Learners in the Data Mist: Visualizations for Optimization of Learning Environments

Abstract

This master thesis presents an Learning Analytics artifact designed to support learning environment optimization. The Design Science research project developed an artifact through iterative design processes that were informed by both quantitative and qualitative evaluation methods. Through five iterations, ranging from a proof of concept artifact to a functional artifact that could serve visualization data through a REST API application. The case for this research project was a SPOC course managed by OsloMet, which offered their data as a case study to examine it with Learning Analytics processes. Each iteration is detailed in its own chapters, with sections labeled according to Dresch, Lacerda and Antunes Jr’s proposal for a Design Science research model. Each evaluation gathered data using a variety of evaluation methods to ensure multiple perspectives on the visualizations and the artifact’s performance factors. Especially the Stakeholders of the SPOC course were involved in semi-structured interviews to obtain functional requirements and establish user needs. The research project eventually produced a functional artifact, which we could use to test whether it was capable of supporting learning environment optimization. We found that the artifact had promising potential, but would be better evaluated through testing over several course runs and with the consumption of multiple data-sets.