Using Moodle for Learning Analytics

The European Commission recently published a Communication on the Digital Education Action Plan which identified three priorities for action towards making better use of innovation and technology in education and training.... [ view full abstract ]

The European Commission recently published a Communication on the Digital Education Action Plan which identified three priorities for action towards making better use of innovation and technology in education and training. The third of these is Improving education through better data analysis and foresight. Educational organisations now have larger repositories of student data than ever before and are being challenged to use this to improve teaching and learning and related systems and to enable data driven decision making to take place. The research fields of educational data mining and learning analytics have developed to process and analyse this raw data so as to uncover useful insight from it.  

One potentially useful source of data for educational organisations is that contained in Learning Management Systems (also sometimes referred to as Virtual Learning Environments or Course Management Systems) which allow learning providers to manage their e-learning provision through the creation of online learning spaces. These systems provide the ability to quantify students learning behaviour and interactions in the learning space in ways that are more difficult to do in traditional ‘face to face’ learning.   

Data extracted from these systems, once processed and analysed, has the potential to  provide useful information to help improve quality of teaching, identify students at risk of early exit or course failure, to identify course topics that students might be having difficulty with, and to help students take control of their learning among other things. However, identifying useful data and converting this data into meaningful metrics of student engagement can be a challenging task particularly given the variety of ways in which learning management systems may be used to support module delivery. In addition, changes to the data model arising from software upgrades can add to the challenge of training models on historical data.  

This paper reports on early results from one research project focusing on the application of learning analytics to the open source learning management system Moodle as used in a further education and training setting in Ireland. The full data set consists of activity records of thousands of users across multiple courses (400+) over 8 years. This analysis is focused on a subset of the data with the ultimate goal to investigate the feasibility of using Moodle activity data to build an effective early warning system for at risk students on blended learning courses.