Gender Effects in Assessment of Clinical Teaching: Does Concordance Matter?

Abstract

Gender bias is thought to exist in the assessment of clinical teachers, yet its extent in different specialties is not well-documented nor has it been studied at the individual-dyadic level. The authors sought to determine whether gender bias exists in residents' assessments of faculty teaching in 3 clinical departments, and if present, whether this is influenced by gender concordance or discordance between the faculty and resident. Residents' ratings of faculty in internal medicine (800 faculty, 5753 ratings), surgery (377, 2249), and family medicine (672, 3438) at the University of Toronto from 2016-2017 were analyzed using the overall global rating on a 5-point scale. A mixed-effects linear regression analysis accounted for nesting of ratings within each faculty member. Overall scores of teaching effectiveness showed a strong skew to favorable ratings for all faculty and a ceiling effect. However, gender effects differed across departments. In internal medicine (38.5% female faculty), no significant gender effects were detected. In surgery (16.2% female) and family medicine (53.0% female), male faculty received significantly higher scores than female faculty. In surgery this was driven by male residents giving male faculty higher ratings (4.46 vs 4.26, P < .001). In family medicine this was driven by male faculty receiving higher ratings regardless of resident gender (4.65 to 4.57, P < .001). Although effects were very small and inconsistent, with gender concordance mattering only for one department, it suggests that gender is a meaningful source of variance in teaching assessments.