Universities pulling together access agreements for next year should have learning analytics in their sights.
In recent years, the Office for Fair Access (OFFA) – which regulates fair access to higher education (HE) in England – has promoted a “student lifecycle” approach to access. This means, in the words of OFFA’s director, Professor Les Ebdon, that “an effective approach to access should not stop at the front door when a person enters higher education”.
In practice, this means OFFA encourages institutions to ensure that students from non-traditional backgrounds are successful following enrolment. Since students from under-represented groups are more likely to drop out of their studies, there should be a focus on retention – still a significant problem for the UK HE sector. Latest Higher Education Statistics Agency (HESA) data shows that, last year, more than 29,000 full-time students (7.4% of the intake) were no longer in higher education after 12 months.
Boosting retention is an area where learning analytics could have a big part to play in supporting fair access. Evidence from around the world shows that effective use of insights from learning analytics can be used to achieve statistically significant increases in retention.
When learning analytics were employed at Columbus State University College in the US, retention rose by 4.2% (and 5.7% among low-income students.) In Australia, pilot schemes at the University of New England saw drop-out rates fall from 18% to 12%, while in the UK, Open University pilots have resulted in a 2.1% boost in retention. We have produced a more comprehensive briefing on the current evidence on learning analytics and student success.
There are two main ways that learning analytics can help power universities’ efforts in these areas:
1. Identification and intervention
Firstly, analytics are a powerful tool for identifying students at risk. Learning analytics systems draw data from across an institution into a single learning records warehouse. This might include usage data from the library and the virtual learning environment, as well as attendance records and grade data.
A learning analytics processor then compares data on individual learners with current and historical data to identify any students who might be at risk of dropping out, or not meeting their full academic potential. Research has shown that the predictive models used are generally reliable – in one case correctly predicting three out of every four students not progressing to the next academic year.
Interventions can then be put in place to support the students identified. This might be as simple as informing students that they are at risk, or might involve prompting tutors to communicate with students to discuss how they can best be supported.
Secondly, because data on student engagement with learning can be monitored in near real time, the effectiveness of interventions with students can be quickly assessed (and, if necessary, adapted) without having to wait for final examination and/or assessment results.
At a time when there is increasing focus on the efficacy of spending on access and student success, this can help institutions to review and demonstrate the effectiveness of their student support. We expect this will inform improvements to the guidance and support available across the board to whole cohorts of students, as well as interventions offered to individual students at risk.
In addition to the benefits of increasing retention, we think that effective use of learning analytics data insights could also become part of institutions’ commitment to OFFA in their access agreements.
Our analysis of 2017/18 access agreements found that 14 institutions explicitly mention learning analytics. Buckinghamshire New University, for example, highlights that it “intends to introduce learning analytics to inform the support, learning, engagement, retention and success of its students” as part of its efforts to establish a stronger culture and practice of data usage across the institution. Exeter University’s access agreement states that it is “developing effective learning analytics tools to enable both students and tutors to monitor performance more effectively and identify strategies to improve”.
We also believe that there is a compelling case for some of the funds that institutions are spending on learning analytics to be designated as “access agreement expenditure”, where institutions can demonstrate that learning analytics is part of their strategy for improving the outcomes of under-represented and disadvantaged groups.
We are encouraging institutions we are working with to consider how learning analytics data might feed into the development of their annual access agreement with OFFA – as well as more general efforts to promote fair access and student success within their institution.
Jisc has argued that all institutions which do not currently have learning analytics in place should give consideration to adopting it at the earliest opportunity. If your institution is interested in finding out more about learning analytics, please get in touch.