<![CDATA[This study aims to analyze teacher-made summative test items in Indonesian language subject for Grade VIII at SMP Negeri 01 Kota Bengkulu in the 2025/2026 academic year in terms of validity, reliability, difficulty level, discrimination power, and distractor effectiveness. This research employed a descriptive quantitative method. The data consisted of 50 multiple-choice items from the Final Semester Summative Test administered through google form and students’ answer sheets from 37 students of class VIII.3. The data were collected through documentation techniques. The data analysis technique used quantitative item analysis, including validity testing, reliability testing using the K-R 20 formula, analysis of difficulty level, discrimination power, and distractor effectiveness. The results showed that: (1) in terms of validity, 23 out of 50 items were valid, while 27 items were invalid; (2) the reliability coefficient of the test was 0.72, categorized as high; (3) the difficulty level indicated that 41 items were of moderate difficulty and 9 items were easy; (4) the discrimination power revealed that 24 items were categorized as poor, 7 items as moderate, 16 items as good, and 3 items as very good; (5) for distractor effectiveness, 19 items did not require revision, while 31 items needed revision. Based on these findings, it can be concluded that the teacher-made summative test items are fairly good. Some items meet the criteria of validity and reliability with a high reliability level; however, several items are still invalid, have disproportionate difficulty levels, low discrimination power, and ineffective distractors. Therefore, improvements are needed in test construction to ensure higher quality assessment instruments and more accurate measurement of students’ abilities.]]>
