Abstract:

Self-study constitutes approximately half of the workload in several bachelor programmes at the Bern University of Applied Sciences (BFH), yet systematic insight into students’ learning processes during this phase remains limited. This project introduces ELIAS, an interdisciplinary initiative that leverages Learning Analytics (LA) to improve the effectiveness of accompanied self-study in the BSc Agronomy as well as the BSc Physiotherapy and Nursing programmes.

Building on established LA frameworks, the project develops two core components:
(1) an Analytics Engine capable of processing anonymised Moodle activity data to detect behavioural patterns, engagement trajectories, and task-related challenges; and
(2) interactive dashboards designed to support instructors and programme directors in evidence-informed decision-making.

The pedagogical goal is not monitoring, but enabling reflective, data-informed teaching practice and fostering a more transparent understanding of learning pathways. Insights generated through LA provide actionable recommendations for optimising task design, pacing, and instructional support, thereby strengthening learner engagement and learning gains. Developed collaboratively by departments in technology, agronomy, and health, the project combines data science expertise with domain-specific didactic knowledge, offering a scalable model for interdisciplinary learning innovation.

The expected contributions include a functional analytics infrastructure, validated dashboards for educational use, and guidance on ethically grounded pedagogical interpretation. As one of the first BFH-wide initiatives applying LA specifically to self-study, ELIAS advances the institution’s digital education strategy and enriches research-informed practice in learning analytics.

Keywords:

Learning Analytics, Self-Study, Higher Education, Dashboard Design, Educational Data Mining.