Past research suggests that awareness of negative stereotypes about the intellectual inferiority of one's "in-group" can undermine performance on academic tasks, especially among subjects who are strongly identified with the academic domain or highly invested in the academic task. This study included 72 students (n = 42 women, n = 30 men) and examined the effects of stereotype threat on test scores of students. It was hypothesized that students exposed to stereotypes predicting underperformance on the Psychology Major Test for their genders would score lower than students not exposed to these negative stereotypes. A 2 (gender) X 2 (stereotype threat condition) analysis of variance yielded significant interaction effects indicating that men in stereotype threat conditions outscored men in non-stereotype threat conditions. Given that most students experienced low domain identification with males being the least domain identified, the negative effects of stereotype threat were not observed.
When academic achievement has been measured by high-stakes standardized tests, factors including grade point average, class rank, test preparation, test-taking strategies and academic preparedness have been discussed as possible predictors of individual performance on these tests. The issue of test bias has also been discussed as a possible predictor of differential group performance on standardized test scores, especially in instances where background characteristics have been controlled for and group differences in scores have persisted along racial and gender lines (Suzuki, Ponterotto, & Meller, 2000). Further insight into the persistence of group differences in standardized test performance can be gained from a review of stereotype threat studies. These studies have investigated predictors of underperformance for stigmatized groups and found that testing situations, by becoming more anxiety provoking for certain students based on their race, class, and gender, could impede performance (Aronson & Salinas, 1997; Steele, 1997; Aronson, Lustina, Good, Keough, Brown, & Steele, 1999; Inzlicht & Ben-Zeev, 2000; Leyens, Desert, Croizet, and Darcis, 2000; Gonzales, Blanton, & Williams, 2002) Stereotypes can be positive, negative, or neutral beliefs about people based on their group membership and when related to academic achievement, these stereotypes often categorize European-Americans and Asian-Americans as more achievement oriented as compared to African-Americans (Nieman, Jennings, Rozelle, Baxter, Sullivan, 1994), high SES students as more academically advanced as compared to low SES students (Croizet & Claire, 1998), and men as more skilled in math as compared to women (Shih, Pittinsky, & Ambady, 1999).
According to Steele and Aronson (1995) negative stereotypes about one's in-group could produce a type of "stereotype threat" whereby one may feel pressure not to conform to given stereotypes, with this pressure then being seen to impede one's performance on certain tasks. Furthermore, Marx, Brown, and Steele (1999) suggest that stereotype threat could especially influence academic outcomes for groups of people who are strongly identified with an academic domain (concerned about performing well in a given content area) and negatively stereotyped in the same academic domain. The mix of these two conditions are expected to produce significant anxiety during the context of an academic performance that shifts attention away from the task.
PURPOSE OF THIS STUDY
This study investigated the effects of stereotype threat on test performance of male and female college students. Specifically, it was hypothesized that students who were exposed to stereotypes predicting underperformance for their genders, would score lower on an achievement test as compared to similar ability peers not exposed to these stereotypes. Stereotype threat was expected to exert the least influence on the scores of non-majors (less domain identified) and the most influence on the scores of psychology majors (more domain identified).
Seventy-two introductory psychology students, including 42 women and 30 men participated in this study. Thirteen percent of the men in the study were declared psychology majors while twenty percent of the women were declared psychology majors. Near the end of the term, participants were told that they would be given a short, cumulative assessment called the Psychology Major Test that included questions similar to those found on their previous three in-class exams. Furthermore, to ensure that the test was taken seriously, participants were told that they would be given extra credit points to be added to their final grade if they scored 75% or better on the test.
Materials and Procedure
Participants were randomly assigned to one of four conditions: male stereotype threat, male non-stereotype threat (control), female stereotype threat, female non-stereotype threat (control). After being told that they would be taking a 60-minute psychology exam consisting of 40-multiple choice questions, students were given their exams along with answer sheets and a printed one-page summary describing the test's purposes. The Psychology Major Test, created by the experimenter, was described as a measure of general knowledge and content mastery for students completing introductory psychology. To create the perception of test quality and standardization, it was discussed that the PMT had been given to more than 3000 students at over 138 colleges and was widely used. For those in the non-stereotype threat conditions no mention of gender differences in average scores was added to the one-page test description.
