Reliability and Accuracy of Surgical Resident Peer Ratings

Abstract

The reliability and accuracy of peer ratings by general surgery residents were examined. Thirty-two residents anonymously and voluntarily evaluated their peers on ten areas of perfor mance in 1988; twenty-eight residents were evaluated in 1989, and thirty-three were evaluated in 1990. Resident peer ratings were found highly reliable, with no evidence of serious restriction of range or leniency effects. The results indicated a high level of test-retest reliability replicated across three academic years. Halo effects appear to pose the greatest threat to rater accuracy. There was some evidence that chief residents exhibit less halo effect than junior residents when rating peers.