Gender bias in postgraduate year one pharmacy letters of recommendation

Abstract

AbstractIntroductionLetters of recommendation (LORs) are important in pharmacy residency applications. Applicant gender, among other variables, may result in implicit biases that could impact LORs and/or residency attainment. This study hypothesized that LORs have linguistic gender differences in pharmacy residency applications.ObjectivesThe primary objective was to determine if gender‐linguistic differences exist in applicants' LORs. Gender was assessed via the pronouns utilized within the applications. The secondary outcome was to assess LOR linguistic and demographic differences between candidates who did and did not receive interviews.MethodsThis was a multiyear, multicenter study involving postgraduate year one (PGY1) applicants to participating pharmacy residency programs. Demographic data were extracted using the PhORCAS (Pharmacy Online Residency Centralized Application Service) WebADMIT (Admissions Management System) portal, and LORs were analyzed by validated linguistics processing software.ResultsA total of 7529 LORs and 2383 applicants (28.5% men, 71.5% women) were included. Women candidates had higher mean number of awards (4.71 vs. 4.1, p = 0.001) and leadership positions (4.87 vs. 4.48, p = 0.019). Compared with men candidates, women had statistically significantly higher levels of clout (p < 0.001), positive emotion (p = 0.01), social processes (p < 0.001), prosocial behavior (p = 0.002), and social referents (p < 0.001). Women also had lower authenticity compared with men candidates (p < 0.001). Two thousand and one hundred twenty applicants included in the secondary analysis found no difference in offer to interview between women and men candidates (odds ratio [OR] 1.173 [95% confidence interval (CI) 0.895–1.57], p = 0.247).ConclusionsMen and women applicants' LORS differed in specific linguistic variables, although offer to interview was not significantly different based on gender. LOR writers and programs should consider implicit biases that could affect residency offers to interview.