1Assistant Professor, Department of Anatomy, Birat Medical College and Teaching Hospital, Biratnagar, Nepal
2Assistant Professor, Department of Anatomy, Chitwan Medical College, Bharatpur, Nepal, shekhar_np1@hotmail.com
3Lecturer, Department of Statistics, Birendra Multiple Campus, Bharatpur, Nepal, dhunganagovindana2012@gmail.com
*Corresponding author email id: *arund@mail.ru
Online published on 14 August, 2018.
To evaluate multiple-choice questions (MCQs) and structured short-answer questions (SSAQs) items in human anatomy by determining their discrimination and difficulty index, and performing ROC (receiver operating characteristics) analysis to signify their ability to predict and discriminate poor and good performance of students.
First and second-year medical students of 2016 were followed through 1 year to register their internal examination performance (SSAQs and MCQs) of various items in anatomy. Responses were analysed and evaluated for performance.
A total of 30 SSAQs and 100 MCQs of five items were analysed. Thedifficulty index of MCQ and SSAQ was 0.36 and 0.38, respectively, which is considered optimal for both. SSAQ showed higher discrimination index (0.46-considered very good item) than MCQ (0.29-marginal item and subject to improvement), suggesting SSAQ a better tool to discriminate poor and good students than MCQ.
This study suggests that though difficulty index of SSAQ and MCQ are similar, and in terms of discriminating good and poor students SSAQ was a better tool, ROC analysis revealed both the tools to be adequate in terms of discriminating students with a relatively better value to SSAQ. As this study was conducted in a university where SSAQs are the only form of evaluation of theoretical knowledge in anatomy, conducting similar study where both the tools are followed is highly recommended.
Predict, Assessment, MCQ, SSAQ, Human anatomy, Examination, Performance
