ArticlePDF Available

Excuse Me - What Did You Just Say?!: Women's Public and Private Responses to Sexist Remarks


Abstract and Figures

Two studies illustrate women's struggle between their desire to challenge sexism and the social pressures and costs that lead to not publicly responding. In Study 1, 45% of the 108 women confronted a man who made a sexist remark and only 15% did so directly. Confronting was most likely to be chosen by women actively committed to fighting sexism in their daily lives. Private responses illustrate that a lack of responding was not necessarily indicative of complacency about the remarks or a lack of thoughts about confronting. The results from Studies 1 and 2 reveal that diffusion of responsibility, normative pressures to not respond, social pressures to be polite, and concern about retaliation likely suppressed responding. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Content may be subject to copyright.
Excuse Me—What Did You Just Say?!: Women’s Public
and Private Responses to Sexist Remarks
Janet K. Swim and Lauri L. Hyers
Pennsylvania State University
Received August 14, 1997; revised July 20, 1998; accepted October 1, 1998
Two studies illustrate women’s struggle between their desire to challenge sexism and the
social pressures and costs that lead to not publicly responding. In Study 1, 45% of the
women confronted a man who made a sexist remark and only 15% did so directly.
Confronting was most likely to be chosen by women actively committed to fighting sexism
in their daily lives. Private responses illustrate that a lack of responding was not necessarily
indicative of complacency about the remarks or a lack of thoughts about confronting. The
results from Studies 1 and 2 reveal that diffusion of responsibility, normative pressures to
not respond, social pressures to be polite, and concern about retaliation likely suppressed
responding. r1999 Academic Press
As the focus of psychological research on prejudice has widened to include the
target’s perspective, we can no longer view those on the receiving end of
prejudice and discrimination simply as passive victims (Lalonde & Cameron,
1994). Theoretical examinations have described targets as strategic negotiators of
threatening situations (Goffman, 1963), as partners in a dynamic interaction
(Deaux & Major, 1987), and as stress managers who actively cope through
externally and internally focused responses (Fitzgerald, Swan, & Fischer, 1995).
Support for this study came, in part, from an internal grant from the College of Liberal Arts,
Pennsylvania State University. We thank Laurie Cohen for her comments on an earlier version of this
paper and the following individuals for their help in conducting the studies and coding materials:
Robert Alt, JenniferAynardi, Carolann Conforti, John Dimakopoulos, Stephen Dolan, Susan Dvorkin,
Candice Graef, Michell Hall, Cindy Horn, Susan Horvath, Gregory Kelch, Amy Klein, Lisa Levinson,
Debbie Miller, Carla McCulty, Kathleen Murphy, Dania Pazakis, Chris Price, and Kari Shaeffer.
Portions of the research were presented at the Joint European Association for Experimental Social
Psychology and Society for Experimental Social Psychology conference in Washington, DC and at the
Empire State Social Psychology conference in New York and the American Psychological Society
conference in Washington, DC.
Correspondence concerning this article and reprint requests should be addressed to Janet K. Swim,
Department of Psychology, 515 Moore Building, The Pennsylvania State University, University Park,
PA16801. Electronic mail may be sent via Internet to JKS4@PSU.EDU.
Journal of Experimental Social Psychology 35, 68–88 (1999)
Article ID jesp.1998.1370, available online at on
0022-1031/99 $30.00
Copyright r1999 by Academic Press
All rights of reproduction in any form reserved.
jesp 1370
@xyserv2/disk3/CLS_jrnl/GRP_jesp/JOB_jesp35-1/DIV_325a03 cind
No. of Pages—21 First page no.—68 Last page no.—88
These approaches emphasize that targets employ a variety of cognitive as well as
behavioral coping strategies to withstand the injustices that they face (Crocker &
Major, 1989; Feagin, 1991; Fitzgerald et al., 1995; Hyers & Swim, 1998; Lalonde
& Cameron, 1994; Wright, Taylor, & Moghaddam, 1990). Understanding how
targets respond to prejudice is important in order to provide a fuller picture of the
way that targets negotiate social interactions in an era when encounters with
prejudice are still common (Swim, Cohen, & Hyers, 1998). In the present
research, we wished to examine the often underestimated role of targets as active
rather than passive recipients of prejudice, without neglecting to account for their
more personal, private reactions.
Public Responses
In this paper, we focus on how women confront encounters with sexism.
Confronting styles may vary from target group to target group because the nature
of prejudice, its history, and expression are idiosyncratic to different stigmatized
groups (Fiske & Stevens, 1993; Young-Bruehl, 1996). It is important to study the
particular aspects of women’s confronting styles for a number of reasons.
Encountering sexism is commonplace for women and making decisions about
whether and how to respond is a part of their everyday lives (Swim et al., 1998).
Personal benefits of responding include altering or reducing this form of daily
hassle and the self-satisfaction of acting on one’s beliefs rather than being
overpowered by a prejudiced individual (Crosby, 1993). Confronting can also aid
in improving the condition of women in general by altering specific perpetrator’s
beliefs, altering bystanders’ perceptions of events, or altering social norms as to
what is considered appropriate behavior (Blanchard, Crandall, Brigham, &
Vaugh, 1994; Lalonde & Cameron, 1994). Yet, women may not respond because
of social influence pressures to not respond, social pressures against identifying
oneself as a feminist, fears of retaliation, or fears of being perceived as impolite or
overly aggressive. These latter concerns might be especially heightened for
women because confronting could be seen as inconsistent with the female gender
role (Jack, 1991; Swim, Ferguson, & Hyers, 1999).
There has been very little, if any, high-impact, nonretrospective empirical work
on response styles women make to encounters with prejudice. Therefore, in our
first study, we placed women in a group discussion where they heard a male
confederate make several sexist remarks, which allowed us to observe women’s
immediate responses to a specific encounter, rather than relying on long-term,
retrospective recall. The sexist remarks made by our male confederate openly
endorsed traditional gender roles (viewing women as sex objects and as respon-
sible for domestic chores). The expression of prejudice directed toward women as
a group, rather than directed at a specific woman, is pervasive and typical of
women’s everyday experiences with sexism (Fitzgerald et al., 1988; Swim et al.,
1998). We also included a condition where women heard parallel nonsexist
jesp 1370
@xyserv2/disk3/CLS_jrnl/GRP_jesp/JOB_jesp35-1/DIV_325a03 cind
comments in order to obtain baseline information about women’s public and
private responses to the nonsexist component of these remarks.
Predicting Public Responding
Understanding public responding includes knowledge about factors that might
influence one’s decision to respond. Responding to events involves labeling an
event as prejudicial, being motivated to respond to the event, and then deciding to
act on that motivation. There are a number of factors that can influence labeling an
event as prejudiced (e.g., Crosby, 1984; Feldman Barrett & Swim, 1998; Rug-
gierio & Taylor, 1997) and greater certainty that an incident is prejudicial can
influence the likelihood and style of responding (e.g., Wright et al., 1990).
However, labeling an event as prejudicial may not be sufficient to result in
confronting. The motivation to respond is likely a function of how offensive the
prejudicial incident is perceived to be (Kowalski, 1996). The decision to act on
one’s motivation likely involves assessing possible costs and benefits of publicly
responding (Kowalski, 1996).
Endorsement of gender-related beliefs should be related to perceptions of
events, motivation to respond, and decisions to respond. More specifically,
identification with women as a group, such as holding gender central to the
self-concept, feeling a sense of common fate with other women, and being
concerned about the well-being of women in general (Branscombe, Owen, &
Kobrynowicz, 1993), may increase the likelihood that a woman will label an
incident as sexist (Johnston, Swim, & Stangor, 1998) and take action in response
to the incident (Sauders, 1992, as cited by Fitzgerald et al., 1995). Women who
have traditional attitudes about gender roles and hold modern sexist attitudes,
such as believing that sexism is no longer a problem in society, may also be less
likely to label an event as sexist or to publicly respond to an incident (Brooks &
Perot, 1991; Jensen & Gutek, 1982; Mazer & Percival, 1989; Swim & Cohen,
1996). Finally, an active commitment to fighting sexism may be required before a
woman feels the responsibility or courage to confront sexism (Crosby, 1993).
A factor that has been overlooked when considering whether people confront
offensive behavior is the effect of social influence. Observations of others’
behaviors after an offensive incident might influence one’s interpretation of the
extent to which an event is sexist and offensive.Also, group norms and diffusion
of responsibility are likely to influence one’s perceptions of the appropriateness of
responding, motivation to respond, and assessment of the costs and benefits of
responding. In the present study, the effect of social influence was examined by
testing the impact of gender composition on women’s public verbal responses to
sexist remarks. Informational and normative social influence processes should be
more powerful when bystanders are more similar to the target than when they are
less similar (Abrams, Wetherell, Cochrane, Hogg, & Turner, 1990). Thus, if
bystanders do not respond (as in the present study), then women’s responses
would be suppressed more when the nonresponding bystanders are women rather
than men. Research on helping in emergencies suggests similar outcomes (Latane
jesp 1370
@xyserv2/disk3/CLS_jrnl/GRP_jesp/JOB_jesp35-1/DIV_325a03 cind
& Darley, 1970). The presence of bystanders creates diffusion of responsibility
and lack of responding. People may feel that women are more responsible than
men for confronting sexism directed at women, perhaps because women should
be most qualified to decide whether the remarks need to be confronted or because
they have more to gain by curbing sexism directed at women. Thus, women may
believe that fellow bystanders share more responsibility for responding when the
bystanders are women than when they are men.
