Bildungsdatenwissenschaft: ein Paradigmenwechsel für die Methodologie der Erziehungswissenschaft?

This article analyzes educational research approaches in the context of educational data science. Based on traditional research paradigms, it examines the extent to which educational data science implies a potential paradigm shift in educational science. The increasing digitalization and the resulting availability of large amounts of data open up new possibilities for gaining insights into educational processes. Educational data science is defined as the application of methods from computer science, statistics, and related disciplines to pedagogical phenomena. Central constructs such as big data, machine learning, data mining and data analytics are explained, with their interdisciplinary basis emphasized. The article also highlights the diversification of educational data science into research strands such as educational data mining and learning analytics. Despite the potential offered by the analysis of large data sets in the context of education, an analysis of educational science publications shows a low presence of data science methods. The article therefore argues for a stronger integration of educational data science into educational research by expanding curricula, continuing education programs, and promoting interdisciplinary cooperation in order to enhance the methodological diversity of educational science while addressing methodological and ethical challenges.