ABSTRACT
PURPOSE: This study examined medical students' stress and certification exam preparation practices in a reformed clerkship curriculum that excluded high-stakes knowledge testing from end-of-rotation performance evaluation. METHOD: Stress and exam preparation practices were assessed via a survey comprising locally developed questions, three subscales of the Motivated Strategies for Learning Questionnaire, and two subscales of the Medical Student Stressor Questionnaire. The association between stress, learning self-regulation, and certification exam scores was evaluated retrospectively using non-parametric tests of association (Spearman's rho). RESULTS: Forty students responded to the survey and consented to use of academic performance data (57% participation rate). Mean certification exam scores were indistinguishable from historical controls. Exam preparation practices resembled those of pre-clinical students: exam-related worrying and time devoted to studying were high, increasing as the exam drew near; preferred study resources were directly analogous to exam questions; and study involved relatively few generative strategies (e.g., concept mapping). Sustaining effort and creating time and space to study were associated with better exam performance, as was participation in this study. DISCUSSION: On the surface, the absence of regularly spaced, high-stakes testing from clerkship performance evaluation appears to "do no harm" to students' certification exam scores. Students already performing better academically may excel due in part to effective learning self-regulation strategies. However, a clerkship curriculum that does not scaffold self-regulation via cumulative knowledge assessment could further disadvantage students already earning lower scores. Evaluating the impact of curriculum reforms should continuously examine changes to learners' experience in context.
