time. A classification made as a result of the
discriminant analysis of her data which put her In the group that would
Indicate she would tend to transfer out of the College of Business Adminis
tration could be said to be correct 62.5 percent of the time. This


37
Table 3
Marital Status of Students
In Groups I, II, III, and IV
Number of Students and Percentages of Groups
Group Ia Group II*3 Group 111C Group IV^
Single/Divorced
Number
27
29
12
48
Percent
81.8
72.5
52.2
76.2
Married
Number
6
9
9
15
Percent
18.2
22.5
39.1
23.8
Widowed
Number
0
2
2
0
Percent
0
5.0
8.7
0
tornen students enrolled In the College of Business Administration.
^Women students who had transferred out of the College of Business
Administration Into another college within the University.
S/omen students enrolled In the College of Education.
^Men students enrolled In the College of Business Administration.


90
Table 2
T-Scores for the Femininity
minus Masculinity Difference
Short
Dlff.
Or Ig.
Short
Dlff.
Orlg.
Form
SS
Form
Form
SS
Form
T-Score
(a-b)
T-Score
T-Score
(a-b)
T-Score
12
-50
17
51
+ 1
51
13
-49
17
52
+2
51
13
-48
18
52
+3
52
14
-47
19
53
+4
53
15
-46
19
54
+5
53
"75
" -^5
20
55
+6
~W~
17
-44
21
55
+7
55
17
-43
21
56
+8
55
18
-42
22
57
+9
56
19
-41
23
58
+10
57
20
-40
23
IF"
+ 11
57
20
-39
24
59
+12
58
21
-38
25
60
+13
59
22
-37
25
61
+14
59
23
-36
26
61
+15
60
23
-35
~rr
62
+ 16
61
24
-34
27
63
+ 17
61
25
-33
28
64
+ 18
62
26
-32
29
64
+19
63
26
-31
29
65
+20
63
27
-30
30
66
+21
64
28
-29
31
67
+22
65
29
-28
31
67
+23
65
29
-27
32
68
+24
66
30
-26
33
69
+25
67
31
-25
33
70
+26
~TT
32
-24
34
71
+27
68
32
-23
35
71
+28
69
33
-22
35
72
+29
69
34
-21
36
73
+30
70
35
-20
37
~TT
+31
7l
36
-19
37
74
+32
71
36
-18
38
75
+33
72
37
-17
39
76
+34
73
38
-16
39
77
+35
73
39
-15
40
77
" +36
39
-14
41
78
+37
75
40
-13
41
79
+38
75
41
-12
42
80
+39
76
42
-11
43
80
+40
77


25
... in the eleven-year periods between 1950-51 and 196061,
the number of women serving as junior high and senior high
school principals dropped from 18$ to 3*8$ for all secondary
schools. (Van Meir, 1975, p. 163)
More recent statistics cited by the National Center for Education
Statistics revealed that In the State of Florida In 1977 23 percent
of public school principals and assistant principals were women
(National Center for Education Statistics, 1978, p. 19).
Young (1976) 'researched popular myths in public education con
cerning women in the field of educational leadership at the elementary
and secondary level. For each myth cited, contrary evidence evolved
from the study. Several of the most quoted myths that were disproved
are (1) women rarely possess the credentials for administrative
positions, (2) women seldom aspire or apply for administrative posi
tions, (3) women will not make the required commitment to an adminis
trative position, and (4) women do not have to work because they are
merely after a second income (Young, 1976, p. 83).
Borgstrom, Whiteley, and Rudolph studied the Impact of recent
societal changes on women's educational aspirations. The researchers
utilized data from a student survey administered by the American
Council on Education. They compared the data with survey responses
of freshman students at the University of California, Santa Barbara
(UCSB), from 1968 to 1978. They were attempting to characterize the
University of California, Santa Barbara, campus culture today and show
how and when UCSB students have changed over the past ten years.


