The Interference of Stereotype Threat With Women's Generation of Mathematical Problem-Solving Strategies

Journal of Social Issues, Spring, 2001 by Diane M. Quinn, Steven J. Spencer

Diane M. Quinn[*]

At the highest levels of math achievement, gender differences in favor of men persist on standardized math tests. We hypothesize that stereotype threat depresses women's math perfomance through interfering with their ability to formulate problem-solving strategies. In Study 1, women underperformed in comparison to men on a word problem test, however, women and men performed equally when the word problems were converted into their numerical equivalents. In Study 2, men and women worked on difficult problems, either in a high- or reduced-stereotype-threat condition. Problem-solving strategies were coded. When stereotype threat was high, women were less able to formulate problem-solving strategies than when stereotype threat was reduced. The effect of stereotype threat on cognitive resources and the implications for gender differences in mathematical testing are discussed.

"I have no idea how to do this one, it just doesn't seem like there is enough information. I have no idea how to set up a formula for this. I really don't know what to do with that one.

"Hmm, I'm thinking that I don't know where to start...I don't know, I don't see, I don't understand any of them, so I'll skip it."

The above quotes illustrate how difficult and frustrating trying to solve mathematical problems can be. What makes these quotes particularly disheartening is that they all come from college women with strong mathematical backgrounds who are trying to solve high-level Scholastic Assessment Test (SAT) problems. Why are these women having a difficult time? What is interfering with their abilities to formulate problem-solving strategies? We propose that these women are in a situation in which cultural stereotypes about their math abilities are depressing their performance and interfering with their problem solving.

Stereotypes about academic skills are well known in our culture (Eagly & Wood, 1991; Eccles, Jacobs, & Harold, 1990; Levine & Ornstein, 1983; Maccoby & Jacklin, 1974; Swim, 1994). According to these stereotypes, men/boys are better at math and science domains, whereas women/girls are better at English and reading domains. These stereotypes are transmitted in the culture in a variety of ways, including through mass media, books, parents, peers, and teachers. They may affect children through the toys they play with, the books they choose to read, the way they are treated in class, and, eventually, the classes they choose to take and the careers they pursue (Biernat, 1991; Constantinople, Cornelius, & Gray, 1988; Eccles, 1987; Hewitt & Seymour, 1991; Leinhardt, Seewald, & Engel, 1979; Martin, Wood, & Little, 1990; Meece, Eccles, Kaczala, Goff, & Futterman, 1982).

One of the most basic ways in which stereotypes about women's math abilities are promulgated is through parents' and teachers' expectations. Jacobs and Eccles (1992) found that the stereotypic beliefs of mothers tended to color the mothers' perceptions of their daughters' and sons' math abilities. In addition, work by Frome and Eccles (1998) showed that mothers tended to underestimate the mathematical abilities of their sixth-grade daughters and overestimate the math abilities of their sons. In a study of high school students, Hyde, Fennema, Ryan, Frost, and Hopp (1990) found that high school boys reported greater favorable attitudes from their mothers, fathers, and teachers about their math ability than high school girls. These stereotypical expectancies and beliefs from parents, teachers, and the culture certainly have effects on girls' attitudes toward math. Jacobs and Eccles (1992) found that mothers' perceptions of their children's abilities have a greater influence on children's perception of their own abilities than do past grades. In addition, mothers' low expectations for their daughters are related to girls' having lower math performance expectations, which in turn leads to intentions to take fewer math courses (Eccles & Jacobs, 1986).

All of this research highlights the many powerful ways that stereotypes about gender and math are transmitted in the culture. What is the effect of these stereotypes on girls' and women's actual math performance and achievement? Examination of gender differences in math achievement largely contradicts the stereotypes. In the elementary school years, girls are performing better than boys on tests of computation and equally with boys on word problem tests (Hyde, Fennema, & Lamon, 1990). Again, at middle school age, girls are performing higher on computation tests and equally on word problem tests (Armstrong, 1981; Hyde et al., 1990). Girls are also getting higher grades in math classes than boys (Frome & Eccles, 1998; Kimball, 1989). Starting approximately at junior high, however, and continuing more strongly in high school, the pattern of achievement reverses. Boys and girls perform equally on computational tests, but boys outperform women on math word problem tests. This gap on math word problems continues and widens into college and adulthood (Armstrong, 1981; Hyde et al., 1990). Although recent work that has looked at national samples has found the achievement gap between high school men and women has narrowed over the past 20 years (Cole, 1997; Feingold, 1988), significant gaps are still quite evident when the highest achievers are examined. Indeed, on standardized tests such as the SAT and GMAT (Graduate Management Admissions Test), it is at the highest level, within the group of high scorers, that women are underrepresented in comparison to men, even though they may have equal math backgrounds and class grades (Cole, 1997; Stockard & Wood, 1984). In addition, women are still underrepresented in college majors and professional careers that stress mathematical skills, such as engineering and computer science (Ramist, Lewis, & McCamley-Jenkins, 1994; Strenta, Elliot, Adair, Scott, & Matier, 1993).