Under the term “learning analytics”, a field of research has developed internationally in recent years that focuses on the collection, evaluation and application of complex, often multi-modal and digital behavioral traces. These behavioral traces, which learners and teachers leave behind in digital contexts, are evaluated with the help of computer-based models or machine learning methods in order to gain insights into teaching and learning processes. Teaching and learning research is currently slowly approaching the field of learning analytics. The potential of learning analytics for teaching and learning research has therefore already been recognized, so that it seems extremely worthwhile to think further at this point. This article therefore outlines the research on learning analytics, which is already well advanced, and uses specific examples to illustrate the potential of learning analytics approaches for the further development of learning and instructional theories. In particular, self-regulated and cooperative learning as well as the design of the learning environment and support for teachers are discussed. The risks and challenges of learning analytics (such as the lack of coupling between theory and empiricism and ethical aspects) as well as the opportunities (such as capturing the complexity and temporality of teaching and learning processes) of learning analytics for teaching science and practice are also considered.
Eberle, J., Strauß, S., Nachtigall, V., & Rummel, N. (2024). The potential of learning analytics for research on behavioral learning processes: current and future significance for research on learning and instruction. Unterrichtswissenschaft. https://doi.org/10.1007/s42010-024-00205-5