Gendered Math: Stereotype Threat Susceptibility and Women’s Choices in College Majors
Even though some critics argue that the gender gap in mathematics has closed, the majority of people knows and accepts the gender differences in college major departments. “In 2011, only twenty-six percent of workers in science, technology, engineering, and math (STEM) fields were women, according to the US Census Bureau.” Typically, more men pursue studies in physics, mathematics, and computer science than women. Though some critics argue that overrepresentation of males in these fields can explain this phenomenon, it cannot explain how other historically male dominated fields such as medicine and law have managed to gain a greater proportion of women over time. Therefore, the environments of mathematics and hard science departments must contain qualities that attract men to and repel a lot of women from their fields.
Stereotype threat is one possible explanation for this gender gap; it involves an unconscious fear of confirming a negative stereotype by performing badly in the subject area. Society has accepted a stigma of women possessing less math ability than men. “Junior high school boys outperform girls on advanced quantitative assessments, and high school and college men perform better than women on tests of advanced mathematical ability.” If young girls get exposed to and accept this stigma, then they will feel pressure in the mathematics environment. Their working memories become compromised, prohibiting their ability to think clearly on tests, and their chances for competing for grades and future job positions diminish because of an accepted inferiority to their male peers. If stereotype threat greatly affects performance, then women’s perspectives in mathematics can shift due to a change in the stereotype.
Results for effects of stereotype threat scatter across the board, from no gender differences to large gaps between male and female test scores. Some results demonstrate the highest performing boys achieve significantly higher on math tests than the highest performing girls in a threat-induced environment. This counters the results showing that “girls perform similarly or better than boys across all years of schooling,” which can be attributed to stereotype threat needing harder problems to activate or teachers using more than only test results to grade their students. Ganley’s study looks at elementary and middle school, high achieving students, but it does not find any supporting evidence for stereotype threat effects. No matter if they made the tests harder, increased the age range, or altered the explicit stereotype activation method, the results stayed consistent with no induced susceptibility to stereotype threat.
Nevertheless, Ganley’s study does not look into implicit stereotypes. All their tests directly address the stereotype in the tests by openly informing participants that girls are worse than boys at math. Their results show significant “gender differences on the mathematics tests regardless of condition.”In other words, girls underperformed compared to boys no matter if they received the explicit threat condition or took the test in the neutral condition. The research leads to an ambiguous conclusion if the gender gap comes from some unknown source or if Ganley did not correctly separate a neutral from a threat-induced environment.
Ganley concludes that future studies need to figure out what triggers this gap besides stereotype threat, but Amy Keifer argues that “explicit concerns about being stereotyped or stereotype-consistent performance, such as evaluation apprehension, anxiety, and performance expectations, do not reliably mediate stereotype threat effects,” debasing Ganley’s research methods. Keifer associates correct conditions with “implicit gender-math stereotypes,” subconscious links of men more than women with mathematics. If people unknowingly label mathematics as masculine, women’s susceptibility to stereotype threat increases because they cannot relate as well to the field. So women in college will see mathematics as a masculine major that contradicts their self-identified femininities and will chose to avoid it.
To test implicit gender-math stereotypes, Kiefer had subjects categorize words based on two categories: self or other. This test resulted to math being more associated with men, and it gave scores to woman participants based on implicit identification with femininity. Then all participants took a math test, where women’s math test scores were lower in the threat than the reduced threat condition. The findings also concluded that “women with strong implicit gender-math stereotypes chronically activate these stereotypes and thus experience stereotype threat even under so-called ‘reduced threat’ conditions, whereas women with weak or counter implicit gender-math stereotypes are most benefitted by threat reduction.” Women who subconsciously believe that mathematics is more a man’s field will therefore activate the stereotype regardless of an outside stimulation, but women with weak gender stereotypes can greatly improve their math scores when in supportive environments. To alter this view, societal messages and images of subjects like math and physics need to stop associating strictly with males. Since previous research supports the malleability of such stereotypes with counterexamples, a change in presentation can lead to a change in gender representation in these fields.
