Evaluasi Kualitas Butir Soal Asesmen Sumatif Pendidikan Agama Islam Kelas VIII di SMP Negeri 2 Kaligondang (Analisis Validitas, Reliabilitas, Tingkat Kesukaran, Daya Beda, dan Efektivitas Pengecoh)

Abstract

Evaluation is an essential component in the learning process, as it determines the effectiveness of instruction. This study aimed to analyze the quality of test items in the final semester of Islamic Religious Education for Grade VIII. A descriptive quantitative approach was employed to examine test quality, including construct and empirical validity, reliability, difficulty level, discrimination index, and distractor effectiveness. The subjects were 179 Grade VIII students. Data were collected by documenting items and answer sheets and analyzed using IBM SPSS Statistics 27 and ANATES V4. The results indicated that cognitively, items were dominated by levels C3 (45%), C2 (35%), and C4 (25%), were no items at levels C1, C5, and C6. Empirically, all items were 100% valid, with a value of 0,146. The reliability coefficient was 0,717, indicating acceptable reliability. The difficulty level was mostly moderate (65%) and easy (35%), with no difficult items. The discrimination index ranged from good to very good (50% each). Half of the distractors functioned effectively, while the rest required improvement. Therefore, the test is feasible for use with revisions in distractor quality and cognitive level distribution.

Keywords: Assessment Sumative Evaluation, Item Quality, Islamic Religious Education