In contrast, expectations about women’s versus men’s reactions to confronta-
tions suggests that public confrontation may be more likely to occur when other
women are present. A woman may anticipate social support from women and
hostility from men if she publicly responds because of an expectation that other
women would be more likely than men to share her perceptions. Thus, the cost of
responding would be smaller and the likelihood of publicly responding would be
greater when a woman is with other women than when she is the only woman in a
Private Responses
Choosing not to respond does not mean a lack of private responses (Kowalski,
1996). The types of private responses to prejudice described by previous research-
ers primarily represent different ways of coping with prejudice (e.g., denial of
prejudice or making intragroup rather than intergroup comparisons, Fitzgerald et
al., 1995; Tajfel & Turner, 1979). However, private responses can also be
characterized by immediate thoughts and feelings about the offensiveness of the
incident and whether to confront. In the present study we examine the content of
women’s private evaluations of the sexist remarks they encounter and the person
making the remark and their thoughts about how they would like to respond to
these remarks.
A final purpose of the first study was to examine how encountering and
confronting prejudice might affect state self-esteem. Self-blame is a common
response to sexual harassment (Fitzgerald & Ormerod, 1993) and it can be
associated with negative feelings about the self. Similarly, women may internalize
the disempowering messages of sexist remarks (Crocker, Voelkl, Testa, & Major,
1991; Dion, 1975; Dion & Earn, 1975). However, there are ways to buffer oneself
from the negative impact of prejudice (Feldman Barrett & Swim, 1998). Women
who hold gender-related attitudes that would predispose them to identifying the
remarks as sexist may be more likely to make self-protective attributions (Crocker
& Major, 1989; Crocker et al., 1991; Landrine & Klonoff, 1997). Further, if a
woman confronts a sexist person, the confronting may be self-affirming and
self-esteem enhancing.
In sum, in Study 1 we examined public and private responses women make
when encountering sexism directed at women. We were particularly interested in
assessing the frequency and style of women’s public and private responses that
jesp 1370
@xyserv2/disk3/CLS_jrnl/GRP_jesp/JOB_jesp35-1/DIV_325a03 cind
communicated displeasure about the sexist remark. Although we predicted that
women would be more likely to publicly respond to sexist than parallel nonsexist
responses, we predicted that a substantial number of women would not publicly
respond to the sexist condition in contrast to the substantial number of women
who would report unfavorable private thoughts and feelings about the sexist
comment and person. We also predicted that those who were more gender
identified, had more feminist-related beliefs, and reported being actively commit-
ted to fighting sexism would be more likely to publicly confront sexist comments.
We did not make directional predictions about the effect of gender composition
due to the social influence and diffusion of responsibility explanation yielding one
prediction and the social support explanation providing an opposite prediction.
Further, we predicted more confrontations when our female participants were the
solo member of their gender in the group than when their gender was in the
majority. Finally, we predicted that public confrontations, gender identification,
nonsexist beliefs, and having an activist orientation would buffer women against
the possible negative effects of sexist comments on self-esteem.
Participants were 108 women recruited by phone from a group of students who completed a
departmental prescreening questionnaire in their Introductory Psychology class. We excluded 9
participants who expressed suspicion during debriefing that the confederates were not naı¨ve partici-
pants. In addition, thought and feeling listings and data about verbal confrontations were not available
from 4 participants in the sexist condition due to technical difficulties with the video camera.
The design was a 2 (gender composition: solo vs. nonsolo woman) 2 (type of remark: sexist vs.
nonsexist) between-subjects factorial. Participants were randomly assigned to conditions. In addition,
the impact of several gender-related measures on the dependent measures was assessed.
Gender-Related Beliefs
As part of a departmental prescreening of Introductory Psychology classes, students completed
several scales relevant to the present study. These included Spence, Helmreich, and Stapp’s (1973)
shortened version of the Attitudes Toward Women Scale and Swim,Aikin, Hall, and Hunter’s (1995)
Modern Sexism Scale. The Attitudes Toward Women Scale measures participants’ attitudes about
gender roles and gender equality (Cronbach’s ␣⫽.83). The Modern Sexism Scale measures beliefs
that discrimination against women is not a problem, women complain too much about sexism, and
women have gotten too much special treatment (Cronbach’s ␣⫽.80), and can be considered a
measure of sensitivity to sexism (Swim & Cohen, 1996). Responses to items on these scales were on 0
(strongly agree) to 6 (strongly disagree) scales. Higher scale scores indicate more unfavorable
attitudes toward women’s issues on both scales.
Participants also completed the activism portion of O’Neil, Egan, Owen, and Murry’s (1993)
Gender Role Journey Scale and Branscombe et al.’s (1993) Gender Identity Scale. The activism
subscale of the Gender Role Journey Scale included items such as ‘‘I have taken some actions in my
personal life to reduce sexism’’and ‘‘I am responsible for changing restrictive gender roles.’’ Higher
scores indicate greater activism (Cronbach’s ␣⫽.91). We used four subscales from the Gender
Identity Scale. Higher scores on these subscales indicate that respondents (1) hold biases in favor of
their own gender group (Cronbach’s ␣⫽.86), (2) have a strong emotional attachment to their own
gender group (Cronbach’s ␣⫽.80), (3) feel they are a typical member of their own gender group
jesp 1370
@xyserv2/disk3/CLS_jrnl/GRP_jesp/JOB_jesp35-1/DIV_325a03 cind
(Cronbach’s ␣⫽.84), and (4) feel they share a common fate with members of their own gender group
(Cronbach’s ␣⫽.82).
Participants were recruited to be in a study ostensibly concerning group decision making. The
experimenter escorted one participant and the three confederates to a group discussion room where
they were asked to sit in preassigned seats. In the solo female participant condition, there were three
male confederates. In the nonsolo female participant condition, there were two female confederates
and one male confederate. The experimenter explained that the group was to select 12 individuals,
from a list of 15 women and 15 men with different occupational titles, who would be best suited for
survival on a deserted island. The experimenter asked group members to take turns in a clockwise
order indicating their suggestion and a reason for their suggestion and asked one of the confederates to
record the group’s selections. This procedure allowed the participant to take part in the group
discussion while the confederates said scripted comments. The script included comments about other
confederates’ remarks. However, the confederates were instructed to keep extraneous conversation to
a minimum and always to agree with the participant’s selections.
The male confederate sitting to the participant’s right made either three sexist or three parallel
nonsexist statements at different points during the group discussion. First, in response to another
confederate’s selection of an athlete/trainer, the confederate to the right of the participant said ‘‘Yeah,
we definitely need to keep the (women/people in shape).’’Second, during his own turn, he said ‘‘Let
me see, maybe a chef? No, one of the (women/others) can cook.’’ Third, during his final turn, he
selected a female musician and said ‘‘I think we need more (women on the island to keep the men
satisfied/entertainers to keep everyone happy).’’The confederate made these comments to the group as
a whole and did not look directly at the participant when he said them. The other two confederates
were instructed not to respond to the sexist remarks or to any comments the participant might make
about the remarks. After the selection of the 12 individuals was completed, one of the confederates
informed the experimenter that the group had finished the task. The experimenter then led the
participant and ostensibly the confederates to private rooms to complete a questionnaire.
When the participant completed the questionnaire, the experimenter led her to another room where
she was informed that the group had been videotaped during the group task. In accord with Ickes,
Robertson, Tooke, andTeng’s (1986) naturalistic social cognition procedure, the participant was asked
to view the videotape of her group task in order to help her recall the thoughts and feelings she had
during the session. She was instructed to stop the tape whenever she remembered having a thought or
feeling during the session and was asked to write down the thought or feeling and the time it occurred
from a clock appearing on a videotape display.After she finished viewing the videotape, she was asked
to rate her thoughts and feelings on several dimensions, as described below. She was then given the
final debriefing and thanked for her participation in the study.
Participants rated their impressions of each of the confederates.1Using 7-point scales, they rated the
extent to which each person was cooperative, friendly, prejudiced, flexible, involved, active, respon-
sible, expressive, comfortable, and supportive. Of particular interest were ratings of the confederates’
level of prejudice. This rating was used as a single item scale with higher numbers indicating more
prejudice. Based on factor analyses, three scales were formed from the remaining attributes, with
higher scores representing greater perceived cooperativeness (cooperative, flexible, friendly, support-
ive; Cronbach’s ␣⫽.84), participation (involved, active, expressive, and comforting; Cronbach’s
1Participants also completed four questions designed to assess their perceptions of social support in
the group. Participants reported feeling less support from the group when a sexist remark was made.
This is likely a function of perceptions of the sexist confederate rather than all members of the group
because type of remark did not affect other ratings of the bystanders. Group composition did not affect
the social support ratings.
jesp 1370
@xyserv2/disk3/CLS_jrnl/GRP_jesp/JOB_jesp35-1/DIV_325a03 cind
␣⫽.75), and responsibility (measured by one item, ‘‘responsibility’’). After these ratings, participants
rated the extent to which they would like to work with the person in the future, they would like the
person as a friend, and they perceived that the person had attitudes similar to their own.
Participants’ feelings about themselves during the task were assessed with a modified version of
Heatherton and Polivy’s (1991) state self-esteem scale. We modified the scale so that it referred to
participants’ feelings during the group task rather than during the completion of the scale itself. The
questions asked how they felt about their performance (e.g., I felt confident about my abilities), how
they felt socially, (e.g., I felt concerned about the impression I was making), and how they felt about
their appearance (e.g., I was pleased with my appearance). Responses to these items were recoded
such that higher numbers indicated higher state self-esteem on the performance (Cronbach’s ␣⫽.84),
social (Cronbach’s ␣⫽.88), and appearance (Cronbach’s ␣⫽.93) subscales.
