What Should We Write on End of Rotation Evaluations? A Mixed-methods Exploration of General Surgery Resident and Faculty Perspectives

BACKGROUND: Narrative comments on resident end-of-rotation (EOR) evaluations are essential to competency-based assessment in surgical training, yet residents often perceive them as poor quality and variably useful. We explored resident and faculty perspectives on satisfaction with EOR comments, what constitutes high-quality comments and barriers to producing them. METHODS: This is a sequential mixed-methods study at a single academic general surgery residency. Semi-structured interviews with residents and faculty were thematically analyzed using directed content analysis guided by the Relationships, Autonomy, Mastery, and Purpose (RAMP) framework. Findings informed development of resident and faculty surveys assessing perceived purpose, content, barriers, and satisfaction; responses were summarized using medians and interquartile ranges, and dichotomized Likert-scale responses were compared between residents and faculty using chi-square tests. RESULTS: Of 9 faculty and 8 residents completed interviews, identifying high-quality comments as specific, actionable, balanced, individualized, behaviorally anchored, timely, and purpose-driven. Faculty described relational risk and perceived repercussions as drivers of self-censorship (relationships and autonomy); residents emphasized limited autonomy due to delayed, 1-way feedback (autonomy). Both groups reported feedback quality was often poor (mastery) and did not have a consensus of the goal of EOR comments (purpose). Of 22 residents and 22 faculty (39% response rate each) completed surveys. Most residents and faculty agreed EOR comments are important for resident growth (73% and 68%) and progress (73% and 63%). However, most (86% residents; 78% faculty) reported receiving/writing high-quality comments half the time or less. 95% of residents reported suggestions for improvement were present half the time or less, while 33% of faculty reported including suggestions most of the time/always (p = 0.02). CONCLUSIONS: Although both residents and faculty value EOR narrative comments, there is a gap between desired and actual comments. A motivation-informed (RAMP) framework highlights relational and autonomy constraints that may limit candor and actionability.