Abstract

The test is a tool to collect information about the achievement of learning objectives so that the test must have good quality in order to be able to measure students' abilities accurately. To determine the quality of the test, it is necessary to do an analysis, one of which is the Rasch model approach. This study aims to analyze the quality of the final semester test results in learning mathematics—sampling using a purposive sampling technique. The subjects of this study were 56 students of Junior High School (SMP) in Yogyakarta, consisting of 25 males and 31 females. The test instrument consists of 40 multiple choice questions with the correct answer given a score of 1 and the wrong answer being given a score of 0. Therefore, the data obtained is dichotomous. The test kits were analyzed by applying the Rasch model. The study results obtained an estimate of the validity of item fit, where the questions analyzed had an INFIT MNSQ value between 0.94 - 1.11 and all questions with an OUTFIT value t ≤ 2.00; thus, 40 items fit according to the Rasch model. The characteristics of the test questions based on the Rasch model have a very difficult difficulty level of 0 items by 0%. The difficult category is three items at 7,5%, the medium category is 20 items at 50%, the easy category is 17 items at 42.5%, and 0% for the very easy category. The reliability of the case estimate value is 0.00 in the medium category, and the reliability of the item estimate value is 0.85 in the very high category. The implication of this research is additional knowledge regarding the Rasch analysis of mathematical problem models as a graduation standard.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call