Evaluating Students' Academic Resilience in Chemistry Learning: Insights from a Rasch Model Analysis

Teaching chemistry is a complex subject that requires a certain level of knowledge and skills to understand. Students must overcome this challenge because teaching chemistry involves abstract ideas (atoms, molecules, and electrons) and principles, laws, reaction equations, and mathematical operations. Increasing academic resilience is important in enhancing students' understanding and well-being in learning. This research aimed to test the validity and reliability of students' chemistry academic resilience tests using the Rasch model. Data collection was conducted using Google Forms. Data analysis utilized the Rasch model, assisted by the Winstep application, to reveal various aspects of the assessment. Based on the research findings, a Cronbach's alpha coefficient of 0.78 indicates strong internal consistency. In addition, item reliability reached a significant value of 0.99, while person reliability of 0.78 confirmed the consistency of respondents in providing accurate answers in the assessed categories. Furthermore, the OUTFIT and INFIT MNSQ (persons) have average values of +1.02 and +1.01, respectively (the closer to 1, the better). In contrast, the INFIT and OUTFIT ZSTD values are 0.1 and 0.3, respectively (the closer to 0, the better). The most difficult question is coded ra11 with a logit score of 0.61, and the easiest is coded ra4 with a logit score of -0.05. Therefore, the student academic resilience instrument is an effective measuring tool. In the future, chemistry educators and researchers can benefit from the potential impact of this research on Indonesian education.