Abstract
This study develops and validates an Indonesian mathematical resilience scale for elementary students through robust psychometric testing. Addressing the absence of culturally appropriate measurement tools, the research employs a mixed-method validation design combining exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) using JASP software. Participants included 405 students (aged 10-12) from five Central Java regencies, selected through stratified random sampling. The 15-item Likert-scale questionnaire underwent principal component analysis (EFA) followed by maximum likelihood estimation (CFA), evaluating four theoretical dimensions: Value, Struggle, Growth, and Perseverance. Results demonstrated excellent model fit (CFI=0.912; RMSEA=0.047; SRMR=0.045) with strong composite reliability (CR=0.78-0.92). While three dimensions exhibited adequate convergent validity (AVE>0.50), Perseverance showed marginal acceptability (AVE=0.47), suggesting potential cultural nuances in persistence measurement. The validated instrument enables precise identification of resilience profiles, informing targeted pedagogical interventions. This study contributes methodologically by demonstrating the sequential EFA-CFA approach’s effectiveness in educational instrument development, and practically by providing teachers with a diagnostic tool for fostering mathematical resilience. Findings highlight the importance of integrating affective constructs into mathematics pedagogy and support the development of resilience-based instructional practices. Future research should expand validation to other Indonesian regions and examine the scale’s predictive validity for academic achievement across different instructional contexts.
Keywords: Confirmatory Factor Analysis (CFA), Exploratory Factor Analysis (EFA), Mathematical Reciprocity, Mathematical Resilience, Psychometric Validation.