Why student engagement data should be at the heart of higher education policymaking

There's a wealth of data out there on student engagement, says Emilie Sundorph of the Reform think tank – measuring it should be a compulsory part of the government's new Teaching Excellence Framework


By Emilie Sundorph

08 Sep 2016

Since the Coalition’s decision to treble tuition fees, the argument that university is too expensive has been a permanent fixture of the higher education debate. In recent years, these calls have grown louder. 

The so-called "graduate premium" is increasingly coming into question, with research now suggesting that high-value apprenticeships can be a more profitable route than some degrees. In a survey this year, only 37 per cent of students agreed that they are receiving value for money, a significant decrease from 53 per cent in 2012.

The government has set out to address the perception that some degrees are a poor investment. By introducing the Teaching Excellence Framework (TEF) – the new system of performance evaluation for universities that will determine the level at which fees can be charged – universities minister Jo Johnson is hoping to “wipe out mediocre teaching and drive up student engagement.” If universities fail to do so, they could face penalties in the form of decreasing tuition fees.


Theresa May's higher education shake-up makes BIS Sheffield closure look "perverse" – former BIS adviser
Department for Business names first chair of universities' Teaching Excellence Framework
Brexit shake-up lifts redundancy threat for some BIS Sheffield staff


This will hit them at a difficult time. Although the cap on student numbers has been lifted and applications have continued to increase, the pool of 18-year-olds is set to fall dramatically over the coming years. Meanwhile, Brexit is creating worries about funding opportunities and the future status of EU applicants.

While big data is often offered as a panacea to public policy questions, the potential to improve standards in higher education is real. By matching former students’ academic activities – from how much they access their institution’s virtual learning environment to how often they use the library – with their outcomes, algorithms can now predict future performance with a startling degree of accuracy. 

From before their studies commence to the day they leave university, the resources accessed by students can generate week-by-week estimations of how likely they are to drop out, and even their final grade.

Universities have put these tools into practice in a number of ways. Some give tutors sole access to this information; others feed the data straight back to students via apps. Each approach has its benefits, but both have led to positive outcomes. 

At the Open University, a dashboard of engagement indicators has allowed tutors to make timely interventions for struggling students, increasing retention rates by 2.1% on the previous year. At Nottingham Trent University, a pilot providing students with learning analytics apps saw 27% change their behaviour – for example increased attendance. In the US, where learning analytics is more widely applied, a survey showed that two thirds of students with access to learning analytics found this had a positive effect on their academic performance.

"From before their studies commence to the day they leave university, the resources accessed by students can generate week-by-week estimations of how likely they are to drop out, and even their final grade"

This is self-evidently beneficial to students. Once embarked on a degree, most have a strong interest in completing with the best result possible. Having access to a personal app with information on current performance and predicted outcomes also mirrors the service many are used to in other walks of life. When it comes to exercise they can set targets, follow progress and compare themselves to their friends – why should the same opportunities not be part of their university experience?

The benefits to universities are also clear. Increased retention rates will result in higher revenue from tuition fees. And as competition for students rises, no university wants to fall behind in the services they offer. Policymakers, who have been set the unenviable task of determining the quality of university tuition, will benefit too. The extent to which students utilise facilities and resources provides a measure of engagement, a crucial blind spot in the current performance framework. 

In a report published today, Reform recommends that measuring student engagement through learning analytics should be a compulsory part of the TEF. Such a move would sit neatly with the government’s ambition to widen university participation. 

The existing focus on employment disadvantages institutions with a high intake of women and students from black and ethnic minority backgrounds, as their graduate earnings tend to be lower – an engagement metric derived from learning analytics can help overcome that.

Whilst the absolute value of higher education is impossible to narrowly define, something is indisputably wrong if a majority of students leave with less than they expected. Learning analytics promises a new future where universities have better insights into the motivation of their students and as a result can create more personalised learning experiences. 

By including this data in the TEF, students will be better placed to judge the value of investing in higher education and government will be better equipped to penalise poor performance.

Share this page