Scholarlink

<p>This study investigates the influence of interactive learning modules on self-directed learning (SDL) among junior high school students in Jakarta, with a particular focus on learning autonomy. As education increasingly adopts technology-enhanced approaches, understanding how digital instructional tools shape student agency and independent learning becomes crucial. Despite growing interest in self-directed learning, few studies have examined the mechanisms through which interactive modules contribute to students’ autonomy and their capacity for self-regulation, particularly within the Indonesian context. Using a quantitative, cross-sectional design, data were collected from 260 junior high school students selected through stratified random sampling across multiple schools in Jakarta. Three validated instruments were used to measure interactive learning module usage, learning autonomy, and self-directed learning. Partial Least Square Structural Equation Modeling (PLS-SEM) using SmartPLS was employed to analyze the hypothesized relationships, while Random Forest and XGBoost were used to complement the analysis by predicting self-directed learning outcomes and identifying the most influential factors. The results reveal that interactive learning modules have a significant positive effect on both learning autonomy and SDL. Learning autonomy plays a central role, indicating that students with higher autonomy derive greater benefits from interactive learning environments. Predictive analyses demonstrate that XGBoost outperforms Random Forest in predicting SDL outcomes, highlighting the superior predictive power of XGBoost and reinforcing the robustness of learning autonomy as a key determinant. These findings highlight the importance of autonomy in enhancing self-directed learning. By reinforcing students’ capacity for independent learning, interactive modules also align with the broader vision of education sustainability, equipping learners with lifelong skills needed to adapt to evolving educational and societal challenges. This study contributes to the growing body of knowledge on digital pedagogy by clarifying how technology fosters independent learning through autonomy development. It also offers practical implications for educators and policymakers to design and implement interactive learning environments that promote student self-regulation. The originality of this research lies in its integration of PLS-SEM with predictive modeling approaches, Random Forest and XGBoost, to provide both explanatory and predictive insights into technology-based learning in Indonesian secondary education.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Education sustainability through interactive modules and learning autonomy to improve self directed learning among middle school students with PLS SEM random forest and XGBoost

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年12月30日

查看全文
  • Nurzengky Ibrahim,
  • Desy Safitri,
  • Arita Marini,
  • Iqbal Pratama,
  • Asep Marfu

摘要

This study investigates the influence of interactive learning modules on self-directed learning (SDL) among junior high school students in Jakarta, with a particular focus on learning autonomy. As education increasingly adopts technology-enhanced approaches, understanding how digital instructional tools shape student agency and independent learning becomes crucial. Despite growing interest in self-directed learning, few studies have examined the mechanisms through which interactive modules contribute to students’ autonomy and their capacity for self-regulation, particularly within the Indonesian context. Using a quantitative, cross-sectional design, data were collected from 260 junior high school students selected through stratified random sampling across multiple schools in Jakarta. Three validated instruments were used to measure interactive learning module usage, learning autonomy, and self-directed learning. Partial Least Square Structural Equation Modeling (PLS-SEM) using SmartPLS was employed to analyze the hypothesized relationships, while Random Forest and XGBoost were used to complement the analysis by predicting self-directed learning outcomes and identifying the most influential factors. The results reveal that interactive learning modules have a significant positive effect on both learning autonomy and SDL. Learning autonomy plays a central role, indicating that students with higher autonomy derive greater benefits from interactive learning environments. Predictive analyses demonstrate that XGBoost outperforms Random Forest in predicting SDL outcomes, highlighting the superior predictive power of XGBoost and reinforcing the robustness of learning autonomy as a key determinant. These findings highlight the importance of autonomy in enhancing self-directed learning. By reinforcing students’ capacity for independent learning, interactive modules also align with the broader vision of education sustainability, equipping learners with lifelong skills needed to adapt to evolving educational and societal challenges. This study contributes to the growing body of knowledge on digital pedagogy by clarifying how technology fosters independent learning through autonomy development. It also offers practical implications for educators and policymakers to design and implement interactive learning environments that promote student self-regulation. The originality of this research lies in its integration of PLS-SEM with predictive modeling approaches, Random Forest and XGBoost, to provide both explanatory and predictive insights into technology-based learning in Indonesian secondary education.