Publication:

Item analysis of multiple choice questions from an assessment of medical students in Bhubaneswar, India

Background Multiple choice questions (MCQs) are usually used to assess students in different educational streams. However, the MCQs to be used should be of quality which depends upon its difficulty index (DIF I), discrimination index (DI) and number of Non-functional distracter (NFD). Objective of the study is to evaluate the quality of MCQs, for creating a valid question bank for future use and to identify the low achievers, whose problems can be corrected by counselling or modifying learning methods. This study was done in Kalinga Institute of Medical Science (KIMS) Bhubaneswar.

Methods A part completion test in the department of pediatrics was done. Total 25 MCQs and 75 distracters were analyzed. Item analysis was done for DIF I and DI and presence of number of NFD.

Results Difficulty index of 14 (56%) items was in the acceptable range (p value 30-70%), 8 (32%) items were too easy (p value >70%) and 2 (8%) items were too difficult (p value 0.35), 3 (12%) items was good (d value 0.20-0.34) and 8(32%) items were poor (d value<0.2%). Out of 75 distracters, 40 (53.4%) NFDs were present in 22 items. 3 (12%) items had no NFDs, whereas 8 (32%), 10 (40%), and 4 (16%) items contained 1, 2, and 3 NFD respectively.

Conclusion Item analysis is a simple and feasible method of assessing valid MCQs in order to achieve the ultimate goal of medical education.