Eagly 1984

Journal or Personality and Social Psychology Copyright 1084 bv the
1984, Vol. 46, No. 4, 735-754 i Psychological Association, Inc.
Gender Stereotypes Stem From the Distribution of

Women and Men Into Social Roles
Alice H. Eagly and Valerie J. Steffen
Purdue University
According to stereotypic beliefs about the sexes, women are more communal (selfless
and concerned with others) and less agentic (self-assertive and motivated to master)
than men. These beliefs were hypothesized to stem from perceivers' observations
of women and men in differing social roles: (a) Women are more likely than men
to hold positions of lower status and authority, and (b) women are more likely
than men to be homemakers and are less likely to be employed in the paid work
force. Experiments 1 and 2 failed to support the hypothesis that observed sex
differences in status underlie belief in female communal qualities and male agentic
qualities. Experiment 3 supported the hypothesis that observed sex differences in
distribution into homemaker and employee occupational roles account for these
beliefs. In this experiment, subjects perceived the average woman and man stereo-
typically. Female and male homemakers were perceived as high in communion and
low in agency. Female and male employees were perceived as low in communion
and high in agency, although female employees were perceived as even more agentic
than their male counterparts. Experiments 4 and 5 examined perceptions that
might account for the belief that employed women are especially agentic: (a) A
double burden of employment plus family responsibilities did not account for this
belief, and (b) freedom of choice about being employed accounted for it reason-
ably well.
Gender stereotypes, like other social stereo- dren, they are likely to believe that charac-
types, reflect perceivers' observations of what teristics thought to be necessary for child care,
people do in daily life. If perceivers often ob- such as nurturance and warmth, are typical
serve a particular group of people engaging in of women. Because most of people's activities
a particular activity, they are likely to believe are determined by their various social roles,
that the abilities and personality attributes re- stereotypes about groups of people should re-
quired to carry out that activity are typical of flect the distribution of these groups into social
that group of people. For example, if perceivers roles in a society. Furthermore, certain ste-
consistently observe women caring for chil- reotypes may reflect the distribution of groups
into broader aspects of social structure such
as social class. For example, beliefs about racial
The research reported in this article was supported by differences may be based at least in part on
National Science Foundation Grants BNS-7711671, BNS- observations that racial groups differ in social
7924471, and BNS-8023311. A preliminary report was class (Feldman, 1972; Smedley & Bayton,
presented at the meetings of the Midwestern Psychological
Association, May 1982, and the American Psychological 1978; Triandis, 1977).
Association, August 1982. In applying this social structural analysis to
The authors thank Wendy Wood for help with the design people's beliefs about gender, we faced two
and statistical analyses of Experiment 1; Vathsala Venu- issues: (a) What is the content of stereotypes
gopalan for help with the statistical analyses of Experiments about women and men? (b) What are the major
1 and 2; and Barry Brumer, Deborah Rugs, Jan Schafer,
Erik Bronner, and Gail Winbury for assistance in admin- differences in the ways that women and men
istering materials to subjects. The authors also thank Shelly are distributed into social roles? Concerning
Chaiken, Kay Deaux, Chester Insko, Dafna Izraeli, Laurie content, we decided to restrict our focus to
Lewis, Myron Rothbart, Wendy Wood, and two anony- the beliefs about gender that, by virtue of the
mous reviewers for their comments on a draft of the article.
Requests for reprints should be sent to Alice H. Eagly, frequency with which they have been docu-
Department of Psychological Sciences, Purdue University, mented by research and amplified in theoret-
West Lafayette, Indiana 47907. ical discussions, appear to be most important.
735
736 ALICE H. EAGLY AND VALERIE J. STEFFEN
These beliefs concern communal and agentic more easily than do women (e.g., Broverman
personal qualities: Perceivers generally assume et al., 1972; Spence & Helmreich, 1978; Taylor,
that men are oriented toward agentic goals Fiske, Etcoff, & Ruderman, 1978) stem from
and women toward communal goals (e.g., perceivers' inferences that (a) women occupy
Bern, 1974; Block, 1973; Broverman, Vogel, lower status positions than men and (b) the
Broverman, Clarkson, & Rosenkrantz, 1972; lower an individual's status relative to other
Spence & Helmreich, 1978). Following Bak- persons, the more that individual yields to their
an's (1966) discussion of this distinction, influence. To extend this analysis to the com-
agentic qualities are manifested by self-asser- munal and agentic aspects of gender stereo-
tion, self-expansion, and the urge to master, types, we hypothesized that people who are
whereas communal qualities are manifested higher in status and authority have been ob-
by selflessness, concern with others, and a de- served to behave with less communion (self-
sire to be at one with others. This distinction lessness and concern with others) and more
has been accorded considerable importance in agency (self-assertion and urge to master) than
theoretical discussions of gender (Bakan, 1966; those who have lower status positions. There-
Parsons, 1955) and in the development of fore, perceivers' observations that women oc-
measures of sex-typed and androgynous per- cupy lower status positions than men may lead
sonalities (e.g., Bern, 1974; Spence & Helm- them to believe that women are more com-
reich, 1978). Furthermore, these beliefs about munal and less agentic than men.
gender appear to be cross-culturally general By a similar logic, the differing distributions
(Williams & Best, 1982). of women and men into the roles of home-
To examine why women are perceived as maker and employee may account for the ste-
communal and men as agentic, we considered reotypic beliefs that women are communal and
two major differences in the distribution of men are agentic. Because the labor-force par-
females and males into social roles. The first ticipation rates of women (51.2%) and men
of these differences is that women are more (77.2%) still differ considerably (U.S. Depart-
likely than men to hold positions at low levels ment of Labor, 1980), perceivers are likely to
in hierarchies of status and authority and are have observed fewer women than men in em-
less likely to hold higher level positions. The ployee roles and almost exclusively women in
second difference is that women are more likely the homemaker role. The perception of women
than men to be homemakers and are less likely as less agentic and more communal than men
to be employed in the paid work force. would follow if employees have been observed
Given the pervasiveness in natural settings to behave more assertively and masterfully
of sex differences in status, it seems plausible than homemakers as well as less selflessly and
that gender stereotypes stem from the tendency supportively toward others. According to this
of perceivers to observe women in lower status analysis, the stereotypic differences between
roles than men. Such observations would be women and men should parallel the differences
made in organizational settings in which the that people perceive between homemakers and
positions held by men tend to be higher in employees. Some empirical support for this
status and authority than the positions held hypothesis is provided by Clifton, McGrath,
by women (e.g., Brown, 1979; England, 1979; and Wick's (1976) finding that the communal
Kanter, 1977; Mennerick, 1975). Also, in attributes ordinarily ascribed to women were
family settings husbands tend to have an over- assigned only to housewives and not to four
all power and status advantage over wives other categories of women (female athlete, ca-
(Blood & Wolfe, 1960; Gillespie, 1971; Scan- reer woman, club woman, and "bunny").
zoni, 1982). Such differences in the status of The experiments that we have carried out
men's and women's roles may be determining to test these ideas share several features of
factors in beliefs about gender. design. To minimize demand characteristics
Another reason for examining status dif- stemming from subjects' knowledge of our hy-
ferences is that recent research by Eagly and potheses, each subject read a description of
Wood (1982) demonstrated that the stereotypic only one woman or man. In addition, the as-
beliefs that women are more easily influenced pect of social roles presumed to account for
than are men and that men exert influence gender stereotypes (hierarchical status, or oc-
GENDER STEREOTYPES 737
cupation as a homemaker or employee) was locations agreed to participate. Subjects' mean age was
varied: (a) In Experiments 1 and 2, which ex- 22.30 years in Experiment 1 and 21.81 years in Experi-
ment 2.
amined hierarchical status, some stimulus
persons had high-status job titles and some
had low-status job titles, and (b) in Experiment Procedure
3, which examined the homemaker-employee Each subject read a brief description of an employee
distinction, some stimulus persons were (e.g., "Phil Moore is about 35 years old and has been
homemakers and others were employees. For employed for a number of years by a supermarket. He is
one of the managers") and rated this stimulus person.
other stimulus persons,, this stereotype-relevant The descriptions in Experiment 1 varied according to a
aspect of the social role was omitted: (a) In 2 X 3 X 2 (female vs. male X high-status job title vs. low-
the status experiments, the job title was omit- status job title vs. no job title X bank vs. supermarket
ted, and (b) in the homemaker-employee ex- setting) factorial design. The design of Experiment 2 dif-
fered with respect to the setting variable, which had four
periment, designation as a homemaker or em- levels because a medical clinic and a university department
ployed person was omitted. It was in these of biology were added.
experimental conditions, in which the role de- In the laboratory sessions of Experiment 1, a female
scriptions were omitted, that subjects should experimenter administered materials to subjects in groups
have manifested gender stereotyping. Accord- of about 25. Subjects first indicated their age and sex. To
ensure that subjects thought carefully about the stimulus
ing to our analysis, perceivers view men and person, the experimenter had them "spend a moment
women stereotypicaUy in the absence of role thinking about" the stimulus person (after reading the
information, because under such conditions description) and then write a few sentences about the per-
the attributes ascribed to women and men re- son. Subjects then responded to the measures described
below.
flect the differing social roles that underlie the At the public campus locations in both experiments,
stereotypes. In contrast, the addition of role an experimenter approached each subject by asking her
descriptions to female and male stimulus per- or him to participate in a study on "impressions of other
sons prevents gender-stereotypic judgments if people." After the subject had completed the questionnaire,
such descriptions (e.g., job titles) provide clear- the experimenter asked her or his-age and recorded this
information along with the subject's sex.
cut information about the aspect of social roles
that ordinarily covaries with sex (e.g., hier- Manipulation of Independent Variables
archical status). In the presence of such role
information, the covariation of sex and role Sex of stimulus person. The stimulus persons were
that is the implicit basis of gender stereotypes either female or male. Sex was identified by sex-typed
names (e.g., Sue Fisher, Phil Moore).
is removed, and role would determine per- Status of job title and setting of jab. The stimulus
ceivers' beliefs about people's attributes. persons had either a high-status job title, a low-status job
Women and men who have the same role title, or no job title. To provide an internal replication of
would be perceived equivalently. the design, the stimulus persons were described as employed
by a bank or a supermarket and, in Experiment 2, also
by a medical clinic or a university department of biology.
Experiments 1 and 2 The high- and low-status job titles for these settings were
vice-president and teller, manager and cashier, physician
Method and x-ray technician, and professor and lab technician,
respectively.
Subjects
In Experiment 1, 276 females and 208 males partici- Measuring Instruments1
pated. Of these subjects, 256 were University of Massa-
chusetts psychology students who participated in a lab- Beliefs about stereotypic attributes. Using 5-point
oratory setting to obtain extra-credit course points. An scales, subjects rated the stimulus persons on 18 attributes,
experimenter randomly selected the remaining 228 subjects presented either as personality characteristics (Experiment
by choosing on repeated occasions every fourth person 1) or as.attributes of on-the-job behavior (Experiment 2).
seated in a University of Massachusetts coffee shop. In Each on-the-job rating scale was preceded by a question
Experiment 2, 237 females and 243 males participated. asking how much of the attribute the stimulus person
One female and one male experimenter each randomly exhibited on the job (e.g., "How competitive do you think
selected half of the subjects by choosing every fourth person this person is on the job?").
seated in a coffee shop or general library at Purdue Uni-
versity. Especially in Experiment 1, the subjects sampled
from the public campus locations included university staff
as well as students. la the experiments reported in this ' For all experiments, measures are listed in the order
article, 80% or more of the persons selected from such in which they were administered.
738 ALICE H. EAGLY AND VALERIE J. STEPHEN
Attributes were selected primarily from the Personal ported here. Also, analyses including subject
Attributes Questionnaire (Spence & Helmreich, 1978) to sex as an additional variable yielded very few
ensure that they (a) represented gender stereotypes and
(b) included both communal and agentic qualities. A factor differences between female and male subjects
analysis (varimax orthogonal rotation) of subjects' ratings for any of the experiments in this series.3
was performed for each of the five experiments in this Therefore, this variable was also dropped from
series. All analyses yielded similar two-factor solutions. all analyses reported in this article. Finally,
One factor, labeled communal, accounted for an average
of 17.4% of the variance in the five experiments, and an-
because analyses including experimenter (fe-
other factor, labeled agentic, accounted for an average of male vs. male) as an additional factor yielded
23.9% of the variance. Although various other labels have very few effects in Experiment 2 (or in Ex-
been used by researchers to characterize these dimensions periments 3 and 5, which also used one ex-
(e.g., expressiveness vs. instrumentality, social orientation perimenter of each sex), this variable was
vs. task orientation, and femininity vs. masculinity), none
provided as close a match to the actual content of the dropped from all reported analyses.
factors as did Bakan's (1966) terms, communion and
agency. Inferred Job Status
The measure of perceived communion was the mean
of each subject's ratings on the attributes that (in the five Salary estimates. On subjects' estimates of
experiments) consistently loaded highly on the communal the stimulus persons' salaries, the main effects
factor: kind, helpful, understanding, warm, aware of others'
feelings, and (in all but Experiment 1, which omitted this of sex of stimulus person and status of job
attribute) able to devote self to others. The measure of title were highly significant for both Experi-
perceived agency was the mean of each subject's ratings ments 1 and 2.4 Women were judged to have
on the attributes that consistently loaded highly on the lower salaries than men: For Experiment 1,
agentic factor: active, not easily influenced, aggressive, in- Ms = $12,154 versus $15,425, respectively,
dependent, dominant, self-confident, competitive, makes
decisions easily, never gives up easily, and (in all but Ex- F(\, 244) = 27.24, p < .001, and for Exper-
periment 1, which omitted this scale) stands up well under iment 2, Ms = $19,749 versus $23,535, re-
pressure. These measures had satisfactory internal con- spectively, F(\, 449) = 15.29, p < .001. Con-
sistency: The mean values of coefficient alpha (Cronbach, sistent with the significant main effects of sta-
1951) in the five experiments were .84 for communion
and .86 for agency. The findings of all five experiments
tus—for Experiment 1, F(2, 244) = 50.58,
are presented in terms of these measures. Because few and for Experiment 2, F(2, 449) = 137.28,
significant effects were obtained on several rating scales ps < .001—persons with high-status job titles
not included in these measures, these findings will not be were judged to earn considerably more than
reported. persons whose job titles were not given, who
Inferred job status. In the laboratory portion of Ex-
periment 1 and in Experiment 2, subjects estimated (in
dollars) the stimulus person's annual salary. In Experiment
2, for stimulus persons who had no job title, subjects also 2
gave their "best guess" concerning the individual's job Because this rating-scale measure proved insensitive
title. Coders blind to the experimental conditions divided to differences between conditions, it was not included in
the job titles into (a) a high-status category consisting of additional studies and will not be discussed further.
3
jobs that either included an administrative or managerial The absence of Sex of Subject X Sex of Stimulus Person
component or required high-level technical skills and (b) interactions is noteworthy in view of research suggesting
a low-status category of other jobs. Coders agreed on ap- that in-group members perceive in-groups more favorably
proximately 95% of these relatively objective judgments, and less stereotypically and homogeneously than they per-
and disagreements were resolved by discussion. In the lab- ceive out-groups (e.g., Brewer, 1979; Tajfel, 1981; see also
oratory portion of Experiment 1, subjects also rated the Park & Rothbart, 1982, regarding perceptions of women
status of the stimulus person's job on a 15-point scale and men). Perhaps this in-group-out-group bias is a man-
ranging from low status to high status.2 ifestation of the self-enhancing tendency in person per-
ception (e.g., Zuckerman, 1979) and occurs when re-
spondents rate in-group members as they would rate
Results themselves. It is likely that in the present experiments,
subjects merely retrieved their concepts of various groups
of people (e.g., male bank tellers, employed women, average
The principal data analyses were Sex of men) and did not treat themselves as exemplars of the
Stimulus Person X Status of Job Title X Setting same-sex categories.
4
analyses of variance (ANOVAS). Because no sig- Because the variance of subjects' salary estimates was
nificant effects occurred in Experiment 1 for extremely heterogeneous, with larger means associated with
larger variances, analyses were performed on the logarithm
locale in which stimulus materials were ad- of the salaries in all experiments in this series. There was
ministered (coffee shop vs. laboratory), this no serious heterogeneity on the other dependent variables
variable was dropped from the analyses re- in these experiments.
in turn were judged to earn more than persons Table 1

