Use of Item analysis to improve quality of Multiple Choice Questions in II MBBS

Item analysis helps in assessing the quality of the multiple choice items and of the test as a whole.

To analyze the quality of MCQs & to determine internal consistency reliability of test.

To evaluate perception of teachers towards item analysis.

In a cross-sectional study, total 40 items from Microbiology First Continuous Evaluation Test of 83 students of 2^nd year M.B.B.S. were analysed. Item & test analysis was done by calculating Difficulty index (DIF I), Discrimination index (DI), Distractor effectiveness (DE), Point Bi-serial correlation (PBS) and Cronbach's Alpha. Feedback of faculties was collected on a 5 point Likert scale.

Difficulty index of 12(30%) & 18(45%) items were in the ideal (50–60%) & acceptable range (30–70%) respectively, 7(17.5%) items were easy (>70%) and 3(7.5%) items were difficult (<30%). Discrimination index of 17(42.5%) items was excellent (=>0.36), 7(17.5%) items was good (0.25–0.35), 1(2.5%) item was acceptable (0.20–0.24) and 15(37.5%) items was poor (=<0.20). Out of 15 items with poor DI, 6 (40%) items were of easy and difficult level of DIF I and 3 items (20%) were poorly framed but rest of the items were appropriately framed testing higher level of cognitive domain. Out of 120 distractors analysed, 102(85%) were functional & 18(15%) were non-functional. Thirty (75%) items had acceptable PBS values (>0.15). Cronbach's Alpha improved from 0.702 to 0.794 on removal of 15 items with poor DI. Most of the faculties found item analysis useful to improve quality of MCQs.

Majority of the items had acceptable level of difficulty & discrimination index. Most of distractors were functional. Item analysis helped in revising items with poor discrimination index and thus improved the quality of items & a test as a whole.