Improving learning and teaching by using digital trace data in higher education (digiVL)

We use data from about 8000 undergraduate students in science lectures at the University of California, Irvine. Data includes:

• Digital trace data from learning management systems, that provide objective measures of students’ learning activities, duration of learning activities, and performance,
• Self-reported data on course-specific motivation and goals
• Detailed information on course design and course contexts (course syllabi)

We use quantitative statistical analyses (e.g. multilevel models, latent profile analysis) to investigate interindividual and intraindividual differences in learning activtities and success.

Students’ learning activities, for example, can differ in the extent to which students use course materials in their learning management systems, if they study regularly throughout the semester or cram until shortly before exam dates, or if they reach out to peers and instructors for help. We Investigate, which individual student characteristics (e.g., motivation, prior achievement, goals, etc.) and which course-design related factors (regular assigments, discussion forums, etc.) support successful learning.

Results of this project will inform a follow up study to investigate self-regulated learning and student success with innovative data at TU Dortmund university in collaboration with researchers of the TUD profile FAIR. 

• Prof. Richard Arum (University of California Irvine, School of Education)
• Jun.-Prof. Charlott Rubach (Universität Rostock, Institut für Schulpädagogik und Bildungsforschung)