CHAPTER IV
ANALYSIS OF DATA
A total of 159 subjects were Included In the study. These subjects
were divided Into four groups as follows: Group I (thirty-three subjects)
was composed of a sample of women enrolled In the College of Business
Administration at the University of Central Florida during the spring
quarter, 1980, all of whom had completed ninety quarter hours of college
study, Including at least two courses taught In the College of Business
Administration. Group II (forty subjects) was made up of a sample of
women students enrolled at the University of Central Florida during the
spring quarter, 1980, all of whom had taken at least two courses In busi
ness administration, and who had transferred out of the College of Business
Administration at some point between January, 1979 and June, 1980. Group
III (twenty-three subjects) was a sample of women students enrolled In
selected classes In the College of Education who had never been enrolled
In the College of Business Administration. Group IV (sixty-three subjects)
was a sample of male students enrolled In classes in the College of Busi
ness Administration during the spring quarter, 1980.
Groups I and II were the groups upon which this study was focused.
Groups III and IV were Included In the study as a means of providing com
parative data; consequently, Groups III and IV were not used In the dis
criminant analysis portion of the study, but only In the descriptive
analysis. Groups III and IV were Included to address questions surrounding
31


58
Table 14
Means and Standard Deviations
for Discriminating Variables at .05 Level
of Probability for Groups I and II
Di scrtminating
Variables
Group
Mean
Ia
S.D.C
Group
Mean
h
S.D.
Age, (range 18 to 55)
23.21
6.10
23.70
7.71
Marital Status^
1.18
0.39
1.33
0.57
Employment Statusc
2.24
0.56
2.23
0.73
Grade Point Average^
2.98
0.44
2.91
0.56
Student Classification
3.73
0.52
3.03
1.07
SAT Scores9
987.12
141.42
930.50
183.35
0
Women students enrolled in the College of Business Administration.
^Women students who had transferred out of the College of Business
Administration into another college with the University.
CS.D. Is the abbreviation for standard deviation.
^Marital status was recorded as single or divorced (Value 1),
married (Value 2), or widowed (Value 3).
0
Employment status was measured as full time (Value 1), part time
Value 2), or unemployed (Value 3).
^Grade point averages were computed on a four point scale (4.0 A grade).
^Scholastic Aptitude Test scores are normally computed on a maximum
range from 500 to 1500 points or higher. The current minimum
score for admission to the University of Central Florida is 800.


15
Because the reliability of the Androgyny t ratio could not
be calculated directly, coefficient alpha was computed for
the highly correlated Androgyny difference score, femininity
masculinity, using the formula provided by Nunnally (1967)
for linear combinations. The reliability of the androgyny
difference score was .85 for the Stanford sample and .86
for the Foothill sample. (Bern, 197**, p. 159)
In an effort to determine test-retest reliability, the Bern
Sex-Role Inventory was administered for a second time to 28 males and
28 females from the Stanford normative sample. The second adminis
tration of the Inventory took place approximately four weeks after
the first administration.
During this second administration, subjects were told that
we were Interested In how their responses on the test might
vary over time, and they were explicitly Instructed not to
try to remember how they had responded previously. Product-
moment correlations were computed between the first and
second administrations for the Masculinity, Femininity,
Androgyny, and Social Desirability scores. All four scores
proved to be highly reliable over the four-week Interval
(Masculinity r 90; Feminintty r .90; Androgyny r 93;
Social Desirability r .89). (Bern, 197**, p. 160)
Demographic Profile
The demographic profile data sheet was designed to elicit perti
nent background Information from the subjects. Most of the information
requested on the demographic profile data sheet also was collected by
the Investigator from the student's permanent file and University
data tapes. Collecting the same Information from two sources was
Intended to provide a cross check on the data.
The following Information was available only from the demographic
profile data sheet: Address for Sumer Quarter; marital status,
employment outside the home; occupation; and number of children.


57
"The ratio level of measurement has all of the properties of an
Interval level with the additional property that the zero point Is In
herently defined by the measurement scheme" (Nie, 1975, p. 5). Age In
dicates the same or fixed distance between age 18 and 19 or age 25 and
26, and has the additional property of a zero point. Table 13 shows the
values or scores assigned to each variable and the level of measurement
used in each assignment.
The eleven variables subjected to the direct method Included seven
Items from the demographic profile data sheet (1) age, (2) marital status,
(3) employment status, (A) grade point average, (5) student classification,
(6) number of children, and (7) year of high school graduation, one item
from the permanent University file (Scholastic Aptitude Test score or Its
equivalent), and the three standard scores derived from the personality
characteristics on the Bern Sex-Role Inventory (masculine standard score,
feminine standard score, and difference standard score). Items six and
seven from the demographic profile data sheet were rejected, leaving nine
variables contributing to the discriminant function.
The nine discriminating variables which surfaced in the analysis as
contributing to the discriminant function were (1) age of the participant,
(2) marital status, (3) employment status, (4) grade point average, (5)
student classification, (6) Scholastic Aptitude Test score, (7) Bern Sex-
Role Inventory masculine standard score, (8) Bern Sex-Role Inventory feminine
standard score, and (9) Bern Sex-Role Inventory difference standard score.
The means and standard deviations for the nine discriminating vari
ables are presented in Tables 1k and 15.


