Anticipated Discrimination and Gender Stereotypes in Job Applications - Experiment 3

A well-established literature documents that women are, on average, less likely than men to enter competitive environments. This gender gap may be exacerbated by beliefs about biased evaluation of competitive performance. Most existing experimental studies documenting gender differences in competition entry assume that performance is evaluated objectively based on a measurable ranking. However, in many real labour-market settings, evaluation is, to some extent, subjective and carried out by human assessors who may hold stereotypes or exhibit biases. Such subjectivity can generate anticipated discrimination, which may further distort competition-entry decisions. Building on a previous study of ours in which we measure how workers adjust their willingness to apply to competitive jobs by job domain and mandatory gender revelation, this experiment measures the actual evaluation decisions of human assessors in a male-stereotyped mathematics task. These assessors, in the role of managers, assess groups of four workers and choose the worker they predict will perform for the best in a future task. To make their decision, managers are given worker-specific signals of this future performance and an avatar of each worker which is either neutral or displays their gender. In the role of recruiters, assessors are tasked with selecting a worker who they believe a manager will select. In this sense, the recruiter acts as a third party decision maker that must accommodate to their belief of another's preferences.