Development of Student Academic Performance Determinants Using Rasch Model Analysis

Abstract

This study aims to develop an instrument to determine students' academic performance. In this case, a quantitative design was used, with a random sampling technique utilized in selecting 415 students from the University of Muhammadiyah Malang. Data analysis was also carried out using Rasch Model. Based on the fit item criteria analysis, 5 misfit elements were each observed for the Mean and Standard Deviation logit and MNSQ (Mean Square) outfit. This indicated that the most difficult and easiest questions were numbers 47 and 10, respectively. For regional and school origins, the items containing differential function questions were 2 and 9, respectively. The reliability of the measuring instrument, item, and person was also 83, 98, and 79. In addition, only 7 of the 67 utilized items did not meet the required criteria and the remaining were completely accepted.