This study examines the role of sustainable lifestyle in the relationship between environmental education and ecological intelligence among junior high school students in the Province of Bangka Belitung, Indonesia. Grounded in environmental psychology and education theories, the research explores how structured environmental learning influences students’ sustainable behaviors and ecological reasoning. A quantitative approach was employed using Partial Least Squares Structural Equation Modeling (PLS-SEM) via SmartPLS to assess the structural relationships between constructs, while Support Vector Machine (SVM) and Light Gradient Boosting Machine (LightGBM) via Python were applied to classify and predict ecological intelligence levels based on environmental education and sustainable lifestyle factors. Data were collected through structured questionnaires administered to a stratified random sample of 250 students, measuring Environmental Education, Sustainable Lifestyle, and Ecological Intelligence using validated five-point Likert scale items. Findings revealed that environmental education significantly enhances sustainable lifestyle and ecological intelligence. Sustainable lifestyle also positively contributes to ecological intelligence and explains part of the link between education and intelligence. Machine learning analyses supported these findings: SVM accurately classified students’ ecological intelligence (EI) levels with 85.3% accuracy, while LightGBM highlighted the most influential predictors, including SL2 (0.20) and SL1 (0.18) from Sustainable Lifestyle (SL) items, and EE3 (0.15), EE1 (0.10), and EE2 (0.08) from Ecological Engagement (EE) items. This study contributes to the literature on environmental education in Indonesian schools by integrating structural modeling with advanced predictive analytics. The results underscore the importance of embedding behaviorally oriented environmental learning in school curricula to support the development of ecological intelligence in youth.
Safitri et al. (Sun,) studied this question.