Arika Hoffman1, Rachel Ghoubrial2, Melanie McCormick2, Praise Matemavi2, Robert Cusick3. 1. University of Nebraska Medical Center S 42nd St & Emile St, Omaha, NE, 68198, USA. Electronic address: arika.hoffman@unmc.edu. 2. University of Nebraska Medical Center S 42nd St & Emile St, Omaha, NE, 68198, USA. 3. Children's Hospital and Medical Center 8200 Dodge St, Omaha, NE, 68114, USA.
Abstract
BACKGROUND: Letters of recommendation (LoR) are considered one of the most important predictors of matching into a pediatric surgery fellowship. We determined if gendered differences exist in LoR written for resident candidates. METHODS: A retrospective review of blinded LoR to a fellowship program between 2015 and 2017 was performed. RESULTS: Of the 364 LoR reviewed for 49 female and 48 male applicants, male surgeons wrote 82.5% of letters. Male LoR contained more agentic terms (p = 0. 042), first name occurrences (p = 0.0082) and phrase "future success" (p = 0.015). Female letters included more socio-communal phrases (p = 0. 010) and 5% referenced a spouse's accomplishments vs. 0% of male letters. Male LoR contained more active possessive language (p-0. 027); ie: "he published", "he presented". We found no difference in an applicant's research experience (p = 0.06) or leadership qualities (p = 0. 067). CONCLUSION: Gender differences exist in LoR written for fellowship applicants applying to a highly competitive subspecialty.
BACKGROUND: Letters of recommendation (LoR) are considered one of the most important predictors of matching into a pediatric surgery fellowship. We determined if gendered differences exist in LoR written for resident candidates. METHODS: A retrospective review of blinded LoR to a fellowship program between 2015 and 2017 was performed. RESULTS: Of the 364 LoR reviewed for 49 female and 48 male applicants, male surgeons wrote 82.5% of letters. Male LoR contained more agentic terms (p = 0. 042), first name occurrences (p = 0.0082) and phrase "future success" (p = 0.015). Female letters included more socio-communal phrases (p = 0. 010) and 5% referenced a spouse's accomplishments vs. 0% of male letters. Male LoR contained more active possessive language (p-0. 027); ie: "he published", "he presented". We found no difference in an applicant's research experience (p = 0.06) or leadership qualities (p = 0. 067). CONCLUSION: Gender differences exist in LoR written for fellowship applicants applying to a highly competitive subspecialty.