Claiming the Validity of Negative In-group Stereotypes When Foreseeing a Challenge: A Self-handicapping Account

Abstract

This research proposes a self-handicapping process in which people proactively endorse negative in-group stereotypes when there is the prospect of failure in a task. In Experiment 1, we found that women were more likely to endorse the math-gender stereotype stigmatizing their gender group when they anticipated a difficult versus easy math task. In Experiment 2, the same pattern was observed among men stigmatized with relatively poor verbal skills. In Experiment 3, we found that such a self-handicapping tendency was most prominent among individuals with high trait self-esteem, who are presumably more motivated to maintain self-esteem versus those with low trait self-esteem. All together, these results suggest that endorsing negative in-group stereotypes can be used as an anticipatory coping mechanism, occurring even before receiving failure feedback in the presence of a high risk of failure.

Keywords:

• Self-esteem
• Self-handicapping
• Stereotype
• Stereotype endorsement

Notes

1. In Experiment 2, women did not lower their endorsement for the verbal-gender stereotype when faced with a difficult verbal task. That is, unlike the findings in Experiment 1 (e.g., men for superior math skills), in Experiment 2, people whose group was associated with positive traits (e.g., women for superior verbal skills) did not self-handicap by denying the positive in-group stereotype. The first possible way of understanding the difference in the results could be based on gender differences in self-handicapping tendencies. As noted in previous research, men are generally more likely to self-handicap prior to a threat of failure than women (Berglas & Jones, 1978; Harris & Snyder, 1986; Smith, Snyder, & Handelsman, 1982). Thus, men could have more readily engaged in self-handicapping behavior by denying their positive in-group stereotype (i.e., the math-gender stereotype) relative women, who had the same opportunity in Experiment 2. In addition, this difference might be indicative of a differential threshold of denying positive traits between the genders (the math skills of men vs. the verbal skills of women), given different social status. As McGuire and Padawer-Singer (1976) noted, the features of the self that are unique and rare tend to become particularly salient features of one's self-concept. Given the relatively low status among women, denying one of a smaller number of positive traits would not be as easy, compared to men, who possess a variety of positive features in addition to superior math skills. Thus, men, presumed to be of high social status, may carry a higher readiness to self-handicap by denying positive in-group stereotypes than women. We thank a reviewer for this perspective. Overall, future research on gender differences in self-stereotyping or a comparison of high- and low-status groups regarding the propensity to accept or deny positive or negative in-group stereotypes will greatly enhance our understanding as to how individuals strategically use stereotypes as a coping mechanism against future threat.

2. This pattern is not the same as the one obtained in Experiment 1. The stereotype endorsement of male participants is not a main focus of the present research. However, these results seem to echo the notion that men are not the stigmatized target of the math-gender stereotype. As shown in other research (Johns et al., 2005; Steele & Aronson, 1995), superior out-group members (e.g., White or male participants) are not as sensitive to threats posed by stereotypes as are the members of stigmatized groups (e.g., Black or female participants). At the same time, members of “superior” groups, according to certain stereotypes, tend to be differentially influenced by stereotypes (see Shih, Ambady, Richeson, Fujita, & Gray, 2002, for diverse impacts of stereotypes; Vick, Seery, Blascovich, & Weisbuch, 2008). In fact, women, a “superior” group along the verbal-gender stereotype, did not exhibit self-handicapping patterns in Experiment 2, either. Taken together, members of groups who are not stigmatized seem to be less sensitive toward an upcoming threat to the self, and are therefore less likely to strategically endorse or deny stereotypes.

3. We acknowledge that it would be a better design to invite both male and female subjects to participate in the study. However, unfortunately because of limited funding, we could only afford to invite female participants, who would allow us to test our crucial predictions.

4. The low response rate could be attributed to the fact that the survey was conducted during examination week, and that the participants had to come to the department office to pick up the cash after they had participated in the online survey. We suspect that such time and location inconvenience lowered the response rate.

Additional information

Notes on contributors

Kyoungmi Lee

All authors contributed equally to this manuscript. Portions of this research were presented at the Society for Personality and Social Psychology Conference, Tampa, Florida, February 2009.