Abstract

SARRIAS-RAMIS, R.; MATEU, L.; BAILLES, E.  and  PEREZ, J.. Length (number of questions) and exams results. Educ. méd. [online]. 2010, vol.13, n.3, pp.187-192. ISSN 1575-1813.

Aim. This paper resents two studies about the relationship between the number of questions that appear in an exam and their results. These studies have been done in response to two concrete problems found in two Spanish universities. Subjects and methods. The first study, done at the Faculty of Medicine of the Autonomous University of Barcelona, compared the real results achieved in three subjects which use MCQ-tests with the theoretical results which the students would have achieved if half of the questions had been used. The difficulty of the exams in both situations was practically the same and the student's academic results were very similar in both situations. The second study, carried out at the Faculty of Health and Life Sciences at the Pompeu Fabra University of Barcelona, compared the real academic results in two subjects which use free response questions with the supposed ones with half of the questions of the exam in two subjects which use many essay questions. The results obtained by the students were practically identical. Conclusion. Our general conclusion is that in the evaluation of academic performance, it is more important how relevant and representative the questions are than the number of them.

Keywords : Academic assessment; Assessment methods; Length of exams.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )