An analysis of sex differences in computing teaching evaluations

Anonymous student teacher evaluations are commonly used to evaluate the quality of computing instructors at the university level. However, such teaching evaluations are subject to gender and sex-based biases, calling into question their utility and scope. In this paper, we first use data from a large public American university to replicate previous findings showing that significant sex-related differences persist in computing teaching evaluations. Intriguingly, we find that the sex-differences in computing teaching evaluations are primarily driven by bias involving professors, while significant sex-based differences for student-instructors are not observed. Finally, we place the magnitude of the sex-based differences we observe into a broader engineering context.