63
Table 16-continued
Subjects
Actual Group
Membership
Predicted Group
Membership
Discriminant
Scores***
59
2**
1****
O.A533
60
2
2
-1.0115
61
2
]****
0.1311
62
2
2
-0.7055
63
2
2
-1 .A451
61*
2
2
-0.2870
65
2
2
-0.3365
66
2
2
-1.21*36
67
2
2
-0.7151
68
2
2
-2.1282
69
2
2
-2.5279
70
2
] ****
1.0655
71
2
2
-0.591*8
72
2
2
0.2116
73
2
1 ****
0.9951
*Women students enrolled In the College of Business Administration.
**Women students who had transferred out of the College of Business
Administration.
***Scores obtained when coefficients are applied to the nine discri
minating variables. Zero to +1 drives the case to Group I, and
a 0 to -1 drives the case to Group II.
****These subjects were not properly classified In their original group.
NOTE: Age, marital status, employment status, grade point average,
students classification, Scholastic Aptitude Test scores, and
Bern Sex-Role Inventory masculine, feminine and difference scores
were the nine discriminating variables.


Approximately 43 percent of the forty students In Group II, those
women students who transferred out of the College of Business Adminis
tration, responded to the open-ended question. Four broad categories of
answers emerged In the process of grouping the participants' responses,
and all responses fell Into these four categories. They were (1) Inde
cision as to field of study, (2) change of career Interests, (3) expediency
of completion of a degree In another field of study, and (4) boredom or
lack of enthusiasm for the business curriculum.
The category receiving the greatest response was "boredom or lack of
enthusiasm for the business curriculum." Forty-one percent, seven out of
the seventeen responding to the question, were assigned to this category
(Table 10). None of the respondents stated that the academic work in
business administration was too difficult or that they transferred out of
business administration because they were getting low grades In the busi
ness courses.
Descriptive Analysis of
Bern Sex-Role Inventory
Sixty personality characteristics are listed on the Bern Sex-Role
Inventory. Twenty of the characteristics are stereotypically feminine,
twenty are stereotypica1ly masculine and twenty serve as filler Items.
Participants in the study indicated how well each of the sixty personality
characteristics described herself or himself on a scale from one to seven.
The scale ranged from one, "Never or almost never true," to seven, "Always
or almost always true." The sixty personality characteristics are placed
on the Bern Sex-Role Inventory in the following manner: (1) The first


THE CHARACTERISTICS OF MASCULINITY,
FEMININITY, ANDROGYNY, AGE, AND ACADEMIC ABILITY
AS FACTORS IN THE RETENTION OF WOMEN STUDENTS
IN COLLEGE PROGRAMS IN BUSINESS ADMINISTRATION
BY
CAROL POPEJOY WILSON
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
1981


I certify that I have read this study and that In tny opinion It
conforms to acceptable standards of scholarly presentation and is fully
adequate, In scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Glenna D. Carr, Chairman
Professor, Subject Specialization
In Teacher Education
I certify that I have read this study and that In my opinion It
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Ei^ene A. Todd
Professor, Subject Specialization
In Teacher Education
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, In scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Arthur J. Lewis
Professor, Instru
and Support
Ij
tional Leadership


43
Table 7
Scholastic Aptitude Test Scores
for Students in Groups I and II
Percentages of Groups
Ranges for .
SAT Scores3 Group lD Group llc
590
- 800
03
18
801
- 1000
51
50
1001
- 1200
36
24
1201
- 1350
09
08
aSAT Is the abbreviation for Scholastic Aptitude Test. The highest
possible score is 1600.
^Women students enrolled in the College of Business Administration.
S/omen students who had transferred out of the College of Business
Administration Into another college within the University.


