Will III I III III/ll Ill! Ill/l lfllllfllflflfll z. I n R A R y

1293 1035

Mich {gm Scam
University ‘

 

This is to certify that the
thesis entitled
AN ANALYSIS OF THE COMPONENTS OF
SEX-BIASED OCCUPATIONAL BELIEFS
presented by

Phyllis Mary Mellon

has been accepted towards fulfillment
of the requirements for

_Bh_._D_.__ degree in W

«Deli... A

Major professor

DateiulLL—JJLL—

0-7639

 

AN ANALYSIS OF THE COMPONENTS OF

SEX-BIASED OCCUPATIONAL BELIEFS

BY

Phyllis Mary Mellon

AN ABSTRACT OF A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Psychology

1978

ABSTRACT

AN ANALYSIS OF THE COMPONENTS OF
SEX-BIASED OCCUPATIONAL BELIEFS

BY

Phyllis Mary Mellon

This study investigated the role and trait components of sex-
biased occupational attitudes. Using a repeated-measures analysis of
variance design, it was found that subjects differ in terms of their
agreement to statements about traits necessary for certain jobs, the
possession of these traits by men and women, and the suitability of
men and women for those occupational roles. Secondly, it was found
that agreement regarding the suitability of women for various occupa-
tional roles was a joint function of the socioeconomic status of the
job, the sex dominance of the job, and the sex of the subject.

A test ochGuire's (1960) syllogistic framework for measuring
occupational attitudes indicated that agreement with job, role, and
trait statements was a function of the type of job being Considered.
Subjects were most consistent in their agreement with female dominated
jobs, and female subjects tended to agree consistently more often than
did males.

It was concluded that looking at sex-biased occupational

attitudes in terms of trait and role components, as well as considering

Phyllis Mary Mellon

the impact of job classification on sex-biased attitudes, were viable
means for understanding the phenomenon of sex discrimination in

occupational settings.

Approved:

 

Date:

 

Dissertation Committee:
William Crano, Chairperson
Neal Schmitt

Jeanne Gullahorn

Eugene Jacobson

AC KN OWLE DGME NT 8

I wish to thank Dr. William Crano, my chairperson, for the
assistance that he gave me on this dissertation as well as for the
influence that he has had on my development as a social psychologist.

I am also very grateful for the substantial time and energy
that were given to me throughout my graduate career by Dr. Neal
Schmitt. His contributions have added to the utility of the
research.

Thanks are also due to my other committee members, Drs.
Jeanne Gullahorn and Eugene Jacobson. Their influence on my graduate
career cannot be overstated.

Finally, there are always special people who make the pursuit
of any goal more pleasurable and more meaningful. I am very grateful
to my parents, who always offered love and support; and to Bryan,
my husband, whose love, knowledge, and mutual career interests have

added immeasurably to my life.

ii

TABLE

LIST OF TABLES . . . . . . . . . .
LIST OF FIGURES . . . . . . . . .
Chapter

I. REVIEW OF THE LITERATURE .

The Present Research . .
The Syllogistic Approach

OF CONTENTS

Experimental Design and Hypotheses .

II. METHOD . . . . . . . . . .
Subjects . . . . . . . .
Measures . . . . . . . .
Occupations . . . . . . .
Traits or Characteristics
Other Measures . . . . .
Alternate Forms . . . . .
Administration . . . . .

III. RESULTS . . . . . . . . . .
Analysis of Forms . . . .
Hypotheses I and Ia . . .
Hypotheses II and IIa . .
Hypothesis III . . . . .
Summary of Hypothesis

IV. DISCUSSION . . . . . . . .

APPENDICES

Appendix
A. List of Traits
B. Questionnaire . . . . . .

BIBLIOGMPHY O O O O O O O O O 0

iii

Testing

and Occupations

Used in

Questionnaire

Page

iv

16
18
19

21

21
21
22
22
23
23
24

25
25
26
28
34
39

41

45

46

60

LIST OF TABLES

Comparisons Between Forms . . . . . . . . . . . .

Analysis of Variance for Sex of Subject by Sentence

Type . . . . . . . . . . . . . . . . . . . . .

Cell Means for Two by Three Analysis of Variance

Analysis of Variance for Sex by Job Sex Dominance by

JOb status 0 O O O O I O O O O O O O O O O O 0
Cell Means for 2 x 3 x 2 Analysis of Variance . .
Chi-Square Analysis of Job by Agreement Pattern .

Separate Chi-Square Analyses for Females and Males

iv

Page

25

26

27

29
29
36

38

LIST OF FIGURES

Figure Page

1. Plot of Cell Means Representing Three-Way
Interaction . . . . . . . . . . . . . . . . . . . . . . 3O

CHAPTER I

REVIEW OF THE LITERATURE

The effects of sex stereotyping and sex discrimination have
been documented in a large number of research studies (see Terborg,
1977, for a review). While Title VII of the 1964 Civil Rights Act
forbids discrimination on the basis of sex, race, or religion, employ-
ment practices which preclude the integration of women into all levels
of the work force still persist. Instead, "ghettos" of low-status,
stereotypic female occupations (waitresses, nurse’s aides, clerical
aides, etc.) have become the repository for a large percentage of the
female labor force (Gullahorn, 1977), and the high-status, executive
positions remain a bastion of the male sex.

Although the sociopolitical and economic implications of sex
discrimination are beyond the scope of this research, the psychological
basis of sexually biased belief systems, as investigated within the
research setting, are relevant in providing explanatory power for the
phenomenon of sex discrimination.

Since this research differentiates sex typed characteristics
from sex role stereotypes, I shall first explain the cognitive
distinctions. While sex role stereotypes encompass the belief that
it is appropriate for women to occupy only certain, limited professions

or to participate in designated "female" activities, the concept of

sex typed characteristics implies that possession of various abilities
or traits is contingent upon the sex of the person. The issue in the
latter case, therefore, is not one of appropriateness, but one of
ability. Thus, the belief that medicine is not a female's profession
(sex role stereotype) differs from the belief that women are too emo—
tional (sex typed characteristic) to be good doctors.

The studies reviewed in the following section fail to make
those distinctions and often liberally exchange one term for the other.
In an effort to present adequately the content of the various studies,
the literature review employs the terminology that was specific to
each study, regardless of its agreement or disagreement with this
author's nomenclature. Thus, if the authors of a particular study
purport to be measuring stereotypes, it will be reported as such.

The present research adopts a cognitive orientation in order
to assess the underlying components of the beliefs which may comprise
the more general sexually biased attitude. "Sex-bias" is used to
label any relationship in which the distribution of the sexes is
non-equivalent. In addition, it follows the recommendation of
Terborg (1977) and attempts to demonstrate the cognitive differences
between (a) sex-role stereotypes and (b) sex-typed characteristics
as well as their respective contributions to sex bias.

The available literature demonstrates the pervasive influence
of the sex of personnel in the occupational setting. Many findings
illustrate the extent to which sex influences personnel decisions and
prevents full integration of women into the professional job arena.
More germane to the current investigation, the research indicates

that both sex role stereotypes and sex typed characteristics can account

for a variety of discriminatory behaviors and attitudes; that often
these two distinct concepts are confused, confounded, or ignored; and
that separate manipulations of both concepts may furnish additional
information on the phenomenon of sex discrimination.

The presence of sex bias, which often serves to differentiate
assumed ability levels or the appropriateness of various activities
on the basis of sex, can lead to certain expectations as to what
behaviors, emotional expressions, or goals are appropriate for either
sex. For example, Miner (1965) contends that women are expected to be
concerned for others, spiritual, and artistic, whereas men are seen as
being power oriented, initiating, and prestige seeking. Broverman,
Vogel, Broverman, Clarkson, and Rosenkrantz (1972), in their review of
the stereotype literature, report that women are perceived as less
objective, logical, and competent, and as more dependent than men
(Rosenkrantz, Vogel, Bee, Broverman, and Broverman, 1966). In addition,
stereotypically masculine traits generally are perceived to be more
desirable for adults than stereotypically feminine traits. Thus,
women may be in a no-win situation in which they are viewed as
different from men, and lacking in "typically" masculine traits that
are considered appropriate for the adult person.

Bem (1974) has departed from the dichotomous orientation
toward masculine and feminine characteristics and has developed a
measure which allows individuals simultaneously to favor a variety of
masculine agd_feminine traits. Her test (Bem Sex-Role Inventory) of
androgyny measures "the relative amounts of masculinity and femininity

that the person includes in her or his self description, and, as such,

it best characterizes the nature of the person's total sex role"

(p. 158). With her inventory, she demonstrated that masculinity and
femininity are independent dimensions and that the androgyny score
(femininity scale score - masculinity scale score) is not correlated
with social desirability. Bem concluded with a recommendation to
depart from traditional conceptualizations of "masculinity" and "femi-
ninity" as orthogonal and to investigate the implications of greater
role flexibility. However, Bem's measure of androgyny, considered by
her as a £912 inventory, consists of a list of traits, and as such

is more a measure of characteristics, rather than roles.

The behavioral correlates of sex role stereotypes were investi-
gated by Clarkson, Vogel, Broverman, Broverman, and Rosenkrantz
(1970), who found that women with higher self-concepts (in terms of
competence) had fewer children. In addition, a female who perceived
herself as more competent tended to combine employment with child-
rearing activities. They inferred from this that women who succumbed
to the stereotypic notions regarding female competence were less apt
to depart from traditional behaviors than their less "constrained"
female counterparts. Thus, the internalization of such stereotyped
attitudes had powerful influences in terms of directing women's
behaviors and providing the children of these women with role models
who demonstrate the patterns of traditional female lifestyles.

