Abstract:

Improving student performance on exams, particularly in open-ended (OE) questions compared to multiple-choice tests, involves a combination of instructional strategies, assessment design, and targeted analysis. Students claim that TF questions compromise their success in the final assessment. For this reason, our objective is to analyse the qualifications of TF tests compared to OE questions because it can provide valuable insights into student performance and preferences.

To achieve our goal, we have analysed the results of the last five academic courses (2017-18 to 2022-23) of the subject Physiology and Pathophysiology I, a second-year core subject of the Pharmacy degree of the Universitat de Barcelona. Student performance was assessed through three exams (A1: quarter-term partial exam, A2: mid-term partial exam; A3: complete final exam) that included both TF and OE questions, which were distributed through the semester. To compare the scores across exams and years, the Pearson correlation coefficient was studied using the SPSS® program (SPSS Inc., Chicago, IL, USA).

The number of subjects analysed ranged from 330 to 430 depending on the year. In the case of A1 and A2, in general we found better scores for OE questions being the average rating 0.2 to 2.0 points higher than that of test scores, and considering that the maximal qualification was 10. In contrast, students presented better performances for the TF questions in the case of A3. When we studied the correlation between the scores obtained in OE and the TF questions, we found positive correlations in all cases. Interestingly, in the case of A1 and A2, approximately a 5% of students did not answer some of the OE questions proposed while this percentage increased up to 20% when analysing A3 results.

In general, both types of questions are equivalent for evaluating the knowledge of the students, indicating some level of consistency in student performance. However, at the final assessment (A3), a large number of students only answered the TF questions and did not to attempt the OE questions, suggesting differing levels of engagement or difficulty with these question types.

In conclusion, although both question types have their merits, there are differences in student performance and behaviour that warrant attention when designing assessments and interpreting results.

Keywords:

Student assessment, open-ended questions, true/false questions.