For those in the stereotype threat conditions, additional sentences were added to the one-page test description to produce the manipulation. In the male stereotype threat condition, a sentence was added stating that women tended to outscore men on the test by an average of 6 points. In the female stereotype threat condition, a sentence was added that stated that men tended to outscore women on the test by an average of 6 points. The range of test scores was reported as 0-40 and no mean test score was presented to the students.
To investigate the influence of stereotype threat on test performance, a 2 (male and female) X 2 (stereotype threat vs. non-stereotype threat) Anova was performed. Performance on the Psychology Major Test (PMT) was the dependent measure.
It was hypothesized that students exposed to stereotypes predicting underperformance for their genders, would score lower on an achievement test as compared to similar ability peers not exposed to these stereotypes. It was also hypothesized that in the stereotype threat condition, psychology majors would score lower than non-psychology majors due to the increased threat of performing poorly in an area important to their academic self-concepts. The two-way ANOVA indicated no significant main effects but did identify a significant interaction between gender and stereotype threat conditions ([F.sub.1,71] = 6.29, p < .05). Interestingly, this interaction indicated that men in the stereotype threat condition (X = 26.46, SD = 1.9) significantly outscored men in the non-stereotype threat conditions (X = 22.41, SD = 5.6) on PMT scores, as illustrated in Figure 1. As shown in figure 2, there were no significant differences between PMT scores of women in the stereotype (X = 25.10, SD = 3.5) and non-stereotype threat conditions (X = 26.0, SD = 4.06). As illustrated in Table 1, men in the stereotype threat condition (X = 26.46, SD = 1.90) outscored all other comparison groups and the mean score for all PMT test takers was 24.93 (SD = 4.29).
[FIGURES 1-2 OMITTED]
In addition to the 2 X 2 ANOVA, a 2 X 2 ANCOVA with current course grade as the covariate was employed and produced similar results ([F.sub.1,71] = 4.14, p < .05). Although subjects were randomly assigned to stereotype threat conditions it was deemed necessary to control for course grade so that disproportionate numbers of high or low performing psychology students would not end up in the stereotype threat or non-stereotype threat conditions.
Although initially surprising, the findings do support current research involving stereotype threat. Stereotype threat requires a high-degree of domain identification (concern about performing well in a specific content area) as well as being negatively stereotyped for performance in this content area (Steele, 1997). Students in this study did not have a high-degree of domain-identification given that 13% of males (4 out of 30) and 20% of females (8 out of 40) in the study were declared psychology majors. Male and female students exposed to negative stereotypes about performance on the Psychology Major Test, without being highly identified with the field of psychology did not suffer the negative effects of stereotype threat. In fact, negatively stereotyped men outscored all other groups on the Psychology Major Test. Given that men were least likely to be psychology majors and consequently least likely to be domain identified, they may not have experienced stereotype threat but may in fact have received a boost in performance under the "threat" manipulation because of a motivation to prove the test wrong (Wheeler and Petty 2001). In addition, the administration of the PMT by a male psychology professor may have produced stereotype "disconfirming" information by providing an example of male achievement in psychology minutes before taking the PMT. This disconfirming information may have offset the detrimental effects of stereotype threat on PMT scores for males, and in fact, could also have produced feelings of increased self-efficacy in negatively stereotyped males leading to higher PMT scores for this group. Marx and Ruggiero (1999) have postulated such a theory, finding that when women's math ability was negatively stereotyped and a math assessment was administered by both male and female experimenters, women underperformed when in the male administrative but not the female administrative conditions.
