Learning Analytics – getting a picture of student engagement and performance


Learning Analytics (LAs) is a rapidly growing area in formal education that is worth paying attention to because of its rising prevalence and its power to help both students and teachers maximise course performance. The University of Brighton is expanding its successful learning analytics pilot. Here’s a quick overview for you:

What are Learning Analytics?

LAs piece together separate streams of information on learners’ activities to paint an overall picture of engagement with, and progress through, a taught course. This information is presented digitally (usually via a dashboard) in a meaningful way to help students assess where they are on the course (often in comparison to cohort averages) and to help teachers monitor their students’ engagement and the effectiveness of teaching interventions.

Why do it?

There is increasing evidence that providing more meaningful information to students on their engagement and progress earlier on in a course can help prevent students falling behind and dropping out, increases engagement, and improves performance. Gathering metrics on how students are feeling (self-reported by students) as well as performing can be extremely beneficial in highlighting disengagement early enough for it to be addressed. They can be used to help teachers assess the effectiveness of their interventions and respond to emerging learning needs. There are also institutional benefits such as quality assurance and monitoring.

How is it done?

LAs seek to gather any measurable data indicating student engagement or performance:

  • Assessments – any quizzes, tests, or assessments that produce scores
  • Attendance – records of attendance at lectures or class
  • Online activity – log-ins to studentcentral or engagement with online activities e.g. discussion boards
  • Library activity – student activity on borrowing or access to online resources

What are the challenges?

The idea is simple but the execution is much more complex. LAs are only as useful as the quality of the data they combine and the more data streams they combine, the clearer the picture is. However, any data that are collected must be consistent and valid, otherwise it risks presenting skewed information. LAs are dependent on:

  • Consistent data – gaps in data points weakens its validity so whole course and institutional systems often have to be updated and strengthened (e.g. attendance recording processes)
  • Multiple data points – the more data points the more accurate the picture
  • Accurate interpretation – students and teachers must be informed of the limitations of the information so as not to draw erroneous conclusions
  • Technical expertise – the technical demands to combine data feeds and present them meaningfully are highly demanding and require large-scale institutional system changes

Isn’t this just increased snooping or competition??

The purpose of LAs is to help students progress through their course and achieve their greatest potential. In these days of GDPR and information governance, it is important that any data-gathering is consented for and only used in a way that the person has agreed to. Students need to be informed of the LAs process and give consent as part of their course agreement. While it sound like increased monitoring, it is usually the case that these data are already in existance but LAs bring them together and present them in a meaningful way for learning.

Is BSMS doing it?

BSMS is not currently engaged in the University of Brighton pilot, largely due to the complexity of the data gathering on the undergraduate course. We will continue to work with both partner universities so as to benefit from Learning Analytics in the future.

Further info:

University of Brighton Learning Analytics pilot

JISC briefing on learning analytics and student success from January 2017 (pdf)

EDUCAUSE Institutional Analytics and the Data Tsunami

Print Friendly, PDF & Email

Leave a Reply

Your email address will not be published. Required fields are marked *