with low-status job titles: For Experiment 1, Mean Ratings of Stereotypic Attributes of Female
Ms = $19,236 versus $11,773, p < .001, and and Male Employees Who Varied in Status
$11,773 versus $10,120, p < .025; for Exper- of Job Title: Experiment 1
iment 2, Ms = $32,057 versus $18,548, p< Stimulus High Low No
.001, and $18,548 versus $14,359, p < .001. person status status title
The Sex X Status interactions were nonsignif-
icant, F(2, 244) = 1.64 and P(2, 449) = 1.61. Female
Communal 3.55 3.59 3.69
The findings that women were judged to Agentic 3.74 2.63 2.85
have lower salaries than men, regardless of Male
whether job titles were indicated, suggest that Communal 3.41 3.58 3.48
the salary estimates may have functioned Agentic 3.52 2.43 2.78
largely as a measure of perceived wage dis-
crimination. To provide clear-cut evidence that Note. Means are on a 5-point scale; larger numbers indicate
greater communion or agency. Cell ns ranged from 78 to
lower status was associated with women more 85. For communal, MS, = 0.54; for agentic, MS, = 0.44.
than with men, the tendency to ascribe lower
salaries to women should have been especially
strong in conditions omitting job titles, in 3.34, F(l, 456) = 9.95, p < .002. Consistent
which the typical status difference was not with a significant Sex X Status interaction in
countered by information that equated wom- Experiment 2 only, F(2,456) = 3.46, p < .05,
en's and men's jobs. Thus the absence of the this stronger communal tendency was ascribed
expected Sex X Status interaction in Experi- to women (vs. men) with low-status or no job
ment 1 led us to include a purer indicator of titles (ps < .01 or smaller) but was nonsig-
status, job title guesses, in Experiment 2. nificant with high-status job titles.
Job title guesses. For stimulus persons On agency, the main effects of sex and status
without job titles in Experiment 2, a greater were highly significant in both experiments.
proportion of subjects' job-title guesses were Women were perceived as more agentic than
categorized as high status (vs. low status) for men: For Experiment 1, Ms = 3.08 versus
male than for female stimulus persons. High- 2.92, respectively, F(\, 472) = 7.80, p < .005;
status job titles were ascribed to 48 men and for .Experiment 2, Ms = 3.53 versus 3.37, F[l,
30 women, and low-status job titles were as- 456) = 9,99, p < .002. Consistent with the
cribed to 30 men and 39 women. A loglinear significant main effects of status—for Exper-
analysis of the cell frequencies (Bishop, Fien- iment 1, F(2, 472) = 121.01, and for Exper-
berg, & Holland, 1975; Davis, 1974) revealed iment 2, F(2,456) = 21.33.ps < .001—persons
a significant likelihood ratio chi-square (G2 = with high-status job titles were perceived as
4.82, p < .05) for the interaction between sex
of stimulus person and judged job-title status.
Thus the job-title measure of inferred status Table 2
yielded clear evidence that without knowing Mean Ratings qfOn-the-Job Behavior of Female
job titles, subjects inferred that women held and Male Employees Who Varied in Status
lower status positions than men. of Job Title: Experiment 2
Stimulus High Low No

