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.