Student-written single-best answer questions predict performance in finals

Abstract

Single-best answer (SBA) questions are widely used for assessment in medical schools; however, often clinical staff have neither the time nor the incentive to develop high-quality material for revision purposes. A student-led approach to producing formative SBA questions offers a potential solution. Cardiff University School of Medicine students created a bank of SBA questions through a previously described staged approach, involving student question-writing, peer-review and targeted senior clinician input. We arranged questions into discrete tests and posted these online. Student volunteer performance on these tests from the 2012/13 cohort of final-year medical students was recorded and compared with the performance of these students in medical school finals (knowledge and objective structured clinical examinations, OSCEs). In addition, we compared the performance of students that participated in question-writing groups with the performance of the rest of the cohort on the summative SBA assessment. Often clinical staff have neither the time nor the incentive todevelop high-quality material for revision purposes Performance in the end-of-year summative clinical knowledge SBA paper correlated strongly with performance in the formative student-written SBA test (r=~0.60, p<0.01). There was no significant correlation between summative OSCE scores and formative student-written SBA test scores. Students who wrote and reviewed questions scored higher than average in the end-of-year summative clinical knowledge SBA paper. Student-written SBAs predict performance in end-of-year SBA examinations, and therefore can provide a potentially valuable revision resource. There is potential for student-written questions to be incorporated into summative examinations.