Universities must turn to analytics to retain students

By: Chris Brown, Data Analytics Consultant, OCF
Published: Friday, February 20, 2015 - 11:39 GMT Jump to Comments

Individual students now contribute, on average, £27,000 to their university’s income. Retaining these students is key to running a financially viable operation, which can support student-learning needs.

Higher education in the UK has operated as a business for many years. Income sources are numerous, but the Higher Education Statistics Agency [HESA] suggests £11.7bn of income in 2012/2013 came from tuition fees - that’s over 40 per cent of the total.

Frustratingly, HESA says that more than 26,000 students dropped out of university during 2011/12 – the latest available figures. That means around one-in-15 undergraduates failed to complete the first year of their degree and numbers grew to almost a fifth at the worst-performing institution.

The stats also show 18.5 per cent of students – around 73,500 – are projected to fail to complete the course they started. They are either dropping out, transferring to another university or graduating with an alternative qualification.

Losing a student through drop out will have a financial effect on an education institution’s sustainability. Not only do they lose the income stream, they have to spend money replacing it.

So it’s essential for education professionals to get early identification of student issues and to know if, and when, a student is likely to leave. It’s a process that involves tracking pain points, monitoring behaviours that indicate there’s something amiss and analysing the whole mix.

A few years ago this would either be impossible or require high-level use of excel spread sheets and rules. Now, using High Performance Data Analytics [HPDA], this can happen more easily.

HPDA software, such as IBM SPSS or R, connects with existing university IT systems and data sources to retrieve data on demographics, pre-university life, academic status, pastoral/welfare concerns and social habits to build up a picture of the student.

This data is then modelled, following which it can be accessed and evaluated by a range of people.

For example:

- Deans of specific schools can identify at-risk students and receive data-driven recommendations of the most successful intervention strategies for preventing students from dropping out

- Professors and curriculum specialists can review predicted grades and get insight into which program changes should improve student performance and retention

- Professors and academic advisors can analyse and predict each student’s strengths and weaknesses within a subject area, as well as their final exam grades, providing more support if necessary

- Advisors can recommend a list of data-supported intervention strategies to professors for under-performing students

- Registrars, chancellors and vice-chancellors can track predicted retention and graduation rates, along with related financial information, to keep academic and financial targets aligned

This is already happening in the UK - Brockenhurst College in the New Forest is investing in Exceptional Student Experience (ESE) technology from IBM. The technology uses a mixture of the latest cloud, mobile, social and analytics solutions, which combine to help personalise the experience a student gets from the moment they enroll on a course.

The college is hoping to achieve a 15% reduction in students who are at risk of dropping out over the next five years.

It would be great to think that course content and inspiring lecturers alone are enough to keep students for the duration, but it’s just not the case. So many outside factors can turn contentment into disengagement driving up dropout rates.

Understanding these factors and acting upon them using analytics is essential for all universities.

Universities should be skilling up now to ensure they can capitalise on HPDA – retain students longer and not lose out to other universities. They’ll need individuals with a statistics background, an understanding of mathematical modelling, and competence using software.

Plus, they may also need skills for thinking and writing in parallel code, understanding and knowing how to extract value from unstructured data with its different formats and sources of information.

These skills may already exist within the University, but they may need to recruit, train or seek out willing students to help. Skills are in very short demand though.

Therefore, one of the quickest routes to gaining these vital skills is turning to industry and commercial suppliers that can demonstrate proven deployment skills and close relationships with the primary analytics vendors, such as IBM, to help universities understand their analytics challenges and needs, plus provide the necessary hardware and software solutions.

A good supplier should be capable of developing a bespoke analytics solution depending on your requirements and support the deployment with good project management and proof of concept tests.

Everything is now in place for 2015 to be the year of analytics. Its pretty clear universities have the student data available to analyse, the analytics technology and supplier community is ready to help; we just need bold universities ready to make the leap.

To Retrieve their data, use analytics to Route out and identify potential challenges, and use knowledge to Retain their students for the whole duration. You might call that the three Rs of analytics?

Chris Brown is Data Analytics Consultant at OCF.

The views and opinions expressed in this article are those of the author(s) and do not necessarily reflect the official policy or position of The Information Daily, its parent company or any associated businesses.



Outdated infrastructure and an increasingly fragmented market threaten the future of technology-enabled integrated care.

County Durham voters back devolution in the North-East, Sir Digby Jones considers run for West Midlands mayor…

The recent launch of The Mayoral Tech Manifesto 2016 on London’s digital future, sets out a clear agenda…

The manufacturing industry is currently facing scrutiny from parties concerned for its survival. Far from facing…

Almost a year ago, I made some predictions for what would take place in government and public sector customer…

Sheffield, Warrington and Doncaster announce cuts, Lincolnshire is held to data ransom, fight begins for West…

Working for an education charity delivering numeracy and literacy programmes in primary schools, I’m only…

Northamptonshire County Council recently received the maximum four star rating from Better connected after putting…

Historically, the entrance of new generations into the workplace has caused varying levels of disruption. The…

Following another commendation for digital services, Surrey County Council's Web and Digital Services Manager,…

We cannot carry on spinning the roulette wheel that is cyber security, knowing that the “castle and moat”…

This week David Cameron wades into row over £69m of cuts planned by Oxfordshire CC; Stoke on Trent plans…