Beliefs About Stereotypic Attributes person status status title
Subjects' mean ratings of the stimulus per- Female

sons' communion and agency appear in Tables Communal 3.48 3.53 3.57
1 and 2. On communion, the main effect of Agentic 3:77 3.37 3.46
sex of stimulus person was marginally signif- Male
icant in Experiment 1 and significant in Ex- Communal 3.51 3.21 3.30
Agentic 3.61 3.23 3.25
periment 2. Women were perceived as more
communal than men: For Experiment 1, Ms = Note. Means are on a 5-point scale; larger numbers indicate
3.61 versus 3.48, respectively, F(l, 472) = 3.71, greater communion or agency. All cell «s = 80. For com-
p < .06; for Experiment 2, Ms = 3.53 versus munal, MS, = 0.41; for agentic, MS, = 0.34.
more agentic than persons whose job titles were of the findings across four organizational set-
not given or persons with low-status job titles: tings reduces the likelihood that artifacts arose
For Experiment 1, Ms - 3.63 versus 2.81 for from beliefs about particular work environ-
high-status versus no job title and 3.63 versus ments. Third, the generalizability of the find-
2.53 for high- versus low-status job title, re- ings across two different types of ratings (per-
spectively, ps < .001; for Experiment 2, Ms = sonality traits and on-the-job behavior) reduces
3.69 versus 3.36, and 3.69 versus 3.30, re- the likelihood that artifacts arose from insuf-
spectively, ps < .001. In Experiment 1, the ficient sensitivity of global trait ratings. Be-
contrast between the no job title and low- cause inferences from job titles to attributes
status job title conditions was also significant of on-the-job behavior should be easier and
(p < .001).5 more direct than inferences from job titles to
personality traits, the job behavior ratings of
Discussion Experiment 2 should have maximized the
Several aspects of these findings are unfa- possibility of obtaining any effects of explicit
vorable to the hypothesis that the stereotypes and implicit variation of job status.
of female communion and male agency stem Our findings may be more consistent with
from having observed women in lower status the hypothesis that distributions of women and
roles than men. First, our status-difference ex- men into homemaker and employee roles un-
planation of gender stereotypes implies that derlie gender stereotypes. Because all of the
perceptions of lower status persons resemble stimulus persons presented in Experiments 1
those of average women and that perceptions and 2 were described as employed, the relative
of higher status persons resemble those of av- absence of gender-stereotypic perceptions may
erage men. This expectation was only partly reflect the inclusion of this information about
confirmed: Persons with low-status job titles occupational role. Therefore, our third ex-
were perceived as considerably less agentic than periment tested the hypothesis that gender ste-
persons with high-status job titles, but there reotypes stem from perceivers' observations of
was no difference in communion. the distribution of women and men into the
The second and most surprising aspect of roles of homemaker and employed person. In
the findings that discount the status hypothesis this experiment, some female and male stim-
is the counterstereotypic effect that the sex of ulus persons were described as employees and
the stimulus persons had on the ascription of some as homemakers, and the occupation of
agentic traits. Women were perceived as more others was not indicated.
agentic than men despite the perceptions that One implication of the homemaker versus
(a) women earn less than men and hold lower employee explanation of gender stereotypes is
status positions and (b) persons in lower status that the differences perceived between persons
positions are less agentic than persons in higher in these two occupational roles parallel the
status positions. Furthermore, women were stereotypic differences between women and
perceived as more communal than men, even men. Therefore, homemakers were expected
though persons in lower and higher status po- to be perceived as more communal and less
sitions did not differ in perceived communion. agentic than employed persons. It also follows
These perceived sex differences were rela- that women and men are perceived stereotyp-
tively small, despite their statistical signifi- ically when their role assignment as home-
cance. The differences were no larger than .19
on a 5-point scale, with an effect size (d) of
.30 (Cohen, 1977). Nevertheless, several as- 5
In Experiments 1 and 2, several effects of employment
pects of the experiments promote confidence setting were also obtained on beliefs and on inferred status.
in the reliability of the findings. First, the gen- Only one of these effects involved the sex of the stimulus
eralizability of the findings across subjects re- persons. Consistent with this Sex X Setting interaction
cruited at two universities and by two different obtained in Experiment 2 on perceived agency (p < .01),
methods (subject pool and random sampling subjects rated women's behavior as more agentic than men's
in the bank and the supermarket (ps < .01), as marginally
at public campus locations) reduces the like- more agentic than men's in the university department of
lihood that artifacts arose from particular sub- biology (p < .09), but not different from men's in the
ject populations. Second, the generalizability medical clinic.
maker or employee is unknown, because per- employees) allowed us to examine these se-
ceivers have observed that more women than lection considerations.
men are homemakers and fewer women are The greater agency ascribed to employed
employees. Furthermore, the homemaker- women (vs. men) is not plausibly explained
employee analysis implies that women and in terms of a belief that discrimination makes
men are perceived similarly if they have the it necessary for female employees to be more
same occupational role, that is, if both are qualified than their male counterparts. People
homemakers or both are employees. Yet, con- are likely to believe that discrimination exists
sistent with the agency findings of the first two in relation to high-status positions or other
experiments and inconsistent with our social male-dominated jobs, for which traditionally
role analysis, we expected that female em- there were barriers excluding or discouraging
ployees would be perceived as somewhat more women. However, in Experiments 1 and 2,
agentic than male employees. We were less subjects were found to believe in women's su-
confident that female employees would be periority in agentic qualities when low-status
perceived as more communal because of the as well as high-status job titles were given. It
marginal significance of this finding in Ex- seems unlikely that subjects believed that
periment 1. women face discrimination in obtaining the
The design of our third experiment also al- low-status positions used in our research (e.g.,
lowed us to examine two possible explanations bank teller, supermarket cashier).
of the relatively high agency ratings of em-
ployed women. One explanation is that re- Experiment 3
pondents are no longer willing to derogate
women on stereotype questionnaires because Method
of changes in attitudes toward women, greater Subjects
wariness about revealing one's stereotyping,
or possibly other causes. This explanation im- A total of 108 females and 132 males participated. One
female and one male experimenter each randomly selected
plies that it would be impossible to replicate half of the subjects by choosing persons seated in a coffee
the stereotypes of women and men obtained shop or general library at Purdue University. The subjects'
by other investigators (e.g., Broverman et al., mean age was 21.61 years.
1972; Spence & Helmreich, 1978). The inclu- Procedure
sion of average woman and average man cues Each subject read a brief description (e.g., "an average
in our third experiment allowed us to examine man" or "an average woman who is employed full-time")
this issue. and rated this stimulus person. The descriptions varied
According to another explanation for the according to a 2 X 3 (female vs. male X employee vs.
high level of agency ascribed to employed homemaker vs. no occupational description) factorial de-
sign.
women, subjects believed that women are less
likely to be employed than men and therefore Manipulation of Independent Variables
inferred that higher standards are applied to Sex of stimulus person. The stimulus persons were
women than to men by employers (or by described as an average woman or as an average man.
women themselves) when women are selected Occupation of stimulus person. The stimulus persons
for jobs. This selection hypothesis implies that were described as employed full-time, or as caring for a
people who are thus highly selected (or self- home and children and not employed outside of the home,
selected) for a role are believed to be more or no occupational description was provided.
extreme in role-relevant characteristics. If Measuring Instruments
agentic qualities are believed to be relevant to
job success, employed women would be per- Beliefs about stereotypic attributes. The measures de-
scribed for Experiment 1 were used.
ceived as being more agentic than employed Inferred likelihood of employment. For the stimulus
men. It also follows that male homemakers persons for whom no occupational description was pro-
would be perceived as being more communal vided, subjects indicated on an 11-point scale, ranging
than female homemakers (provided that com- from 0% chance to 100% chance, the likelihood that the
person was employed full-time.
munal characteristics are believed to be rel- Inferred job status. For stimulus persons described as
evant to the homemaker role). The inclusion employed full-time, the salary measure described in Ex-
of homemaker stimulus persons (as well as periment 1 was used.
Results Table 3
Mean Ratings of Stereotypic Attributes of
The principal data analyses were Sex of Females and Males Who Varied in
Stimulus Person X Occupation of Stimulus Occupation: Experiment 3
Person ANOVAS.
No
Stimulus occupational
Inferred Likelihood of Employment and person Employee Homemaker description
Inferred Job Status
Female
Subjects who received no occupational in- Communal 3.31 4.22 3.81
formation about the stimulus person inferred Agentic 3.69 3.02 3.00
that the woman was less likely than the man Male
to be employed full-time (Ms = 56.50% vs. Communal 3.39 4.11 3.03
Agentic 3.40 2.90 3.46
79.75%, respectively), F(l, 78) = 30.39, p <
.001. Subjects who rated employees ascribed Note. Means are on a 5-point scale; larger numbers indicate
lower salaries to the woman (M = $15,615) greater communion or agency. All cell ns = 40. For com-
than the man (M= $21,193), F(\, 78) = 13.06, munal, MS, = 0.33; for agentic, MSe = 0.31.
p< .001.
Beliefs About Stereotypic Attributes planned contrasts, employees (regardless of