Coding of Thought and Feeling Listings
Participants wrote an average of 12 thoughts and feelings. After writing their recalled thoughts and
feelings, participants were asked to number them and rate each thought and feeling as to whether it
was (a) a thought or a feeling, (b) related to the group task, (c) negative or positive on a scale from 1 to
7, and (d) written about themselves personally, each of the confederates, or something or someone
else.2We selected and then aggregated the ratings of those thoughts and feelings that participants had
identified as being directed at the confederate sitting to their right (the confederate saying either the
sexist or the nonsexist remark). There were no notable predictors of the thoughts and feelings directed
at others or the participants themselves.
After the study was completed, two female coders independently rated the thoughts and feelings.
Coders rated whether each thought or feeling mentioned one of the three sexist or parallel nonsexist
remarks or the person saying the remarks (␬⫽.88). The thoughts and feelings in this subset were rated
as to whether they included any mention of ways to confront the sexist comment or person (␬⫽.55).
Private confrontations were defined as comments participants would have liked to have made or
actions they would have liked to have done during the group task that would express their
disagreement or displeasure with the remark.
Coding of the Videotapes
Transcribers were instructed to list all comments made by the participant immediately after each
sexist remark until the next scripted comment was made. Then, two female coders independently
noted any confrontations, defined as verbal expressions of displeasure or disagreement with the sexist
remark. The correlation between the number of confrontational responses each coder noted during
each session was 0.98.
After reconciling discrepancies as to whether the response was confrontational, the responses were
coded to capture their specific style of delivery, details not accounted for in most broad conceptual
classifications (e.g., Lalonde & Cameron, 1994; Tajfel &Turner, 1979; Wright et al., 1990). Because
coders could select more than one style for each response, interrater agreement was assessed
separately for each style. The response styles and corresponding reliabilities were direct confrontation
(e.g., saying that the person or remark was sexist or telling the perpetrator to change his behavior;
␬⫽.77), humor or sarcasm (␬⫽.73), questioning of the confederate (␬⫽.78), giving a task-related
response that contradicted the confederates remark (␬⫽.86), surprised exclamations (␬⫽.77),
grumbling noise (␬⫽.66), and resigned acceptance (␬⫽.88). Coders reconciled any discrepancies
through discussion. The 10% of responses that indicated resigned acceptance (e.g., ‘‘Whatever’’) were
excluded from analyses of confrontational responses because they did not clearly indicate disagree-
2Each participant also rated the extent to which the thought took a direct-perspective (a perception
of herself, other people, the task, or the room) or a meta-perspective (a perception of other people’s
perception of her, other people, the task, or the room). There was some confusion for participants as to
what a meta-perspective was and few participants selected this option, so we did not examine this
jesp 1370
@xyserv2/disk3/CLS_jrnl/GRP_jesp/JOB_jesp35-1/DIV_325a03 cind
ment. It could be argued that several of the other behaviors (such as surprised exclamations or
grumbling) represent immediate emotional reactions to the event rather than confrontational reactions.
However, these remarks can be considered confrontational because they were verbal indicators of
disagreement or displeasure and were controllable.
Manipulation Check
A 2 (gender composition) 2 (type of remark)ANOVA was performed on the
questionnaire ratings. This analysis revealed a main effect for type of remark on
all but one rating. When the confederate made sexist remarks, he was perceived as
more prejudiced, less responsible, and less cooperative than when he made
nonsexist remarks (see Table 1). Participants also indicated a lesser preference to
work with the confederate making the sexist remarks in a future group, indicated a
lesser preference to have him as a friend, and perceived that his attitudes were
more dissimilar from their own attitudes compared to when he made nonsexist
remarks. The only rating of this confederate that was not affected by the type of
remark was his level of participation in the group. There were also no main effects
or interactions with gender composition.
Public Responses
Of the 44 women in the sexist condition, 4 confronted all three sexist remarks, 5
confronted two of the remarks, 11 confronted one remark, and 24 made no
confrontations. Thus, 45% (n20) gave at least one verbal confrontational
response in the sexist condition. In contrast, 7 of the 51 women in the nonsexist
condition expressed disagreement or displeasure with the parallel nonsexist
remarks. Because the parallel nonsexist remarks merely expressed a preference
for certain occupations and were not designed to be offensive and few responded
Rating Sexist remark
remark (M)F(1, 95) p
Prejudiced 5.06 (2.06) 1.86 (1.41) 83.57 .001
Responsible 5.12 (1.68) 6.22 (1.01) 15.52 .001
Cooperative 5.28 (1.28) 6.12 (.84) 14.86 .001
Participation 5.21 (1.08) 5.31 (1.31) .21 .70
Work with in the future 3.50 (1.98) 5.04 (1.48) 18.17 .001
Be their friend 3.37 (1.78) 4.53 (1.41) 12.10 .001
Perceive similarity 2.67 (1.78) 4.26 (1.44) 24.52 .001
Note. Numbers in parentheses indicate standard deviations. Higher means represent perceiving the
confederate as more prejudiced, responsible, cooperative, participatory, having more favorable
reactions to working with the person in the future, having more favorable reactions to being their
friends, and perceiving more similarity between themselves and this person. There were 48 women in
the sexist condition and 51 women in the nonsexist condition.
jesp 1370
@xyserv2/disk3/CLS_jrnl/GRP_jesp/JOB_jesp35-1/DIV_325a03 cind
in the nonsexist condition, the confrontation-like responses in the nonsexist
condition are used only for basic comparison purposes and will not analyzed
The most frequent style of confrontation made in the sexist condition was
questioning the confederate (n11, 25%). These questions included asking the
confederate to repeat himself and asking the confederate rhetorical questions
(e.g., ‘‘What did you say?!’’). The second most frequent style of response was a
task-related response (n9, 20%) (e.g., by selecting an occupation or providing
an explanation that contradicted the suggestion made by the sexist person). Next,
the following three styles of responses were equally likely to occur (n7, 16%,
for each): direct comments (e.g., ‘‘You can’t pick someone for that reason. Pick
another person’’), humor or sarcasm (e.g., making an anti-male, sexist comment),
or surprised exclamations (e.g., ‘‘Oh my God. I can’t believe you said that!’’). The
least frequent confrontational response was grumbling (n1, 2%).
Predicting Public Responses3
Gender composition of the group. We used log-linear regressions to test
whether gender composition of the group predicted publicly confronting the
sexist remarks at least once during the entire task as well as at least once to each of
the three sexist remarks. We were interested in responses to each remark because
each remark might be perceived differently at different points during the group
task. For instance, there is a possible cumulative impact of hearing several
remarks and observing others not responding to the remarks, as our confederates
were instructed to behave.
While not significantly different, women were 14% more likely to confront at
least once when they were the only member of their gender in the group (n12,
52%) than when there were other women present (n8, 38%; 2(1) .88,
p.35). Examining the number of women who gave at least one style of
response after each remark revealed that gender composition had a statistically
significant impact for the first remark (2(1) 4.24, p.04), but not for the later
remarks. Participants were more likely to confront after the first remark when they
were the solo woman present (n8, 35%) than if there were other women
present (n2, 9%). There was no statistically significant effect of gender
composition on responding to the second remark (n7, 31% versus n5, 22%,
3We wanted to know whether perceptions of the sexist remarks would mediate the relationship
between activism and confronting. However, we did not have a measure of perceptions of the sexist
remark because we did not want to reveal the purpose of the study by asking the question. As a proxy
for participants’ perceptions of the sexist remark, we used participants’ratings of how prejudiced the
sexist individual was perceived to be. The results indicate that these perceptions did not mediate the
results. A possible reason this variable did not mediate the relationship is that being more likely to
perceive the person as sexist, in contrast to perceiving the remark as sexist, may decrease confronting
because confronting might be perceived as less effective and more risky with a more sexist person.
Consistent with this explanation, log-linear regressions revealed that perceiving that the confederate
was more prejudiced lead to being less likely to confront after the first remark (r⫽⫺.23, 2(1) 4.37,
jesp 1370
@xyserv2/disk3/CLS_jrnl/GRP_jesp/JOB_jesp35-1/DIV_325a03 cind
solo woman versus nonsolo woman, respectively), nor to the third remark (n6,
27% versus n5, 20%, solo woman versus nonsolo woman).
Gender-related beliefs. For those in the sexist remark condition, log-linear
regressions were conducted to test whether any of the gender-related beliefs
predicted publicly confronting the sexist remarks at least once during the study or
to each remark. Separate log-linear regressions were computed for each gender-
related belief. Women who reported actively confronting sexism in their lives
were more likely to confront after the second (r.17, 2(1) 3.77, p.05) and
third remarks (r.25, 2(1) 5.75, p.02). None of the other gender-related
beliefs predicted confronting.
Private Responses
Participants’ ratings of the thoughts and feelings directed at the person sitting to
their right were analyzed with 2 (group composition) 2 (type of remark)
ANOVAs. A higher percentage of the thoughts and feelings were about the
confederate sitting to their right when he made the sexist remarks (M35.73,
SD 21) than when he made the parallel nonsexist remarks (M14.56,
SD 14.38), F(1, 94) 34.84, p.001. With regard to the thoughts and
feelings listed about the person sitting to their right, participants were more likely
to list fewer thoughts, more feelings, fewer task-related thoughts and feelings, and
more non-task-related thoughts and feelings in the sexist than the nonsexist
remark condition (see Table 2). Further, the valence of these thoughts and feelings
was more negative in the sexist (M2.22, SD 1.28) than the nonsexist
condition (M4.02, SD 1.74), F(1, 72) 26.44, p.001. There were no
main effects or interactions involving gender composition.
