Self-reported and digital-trace measures of computer science students’ self-regulated learning in blended course designs

Abstract This study investigated the extent to which self-report and digital-trace measures of students’ self-regulated learning in blended course designs align with each other amongst 145 first-year computer science students in a blended “computer systems” course. A self-reported Motivated Strategies for Learning Questionnaire was used to measure students’ self-efficacy, intrinsic motivation, test anxiety, and use of self-regulated learning strategies. Frequencies of interactions with six different online learning activities were digital-trace measures of students’ online learning interactions. Students’ course marks were used to represent their academic performance. SPSS 28 was used to analyse the data. A hierarchical cluster analysis using self-reported measures categorized students as better or poorer self-regulated learners; whereas a hierarchical cluster analysis using digital-trace measures clustered students as more active or less active online learners. One-way ANOVAs showed that: 1) better self-regulated learners had higher frequencies of interactions with three out of six online learning activities than poorer self-regulated learners. 2) More active online learners reported higher self-efficacy, higher intrinsic motivation, and more frequent use of positive self-regulated learning strategies, than less active online learners. Furthermore, a cross-tabulation showed significant ( p < .01) but weak association between student clusters identified by self-reported and digital-trace measures, demonstrating self-reported and digital-trace descriptions of students’ self-regulated learning experiences were consistent to a limited extent. To help poorer self-regulated learners improve their learning experiences in blended course designs, teachers may invite better self-regulated learners to share how they approach learning in class.