Analysis of Students’ Mathematical Self-Efficacy Instruments Using Rasch Model

Abstract

The purpose of this study was to determine the quality of mathematics self-efficacy instruments in terms of validity, reliability, undimensionality, ability distribution, and suitability of student responses. Data were obtained by distributing mathematics self-efficacy instruments consisting of 20 statements to 45 seventh-grade students of a private junior high school in Yogyakarta, Indonesia, at the 2022/2023 school year selected through purposive random sampling technique. The data were analyzed using Rasch model with the help of the Ministep application. The results of the analysis showed that the mathematics self-efficacy instrument used was declared valid, with an outfit Z standardized (outfit ZSTD) value of -0.16 for respondents and 0.20 for items, and could be used to measure students' mathematics self-efficacy level with a reliability level reaching a score of 0.54 or in the medium category. The results of the distribution of item abilities show that most students admit to having the fighting power to solve problems that occur, especially in learning mathematics. From the analysis of the suitability of student responses, 18 statements show the suitability between the level of self-efficacy and the responses given by students. While two other statements did not show suitability. Therefore, teachers or other academics can use the self-efficacy questionnaire that has been prepared by researchers because it has met the four criteria for testing instruments.