Analyse prozessbezogener Verhaltensdaten mittels Learning Analytics: Aktuelle und zukünftige Bedeutung für die Unterrichtswissenschaft

For several years, “learning analytics” have been growing as an international research field that focuses on collecting, analyzing and using complex and often multi-model digital trace-data produced by learners and teachers in digital learning settings; employing computational analyses or machine learning tools, these data are used to generate insights into processes of learning and instruction. The scientific community in the area of learning and instruction is currently exploring these developments. However, as researchers have started to recognize the potential of learning analytics, it seems worthwhile to think further about how adopting learning analytics approaches may benefit the field. In this paper, we provide insights into the flourishing area of learning analytics research and provide concrete examples how such approaches can help to expand existing theories on learning and instruction. We focus on self-regulated and collaborative learning on the one hand, and on the other hand on instructional design and teacher support based on learning analytics. We also consider risks and challenges that come with learning analytics (such as missing links between available data and scientific constructs, as well as ethical issues) but also benefits for research on learning and instruction as well as practitioners (such as ways to account for the complexity and temporality of processes during learning and instruction).