Usually the stereotype threat affects women who most identify with the mathematics field. “Mathematics identification involves two components: feeling that you are good at mathematics and feeling that it is important to you to be good at mathematics.” So women who already associate themselves with math will more likely perform worse on a test in a threatening environment than women who do not regard their scores as important. Based on Keifer’s gender-math identification theory, a woman who self identifies with math distances herself from feminine traits. She has weaker identification to her femininity so that she can fit into the math world. Stereotype threat may lead to worse performance in math courses for women or less identification as a woman to succeed in math, but what draws the other women away before they even show interest in mathematical fields?
College students choose majors based on factors dealing with interest, success rates, graduation time, and availability. Basit Zafar argues that “the gender gap in mathematics achievement and aptitude is small and declining, and gender differences in mathematical achievement cannot explain the higher relative likelihood of majoring in sciences and engineering for males.” He thinks that stereotype threat and low self-confidence do not significantly link to woman underrepresentation in the sciences and math. For female students, the factors that matter most for choosing a major are gaining parent approval and enjoying coursework or work in future jobs; oppositely, male decision relies on outcomes associated with their future workplace, such as social status and salary. Zafar thinks that women stay away from math and engineering, not because of future salary differences or their beliefs in lack of ability, but because they will not enjoy the coursework in that subject area. Most women simply do not take interest in mathematics.
Nonetheless, stereotype threat and disinterest correlate, especially with women who have not already committed to mathematical fields. Women do not even need other people to elicit stereotype threat and avoid mathematically-based courses. Sapna Cheryan and her team explore the relationship between ambient belonging and how this can impact interest and participation in the computer science major. Ambient belonging is the feeling of fitting into an environment. Through four studies, the group determines that “women are more drawn to an employment opportunity at a company whose environment contained objects not stereotypically associated with computer scientists, whereas men were drawn to a company whose environment contained objects that were associated with the stereotypical computer scientist.” In other words, women who think their environments contain masculine objects stereotypically associated with computer science feel uncomfortable and unwelcome in them; in contrast, men want their environments to have masculine characteristics. The physical cues in an associated environment, even without someone present from the environment, can communicate stereotypes of women as inferior to their male peers.
Unlike the previous studies, Cheryan focuses on what prevents women from gaining interest in the field before even taking the classes. Instead of implicit stereotypes originating from retention of women in technical fields or their ability to work well in a stereotype threat environment, the research shows that women immediately activate stereotypes when entering a room. Women in the study chose between two teams, one with neutrally-rated objects in the room and one with masculine-rated objects. Even if the teams included all women, the test subjects still preferred the environment with non-stereotypical objects. Those objects led to concern about the women being devalued because of their gender. If they rated the objects in the room as more masculine, they expressed more feelings of discomfort and precluded interest in computer science. Thus gender representation does not cause interest in a field; the woman needs a sense of belonging to generate interest. In a college setting, the math and physics departments hiring more women professors may not lead to more woman math majors.
Cheryan and her team separated stereotype threat and ambient belonging by saying that the threat comes from expected impressions others may have of the individual, which differs from belonging, impressions of the field. However both can work off of each other, so that the stereotype threat actually influences the level of comfort and belonging of a woman in a work or learning environment. The objects could be sending the message of implicit stereotype threat. If a woman goes into a room covered in objects that she has previously labelled as masculine, those objects, like the situational conditions for stereotype threat, activate her preexisting stereotype. This leads to concerns about being a target of gender-based discrimination or being devalued by coworkers for her femininity. Before she can reinforce the stereotype of inferiority, she chooses not to participate. Cheryan’s research shows that if women think they would be accepted and welcome in the environment, they may choose to join that group. Conversely, if the department exploits a masculine environment, women will lack interest in that subject.
