Beware the Idiosyncrasies, or, Why Universal Learning Analytics Solutions May Fail

• In Event: 72.066 - Poster Session 16
  In Poster Session: 72.066-3 - Instructional Technology SIG Poster Session

Abstract

Higher education leaders are recognizing the value of learning analytics for improving not only learning and teaching but also the entire educational arena. However, well-established empirical evidence within the emerging field of learning analytics is lacking. This study uses Support Vector Machines as an alternative educational data analytics approach and investigates the idiosyncrasies of higher education institutions and area of studies when applying a learning analytics framework. Findings suggest that educational data is context specific and identical variables and indicators carry different meanings and can have different implications across educational institutions and area of studies. Future work includes further empirical validation and a full implementation of the learning analytics framework as a dynamic plug-in for digital learning environments.

Authors

• Dirk Ifenthaler, University of Mannheim

• Chath Widanapathirana