We were particularly interested in whether women’s thought and feeling
listings included ways to confront the sexist remarks. In the sexist condition, 43%
(n19) of the participants mentioned at least one confrontational response. Of
the 44 women in this condition, 1 reported four confrontational comments, 1
listed three confrontational comments, 5 listed two confrontational comments, 12
listed one confrontational comment, and 25 listed none. The confrontational
comments included responses they made during the study, responses they would
like to have made (e.g., arguing with him, name calling, scolding, leaving, and
questioning the confederate), and responses they were not likely to have seriously
considered (e.g., hitting or killing him). The two most common responses listed
were arguing and taking some type of violent action like hitting the confederate.
In contrast, only 3 of the 51 women in the nonsexist condition reported ways to
confront the parallel nonsexist comment or person saying the comment. The
confrontational responses they listed were simply thoughts about not wanting to
openly disagree or argue with the confederates. Listing at least one confrontation
in the thought and feeling listings was not related to whether participants made
similar public responses during the study (2(1) 1.62, p.20).
jesp 1370
@xyserv2/disk3/CLS_jrnl/GRP_jesp/JOB_jesp35-1/DIV_325a03 cind
We predicted that hearing the sexist remark might affect how participants felt
about themselves during the task in terms of their state self-esteem and this might
be moderated by whether they confronted. Two (type of remark) 2 (group
composition) ANOVAs revealed that the type of remark and gender composition
of the group did not affect performance, appearance, or social state self-esteem.
Contrary to predictions that confronting would be self-affirming, a 2 (confronted
at least once vs. never) 2 (group composition) ANOVA for those in the sexist
condition revealed no significant effects on state self-esteem.
We also predicted that gender-related beliefs would moderate the effect of
hearing sexist remarks on self-esteem. Moderated regression analyses supported
this prediction. First, a gender-related belief measure (kept as a continuous
measure) and the type of remark made (dummy coded) were entered together in
the regression equations. Second, we entered the interaction between the indi-
vidual difference measure and the type of remark. We ran separate regressions for
each of the gender-related beliefs and for each of the types of state self-esteem.
These regressions revealed an interaction between level of activism and type of
sexist remark for performance (b.16, t(86) 2.62, p.01), appearance
(b.32, t(86) 2.63, p.01), and social (b.28, t(86) 2.49, p.01) state
self-esteem. When a sexist remark was made, less activist women reported that
they had lower performance, (r.48, p.01), appearance, (r.45, p.01),
and social (r.46, p.01) state self-esteem during the group session. When a
nonsexist remark was made, level of activism was not related to performance,
(r.01, p.93), appearance, (r⫽⫺.07, p.64), and social (r.02, p.91)
state self-esteem. Similar results predicting social state self-esteem were found for
the Attitudes Toward Women Scale and the emotional attachment to one’s own
gender group subscale of the Gender Identity scale (interaction b⫽⫺.43,
t⫽⫺2.75, p.01; interaction b⫽⫺.23, t2.16, p.03, respectively). When
Rating Mean sexist
remark Mean nonsexist
remark F(1,72) p
Thought 58.99 (40.03) 82.16 (34.77) 7.81 .01
Feeling 41.01 (40.03) 17.84 (34.77) 7.81 .01
Related to task 48.98 (40.79) 79.28 (35.30) 11.63 .001
Not related to task 51.02 (40.79) 20.72 (35.30) 11.63 .001
Note. Numbers in parentheses indicate standard deviations. Higher means indicate a higher average
proportion of thoughts and feelings that the participants rated as having each characteristic. There
were 43 women in the sexist condition and 31 in the nonsexist condition. The means excluded people
who did not identify any of their own thoughts and feelings as being directed toward the confederate
who made the sexist or nonsexist remark.
jesp 1370
@xyserv2/disk3/CLS_jrnl/GRP_jesp/JOB_jesp35-1/DIV_325a03 cind
a sexist remark was made, those with more traditional gender-role attitudes and
less emotional attachment to their gender group had lower social state self-esteem
(r⫽⫺.36, p.02, r.38, p.02, respectively), whereas when a nonsexist
remark was made, these attitudes did not predict their social state self-esteem
(r.18, p.22; r.05, p.69, respectively).
Study 1 results revealed that only 16% of the women confronted the sexist
person with direct verbal comments such as indicating that his remarks were
inappropriate or that he should retract them. However, when including additional
confronting styles, 45% of participants openly expressed their displeasure. In
comparison to the near lack of confronting in the nonsexist condition, this reveals
that women did not behave as passive victims to this encounter with sexism.
An examination of the private thoughts and feelings among the remaining 55%
who did not publicly confront reveals that their lack of confrontation was not
indicative of acceptance of the remark. First, many women privately mentioned
confronting and these thoughts were unrelated to having actually confronted.
Second, perceptions of the sexist confederate indicated that most participants did
not think favorably about the confederate making the sexist remarks and thought
that he was prejudiced. More specifically, three-quarters of those who did not
publicly respond indicated that they saw him as prejudiced in their private ratings.
Similarly, 91% of the women who did not confront publicly nonetheless had
private negative thoughts and feelings about the confederate (i.e., the average
rating of their thoughts and feelings were below the midpoint of the favorability
As hypothesized, activism predicted public confrontation. This relationship
was found for publicly confronting the second and third remark. This pattern
suggests that after a certain amount of time had elapsed and after observing others
not publicly responding one’s sense of personal responsibility became a motiva-
tor. None of the other gender-related beliefs predicted confronting, suggesting
that a particularly committed stance toward fighting sexism is more predictive of
confronting then either being gender identified or having certain feminist beliefs.
The importance of an activist orientation in women’s reactions to sexism was also
revealed through the impact of the sexist remarks on women’s state self-esteem.
In the sexist condition, women who had an activist orientation reported higher
performance, appearance, and social state self-esteem than women who did not
have an activist orientation. This effect is not likely a function of activist women
confronting more than nonactivist women because, contrary to predictions,
confrontation was not related to self-esteem.
The effect of gender composition on public confronting indicates that social
influence processes affected women’s choice of whether to respond more than
social support. Women were more likely to confront the first remark when they
were the solo female than when their gender was in the majority and, while not
statistically significant, this trend existed after the other two remarks. The lack of
jesp 1370
@xyserv2/disk3/CLS_jrnl/GRP_jesp/JOB_jesp35-1/DIV_325a03 cind
effect of gender composition on private ratings and thoughts and feelings about
the confederate and his sexist remarks suggest that social influence did not
influence perceptions. Thus, the effect of group composition was not likely a
result of informational influence. Instead, ingroup identification likely caused
women to feel that other women’s lack of response provided a better role model
than the other men’s lack of response. Further, they likely felt less personal
responsibility for confronting when other women were present than when other
men were present.
There were two purposes for conducting Study 2. First, we tested whether
women would be overconfident about the likelihood that they would respond to
sexist comments. As has been long documented, people are often unaware of the
impact of situations on their behaviors (Milgram, 1974). Further, people are often
overconfident of their ability to predict their own and others’ behaviors due, in
part, to their tendency to underestimate the impact of situational forces on
behavior (Ross & Nisbett, 1991). This overconfidence has important implications
for assumptions about and sensitivity to women who do not respond. Demonstrat-
ing the assumptions people might make, one woman noted in her thought–feeling
listings ‘‘The other two women didn’t appear to find anything wrong with these
comments. The two girls, I realized, have a lot to learn if they [don’t] defend
themselves and their gender.’We predicted that women would be more likely to
indicate that they would take the direct action of publicly confronting sexist
remarks than women actually did in Study 1.
A second purpose of Study 2 was to test possible reasons for women’s selection
of responses in Study 1. After one is motivated to respond, the decision of whether
and how to confront is likely a function of the anticipated costs and benefits of
confronting (Kowalski, 1996). Thus, in Study 2, we examined whether the
perceived costs associated with different styles of confronting might explain the
responses participants selected. While there are many different costs associated
with identifying events as sexist (e.g., Crosby, 1984; Feldman Barrett & Swim,
1998; Ruggiero & Taylor, 1997), we focus on two immediate costs which are
particularly associated with publicly confronting sexist remarks in a group
setting. Complaining can be seen as impolite because it is a violation of the social
standard that says ‘‘If you don’t have anything nice to say, don’t say anything at
all’’(Crosby, 1993). This may be particularly true for women for whom assertive
responding can be perceived as inconsistent with their gender role. Thus, women
who confront will be viewed as impolite. Another perceived cost may be possible
retaliation, ranging from mild social rejection to overt aggression. For instance,
one woman noted in her thought–feeling listing, ‘‘Should I yell at a room full of
guys and give them a women’s lib lecture? They would laugh at me.’ Thus, we
predicted that the most frequently selected styles of responses women made in
Study 1 would most likely be those that were perceived to be the least costly (i.e.,
the most polite and the least risky).