The issue of dimensionality of the female stereotype was
addressed by Clifton, McGrath, and Wick (1976). Using an adjective
check list, they found that the notion of a general stereotype for

women should be changed to a multifaceted one. Their study yielded

two distinct stereotypes, that of housewife and "bunny" (the authors'
term for sex object), and a third, more diffuse stereotype which was

a composite of the roles of clubwoman, career woman, and woman athlete.
These latter three roles failed to emerge as a clear stereotype (as
measured in terms of the adjectives that subjects selected to describe
each of the roles), but they did share traits which were seen as stereo-
typically masculine, e.g., aggressiveness, assertiveness, independence,
and competitiveness. Thus, the evidence points to a multidimensional
orientation toward the female stereotype and necessitates the inclusion
of modifiers to express the different kinds of female stereotypes that
may be held by individuals.

Schein (1973, 1975) investigated the relationship between sex
role stereotypes and requisite management traits. By asking male and
female managers which descriptions were characteristic of either women,
men, or managers in general, Schein assessed the degree of overlap
among the three categories as well as the traits on which they differed.
Men and women alike attributed more similarity between men and managers
than between females and managers. There appeared to be a consensus
among all managers, irrespective of their sex, that the traits
necessary for management duties are typically masculine. This demon-
strated acceptance on the part of both men and women of male characteri-
stics as the "necessary" requirements for a managerial position indi-
cates that a woman can be at a disadvantage in the managerial role
unless she assumes the stereotyped male model of managerial behavior
in order to be successful in the organizational setting. Whether or

not the perception of "maleness" of the managerial role is a function

of the past male dominance of that position and the resulting paucity
of female role models cannot be determined from these results; however,
future research into areas with changing sex distributions will more
accurately address the issue of perceptual bias.

Rosen and Jerdee (1973) found that sex role stereotypes
affected the evaluations of male and female supervisors. Following
from the research of Gilmer (1961), who found that male managers
believed that female supervisors would be inferior to male supervisors,
they investigated subjects' (both undergraduates and "real-world"
bank supervisors) evaluations of different styles of leadership
behavior for hypothetical male and female supervisors. The super-
visory styles included (1) threat, (2) reward, (3) friendly-dependent,
and (4) helping. Four separate analyses of variance (AOV) were done to
test the effects of the sex of the hypothetical supervisor, sex of the
hypothetical subordinate, and status of the subject (student or bank
supervisor) on "good-bad" semantic differential ratings of the four
different supervisory styles.

The results revealed no main effect for sex of the hypothetical
supervisor; that is, male and female supervisors were not evaluated
differently. However, when comparing across the four different super-
visory styles, there was a significant interaction for the sex of the
supervisor and the supervisory styles: the reward style was seen as
more appropriate for male supervisors than for female. No significant
differences were evidenced for the other three types of style. With
respect to the "friendly-dependent" style, however, one unpredicted
finding disclosed a significant difference between ratings of

different sexed supervisor-subordinate pairs and same sexed pairs.

Subjects perceived the friendly-dependent approach as more appropriate
for supervisors in different sexed pairs than in same sexed dyads.
There was no main effect for status of subject, indicating that students
and supervisors shared similar perceptions of supervisory styles.

The implications of the findings are that some stereotyping
still exists, in that the "reward" style (recommendation of the subor-
dinate for a raise) was rated more appropriate for male supervisors
than for females, reflecting expectations possibly based on men's more
frequently being in positions legitimizing reward or coercive power.
The fact that the "friendly-dependent" approach was viewed as more
appropriate for different sexed supervisor-subordinate pairs may illu-
strate subjects' perceptions of cross-sex games where the establishment
of a friendly relationship is followed by solicitation of some form of
assistance from each other.

Another study by Rosen and Jerdee (1974a) used an in-basket
technique to evaluate personnel decisions of male bank supervisors
regarding hypothetical female employees. The in-basket task dealt
with four different personnel decisions: (1) promotion of the employee
to bank manager: (2) selection of a male or female employee to attend
a personnel conference; (3) resolution of a personnel problem in which
a transfer or termination was requested for the problem person by the
person's supervisor; and (4) approval of a leave-of—absence request
for child care duties.

The results for each of the four decisions are as follows:
There was a significant difference between the number of promotions

offered to hypothetical male and female personnel. Thirty—two out of

forty-four male candidates (hypothetical) were promoted by the sub-
jects; whereas only nineteen of the forty-nine females were offered
promotions. With respect to career development opportunities, women
were again discriminated against; when the choice for conference
attendance was between a young, highly promotable male and an older,
unpromotable (i.e., working at a dead-end job) female,subjects
selected the male 76 percent of the time. When the sexes of the young
and older employees were reversed, the promotable female was selected
over the unpromotable male only 56 percent of the time.

For resolutions of personnel problems, subjects had greater
confidence in the male supervisor's recommendation to terminate the
problem person; subjects were more apt to prefer a transfer, rather
than termination, as the mode of resolving the problem for the female
supervisor. However, they generally had no trouble accepting the male
supervisor's recommendation to fire the employee. This sentiment,
that it is unacceptable and demeaning to be "fired" by a female, has
been expressed elsewhere (Komarovsky, 1976).

Their last finding indicated that in one situation male per-
sonnel experienced the consequences of sex discrimination. In the
request for a child-rearing leave of absence, subjects rated the
female's request as significantly more appropriate than the male's.
This finding is indicative of the cultural expectation that men are
to be committed to their careers, while women, perhaps being perceived
as secondary members of the work force, are assumed to have career
discontinuities for childcare duties. This then becomes a Catch-22

situation whereby women are expected to require leaves for child

rearing and are therefore assigned to the more unpromotable, less
important jobs because of that assumption.

Rosen and Jerdee (1973, 1974a, 1974b) have contributed to the
specifications of the different realms in which sex stereotypes and
discriminatory behaviors operate, given a hypothetical framework.
However, aspects of their analyses (e.g., a persistent use of several
univariate AOVs within a single sample, along with the failure to
report the subsequent intercorrelations among their dependent measures
or to attempt multivariate analyses) render some conclusions less
powerful. Moreover, their continued reliance on the existence of sex
role stereotypes as the explanation for differential behaviors of
subjects toward male and female stimulus peOple, in the absence of
any attempt to measure such stereotypes, again reduces the impact of
their findings.

Terborg and Ilgen (1975) also criticized the research of
Rosen and Jerdee (1974a) for their use of post hoc explanations
(sex stereotypes) for differential behavior toward males and females.
To use discriminatory behavior as the operationalization of sex
stereotypes, and then to infer that a stereotypic attitude is present
on the basis of that same behavior is tautological.

An attempt at a more theoretical approach to occupational
sex discrimination can be found in the research of Terborg and Ilgen
(1975). They examined discriminatory behavior, using in-basket
simulation tasks, within three frameworks: attribution theory,
equity theory and stereotypes. Undergraduate subjects simulated

administrative decisions involving four hypothetical engineers,

10

either (1) three males and one female, or (2) four males. In the
case involving the sex of the fourth stimulus person, only the first
name of the male or female engineer was changed; all qualifications,
characteristics, etc., remained identical. The dependent variables
were two measures of access discrimination (discrimination at the entry
level of the organization) which were hiring decision and starting
salary; and five measures of treatment discrimination (discrimination
against an employee once he or she has been hired and has been with
the organization for some time period) which were delegation of work
assignments, decisions regarding employee development opportunities,
personnel evaluations, promotions, and allocation of bonus money. A
measure of subjects' stereotypes of women in business was taken
following the in-basket tasks using a 21-item Likert scale, the Women
as Managers Scale (WAMS) (Peters, Terborg, and Taynor, 1974).

This measure (WAMS) consisted of statements about women's
characteristics and their abilities to function in management posi-
tions. Although it appears to be a rather transparent measure of
attitudes toward women, the authors' research has shown that it is
related to a more general sex-biased attitude. The results supported
the notion that females were discriminated against in terms of both
access and treatment variables. Although there was no difference
in the mean hiring rating for the male or female engineering applicant,
the female received a starting salary significantly lower than the
hypothetical male applicant. The female stimulus person was also
assigned routine tasks more often than challenging tasks, earned

significantly less after two years of working, had her performance

. .11,
"x

11

attributed to luck more often than to skill, and was viewed as
operating under greater internal and external constraints when compared
to her male counterpart.

The measure of stereotypes (WAMS) was significantly correlated
with the decision to hire the female applicant as well as to the
overall performance evaluation of the male. However, it was not signi-
ficantly related to the other dependent measures. Although the lau-
tluxrs failed to question the possible reactive effect of the in-
basket tasks on the questionnaire responses, they still concluded that
processes in addition to stereotypes affect sex discriminatory
behavior and that more research on the psychological aspects which
underlie these differential evaluations of males and females is needed
to explore and understand the phenomenon more adequately.

Bass, Krusell, and Alexander (1971) administered a 56-item
questionnaire to male managers in order to assess their attitudes
toward working women. In general, the managerial subjects felt that
women prefer to work for a man; women should defer to men's abilities;
and men should defer to the needs of the "weaker sex." In addition,
those men who were married regarded women as having no career
potential. The higher-level male subjects viewed women as less
dependable than did lower-level managers; and men who worked with
women generally expressed less favorable attitudes than those who
worked exclusively in male groups.

Although their questionnaire was of questionable psychometric
adequacy, it did furnish some descriptive information on male

managers' attitudes toward managerial women. It also indicated the

12

need to orient men as well as women to the mixed-sex work group
rather than merely to place women in jobs amidst nonsupportive, or
possibly resentful, males. If indeed negative attitudes toward
working women persist to this extent today, then the integration of
women into the work force will be impeded by such attitudes.

The belief that a "woman's place is in the home" has been
modified to express the current prevalent view that a woman's place is
either in the home or performing a few low-status, sex-typed jobs. An
example of this is found in Oppenheimer (1968) who surveyed occupations
from 1900 to 1960 in which 70 percent or more of the personnel were
women. These jobs were characterized as having traditional female
"roots" (e.g., sewing, serving food, caring for the sick, etc.),
depending upon cheap and untrained labor, lacking any career continuity,
and having very little power or authority. In addition, half of the
total female labor force was concentrated in these sex-typed occupa-
tions.

