Peer and Self-assessment of Professionalism in Undergraduate Medical Students at the University of Calgary

Résumé

Background: Peer and self assessment processes are integral to the development of professional behaviours. The purpose of this study was to assess the Rochester Peer Assessment Tool (RPAT) among a group of volunteer first year students.

Methods: We assessed feasibility through participation rates. The evidence for the validity of instrument scores was ascertained through an exploratory factor analysis, MANOVA to determine age and gender differences, and a discrepancy analysis between the self and peer data. Reliability analyses included the Cronbach's alpha analysis and G- and D-studies. Students completed a feedback questionnaire to provide data about acceptability.

Results: Self and peer data were collected for 46 and 44 students, respectively. Each student had a mean of 7.2 peer assessments (out of a possible 8).  The factor analysis identified two factors, interpersonal skills and work study habits. The discrepancy analysis showed students in the lowest/highest quartiles, as assessed by peers, had higher/lower self means than peer means. The G-coefficient was Ep² = 0.77. Student feedback was positive.

Conclusions: RPAT was feasible in our setting, was acceptable to the students, and has been adopted as a mandatory part of our program for first and second year students. The study added to the evidence base for the reliability and validity of the RPAT instrument scores as a method of assessing professional behaviours.