jesp 1370
@xyserv2/disk3/CLS_jrnl/GRP_jesp/JOB_jesp35-1/DIV_325a03 cind
As part of a mass screening packet distributed in Introductory Psychology
classes, 113 women responded to the scenario and activism measures described
The study was a 2 (group composition: solo vs. nonsolo woman) 3 (first,
second, or third sexist remark) mixed factorial with repeated measures on the
second factor and random assignment to one of the two group composition
Scenarios and Dependent Measures
The scenarios described the situation that participants experienced in the sexist
remark condition in Study 1. Study 2 participants were instructed to imagine that
they were participating in the deserted island group task along with three men
(solo condition) or along with a mixed-sex group (two women and a man, the
nonsolo condition). The gender composition manipulation was accomplished by
providing a diagram of the seating arrangement indicating the names of the other
group members and the place where the participant was to imagine sitting. The
next page of the scenario described the scripted dialogue from Study 1 and
specifically noted the male confederate’s remark that they needed an athlete/
trainer to keep the women in shape on the island. Participants were first asked to
rate how offensive (0, not very offensive; 6, very offensive) and sexist (0, not very
sexist; 6, very sexist) they perceived this first remark to be. Next they were asked
to indicate whether they definitely would not, probably would not, probably
would, or definitely would give different styles of responses to the remark. The
response styles represented the confrontational responses participants actually
made during Study 1. (See Table 3 for the list of responses.) Violent responses
(hitting and punching) were also included because they were frequently men-
tioned in the thought and feeling listings of Study 1.5After rating the likelihood
that they would give each response, the second remark made by the sexist
confederate was described and the same questions were asked. Then the third
4Because we anticipated that women would not be sensitive to situational cues, we anticipated that
they may not be sensitive to the impact of group composition on responding. However, in contrast to
their overconfidence in estimating their likelihood of responding, women accurately anticipated that
they would be more likely to say they would publicly confront the sexist person in the solo status
condition than the nonsolo status condition. This effect occurred for all three sexist remarks. Women in
Study 2 may have assumed that they would not have needed to make as many responses because other
women would be helping them. Thus, this finding could be indicative of diffusion of responsibility or
expectations of social support.
5Three additional responses were included on the form (resigned acceptance, glaring, and laughing)
as other options participants might choose. These responses were not included as confrontational
responses in Study 1, so they are not used in the analyses presented below.
jesp 1370
@xyserv2/disk3/CLS_jrnl/GRP_jesp/JOB_jesp35-1/DIV_325a03 cind
remark was described and the same questions were asked. Participants perceived
the sexist remarks as offensive (M3.72, SD 1.84) and sexist (M4.40,
SD 1.69).6Following the response ratings after the third remark, participants
rated, from 0 (not at all) to 6 (very), the extent to which each response directly
communicated disagreement or disapproval of the remark made, represented a
6We examined the impact of gender-related beliefs on perceptions of the comments and anticipated
responses. Being gender identified (emotional attachment to one’s gender and perceiving a common
fate with other women, holding less sexist beliefs (Modern Sexism and Attitudes Toward Women), and
being an activist predicted perceiving the remarks as sexist and offensive. However, being an activist
predicted women being more likely to confront. Perceptions of how sexist and offensive the remarks
were perceived to be mediated the effect activism, but not group composition, on the likelihood of
confronting. Finally, because the questionnaire was included in a mass screening packet for all
students, 45 male participants also completed all questions in the survey. Men and women found the
remarks to be equally sexist but men found them to be less offensive. Men were less likely than women
to anticipate that they would definitely confront the sexist comment (50% vs 81%). This effect
remained significant after covarying out perceptions of the offensiveness of the remarks. In contrast to
the women, they were less likely to anticipate confronting when the other three group members were
men (33%) than when the other three group members were composed of two women and one man
(62%). Even though the group composition variable resulted in the condition being all men versus an
equal number of women and men, this is consistent with a diffusion of responsibility explanation, with
people taking on less responsibility when other people of their gender are available to confront the
sexist person. It should be noted that it is not necessary to make comparisons with men in order to
understand that women did not respond as much as they wanted to and that individual differences,
perceived costs, and social influence processes affected women’s responding in the present study.
Percentage who
actually gave
the response
(Study 1)
Percentage who
anticipated definitely
giving a response
(Study 2)
No response
1. Ignore the commenta55 1
2. Wait to see what others do 55 4
Confrontational responses
3. Question the response 25 47
4. Task-related response 20 22
5. Comment on inappropriateness 16 48
6. Sarcasm or humor 16 37
7. Surprise exclamation 16 40
8. Grumbling 2 10
9. Hit or punch 0 8
Gave at least one confrontational response 45 81
aWhether the lack of responses was a result of ignoring the comment and waiting to see what others
do was not differentiated in Study 1.
jesp 1370
@xyserv2/disk3/CLS_jrnl/GRP_jesp/JOB_jesp35-1/DIV_325a03 cind
polite or socially acceptable response, and represented a risky response in terms
of how the sexist person or others would react to that response.
In order to obtain a conservative estimate of people’s anticipated responses to
each remark, the results were based only on the percentage of individuals who
reported that they would definitely give each response (as opposed to those that
probably would, probably would not, or definitely would not give each response).
We calculated the percentage who would definitely give any of the confronta-
tional responses at least once after each remark, the percentage who would
definitely give any of the confrontational responses at least once across all three
remarks, and the percentage who would definitely give each of the seven
confrontational responses at least once across all three remarks. Lack of respond-
ing was calculated by the percentage of women who indicated that they would
definitely not respond to all three remarks. Hence, the percentage of participants
not responding was based upon the percentage of individuals who would ignore
all three remarks or wait to see what others would do for all three remarks. These
calculations were done to make the results from Study 2 comparable to the results
from Study 1.
Results and Discussion
Confidence in Likelihood of Publicly Confronting
Study 2 illustrated that women were optimistic about the likelihood that they
would take the direct approach of responding publicly to the sexist remarks. The
total percentage of individuals predicting that they would definitely give at least
one confrontational response far exceeded the number who gave at least one
confrontational response during Study 1 (see Table 3). With the exception of
task-related responses, the percentage of participants in Study 2 who anticipated
definitely engaging in each of the confrontational behaviors at least once was
greater than the percentage who actually displayed such responses in Study 1.
Further, in direct opposition to the results of Study 1, very few women in Study 2
anticipated that they would not respond at all. Not responding was rated as the
least direct response option (see Table 4). Thus, women anticipated being more
direct than they likely would have been in a real situation.
The Cost of Responding
The difference in preferred style of confronting between Study 1 and Study 2 is
likely a function of avoiding possible costs associated with confronting. Not
responding, which was the least frequent public response women anticipated in
Study 2, but the most frequent public response actually given in Study 1, was
perceived to be the least risky as well as the least direct.
We wanted to examine whether the particular style of public confrontational
responses women in Study 1 actually made and the women in Study 2 anticipated
making were related to the possible costs of responding. We also tested whether
their confidence that they would be direct would be associated with the confronta-
jesp 1370
@xyserv2/disk3/CLS_jrnl/GRP_jesp/JOB_jesp35-1/DIV_325a03 cind
tion style they used or would anticipate using. We correlated the mean ratings of
how risky, polite, and direct each confrontational response was perceived to be,
first with the percentage of women who actually gave each confrontational
response at least once in Study 1 and second with the percentage who anticipated
definitely giving each confrontational response at least once in Study 2. The less
risky, more polite, and the more direct a response was perceived to be the more
women actually gave the response (r(6) ⫽⫺.30, .88, .36, respectively) and the
more women anticipated giving the response (r(6) ⫽⫺.11, .64, .46, respec-
tively). However, the only significant correlation was between how polite the
response was perceived to be and the percentage of women in Study 1 who
actually gave the response (p.01).7
7We also conducted repeated measures analyses to have a more statistically powerful test of the
effect of costs on public responding. The dependent variable in these analyses was the likelihood that
participants either gave a particular style of response in Study 1 or anticipated giving a particular style
of response in Study 2. We excluded hitting or punching from the analyses because there was no
variance on this variable for Study 1 because no participant gave this responses. We used planned
contrasts with weights derived from Study 2 participants’ perceptions of how direct, polite, and risky
the responses were. The results revealed significant linear trends for all the variables such that the
more direct, polite, and risky behaviors were perceived to be more likely to be done and anticipated.
The effect for riskiness was therefore in the opposite direction to that found with the correlations.
Further, higher order effects qualified all analyses with the exception of the analyses for actual
responding with contrasts based upon how polite the behavior was perceived to be. Thus, like the
analyses using correlations, these analyses revealed that politeness was the most parsimonious
predictor of the participants’ choice of response style.
Response Direct
No response
1. Ignore the comment1.81a(1.43) .77a(1.49) 4.31ab (2.00)
2. Wait to see what others do .73a(1.14) .61a(1.10) 4.15b(1.81)
Confrontational responses
3. Question the response 4.27c(1.29) 3.61c(1.45) 4.17b(1.38)
4. Task-related response 4.19c(1.48) 3.41c(1.59) 4.67a(1.43)
5. Comment on inappropriateness 5.67d(.85) 4.53d(1.54) 3.74b(1.96)
6. Sarcasm or humor 2.89b(1.48) 3.20c(1.45) 3.13c(1.54)
7. Surprised exclamation 3.51b(1.38) 3.32c(1.42) 3.45c(1.37)
8. Grumbling 2.12b(1.40) 2.65b(1.38) 2.47d(1.43)
9. Hit or punch 4.28c(2.34) 5.60d(1.23) .28e(.84)
Note. Numbers in parentheses indicate standard deviations. Higher numbers indicate that the
response was rated as more direct, risky, and polite. Means with different superscripts within a column
are significantly different from each other at p.01, with Newman–Kuels posthoc tests. Nranges
from 109 to 113.
jesp 1370
@xyserv2/disk3/CLS_jrnl/GRP_jesp/JOB_jesp35-1/DIV_325a03 cind
These studies illustrate that women are not responding to sexist remarks in the
manner that they would like to respond. Most women in our studies found the
sexist remarks objectionable, perceived the person saying the remark as preju-
diced, and would like to have publicly responded. However, the constraints of the
situation impeded their responding. The effects of social constraints are illustrated
by the response styles women chose and by the effect of social influence
processes. First, women preferred the least risky choice of not responding.
