Reducing adolescent girls ’ concerns about SteM stereotypes : When do female teachers matter ?

In two experiments, we examined how teacher gender and stereotype threat cues affected adolescents’ self-reported concerns about being negatively stereotyped in computer science courses. High-school students (Experiment 1: N = 218; Experiment 2: N = 193) read about two computer science courses, one with a competent male teacher and one with a competent female teacher, and were randomly assigned to one of three experimental conditions. In the stereotype threat condition, they read a paragraph that introduced negative stereotypes about girls’ perforRésumé Dans deux expériences, nous avons examiné comment le sexe des enseignants et la menace du stéréotype affectaient les craintes rapportées par les adolescents d’être négativement stéréotypés lors de cours en sciences informatiques. Des élèves du secondaire (Expérience 1 : N = 218 ; Expérience 2 : N = 193) ont lu la description de deux cours de sciences informatiques, l’un dispensé par un enseignant compétent de sexe masculin et l’autre par un enseignant compétent de sexe féminin, et ont été assignés au hasard à l’une des trois conditions expérimentales suivantes. Dans la condition de Key-words Adolescents, stereotype threat, STEM, stereotypes, identity, gender Mots-clés CAdolescents, menace du stéréotype, STEM (science, technologie, ingénierie et mathématiques), stéréotypes, identité, genre This work was supported by grants from the National Science Foundation, SMA-0835854 to ANM and DRL-0845110 to SC. We thank the participating teachers and students; C. Handron, M. Vichayapai, and E. Wu for assistance with data collection; and D. Apablasa, D. Bourgin, H. Earle, T. Hendershot, B. Hyland, H. Jin, W. Liu, E. Matskewich, A. Montoya, A. Pena, and N. Soon for assistance with data entry and manuscript preparation. Correspondence concerning this article should be addressed to Allison Master, Box 351525, University of Washington, Seattle, WA, United States of America, 98195. * Institute for Learning & Brain Sciences, Box 351525, University of Washington, Seattle, WA, United States of America, 98195. E-mail: [email protected] ** Department of Psychology, Box 351525, University of Washington, Seattle, WA, United States of America, 98195. E-mail: [email protected] *** Institute for Learning & Brain Sciences and Department of Psychology, Box 357988, University of Washington, Seattle, WA, United States of America, 98195. E-mail: meltzoff@ uw.edu RIPS_3-4_2014.indd 79 16/01/15 17:49 REDUCING GIRLS’ CONCERNS ABOUT STEREOTYPES 80 Gender disparities in STEM (science, technology, engineering, and math) fields remain a troubling problem in American education. The presence of female teachers as role models for girls has been suggested as a way to help increase women’s participation in STEM (Dasgupta, 2011). However, research on this topic has shown mixed results. While some studies demonstrate benefits of female teachers for female students (Asgari, Dasgupta, & Gilbert Cote, 2010; Bettinger & Long, 2005; Stout, Dasgupta, Hunsinger, & McManus, 2011), others show that male teachers can be equally effective (Canes & Rosen, 1995; Dynan & Rouse, 1997; Gilmartin, Denson, Li, Bryan, & Aschbacher, 2007). In the current work, we investigate the situational factors that influence when female and male teachers have differing consequences for girls in STEM fields. In two experiments, we examined how teacher gender and stereotype threat cues affected adolescents’ self-reported concerns about being negatively stereotyped in mance; in the no gender difference condition, they read a paragraph that countered negative stereotypes; and in the baseline control condition, they read neither paragraph. In both experiments, girls reported more concerns about being negatively stereotyped than boys when the teacher was male versus female, and this effect was specifically driven by significant differences in the stereotype threat condition. When situational cues are threatening, female teachers (compared to male teachers) reduce girls’ concerns about being negatively stereotyped, with implications for both theories of identity and educational practice. menace du stéréotype, ils lisaient un paragraphe qui décrivait des stéréotypes négatifs sur les performances des filles ; dans la condition d’absence de différence entre les sexes, ils lisaient un paragraphe qui réfutait les stéréotypes négatifs ; dans la condition contrôle, ils ne lisaient aucun paragraphe. Dans les deux expériences, les filles ont rapporté plus de crainte que les garçons d’être stéréotypées négativement lorsque l’enseignant était un homme plutôt qu’une femme, et cet effet était spécifiquement lié aux différences dans la condition de menace du stéréotype. Lorsque la situation est menaçante, les enseignants de sexe féminin réduisent, par rapport aux enseignants de sexe masculin, les craintes des filles d’être stéréotypées négativement, ce qui a des implications pour les théories de l’identité et les pratiques éducatives. RIPS_3-4_2014.indd 80 16/01/15 17:49 Revue InteRnatIonale de PSycholoGIe SocIale 2014 n° 3-4 81 computer science courses. We hypothesize that female teachers may be particularly important compared to male teachers in situations that activate stereotype threat for girls. Stereotype threat and teacher gender Stereotype threat refers to students’ concerns about being judged through the lens of a negative stereotype (Steele, 1997). According to a pervasive North American stereotype, females have less ability and interest than males in STEM domains (Good, Aronson, & Inzlicht, 2003). One