Developing an Automated Tool for Psychometric Evaluation and Exam Quality Indexing of Multiple-Choice Questions (MCQs): Phase I Study

Abstract

Accurate educational assessment is critical for evaluating student learning and advancing instructional practices. Despite the robustness of Classical Test Theory (CTT) in assessing exam quality, its statistical complexity often limits its practical application by educators. This study presents the design and implementation of an innovative bilingual software tool (English/Arabic) that automates psychometric analysis of multiple-choice examinations. Developed using Python and leading data science libraries, the tool streamlines the computation of essential metrics such as item difficulty, discrimination indices, KR-20 reliability, and distractor efficiency. These parameters are synthesized into an intuitive Exam Quality Index (EQI), accompanied by automated narrative interpretations and targeted recommendations. This user-friendly application supports educators in enhancing assessment quality, promoting fairness, and fostering evidence-based educational practices.