Learning analytics are a powerful tool to help learners and educators to act when action is most needed.
Difficulties and learning patterns can be visualized in real time, giving educators a personalized, more precise interaction with individual students. They are also a powerful change agent because credible, empirical facts bring legitimacy to decisions and allow decision-makers to target limited resources for greatest impact.
Today, learning analytics are most commonly used as an early warning system to prevent dropout and as means to gain greater insights into the factors that promote, and can be used to predict, student learning success. An aspiration in many of the initiatives currently underway, is to use analytics as a means to personalize learning whereby the artificial intelligence within the analytics system predicts and signals what the learner most needs to advance.
Today, it is possible to leverage learning analytics at reasonable cost. Most of what we do is already or can be digitized; cloud technologies enable new algorithms to be applied to huge amounts of data and the legislative requirements around data use are easier to meet.
This paper draws on four international case studies to explore what has driven learning analytics use, the outcomes that have resulted and some of the lessons learnt