We tried to predict how students would impact the general election, and this is how we did

Did students win it? 

With the notable exception of YouGov and their innovative Multilevel Regression and Post Stratification approach, it hasn’t been a great election for the pollsters.

Now, we at Times Higher Education are not pollsters. Nevertheless, in the week after the general election was announced, we decided to see if we could do something a bit more useful than provide a prediction – we would try to understand which universities had the biggest potential influence on the election. Here’s what we produced. 

Our approach was to model the number of students who might be able to vote in a constituency who were associated with specific universities, and then combine it with the marginality of the seat.

Clearly this resulted in some usual suspects. It surprised no one that the University of Cambridge (Labour won the Cambridge constituency in 2015 by a margin of 599 votes) would be in the list. Or Derby University (Derby North, Conservative majority of 41 in 2015). Perhaps more interesting were such universities as Kent (Canterbury had a a Tory majority of 9,891 votes in 2015). 

So how did we do? 

At this stage let’s focus on the constituencies, rather than the universities. Well, Cambridge was a solid Labour hold. Labour voters should be saying “well done” to Anglia Ruskin! 

Next on our list were the constituencies of: Plymouth Sutton and Devonport (a Labour gain from the Tories); Leeds North West (another Labour gain); and Derby North (yet another Labour gain). Kudos to the students and staff of Plymouth University and the University of St Mark and St John; University of Leeds, Leeds Trinity, Leeds College of Art and Leeds Beckett; and the University of Derby. 

Of course, these were already very marginal seats. It's when we get further down our list that things get really exciting. Stand up Bath (where the Tories had a 3,833 vote majority in 2015); Brighton Kemptown (a 690 vote Conservative majority in 2015); and Portsmouth South (a 5,241 majority for the Conservatives in 2015). Not only did these all change hands – slightly unusually to the Lib Dems in Bath – but the swings were 9.8 per cent, 10.8 per cent and 9.4 per cent.

If these swings had been repeated across the country then prime minister Jeremy Corbyn would already be into his first reshuffle. 

The plaudits here, if you are not a Conservative voter, should go to the University of Bath and Bath Spa University; University of Sussex and University of Brighton; and University of Chichester and University of Portsmouth. 

But probably the biggest surprise of the evening (unless you are Theresa May) came in Canterbury, where the University of Kent helped to overturn that near 10,000 vote majority. 

We found things a little harder to call in Scotland, although North East Fife came within two votes of making us look good. Let's hope there aren't a couple of student members of the Lib Dems who forgot to vote (they could be from St Andrews, Abertay, Dundee, SRUC or Highlands and Islands). 

Of course, there are some caveats. We don't know if young people actually did turn out in significantly higher numbers yet – the evidence is that it is likely (perhaps jumping from 48 per cent to about 70 per cent), but overall turnout is only up 2.3 per cent on 2015. And it's also hard to separate out the effect of the collapse of the Ukip and Liberal Democrat vote in much of the UK. And, of course, we focused on Great Britain rather than the whole of the UK. 

Will we do it again? Yes, absolutely. Possibly as early as this autumn.

Duncan Ross is director of data and analytics at Times Higher Education.