AI-based Learning Analytics in Higher Education
The use of AI-based Learning Analytics in Higher Education to improve Teaching, address Challenges, and keep Pedagogy central.
Duration: June 2025 – March 2026
Status: Ongoing
Educational Level: Tertiary Level
Topic: Artificial Intelligence AI
Keywords: Artificial Intelligence, Adaptive Learning System, Learning Analytics
Initial Situation
The rapid digitization of higher educatiom – especially the rise of AI – creates new possibilities and risks for using learning analytics to optimize or even redesign teaching and learning. Universities struggle to harness this potential to solve real issues in teaching-learning contexts while keeping pedagogical relevance at the center. We address with this project the following basic question: “Which concrete measures should a forward-looking university initiate to proactively meet needs regarding AI-based learning analytics?”
Objectives
- Enable participating universities to make informed decisions about integrating AI-based learning analytics, with clear benefits for learners
- Clarify the functionalities and distinguishing features of AI-based learning analytics
- Create a criteria catalog for pedagogically meaningful AI-based learning analytics and identify concrete teaching/learning scenarios where it adds value
- Provide orientation on existing systems and near-term developments
- Compile lessons learned from use cases and good-practice examples
- Foster intensive exchange among partners and enable scaling of results
Method
We conduct expert group discussions combining both technical and pedagogical perspectives on AI-based learning analytics. We plan two rounds per group (3–5 experts each): a first session (2h) and a follow-up (1.5h); sessions are recorded with consent and guided by a semi-structured question framework. We analyze transcripts via qualitative content analysis using MAXQDA. We compile a results report that answers the research questions; the project is explicitly framed as qualitative research.
Planned Translation
Insights from expert discussions are explicitly “translated into a concept for implementation,” not just a report. Therefore we produce practical outputs universities can act on: a) criteria catalog for pedagogically meaningful AI-based learning analytics, b) collection of use cases & good practices, and c) actionable recommendations for introducing AI-based learning analytics. Recommendations include a selection and prioritization of relevant project fields so institutions can pilot and scale with resources aimed at the highest-value areas. We address enablers and guardrails and we foster uptake through a partner network. One main goal is to enable intensive exchange between partners and scaling of results across institutions, giving the findings a direct path into practice. This project equips universities with a practical, prioritized playbook to implement AI-based learning analytics responsible by maximizing pedagogical value.