Hesselbart (1977) investigated the ratings of hypothetical
female or male nursing students by male and female high school students.
She found that the female stimulus person was rated as more feminine,
ambitious, attractive, interesting, and lg§§_emotional than the male.
The sample of students also preferred the medical student over the
nursing student and least preferred the male nursing student; he was
rated the least attractive, least likely to be good at his job, least
likely to obtain the job of his choice, and the person whom they least
wanted to meet. The general conclusion was that a man entering a

predominantly female occupation was viewed even more negatively than

13

a woman entering a male occupation, at least according to high school
students. Whether this is due to the crossing of sexual barriers or
status differences in the two jobs, or an interaction of the two cannot
be determined in this study, since the author confounded job status with
sex stereotype of the job.

The expectation that an increased number of women will be
engaged in a given male-dominated occupation lowered both the prestige
and the desirability of that occupation (Touhey, 1974). Cohen and
Bunker (1975) found that when two hypothetical positions were vacant,
subjects were more likely to recommend the female applicant for the
"editorial assistant" job and the male applicant for the job of
"personnel technician." Pre-tests had indicated that these jobs were
perceived as being equal in status and sex-typed in the same manner as
subjects perceived them. Thus, given equal statuses, subjects pre-
ferred hiring individuals on the basis of congruent sex and sex-typed-
ness of the job combinations rather than on other characteristics.

Variables which were used for the evaluation of job applicants
had a different impact on male and female candidates (Cecil, Paul, and
Olins, 1973). Although males and females were described as applying
for the same "white-collar job," subjects imposed their own view of
the male candidate as an administrative management applicant, and the
female candidate as clerical applicant. The subjects' views were post
hoc determinations by the authors based on the traits which subjects
perceived as being important for either the male or female applicant
to have in order to be hired for the "white-collar" position. Those
traits which were seen as more important for males were the abilities

to be persuasive, aggressive, motivated, withstand pressure, and to

14

change his mind. For the female applicants, the important traits were
a pleasant voice, excellent clerical skills, a high school diploma,
good self-expressions, and being "immaculate in dress and person."

Gordon and Hall (1974), in a study using women graduate
students, found that females' perceptions of men's stereotype of femi-
ninity could be sources of conflict for the females. They further
suggested that the perceived male standard of femininity may produce
difficulties for women who work in male-dominated work groups. Thus,
perceptions of other's stereotypes, irrespective of their validity,
can be a powerful conflict source for women and may deter their pro-
fessional development.

Hamner, Kim, Baird, and Bigoness (1974) found that ratings of
overall task performance were influenced by both the race and the sex
of the stimulus person being rated. Using a videotaped work-sample
task, undergraduate subjects rated all combinations of high-low
performing males and females and blacks and whites, resulting in
eight different ratings from each subject. In general, females were
rated higher than males, and this difference was accentuated for
high-performing ratees; black raters rated black ratees higher than
white ratees, while white raters rated black ratees lower than white
ratees. Although previous findings have shown a tendency to devalue
female performance (e.g., Rosen and Jerdee, 1973), females in this
study were rated higher than equally performing males. This reversal
was attributed to the fact that the job (grocery store stockperson)
was in the "unskilled labor" category, while jobs in other studies

have usually been at the professional level. Thus, the authors are

15

suggesting a sex by job status interaction in which females "win"
at the low end and lose at the high end of the work-status continuum.
This sex by status interaction needs further investigation to address
more definitively the effect of different job status levels on evalu-
ations of males and females.

Sex and attractiveness of hypothetical job applicants were
found to influence hiring recommendations for the job of head of a
furniture department in a large department store that were made by
male undergraduates as well as those of professional male inter-
viewers (Dipboye, Fromkin, and Wiback, 1975). In general, attractive
applicants were preferred over unattractive ones (a pre-tested
photograph was included in each applicant's folder), males were per-
ceived as more suitable than females, and those applicants with high
scholastic standing were preferred over those with low standing.
These findings indicate that both undergraduates and professional
interviewers discriminated on the basis of two non job-related
applicant characteristics (attractiveness and sex), in the case of a

white-collar occupation. Moreover, to have demonstrated that both

 

undergraduates and professional interviewers are subject to the same
perceptual biases, not only supports the generalization of results
obtained with student samples to the general population, but also
causes one to be suspect of the degree of training that is theoreti-
cally part of the development of a professional interviewer.

Sex has also been found to influence leadership behavior
(Megargee, 1969), managers' evaluations of employees' grievances
(Rosen and Jerdee, 1975), and supervisory behavior and subordinate

satisfaction (Petty and Lee, 1975). Attributions made to luck or

16

skill also tended to be dependent upon sex; Deaux and Emswiller (1974)
found that performance on a masculine task is attributed to skill for
males and to luck for females. Taynor and Deaux (1973) demonstrated
that sex can influence reward allocation because it is seen as a
"nonvoluntary constraint" when performing a masculine task. If a
female is not expected to perform as well as a male, but she does
perform as well, she will then be over-rewarded with respect to the
male.

Thus, the influence of sex on evaluations and reward allocations
has been demonstrated to be pervasive, and the domains in which that
influence operates are varied. Moreover, other variables (e.g., job
status, race, attractiveness, etc.) have been shown to interact with
sex, and no clear-cut picture of the roles of various person and/or
job characteristics has emerged. Thus, the literature is replete with
studies of masculine or feminine traits or studies which manipulated
the sex-typedness of the occupation, but apart from demonstrating
that bias exists and offering a few conjectures as to why, the area
lacks cohesiveness. If the emphasis were to be shifted from demon-
strations of behaviors to the exploration of the underlying variables
which direct the behaviors, then a more parsimonious and cognitive set

of research findings may emerge.

The Present Research

The studies reviewed dealt with experimental simulations of
personnel decisions, attitude questionnaires administered in the
field or in the lab, and numerous adjective check-lists which were all

used to demonstrate bias attributable to sex. Because of this bias,

17

researchers usually invoked the existence of sex role stereotypes as
the "villainous" cause of such differential response patterns. How-
ever, given either the applied nature of some of the research, or the
short-sightedness of other studies, most researchers have failed to put
stereotyping within a framework that would serve as a precursor to the
development of a more theoretical orientation toward the problem of
sex bias. Apart from Terborg and Ilgen (1975) or Deaux and her
colleagues (Deaux and Emswiller, 1974; Taynor and Deaux, 1973), most
studies merely took a descriptive approach to the phenomenon which has
resulted in a plethora of attitudinal/behavioral data lacking con—
ceptual organization.

Terborg (1977), in his review of the research on sex discrimi-
nation, noted the need for a new approach in the area. He pointed out
that past studies have confused, overlooked, or ignored the distinction
between sex role stereotypes and sex typed characteristics. The former
deals with the notion that it is appropriate for females to engage
in only certain activities, irrespective of abilities or desires;
whereas sex typed characteristics refer to the belief that women
possess certain traits, men possess certain other traits, and women
just are not endowed with many traits that are requisites for certain
jobs. Thus, the justification for why women should not be engineers
could be that (1) engineering is "man's work and no place for a lady"
if one were to invoke a role stereotype; or that (2) women lack the
mathematical abilities that are necessary for good engineers, if
one relied upon sex typed characteristics as the explanation or

justification.

18

Although neither of the two beliefs has demonstrated validity,
their cognitive distinctions, as well as their sensitivities to dif-
ferent manipulations need to be investigated. The already noted ten-
dency in so many different studies to interchange these terms, con-
cepts, and measurements dealing with sex role stereotypes and sex typed
characteristics may be a disservice to both theoretical and applied

concerns .

Syllogistic Approach

 

McGuire (1960) formulated a logical model for evaluating the
consistency among a person's cognitions. He contended that a person's
"belief space" could be determined by a set of statements which
represent logically interrelated cognitions about a given belief.

The syllogistic structure typically consists of two premises,
a_and b, and a statement 9, which is a logical conclusion following
from a_and b. In the event that a person agrees with E, but not 3_and
b, McGuire designates another proposition, 5, which would explain the
apparent inconsistency between agreement with g_and disagreement with
a_and b, An example of a syllogism, in the context of this research,

can be given by the following statements:

(a) Good managers should be responsible.
(b) In general, men are more responsible than women.

(3) In general, men are better managers than women.
Thus, within a logical framework, agreement with a_and b_wou1d imply
agreement with 3 unless there exists an unspecified statement k_whereby
a person could disagree with a_and b_and agree with g_for a reason (5)

having nothing to do with the trait of responsibility. Agreement with

19

a_and b, but not 9, would imply that a person believes that characteri-
stics are sex typed, but does not apply that to a role stereotype.
Thus, the syllogistic approach allows us to measure sex role

stereotypes (as shown in statement 3), sex typed traits (statement 2),

and job characteristics (statement 3).

Experimental Design and Hypotheses

In order to assess the "components" of sexually biased occupa-
tional beliefs, several different frameworks were used.

First, to determine whether or not males and females respond
differently to statements about jobs, roles, and traits (types a, g,
and b), a 2 x 3 between-subjects design was used in which sex (male,
female) was the between-factor, and sentence type (a, b, g) was the
within factor. This design tested the following hypotheses:

Hypothesis I: There are cognitive distinctions among the
components of sexually biased beliefs.

Hypothesis Ia: There are sex differences in the manner in
which males and females respond to these belief statements.

Secondly, the effects of job status (SES) and sex dominance
were tested in a 2 x 3 x 2 AOV design in which sex of subjects was
the between factor, and sex dominance of job (male, female, or
sex-neutral) and job status (low, moderate) were the within-factors.
This design tested the notion that subjects'agreement with role
statements (type 3) was contingent upon the type of occupation men-
tioned in the statement. Thus, perceptions of role appropriateness
may be determined by the job status and/or sex dominance of the job.

