This study aims to analyze the quality of test items used to measure students' abilities. Employing a descriptive quantitative approach with an evaluative method, the subjects of the study consisted of 23 test participants selected through purposive sampling. The object of the research was 10 multiple-choice questions with five answer options. Data were collected through documentation methods, including participants' answer sheets, answer keys, and total score data. The results show: (1) Test reliability is low (0.34); (2) Item validity varies with 30% moderate, 40% low, and 30% very low; (3) Item difficulty levels are disproportionate with 40% very easy, 50% easy, and 10% difficult; (4) Discrimination power shows 20% good items, 30% adequate, and 50% poor; (5) Distractor effectiveness reveals 40% non-functioning distractors, 45% minimally functioning, and 15% effectively functioning. Based on the analysis, the test instrument requires substantial improvement, particularly in reliability, difficulty level proportion, and distractor quality. Improvement recommendations include total revision for 30% of items, partial revision for 40% of items, and minor improvements for 30% of items.
Copyrights © 2025