their sex) were perceived as more agentic than
Subjects' mean ratings of communion and homemakers (ps < .001 for female and male
agency appear in Table 3. On communion, stimulus persons). In addition, for stimulus
the significant main effects of sex, F(\, 234) = persons without an occupational description,
13.32, p < .001, and occupation, F(2, 234) = the traditional gender stereotype of the man
49.08, p < .001, should be interpreted in the as more agentic than the woman was obtained
context of a significant Sex X Occupation in- (p < .001). The female employee was perceived
teraction, F(2, 234) = 12.34,p < .001. These as more agentic than the male employee (p <
effects are best described in terms of the .025), whereas the female and the male home-
planned contrasts implied by the hypotheses. maker were not perceived to differ. The agentic
As expected, homemakers, regardless of their tendency of the woman without an occupa-
sex, were perceived as more communal than tional description did not differ from that of
employees (p < .001 for female stimulus per- the female homemaker but was less than that
sons; p < .005 for male stimulus persons). In of the female employee (p < .001). The agentic
addition, for stimulus persons without occu- tendency of the man without an occupational
pational descriptions, the traditional gender description was greater than that of the male
stereotype of the woman as being more com- homemaker (p < .001) but not different from
munal than the man was obtained (p < .001). that of the male employee.
The female and male employees were not per- These ANOVA findings are generally consis-
ceived to differ in communion, nor were the tent with the theory that gender stereotypes
female and male homemakers. The communal stem from the observed distribution of women
tendency of the woman whose occupation was and men into homemaker and employee roles.
not given was less than that of the female Therefore, it is worthwhile to examine the
homemaker (p < .001) but greater than that correlations between (a) inferred role distri-
of the employed woman (p < .001). The com- butions of the stimulus persons who lacked
munal tendency of the man whose occupation occupational descriptions and (b) the ascrip-
was not given was less than that of the male tion of gender-stereotypic attributes to them.
homemaker (p < .001) or of the employed Overall, the higher the inferred likelihood that
man (p < .005). the average woman or man was employed, the
On agency, the significant main effect of lower was her or his communion, r(78) = —.34,
occupation, F(2, 234) = 21.88, /?<.001, p < .01, and the higher was her or his agency,
should be interpreted in the context of the r(78) = .41, p < .001. Yet inferred likelihood
significant Sex X Occupation interaction, F(2, of employment should have been a stronger
234) =10.14, p<m\. According to the predictor of Stereotypic perceptions for the
woman than for the man. This differential butes of the stimulus persons whose occupa-
predictability would be expected because like- tions were described reveal even more clearly
lihood of employment for the average man that social roles underlie gender stereotypes.
was relatively invariant (SD = 15.61 for av- Subjects believed that, regardless of sex, per-
erage man vs. 21.31 for average woman), F(39, sons described as homemakers differed from
39) = 1.86, p < .05, no doubt because this persons described as employees. Female and
likelihood was generally quite high (see the male homemakers were perceived to be like
previous subsection of results). For the woman, stereotypic women (high in communion and
greater inferred likelihood of employment was low in agency), whereas female and male em-
associated with slightly less communion, ployees were perceived to be like stereotypic
r(38) = -.19, p < .25, and greater agency, men (low in communion and high in agency).
r(38) = .43, p < .01. For the man, the cor- Another important aspect of subjects' beliefs
relations between inferred likelihood of em- about the stimulus persons with occupational
ployment and perceived communion, r(38) = descriptions is that the females and males were
.13, and agency, r(38) = .01, were very small perceived relatively equivalently once their so-
and nonsignificant. cial role as employee or homemaker was spec-
ified. The only exception to this pattern, and
Disctission an exception that demands explanation, is the
significantly greater perceived agency of em-
The findings of Experiment 3 were generally ployed women compared with employed men.
favorable to the hypothesis that a sex difference The comparisons between same-sex persons
in the distribution of women and men into who differed in occupation were also moder-
homemaker and employee roles underlies the ately consistent with our role-distribution the-
stereotype that women are communal and men ory of gender stereotypes. Because subjects
are agentic. Two findings confirmed prereq- judged that there was roughly a 50-50 chance
uisite conditions for testing this hypothesis: (a) that the average woman was employed, per-
The average woman and man without occu- ceptions of her personal attributes should have
pational descriptions were perceived stereo- fallen between those of the employed woman
typically (i.e., the woman was perceived as and the female homemaker. Although this ex-
more communal and the man as more agentic), pectation was fulfilled (see Table 3), the average
and (b) the average man was judged as con- woman's perceived agency did not differ sig-
siderably more likely than the average woman nificantly from that of the homemaker. Be-
to be employed. cause subjects considered it quite likely that
The correlational analyses relating likeli- the average man was employed, their beliefs
hood of employment to the attributes of the about his personal attributes should have been
average woman and man provided one test of similar to their beliefs about the employed
whether observations of women's and men's man. This expectation was fulfilled for
differing occupational roles account for ste- agency, but the average man was perceived as
reotypes. The likelihood that the average significantly less communal than the em-
woman was employed related positively to her ployed man.6
agency and negatively (albeit weakly) to her
communion. In other words, subjects who re-
ported that women are often employed tended
to view women counterstereotypically, and 6
In the interpretation of these findings, subjects' judg-
those who reported that women are less often ments of the probability that the average woman and man
employed (and presumably more often home- are employed should not be treated as exact estimates of
makers) tended to view women stereotypically. the observed distributions of people into employee and
homemaker roles. One reason for our caution is that males
Comparable findings were not obtained for who are not employed have probably been observed in
men, probably because they are generally em- largely different roles than females who are not employed.
ployees and rarely homemakers in the natu- Whereas such females are usually homemakers, probably
rally occurring situations in which subjects such males are retired or seeking employment. Therefore,
had made their previous observations. the subjects' judgment that there is an 80% chance that
the average man is employed does not imply that 20% of
The ANOVA findings concerning the attri- men have been observed as homemakers. It also does not
Although the findings of this experiment rather than a difference in how they are per-
are generally supportive of the idea that perceived. If stimulus persons are implicitly com-
ceived sex differences stem from perceivers' pared with same-sex reference groups, this se-
observations of women and men in differing mantic interpretation suggests that female
occupational roles, they do not explain the employees are implicitly compared with other
tendency for employed women to be perceived females and therefore are judged as very agen-
as more agentic than employed men.7 It is tic, whereas male employees are implicitly
noteworthy that this finding was obtained in compared with other males and therefore are
Experiments 1 and 2 for the stimulus persons judged as not especially agentic. This inter-
without job titles as well as in Experiment 3 pretation also suggests that male homemakers
for the average employed persons because these are implicitly compared with other males and
stimulus persons were probably thought to therefore are judged as very communal,
have lower status positions if they were female. whereas female homemakers are implicitly
As demonstrated in Experiments 1 and 2, compared with other females and therefore
lower status employees were believed to be less are not judged as especially communal. Al-
agentic than higher status employees; there- though female employees were perceived as
fore, any tendency to ascribe lower status to more agentic than male employees, male
women would tend to counteract any factors homemakers were not perceived as more com-
making them appear more agentic than men. munal than female homemakers. The fact that
Both of the explanations that we introduced female and male homemakers were perceived
earlier for this sex difference in agentic qualities equivalently, then, does not support the ex-
were discounted by the findings from Exper- planation that the perceived sex differences
iment 3. The idea that subjects would be un- are an artifact of differences in the reference
willing to derogate women on stereotype ques- group implicitly used by subjects when rating
tionnaires was discounted by their perception females and males. Yet so little is known about
of the average woman as significantly less how people use implicit comparison standards
agentic than the average man. The other ex- that we cannot be completely certain that such
planation of female employees' greater agency a process has no relevance to our findings.
was that categories of people (e.g., women) Because our experiments so consistently re-
who are relatively uncommon in a social role vealed that female employees were perceived
(e.g., employee) are believed to be more strin- as more agentic than their male counterparts
gently selected in terms of the requirements and because this finding is counterstereotypic
of the role than people who commonly occupy and perhaps important, we considered another
the role. This explanation was discounted by explanation of the finding. Accordingly, the
the finding that the male homemaker, who is fact that employed women often balance
considerably rarer than the female employee,
was not perceived to differ from the female
homemaker in communion, the quality that
subjects believed typical of homemakers.8 'Other investigators have reported higher perceived
Yet another explanation of employed wom- agency and competence in females than males, yet in these
en's higher agency is that this perceived gender studies the situation faced by the stimulus person was
judged to be especially difficult or demanding for women
difference reflects a semantic or response-lan- (Abramson, Goldberg, Greenberg, & Abramson, 1977;
guage difference in the way males and females Taynor & Deaux, 1973), or the stimulus materials may
are judged (e.g., Manis, 1971;Upshaw, 1969) have implied employment or special competence (Deaux
& Lewis, in press). Also, in Experiment 3 the communal
ratings of employed women and men did not differ (al-
though they did differ significantly in Experiment 2 and
marginally in Experiment 1). Therefore, only the sex dif-
imply that the average man's personality attributes would ference in employees' agency appears reliable and con-
be perceived as between those ascribed to employees and sequently requires explanation.
8
those ascribed to homemakers. Perhaps the low level of Yet it could be that this selection hypothesis pertains
communion ascribed to the average man reflects the un- to the employee role (because perceivers know that people
favorably evaluated roles ascribed to the 20% of men not are screened for employment) but not to the homemaker
thought to be employed (e.g., being retired, seeking em- role (because perceivers are uncertain about whether people
ployment). are screened for the occupation of homemaker).
two demanding roles—homemaker and em- Parental status of stimulus person. The stimulus per-
ployee—may account for the relatively high sons whose parental status was provided were described
as having children at home or not having children. (The
agency ratings of such women. Perceivers may unmarried woman or man with children at home was
have observed these agentic qualities among further described as a "single parent," to aid subjects in
the women who experience this potential role interpreting this potentially ambiguous description.) For
overload and role conflict. the remaining stimulus persons, no information about
children was provided.
To enable us to test this "double-burden"
explanation, subjects in a fourth experiment
rated the personality attributes of an employed Measuring Instruments
woman or man described as either married or Subjects' beliefs about stereotypic attributes and job
single and as either having or not having chil- status were assessed using measures described in Exper-
dren at home. Other subjects rated an em- iment 1. The subjects also estimated the number of hours
that the stimulus person worked per day doing household
ployed woman or man whose marital and pa- and family-related work. For stimulus persons whose mar-
rental statuses were hot described. Should the ital and parental statuses were not provided, subjects Used
double-burden explanation of women's greater an 11-point rating scale, ranging from 0% to 100% chance,
agency be correct, employed women with to rate the likelihood that the person was married. On a
second, similar scale, the subjects rated the likelihood that
family responsibilities, especially involving the person had children at home.
children, would be regarded as particularly
agentic. Family responsibilities would increase Results and Discussion
the agency ascribed to employed men only if
such men did not have wives who typically The principal data analyses were Sex of
would carry out the household duties. The Stimulus Person X Marital Status of Stimulus
single father would fulfill this criterion and, Person X Parental Status of Stimulus Person
as a consequence, might also be perceived as ANOVAS, with the error term including the two
especially agentic. control conditions, which omitted the infor-
mation about marital and parental statuses.
Experiment 4
Inferences About Marriage, Children,
Method Household Work, and Job Status
Subjects For employees whose marital and parental
A total of 108 females and 135 males participated in statuses were omitted, the woman was judged
groups of about 15 in a laboratory setting to fulfill a psy- less likely to be married (M = 54.17%) than
chology course requirement at Purdue University. One the man (M = 76.00%), F(\, 47) = 20.50, p <
female experimenter administered all stimulus materials. .001. She was also judged less likely to have
Subjects' mean age was 19.33 years. children at home than her male counterpart
(Ms = 43.75% versus 66.00%, respectively),
Procedure F(l, 47) = 16.22, p< .001. Thus subjects did
Each subject read a brief description of a full-time em- not consider it highly likely that the average
ployee (e.g., "an average man who is employed full-time, employed woman faced a double burden of
is married, and has no children") and rated this stimulus family and employment responsibilities.
person. The descriptions varied according to a 2 X 2 X Analyses of subjects' estimates of the num-
2 (employed female vs. employed male X married vs. not
married X children at home vs. no children at home) ber of hours per day the employed persons
factorial design. Two control conditions were provided: a spent doing household and family-related work
female for whom neither marital nor parental status was revealed only one significant effect: Persons
provided and a male for whom neither status was provided. with children were judged to do more work
than persons without children (Ms = 4.91 ver-
Manipulation of Independent Variables sus 3.34, respectively), F(l, 233) = 6.96, p <
Sex of stimulus person. The stimulus persons were .01. Subjects failed to acknowledge that mar-
described as an average employed woman or an average ried female employees spend more time on
employed man. household work than married male employees,
Marital status of stimulus person. The stimulus persons
whose marital status was provided were described as mar- even though this sex difference has been well
ried or as not married. For the remaining stimulus persons, documented in sociological literature on
no information about marital status was provided. housework (e.g., Hartmann, 1981).
Analyses of subjects' salary estimates rep- nation, subjects in a fifth experiment rated the
licated the effect of sex obtained in our pre- personal attributes of a female or male em-
vious experiments: Women were judged to ployee described as employed by choice or
have lower salaries than men (Ms = $ 18,074 employed out of necessity. Other subjects rated
versus $25,055, respectively), F(l, 233) = a female or male employee about whom no
38.29, p<. 001. choice information was provided.
Beliefs About Stereotypic Attributes Experiment 5

