Evaluasi Psikometrik Instrumen Kemampuan Berpikir Tingkat Tinggi pada Materi Sistem Organ Manusia

Abstract

This study aims to evaluate the psychometric quality of a higher order thinking skills instrument in the topic of the human organ system for junior high school science learning. A descriptive quantitative approach was applied to analyze the validity, reliability, difficulty index, discrimination power, and distractor effectiveness. The instrument consisted of ten multiple-choice items developed based on the revised Bloom’s taxonomy at levels C4-C6. Content validity was examined by two experts using Aiken’s V, while empirical validity was tested with the Pearson Product-Moment correlation. The reliability coefficient was calculated using KR-20. The results show that the content validity values ranged from 0.83 to 0.94 (very valid), and the reliability coefficient reached 0.78 (high category). Eight items demonstrated moderate difficulty, seven items had good discrimination power, and 70 percent of distractors functioned effectively. These findings indicate that the developed HOTS instrument is psychometrically sound and feasible for strengthening 21st-century science assessment, although several items require minor revision to improve balance in difficulty and distractor performance.