Analysis of Question Items Using the Differentiating Power Method

Fariz Zakaria (Amikom Yogyakarta)
Wasis Haryo Sasoko (Faculty of Informatics Engineering, Amikom University)
Ema Utami (Faculty of Informatics Engineering, Amikom University)

International Journal of Informatics, Information System and Computer Engineering (INJIISCOM)
Vol. 6 No. 1 (2025): INJIISCOM: VOLUME 6, ISSUE 1, JUNE 2025

Multiple-choice questions (MCQs) remain popular in education due to their assessment efficiency, perceived objectivity, and capacity to provide rapid feedback. However, ensuring the quality of MCQs requires an in-depth item analysis to maintain assessment validity, reliability, and fairness. This study employs a quantitative descriptive method to evaluate the Differentiating power of MCQ items from an exam dataset. The analysis reveals that a vast majority of the test items require revision, while a smaller portion must be completely rejected. The findings suggest that rejected items should be discarded, whereas items needing correction must be improved to enhance the overall exam quality. Evaluating question quality is crucial for maximizing assessment effectiveness, particularly in high-stakes higher education environments