Evaluation for Analyzing Self-Regulated Learning Processes using Trace-Data of Learning Log from Digital Textbooks

Abstract

SRL (Self-Regulated Learning) is currently the subject of much research. In our previous studies, there are systems (STELLA) that can accumulate learners' learning history in digital textbooks and systems (SELFY) that support SRL by providing feedback to learners on their learning history collected by STELLA. However, since the evaluation method is in the form of a questionnaire, the evaluation is inherently subjective and often lacks real-time behavioral insight. Furthermore, existing methods are frequently domain-dependent, making it difficult to generalize findings across different subjects. We had previously proposed this based on trace data from digital textbooks. We compare high and low groups of SRL scores measured by the SRL-SRS scale revealed that the high group tended to exhibit behavioral cycles consistent with theoretical SRL characteristics. These findings suggest the proposed method may be effective for evaluating SRL based on trace data.