The Presence of Gender Bias in Letters of Recommendations Written for Urology Residency Applicants

OBJECTIVE To determine whether implicit gender bias exists in the urology residency application process, we compared linguistic differences in letters of recommendation (LOR) submitted for male and female applicants. METHODS LOR were abstracted from residency applications to a urology residency program. Linguistic Inquiry and Word Count, a validated text analysis software program, characterized the linguistic content of the letters. Analyzed letters were compared according to gender of the applicant using multivariable analysis, examining the association of applicant gender, letter writer, and letter characteristics. Multivariable analysis was also performed to determine the association of letter characteristics with matching into a urology residency. RESULTS Of 460 letters evaluated, letters for male applicants are written in a more authentic tone compared to letters written for female applicants. Letters written for male applicants contain significantly more references to personal drive, work, and power than letters written for female applicants. Significant differences are maintained on multivariable analysis when controlling for race and Step 1 score of the applicant. Letters with references to power were significantly more likely to be associated with an applicant who matched into urology than an applicant who didn't match. CONCLUSION Significant linguistic differences exist among LOR written for men and women applying into urology, suggesting that gender bias may permeate resident recruitment. These differences may affect the likelihood of women matching into urology. (C) 2019 Elsevier Inc.