‘Sorting’ Out Gender Discrimination and Disadvantage: Evidence from Student Evaluations of Teaching

How should gender discrimination and systemic disadvantage in teaching evaluations be addressed when more discriminatory and less generous students sort into fields, courses, and instructors’ sections? We estimate student-level measures of gender bias and generosity by comparing how the same student rates male and female instructors, controlling for professor fixed effects. Accounting for measurement error, we document substantial heterogeneity in bias and generosity across students. Bias varies systematically by student gender, and sorting patterns disadvantage female faculty in some courses and male faculty in others. We propose Empirical Bayes–inspired adjustments that correct evaluations for predictable disadvantages caused by student sorting.