Thus, the following hypotheses were tested:

20
Hypothesis II: Occupational status and sex dominance influence
subjects' agreement to role statements.

Hypothesis IIa: There are sex differences in the manner in
which subjects are influenced by these role statements.

In order to test the degree of consistency of within-syllogism
agreement, a chi-square test of four patterns of agreement (agreement
with a, b, and g; agreement with a_and b_but not 3; agreement with 9
only; and all other agreement patterns) was done across the six levels
of combinations of status and sex dominance of the occupations. This
assessed the extent to which agreement varied with the different occu-
pations, and tested the following hypothesis:

Hypothesis III: The patterns of agreement across occupations
vary with the (1) status of the job and (2) the sex dominance.

CHAPTER II

METHOD

Subjects

One hundred and thirty-two female and sixty-six male subjects
were recruited from introductory psychology classes and received class

credit for their participation.

Measures

Measures of sex role stereotypes, sex typed traits, and per-
ceived job characteristics consisted of five-point Likert statements
to which subjects indicated the extent of their agreement or disagree-
ment. The five response categories were (1) Strongly Agree, (2) Agree,
(3) Neither Agree nor Disagree, (4) Disagree, and (5) Strongly Dis-

agree.

Job Requisites.--The extent to which subjects felt that certain
traits are necessary for certain jobs was measured by statements using
the jobs and traits that were employed in the other two categories of
statements. Statements in this category represent those of type a_

presented previously (e.g., Good doctors should be logical).

Sex Typed Traits.--The extent to which subjects believed that

certain traits are possessed to a greater degree by either males or

21

22

females *was measured by statements (type 2) like the following:

In general, women are more ambitious than men.

In general, men are more logical than women.

Half of the statements had "women" as subjects; half had "men."

Sex Role Stereotypes.--A number of statements dealing with the

 

relative appropriateness of either men or women for a number of
different jobs served as measures of the degree of role stereotypedness
(type S):

In general, women are better doctors than men.

In general, men are better nurses than women.
Again, half of the statements had "women" as subjects and half had

"men."

Occupations

 

Occupations used in the attitude statements were combinations
of pretested moderate or low status, and male, female, or sex-neutral
professions. Occupations classified as "Low" or "Moderate" socio-
economic status were obtained from the Duncan Socio-Economic Status
S2213 (Reiss, Duncan, Hatt, and North, 1971). The sex dominance
of the occupations (male, female, or neutral) was obtained from the
United States Bureau of the Census (1960). Those occupations having
fewer than 25 percent women were classified as "male"; those with
75 percent or more women were classified as "female"; sex-neutral

occupations had between 40 percent and 60 percent female composition.

Traits or Characteristics

 

Traits used in the attitude statements consisted of nine

traits-~three masculine, three feminine, and three sex-neutral. Traits

 

23

were obtained from previous research studies (Anderson, 1968;
Broverman, et al., 1972; Cecil et al., 1973; and Clifton et al.,
1976). (See Appendix A for the complete list of occupations and

traits used in this study.)

Other Measures

 

After the randomized pages of role, trait, and job statements,
the Women As Managers Scale (WAMS) (Peters et al., 1974), the Bem
(1974) Sex Role Inventory (BSRI), and measures of age, sex, and
other subject characteristics were included in the questionnaire.

The WAMS, BSRI, and other demographic information were collected for
use at a later time and not part of this research. Thus, they were
included at the gnd_of the questionnaire so as not to influence

responses to job, trait, and role statements.

Alternate Forms

 

Two different forms were written for the attitude statements
measuring role stereotypes (type 3) and sex typed characteristics
(type b). Form One consisted of a random half of both type b_and g
statements beginning with "In general, men are . . . ," while the
other half began with "women." Form Two statements were the opposite
of Form One's--"women" were substituted for "men," and "men" for
"women." This was done to insure no bias due to the sex of the
subject of each attitude statement as well as to determine if
responses were dependent upon the sentence subject's sex.

One-half of the subjects received each form, with an equal

number of male and female subjects in each half.

24

Administration

 

Questionnaire administration was conducted in group settings.
The questionnaire booklet (see Appendix B) consisted of randomized
pages of the three different sentence types, WAMS, BSRI, and

demographic questions.

CHAPTER III

RESULTS

Analysis of Forms

 

Both forms were recoded such that "1" indicated agreement with
statements having "women" as subjects, and "5" indicated agreement with
statements having "men" as subjects of the sentence. Thus, the:low
scores indicated a positive attitude toward women in terms of being more
suitable for the various roles or possessing more of a given trait than
men. Analysis of the two forms indicated that they were comparable
for both the role and trait statements. Tests of these differences
across forms for both role and trait statements were done using Dunn's
(1961) test for multiple comparisons, thereby insuring against alpha
inflation due to nonindependent Ertests. Table 1 presents the results
of this analysis. Both Eftests for role and trait forms were nonsigni-
ficant, 5(196) = -.099 and .126, respectively. Thus, both forms were
combined and treated as one for the remaining analyses.

Table l.--Comparisons Between Forms.

 

 

 

 

Form 1 Form 2
Role Trait Role Trait
Means 25.57 46.96 26.17 44.72
Standard Deviations 2.17 3.25 2.45 3.40
Sample Sizes 99 99

 

25

26

Hypotheses I and Ia

 

A 2 x 3 analysis of variance (AOV) was done to test the effects
of sex of subject and sentence type (a, b, and g) on overall responses.

Table 2 presents the summary of this repeated measures analysis.

Table 2.--Ana1ysis of Variance for Sex of Subject by Sentence Type.

 

 

 

Source MS df F
Sex of Subject (A) .051475 1 .736
Error .069963 196
Sentence Type (B) .045194 2 55.721*
A x B .620993 2 7.438*
Error .083486 392
Total 593

* p < .001

The sex by sentence type interaction (A x B) was significant,
indicating that the manner in which males and females respond to job,
trait, and role statements varies with sex and levels of sentence type.
Tests of the simple main effects (Winer, 1971) were done for both sex
and sentence type. Results for the test of the simple main effect for
sentence type at both levels of sex revealed that sentence type was
highly significant for males, §f2,130) = 15.938, p_< .001 and for
females, 2(2,262) = 47.688, p_< .001. The test of the simple main
effect for sex at the three levels of sentence type yielded differing
results. For the trait statements (type b), there was no significant
effect due to sex, §(1,196) = 1.003. However, tests for sex at the

other two levels were significant: for job statements (type a),

27

§(1,196) = 4.431, p_<:.037, and for role (type c) statements, §(l,196)
= 12.682, p < .001.

Thus, this analysis indicated that males and females exhibit
different patterns of agreement in terms of the syllogistic framework.
They differ on agreement with the major premise (type §_or job state-
ments), fail to differ on the minor premise (type b_or trait state-
ments), and again differ on their agreement with the conclusion (type
g_or role statements). From the pattern of female-male means presented
in Table 3 for the three sentence types, it appeared that males agree
slightly more than females with statements about characteristics that
are necessary for the jobs; women agree more often than do men with
statements about roles; and, of course, the sexes exhibited no dif-
ferences in their agreement with trait statements. The data that were
analyzed consisted of scale means for all three sentence types, since
each was comprised of different numbers of items. Thus, the cell means
in Table 3, as scale means, indicated that on a five-point Likert
scale, the mean responses were closer to neutral, rather than agree

or disagree.

Table 3.--Ce11 Means for Two by Three Analysis of Variance.

 

 

Sentence Type Males (N=66) Females (N=132)
Job (3) 3.12 3.21
Trait (E9 2.88 2.87

Role (2) 3.15 3.01

 

28

Hypotheses I and Ia were supported from these findings. The
significant sex by sentence type interaction indicated that both com-
ponents influence agreement. Furthermore, tests of the simple main
effects demonstrated that sentence type was significant at both levels
of sex, and the sexes differed on all but the trait statements. The
significant simple main effect test for sentence type indicated that
there are differences in the manner in which subjects respond to the
three sentence types (Hypothesis I), and the test of the simple main
effect for sex, significant for two of the three sentence levels,

supported Hypothesis Ia.

Hypotheses II and IIa

 

The test of the effects of sex of subject, job status, and the
sex dominance of the job on overall agreement to twelve role (type c)
statements was done using a 2 x 2 x 3 repeated-measures analysis of
variance design, where sex was a between-subject factor, and the six
combinations of job status (low, moderate) and sex dominance of the
job (female, male, or sex-neutral) were the repeated measures. Since
there were only two occupations at each level, cell scores were based
on the ggm_of the responses to the two role statements at each of the
six levels. The occupations used in each level are listed in Appendix A.

Table 4 presents the summary table for this analysis. The
three-way interaction was significant, Fj2,392) = 6.11, p_< .002,
indicating that agreement to role statements varied as a joint function
of sex, job status, and job sex dominance. A plot of this interaction
is displayed in Figure l. The cell means for all levels of the three

factors are presented in Table 5.

29

Table 4.--Analysis of Variance for Sex by Job Sex Dominance by Job

 

 

 

Status.

Source MS df F
Sex of Subject (A) 13.7883 1 12.289 *
Error 1.1220 196
Job Status (B) 1.4150 1 1.584
A x B 3.0408 1 3.405
Error .8931 196
Sex Dominance (C) 164.2790 2 124.968 *
A x C 2.5661 2 1.952
Error 1.3146 392
B x C 21.6069 2 25.113 *
A x B x C 5.2580 2 6.111 *
Error .8604 392
Total 1187

* p < .002

Table 5.--Cell Means for 2 x 3 x 2 Analysis of Variance.