Method
Subjects' mean ratings of communion and
agency appear in Table 4. On communion, Subjects
the main effects of marital and parental sta- A total of 110 females and 132 males participated. One
tuses were both significant. Married employees female and one male experimenter each randomly selected
half of the subjects by choosing persons seated in a coffee
were judged as being more communal than shop or general library at Purdue University. Subjects'
single employees (Ms = 3.54 versus 3.26, re- mean age was 20.59 years.
spectively), F(l, 233) = 9.60, p < .01. Em-
ployees with children at home were judged as Procedure
more communal than employees without chil-
dren at home (Ms = 3.68 versus 3.11, respec- ployee Each subject read a brief description of a full-time em-
9 [e.g., "an average woman who is employed full-
tively), F(\, 233) = 41.75, p < .001. time and who is employed because she wants to work (and
On agency, only the sex effect proved sig- not because she has to work)"] and rated this stimulus
nificant: Replicating the three previous ex- person. The descriptions varied according to a 2 X 3 (female
periments, female employees were perceived vs. male X employed by choice vs. employed out of ne-
as more agentic than male employees (Ms = cessity vs. no choice information) factorial design.
3.75 versus 3.59, respectively), F(\, 233) =
7.27, p < .01, calculated on the basis of all Manipulation of Independent Variables
10 conditions.10 Thus, contrary to the double- Sex of stimulus person. The stimulus persons were
burden hypothesis, neither marital nor paren- described
employed man.
as an average employed woman or an average
tal status influenced the agency ascribed to Choice of stimulus person to be employed. The stimulus
employed women or men. persons were described as employed because they want to
In view of this nonconfirmation of the dou- work and not because they have to work or employed
ble-burden explanation of women's perceived because they have to work and not because they want to
work, or no information about employment choice was
agency and in view of the continued reliability provided.
of the finding itself, we considered yet another
explanation: Perhaps employed women are Measuring Instruments
considered more agentic than their male Subjects' beliefs about Stereotypic attributes and job
counterparts because they are more likely to status were assessed with the measures described in Ex-
have chosen to be employed. Because agency
evidently typifies employees (perhaps because
most jobs require and reward agentic behav- 9
ior), observing that women often are employed Consistent with a Sex of Stimulus Person X Parental
Status of Stimulus Person interaction, F(\, 233) = 5.86,
by choice may have led perceivers to infer that p < .05, male employees with children at home were per-
such women are agentic. This explanation of ceived as considerably more communal than male em-
the effects of choice on stereotype formation ployees without children at home (Ms - 3.76 versus 2.97,
follows from correspondent inference theory respectively, p < .001), whereas female employees with
(Jones & Davis, 1965), which suggests that children at home were not perceived as so greatly different
(M = 3.59) from female employees without children at
perceivers more often infer that an employee home (M - 3.25), although this difference remained sig-
possesses the personal qualities she or he man- nificant for the females (p < .01).
10
ifests behaviorally if it appears that the em- For all five role descriptions, the female agency mean
ployee has freely chosen to work and has not was larger than the male mean. Yet this agency sex dif-
ference became quite small for the stimulus persons for
been required to work by virtue of her or his whom marital and parental statuses were not described.
life situation or sex role. However, in this female stimulus-person condition, there
In a test of this freedom-of-choice expla- were two unusually low agency scores.
Table 4
Mean Ratings of Stereotypic Attributes of Female and Male Employees Who Varied in Marital and
Parental Statuses: Experiment 4
Stimulus Married/ Married/no Single/ Single/no No
person children children children children description
Female
Communal 3.77 3.38 3.42 3.11 3.30
Agentic 3.70 3.83 3.91 3.84 3.49
Male
Communal 3.86 3.10 3.66 2.87 3.40
Agentic 3.59 3.54 3.75 3.60 3.45
Note. Means are on a 5-point scale; larger numbers indicate greater communion or agency. Cell ns ranged from 22 to
27. For communal, A/5e = 0.37; for agentic, MS, = 0.24.
periment 1. For the stimulus persons about whom no choice ployees working out of necessity (M = 3.10,
information was provided, subjects used an 11-point scale, ps < .01 or smaller).
ranging from 0% to 100% chance, to rate the likelihood
that the person worked "because she [he] has to and not On agency, the main effects of sex and choice
because she [he] wants to." were both significant, as was the interaction
between these variables. Women were per-
Results ceived as more agentic than men (Ms = 3.54
The principal data analyses were Sex of vs. 3.30, respectively), J^l, 236) = 12.34, p <
Stimulus Person X Choice of Stimulus Person .001. Consistent with the choice main effect,
F(2t 236) = 89.54, p < .001, employees work-
to Be Employed ANOVAS.
ing by choice were perceived as more agentic
than employees without choice information
Inferred Likelihood of Employment Out of (Ms = 3.86 versus 3.59, p < .002, respectively),
Necessity and Inferred Job Status who in turn were more agentic than employees
The average employed woman was judged working out of necessity (M = 2.80, p < .001).
less likely to be employed out of necessity than Consistent with the Sex X Choice interaction,
the average employed man (Ms = 48.75% ver- F(2, 236) = 3.46, p < .05, the tendency for
sus 67.33%, respectively), F(l, 78) = 11.50, women to be perceived as more agentic than
p< .01. Also, women were judged to have men was significant for employees without
lower salaries than men (Ms = $16,330 versus choice information (p < .001) and employees
$23,983, respectively), F(l, 235) = 55.81,;? < working out of necessity (p < .025) and non-
.001. The main effect of the stimulus person's significant for employees working by choice.
choice to be employed, F(2,235) = 17.78, p < Note also that for women, the agency of the
.001, revealed that higher salaries were ascribed employee working by choice did not differ
to employees working by choice (M = from that of the employee without choice in-
$23,265) or to employees without choice information, and these two employees were more
formation (M - $2-1,304) than to employees agentic than the employee working out of ne-
working out of necessity (M=$ 15,977, cessity (ps < .001). For men, the agency of
ps < .001). the employee working by choice was greater
than that of the employee with no choice in-
Beliefs About Stereotypic Attributes formation (p < .001), and these two employees
were more agentic than the employee working
Subjects' mean ratings of communion and out of necessity (ps < .001).
agency appear in Table 5. On communion, These ANOVA findings generally support the
only the main effect of choice proved signif- hypothesis that employed women are perceived
icant, F(2, 236) = 7.41, p < .001: Greater as more agentic than employed men, because
communion was ascribed to employees work- women have been observed to exercise greater
ing by choice (M = 3.43) or employees without choice about being employed. Thus it is
choice information (M = 3.35) than to em- worthwhile to examine, for employees about
Table 5 gardless of their sex, employees working by

