5 Learning analytics

Learning analytics is the collection and analysis of usage data associated with student learning (Brown, 2011). The purpose of learning analytics is to observe and understand learning behaviours and to improve completion rates, pass rates and student satisfaction.

Educational institutions gather a large amount of student data such as demographics, modules/courses a student is studying, when and how students engage with online activities, borrow resources from the library, about their workload, formative and summative assessment results, and so on. However, the amount and types of data collected by each institution may differ depending upon the nature of the institution (e.g. an online distance-education university compared to a traditional face-to-face university) and the institution’s strategy on what data to collect and use.

Any interaction with a web-based system results in ‘digital breadcrumbs’, which can be tracked, and this data could be used to better understand what students do and to personalise learning experiences (Wagner and Ice, 2012). With the widespread use of virtual learning environments (VLEs) – also known as learning management systems (LMSs) – educational institutions now collect increasingly large sets of data (commonly referred to as ‘big data’) of student interactions with institution-based computer systems. The data not only tracks whether students are using a VLE, but what specific resources are being accessed, such as whether they are using other online tools such as blogs, wikis, discussion forums, recorded lectures, etc.

Big data describes the significant growth in volume and variety of data that is no longer possible to manage using traditional databases. Analytics refers to a set of software tools, machine-learning techniques and algorithms used for capturing, processing, indexing, storing, analysing and visualising data (Daniel, 2017). Learning analytics provides the means to extract information from learning-related big data.

The Society for Learning Analytics Research (SoLAR, 2011) defines learning analytics as: ‘The practice of developing actionable insights through the collection, analysis and reporting of data about learners and their contexts, for purposes of understanding and optimising learning and the environments in which it occurs’. Learning analytics is an innovative research area that is increasingly exploiting data analysis techniques being offered by emerging disciplines of data science and artificial intelligence.

Institutions make use of tools, dashboards and reports for learning analytics. Learning analytics provide practice-based evidence to help to answer questions related to:

• student progress and monitoring; for example, how many students completed an online quiz; which students didn’t submit their assignment; student withdrawals from a module in a particular week
• learning design; for example, lower than expected engagement with an online discussion forum; students not attempting certain online activities in the order that was intended.

The potential benefits of evidence from learning analytics are to influence:

• student retention and progression (Bronnimann et al., 2018), such as addressing knowledge and skills gaps and intervening to provide additional support at strategic points before an issue becomes too serious
• student mental health and wellbeing (Jisc, 2020)
• learning design: knowing how resources are being used and how learning design could be changed to improve student engagement and student performance
• curriculum design and development, and curriculum evaluation processes.

In the next section, you will learn about the relationship of learning analytics with SoTL.