US tab

Self-silencing in Response to
Sexist Behavior: Exploring Women's
Willingness to Confront Sexism

By Marie Sabbagh, Tess Hare, Erika Wheelhouse, and Holly McFarland | Mentor: Dr. Erin Murdoch


Participants' responses to the second sexist comment proved difficult to label as simply confronting or nonconfronting, forcing us to devise a rating system for "confronting behavior" in order to assign a number to how forceful the confrontation was. For example, we needed to decide as a group whether nervous laughter could be interpreted as some kind of mild confrontation, or whether it was just a reaction. We had to determine whether refusing to choose the stay-at-home mother as the main housekeeper could be interpreted as confronting the "cleaning is woman's work" comment. Ultimately we chose to interpret participants' behavior as either 1 - Forcefully Confronting (n = 9), 2 - Mildly Confronting (n = 10), or 3 - Non-Confronting (n = 10). Higher numbers on this scale indicated greater self-silencing. "I don't appreciate that comment" and "we shouldn't assume she can cook just because she's female" are examples of statements identified as Forcefully Confronting.

We conducted a 2 (Condition: modeling, no modeling) x 2 (Comment: comment 1, comment 2) mixed design ANOVA on confrontational behavior. There was no evidence of a significant difference between conditions, F(1, 27) = .004, p > .05. That is, participants in the presence of confederates who confronted a sexist statement were no more likely to confront (M = 2.27, SE = 0.15) than were participants in the condition without a confronting confederate (M = 2.25, SE = 0.18). However, participants were more likely to confront the second sexist comment (M = 2.03, SD = 0.82) than the first (M = 2.48, SD = 0.69), t(28) = 2.78, F(1, 27) = 6.76 p < .05. In other words, this analysis shows that, although modeling confronting behavior was not more likely to increase confronting behavior, a higher number of sexist remarks will result in unmistakable confronting behavior.

The Condition X Comment interaction was not significant, F(1, 27) = .35, p > .05. However, although these results were not significant at the multivariate level, exploratory analysis revealed that increased confrontations after the second sexist statement occurred only in the modeling condition. Participants in the modeling condition were significantly more likely to confront the second sexist statement (M = 2.00, SD = 0.87) than the first sexist statement (M = 2.53, SD = 0.62), p = .02. A significant increase in confronting did not occur when the confederate remained silent. Next, we conducted analyses of our individual differences measures to determine whether any of these variables contributed to participants' confronting behavior. Results indicated that only self-monitoring made a significant difference in confronting. To categorize participants' level of self-monitoring and present a clear relationship between confronting and self-monitoring, a median split was used to create high and low self-monitoring groups. An independent samples t-test indicated that high selfmonitors (M = 2.53, SD = 0.64) were less likely to confront than were low self-monitors (M = 1.50, SD = 0.65), t(27) = -1.53, p < .05. In other words, participants who are more careful with how they appear to others are less likely to confront sexist statements.

We also explored whether the age of participants could be a factor. We used a median split for age that resulted in a mean of 21.85 years (SD = 1.41) for younger participants and 34.00 years (SD = 8.60) for older participants. An independent samples t-test showed that older participants (M = 1.73, SD = 0.80) were somewhat more likely to confront sexist behavior than were younger participants (M = 2.31, SD = 0.75), t(26) = 1.08, p = .06.

Careful review of the Comments section of the demographic survey yielded the curious finding that 10% of the participants who were labeled as not verbally confronting the sexist comments attested in writing that they did in fact confront.

Discussion >>