Rasch validation of constructivist learning environment in Indonesian primary school mathematics

Abstract

Despite widespread use of the constructivist learning environment survey (CLES), there is limited evidence regarding its validity in primary mathematics classrooms in Indonesia. This gap raises concerns about the accuracy of measuring students’ perceptions of constructivist practices in this specific context. This study aims to examine the psychometric properties of the Indonesian version of the CLES using the Rasch measurement model. A total of 246 fifth- and sixth-grade students from four districts participated in the study. The results indicate that the instrument demonstrates adequate psychometric quality, including good reliability (Cronbach’s α=0.83), strong person reliability (0.81), and excellent item reliability (0.97). The scale also satisfies unidimensionality assumptions, and the 5-point Likert response format functions effectively. These findings suggest that the instrument consistently captures students’ perceptions and the items are well-targeted to the sample. However, differential item functioning (DIF) was identified across gender and district groups, indicating potential bias in several items. Overall, the Indonesian CLES is a reliable and valid tool, with minor refinements needed to improve measurement invariance across groups. This study contributes to both theory and practice by providing a context-sensitive, Rasch-validated instrument that supports accurate evaluation and improvement of constructivist mathematics teaching in Indonesian primary schools.