Mean Ratings of Stereotypic Attributes of Female choice were perceived as more agentic than
and Male Employees Who Varied in Choice to employees working out of necessity. Further-
Be Employed: Experiment 5 more, perceived agency was similar for the
Employed
women and men about whom choice infor-
Stimulus Employed out of No choice mation was provided. When employed by
person by choice necessity information choice, the woman and the man did not differ
in agency, although when employed out of ne-
Female cessity, the woman was perceived as somewhat
Communal 3.38 3.23 3.41
Agentic 3.86 2.95 3.80
more agentic than the man."
Male In addition, it should be noted that the
Communal 3.45 2.98 3.27 comparisons between stimulus persons of the
Agentic 3.85 2.66 3.38 same sex are relatively consistent with our
Note. Means are on a 5-point scale; larger numbers indicate
freedom-of-choice hypothesis. Because sub-
greater communion or agency. Cell ns ranged from 40 to jects believed that it was moderately likely that
41. For communal, MS, = 0.31; for agentic, MS, = 0.27. the average employed woman had freedom of
choice, her agency should have been perceived
as similar to that of the woman employed by
whom no choice information was provided, choice. As expected, the agency of the average
the correlation between freedom of choice and employed woman did not differ from that of
agency. Overall, the more likely an employee the woman employed by choice, and it was
was to be regarded as working out of necessity greater than that of the woman employed out
rather than choice, the lower was her or his of necessity. Because subjects believed that it
agency, r(78) = -.40, p < .001. However, this was moderately unlikely that the average em-
relation was considerably stronger for the fe- ployed man had freedom of choice, his agency
male employee, r(38) = -.55, p < .001, than should have been perceived as lower than that
the male employee, r(38) = -.09, ns. of the man employed by choice. As expected,
the agency of the average employed man was
Discussion lower than that of the man employed by choice
although higher than that of the man employed
In Experiment 5, the average employed out of necessity.
woman was once again rated as more agentic Even though the communion findings are
than the average employed man, and the find- not relevant to our hypothesis, it should be
ings of this experiment favored the hypothesis noted that lack of choice also lowered the
that this sex difference stems from perceivers' stimulus persons' communal tendency. The
observations that women are more likely than fact that the effect of choice on communion
men to have chosen to be employed. One of was weaker than its effect on agency may ex-
the findings favoring this hypothesis is that the plain why women's greater freedom of choice
average employed woman was judged as more did not cause female employees to be perceived
likely than the average employed man to be as more communal than male employees, as
working by choice rather than necessity. Fur- well as more agentic. Consistent with the weak
thermore, on a correlational basis, the less
likely subjects believed it was that the average
employed woman works out of necessity, the
more agency they ascribed to her. "The meaning of these comparisons is necessarily
somewhat ambiguous because the choice information may
Stronger evidence that sex differences in have had different implications for males than for females.
choice to be employed underlie the greater For example, a man employed by choice may be perceived
perceived instrumentality of female (vs. male) as independently wealthy, whereas a woman employed by
employees is provided by subjects' beliefs choice may be perceived as married and merely in com-
about the agency of employees described as fortable financial circumstances. Similarly, a woman em-
ployed out of necessity may be perceived as single or in
either having or lacking freedom of choice. especially poor financial circumstances, whereas a man
The choice information strongly affected employed out of necessity may be perceived as fairly typical
agency ratings in the expected direction: Re- of all men (as suggested by the findings presented above).
effect of choice on communion, in this series (operationally defined as beliefs that certain
of experiments, nonsignificant tendencies in attributes differentiate women and men; see
the female direction were found in the majority Ashmore & Del Boca, 1981) arise when
of communion comparisons, and in Experi- women and men are observed typically to
ment 2, which had a large number of subjects, carry out different social roles.
the comparison reached significance. Any theory of the content of gender stereo-
As we noted earlier, the relation between types should account for the major perceived
freedom of choice and agency may be attri- gender differences documented in past re-
butionally mediated. Because jobs are thought search—namely, the agentic qualities ascribed
to require agentic behavior, observations that to men and the communal qualities ascribed
members of a particular group are generally to women. These differences should be ac-
employed by choice may favor the correspon- counted for by one or both of the major dif-
dent inference that such people possess agentic ferences in the way women and men are dis-
personality attributes. It is also possible that tributed into social roles—namely, the con-
employed women's greater freedom of choice centration of women in lower positions in
implies that they are more qualified because hierarchies of status and authority, and in the
of selection or self-selection in terms of agentic homemaker rather than the employee occu-
qualities. That is, mainly those women who pational role. In our research, only the home-
are agentic have chosen to worjt or have been maker-employee difference appeared to ac-
specially selected to work (see also Footnote count for the subjects' beliefs that women are
8). An additional explanation stems from the especially communal and men especially
finding that activities perceived to be voluntary agentic. Although status differences did not
rather than required are associated with more account for these beliefs, it would be surprising,
positive affect and higher involvement (Csik- in view of the importance that sociologists have
szentmihalyi & Figurski, 1982). Therefore, the accorded to hierarchy as a fundamental aspect
higher ratings of the agency (and communion) of social roles (e.g., Blau, 1964; Weber, 1947),
of employees who freely chose to work might if observed status differences did not underlie
be a consequence of the positive affect that any beliefs about gender-differentiating traits
has been observed to characterize this major and behaviors. On the contrary, status differ-
portion of such individuals' lives. ences have been shown to account for the belief
that women are more compliant than men
General Discussion (Eagly & Wood, 1982).12
In documenting that occupational roles un-
According to our framework, social struc- derlie belief in female communal qualities and
ture accounts for the content of stereotypes, male agentic qualities, our research has shown
or more exactly, the observed distribution of that beliefs about women resemble those about
groups into various aspects of social structure homemakers and that beliefs about men re-
underlies stereotypes. Although such an ex- semble those about employed persons. In ad-
planation may account for stereotypes about dition, our research has shown that beliefs
other subgroups within societies (e.g., races), about what is typical of homemakers and em-
we have confined our investigation to gender ployees override beliefs about what is typical
stereotypes by hypothesizing that the observed of women and men, whereas the converse does
distribution of women and men into social not occur. Beliefs about what is typical of
roles underlies these stereotypes. In particular,
our experiments investigated whether per-
ceivers' beliefs about women and men stem 12
Indeed, still other aspects of the distribution of men
from their previous observations of women and women in society may affect beliefs about the sexes.
and men in differing statuses within work For example, Kiesler (1975) suggested that because men
hierarchies and in differing occupational roles. outnumber women as successful achievers in many oc-
Because the content of gender stereotypes cupations, perceivers tend to evaluate an individual wom-
arises from perceivers' observations of people's an's achievements less favorably than an individual man's,
even when their products are objectively equal. Perceivers'
activities and these activities are determined judgments of an individual's success may thus reflect the
primarily by social roles, gender stereotypes probability of success for persons of the individual's gender.
women and men do not override beliefs about ness. This information caused the prior prob-
what is typical of homemakers and employ- abilities of women's and men's assertiveness
ees.13 Instead, except for the greater agency of to be revised. Similarly, in our research, even
female employees, women and men were per- extremely general information about a person's
ceived equivalently once their occupational employment status caused subjects to revise
role as homemaker or employee was specified. their estimates of women's and men's com-
The underlying reason that occupational role munal and agentic qualities.
is more important than sex in determining Our findings, however, highlight the incom-
beliefs about communal and agentic charac- pleteness of the Bayesian analysis, which does
teristics is that stereotypes concerning these not explain why stereotypes have certain con-
qualities have become associated with the sexes tent, even though it does model a process by
mainly because sex has been observed to co- which past behavior or other characteristics
vary with occupational role. that are believed to predict future behavior
Although the finding that female employees override information about sex. To account
are perceived as more agentic than their male parsimoniously for the content of gender ste-
counterparts was initially serendipitous, it is reotypes, an investigator must find a behavioral
compatible with our social-structural per- or personal attribute that is diagnostic of the
spective if it is interpreted in terms of perceived particular set of attributes believed to char-
choice. The perception that employed women acterize womeji and men—namely, the strong
are likely to be employed by choice arises not communal and weak agentic tendencies as-
only from perceivers' observations that women cribed to women and the weak communal and
(and not men) are homemakers but also from strong agentic tendencies ascribed to men. To
their belief that women's primary obligation explain these beliefs, an attribute must (a) dif-
is to the domestic role, a belief documented ferentiate the sexes and (b) relate to perceived
by public opinion research (e.g., Mason, 1973; agency and communion in opposite directions.
Mason, Czajka, & Arber, 1976). Because tra- As we have shown, these criteria are fulfilled
ditionally the role of homemaker did not in- by the occupational roles of homemaker and
clude any obligation to seek employment out- employee. Yet we have shown that one's po-
side the home, employment tends to be re- sition in hierarchies of status and authority
garded as an optional or freely chosen aspect does not meet these requirements.
of married women's lives. Therefore, even the Although most social cognition research has
perception of employed women as more agen- little power to account for the content of ste-
tic than employed men reflects previous ob- reotypes, research on "illusory correlation"
servations of women's and men's different dis- (e.g., Hamilton, 1981; Hamilton & Gifford,
tributions into (and obligations in relation to) 1976) pertains to content. It has shown that
domestic and paid employment roles. minority group members may be perceived to
have characteristics with low probability of
Relation to Research on Cognitive occurrence merely because the group members
Bases of Stereotyping and such characteristics are both rare and
therefore distinctive. In our research women
Our research has implications for several constitute minorities in some social roles (e.g.,
aspects of stereotyping that have been ex- as high-status employees), as do men in other
amined recently in the research literature on social roles (e.g., as homemakers). Neverthe-
social cognition. One highly relevant analysis
has treated sex stereotypes as prior probabil-
ities in a Bayesian model (Locksley, Borgida,
Brekke, & Hepburn, 1980; Locksley, Hepburn, 13
Deaux and Lewis (in press) have also reported that
& Ortiz, 1982). Locksley and her colleagues role information is more important than sex in determining
found perceived sex differences in a target per- beliefs about women and men, although their role de-
scriptions included more than occupational cues (e.g., the
son's assertiveness only when subjects did not masculine role was described as "head of the household,
have diagnostic, individuating information financial provider, a leader, and responsible for household
about the target's previous levels of assertive- repairs").
less, despite their rarity, such women and men information because of the origin of these ste-
appear to be assigned primarily the attributes reotypes in observed variation in role assign-
that correspond to the roles they occupy. ments. Gender stereotypes are not applied
Our research also relates to the issue of when typical role arrangements are invalidated
whether there is a basic level of categorization by these cues.
that people commonly use in representing
other people and ascribing attributes to them Conclusion
(Brewer, Dull, & Lai, 1981; Cantor & Mischel,
1979; Taylor, 1981). In particular, categori- To turn from issues of cognitive mediation
zations at this basic level are hypothesized to toward more general issues addressed in the
maximize the richness, differentiation, and stereotyping literature, we note that our anal-
vividness of subjects' perceptions of people. ysis bears on the classic "kernel-of-truth"
Using such criteria, Cantor and Mischel (1979) question pertaining to the validity of stereo-
concluded that categorization at the level of types. In assuming that stereotypes validly
social roles or "persona" is basic in person represent the social structure and division of
perception. It is notable that the research re- labor in a society, our approach is consistent
ported in the present article and our earlier with that of other investigators who have im-
research on beliefs about social influence plicated these factors as the kernel of truth of
(Eagly & Wood, J 982) point to the importance ethnic stereotypes (Brewer & Campbell, 1976)
of categorizations at the level of social roles and gender stereotypes (Williams & Best,
