Learning Analytics and MOOLs: there is much more than equipment usage logs

When dealing with online laboratories (either remote or virtual), it is crucial to monitor experiment activities and their outcomes closely, as well as to control how students access to these labs. A similar need also exists for evaluating student learning patterns and results, at both the individual and group level, so that customized study paths and contents can be proposed. To such aim, the research in the field of Learning Analytics has been very prolific in the recent decade, especially in the STEM area. This scenario becomes more challenging when online labs are offered as Massive Open Online Laboratories (MOOLs), large-scale and cloud-based infrastructure allowing students to access catalogs of online experiments on several topics as normally happens with traditional courses accessible via Massive Open Online Courses (MOOCs). This paper aims at focusing on the most relevant aspects of LA for MOOLs in terms of data requirements and data-related challenges, by examining four aspects that are tightly related to LA: data models and catalogs, data quality and scope, data privacy and ethics, and data visualizations. The resulting considerations can be used as a set of guidelines to take into account when designing a MOOL.