International university data

Posted by Duncan Ross On 11 December 2018

I was lucky enough to be invited to talk to the All Party Parliamentary University Group on the subject of big data – something of a dream invitation from my perspective: data, politics and universities in one go!

Of course, working for an organisation that creates university league tables puts me in an odd position. Our role isn’t to focus on the UK sector, but rather to look at higher education around the world.

Now obviously that’s of interest to politicians – how can you tell if your universities are delivering? And how can you evaluate if they are achieving whatever aims you might set for them? In theory you could look at them only within a UK context, but that would be limiting. After all the marketplace for jobs and businesses is international, and if your HE sector isn’t delivering then this is one factor that might impact your economic plans…

It’s also increasingly the case that higher education itself is international. It has become normal practice to have staff recruitment strategies that focus on finding the best talent from all around the world, and we’re happy to subsidise our institutions by recruiting international students, paying international rates. Often research grants are international.

Students are also increasingly willing to travel for the best education. One of the side effects of charging £9,000 per year (in England at least!) is that a US or European education now looks much more affordable than it once did – it certainly opens the mind. Why not travel to the Netherlands for a degree? If you live in Maidstone you’re closer to Maastricht University than the University of Durham.

But what resources are there to evaluate universities internationally? Precious few. There are some good datasets available – although most aren’t public or open data. The UK performs remarkably well here: HESA is fantastic. The LEO dataset is great. The NSS has stunning response rates. And sorry for the alphabet soup there.

But outside the UK the picture is much more… mixed. The US has some good data. College Scorecard is an initiative that should definitely be supported, for example. But other datasets are less good. And many are simply not present, or provide data that is less than consistent.

Even more challenging, of course, is that where this data exists it is to whatever definitions that country has decided, and these can be surprisingly challenging to compare across international boundaries. And that’s what we try to do with our rankings. Get data that is as consistent as possible, and compare universities.

Of course, we have our biases. Our biggest ranking, the World University Ranking, covers 1258 universities from 87 countries, but is focused on research. We also generate a set of teaching rankings – exploring how universities compare for the teaching mission across 12 countries (including the UK, US, and Japan). And early next year we will explore how universities are working towards the UN’s Sustainable Development Goals.  

These are never perfect comparisons, but they do, perhaps, shine some light on the worldwide nature of higher education.

Duncan Ross is the Data and Analytics Director at Times Higher Education (THE)