On the importance of transparency and truncation to reduce gender and nationality bias in jury decisions

01/04/2024 to 30/09/2026

Subjective performance evaluations (SPEs) are commonly used by organizations to assess, select, and rank candidates in different competitive situations; for instance, in education for assessing and selecting students, in research for assessing and selecting grant applicants, in the labor market for assessing and selecting job applicants, or in sports for judging the performance of athletes and determining champions. Since many of such SPEs have serious economic consequences for both candidates and organizations, individual or panels of experts are usually appointed to ensure the quality of evaluation. There is, however, substantial evidence for the existence of different biases in SPEs leading to favoritism or discrimination against candidates. In this project, we intend to exploit rich panel data from real competitive environments with high stakes, i.e., professional sports, in order to test whether (i) nationalistic bias might be crowding out gender bias. Moreover, we intend to explore whether and how much (ii) transparency, i.e., the disclosure of judges’ identity, affects these biases by exerting external pressure and control. Likewise, we intend to explore whether and how much (iii) truncation of extreme scores affects these biases by changing the incentive structure for judges.