The focus of the project ‘AI-based Learning Analytics in Higher Education’ is to explore the pedagogical potential of AI-based learning analytics. To obtain results, group discussions are held with relevant experts and evaluated qualitatively.
Project description
Project Description
The main objective of the project is to enable universities to make informed decisions about the integration and use of AI-based learning analytics. It is essential that this creates added value, especially for the learners themselves. The project is based on the following target group and specific objectives:
Target group
The project has an orientation perspective in higher education teaching and is therefore aimed at people who organise and carry out teaching at universities.
Objectives
- Gain clarity about the functionalities and special features of AI-based learning analytics.
- Create a catalogue of criteria for pedagogically meaningful AI-based learning analytics and examine the question of in which specific teaching and learning scenarios AI-supported learning analytics makes sense and how it may influence teaching, learning and learning behaviour.
- Provide orientation knowledge on existing systems and planned developments in the near future.
- Compile previous experience based on use cases and good practice.
- Develop recommendations for action for universities in the context of introducing AI-based learning analytics. This also means prioritising project areas for investigating AI-based learning analytics at universities so that resources can be used in a targeted manner.
- Greater clarity on the issues of data availability and data protection.
- Intensive exchange of experience between the cooperation partners and scaling of the results.
Research Question
Increasing digitalisation in education, including the use of artificial intelligence (AI), opens up new opportunities and risks for the use and integration of learning analytics, not only to optimise the teaching and learning process, but also to redesign it in some circumstances. However, educational institutions face the challenge of exploiting the potential for solving existing problems in the teaching-learning context while focusing on pedagogical relevance. The overarching question underlying this project is therefore:
What concrete measures should a forward-looking university take to proactively meet the needs of AI-based learning analytics?
To this end, the following aspects are being examined:
- Unique points of AI-based learning analytics: What distinguishes AI-based learning analytics? How does AI-based learning analytics differ from previous approaches? What are the potentials and risks of AI-based learning analytics?
- Pedagogically meaningful AI-based learning analytics: What distinguishes pedagogically meaningful AI-based learning analytics? What scenarios are conceivable? What scientific studies (including meta-studies) already exist on this topic?
- Use cases and good practice: What examples of AI-based learning analytics can serve as use cases and good practice? What lessons can be learned from them? What experience has been gained so far?
- Existing systems and future developments: What systems are currently available? What systems and functionalities are currently under development? What theoretical approaches are there for future developments?
- Data availability and data protection: What data is necessary to implement AI-based learning analytics? What challenges arise from a technical and legal perspective? What current developments and solutions exist in the field of data protection?
Methodology
Group discussions with experts will be held to answer the research questions. The experts will either specialize in the technical or educational aspects of AI-based learning analytics or relevant sub-aspects thereof. Experts are defined as individuals who conduct research or development in these areas and therefore ideally have many years of expertise. Two group discussions are planned, each consisting of three to five experts, with both ‘technicians’ and ‘educators’ participating (i.e., both areas of expertise should be represented within a group discussion). A group discussion lasts around 2.5 hours and is facilitated by two moderators. The discussions will be recorded with the participants' consent and conducted using a semi-structured questionnaire. The evaluation will be carried out using qualitative content analysis, with the MAXQDA software being used. A results report will answer the research questions and summarize the findings. Overall, the project can be classified as qualitative research.
