Item Analysis Of Arabic Language Examination

Abstract

Assessment and evaluation are needed to determine how student competence is compatible with predetermined educational goals. This study aims to analyze the Midterm Exam for odd-semester Arabic subjects at Islamic Junior High. The type of research used in this research is descriptive quantitative research. This study uses the documentation method for data collection. In addition, researchers will use the help of a computer program, namely Anathes 4.0.2. The results showed that: (1) The validity test of 15 multiple choice test questions was declared valid (80%). As well as for the validity test of 5 matching test questions they were declared valid (80%). (2) The reliability test of multiple choice test items is 0.87. In contrast, the results of the reliability of matchmaking test items were 0.76. (3) The difficulty level test of the multiple test questions categorized as hard is 0 items (0%), medium is three items (20%), and easy is 12 items (80%). The difficulty level test of the matching test questions categorized as hard is 0 items (0%), medium is four items (80%), and easy is 1 item (20%). (4) The difficulty level test of the Multiple test questions categorized as bad is 1 item (6,66%), enough is two items (13,3%), good is nine items (60%), and Very Good is three items (20%). The difficulty level test of the matching test questions categorized as bad is 0 items (0%), enough is 1 item (20%), good is 0 items (0%), and Very Good is four items (80%). With these results, the test takers or teachers must increase their attention to the analysis of the items, namely in terms of validity, reliability, difficulty level, and discriminatory power.