Stereotype threat in men on a test of social sensitivity

Sex Roles: A Journal of Research,  April, 2005  by Anne M. Koenig,  Alice H. Eagly

• E-mail
• Print
• Link

Theorists define stereotype threat as a state of self-evaluative threat, whereby anxiety about confirming a negative stereotype in others' eyes, or in one's own, produces behavior that is consistent with and confirms the stereotype (Steele, 1997; Steele & Aronson, 1995; for a review see Steele, Spencer, & Aronson, 2002). Stereotype threat is usually explained as assimilation to a negative group-relevant stereotype through motivational processes (Wheeler & Petty, 2001). Theoretically, stereotype threat can happen to anyone with a group identity laden with a negative stereotype, even if the group is historically advantaged or of high status (Steele & Aronson, 1995). Thus, men, as a high status group, should show stereotype threat when they are presented with the negative stereotype that men are not as socially sensitive as women, an ability ordinarily operationalized as the ability to decode others' nonverbal cues.

In general, there are many demonstrations of performance decrements when members of a stereotyped group are reminded of or threatened by a negative stereotype. For example, women were threatened by gender stereotypes on math ability tests (e.g., Keller, 2002; Spencer, Steele, & Quinn, 1999). African-Americans by racial stereotypes on intelligence tests (e.g., Steele & Aronson, 1995), Latina women by ethnic stereotypes on intelligence and spatial ability tests (e.g., Gonzales, Blanton, & Williams, 2002), and low socioeconomic status individuals by social class stereotypes on intelligence tests (e.g., Croizet & Claire, 1998).

There is also evidence that nonstigmatized, high status, advantaged groups, such as White men, can show performance decrements when threatened by negative stereotypes. These negative stereotypes include the superiority of Asians to Whites in mathematics (e.g., Aronson et al., 1999; Smith & White, 2002) and Blacks to Whites in athletic domains (e.g., Stone, Lynch, Sjomeling, & Darley, 1999). Stereotype threat studies using these stereotypes involved racial comparisons that theoretically have negative consequences for both White men and White women.

There are also a few demonstrations of stereotype threat in men on the basis of gender stereotypes. For example, Leyens, Desert, Croizet, and Darcis (2000) found that men who were reminded that men are not as good as women in processing affective information made more errors in classifying affective than nonaffective words in a lexical decision task than did men who were not reminded of this stereotype. Men have also been threatened on a verbal analogies test (Keller & Bless, 2003) and on a golf-putting task (by the claim that women are better short game players than men are; McConnell, Beilock, Rydell, Jellison, & Carr, 2004).

Although these studies do indeed show that stereotype threat can occur in men, the abilities used by Leyens et al. (2000) and McConnell et al. (2004) were not explicitly shown to have a feminine stereotype. The negative stereotype may not have been well known by the participants and thus may have lacked validity. These abilities are also somewhat specialized because they are not frequently used skills. In addition, Leyens et al. (2000) and McConnell et al. (2004) showed stereotype threat only in particular segments of the population (students specializing in psychology and expert golfers, respectively). Thus, these studies do not make a strong case that threat on the basis of gender stereotypes commonly has negative consequences for a majority of men in the natural environment.

In the current study, we extended the study of stereotype threat in men to a domain that has more obvious daily life consequences. Social sensitivity, or the ability to interpret accurately the expressive behavior of others and to decode others' nonverbal cues, is an aspect of social intelligence or social competence (Archer, Costanzo, & Akert, 2001). Social sensitivity is a skill used by everyone in daily conversation. Thus, stereotype threat on a social sensitivity test has the potential to affect virtually all men.

For stereotype threat to occur on a social sensitivity test, men need to think that women are better at encoding and decoding nonverbal communications than men are. Women are in fact perceived by both men and women to be more fluent, skilled, and involved communicators, whereas men are perceived as louder, more restless, and less skilled (Briton & Hall, 1995). The feminine stereotype for social sensitivity is also to some extent true. Meta-analysis of accuracy in decoding nonverbal cues revealed a mean effect size (d) of 1.02 in favor of women for visual plus auditory modes of communication (Hall, 1978). The test of social sensitivity used in the current study, the Interpersonal Perception Task-15 (Costanzo & Archer, 1993a, 1993b) also typically shows a sex difference (3) that favors women (Costanzo & Archer, 1989, 1993b; Smith, Archer, & Costanzo, 1991; but see Patterson, Foster, & Bellmer, 2001; Patterson & Stockbridge, 1998). Therefore, much like abilities such as mathematics that have been used in past stereotype threat research with women (e.g., Spencer et al., 1999), social sensitivity is an ability on which women stereotypically and actually excel relative to men.