(i.e., occupation and hierarchical status), which 1982). As such, our approach deviates from
are highly diagnostic of people's traits and psychologists' more typical assumptions that
abilities. One's sex can be considered less basic stereotypes stem from psychological factors
than roles because it functions as a cue that, such as perceivers' distortions (e.g., Katz &
due to its previous association with social roles, Braly, 1933; Lippmann, 1922) and biases in-
provides indirect access to occupation and hi- herent in perceivers' cognitive processing (e.g.,
erarchical status. Hamilton, 1979). However, the kernel-of-truth
Finally, our framework is not inconsistent question is considerably broader than our so-
with the idea popular in the social cognition cial structural hypotheses, and as Brigham
literature that stereotypes function as proto- (1971), Campbell (1967), and others have
types (e.g., Cantor & Mischel, 1979) or sche- noted, there is ample reason to believe that
mata (e.g., Taylor & Crocker, 1981). Our claim stereotypes both represent and distort reality.
that observations of social roles underlie gender Our research, by relating stereotypes to per-
stereotypes is not meant to imply that the ceived distributions into social roles, provides
stimulus person's social role is ordinarily re- a promising avenue for addressing the kernel-
trieved for perceivers to infer her or his attri- of-truth question.
butes. Instead, perceivers' prototype or schema This kernel-of-truth issue raises the question
of a typical woman or man is retrieved, and of whether the various attributes associated
judgments are made in terms of the personal with social roles reflect ingrained personality
attributes already associated with it. Because traits and .abilities characteristic of the typical
people's activities are determined primarily by occupants of the roles. In relation to home-
their social roles, the prototypes or schemata maker and employee roles, for example, one
that perceivers possess of woman and man may wonder whether girls are socialized to ac-
consist largely of attributes that covary with quire communal traits and boys to acquire
the role assignments of women and men. agentic traits, with the result that women as
Therefore, our analysis primarily elaborates a group are more suited to perceivers' concept
the process by which stereotypes have acquired of the homemaker role and men as a group
particular content rather than the processes to perceivers' concept of the employee role.
that occur when stereotypes are used to make Although we believe that socializing agents
judgments. Yet our analysis includes the im- tend to prepare girls and boys for the social
portant point that the use of gender stereotypes roles that they believe these girls and boys will
is under the control of cues that carry role probably occupy as adults (Eagly, 1983), our
theory of stereotypes does' not address these Another implication of our idea that social
issues of how people are prepared for roles structure underlies beliefs about gender is that
and recruited into them. According to our change in these beliefs probably must await
framework, the proximal cause of gender ste- social change. Our theory and findings suggest
reotypes is the differing distributions of women that gender stereotypes—the beliefs that
and men into social roles, whatever the cau- women in general differ from men in general—
sation that lies behind these differing distri- will not disappear until people divide social
butions. roles equally, that is, until child care and
We also note that our findings have some household responsibilities are shared equally
implications for the issue of gender equality. by women and men and the responsibility to
The gender or "sex role" stereotypes that psy- be employed outside the home is borne equally.
chologists have made famous and enshrined Interventions designed to affect ideas about
in their theories about gender and the methods gender through education and exposure to the
of gender research appear to be eradicated and media (e.g., ensuring that textbooks have non-
even partly reversed by information as general sexist portrayals of women and men) would,
and basic as the fact that the persons being of course, have some impact in terms of our
judged are employed. A literal interpretation theory because beliefs about the distribution
of our findings suggests that employed women, of people into social roles derive from indirect
who constitute about half of American women sources such as textbook portrayals as well as
(U.S. Department of Labor, 1980), are prob- from direct experience. Yet daily life provides
ably not ordinarily perceived in terms of the abundant direct experience with women and
female stereotype of low agency and high men. Therefore, efforts to remove gender ste-
communion, because it is usually quite salient reotypes educationally may have relatively little
to an employed woman's friends, associates, impact, compared with actual changes in the
and family members that she is employed. Yet distribution of the sexes into social roles.
it is no doubt incorrect to think that employed
women escape the traditional female stereo- References
type altogether, because we do not fully un-
derstand the conditions under which people Abramson, P. R., Goldberg, P. A., Greenberg, J. H., &
apply occupational stereotypes in preference Abramson, L. M. (1977). The talking platypus phe-
to gender stereotypes. It is very likely that there nomenon: Competency ratings as a function of sex and
professional status. Psychology of Women Quarterly, 2,
are contexts in which an employed woman is 114-124.
perceived traditionally, even by people who Ashmore, R. D., & Del Boca, F. K. (1981). Conceptual
know that she is employed. approaches to stereotypes and stereotyping. In D. Ham-
It is also important to note that the stereo- ilton (Ed.), Cognitive processes in stereotyping and in-
tergroup behavior (pp. 1-35). Hillsdale, NJ: Erlbaum.
types we obtained of employees do not portray Bakan, D. (1966). The duality of human existence: An
women and men equally. Instead, these ste- essay on psychology and religion. Chicago: Rand
reotypes favor the formerly excluded group: McNally.
Employed women are perceived as more agen- Bern, S. L. (1974). The measurement of psychological
tic than employed men (even though women androgyny. Journal of Consulting and Clinical Psy-
chology, 42, 155-162.
are believed to have lower wages). Yet it should Bishop, Y. M. M., Fienberg, S. E., & Holland, P. W. (1975).
be kept in mind that the source of the extra Discrete multivariate analysis: Theory and practice.
agency appears to be the belief that women Cambridge, MA: MIT Press.
have freely chosen to be employed. Free choice Blau, P. (1964). Exchange and power in social life. New
York: Wiley.
is, of course, absent for many women, a fact Block, J. H. (1973). Conceptions of sex roles: Some cross-
acknowledged by our respondents in their es- cultural and longitudinal perspectives. American Psy-
timates of the likelihood that employed women chologist, 28, 512-526.
have freedom of choice. Enhanced agency Blood, R. O., Jr., & Wolfe, D. M. (1960). Husbands and
would be ascribed mainly to relatively affluent, wives: The dynamics of married living. Glencoe, IL:
Free Press.
married women, because they could most Brewer, M. B. (1979). In-group bias in the minimal in-
plausibly be regarded as having freely chosen tergroup situation: A cognitive-motivational analysis.
to enter the paid work force. Psychological Bulletin, 86, 307-324.
Brewer, M. B., & Campbell, D. T. (1976). Ethnocentrism Hamilton, D. L., & Gifford, R. K. (1976). Illusory cor-
and intergroup attitudes: East African evidence. New relation in interpersonal perception: A cognitive basis
York: Sage. of stereotypic judgments. Journal of Experimental Social
Brewer, M. B., Dull, V., & Lui, L. (1981). Perceptions of Psychology, 12, 392-407.
the elderly: Stereotypes as prototypes. Journal of Per- Hartmann, H. I. (1981). The family as the locus of gender,
sonality and Social Psychology, 41, 656-670. class, and political struggle: The example of housework.
Brigham, J. (1971). Ethnic stereotypes. Psychological Bul- Signs, 6, 366-394.
letin, 76, 15-38. Jones, E. E., & Davis, K. E. (1965). From acts to dispo-
Broverman, I. K., Vogel, S. R., Broverman, D. M., Clark- sitions: The attribution process in person perception.
son, F. E., & Rosenkrantz, P. S. (1972). Sex-role ste- Advances in Experimental Social Psychology, 2, 219-
reotypes: A current appraisal. Journal of Social issues, 266.
28, 59-78. Kanter, R. M. (1977). Men and women of the corporation.
Brown, L. K. (1979). Women and business management. New York: Basic Books.
Signs, 5, 266-288. Katz, D., & Braly, K. W. (1933). Racial stereotypes of
Campbell, D. T. (1967). Stereotypes and the perception one hundred college students. Journal of Abnormal and
of group differences. American Psychologist, 22, 817- Social Psychology, 28, 280-290.
829. Kiesler, S. B. (1975). Actuarial prejudice toward women
Cantor, K, & Mischel, W. (1979). Prototypes in person and its implications. Journal of Applied Social Psy-
perception. Advances in Experimental Social Psychology, chology, 5, 201-216.
12, 3-52. Lippmann, W. (1922). Public opinion. New \ork: Harcourt,
Clifton, A. K., McGrath, D., & Wick, B. (1976). Stereo- Brace.
types of women: A single category? Sex Roles, 2, 135- Locksley, A., Borgida, E., Brekke, N., & Hepburn, C.
148. (1980). Sex stereotypes and social judgment. Journal
Cohen, J. (1977). Statistical power analysis for the be- of Personality and Social Psychology, 39, 821-831.
havioral sciences (2nd ed.). New \brk: Academic Press. Locksley, A., Hepburn, C., & Ortiz, V. (1982). Social ste-
Cronbach, L. J. (1951). Coefficient alpha and the internal reotypes and judgments of individuals: An instance of
structure of tests. Psychometrika, 16, 297-334. the base-rate fallacy. Journal of Experimental Social
Csikszentmihalyi, M., & Figurski, T. J. (1982). Self-aware- Psychology, 18, 23-42.
ness and aversive experience in everyday life. Journal Manis, M. (1971). Context effects in communication: De-
of Personality, 50, 15-28. terminants of verbal output and referential decoding.
Davis, J. A. (1974). Hierarchical models for significance In M. H. Appley (Ed.), Adaptation-level theory (pp. 237-
tests in multivariate contingency tables: An exegesis of 255). New York: Academic Press.
Goodman's recent papers. In H. L. Costner (Ed.), So- Mason, K. Q (1973). Studying change in sex-role defi-
ciological methodology, 1973-74 (pp. 189-231). San nitions via attitude data. Proceedings of American Sta-
Francisco: Jossey Bass. tistical Association, Social Statistics Section, 138-141.
Deaux, K., & Lewis, L. L. (in press). The structure of Mason, K. Q, Czajka, J. L., & Arber, S. (1976). Change
gender stereotypes: Interrelationships among compo- in U.S. women's sex-role attitudes, 1964-1974. American
nents and gender label. Journal of Personality and Social Sociological Review, 41, 573-596.
Psychology. Mennerick, L. A. (1975). Organizational structuring of
Eagly, A. H. (1983). Gender and social influence: A social sex roles in a non-stereotyped industry. Administrative
psychological analysis. American Psychologist, 38, 971- Science Quarterly, 20, 570-586.
981. Park, B., & Rothbart, M. (1982). Perception of out-group
Eagly, A. H., & Wood, W. (1982). Inferred sex differences homogeneity and levels of social categorization: Memory
in status as a determinant of gender stereotypes about for the subordinate attributes in in-group and out-group
social influence. Journal of Personality and Social Psy- members. Journal of Personality and Social Psychology,
chology, 43, 915-928. 42, 1051-1068.
England, P. (1979). Women and occupational prestige: A Parsons, T. (1955). Family structure and socialization of
case of vacuous sex equality. Signs, 5, 252-265. the child. In T. Parsons & R. F. Bales (Eds.), Family:
Feldman, J. M. (1972). Stimulus characteristics and subject Socialization and interaction process (pp. 35-131).
prejudice as determinants of stereotype attribution. Glencoe, IL: Free Press.
Journal of Personality and Social Psychology, 21, 333- Scanzoni, J. (1982). Sexual bargaining: Power politics in
340. the American marriage (2nd ed.). Chicago: University
Gillespie, D. L. (1971). Who has the power? The marital of Chicago Press,
struggle. Journal of Marriage and the Family, 33, 445- Smedley, J. W, & Bayton, J. A. (1978). Evaluative race-
458. class stereotypes by race and perceived class of subjects.
Hamilton, D. L. (1979). Acognitive-attributional analysis Journal of Personality and Social Psychology, 36, 530-
of stereotyping. Advances in Experimental Social Psy- 535.
chology, 12, 53-84. Spence, J. T., & Helmreich, R. L. (1978). Masculinity &
Hamilton, D. L. (1981). Illusory correlation as a basis for femininity: Their psychological dimensions, correlates,
stereotyping. In D. Hamilton (Ed.), Cognitive processes & antecedents. Austin: University of Texas Press.
in stereotyping and intergroup behavior (pp. 115-144). Tajfel, H. (1981). Human groups and social categories.
Hillsdale, NJ: Erlbaum. New York: Cambridge University Press.
Taylor, S. E. (1981). A categorization approach to stereo- Upshaw, H. S. (1969). The personal reference scale: An
typing. In D. Hamilton (Ed.), Cognitive processes in approach to social judgment. Advances in Experimental
stereotyping and intergroup behavior (pp. 83-114). Social Psychology, 4, 315-371.
Hillsdale, NJ: Erlbaum. U.S. Department of Labor, Bureau of Labor Statistics.
Taylor, S. E., & Crocker, J. (1981). Schematic bases of (1980). Perspectives on working women: A databook
social information processing. In E. T. Higgins, C. P. (Bulletin 2080). Washington, DC: Government Printing
Herman, & M P. Zanna (Eds.), Social cognition: The Office.
Ontario symposium (Vol. 1, pp. 89-134). Hillsdale, NJ: Weber, M. (1947). The theory of social and economic or-
Erlbaum. ganization. Glencoe, IL: Free Press.
Taylor, S. E., Fiske, S. T., Etcoff, N. L., & Ruderman, Williams, J. E., & Best, D. L. (1982). Measuring sex ste-
A. J. (1978). Categorical and contextual bases of person reotypes: A thirty-nation study. Beverly Hills, CA: Sage.
memory and stereotyping. Journal of Personality and Zuckerman, M. (1979). Attribution of success and failure
Social Psychology, 36, 778-793. revisited, or: The motivational bias is alive and well in
Taynor, J., & Deaux, K.. (1973). When women are more attribution theory. Journal of Personality, 47, 245-287.
deserving than men: Equity, attribution, and perceived
sex differences. Journal of Personality and Social Psy-
chology, 28, 360-367.
Triandis, H. C. (1977). Interpersonal behavior. Monterey, Received March 28, 1983
CA: Brooks/Cole. Revision received November 9, 1983
New Journal on Aging: Call for Nominations
By action of APA's Publications and Communications Board and Council of Rep-