Further research should investigate the impact of stereotype threat on standardized test performance for those who are minimally, moderately, and highly identified with a given academic domain. These varied levels of identification or ego-involvement may produce advantages or disadvantages during testing situations depending on whether one is invested enough in a task to try their best or overly invested causing distraction while performing a task.
Aronson, J., Fried, C., & Good. (2002). Reducing the effects of stereotype threat on African-American college students by shaping theories of intelligence. Journal of Experimental Social Psychology, 38, 113-125.
Aronson, J., Lustina, M., Good, C., Keough, K., Brown, J. L., & Steele, C. M. (1999). When white men can't do math: Necessary and sufficient factors in stereotype threat. Journal of Experimental Social Psychology, 35, 11-23.
Aronson, J. & Salinas, M. F. (1997). Stereotype threat and women's calculus performance. Unpublished manuscript, Univ. of Texas.
Ben-Zeev, T., Fein, S., & Inzlicht, M. (2004). Arousal and stereotype threat. Journal of Experimental Social Psychology, 12, 409-415.
Croizet, J. C., & Claire, T. (1998). Extending the concept of stereotype threat to social class: The intellectual underperformance of students from low socioeconomic backgrounds. Personality and Social Psychology Bulletin, 24, 588-594.
Geen, R. G. (1991). Social motivation. Annual Review of Psychology, 42, 377-399.
Gonzales, P. M., Blanton, H., & Williams, K. J. (2002). The effects of stereotype threat and double-minority status on the test performance of Latino women. Personality and Social Psychology Bulletin, 28, 659-670.
Inzlicht, M., & Ben-Zeev, T. (2000). A threatening intellectual environment: Why females are susceptible to experiencing problem-solving deficits in the presence of males. Psychological Science, 11, 365-371.
Leyens, J.-Ph., Desert, M., Croizet, J. C., & Darcis, C. (2000). Stereotype threat: Are lower status and history of stigmatization preconditions of stereotype threat? Personality and Social Psychology Bulletin, 26, 1189-1199.
Marx, D., Brown, J. L., & Steele, C. M. (1999). Allport's legacy and the situational press of stereotypes. Journal of Social Issues, 55 (3), 491-502.
Marx, D., & Ruggiero, K. M. (1999). Role-conferred competence: How minority experimenters affect minority students' test performance. Manuscript in preparation, Harvard University, Cambridge, MA.
Niemann, Y. F., Jennings, L., Rozelle, R. M., Baxter, J. C., & Sullivan, E. (1994). Use of free responses and cluster analysis to determine stereotypes of eight groups. Personality and Social Psychology Bulletin, 20, 379-390.
Shih, M., Pittinsky, T. L., & Ambady, N. (1999). Stereotype susceptibility: Identity salience and shifts in quantitative performance. Psychological Science, 10, 80-83.
Steele, C. M. (1997). A threat in the air: How stereotypes shape intellectual identity and performance. American Psychologist, 52, 613-629.
Steele, C. M., & Aronson, J. (1995). Stereotype threat and the intellectual test performance of African Americans. Journal of Personality and Social Psychology, 69, 797-811.
Suzuki, L. A., Ponterotto, J. G., & Meller, P. J. (Eds.). (2000). Handbook of multicultural assessment (2nd ed.). San Francisco: Jossey-Bass.
Wheeler, S.C., & Petty, R.E. (2001). The effects of stereotype activation on behavior: A review of possible mechanisms. Psychological Bulletin, 127, 797-826.
KEITH B. WILLIAMS
Richard Stockton College of New Jersey
Table 1 Mean PMT Scores by Gender and Stereotype Threat Condition Group M SD N Major Men stereotype threat 26.46 1.90 13 2 Men non-stereotype threat 22.41 5.80 17 2 Women stereotype threat 25.08 3.51 24 4 Women non-stereotype threat 26.00 4.06 18 4 Total 24.93 4.29 72 12 Note. Maximum score on the PMT (Psychology Major Field Test) is 40 points therefore the range of possible individual scores is 0-40.…