A Female Flight from Math
At my high school back in the 1980s, some teachers in the math department became concerned that many of their high-achieving female students were dropping out of the honors math sequence before they even got to precalculus. These teachers wondered whether this might have something to do with the environment the girls faced in their classes. They created a girls-only honors math class in the hope that this would stave off the drop-out rate. My experiences with this classroom experiment sparked my interest in the problem of female underrepresentation in quantitative fields.
The gender disparity in quantitative fields grows from high school to college to graduate school, and is most pronounced in quantitative professions such as university-based mathematics research. The reasons for this are, of course, a subject for much debate. But it is worth noting that women who drop out of quantitative majors do not tend to have lower scores on college entrance exams or lower freshman grades than their male peers. If females are leaving math fields when they are performing just fine, it is worth considering what might induce them to leave.
As an experimental social psychologist, I have studied this question by conducting laboratory studies with female undergraduate math students. In these studies, I and my colleagues, Claude Steele and Lee Ross, investigated how women in math are affected by negative stereotypes about their mathematical ability and potential. In particular, we looked at how these stereotypes can create psychological pressures that may drive women out of math.
The Impact of Stereotypes on Female Students’ Test Performance
Do stereotypes really matter that much? Research shows that they do. The performance of high-achieving female math students on challenging math tests can be impaired by a social-psychological experience known as stereotype threat. This “threat” occurs when a female math student is taking a difficult math test, and the challenges she experiences with it bring to her mind negative stereotypes about female math ability. Such stereotypes are widespread: 45% of our research participants report believing that men are “better at math” than women—less than 1% report that women are better! Once she begins to consider these stereotypes, our female test-taker may become concerned that she will be judged according to them, or even that her performance may confirm them. Numerous experiments have found that the experience of “stereotype threat” is sufficiently distracting and upsetting to cause women to score lower on difficult math tests than equally skilled men.
Stereotype threat is likely to emerge during everyday experiences. For example, Michael Inzlicht and Thalia Ben-Zeev found that the performance decrements associated with stereotype threat arise when a woman takes a math test in a room with two male test-takers rather than two other women. (The implications of this finding for all-girls math classes like the one I took are provocative.) Paul Davies and his colleagues have found similar reductions in test performance among women who, before taking a difficult math test, are asked to watch TV commercials that depict women in a trivializing way (and a way that is, presumably, inconsistent with stereotypes about being good at math).
Accumulating experimental evidence thus suggests that the experience of stereotype threat contributes to women’s lower standardized math test scores (the test questions used in these experiments are typically taken directly from the GRE), and to their decreased persistence in quantitative fields.
How Stereotypes Pressure the Identities of Female Math Students
My research has considered a further question: Does stereotype threat also induce pressures on the identities of women in quantitative fields? That is, do female quantitative students feel pressure to adapt their feminine identities to these environments? Again, the answer is yes. My colleagues and I have found that in responding to stereotype threat, women in math often “bifurcate” their feminine identities, cutting off “feminine” traits they view as related to negative stereotypes about female math ability and potential, and continuing to identify with “feminine” traits that they view as relatively unrelated to the stereotypes.
Women students in quantitative fields (such as math, physics, electrical engineering, and statistics) report a bifurcated pattern of feminine identity relative to their female peers. They report fewer feminine traits that are likely to lead to their being negatively judged in math environments (e.g., flirtatiousness, emotionality, being gossipy, a stereotypically feminine appearance, and interest in having children), but no fewer feminine traits seen as unlikely to lead to such negative judgment (e.g., sensitivity, nurturance, and even fashion-consciousness).
Do women in mathematical arenas bifurcate their identities in response to prolonged exposure to threatening stereotypes in these environments? Or are women with bifurcated identities simply more likely to study mathematics? To determine whether exposure to stereotype threat could induce women to bifurcate their identities, we conducted a series of experiments. We led women to experience the “threat” of negative stereotypes about female math ability by exposing them to an article from Science purporting to show innate sex differences in math ability.
In response to the article, our female math students (but not their peers in other fields) bifurcated their feminine identities. On a written survey, they specifically disavowed feminine qualities that might be seen as related to negative stereotypes about female math ability. This bifurcation is a testament to female math students’ ability to adapt to the threats posed by math environments. However, it is not difficult to imagine how the stress of engaging in such adaptation could constitute yet another deterrent to women’s persistence in quantitative fields. Although sacrificing flirtatiousness as an aspect of one’s identity may seem trivial, sacrificing an interest in having children does not.
Keeping Women in Math by Removing Stereotype Threat
One important way to address women’s underrepresentation in mathematical pursuits is to create environments without stereotype threat—environments in which females are not concerned about being judged according to negative stereotypes. But how can this be done? Geoffrey Cohen and his colleagues have demonstrated the importance of “wise” mentoring: that is, mentoring in which students are assured that although the work which they are being asked to do is challenging, they do indeed have the ability to succeed at it. Joshua Aronson and his colleagues have emphasized the importance of stressing the “expandability” of intelligence: the idea that intellectual ability is not something that one has a finite amount of, and that it can be increased with experience and training. Finally, increases in female gender representation are likely to, in themselves, lead to further increases. The more our female math students are exposed to women role models who can show them that not only can women “do math” but also that their feminine identities need not be viewed as a liability, the more they are likely to view math environments as places where they can belong and succeed.
Aronson, J., C. B. Fried, and C. Good. “Reducing the Effects of Stereotype Threat on African American College Students by Shaping Theories of Intelligence.” Journal of Experimental Social Psychology issue no. (2002): page nos.
Cohen, G. L. “The Mentor’s Dilemma: Providing Critical Feedback Across the Racial and Gender Divides.” Dissertation Abstracts International (1999).
Davies, P. G., S. J. Spencer, D. M. Quinn, and R. Gerhardstein. “Consuming Images: How Television Commercials that Elicit Stereotype Threat Can Restrain Women Academically and Professionally.” Personality and Social Psychology Bulletin (2002).
Inzlicht, M., and T. Ben-Zeev. “A Threatening Intellectual Environment: Why Females Are Susceptible to Experiencing Problem-Solving Deficits in the Presence of Males.” Psychological Science (2000).
Spencer, S. J., C. M. Steele, and D. M. Quinn. “Stereotype Threat and Women’s Math Performance.” Journal of Experimental Social Psychology (1999).
Steele, C. M. “A Threat in the Air: How Stereotypes Shape Intellectual Identity and Performance.” American Psychologist (1997).
Princeton, New Jersey