Multiple choice questions are superior to extended matching questions to identify medicine and biomedical sciences students who perform poorly

In recent years, medical faculties at Dutch universities have implemented a legally binding study advice to students of medicine and biomedical sciences during their propaedeutic phase. Appropriate examination is essential to discriminate between poor (grade <6), moderate (grade 6-8) and excellent (grade ≥8) students. Therefore, we compared the discriminatory properties of extended matching questions (EMQs) versus multiple-choice questions (MCQs) and identified the role of sex, age and examination preference on this score. Data were collected for 452 first-year medical and biomedical science students during three distinct course examinations: one examination with EMQ only, one with MCQ only and one mixed examination (including EMQ and MCQ). Logistic regression analysis revealed that MCQ examination was 3 times better in identifying poor students compared with EMQ (RR 3.0, CI 2.0-4.5), whereas EMQ better detected excellent students (average grade ≥8) (RR 1.93, CI 1.47-2.53). Mixed examination had comparable characteristics to MCQ. Sex and examination preference did not impact the score of the student. Students ≥20 years had a 4-fold higher risk ratio of obtaining a poor grade (<6) compared with students ≤18 years old (RR 4.1, CI 2.1-8.0). Given the strong discriminative capacity of MCQ examinations to identify poor students, we recommend the use of this type of examination during the propaedeutic phase of medicine and biomedical science study programmes, in the light of the binding study advice.