Table 11
Interpretation of Categories
Derived from the
Bern Sex-Role Inventory
Masculine Score
Below Median
Above Median
Below
Median
Undifferentiated
(low feminine low masculine)
Masculine
(low; feminine -
high masculine)
Feminine
Score
Above
Femlnlne
Androgynous
Median
(high feminine low masculine)
(high feminine
- high masculIne)
Medians Based on Normative Data
Feminine Raw Score 4.90
Masculine Raw Score A.95
vn
o


49
Each subject was classified into one of four categories based on the
median raw scores of the normative sample and the subject's mean raw scores.
All participants In the study were placed In one of the following four cate
gories: (1) masculine, which denotes a low feminine score and high mas
culine score, (2) feminine, which denotes a low masculine score and a high
feminine score, (3) androgynous, which denotes a high masculine score and
a high feminine score, or (4) undifferentiated, which denotes a low masculine
score and a low feminine score (Table 11). "High difference scores In
either direction Indicate a tendency to be strongly sex-typed, positive
scores Indicating femininity and negative scores indicating masculinity"
(Bern, 1979 p. 3). Positive scores indicate a generally more feminine
orientation, negative scores a more masculine orientation, and scores near
zero an androgynous orientation.
Thirteen and three tenths percent of the subjects In Group I (women
students enrolled In the College of Business Administration), 15.0 percent
of the subjects In Group II (women students who had transferred out of the
College of Business Administration), and 13.6 percent of the subjects In
Group III (women students enrolled in the College of Education) were placed
In the masculine category as a result of their scores on the Bern Sex-Role
Inventory. Sixty-two and one tenth percent of the subjects In Group IV
(men students enrolled in the College of Business Administration) were
placed In the masculine category (Table 12).
Thirty-six and seven tenths percent of women students enrolled in the
College of Business Administration (Group I) were classified as feminine
as a result of their scores on the Bern Sex-Role Inventory. Thirty-seven


14
were selected as feminine or masculine on the basis of
cultural definitions of sex-typed social desirability and
not on the basis of differential endorsement by females and
males, i.e., a characteristic qualified as feminine If it
was Judged to be traditionally more desirable In American
society for a woman than for a man, and It qualified as
masculine if It was Judged to be more desirable in American
society for a man than for a woman.
The BSRI is essentially self-administering and may be
given to large groups as well as to Individuals. It has
also been administered by mail in several studies. The
inventory consists of 60 adjectives and phrases printed
on a single sheet with Instructions and space for personal
Information about the subject on the reverse side. (Bern,
1979, PP. 1-3)
The Bern Sex-Role Inventory initially was administered to 444
male and 279 female students In Introductory psychology at Stanford
University during the Winter and Spring of 1973. It was also adminis
tered to an additional 117 male and 77 female students at Foothill
Junior College. The data that these students provided represent the
normative data for the Bern Sex-Role Inventory.
Subjects were classified as sex typed, whether masculine or
feminlnine, If the androgyny t ratio reached statistical significance
(/t/ _> 2.025, df 38, p < .05), and they are classified as androgynous
If the absolute value of the t ratio Is less than or equal to one
(Bern, 1974, p. 161).
Dr. Bern estimated the Internal consistency of the Bern Sex-Role
Inventory by computing the coefficient alpha separately for the mas
culinity, femininity, and social desirability scores of the siijects
In each of the two normative samples. The results Indicated all
three scores to be highly reliable, both In the Stanford sample
"masculinity a .86; femininity a .80; social desirability a
.70" (Bern, 1974, p. 159).


89
Table l-continued
Femlnlnfty(a)
Mascullnlty(b)
Femlnlnlty(a)
HasculInlty(b)
Orlg
Short
^ Raw
Score
Short
Orlg
Orlg
Short
Raw
Score
Short
P-fo
lb
T7TT
18
66
5.75
62
16
17
2.80
25
18
67
55
5.80
62
63
17
2.85
19
67
5.85
63
17
19
2.90
26
20
68
57
5.90
63
64
18
2.95
21
69
5.95
65
"T9
20
3.00
27
TT
~Jo~
58
6.00
65
65
20
3.05
22
71
6.05
66
21
21
3.10
28
23
72
59
6.10
66
67
22
3.15
24
73
6.15
68
23
22
3.20
30
24
73
60
6.20
67
68
23
3.25
25
74
6.25
69
24
24
3.30
31
IT-
75
2
6.30
68
70
25
3.35
26
76
6.35
71
26
25
3.AO
32
27
77
63
6.40
70
71
27
3.45
28
78
6.45
72
27
26
3.50
33
29
78
64
6.50
71
73
28
3.55
29
79
6.55
~W
29
27
3.60
35
30
80
65
6.60
72
74
30
3.65
31
81
6.65
75
31
28
3.70
36
32
82
67
6.70
73
76
32
3.75
32
83
6.75
76
33
30
3.8o
37
33
Si-
68
6.80
75
77
34
3.85
34
84
6.85
78
34
31
3.90
38
35
85
69
6.90
76
79
35
3.95
35
86
6.95
79
87
71
7.00
77
80


