Islamiani Safitri, R.A. Ansyari Rizli
The problem of this study is that the making of science achievement tests based on the Integrated Islamic Curriculum by science teachers at Islamic Junior High Schools does not go through a calibration process. Hence, the test items' validity, reliability, and characteristics are unknown. Therefore, it is necessary to calibrate the science achievement test at Islamic Junior High School to see the quality and characteristics of the items. This research uses a quantitative approach to analyze science achievement test items through the Racsh Model. The sample used was class III students of Junior High School, totaling 100 people. The accuracy of the Rasch Model calibration method in this study is determined by content validity, reliability, item analysis, and student ability. To determine the content validity using the Aiken formula, the test data is processed with Jmetrik to analyze the items. The results of this study are 40 items declared valid with medium (27 items) and strong (13 items) categories. Furthermore, items with difficulty levels in the range -2 ≤ b ≤ +2 were 22 items (55%), and at difficulty levels b < -2 or b > +2 were 18 items (45%). The fit of the items with the Rasch model, 26 items (65%) fit the model, and 14 items (35%) do not fit the model. Item reliability was 0,9779, and person reliability was only 0,6488. Therefore, it is of utmost importance to improve the items of the science achievement test based on the Integrated Islamic Curriculum so that they can measure students' science learning outcomes appropriately. © 2024 Copyright held by the owner/author(s).
Department of Mathematics Education, Universitas Labuhanbatu, North Sumatera, Indonesia; Department of Management Education, Universitas Negeri Yogyakarta, D.I. Yogyakarta, Indonesia