Second, when women did respond, they preferred more polite responses. Third,
women were less likely to publicly confront when other women were present than
when they were a solo member their gender in the group. The effect of group
composition was likely a result of diffusion of responsibility or female bystanders
providing a stronger normative standard about how women should respond than
male bystanders.
Women who are most likely to respond are those with an activist orientation.
While not endorsing traditional roles or modern sexist beliefs would be indicative
of a feminist orientation, a more well developed feminist identity is associated
with taking an active stance to fight sexism (Bargard & Hyde, 1991). Given that
the other gender-related beliefs did not predict public responding in Study 1, the
results indicate that it takes a person who is particularly committed to ending
sexism to overcome the social influence processes constraining behavior.8
It is likely that the process leading to public responding is similar for targets of
prejudice other than women (e.g., African Americans) as well as for bystanders
who are not the targets (e.g., men overhearing sexist comments about women).
Specifically, this process would consist of labeling an event as prejudicial, being
motivated to respond to the event, and then deciding to act on the motivation.
Further, an activist orientation and social influence processes are likely to affect
outcomes of each of the steps in this process. Similarly, group differences may
emerge due to different tendencies to label events as prejudice, motivations to
respond to prejudice, and assessments of costs and benefits. For instance,
compared to women, men may be less likely to label events as sexist (Russo-
Devosa & Swim, 1997), feel less compelled to confront sexism, and be less afraid
of appearing impolite.
Future research should explore the ramifications of confronting. People’s
8The fact that gender-related beliefs predicted perceptions of the sexist remark but only activism
predicted publicly responding (see footnote 5) is consistent with the argument that labeling an event as
prejudicial is not sufficient for understanding public responses to prejudice. It is possible that the
greater conceptual similarity between the activism measure and confronting accounts for its greater
predictive power. However, the temporal pattern of findings from Study 1 suggest that the scales’
ability to identify people more strongly committed to fighting sexism is also a viable explanation for
its greater predictive power. That is, the pattern of findings suggests that participants’assessment of the
situation as one in which others were not taking the responsibility to respond influenced the greater
predictive power of activism as the study progressed.
jesp 1370
@xyserv2/disk3/CLS_jrnl/GRP_jesp/JOB_jesp35-1/DIV_325a03 cind
anticipation of the costs and benefits about responding may not be correct. The
discrepancy between public and private responses suggests that if a woman
dissents from the social pressure not to confront, it is likely that the dissent will
receive support from other women because many women would privately agree
with the dissenter. The dissenter may therefore not be perceived as impolite and
the response might not be as risky as assumed. Further, the benefits of confronting
could be examined in future research. Confronting may serve a social role by
altering other people’s perceptions. Past research has shown that awareness that
others do not tolerate prejudice can decrease a bystander’s prejudice and tolerance
of comments (Blanchard et al., 1994; Citron, Chein, & Harding, 1950). While not
all the verbal confrontations in the present study are likely to have been
strategically planned to influence others’ perceptions, they still may have that
impact. Other perhaps unintended effects of confronting include being a role
model for others to express their displeasure with sexist remarks, altering social
standards for what is acceptable behavior, and educating perpetrators of sexist
remarks. Finally, while it may seem that confronting could make women feel
efficacious and therefore heighten their self-esteem, the results from Study 1
indicated that confronting was unrelated to state self-esteem. Instead of having
this personal effect, confronting may help protect the state self-esteem of women
who may be uncertain as to whether the remarks were sexist by perhaps
increasing the likelihood that they would label such remarks as prejudicial.
In sum, the fact that 45% of women did respond in some confrontational
manner to the remarks reveals that women are not necessarily passive recipients
of prejudicial encounters. This is particularly likely to be true for women
personally committed to fighting sexism. However, the present study also illus-
trated women’s struggle between their desire to challenge sexism and the social
pressures that work against direct responding. Merely labeling a remark as
prejudicial and wanting to respond is not likely to be sufficient to predict
responding. For instance, thinking about confronting did not predict actually
public confronting. Further, concerns about the costs associated with confronting
support the idea that it takes courage to complain in public about sexist behaviors.
The low frequency of direct comments about the remarks and the fact that 55%
did not make public confrontations in contrast to the negativity of their private
opinions and the desires of many to confront indicate that observers will not be
privy to women’s personal opinions about sexism and their views of those who
make sexist remarks. This can lead to pluralistic ignorance and normative
pressures not to respond, sexist people not learning that they are behaving
offensively and are disliked because of this, and, in general, the perpetuation of
sexist comments in the culture. We hope that recognition of pluralistic ignorance
helps alleviate this ignorance and that recognition of the social barriers women
face as targets of prejudice will yield more sympathetic responses to the struggles
that they and others face when deciding whether to take active stances against
sexism in their everyday lives.
jesp 1370
@xyserv2/disk3/CLS_jrnl/GRP_jesp/JOB_jesp35-1/DIV_325a03 cind
Abrams, D., Wetherell, M., Cochrane, S., Hogg, M. A., & Turner, J. C. (1990). Knowing what to think
by knowing who you are: Self-categorization and the nature of norm formation, conformity and
group polarization. British Journal of Social Psychology, 29, 97–119.
Bargard, A., & Hyde, J. S. (1991). Women’s studies: A study of feminist identity development in
women. Psychology of Women Quarterly, 15, 181–201.
Blanchard, F.A., Crandall, C. S., Brigham, J. C., & Vaugh, L. A. (1994). Condemning and Condoning
racism: A social context approach to interracial settings. Journal of Applied Psychology, 79,
Branscombe, N. R., Owen, S., Kobrynowicz, D. (1993). Measuring identification with one’s gender
group in women and men. Paper presented at the American PsychologicalAssociation, Toronto,
Brooks, L., & Perot, A. R. (1991). Reporting sexual harassment: Exploring a predictive model.
Psychology of Women Quarterly, 15, 31–47.
Citron, A. F., Chein, I., & Harding, J. (1950). Anti-minority remarks: Aproblem for action research.
Journal of Abnormal Social Psychology, 45, 99–126.
Crocker, J., & Major, B. (1989). Social stigma and self-esteem: The self-protective properties of
stigma. Psychological Review, 96, 608–630.
Crocker, J., Voelkl, K., Testa, M., & Major, B. (1991). Social stigma: The affective consequences of
attributional ambiguity. Journal of Personality and Social Psychology, 60, 218–228.
Crosby, F. (1984). The denial of personal discrimination. American Behavioral Scientist, 27, 371–386.
Crosby, F. J. (1993). Why complain? Journal of Social Issues, 49, 169–184.
Deaux, K., & Major, B. (1987). Putting gender into context: An interactive model of gender-related
behavior. Psychological Review, 94, 369–389.
Dion, K. L. (1975). Women’s reactions to discrimination from members of the same or opposite sex.
Journal of Research in Personality, 9, 294–306.
Dion, K. L., & Earn, B. M. (1975). The phenomenology of being a target of prejudice. Journal of
Personality and Social Psychology, 32, 944–950.
Feagin, J. R. (1991). The continuing significance of race: Antiblack discrimination in public places.
American Sociological Review, 56, 101–116.
Feldman Barrett, L., & Swim, J. K. (1998). Appraisals of prejudice and discrimination. In J. K. Swim
& C. Stangor (Eds.), Prejudice: The target’s perspective. San Diego:Academic Press.
Fiske, S. T., & Stevens, L. E. (1993). What’s so special about sex? Gender stereotyping and
discrimination. In S. Oskamp & M. Costanzo (Eds.) Gender issues in contemporary society:
Applied social psychology annual (Vol. 9, pp. 173–196). Newbury Park, CA: Sage.
Fitzgerald, L. F., & Ormerod, A. J. (1993). Breaking silence: The sexual harassment of women in
academia and the workplace. In F. Denmark and M. Paludi (Eds.) The psychology of women:
Handbook of issues and theories (pp. 553–581). Westport, CT: Greenwood.
Fitzgerald, L. F., Shullman, S. L., Bailey, N., Richards, M., Swecker, J., Gold, Y., Ormerod, A. J., &
Weitzman, L. (1988). The incidence and dimensions of sexual harassment in academia and the
workplace. Journal of Vocational Behavior, 32, 152–175.
Fitzgerald, L. F., Swan, S., & Fischer, K. (1995). Why didn’t she just report him? The psychological
and legal implications of women’s responses to sexual harassment. Journal of Social Issues, 51,
Goffman, E. (1963). Stigma: Notes on the management of spoiled identity. Englewood Cliffs, NJ:
Prentice Hall.
Heatherton, T. F., & Polivy, J. (1991). Development and validation of a scale for measuring state
self-esteem. Journal of Personality and Social Psychology, 60, 895–910.
Hyers, L. L., & Swim, J. K. (1998). Acomparison of the experiences of dominant and minority group
members during an intergroup encounter. Group processes and intergroup relations.
