Skip to content
The UK's only independent think tank devoted to higher education.

Why do academics keep getting election predictions wrong?

  • 15 May 2017
  • By Simon Goldsworthy

This guest blog has been kindly provided by Simon Goldsworthy, Professor of Public Relations and Advertising and Head of Department of Communication & Media (Communication, Public Relations, Journalism & Film Studies), Richmond University, Kensington Campus.

Shortly before the referendum on the UK’s place in the EU, the Political Studies Association (PSA) polled its members about the outcome. Almost 500 academics came forward, and almost 9 out of 10 forecast a Remain victory. This wasn’t a one-off – a similar failure occurred when they tried to anticipate the outcome of the 2015 general election – nor did their counterparts in the US fare too well when it came to the Presidential campaign. This time around in the UK, it might be hard to get the overall result wrong but, when it comes to the detail, will they miss the target again?

I’m at the academic end of a Public Relations and Communications Association inquiry into the wider failure of political predictions – and what can be done to improve them (so all contributions welcome). Unsurprisingly, a Times Higher Education survey before the referendum revealed that around 9 in 10 academics favoured a Remain victory, an eerily similar percentage to the proportion predicting a Remain outcome. Was there perhaps some connection – however unconscious – between academics’ own political views and what they expected to happen? Did they miss vital clues and fail in their duty to try to understand what was going on, however much it upset them?

One telling finding of the PSA survey was that academics anticipated a referendum turnout of 61%, when in fact it was just over 72%. The under-reported truth about the referendum is that turnout was critical. Voters in Remain majority areas were – overall – less enthusiastic to vote than those in Leave ones. We often hear about people in Scotland and Northern Ireland voting to remain, but are seldom reminded that, of voters in the two areas, only 67% and 63% respectively bothered to vote, as compared with 73% in England and 72% in Wales. Glasgow voted overwhelmingly for Remain, but only 56% of its voters took part. The most pro-Leave area, Boston in Lincolnshire, witnessed a 77% turnout, whereas the turnout in England’s most pro-Remain area, Lambeth, was 68%.

This differential in enthusiasm was surely crucial, but does not feature in the predictions of the PSA’s expert members. Was it because the Leave movement was taking place among people with whom, too often, they are unfamiliar? Sometimes it is said contemporary academics have retreated into a world of their own, a trend exacerbated by the extensive use of social media, which enables them to inhabit virtual gated communities with an increasingly uniform outlook. Universities, with so many staff and students from other EU countries (and such great hopes of continued EU links and funding), may be finding it too much of a challenge to understand the society in which they are based.

There is a pitter-patter of anecdotes to support this. One well-known Vice-Chancellor is rumoured to have told his staff that he didn’t know anyone who supported Brexit. At Sussex, they have discussed how to deal with right wing attitudes in the classroom, while at Cambridge a don staged a nude protest against Brexit.

Wishful thinking may be rearing its head again. At the time of writing, there seems to be a lot of talk about a progressive alliance but more resistance to acknowledging that the shift of UKIP voters towards the Tories – Ukippers, although all but unknown on campuses, outnumbered the combined Lib Dem and Green vote at the last election – is likely to have far more of an impact on the general election result.

What is to be done? The calling of the election means we’ve taken a break from our deliberations, but we have started to consider solutions. Universities would understandably resist formal interference in their affairs, even if it had the laudable aim of opening them up to wider thinking, forcing them to test their ideas and –  surely – making debates less predictable and more stimulating. But, undoubtedly, they should bring in more guest speakers with a wider range of perspectives and expand their reading lists to include more contrasting views. Any temptation to use official social media accounts for political campaigning and invective must be guarded against. And those universities which fail to open themselves up to society must accept that’s a legitimate subject of concern. Once universities start to think that they might damage their reputations and lose good students, and perhaps even the research funding that comes with them, they might start to think again.


  1. The Bunge says:

    An exceedingly welcome breath of fresh air – congratulations! I have long-felt that academics/activists/experts try to ‘talk the market up’ in favour of their personal positions, even when contradictory evidence indicates something very different. The adage ‘none so blind as though who will not see’! So thank you for bringing this important matter into the open.

  2. Morris says:

    This is refreshing. Some teachers seem to forget their job is not to teach students what to think,but to teach them how to think.

  3. John says:

    Most of the blog claims that research among academics failed to predict the size of the leave vote. However, nearly all polls among very different demographics did the same, which is nothing new.
    The final part claims universities are too insular and need to open up to wider thinking. This is a valid point, specially with regard to the blogger’s own. However, would John McDonnell or Dianne Abbot really be invited to address the students? I don’t think so.

  4. Nicholas Bryars says:

    A related theme might be this. Had the UK had a system of proportional representation in 2015, we would be about 6 months away from a general election having left the EU already because we would have had a Conservative/UKIP coalition government. I wonder how many academics would have welcomed and/or predicted that outcome?

Leave a Reply

Your email address will not be published. Required fields are marked *