Effect of number of options in EMQs on the discrimination abililty, difficulty level and solving time of test

Date of Award

2017

Document Type

Thesis

Degree Name

Master of Health Professions Education (MPHE)

Department

Educational Development

Abstract

To see the effect of number of options in EMQs on discrimination ability, difficulty level and solving time of test. Settings: Liaquat National Hospital Medical College Methodology: One hundred fourteen EMQs of final year MBBS level were developed. Each EMQ were developed with three versions with eight, ten and twelve options. Test was administered to students via computer software. Each student has to solve thirty-eight EMQs with eight options, thirty-eight EMQs with ten options and thirty-eight EMQs with twelve options. Data Collection and Analysis: Students responses and solving time for each EMQ was recorded with custom made computer software. Item discrimination and item difficulty indices were calculated for all three versions of EMQs separately and were compared with ANOVA on SPSS version 20. Results: Total 114 students were participated in the study. Crohnbach's alpha was more than 0.9 for each of the three tests. Items with ten options had better discrimination index than items with eight options, while there was no significant difference found between ten and twelve options. There was no significant correlation between number of options and difficulty index. Average solving time for twelve-option version was 76 seconds, for ten-option version was 73.7 seconds and for eight-option version was 73.4 seconds. Conclusion: Ten option EMQs had a higher discriminatory index than eight option versions but there was no statistical significant difference between ten option and twelve option version. EMQs with more options need more solving time. Key Words: EMQ, Difficulty index, Discrimination index, Response time.

This document is available in the relevant AKU library

Share

COinS