Assessment is a critical component of the learning process, serving to determine the achievement of learning objectives. This study aims to evaluate the quality of Arabic language multiple-choice final exam items for eleventh-grade students of class "L" at Madrasah Aliyah Negeri 2 Cilacap in the 2023/2024 academic year, focusing on validity, reliability, difficulty level, discrimination index, and distractor effectiveness. This quantitative descriptive study used data sources including question sheets, student answer sheets, question grids, and answer keys. Data were collected through interviews and documentation and analyzed using the Anates V4 application. The findings revealed that the overall quality of the items was low. In terms of validity, only 7 items (23%) were categorized as high and none as very high; 14 items (47%) were classified as very low. Reliability was high with a coefficient of 0.72. Regarding difficulty level, most items were too easy, with 15 items (50%) in the very easy category. The discrimination index showed 11 items (37%) were not good, while none reached the very good category. Distractor effectiveness showed poor results, with 12 items (40%) in the very bad category. These findings suggest that the exam items require significant revision to meet quality standards.
Copyrights © 2025