Clerkship evaluation - What are we measuring?

Background: As society's expectations of physicians change, so must the objectives of training. Professional organizations involved in training now emphasize multiplicity of roles. But how well do we evaluate these multiple roles? Aims: To investigate the principal components of evaluation in the Internal Medicine clerkship rotation at the University of Calgary. Methods: We performed factor analysis on all evaluation components in the Internal Medicine clerkship rotation, including the in-training evaluation report (ITER), objective structured clinical examination (OSCE), and multiple choice questions (MCQ) examination. Results: We identified three principal components: information processing, professionalism, and declarative knowledge. Both the OSCE and MCQ loaded on a single factor, declarative knowledge. The nine items on the ITER loaded on two factors-information processing and professionalism. Conclusions: Despite using 11 evaluation items on three tools, we identified only three principal components of evaluation. Both our MCQ and OSCE appeared to measure declarative knowledge. The latter may be due to the fact that we use standardized patients without clinical findings-such that evaluations are primarily based upon the demonstration of examination routines. Reasons for the lack of discriminant validity of our ITER include overlapping attributes and constant errors, including a halo effect and an error of leniency.