 

Males Females

 

 

Sex
Dominance Low Status Moderate Status Low Status Moderate Status

 

Male 7.14 6.56 6.69 6.45
Female 5.21 5.61 5.48 5.27
Sex-Neutral 6.00 6.82 5.82 6.27

 

Note. Lower numbers imply more agreement with women being more
suitable for the occupation than are men; means are the
sum of responses to two five-point Likert statements.

3O

. ousma
.coHuomuwucH >oznwmuns mcwuzmmwumom memo: HHwO no uon H .m

”‘02‘2 .500 «19.0 Q01”

 

 

J 32 (cc
magma chkauz wacE misfit 5H m”
d d nt0.m
30.40:
Uktduooi L.
.1!
30... to...
30%..
nrdmhmdfl TEEN...» no.4 whmaw w; 1?. l.

 

31

Given the significant three-way interaction, interpretations of
any two-way interactions or main effects are confounded by this signi-
ficance. Therefore, in order to understand more completely the impact
of the three factors on responses, tests of the simple interactions and
the simple, simple main effects (Winer, 1971) were done.

Tests of the job sex dominance by job status simple interaction
were significant for both levels of sex of subject, {(2,262) = 11.81
and Ff2,130) = 19.05, p_< .001, for females and males, respectively.
Women were thought by all to be more suitable for male dominated,
moderate-status jobs than low-status male jobs, more suitable for
low-status, sex-neutral jobs, rather than moderate; males felt that
women were more suitable for low—status female jobs, while female
subjects felt that women were more suitable for the moderate-status
jobs that were female dominated.

The sex of subject by job sex dominance simple interaction
was only significant for low-status occupations, FfZ,392) = 4.98,

p < .007. For moderate-status occupations, the simple two-way inter-
action was nonsignificant; however, the main effect for job sex
dominance was significant, F(2,392) = 75.93, p_< .001, as was the
main effect for sex of subject, {(1,196) = 15.86, p.< .001. In
general, men and women responded differently to the appropriateness
of women for jobs of different sex dominance ggly_at the low—status
level. For moderate-status occupations, this difference was not
apparent; sex and sex dominance independently influenced perceptions
of appropriateness.

The simple interaction of sex by job status at the three

levels of job sex dominance was significant for female dominated

32

jobs, Fjl,196) = 7.30, p_< .007; as well as for sex-neutral jobs,
£11,196) = 4.09, p_< .045. For neutral occupations, male subjects
agreed that women were much more suitable for the low-status occupations
than for those of moderate status; while females exhibited the same
general pattern, they had much less of a discrepancy in their agreement
between moderate- and low-status jobs. For female dominated jobs,
women reversed their agreement pattern, responding that women were

more suitable for moderate-status jobs than for low-status jobs. Male
subjects were the reverse. For male dominated occupations, the sex by
job status interaction was nonsignificant, F(l,l96) = 3.19. The main
effects for status, §(1,196) = 16.15, p_< .001, and for sex,

Fj1,196) = 4.79, p.< .03, were both significant. Subjects felt that
women were more suitable for moderate-status jobs than those of lower
status, and female subjects agreed that women were more suitable for
these occupations than did males.

Tests of the simple, simple main effects for job sex dominance

 

were significant at all levels of job status-sex of subject, at p_<
.001. All jobs were perceived to be of different suitability for
women, depending upon which sex has primarily filled each job.

Tests of job status at the six levels of sex of subject-job
sex dominance were significant for five of the six levels: female
subjects, male dominated jobs, §(1,131) = 4.66, p_< .033; female
subjects, sex-neutral jobs, Fj1,l3l) = 17.93, p_< .001; male subjects,
female dominated jobs, Ffl,65) = 4.12, p_< .048; male subjects,
male dominated jobs, Fjl,65) = 17.32, p.< .001; and male subjects,
sex-neutral jobs, §(1,65) = 29.36, p_{ .001. The nonsignificant

status effect was for female subjects, female dominated jobs,

33

§(1,131) = 2.86. Apparently, females perceived little difference in
suitability of women in traditional female occupations at either
status level. The mean agreement for both status levels (Table 5)
indicated that female subjects agreed that women are only slightly
more suitable for moderate-status jobs (;'= 5.27) than for low-status
jobs (E': 5.48). Given that these two cell means represented the sum.
of agreement to two role statements, dividing by two indicated that
women's responses fell between agree and neutral on the five-point
Likert scale. But at all other levels of sex of subject--job sex
dominance, there were significant differences in the agreement with
the appropriateness of women for low— vs. moderate-status occupations.

The tests of the simple, simple main effect for sex of subject

 

at the six levels of job status--sex dominance yielded only two sig-
nificant results: there were sex differences for moderate-status,
sex-neutral occupations (store manager and musician), Fjl,l96) = 17.88,
p'< .001, as well as for low-status, male dominated occupations
(janitor and taxi-driver), §(1,196) = 6.58, p_< .02. For the moderate-
status, sex-neutral occupations of store manager and musician, female
subjects agreed more often that women were more suitable for these
occupations than did male subjects (cell means of 6.27 vs. 6.82). For
the low-status, male dominated occupations of janitor and taxi-driver,
female subjects again agreed more with the suitability of women for
these occupations (6.69 vs. 7.14). However, given that these means
represent the sum of responses to two five-point Likert scales, divi-
sion by two indicate that all subject fell slightly to the "disagree"

side of neutral. So, although significant, these differences represent

34

differences in "degree of neutrality" more than they do a real
agree-disagree difference.

The four nonsignificant simple, simple main effect tests for
sex of subject were those of moderate status, female dominated jobs
(elementary school teacher and social worker), Fjl,l96) = 3.72;
moderate-status, male dominated jobs (insurance agent and chiropractor),
§(1,196) = .65; low-status, sex-neutral occupations (textile weaver
and shoe-factory worker), §(1,196) = 2.04; and low-status, female
dominated occupations (hairdresser and licensed practical nurse),
Ejl,l96) = 2.35.

The results of this analysis, although not unequivocal, did
demonstrate support for Hypothesis II and Hypothesis IIa. Job status
and sex dominance simple interactions were significant at both levels
of sex of subject, offering support for Hypothesis II. Tests of the
simple, simple main effect for sex of subject revealed only two sig-
nificant findings of the six possible tests. Thus, Hypothesis IIa

received some support, contingent upon the classification of the job.

Hypothesis III

 

In order to determine the extent of agreement to job, trait,
and role statements (types a, b, and g, respectively) across the six
different levels of job status (low and moderate) and job sex dominance
(female, male, or sex-neutral), as well as to examine the utility of
the syllogistic framework (McGuire, 1960) for measuring sex-biased
job attitudes, a six by four chi-square test was performed. The six
levels of occupations were (1) low-status, sex-neutral; (2) low-status,

female dominated; (3) low-status, male dominated; (4) moderate-status,

35

sex-neutral; (5) moderate-status, female dominated; and (6) moderate-
status, male dominated. The four levels of agreement were (1) agreement
with E, but not a_and b; (2) agreement with §_and b, but not 9; (3)
agreement with a, b, and g; and (4) all other agreement patterns. The
first level of agreement represented, in terms of the syllogistic
framework, agreement with the conclusion, but not with the premises.
The second level represented agreement with both the major (a) and
minor (2) premises, but not the conclusion. The third level represented
the consistent agreement pattern in which subjects agreed to the major
and minor premises and the conclusion that was logically consistent
with the premises. The fourth level represented an aggregation of all
other agreement patterns that were inapplicable in terms of the syllo-
gistic framework. The individual probabilities for these four agreement
patterns are, respectively, (1) .144; (2) .096; (3) .064; and (4) .696,
for Levels One, Two, Three, and Four.

The data were combined for both sexes, and agreement and strong
agreement responses were both considered as agreement for this
analysis. Six sets (six sets of a, b, and g_statements) were analyzed
per job level. Each set consisted of one of the Egg appropriate occu-
pations at that job level and one of thggg_traits: independent, sensi-
tive, and loyal. This subset of traits was selected because it
represented one each of masculine, feminine, and sex-neutral traits
and was considered to be representative of the agreement process that
was being measured. These three traits (these same traits were used
at all_job levels) and two occupations resulted in six syllogisms

per job level, e.g.,

36

(a) Good social workers should be independent.
(2) In general, women are more independent than men.

(c) In general, women are better social workers than men.

The frequencies of the responses to these six syllogisms per
job level were averaged to produce one entry per cell. These cell
entries were ngt_independent.

The result of the six by four chi-square test was significant,
x2(15) = 366.78, p_< .001. Table 6 contains the average frequencies
that were used to compute this statistic. Given that the row
entries are nonindependent (i.e., repeated measures), interpretations

of this significance must be viewed with caution.

Table 6.--Chi-Square Analysis of Job by Agreement Pattern.

 

Agreement Agreement with Agreement with Other
with a a and b a, _b_, and _c_

 

 

gigegzzg Values 28.51 19.01 12.67 137.81
Low-Male 12.17 33.67 8.50 143.66
Mod-Male 7.83 57.33 15.67 117.17
Low-Female 23.33 30.83 41.67 102.17
Mod-Female 16.01 38.33 44.33 99.33
Low-Neutral 23.17 20.17 12.00 142.66
Mod-Neutral 11.33 44.67 18.50 123.50

 

Note: Cell entries represent averages for all six sets of syllogisms

within each job category; N

= 198.

37

In order to understand where the agreement pattern varied, six
one-by-four chi-square tests at the four levels of job status were
done. Results for five of the six jobs were significant at p_< .001.
The only nonsignificant chi-square was found at the low-status, sex-
neutral job level, X2(3) = 1.26. Thus, for the jobs of shoe-factory
worker and textile weaver, agreement patterns were close to chance
levels. For the remaining five job categories, agreement patterns
fluctuated. For both low- and moderate-status female dominated jobs,
the syllogistic effect was strongest, with the frequency of agreement
to a, b, and g_being greater than agreement with a_only or a and b,
Subjects also tended to agree more with a and b_than what was
expected, and agreed much less than chance with a_only or with the
"other" category. For male dominated occupations at both status
levels, large deviations from expected values occurred with respect
to agreement with 37-which was much less than predicted; and with
respect to agreement with a and 27-which was much more than expected.

