Differences in colorectal residency applicant evaluations by evaluator role and residency training environment
Related Experiment Videos
Allie Jin1, Molly Leonard2, Kayvan Barekatain3
1Division of Colorectal Surgery, University of Louisville, Hiram C. Polk Department of Surgery, Louisville, KY. Electronic address: https://twitter.com/alliejin.
Background:
Standardized letters of evaluation are central to colorectal surgery residency selection; yet, variability in evaluator scoring patterns and differences in applicant evaluation by residency training environment remain insufficiently understood. This study evaluated differences in applicant ratings based on residency program type and evaluator role.
Methods:
A retrospective review of colorectal surgery residency applications submitted to a single academic institution across 3 application cycles (2023, 2024, and 2025) was performed. Applicant characteristics, residency training environment (academic versus community), evaluator characteristics, and domain ratings were abstracted. Nonparametric comparisons were performed using Mann-Whitney U tests.
Results:
A total of 342 applications and 1,287 letters of recommendation were analyzed. Across all cycles, ratings were heavily clustered within the top 2 performance categories across domains. Colorectal surgery program directors consistently demonstrated more conservative scoring patterns, particularly in technical ability and select cognitive domains; however, differences were domain dependent and not uniformly significant across all categories. Applicants from academic programs generally received higher ratings across multiple domains compared with community programs, with statistically significant differences observed in select measures, though many domain-level comparisons were not significant. Despite these differences, substantial overlap in score distributions was observed between groups.
Conclusion:
Colorectal surgery program directors apply systematically more conservative scoring than other evaluators in performance-based domains. This pattern has direct implications for how standardized letters are interpreted during selection. Applicant training environments contribute to smaller, domain-dependent differences. However, high ratings remain common across all groups, limiting discrimination in applicant assessment.