10
the growing recognition by women and men that equal rights Is a matter
of simple Justice" (National Commission on the Observance of Inter
national Women's Year, 1976, p. v).
Women students who chose business as their profession, which is
still a nontradltJonal field of study for women today, should be free
to strive for top level positions within the profession. Women stu
dents should be freed from the direct and subtle effects of sex-role
stereotyping In their efforts to complete their academic programs In
such nontradltional fields of study as business administration.
Statement of the Problem
It was the purpose of this study to determine whether the charac
teristics of masculinity, femininity, androgyny, age, and academic
ability are factors which correlate with the retention of women stu
dents In programs of study In the College of Business Administration
at the University of Central Florida.
Definition of Terms
The following definitions apply to the terms as they are used
In this study.
Stereotype: An unvarying form or pattern, having no Individuality,
as though cast from a mold.
Sex-role stereotypes: Socially designated behaviors that differen
tiate between men and women; women commonly possess a common pattern
of conventional expression, notion, character, or mental attitude,
Just as men do, but the attitudes and patterns differ greatly.


88
Table 1
T-Scores for Femininity and Masculinity
Based on the 1978 Stanford Sample (Sexes Combined) N"8l6
Femlnlnlty(a)
Masculinity(b)
Femininity(a)
MasculInity(b)
Orlg
Short
Raw
Score
Short
Orlg
Orlg
Short
Raw
Score
Short
Orlg
-15
-5
1.00
2
-8
36
33
A.00
A0
36
-11
1.05
-7
37
A.05
37
-13
-A
1.10
3
-7
38
3A
A.10
Al
38
-12
1.15
-6
39
A. 15
38
-11
-3
1.20
5
-5
39
35
A. 20
A2
39
-II
1.25
-A
AO
X25
~T0~
-10
-2
1.30
6
-A
A1
36
A. 30
A3
A0
-9
1.35
-3
A2
A.35
Al
-9
0
1 .AO
7
-2
A3
38
A.A0
A5
A2
-7
1.A5
-1
AA
A.A5
A3
-6
1
1.50
8
-1
45
39
X50
-5
1.55
0
A5
A.55
AA
-5
2
1.60
10
1
A6
A0
A.60
A7
A5
-A
1.65
1
A7
A.65
A6
-3
3
1.70
11
2
A8
Al
A. 70
A8
A6
-2
1.75
3
~~W~
4.75
~TT
-1
5
1.80
12
A
50
A3
A.80
50
A8
0
1.85
A
51
A.85
A9
1
6
1.90
13
5
51
AA
A.90
51
A9
1
1.95
6
52
A.95
50
2
7
2.00
15
7
53
A5
5.00
52
51
3
2.05
7
5A
5.05
51
A
8
2.10
16
8
55
A6
5.10
53
52
5
2.15
9
56
5.15
53
6
10
2.20
17
10
56
A8
5.20
55
5A
6
2.25
HRT~
57
5.25
5T"
7
11
2.30
18
11
58
A9
5.30
56
55
8
2.35
12
59
5.35
56
9
12
2.A0
20
13
60
50
5.A0
57
57
10
2.A5
13
61
5.A5
57
II
1A
2.50
~TT
nr-
TT
52
5.50
58
55~
12
2.55
15
62
5.55
59
13
15
2.60
22
15
63
53
5.60
60
60
13
2.65
16
6A
5.65
61
1A
16
2.70
23
17
65
5A
5.70
6l
61


2
"All who assist students in making educational and career choices
are likely to Influence, consciously or unconsciously, the decisions
of some students" (Taylor and Schavlik, 1975, p. 27).
Without a doubt, the thrust of Title IX should provide an incen
tive and push institutions of higher education in that direction of
eliminating Inequality which Is based on sex-role stereotyping. What
took the American society so long to recognize the stunting effects
sex stereotyping has had on women? According to Kate Mueller, women
had to wait until the 1970's for the women's liberation movement to
awaken the public to women's problems and needs. This wait for women
to begin to ask for equal rights, in Mueller's opinion, can be traced
to several factors.
1. In the 1950's many of the women indicated they were not
too unhappy with their own situations.
2. A women's movement would mean full utilization of birth
control, Including the birth control pill, so that maternal
responsibilities could be carefully planned. It took
awhile for the birth control pill to be accepted.
3. It wasn't until the later 1960's that women's groups
learned how to organize, and how to apply verbal protests.
A. New ways of looking at love and new attitudes and behavior
about sex relations have slowly developed during the last
forty years. Along with these new attitudes and behaviors
came planned parenthood and lower birthrates.


