This study aims to analyze the quality of multiple-choice questions used in the Arabic language Mid-Semester Summative Assessment at MTs Tanada, focusing on difficulty level, discrimination power, and distractor function. A descriptive quantitative design was employed, analyzing 20 test items and student answer sheets using spreadsheet software. Difficulty levels were determined based on the difficulty index of each item, while discrimination power was assessed by an item's ability to distinguish between high- and low-performing students. Distractor function was evaluated based on the frequency with which students selected incorrect answer options. The analysis revealed that regarding difficulty levels, 8 items (40%) were categorized as easy, 2 items (10%) as moderate, and 10 items (50%) as difficult. In terms of discrimination power, 6 items (30%) were rated as excellent, 1 item (5%) as good, 3 items (15%) as fair, and 10 items (50%) as poor. Of the 60 distractors analyzed, 43 (71.67%) functioned effectively, while 17 (28.33%) did not. These findings indicate varied item quality; notably, half of the items exhibited poor discrimination power, and the distribution of difficulty levels was dominated by the "easy" and "difficult" categories. The results provide a basis for educators to retain items that meet established criteria and to revise or replace those that do not, thereby improving the quality of the Arabic language assessment instrument
Copyrights © 2026