Psychometric Properties of Multiple-Choice Questions Used in Undergraduate Medical Education in India: A Systematic Review and Meta-Analysis.

Multiple-choice questions (MCQs) are widely used in Indian undergraduate medical education to assess learning. High-quality MCQs are essential for reliable and valid evaluation, yet real-world psychometric performance across Indian institutions remains unclear. This study aimed to systematically review and analyze the psychometric properties of MCQs used in the assessment of undergraduate medical students to establish normative benchmarks. A systematic search of PubMed, Scopus, Web of Science, and Cochrane Central was conducted to identify studies reporting psychometric indices of MCQs in Indian undergraduate medical education. Eligible studies included those reporting difficulty index, discrimination index, distractor efficiency, internal consistency, or point biserial correlation. Data were extracted independently by reviewers and pooled using random-effects meta-analysis. Heterogeneity and publication bias were assessed. Fifteen studies with 17 datasets were included, covering multiple medical subjects and institutions. The pooled mean difficulty index (dataset = 15) was 51.5 (95% CI: 47.3-55.7%), the discrimination index (dataset = 14) was 0.29 (95% CI: 0.25-0.34), and distractor efficiency (dataset = 12) was 61.7% (95% CI: 45.5-77.9%). There was high heterogeneity between studies. However, no significant publication bias was detected for any psychometric parameter. This study provides the first comprehensive synthesis of MCQ performance in Indian medical education, highlighting moderate difficulty, generally acceptable discrimination, and wide variability in distractor effectiveness. These findings offer empirical evidence to establish normative benchmarks for supporting policy-making to enhance the quality of MCQ, aligned with the national average.