Development and validation of a learning analytics rubric for self-regulated learning

Abstract This study presents the development and validation of a Learning Analytics Rubric for Self-Regulated Learning (SRL) in higher education. The rubric aims to measure students’ SRL processes within learning management systems (LMS) in a scalable, consistent, and explainable manner. The research follows a design-based approach, mapping validated SRL scales to LMS data indicators, developing the rubric for a Canvas LMS, and validating it with postgraduate students. The study identifies challenges in measuring SRL, such as the dynamic nature of SRL and the limitations of translating traditional self-report methods to digital environments. By leveraging learning analytics, the study proposes a novel approach to measure SRL behaviors using LMS data. The validation process reveals that five of the seven indicators accurately reflect students’ SRL skills, with strong alignment between student self-assessments and system-generated scores for indicators related to reviewing content, integrating information from multiple sources, following study schedules, pacing learning, and reading assessment instructions. However, significant discrepancies were observed in indicators measuring completion of extra activities and early semester engagement with the LMS, highlighting the need for further refinement. The findings suggest that integrating learning analytics with rubrics can provide valuable insights into students’ learning processes, particularly measuring SRL, while supporting the development of effective educational interventions from a student-centered approach.