Evaluating Gender Biases on Actual Job Performance of Real People: A Meta‐Analysis¹

Abstract

This study examined gender bias on job performance in work settings where confounding variables (e. g., organizational level, experience, education) were cautiously taken into consideration to ensure fair comparisons. Although previous meta‐analyses examined gender biases on evaluations, findings in tightly controlled laboratory environments may differ from those in highly complicated field studies. We found little evidence of overall gender bias in performance appraisals in nonconfounded field studies. However, there were significant pro‐male biases when only men served as raters. Measure‐specific gender stereotypicality, instead of genera! stereotypicality about the job, produced gender bias in performance appraisal. Masculine measures produced pro‐male bias, and feminine measures produced pro‐female bias.

Role congruity theory of prejudice toward female leaders.

“…Second, female leaders became less effective relative to male leaders as the proportion of male subordinates increased, perhaps reflecting male (vs. female) subordinates' construal of leadership in more masculine terms and greater approval of traditional gender roles. Third, the greater the proportion of men among the raters whose data produced the measures of effectiveness (these raters were not necessarily leaders' subordinates), the less was the effectiveness of women relative to men (see Bowen, Swim, & Jacobs, 2000, for a similar effect). Fourth, women were substantially less effective than men in military organizations, a traditionally masculine environment, but modestly more effective than men in organizations in the domains of education, government, and social service.…”

An Investigation of the Underlying Mechanism of Discrimination: The Prototype Matching Model

“…For example, future research may explore whether evaluator-applicant similarity on discriminable characteristics affects the stereotypes and prototypes evaluators use when evaluating applicants. Research suggests that rater x ratee interactions can affect the magnitude of discrimination (Bowen et al, 2000), and these interactions may also be relevant to predicting evaluators' stereotypes and prototypes. In applying the prototype matching model, it is also important that future research explore the role of contextual factors such as how prototypes differ between jobs other than sales and customer service, and how the prototype matching model generalizes to work-related settings beyond that of selection.…”

“…However, meta-analyses of field research have also shown an advantage for females in Two points become salient from the lab-and field-based sex discrimination research described above. First, the practical significance of sex difference findings has typically been described as small (i.e., Bowen et al, 2000;Davison & Burke, 2000;Dean et al, 2008;Olian et al, 1988;Roth et al, 2008;Swim et al, 1989). However, these seemingly small practical differences have been argued to have greater long term effects (Agars, 2004;Eagly, 1995).…”

“…This has led to multiple meta-analyses of the lab-based research (Bowen, Swim, & Jacobs, 2000;Davison & Burke, 2000; Olian, Schwab, & Haberfield, 1988;Swim, Borgida, Maruyama, & Myers, 1989), which have consistently shown a small A sizeable body of evidence comparing males and females also comes from field samples.…”

“…However, existing research has yet to explicitly test this process model. Instead, the theory has been based on the integration 11 of research findings using group membership (e.g., male/female, normal weight/overweight) as a predictor of (a) stereotypical attributes (Crocker et al, 1993;Harris, Harris, & Bochner, 1982;Koenig, Eagly, Mitchell, & Ristikari, 2011;, and/or (b) evaluative outcomes for specific work roles (Bowen et al, 2000;Davison & Burke, 2000;Dean et al, 2008;Finkelstein et al, 2007;Roehling et al, 2013;Swim et al, 1989). In other words, research has not explicitly tested the mediational hypothesis that possessing certain characteristics (e.g., being female or overweight) leads evaluators to ascribe stereotypical attributes to those individuals, which in turn leads to discrimination.…”