The gender of the teacher is another important factor for inducing stereotype threat. Role models provide guidance to help the students with their experience through the field they study. They also create connections for the student outside of the academic world through networking. Men are less likely to mentor women due to rumors of sexual relations, women perceived as less focused, or their more comfort mentoring a male. Men with male role models progress in their career paths more effectively, while women struggle to find support systems in mathematical fields. Helena Holmlund’s study sites that there has not been evidence to support “a larger number of female faculty or a female dissertation chair [that] positively affects future success for female students.”  Similarly to what Cheryan stated, the representation of female professors in a department does not directly correlate with the interest of female students to major in that department or the success of those females after college. Men seem to benefit from male mentors to further their careers, but evidence shows that women do not have that same benefit with female mentors, causing another reinforcement of a gendered stereotype in math.
Though Holmlund’s study found insignificant evidence that student-teacher gender interactions play roles in explaining gender differences in school performance, it did find slight grade differences in mathematics. The first math course, in which students can enroll by choice, shows a gender gap in favor of male students. So if girls do poorly in their first exposure to mathematics outside of class requirements, then they will have less confidence and interest in furthering those studies. If males do better than females, the phenomenon emphasizes the preconceived stereotype.
Holmlund’s study does not support its hypothesis that student-teacher gender similarities would improve student performance, but it sites numerous research that does contain significant evidence for similar topics. Past research found that “having a same sex teacher has substantial positive impacts, both on test scores, student interest in the subject, and teacher assessments of students.” Especially in introductory courses where students learn the basic understandings of the subject, the sex of the teacher can affect how the student will perceive what the next four years of study will entail. If female or male professors encourage and inspire their female students in introductory mathematical courses, the interest and further exploration of that subject increases for women.
Lastly, stereotype threat susceptibility initiates from none other than the parents. Developmental theories such as the data in Carol Tomasetto’s report suggest that “parental attitudes and endorsement of gender stereotypes about math are important for the development of children’s academic attitudes, beliefs and performance.”  Parents influence their kids from the moment they enter the world. Children pick up subtle hints and implicit stereotypes from what their parents say and how their parents act. Stereotype threat susceptibility is no different because parents can enforce the belief that math is a male activity. They can use explicit methods such as verbal comments or implicit methods such as not buying logic-based toys for daughters or intruding in daughter’s math homework more than son’s, implying the girl needs more help than the boy in that subject.
In Tomasetto’s study, elementary school children completed a math test and a survey about their own explicit stereotypes with gender and mathematics, where sixty nine percent of the kids indicated that girls and boys are equally good at math. Though Tomasetto’s hypothesis suggested a link to both parents’ stereotypes, the data supports that “the endorsement of gender stereotype by the child and gender stereotype by both parents were not significantly related to math performance.” The child’s own stereotype views and both parents’ stereotype views did not play a part in the child’s math performance. This implies that explicit methods do not induce the trigger of stereotype threat, similarly to Keifer’s argument. But the data does connect the child’s performance to the mother’s gender stereotype. “ST [stereotype threat] was a significant predictor of performance,” and “ST led to the classic performance deficit when mothers’ stereotypes were relatively stronger, whereas ST had no effect when mothers’ level of stereotypes was lower.” The data indicates that mother’s gender stereotypes moderates the child’s stereotype threat susceptibility; all variables of age, school district, or level of stereotype identification support this trend.
Mothers, particularly, influence their young daughters’ perspectives on life, and they become the first female mentor for their daughters to seek advice on school. When mothers do not reject the stereotypical view of math as masculine, the child’s performance drops in the stereotype threat environment; one the other hand, when mothers strongly reject the stereotype, the daughters’ performances are equal in both the stereotype threat induced and control condition. The parents, especially the mothers, have great power over their kids’ potential abilities in early years of school. Unlike professors in college who have to work around already learned stereotypes, mothers carry the potential to shape positive views of femininity and mathematics before other stereotypes form. With a counter stereotypical enforcement of holding equal expectations for their daughters and sons in mathematics, mothers could potentially protect their daughters from effects of stereotype threat. The daughters could then become more interested in learning mathematics and get more involved in STEM fields.