resentatives, the APA is publishing a quarterly journal called Psychology and Aging,
the first issue of which will appear in 1985. Psychology and Aging will contain original
articles on adult development and aging. The articles may be reports of research or
applications of research, and they may be biobehavioral, psychosocial, educational,
methodological, clinical, applied, or experimental (laboratory, field, or naturalistic)
studies. For more information about the new journal, see the November 1983 issue
of the APA Monitor.
Nominations for the editor of Psychology and Aging are now open. Candidates
must be members of APA and should be available to start receiving manuscripts in
mid-1984 to prepare for issues published in 1985. To nominate candidates, prepare
a statement of one page or less in support of each candidate. Submit nominations
no later than April 2, 1984, to:
Martha A. Storandt
Department of Psychology
Box 1125
Washington University
St. Louis, Missouri 63130
(314) 889-6508

Eagly 1984

Uploaded by

Copyright:

Available Formats

Eagly 1984

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Eagly 1984

Uploaded by

Copyright:

Available Formats

Journal or Personality and Social Psychology Copyright 1084 bv the

1984, Vol. 46, No. 4, 735-754 i Psychological Association, Inc.

Gender Stereotypes Stem From the Distribution of

in turn were judged to earn more than persons Table 1

Stimulus High Low No

Subjects' mean ratings of the stimulus per- Female

Beliefs About Stereotypic Attributes planned contrasts, employees (regardless of

Beliefs About Stereotypic Attributes Experiment 5

Table 5 gardless of their sex, employees working by

New Journal on Aging: Call for Nominations

By action of APA's Publications and Communications Board and Council of Rep-

You might also like