Are anonymous evaluations a better assessment of faculty teaching performance? A comparative analysis of open and anonymous evaluation processes.

OBJECTIVE We compared teaching performance of medical school faculty using anonymous evaluations and open evaluations (in which the evaluator was not anonymous) and examined barriers to open evaluation. METHODS Residents and medical students evaluated faculty using an open evaluation instrument in which their identity was indicated in the evaluation. Following this, they completed anonymous evaluation on the same faculty members. Aggregate outcomes using the two evaluation systems were compared. Outcomes by group of evaluators (residents and students) were analyzed. Trainees were also asked to rate the barriers to the open evaluation process. RESULTS A statistically significant difference between the open and anonymous evaluations was noted across all items, with faculty receiving lower scores on the anonymous evaluations. The mean score for all the items on the open evaluations was 4.45 +/- 0.65, compared to mean score of 4.07 +/- 0.80 on the anonymous evaluations. There was also a statistically significant difference between open and anonymous evaluations in five clinical teaching domains that were evaluated individually. Residents perceived that the three most common barriers to optimal evaluation were an apprehension of possible encounters with the same attending physician in the future, destruction of working relationships with the attending, and a feeling of frustration with the evaluation system. CONCLUSIONS The evaluation of faculty teaching performance is complex. Most academic medical centers use the open evaluation format. This study supports the case for the use of the anonymous evaluation method as a more accurate reflection of teaching performance.