Digital Transformation
Showing results for 
Search instead for 
Did you mean: 

Digital Footprints and big data prediction helps maximize students' performance


Among Big Data’s myriad uses is tracking human behavior, which has obvious benefits for customer service and marketing, but also could be a powerful tool in tracking not just preference and sentiment, but human performance.


Over the past 18 months, Nottingham Trent University has used Big Data analytics to identify students who are underperforming and in danger of failing or dropping out. This lets the university intervene earlier, leading to more successful student outcomes—which, of course, is a university’s key metric.




NTU is one of the largest universities in the UK, with 28,000 students. It already had a rather low dropout rate of 7 percent, but working with HP Software and DTP SolutionPath, an HP partner, the university looked to do even better. NTU leaders wondered whether the university could use students’ existing digital footprints to maximize their achievement, whether a student was on the edge of failure or just not reaching full potential.


Studying the student

Universities know that tutor support is very important in correcting student under-achievement. The earlier the support, the better. But the busy tutors have from 12 to 20 students—so how could modern prediction analysis techniques help tutors single out the students most in need of help?


NTU gave DTP SolutionPath a mass of anonymized historic digital footprint data so that they could build a prediction model. Footprint data came from six sources:

  • Tutorial attendance
  • Online teaching resource usage
  • Assignment e-submissions
  • Library book loans
  • Campus swipes
  • Academic achievement history

 NTU called the resultant prediction score the “student engagement rating”. It’s plotted for each class, and then each student has his or her engagement rating plotted against the class average.


Tutors receive the information as a simple dashboard, and also get email alerts when a student starts to drift below the class engagement average. Mike Day, director of Information Systems at NTU, says that the design of the alerting algorithm is important. “Too much email and we spam the tutors. Too little and they might miss something.”




Mike is a big fan of crowdsourcing. The next-generation alert engine is being crowdsourced by the university itself. While I was visiting Mike’s department, I saw posters encouraging students to go to a web site to crowd-design the university’s new mobile app.


High marks

The prediction system is used not only to help students who are in danger of dropping out. It is also used to help those who could get a 1st who are tracking a lower 2nd, for example—something like helping a C student earn an A.


NTU estimates the tutors are able to intervene six to eight weeks earlier than they would without the system. Eighty-three percent of tutors said the system has resulted in positive changes in workgroup interaction; 100 percent of tutors want to keep the system.


I asked Mike Day about the return on investment. He replied that, sure, he could easily work out the very compelling ROI. But he emphasized that ROI is not NTU’s focus.


“We believe that our primary duty is to maximize the potential of our students,” he said. “Sure, the system saves them, and us, money, but more importantly it can positively affect the lives of our students. For us, this is more important than the savings it brings.”


Balancing privacy

What about privacy? Mike said that there was widespread concern before the system was put into use, but that there is now broad agreement that the right balance has been struck between privacy and benefits. As always, privacy is not absolute—it’s always a balance.


I asked whether NTU planned to use HP Autonomy’s ability to understand human interaction, maybe looking at students’ twitter streams and other postings on social media sites. Mike felt that this was one step too far, and that the invasion of privacy was unlikely to be outweighed by the incremental benefits.


Advanced studies

I asked Richard Gascoigne, The DTP SolutionPath's lead on the project, how else this technology might be used. Needless to say, he’s been thinking about this a lot.


“We may be able to use similar data footprints to predict likely re-offending of prisoners. We might also be able to use digital footprints to predict children or families at risk. Or define better care plans for patients, this is a truly a universal tool.”


This is a situation where “human understanding” analysis may work in conjunction with digital footprint prediction. Already, HP Enterprise Services is working with a local authority in the UK to derive human meaning from police, ambulance, hospital, and social services reports. The analysis is looking for report text that indicates someone might be a victim of abuse. Perhaps digital footprint analysis might further enhance this analysis.


Back at NTU, the system goes from strength to strength. Mike says that they already know from the data that those students with high engagement are twice as likely to get a good degree, while even those students with a good academic background, but a low engagement score, only have a 40% chance of staying the course. He believes that the benefits from further innovation are huge, from helping understand how support might be even better to showing the way to a personalized educational experience.

Mike Shaw
Director Strategic Marketing

linkedin.gifMike Shaw

0 Kudos
About the Author


Mike has been with HPE for 30 years. Half of that time was in research and development, mainly as an architect. The other 15 years has been spent in product management, product marketing, and now, strategic marketing. .

Jan 30-31, 2018
Expert Days - 2018
Visit this forum and get the schedules for online HPE Expert Days where you can talk to HPE product experts, R&D and support team members and get answ...
Read more
See posts for dates
HPE Webinars - 2018
Find out about this year's live broadcasts and on-demand webinars.
Read more
View all