Development of a Learning Analytics-Based Adaptive Curriculum Model to Support Personalized Learning

Abstract

Personalized learning requires curriculum approaches that can respond to differences in students' abilities, learning pace, engagement, and learning needs. However, conventional curricula often provide relatively uniform learning content, activities, pacing, and assessment, making continuous individual adaptation difficult. This study aimed to develop a Learning Analytics-Based Adaptive Curriculum Model to support personalized learning. The study employed a Research and Development/Design-Based Research approach involving students, teachers, and educational experts. Learning analytics data, including assessment performance, task completion, learning activity, engagement, learning pace, error patterns, and learning progress, were analyzed to develop student learning profiles and inform curriculum adaptation. The model was evaluated through expert validation, practicality testing, classroom implementation, and effectiveness assessment. The findings showed that distinct student learning profiles could be identified and translated into differentiated learning pathways involving remediation, standard learning, and enrichment. Expert evaluation indicated that the model was valid, while teacher and student responses demonstrated its practicality for instructional implementation. The implementation also showed positive changes in learning achievement, engagement, mastery, progress, and task completion. The study concludes that integrating learning analytics into curriculum decision-making can provide a systematic mechanism for connecting student learning evidence with adaptive curriculum responses and personalized learning. The model offers practical implications for teachers, curriculum developers, schools, and educational technology developers seeking to implement more responsive and data-informed learning environments.