A Learning Analytics Approach to MicroMasters

The digitalisation of learning platforms has enabled the recording of a huge amount of data about learners and learning in general.

The combination of this source of information and the advances in data processing and machine learning, often labelled ‘learning analytics’, has enabled researchers to approach student learning from a stronger quantitative perspective. The upcoming launch of MicroMasters at the University of Edinburgh provides an opportunity to begin to understand learning trajectories on online courses through data modelling. MicroMasters focus on a wider, more focused participation from the learner and mimics the efforts of pursuing a full MSc degree in a smaller often extended with a capstone project. Furthermore, it has a hybrid pricing structure designed to widen participation. The data modelling approaches allow us to build a predictive model that can be used to discover typical usage data combined with learner attributes. In an initial application, this model can be used to verify the learner’s position, but in a later stage will be beneficial to predict the outcome of new students even at early stages of their learning trajectory. Hence, an underpinning learning recommender system can be implemented which can later be adopted to other courses by retraining the models.

Research areas
Data Society
Research team

PI: Dr Johannes de Smedt (University of Edinburgh Business School)

PI: Dr Michael Gallagher (Centre for Research in Digital Education)

Key contact
Dr Michael Gallagher

Principal’s Teaching Award Scheme (PTAS)