Risk Factors Associated with Poor Computer-based Testing (CBT) Scores - Comparing Students' Performance "Without/With COVID-19" and Backgrounds.

ABSTRACT

Objective: The present study compared students' CBT scores during the 2-year period before ("without COVID-19") and 2-year period during ("with COVID-19") the COVID-19 pandemic, and analyzed factors associated with poor results. Materials A total of 530 students (368 males and 162 females), who had taken CBT within the period between 2018 and 2021. Methods: Analysis was performed based on the questionnaire results, and the students' performance was compared between "without/with COVID-19" to identify the causes of poor CBT scores. Results: The overall mean IRT score was 515.5±85.4. The without and with COVID-19 groups' scores were 495.7±85.9 and 534.4±80.8, respectively (p<0.01). Among all students, 43 (8.1%) had IRT scores lower than 400 as poor CBT scores; 27 (10.4%) without and 16 (5.9%) with COVID-19, revealing a decrease in the latter. The multivariate analysis of the risk of students having poor CBT scores showed that students with poor performance during the third year (odds ratio7.02), starting preparation for CBT late (2.19), and not taking any practice examination (4.58) are more likely to have poor CBT results. Conclusions: Due to the COVID-19 pandemic, students spent more time on online home study, and this may have consequently improved their CBT scores. Such learning performance is desirable for medical students, but they have lost the opportunity to gain valuable experiences that they could have acquired through extracurricular activities, such as club activities. In this respect, we cannot simply be pleased by the improvement in students' CBT scores.