Two separate chi—square analyses were done for each sex. These
are presented in Table 7. Both chi-squares were significant: for
females, x2 (15) = 289.67, and for males, x2 (15) = 93.51, both p_<
.001. The syllogistic effect was strongest for agreement with the
suitability of women for female dominated jobs. on the Part Of
b°th female and male SUPJECtB- The syllogistic pattern was weakest
for male subjects responding to both male dominated and sex-neutral
jobs; all values were lower than expected. For females, the only
syllogistic pattern that was less than expected was agreement with

low-status, male dominated jobs (janitor and taxi-driver). Thus, it

Table 7.--Separate Chi-Square Analyses for Females and Males.

38

 

 

 

 

 

Agreement Agreement Agreement Other
with a_ with a_& b_ with a, b, &'3
Females (N = 132)
Expected Values 19.01 12.67 8.45 91.87
JOB TYPE
Low-Male 8.83 23.83 6.83 92.50
Mod-Male 4.67 39.50 12.17 75.67
Low-Female 13.83 22.67 28.83 66.67
Mod-Female 10.83 24.83 33.00 63.33
Low-Neutral 16.00 14.17 10.00 91.83
Mod-Neutral 8.17 31.00 15.17 77.67
Males (N = 66)

Expected Values 9.50 6.34 4.22 45.94
JOB TYPE
Low-Male 3.33 9.83 1.67 51.17
Mod-Male 3.17 17.83 3.50 41.50
Low-Female 9.50 8.17 12.83 35.50
Mod-Female 5.17 13.50 11.33 36.00
Low-Neutral 7.17 6.00 2.00 50.83
Mod-Neutral 3.17 13.67 3.33 45.83

 

39

appears that the greater response consistency (in terms of the syllo-
gisms) is with the female subjects. In general, the analysis indicated
that both men and women agree less than expected with a_only; females
agree more often than expected with both a_and b, and males agree with
§_and b_more than expected for all jobs except low-status, sex-neutral
(shoe-factory worker and textile weaver). While women agree more than
expected with a, b, and g_for all but low-status male jobs, men only
agree more than expected with female jobs. '

Thus, Hypothesis III received support from these findings,
although somewhat questionable due to the nonindependence of the data.
The agreement pattern did vary with the job categories, and males and
females alike tended to agree m2§£_consistently, in terms of the syllo-
gisms, with female dominated jobs at both status levels. Females, in
general, tended to agree with all jobs except those that were low-status,
male dominated; whereas males only agreed with female jobs. Thus, the
consistency of the agreement pattern tended to vary with job type, as
predicted, with the most marked effect being with female dominated
jobs. Furthermore, an unpredicted sex difference was found where
females tended to agree with a, b, and g_much more often than men for

all job types.

Summary of Hypothesis Testing

 

The results of the two by three analysis of variance for sex by
sentence type supported Hypothesis I and Ia, which stated that subjects
respond differently to job, role, and trait statements, and that there

are sex differences in these responses.

4O

Hypothesis II and IIa received support from the 2 x 3 x 2 AOV
in which subjects responded differently to jobs, depending upon their
status and sex dominance, and in which there were some sex differences
at some levels of job type.

The Chi-square analysis of the agreement pattern by job classi-
fication was significant, indicating that agreement varied as a function
of the type of job being considered. The syllogistic effect was most
pronounced for female dominated jobs of both status levels. Moreover,
separate analyses for male and female subjects revealed a strong
tendency on the part of females to respond more often than expected
with the syllogistic response pattern for all jobs except those that
were low-status, male dominated. Males only agreed with a, b, and g for
female dominated jobs. Thus, Hypothesis III, that agreement patterns
varied as a function of job type, received some support, although the
nonindependence of the data made this support questionable. An unpre-

dicted sex difference in agreement patterns was also found.

CHAPTER IV
DISCUSSION

The results of the analysis for Hypotheses I and Ia supported
the utility of distinguishing between sex typed characteristics or
traits and sex £213§_(Terborg, 1977). That subjects perceived dif-
ferences between roles and traits, and that males and females differed
somewhat in terms of their agreement to these attitudinal statements
indicated that the more global sex biased attitude is comprised of
more specific components. The next logical step would be to test the
differential sensitivity of these components to manipulation in a
laboratory setting.

The analysis of the role statements (tests of Hypotheses II and
IIa) demonstrated that subjects had some interesting perceptions about
male and female appropriateness for different occupations. The sex
dominance of an occupation tended to be a powerful determinant of the
degree to which subjects agreed to statements concerning the appro-
priateness of women over men for various occupations. Figure 1 clearly
illustrated that as sex dominance changed from male, through neutral,
to female, agreement increased. Thus, males and females alike were
invoking the "status quo" for their responses. Another interesting
finding was that, for female dominated jobs, females had no significant

difference in their agreement to low- or moderate-level jobs, although

41

42

they did agree slightly more with moderate-level jobs. Males had a
marginally significant difference between the two status—levels of
female—dominated jobs, and furthermore, felt that women were more
appropriate for low-, rather than moderate-level, female dominated jobs.
The tests of the simple, simple main effect for job status demonstrated
the ascribed "low status" to female occupations: women saw no difference
in terms of low and moderate status levels within female dominated
occupations, and males' perceived difference just reached significance
(with the alpha inflation due to multiple, nonindependent tests, it is
questionable whether this marginal effect is significant). Status levels
were much more differentiable at the other levels of job sex dominance.

In general, the most outstanding sex differences occurred for
moderate status, sex-neutral jobs and for low status, male dominated
jobs. Men felt that women were less acceptable for both of these
occupational levels than did female subjects. Tests of the simple,
simple main effect for sex of subject failed to reach significance
at the other job levels. The fact that women are more agreeable to
women being in various occupations than were their male counterparts
simply reaffirmed the notion that women are their own beSt supporters,
yet women also succumbed to "status-quo" responses. Furthermore, the
finding points out the need for orienting men, who have served as
occupational "gate-keepers? in the past, more toward equal utilization
of individuals throughout the labor force and away from the assumption
of "inherent" aptitudes, abilities, or appropriateness based on
chromosomal composition.

Perhaps the most significant aspect of this research lies with

the finding that the sex dominance and SES classification of the job

43

influences to a large degree the perceived appropriateness of women in
that job. The implications of this finding should caution researchers
against using the classic "one-job" design for measuring any kind of
occupational sex bias. That subjects do view jobs differently in

terms of their sex dominance as well as their 3E5 necessitates that a
more multifaceted theory of occupational sex discrimination be developed
to deal with the "parameter" aspects of the problem.

The significant test of the syllogistic effect reinforced the
notion that the typg_of job influences subjects' responses. Both males
and females agreed with the major and minor premises and the conclusion
more often than expected when the job in question was female dominated.
With regard to male subjects, their responses to all other job types
did not fit the syllogistic pattern, i.e., their agreement to a, b, and
g_was less than expected by chance. For females, however, the syllogi-
stic pattern appeared more often than expected for all other jobs
except those that were low-status, male dominated. The data indicated
that females' belief systems regarding the suitability of women for
various occupations (except low-status, male) are more logically con-
sistent when compared to male subjects. Although the nonindependence
of the data renders the findings suggestive, the findings point to the
utility of using the syllogistic framework for a less obtrusive test
of the structure of biased belief systems.

This research, as does most research, raises questions along
with those that it addresses. It has been demonstrated that subjects
can differentiate the components of their sex—biased attitude in terms

of role and trait aspects. Furthermore, the characteristics of jobs

44

that can differentially affect responses has been pointed out. To
demonstrate the interrelationship among these components with a variety
of different occupations in an attitude-change study would bolster the
validity of such an approach as well as furnish the means to influence
individuals' attitudes within the work setting and eliminate biases
which impede the complete utilization of women into the labor force.

Finally, the lack of female dominated occupations at the high
SES level precluded the possibility of testing a balanced design at
three, rather than two status levels. A11 high-status jobs tended to
be male dominated, so no high, female dominated jobs were available.
Moreover, there was a problem with the selection of sex-neutral, low-
status occupations. Low status tended also to imply sex typing. The
jobs that were chosen for this level were that of textile weaver and
shoe-factory worker-~hard1y very relevant to the average introductory
student. These problems with job selection tend to reflect, or be
reflective of, the sexual stratification of our labor force at the
lower SES levels and the complete absence of female participation at
the high-status end of the continuum.

Approaching sex bias in terms of the traits and roles which
underlie it may provide the means for more adequate understanding of
the phenomenon as well as suggest ameliorative steps. Specifying job
attributes that can interact to influence attitudes forces awareness
of the many subtleties involved in the general problem of sex discrimi-
nation. Thus, dealing with sex bias in terms of its multifacetedness

may be the route to the resolution of occupational sex discrimination.

APPENDICES

APPENDIX A

LIST OF TRAITS AND OCCUPATIONS USED

IN QUESTIONNAIRE

45

cowoflmse
HommcoE wnoum

 

Hmuusmzuxmm .msumum wumumpoz

umxuo3 zuouUMMImonm
H0>003 mafluxou

 

HMHHSOZwam .msumumlzoq

umwcocmwp
o>wumxamu
HmaoH

wauo3 anacom
Honommu Hoonom mhmucmEmHm

Houomumouano
usmmm oocouamcw

 

 

 

HmuustIxmw

mcwcfiEwm .maumum mumuwpoz

owns: HMOfluomum
memwupuwmn

 

mcflcHEmm .msuwumlzoq

mZOHfiflmDUUO

omegafimw
m>wuflmcmm
w>Huomuuum

mcflcfiEwm

WBHde

ocwasommz .msumum oumnopoz

um>flnplwxmu
Houwcmn

 

oceanommz .mpumumlzoq

mcflasomME
m>fiuwummfioo
ucmpcmmwpcfl

mcwasommz

mmH¢ZZOHBmMDO 2H 0mm: mZOHfimmDUOO 92¢ mBHde m0 BmHA

< xHszmm4

APPENDIX B

QUESTIONNAIRE (FORM 1)

APPENDIX B

Occupational Attitudes

We are interested in knowing your thoughts about certain occupations.
There are no right or wrong answers to these questions. You will merely
indicate the extent or degree of your agreement with each of the state-
ments. You will circle the number of the response which best indicates
your attitude about that statement.