potential way to reduce the effects of negative stereotypes is to expose women to female role models. Role models are successful “experts” in the field, ranging from teachers to more advanced peers and other professionals (Lockwood, 2006; Stout et al., 2011). Within contexts in which stereotype threat is salient (such as female math majors about to take a difficult math test), female role models can provide benefits over male role models that include improved performance (Marx & Roman, 2002; Marx, Stapel, & Mueller, 2005; McIntyre, Paulson, & Lord, 2003) and more positive attitudes (Lockwood, 2006; Stout et al., 2011). Competent female teachers show that women can overcome these stereotypes and succeed in STEM (Lockwood, 2006), and may also signal to girls that their teacher will be less likely to endorse negative stereotypes about them (Wout, Shih, Jackson, & Sellers, 2009). However, in other situations, studies have found that teacher gender has little effect on high-school girls’ motivation and achievement in STEM (e.g., Ehrenberg, Goldhaber, & Brewer, 1995; Martin & Marsh, 2005). Having support from either male or female role models prior to college often influences women to choose STEM majors when they reach college (Downing, Crosby, & Blake-Beard, 2005). Similar results have been found during college: College women who viewed their science professors as positive role models were more interested in science careers, regardless of the professor’s gender (Young, Rudman, Buettner, & McLean, 2013). A study across many STEM and non-STEM departments at three universities found that increases in the proportion of female faculty did not predict subsequent increases RIPS_3-4_2014.indd 81 16/01/15 17:49 REDUCING GIRLS’ CONCERNS ABOUT STEREOTYPES 82 in the proportion of female students recruited to that department (Canes & Rosen, 1995). The abovementioned research suggests that female teachers may be beneficial for women in comparison to male teachers specifically in situations where stereotype threat is salient. However, there has not yet been a direct experimental manipulation of both stereotype threat and teacher gender within a single study. Some studies have manipulated role model gender but not threat (Lockwood, 2006; Marx & Roman, 2002; Stout et al., 2011); others manipulated threat but not role model gender (Marx et al., 2005; McIntyre et al., 2003). Thus, we lack direct experimental evidence about how girls respond to female versus male teachers when they are or are not under threat. We suggest that female teachers are particularly helpful for girls compared to male teachers when negative stereotypes are activated, and that this has important implications for identity theory and educational policy and practice. adolescence as a critical age period for identity Adolescence is a particularly critical age at which to examine issues about academic identity. Adolescents begin to make significant career choices (Weisgram & Bigler, 2006), and girls in high school have fewer career aspirations involving computers and technology than boys (Lupart & Cannon, 2002; Schulenberg, Goldstein, & Vondracek, 1991). Because adolescence is a key time to recruit girls into advanced STEM training and career paths (Barker, Snow, Garvin-Doxas, & Weston, 2006), it is useful to understand whether and under what circumstances female teachers can ensure more positive experiences for girls in STEM than male teachers. There is strong evidence that children are significantly influenced by gendered stereotypes about STEM as detected both by explicit and implicit measures (e.g., Cvencek, Meltzoff, & Greenwald, 2011). Although adolescent girls may not confront negative STEM stereotypes daily (Ganley et al., 2013), girls are likely to eventually experience situations in which they must confront negative gender stereotypes, and teacher gender may be particularly influential for girls in these situations. RIPS_3-4_2014.indd 82 16/01/15 17:49 Revue InteRnatIonale de PSycholoGIe SocIale 2014 n° 3-4 83 the current research We focus on computer science due to the particularly low percentage of females; women earn approximately 18% of bachelor’s degrees in computer science, one of the lowest percentages of any STEM field (National Science Foundation, 2013). Previous research has found that stereotypes about computer scientists (e.g., as male, technologically oriented, and socially awkward; Cheryan, Plaut, Handron, & Hudson, 2013) lower high-school girls’ interest in computer science (Master, Cheryan, & Meltzoff, 2014). Here we examine a complementary piece of the puzzle: How stereotypes about girls’ ability in computer science affect their concerns about being negatively stereotyped. These concerns are important because they provide a window into girls’ thoughts about stereotypes and their own identity and may predict other meaningful STEM outcomes such as performance (Cohen & Garcia, 2005; Spencer, Steele, & Quinn, 1999). Because performance in early courses can affect whether students choose to persist in STEM majors (Miyake et al., 2010), investigating factors that encourage girls in introductory courses is crucial to ensuring their future participation. In two experiments, we examine how teacher gender and stereotype threat cues affected adolescents’ self-reported concerns about being negatively stereotyped in potential computer science courses. We hypothesize that when threatening cues become salient, female teachers may elicit fewer concerns than male teachers. But when girls are less concerned about negative gender stereotypes, we predict that female and male teachers should be equally effective.