Ickes, W., Robertson, E., Tooke, W., & Teng, G. (1986). Naturalistic social cognition: Methodology,
assessment, and validation. Journal of Personality and Social Psychology, 51, 66–82.
jesp 1370
@xyserv2/disk3/CLS_jrnl/GRP_jesp/JOB_jesp35-1/DIV_325a03 cind
Jack, D. C. (1991). Silencing the self: Women and depression. Cambridge, MA: Harvard Univ. Press.
Jensen, I., & Gutek, B. A. (1982).Attributions and assignment of responsibility for sexual harassment.
Journal of Social Issues, 38, 121–136.
Johnston, K., Swim, J. K., & Stangor, C. (1998, in preparation). The effects of enhancing gender
identity on the perception of personal gender discrimination.
Kowalski, R. M. (1996). Complaints and complaining: Functions, antecedents, and consequences.
Psychological Bulletin, 119, 176–196.
Lalonde, R. N., & Cameron, J. E. (1994). Behavioral responses to discrimination: Afocus on action. In
M. P. Zanna & J. M. Olson (Eds.). The psychology of prejudice: The Ontario symposium (Vol. 7,
pp. 257–288). Hillsdale, NJ: Erlbaum.
Landrine, H., & Klonoff, E. A. (1997). Discrimination against women: Prevalence, consequences,
remedies. Thousand Oaks, CA: Sage.
Latane, B., & Darley, J. M. (1970). The unresponsive bystander: Why doesn’t he help? New York:
Mazer, D. B., & Percival, E. F. (1989). Ideology or Experience? The relationships among perceptions,
attitudes, and experiences of sexual harassment in university students. Sex Roles, 20, 135–147.
Milgram, S. (1974). Obedience to authority: An experimental view.New York: Harper & Row.
O’Neil, J. M., Egan, J., Owen, S., & Murry, V. M. (1993). The gender role journey measure: Scale
development and psychometric evaluation. Sex Roles, 38, 167–185.
Ross, L., & Nisbett, R. E. (1991). The person and the situation: Perspectives of Social Psychology.
New York: McGraw–Hill.
Ruggiero, K. M., & Taylor, D. M. (1997). Why minority group members perceive or do not perceive
the discrimination that confronts them: The role of self-esteem and perceived control. Journal of
Personality and Social Psychology, 72, 373–389.
Russo-Devosa, Y., & Swim, J. K. (1997). The role of Modern Sexism in perceptions of sexist attitudes
and everyday sexist behaviors. Poster presented at the annual meeting of the Midwestern
Sauders, B. E. (1992). Sexual harassment of women in the workplace: Results from the national
women’s study. Presentation to the 8th Annual NC/SC Labor Law Seminar, Nashville, NC.
Spence, J. T., Helmreich, R., & Stapp, J. (1973).A short version of the Attitudes toward Women Scale
(WAS). Bulletin of the Psychonomic Society, 2, 219–220.
Swim, J. K., Aikin, K. J., Hall, W. S., & Hunter, (1995). Sexism and racism: Old-fashioned and
modern prejudices. Journal of Personality and Social Psychology, 68, 199–214.
Swim, J. K., & Cohen, L. L. (1996). Overt, covert, and subtle sexism: A comparison between the
Attitudes TowardWomen and Modern Sexism scales. Psychology of Women Quarterly.
Swim, J. K., Cohen, L. L., & Hyers, L. L. (1998). Experiencing everyday prejudice and discrimina-
tion. In J. K. Swim & C. Stangor (Eds.) Prejudice: The target’s perspective. San Diego:Academic
Swim, J. K., Ferguson, M. J., & Hyers, L. L. (1999). Avoiding stigma by association: subtle prejudice
against lesbians in the form of social distancing. Basic and Applied Social Psychology.
Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W. G. Austin & S.
Worchel (Eds.), The social psychology of intergroup relations (pp. 33–47). Monterey, CA:
Wright, S. C., Taylor, D. M., & Moghaddam, F. M. (1990). Responding to membership in a
disadvantaged group: From acceptance to collective protest. Journal of Personality and Social
Psychology, 58, 994–1003.
Young-Bruehl, E. (1996). The anatomy of prejudices. Cambridge, MA: Harvard Univ. Press.
jesp 1370
@xyserv2/disk3/CLS_jrnl/GRP_jesp/JOB_jesp35-1/DIV_325a03 cind
... Although people have many options when deciding how to respond after they witness prejudice, targets of prejudice and nontarget bystanders alike generally choose to not confront perpetrators when they have the opportunity (e.g., Dickter & Newton, 2013;Kawakami et al., 2019;Shelton & Stewart, 2004;Swim & Hyers, 1999;Weber & Dickter, 2015;Woodzicka & LaFrance, 2001). ...
... This finding stands in direct opposition to many people's beliefs about how they would behave should they witness prejudice or discrimination (Kawakami et al., 2009;Swim & Hyers, 1999;Woodzicka & LaFrance, 2001) and their personal values of standing up against inappropriate behavior (Rasinski et al., 2013). The Confronting ...
... First, as the present study took place online, our findings may not generalize to realworld behavior. Although intentions are proximal predictors of actions (Ajzen, 1985(Ajzen, , 1987, prospective behavior does not always align with actual confronting behavior (Swim & Hyers, 1999;Woodzicka & LaFrance, 2001). A conceptual replication of the present study in a more naturalistic setting is needed to evaluate the efficacy of antiracism education on actual institutional discrimination confronting behavior. ...
Confrontation is an important mechanism to reduce racial prejudice. Yet, little research has examined White adults’ intended confrontation within the context of discriminatory policies that pose barriers to Black Americans seeking employment. Moreover, less research has investigated reactions to zero-sum and negative-sum anti-Black institutional discrimination. The present study investigated the effects of an antiracism educational exercise on White adults’ intended confrontation of zero-sum and negative-sum institutional discrimination. Participants aged 20–85 years (n = 195; Mage = 54.16) were randomly assigned to one of three conditions: (1) antiracism education with writing reflection, (2) antiracism education alone, or (3) a control condition. Participants then reported how they would respond in situations that described zero-sum (i.e., new hire recruitment exclusively from predominantly White institutions) and negative-sum anti-Black institutional discrimination (i.e., hairstyle discrimination). Participants in the antiracism education alone condition intended to confront more assertively in both scenarios compared to participants in the control condition. Across conditions, participants were more likely to intend to confront, and intended to confront more assertively, in the zero-sum discrimination scenario than the negative-sum discrimination scenario. We conclude that a brief antiracism education exercise may hold promise for increasing White adults’ assertive confrontation of institutional discrimination, but the effects may be contingent on whether a reflective writing task is included and the characteristics of the discriminatory policy.
... Participants who failed the manipulation checks were excluded. Additionally, to obtain insight into the interpretation of the forum comment, participants were asked to rate whether they perceived the commenter as prejudiced, friendly, reasonable and selfish, on a scale from 1 (strongly disagree) to 7 (strongly agree), adapted from Swim and Hyers (1999). ...
... To ensure participants paid attention to the article they were asked two multiple choice questions about the topic of the article, and what it conveyed about women's position relative to men (disadvantaged; advantaged; equal). Additionally, to obtain insight into the interpretation of the forum comment, participants were asked to rate whether they perceived the commenter as prejudiced, friendly, reasonable and selfish, on a scale from 1 (strongly disagree) to 7 (strongly agree), adapted from Swim and Hyers (1999). ...
Information about this project can be found on the OSF through the following link: This work examines strategic factors that impact women’s intention to express anger. Research suggests that women express anger to a lesser extent than they experience it (Hyers, 2007; Swim et al., 2010), and we focus on the role of gender stereotypes in this phenomenon. We differentiate two “routes” by which gender stereotypes can lead women to avoid expressions of anger. First, in the stereotype disconfirmation route, women become motivated to avoid expressing anger because it supposedly disconfirms stereotypical prescriptions for women to be kind and caring. We also identify a stereotype confirmation route, in which women avoid anger expressions because anger confirms the stereotype that women are overly emotional. Across three experimental studies (Nstudy1 = 558, Nstudy2 = 694, Nstudy3 = 489), we show that women experienced anger about gender inequality, but were relatively reluctant to express the anger they felt. That is, there was evidence for an “Anger Gap.” Feminists in particular showed a large Anger Gap when it was suggested that anger might confirm stereotypes. This work demonstrates that stereotype information introduces strategic concerns that women must take into account when deciding whether to express anger about gender inequality. Additionally, this work highlights that the notion that anger confirms a stereotype can be as powerful in discouraging anger expressions as the idea (identified in previous work) that anger may disconfirm stereotypes.
... Even if they do support them, men sometimes believe they lack the standing to get involved (Sherf et al., 2017). Finally, people often feel reluctant to confront sexist conduct because they fear retaliation (Swim & Hyers, 1999;Good et al., 2012;Kaiser & Miller, 2001. ...
To address sexism, people must first recognize it. In this research, we identified a barrier that makes sexism hard to recognize: rudeness toward men. We found that observers judge a sexist perpetrator as less sexist if he is rude toward men. This occurs because rudeness toward men creates the illusion of gender blindness. We documented this phenomenon in five preregistered studies consisting of online adult participants and adult students from professional schools (total N = 4,663). These attributions are problematic because sexism and rudeness are not mutually exclusive. Men who hold sexist beliefs about women can be—and often are—rude toward other men. These attributions also discourage observers from holding perpetrators accountable for gender bias. Thus, rudeness toward men gives sexist perpetrators plausible deniability. It protects them and prevents the first perceptual step necessary to address sexism.
