Alignment of the Blueprint and Item Analysis as Indicators of the Validity of the Multiple Choice Test Instrument

Abstract

Multiple Choice Questions (MCQs) are widely used as cognitive assessment tools in medical education due to their efficiency and objectivity. The quality of MCQs is strongly influenced by the alignment between examination planning through blueprinting and the empirical performance of test items evaluated using item analysis. This study aimed to examine the alignment between the examination blueprint and item analysis results as indicators of MCQ instrument validity. This study employed a descriptive observational quantitative design with a retrospective approach. Data were obtained from examination blueprint documents and MCQ test results from the most recent block examination conducted at the Medical Study Program, Faculty of Medicine, Universitas Muhammadiyah Palembang. A total of 150 MCQ items were analyzed using statistical software to calculate difficulty index, discrimination index, and internal reliability measured by Cronbach’s alpha coefficient. The results demonstrated that most items were classified within the moderate difficulty category (47.33%), with relatively balanced proportions of difficult and easy items. Item discrimination analysis revealed that 70.67% of items showed good discrimination (?0.30), although 29.33% of items demonstrated fair to poor discrimination. The internal reliability of the examination instrument was very high, with a Cronbach’s alpha value of 0.91. In conclusion, the alignment between the examination blueprint and item analysis findings supports the validity and reliability of the MCQ assessment instrument. Nevertheless, continuous evaluation and revision of items, particularly those with low discrimination indices, are recommended to enhance the overall quality of assessment.