The shortcomings of the study deal with some issues it overlooks. Fathers also play roles in inducing stereotype threat susceptibility, but only when their daughters have gotten older. “Children 4—6 years of age are more sensitive to mothers’ than fathers’ attitudes.” So fathers will not affect their daughters’ stereotype beliefs until around the age of nine or ten. However, both parents, through parenting techniques, can influence how their children perform in school. A study on single fathers and daughter stereotype threat susceptibility would further the research in the origins of stereotype threat. Going off Keifer’s data, if a girl bonds closer to her father, she may self-identify as more masculine and perform less of a stereotype threat than a girl who self-identifies as more feminine. Tomasetto also overlooks other socialization techniques such as the media, school officials, and peers. Though the data shows a correlation between the mother and daughter on stereotype threat, other socialization sources contribute to the child’s perception of self and their identification with mathematics.
Overall, studies show significant data validates effects of stereotype threat and its origins on women in mathematics. However, stereotype threat is a complex phenomenon, and the source of activation has not been pinpointed. Some critics argue that implicit cues can spark gender gaps in test performance; parents, teachers, and even just classroom objects can make a female student unwelcome or feel inferior to her male peers. Women need to sense people want them in the work or study environment, even by changing the décor of the room. Gender stereotypes can be prevented by reducing stereotypes in the family environment, stopping media images of unintellectual girls, and creating prevention programs for negative imagery. Women also need supportive roles models who will be able to guide them through the process of demasculinizing the mathematics domain. Hopefully, reducing stereotype threat will close the gender gap once and for all.
 Jonathan Dame, “Choosing the Right Major Key to Closing Gender Pay Gap,” USA Today College, December 13, 2013, accessed December 15, 2013, http://www.usatoday.com/story/news/nation/2013/12/13/gender-pay-gap-stem/4014899/.
 Sapna Cheryan, Victoria C. Plaut, Paul G. Davies, and Claude M. Steele, “Ambient Belonging: How Stereotypical Cues Impact Gender Participation in Computer Science,” Journal of Personality and Social Psychology 97, no. 6 (2009): 1045.
 Amy K. Keifer and Denise Sekaquaptewa, “Implicit Stereotypes and women’s math performance: How implicit gender-math stereotypes influence women’s susceptibility to stereotype threat,” Journal of Experimental Social Psychology (2006): 1.
 Colleen M. Ganley, Leigh A. Mingle, Allison M. Ryan, Katherine Ryan, Marina Vasilyeva, and Michelle Perry, “An Examination of Stereotype Threat Effects on Girls’ Mathematics Performance,” Developmental Psychology 49, no. 10 (2013): 1886.
 Ibid 1887.
 Ibid 1894.
 Kiefer 2.
 Ibid 2.
 Ibid 5.
 Ganley 1888.
 Kiefer 6.
 Basit Zafar, “College Major Choice and the Gender Gap,” Federal Reserve Bank of New York Staff Reports, no. 364 (2009): 1.
 Ibid 15.
 Cheryan 1045.
 Ibid 1056.
 Ibid 1057.
 Julia T. Wood, Gendered Lives: Communication, Gender, and Culture, Ninth Edition (Boston: Wadsworth Cengage Learning, 2011), 242.
 Helena Holmlund and Krister Sund, “Is the gender gap in school performance affected by the sex of the teacher?” Labour Economics 15, no. 1 (2008): 39.
 Ibid 43.
 Ibid 39.
 Carlo Tomasetto, Francesca Romana Alparone, and Mara Cadinu, “Girl’s Math Performance Under Stereotype Threat: The Moderating Role of Mother’s Gender Stereotypes,” Developmental Psychology 47, no. 4 (2011): 944.
 Ibid 946.
 Ibid 946.
 Ibid 947.