Do the standard measurements and results that constitute the foundation of contemporary learning analytics actually help learners? 

Henrik Køhler Simonsen: On the basis of our data, we would argue that conventional learning analytics is not enough. The automatic extraction of primarily quantitative learning analytics data is not adequate if we want to actually help learners. To help learners improve their practice and offer them action-oriented advice on how to improve their working methods, we need denser and more qualitative data.

Does your empirical data shed any light on this? 

Henrik Køhler Simonsen: Our empirical data seem to suggest that the collection, analysis, and interpretation of qualitative data by means of screen recordings do enable the mentor or teacher to offer concrete action-oriented advice to the learner. We argue that both types of data are necessary and that the combination of both quantitative and qualitative data seems to be an important step forward.

How can the use of Learning Analytics qualitatively improve the learning process? 

Henrik Køhler Simonsen: On the basis of our data, we argue that qualitative data in the form of, for example, screen recordings does improve the learning process because qualitative data enables mentors or teachers to first analyse HOW the learner works in practice. They can then offer a learner concrete advice on HOW to improve her or his practice, including, for example, how to effectively use available tools.

What type of additional effort would be required to use Learning Analytic data in a different way if necessary, or let’s say in a more meaningful way? 

Henrik Køhler Simonsen: We would argue that automatic screen recordings or eye tracking recordings would be powerful tools in our quest to learn more about the learner. We are not necessarily arguing that we need integrated screen recording tools, but we argue that the mentor or the teacher and the learner in particular would benefit considerably from a structured way of using qualitative learning analytics data.