LIST OF TABLES
Table Page
1. Sex of Students In Groups I, II, III and IV 35
2. Ages of Students In Groups I, II, III and IV 36
3. Marital Status of Students In Groups I, II,
III, and IV 37
4. Employment Status of Students In Groups I,
II, III, and IV 38
5. Grade Point Averages for Students In
Groups I, II, III, and IV 40
6. Students Classifications In Groups I, II,
III and IV Al
7. Scholastic Aptitude Test Scores for Students
In Groups I and II 43
8. Number of Students In Groups I, II, III, and
IV with Children 44
9. Year of High School Graduation for Students In
Groups I, II, III, and IV 45
10. Responses by Group II Students to the Four
Categories Derived from the Open-ended
Question on the Demographic Profile Data
Sheet 47
11. Interpretation of Categories Derived from
the Bern Sex-Role Inventory 50
12. Comparison of Undifferentiated, Masculine, Androgynous
and Feminine Bern Sex-Role Inventory Classifications
by Group 51
v


69
ability variable, along with the remaining eight discriminating vari
ables, resulted In the successful classification of 69.7 percent of
cases In Group I and 62.5 percent of the cases In Group II (Table 13).


60
The discriminant function coefficients were computed for the nine
discriminating variables. The coefficients represent the relative con
tributions of the nine discriminating variables to the equation (discri
minant function). They serve to identify the variables which contributed
most to the discriminant function. The coefficients for the discriminating
variables are shown on Table 13.
In determining the relative weight of a particular variable, the sign
of the coefficient Is ignored, and the value of the coefficient for any one
variable may be compared to another. For Instance, the student classifi
cation coefficient projects 20.8 times more Influence on the discriminant
function than does the age coefficient. The relative weight in this ex
ample Is determined by dividing the coefficient for the age variable
(.0*1) into the coefficient for the student classification variable (.85*0.
(.85* divided by .0*1 20.8).
Discriminant scores were computed for each one of the participants
in Group I and II. The computation process was as follows: The standard
value for the first discriminating variable was multiplied by its corres
ponding coefficient; this product was added to the product of the second
discriminating variable which likewise was calculated by multiplying the
standard value of the variable by Its corresponding coefficient. This
process took place for each of the nine discriminating variables, re
sulting In the computation of a "discriminant score" for each of the
subjects in Groups I and II (Table 16).


52
and one half percent of the women students who had transferred out of the
College of Business Administration (Group II) and 27.3 percent of women
students enrolled In the College of Education (Group III) were placed In
the feminine category as a result of their scores on the Bern Sex-Role
Inventory. The percent of male business students (Group IV) who were
placed In the feminine category as a result of their scores on the Bern
Sex-Role Inventory was 1.7.
Forty-three and three tenths percent of the subjects In Group I (women
students enrolled In the College of Business Administration), 40.0 percent
of the subjects In Group II (women students who had transferred out of the
College of Business Administration), and ^5.5 percent of the subjects In
Group III (women students enrolled In the College of Education) were placed
In the androgynous category as a result of their scores on the Bern Sex-Role
Inventory. The percent of subjects In Group IV (men students enrolled In
the College of Business Administration) who were placed In the androgynous
category as a result of their scores on the Bern Sex-Role Inventory was 19.0.
Groups III and IV were Included In the analysis of the four categories
of masculine, feminine, androgynous, and undifferentiated in order to address
the following questions. (1) Did women students majoring In education, a
traditionally feminine field of study, produce a more feminine score on the
Bern Sex-Role Inventory than did the other groups in the study? (2) Did
men students majoring In business administration, a traditionally masculine
field of study, produce a more masculine score on the Bern Sex-Role Inven
tory than did the other groups In the study? Women majoring In education
(Group III) did not produce a more feminine score than did the other groups.