Please use the following response categories:

Strongly Agree
Agree

Neither agree nor disagree

Disagree
Strongly Disagree

 

U1 9. w IO Dd
ll

 

Please be sure that your responses are circled under each of the atti-
tude statements and that you answer every statement. Since there are
several pages to your booklet, we ask that you answer every page and
do not skip any.

Thank you very much for your participation and cooperation!

If you are interested in results from the entire group (since your
responses are anonymous we cannot tell you anything about your own
responses) you can contact the person below at the end of Spring Term:
Phyl Mellon

436 Baker Hall
353-8708

46

47

Remember to use these responses:

1 = Strongly Agree

2 = Agree

3 = Neither agree nor disagree
4 = Disagree

5 = Strongly Disagree

Good elementary school teachers should be loyal.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good janitors should be talkative.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good dentists should be feminine.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good janitors should be loyal.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good store managers should be masculine.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good textile weavers should be attractive.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good dentists should be talkative.
1=Strongly Agree 2=Agree 3=Neither =Disagree

Good hairdressers should be loyal.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good shoe-factory workers should be independent.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good store managers should be independent.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good elementary school teachers should be dishonest.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good taxi-drivers should be loyal.
1=Strongly Agree 2=Agree =Neither 4=Disagree
Good musicians should be talkative.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good engineers should be attractive.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good textile weavers should be talkative.
1=Strongly Agree 2=Agree =Neither 4=Disagree

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

(Circle the number of your response)

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

48

Good shoe-factory workers should be dishonest.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good social scientists should be competitive.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good chiropractors should be sensitive.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good engineers should be masculine.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good hairdressers should be sensitive.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good taxi-drivers should be masculine.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good store managers should be attractive.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good practical nurses should be competitive.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good social workers should be independent.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good practical nurses should be masculine.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good hairdressers should be talkative.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good social scientists should be independent.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good insurance agents should be feminine.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good dentists should be independent.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good textile weavers should be independent.
1=Strongly Agree 2=Agree

Good social scientists should be attractive.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good dentists should be competitive.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good elementary schoOl teachers should be masculine.
1=Strongly Agree

2=Agree =Neither 4=Disagree

3=Neither 4=Disagree'

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

 

49

Good textile weavers should be dishonest.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

5=Strongly

Good elementary school teachers should be attractive.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good shoe-factory workers should be talkative.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good textile weavers should be loyal.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good chiropractors should be talkative.
1=Strongly Agree 2=Agree 3=Neither

=Disagree

Good practical nurses should be sensitive.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good chirOpractors should be dishonest.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good chiropractors should be loyal.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good taxi-drivers should be competitive.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good hairdressers should be dishonest.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good chiropractors should be independent.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good janitors should be dishonest.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good social scientists should be sensitive.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good insurance agents should be independent.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good social workers should be masculine.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good musicians should be feminine.
1=Strongly Agree 2=Agree =Neither 4=Disagree
Good musicians should be loyal.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good taxi-drivers should be independent.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

50

Good dentists should be dishonest.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good social workers should be talkative.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good practical nurses should be feminine.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good store managers should be feminine.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good engineers should be dishonest.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good hairdressers should be feminine.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good store managers should be sensitive.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good dentists should be attractive.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good social workers should be loyal.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good engineers should be sensitive.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good practical nurses should be dishonest.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good textile weavers should be feminine.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good engineers should be talkative.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good taxi drivers should be sensitive.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good insurance agents should be masculine.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good social workers should be competitive.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good shoe-factory workers should be loyal.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good insurance agents should be competitive.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

51

Good musicians should be masculine.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good engineers should be competitive.
1=Strongly Agree 2=Agree =Neither 4=Disagree 5=Strongly

Good elementary school teachers should be competitive.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good shoe-factory workers should be feminine.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good social scientists should be loyal.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good taxi-drivers should be attractive.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good musicians should be attractive.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good engineers should be loyal.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good insurance agents should be sensitive.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good chiropractors should be attractive.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good insurance agents should be attractive.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good social scientists should be talkative.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good chiropractors should be feminine.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good musicians should be sensitive.

=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good practical nurses should be loyal.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good textile weavers should be masculine.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good hairdressers should be attractive.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good practical nurses should be talkative.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

52

Good janitors should be masculine.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good elementary school teachers should be sensitive.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good store managers should be competitive.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good shoe=factory workers should be sensitive.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good engineers should be feminine.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly

Good elementary school teachers should be independent.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly

Good store managers should be talkative.

1=Strongly Agree" 2=Agree 3=Neither 4=Disagree 5=Strongly
Good social scientists should be masculine.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good dentists should be sensitive.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good shoe-factory workers should be competitive.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good musicians should be dishonest.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good textile weavers should be sensitive.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good janitors should be feminine.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good store managers should be dishonest.

1=Strongly Agree- 2=Agree 3=Neither 4=Disagree 5=Strongly
Good social workers should be sensitive.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good insurance agents should be talkative.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good chiropractors should be masculine.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly
Good store managers should be loyal.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

53

Good social scientists should be dishonest.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good engineers should be independent.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good janitors should be competitive.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good insurance agents should be dishonest.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good musicians should be independent.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good social scientists should be feminine.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good elementary school teachers should be talkative.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good musicians should be competitive.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good practical nurses should be independent.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good elementary school teachers should be feminine.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good chiropractors should be competitive.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good social workers should be feminine.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good janitors should be sensitive.
=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good insurance agents should be loyal.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good social workers should be dishonest.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good janitors should be independent.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

dishonest.
=Neither 4=Disagree

Good taxi-drivers should be
1=Strongly Agree 2=Agree

Good hairdressers should be
1=Strongly Agree

independent.

2=Agree 3=Neither 4=Disagree

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

 

54

Good textile weavers should be competitive.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good social workers should be attractive.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good taxi-drivers should be talkative.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree
Good hairdressers should be competitive.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good practical nurses should be attractive.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good dentists should be masculine.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good shoe-factory workers should be attractive.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good hairdressers should be masculine.
1=Strongly Agree 2=Agree =Neither 4=Disagree

Good dentists should be loyal.
1=Strongly Agree 2=Agree =Neither 4=Disagree

Good shoe-factory workers should be masculine.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good janitors should be attractive.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

Good taxi-drivers should be feminine.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree

In general, men are

1=Strongly Agree 2=Agree 3=Neither
more masculine than women.

2=Agree 3=Neither 4=Disagree

In general, men are
1=Strongly Agree

In general, men are
1=Strongly Agree

better engineers than women.
2=Agree 3=Neither 4=Disagree

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

5=Strongly

better textile weavers than women.
=Disagree 5=Strongly

5=Strongly

5=Strongly

In general, women are better shoe-factory workers than man.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree

In general, men are better hairdressers than women.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree
In general, men are more dishonest than women.

1=Strongly Agree ~2=Agree 3=Neither 4=Disagree

5=Strongly

5=Strongly

5=Strongly

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

Disagree

55
In general, women are better practical nurses than men.
1=Strongly Agree 2=Agree =Neither 4=Disagree 5=Strongly Disagree

In general, women are more feminine than men.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

In general, women are better elementary school teachers than men.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

In general, women are more competitive than men.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

In general, men are better social workers than women.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

In general, men are better store managers than women.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

In general, women are more talkative than men.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

In general, men are better janitors than women.
1=Strongly Agree 2=Agree =Neither 4=Disagree 5=Strongly Disagree

In general, men are more loyal than women.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

In general, women are better musicians than men.
=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

In general, women are better taxi-drivers than men.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

In general, men are more independent than women.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

In general, women are better dentists than men.
=Strongly Agree 2=Agree 3=Neither =Disagree 5=Strongly Disagree

In general, women are more attractive than men. .
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

In general, women are better chiropractors than men.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

In general, women are better social scientists than men.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

In general, men are more sensitive than women.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

In general, men are better insurance agents than woman.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

56

*It is less desirable for women than men to have a job that requires
responsibility.
1=Strongly Agree 2=Agree =Neither 4=Disagree 5=Strongly Disagree

Women have the objectivity required to evaluate business situations
properly.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

Challenging work is more important to men than it is to women.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

Men and women should be given equal opportunity for participation in
management training programs.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

Women have the capability to acquire the necessary skills to be
successful managers.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

On average, women managers are less capable of contributing to an orga-
nization's overall goals than are men.
1=Strongly Agree 2=Agree =Neither 4=Disagree 5=Strongly Disagree

It is not acceptable for women to assume leadership roles as often as
men.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

The business community should someday accept women in key managerial
positions.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

Society should regard work by female managers as valuable as work by
male managers.
=Strongly Agree 2=Agree =Neither 4=Disagree 5=Strongly Disagree

It is acceptable for women to compete with men for top executive
positions.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

The possibility of pregnancy does not make women less desirable
employees than men.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

Women would no more allow their emotions to influence their managerial
behavior than would men.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

Problems associated with menstruation should not make women less
desirable then men as employees.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

57

To be a successful executive, a woman does not have to sacrifice some
of her femininity.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

On the average, a woman who stays at home all the time with her children
is a better mother than a woman who works outside the home at least
half-time.

