Frequency and determinants of residents' narrative feedback on the teaching performance of faculty: narratives in numbers.

Purpose: Physicians involved in residency training often receive feedback from residents on their teaching. Research shows that learners value narrative feedback, but knowledge of the frequency and determinants of narrative feedback in teaching performance evaluation is lacking. This study aims to identifythe frequency with which residentsgave positive comments and suggestions for improvement to faculty, and the factors influencing that frequency.

Method: From September 2008 through May 2010, the authors collected data, using a validated formative feedback system (System for Evaluation of Teaching Qualities). The authors used univariate and multivariable analysis to investigate the associations between participants' characteristics, including faculty members' teaching performance, and the frequency of the two types of narrative comments.

Results: In total, 659 residents (79% of 839) completed 6,216 evaluations on 917 faculty (95% of 964), resulting in 11,574 positive comments and 4,870 suggestions for improvement. On average, faculty members received 13 positive comments and 5 suggestions for improvement. Multivariable analysis showed that higher teaching performance was associated with higher numbers of positive comments (regression coefficient 0.538; 95% confidence interval: 0.464 to 0.613) and with lower numbers of suggestions for improvement (-0.802; -0.911 to -0.692), both P < .0001. Nonacademic hospitals, participation in teacher training, and female residents' evaluation were statistically significant determinants of receiving more narrative feedback.

Conclusions: Residents provided narrative feedback that paralleled and elaborated on quantitative evaluations they provided; therefore, faculty would be wise to attend to narrative feedback. Analysis of the quality of narrative feedback is needed to understand its effectiveness.