Multiple Choice Questions with Different Numbers of Options in University Putra Malaysia Undergraduate Medical Program: A Comparative Analysis in 2017 and 2018

Abstract

Background: Multiple-choice questions (MCQs) are used for measuring the student’s progress, and they should be analyzed properly to guarantee the item’s appropriateness. The analysis usually determines three indices of an item; difficulty or passing index (PI), discrimination index (DI), and distractor efficiency (DE). Objectives: This study was aimed to analyze the multiple-choice questions in the preclinical and clinical examinations with different numbers of options in medical program of Universiti Putra Malaysia. Methods: This is a cross-sectional study. Forty multiple-choice questions with four options from the preclinical examination and 80 multiple-choice questions with five options from the clinical examination in 2017 and 2018 were analyzed using optical mark recognition machine and Ms. Excel. The parameters included PI, DI, and DE. Results: The average difficulty level of multiple-choice questions for preclinical and clinical phase examinations were similar in 2017 and 2018 that were considered ‘acceptable’ and ‘ideal’ ranged from 0.55 to 0.60, respectively. The average DIs were similar in all examinations that were considered ‘good’ (ranged from 0.25 to 0.31) except in 2018 clinical phase examination that showed ‘poor’ items (DI = 0.20 ± 0.11). The questions for preclinical phase showed an increase in the number of ‘excellent’ and ‘good’ items in 2018 from 37.5% to 70.0%. There was an increase of 10.0% for preclinical phase, and 6.25% for clinical phase, in the number of items with no non-functioning distractors in 2018. Among all, preclinical multiple-choice questions in 2018 showed the highest mean of DE (71.67%). Conclusions: Our findings suggested that there was an improvement in the questions from preclinical phase while more training on questions preparation and continuous feedback should be given to clinical phase teachers. A higher number of options did not affect the level of difficulty of a question; however, the discrimination power and distractors efficiency might differ.