1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

Women are less capable of learning mathematical and mechanical skills
than are men.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

Women are not ambitious enough to be successful in the business world.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

Women cannot be assertive in business situations that demand it.
1=Strongly Agree 2=Agree =Neither 4=Disagree 5=Strongly Disagree

Women possess the self-confidence required of a good leader.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

Women are not competitive enough to be successful in the business
world.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

Women cannot be aggressive in business situations that demand it.
1=Strongly Agree 2=Agree 3=Neither 4=Disagree 5=Strongly Disagree

58

This set of items has a slightly different response category. It consists of a seven-
point rating scale on which you are to indicate the extent to which each of the items
below are true of you. Please mark a number from 1 to 7 to the left of each item.
Use the rating scale noted below.

/ 1 [7 2 / 3 1 4 / 5 / 6 / 7 /
Never of almost Always or almost always

never true Of me true Of me

Remember, please give us responses of l, 2, 3, 4, S, 6, or 7 to the left of each item,
depending upon how much you believe that item is true of you.

Self-reliant Makes decisions easily
Yielding CompasSionate
Helpful Sincere

Defends own beliefs Self-sufficient

_____ Cheerful Eager to soothe hurt feelings
_____Moody Conceited

____ Independent Dominant

____ Shy _____Soft spoken

____ Conscientious ____ Likeable

__ Athletic __ Masculine
_____Affectionate _____Warm

____ Theatrical ____ Solemn

____ Assertive _____Willing to take a stand
____ Flatterable _____Tender

____ Happy ____ Friendly

____ Strong personality ____ Aggressive

____ Loyal ____ Gullible

‘____ Unpredictable ____ Inefficient

____ Forceful ____ Acts as a leader

____ Feminine _____Chi1dlike
_____Reliable _____Adaptab1e

____ Analytical ____ Individualistic

____ Sympathetic ____ Does not use harsh language
____ Jealous _____Unsystematic

_____Has leadership abilities _____Competitive

____ Sensitive to the needs of others ____ Loves children

__ Truthful __ Tactful

‘_____Willing to take risks _____Ambitious

_ Understanding __ Gentle

_____Secretive ____ Conventional

59

Please fill in the following information:

Age: years
**SEX: Male
Female
MAJOR:

 

Year in College

 

Father's occupation

 

Mother's occupation

 

Marital Status: Single

Married
Divorced or Separated

Widowed

What are your educational plans? Check the one which applies most to your plans.

One to three years of college

Graduate from college

Obtain a master's degree

Obtain a Ph.D., M.D., D.Ed., or other advanced degree

Uncertain as to my future education

 

3’
H:
n
(D
’1

college, what occupation do you ideally see yourself in?

 

HAND PREFERENCE: Please check the category that most accurately describes your

hand preference for each task:

 

Indicate hand preference:

Always Usually No Pre- Usually Always
left left ference right right

 

1. To write a letter legibly

 

2. To throw a ball to hit a target

 

3. To hammer a nail into wood

 

A

. To unscrew the lid of a jar

 

Is any member of your immediate family (biological parents or siblings) left-handed?
YES NO

 

 

How good is your sense of direction? (Please check a number from 1 to 7 indicating
the extent to which your sense of direction is poor or excellent)

[7 1 ./ 2 4/ 3 Al 4 / 5 g1;_ 6 / 7 /
POOR EXCELLENT

Note: * The remaining items (excluding sex) in the questionnaire were not used in

this analysis.
** Sex was used in the analysis.

BIBLIOGRAPHY

BIBLIOGRAPHY

Anderson, N. Likableness ratings of 555 personality-trait words.
Journal of Personality and Social Psychology, 1968, 2, 272-279.

Bass, B., Krusell, J., & Alexander, R. Male managers' attitudes toward
working women. American Behavioral Scientist, 1971, 15, 221-
236.

 

Bem, S. The measurement of psychological androgyny. Journal of Con-
sulting and Clinical Psychology, 1974, 22, 155-162.

 

Broverman, I., Vogel, S., Broverman, D., Clarkson, F., & Rosenkrantz, P.
Sex-role stereotypes: A current appraisal. Journal of Social
Issues, 1972, 28, 59—78.

Cecil, R., Paul, R., & Olins, R. Perceived importance of selected vari-
ables used to evaluate male and female job applicants. Per-
sonnel Psycholggy, 1973, 2g, 397-404.

 

Clarkson, F., Vogel, S., Broverman, I., Broverman, D., & Rosenkrantz, P.
Family size and sex-role stereotypes. Science, 1970, 167,
390-392.

Clifton, A., McGrath, D., & Wick, B. Stereotypes of woman: A single
category? Sex Roles, 1976, 3, 135-147.

Cohen, S., & Bunker, K. Subtle effects of sex role stereotypes on
recruiters' hiring decisions. Journal ongpplied Psychology,

Deaux, R., & Emswiller, T. Explanations of successful performance on
sex-linked tasks: What is skill for the male is luck for the
female. Journal of Personalipy and Social Psychology, 1974, 22,
80-85.

Dipboye, R., Fromkin, H., & Wiback, K. Relative importance of applicant
sex, attractiveness, and scholastic standing in evaluation of job
applicant resumes. Journal of Applied Psychology, 1975, pg,
39-43.

Dunn, O. J. Multiple comparisons among means. Journal of the American
Statistical Association, 1961, 5Q, 52-64.

60

61

Gilmer, B. Industrial Psychology. New York: McGraw-Hill, 1961.

Gordon, F., & Hall, D. Self-image and stereotypes of femininity: Their
relationship to women's role conflict and coping. Journal of
Applied Psychology, 1974, 55, 241-243.

 

Gullahorn, J. The choice of challenge: Womanhood and personhood. Paper
presented at the conference on Mental Health for Women: Chal-
lenges and Choices, Rochester, New York, 1977.

Hamner, W., Kim, J., Baird, L., & Bigoness, W. Race and sex as determi-
nants of ratings by potential employers in a simulated work-
sampling task. Journal of Applied Psychology, 1974, 55, 705-711.

 

Hesselbart, S. Women doctors win and male nurses lose. Sociology of
Work and Occupations, 1977, 3, 49-62.

 

 

Ilgen, D., & Terborg, J. Sex discrimination and sex-role stereotypes:
Are they synonymous? No! Organizational Behavior and Human
Performance, 1975, 154-157.

 

 

Komarovsky, M. Dilemmas of masculinity. New York: W. Norton & Co.,
Inc., 1976.

 

McGuire, W. Cognitive consistency and attitude change. Journal of
Abnormal and Social Psychology, 1960, 55, 345-353.

 

 

Megargee, E. Influence of sex roles on the manifestation of leadership.
Journal of épplied Psychology, 1969, 55, 377-382.

 

Miner, J. Studies in management education. New York: Springer, 1965.

 

 

Oppenheimer, V. The sex-labeling of jobs. Industrial Relations, 1968,
7, 307-325. .

Peters, L., Terborg, J., & Taynor, J. Women as Managers Scale (WAMS):
A measure of attitudes toward women in management positions.
JSAS Catalog of Selected Documents in Psychology, 1974, No. 585.

Petty, M., & Lee, G. Moderating effects of sex of supervisor and sub-
ordinate on relationships between supervisory behavior and
subordinate satisfaction. Journal of Applied Psychology, 1975,
55, 624-628.

 

Reiss, D., Duncan, 0., Hatt, P., & North, C. Occupations and social
status. New York: The Free Press of Glencoe, 1971.

 

Rosen, B., & Jerdee, T. The influence of sex-role stereotypes on evalu-
ations of male and female supervisory behavior. Journal of
Applied Psychology, 1973, 51, 44-48.

 

62

Rosen, B., & Jerdee, T. Influence of sex role stereotypes on personnel
decisions. Journal of Applied Psychology, 1974(a), 55, 9-14.

 

Rosen, B., & Jerdee, T. Effects of applicant's sex and difficulty of
job on evaluations of candidates for managerial positions.
Journal of Applied Psychology, 1974(b), 55, 511-512.

 

Rosen, B., & Jerdee, T. Effect of employee's sex and threatening
versus pleading appeals on managerial evaluations of grievances.
Journal of Applied Psychology, 1975, 55, 442-445.

Rosenkrantz, P., Vogel, S., Bee, H., Broverman, I., & Broverman, D.
Sex-role stereotypes and self-concepts in college students.
Journal of Consulting and Clinical Psychology, 1968, 53, 287-295.

Schein, V. The relationship between sex role stereotypes and requisite
management characteristics. Journal of Applied Psychology,
1973, 51, 95-100.

 

Schein, V. Relationships between sex role stereotypes and requisite
management characteristics among female managers. Journal of
Applied Psychology, 1975, 55, 340-344.

 

 

Spence, J., & Helmreich, R. The attitudes toward women scale: An ob-
jective instrument to measure attitudes toward the rights and
roles of women in contemporary society. JSAS Catalog of
Selected Documents in Psychology, 1972, No. 153.

 

 

Taynor, J., & Deaux, K. When women are more deserving than men:
Equity, attribution, and perceived sex differences. Journal of
Personality and Social Psychology, 1973, 35, 360-367.

 

Terborg, J. WOmen in management: A research review. Journal of Applied

 

Psychology, 1977, 52, 647-664.

 

Terborg, J., & Ilgen, D. A theoretical approach to sex discrimination
in traditionally masculine occupations. Organizational
Behavior and Human Performance, 1975, 55, 352-376.

 

Touhey, J. Effects of additional women professionals on ratings of
occupational prestige and desirability. Journal of Personality
and Social Psychology, 1974, 35, 86-89.

 

 

United States Bureau of the Census. 1960 Census of population, alpha-
betical index of occupations and industries. Washington, D.C.:
U.S. Government Printing Office, 1960.

Winer, B. J. Statisticalgprinciples in experimental design. New York:
McGraw-Hill, Inc., 1971.