Investigation of variables predicting academic resilience in mathematics: An explanatory item response modeling application

In this research, it was aimed to examine the variables that predict academic resilience in mathematics. The study group of the research consists of 553 high school students, 355 female and 198 male. The research was carried out according to the relational research design, which is a descriptive research. In the research, data were collected with the help of Academic Resilience Scale in Mathematics and personal information form. The relationship between students' academic resilience in mathematics and the variables of gender, academic success in mathematics, participation in mathematics courses, and educational status of parents were investigated by Explanatory Item Response Modeling (EIRM). In the analysis of the data, the Partial Credit Model was used within the scope of Item Response Theory (IRT) in the first stage. According to the results of the analysis, it was seen that the students' perceptions of resilience in mathematics differed according to their academic success in mathematics. In the second stage of the research, measurement invariance analyzes were made based on the Item Response Theory according to the variables of gender, mother and father education status, and it was observed that there was no measurement invariance in gender subgroups. As the third and final stage, the Differential Item Function (DIF) analyzes of the scale items were performed in the subgroups according to the variables of gender, mother and father educational status. As a result, DIF was determined in only one item according to the father's educational status subgroups.