... (1) agent oppression, where the violence is exercised by a person and (2) structural oppression, in which the institution is the one that acts as the oppressor. The distinction between the two forms is interesting, as oppression is not usually considered at an individual level, but Haslanger advocates same tone, studies have demonstrated that women seem pressured to be polite (Swim and Hyers 1999) and that they fear being labelled feminists (Becker 2007). Thus, I believe that an explanation for the subservience of women in the face of oppression is the fear of being humiliated. ...
The present article will tackle the concept of internalized misogyny by trying to review existing theories and to extract a number of common threads of these theories in order to find some useful insights on the internal mechanisms that make up internalized misogyny, and on how internalized misogyny should be approached by practical action. I start the discussion by exploring oppression and the internalization of oppression, and afterwards move to internalized misogyny itself, charting its place within gender dynamics in general, as well as its impact on gender roles, on women’s actions towards other women, and their actions towards themselves. Using data from the World Value Survey (2017–2020), I will explore how internalized misogyny is reflected in specific sexist attitudes, how it relates to male misogyny, and which aspects of gender relations seem to come to the fore when dealing with internalized sexism. This will allow us to confront and complement the theories on internalized sexism with data on attitudes and beliefs, and develop a clearer picture of the phenomenon, as well as drawing some brief conclusions regarding practical action to mitigate gender oppression.
... If women perceive receiving middle author positions as a norm for collaborative projects, this could explain why they did not cite more cases of dissatisfaction in authorship than men. Second, minorities may worry about being penalized for calling out sexism and racism [33,38,77], making them less likely to mention these efects. ...
Conference Paper
Women remain underrepresented in many fields in computer science, particularly at higher levels. In academia, success and promotion are influenced by a researcher's publication record. In many fields, including computer science, multi-author papers are the norm. Evidence from other fields shows that author order norms can influence the assignment of credit. We conduct interviews of students and faculty in human-computer interaction (HCI) and machine learning (ML) to determine factors related to assignment of author order in collaborative publication. The outcomes of these interviews then informed metrics of interest for a bibliometric analysis of gender and collaboration in research papers published from 1996 to 2016 in three top HCI and ML conferences. Based on our findings, we make recommendations for assignment of credit in multi-author papers and interpretation of author order, particularly in regard to how this area affects women.
Self-confrontation, whereby people become aware of their responses being more biased than their personal standards condone, triggers self-regulation and bias reduction. However, impediments to self-confrontation reduces its occurrence. Other-confrontation, where someone points other another person's biased responses with disapproval, provides an antidote. Research has identified confrontees' reactions and associated moderators, but in a largely descriptive manner. We propose a theoretical framework capturing consequences of other-confrontation for confrontees. The confrontee's perceived validity of the confrontation determines whether they evaluate their response based on their personal standards, which prompts negative self-directed affect and bias reduction. Simultaneously, the confrontee's perception that the confronter is trying to impugn their egalitarian and non-prejudiced image triggers negative other-directed affect and, in turn, the confrontee's generation of social costs (e.g., dislike for the confronter). Moderators affecting bias reduction and social costs operate through their influence on people's answers to the perceived validity and impugnment questions.
Our research centers the underexamined perspective of Black Americans regarding allyship behaviors and investigates their perceptions and experience of a White ally who confronts a White perpetrator of prejudice. In two experimental studies (N = 1176), we found that Black participants reported higher levels of self-esteem after a White ally confronted a White perpetrator of racial prejudice compared to no confrontation. Additionally, we found that White ally confrontations that signaled intrinsic motivations (i.e., driven by personal values) were perceived as less suspicious in motive than those that signaled extrinsic motivations (i.e., driven by image concerns), which related to more target self-esteem. We discuss implications for research on allyship and confrontation as well as practical considerations. Our results strongly suggest that advantaged group members cannot allow overt prejudice to stand unchecked and should consider the motivations they convey in their actions.
Brands are increasingly embracing social activism and adopting positions on controversial issues, prompting some consumers to react by making uncivil comments on social media. How should brands reply to such incivility while maintaining their positions and protecting their reputations? Two common types of reply include either a sarcastic or an assertive tone, but the effects of these types of communication on consumers’ attitudes toward brands remain largely unexplored. Results from a series of five studies exploring different causes (LGBT+ phobia, sexism, and racial equity) show that consumers evaluate brands that reply using an assertive tone more favorably than those using a sarcastic tone, which can be partially explained by the perceived aggressiveness of sarcasm. Additionally, support for a brand's stance acts as a boundary condition on the effect the type of reply adopted by the brand has on consumer attitudes toward the brand. So, the more someone supports a brand's stance, the less their perception of aggressiveness will negatively influence their attitude to that brand. We discuss the implications of these findings for marketing theory and practice.
Sexual harassment and other forms of gendered discrimination are social psychological phenomena, yet the psychological impact of sexual harassment has rarely been examined through a model which considers the role of diverse content of gender identity (i.e. norms). We used an experimental design to investigate how salient norms associated with the social identity of ‘women’ affect coping with sexual harassment. Participants who identified as women ( N = 291) were randomly assigned to either a feminist, traditional feminine, or control norm condition, in which the salience of particular norms associated with womanhood was manipulated. Participants completed measures of personal growth (as a proxy for post-traumatic growth), and help-seeking intentions in response to a hypothetical sexual harassment scenario. Participants in the feminist condition reported significantly greater personal growth relative to those in the traditional feminine and control conditions. Participants in both the feminist and traditional feminine conditions reported significantly greater intentions to seek help from formal supports (e.g. primary care doctor), relative to those in the control condition. The findings suggest that the salience of social identities and their content may be valuable resources in promoting recovery following experiences of gendered discrimination and support the role of social identities in influencing post-trauma trajectories.
Although the effects of confronting sexism have been examined in the past, no research to date has demonstrated the most common dimensions along which confrontations of workplace sexism differ. Furthermore, previous research has not fully examined how different forms of sexism confrontations differentially impact the workplace experiences of targets experiencing these sexist actions. Thus, these two studies examine how confrontations of workplace sexism commonly differ and the workplace implications of these differences. In Study 1, data were collected from individuals who had witnessed a sexist encounter in the workplace that was subsequently confronted. Using a content analysis approach, we found that the most common differences in confrontations of workplace sexism included the identity of the confronter, the tone of the confrontation, the location of the confrontation, the number of confronters, and the timing of the confrontation. In Studies 2 and 3, we examined how these differences impacted job stress, turnover intentions, perceived organizational support, job satisfaction, and workplace diversity climate of women targets. In Study 2, analysis of survey data from women employees determined that workplace confrontations of workplace sexism of any form are equally beneficial for women targets. Study 3, an experimental vignette study, provided support for the Study 2 findings. This study aids researchers in understanding the dimensionality of sexism confrontations as well as the relationships between these dimensions and important workplace outcomes. This study will also prove useful for practitioners intent on educating allies and targets on when and how they should confront sexism at work.
Full-text available
The Attitudes Toward Women Scale (AWS) is routinely used as a general measure of sexism. In this article, it is argued that the AWS (Spence, Helmreich, & Stapp, 1973) actually measures overt or blatant sexism (harmful and unequal treatment of women that is intentional, visible, and unambiguous), whereas the Modern Sexism Scale (MS) measures covert or subtle forms of sexism (sexism that is either hidden and clandestine or unnoticed because it is built into cultural and societal norms). Support for this distinction is shown by way of (a) confirmatory factor analyses, (b) correlations with affective reactions to different categories of women and men (i.e., women and men in general, traditional women and men, feminists, and chauvinists), and (c) correlations with perceptions of sexual harassment. These analyses indicate that the AWS and MS scales measure distinct but related constructs.
Campus racial harassment provided the context for an experiment, replicated over 3 different campus samples, regarding the effects of social influence on Whites' reactions to racism. Hearing some-one condemn racism led Ss to express significantly stronger antiracist opinions than occurred following exposure to a no-influence control condition. Furthermore, hearing someone condone racism led Ss to adopt significantly less strong antiracist positions than when no other opinions were introduced. The robust social influence effects were obtained regardless of whether the source was White or Black or whether Ss responded publicly or privately. A social context approach to interracial settings is discussed.
The question addressed is, when do disadvantaged-group members accept their situation, take individual action, or attempt to instigate collective action? Ss attempted to move from a low-status group into an advantaged, high-status group and were asked to respond to their subsequent rejection. Ss who believed that the high-status group was open to members of their group endorsed acceptance and individual actions. When access to the high-status group was restricted, even to the point of being almost closed (tokenism), Ss still preferred individual action. Disruptive forms of collective action were only favored by Ss who were told that the high-status group was completely closed to members of their group. Ss who believed they were near to gaining entry into the high-status group favored individual protest, while Ss distant from entry were more likely to accept their position. The theoretical and societal implications of these findings are discussed. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
The purposes of this study were (a) to explore the utility of a model for predicting reporting of sexual harassment and (b) to collect data on the incidence of sexual harassment using the Sexual Experiences Questionnaire (SEQ; Fitzgerald et al., 1988). The sample included 214 tenure-track faculty women and 276 women graduate students. The full model postulated that age, marital status, feminist ideology, and frequency of behavior would be directly related to perceived offensiveness of the behavior. In turn, perceived offensiveness, normative expectations for reporting, and perceived outcomes of reporting would directly influence reporting. As predicted, perceived offensiveness showed a direct influence on reporting, and feminist ideology and frequency of behavior were significant predictors of perceived offensiveness. Incidence data showed that the most frequently experienced situations involved gender harassment and seductive behaviors. Results suggest that educating women about the offensiveness of sexual harassment might increase frequency of reporting. Also, further exploration of the model seems warranted.