This study analyzes the quality of a science test instrument on inheritance, based on Higher Order Thinking Skills (HOTS) and scientific literacy, in terms of validity, reliability, difficulty level, discrimination index, and distractor effectiveness. A descriptive quantitative approach with an ex post facto item analysis design was used. The participants were 28 students, and the instrument consisted of 20 multiple-choice items developed according to HOTS and scientific literacy indicators. Validity was tested using point-biserial correlation, reliability using Kuder-Richardson (KR-20), and item difficulty and discrimination indices were analyzed through correct response proportions and comparisons between upper and lower groups. Results showed that 85% of the items were valid and 15% invalid. The reliability coefficient was 0.8827, indicating high consistency. In terms of discrimination, 60% of items were very good, 25% good, 10% fair, and 5% poor. Most items were moderately difficult (90%), and 10% were difficult, with no easy items. Overall, the instrument demonstrates good quality and is suitable for learning evaluation, although some items require revision to further optimize measurement accuracy.
Copyrights © 2026