Abstract
Having been spectacularly wrong about the 2015 general election—when they had predicted that Labour and the Conservatives were too close to call in terms of vote share, only for the latter to end up 7% ahead—the opinion pollsters knew they were on trial in 2017. Since the 2015 debacle, they had enjoyed some success with predictions for Corbyn’s leadership election in 2015 and 2016, Sadiq Khan’s victory in the London mayoralty race, and in the devolved elections in Wales and Scotland. Their performance in the 2016 referendum, however, was more mixed. In the last month of the campaign, marginally more polls predicted a Leave outcome than Remain, although most of the final polls predicted a win for Remain.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
John Curtice, ‘EU Referendum – How the Polls Got it Wrong Again’, The Conversation, 26 June 2016, https://theconversation.com/eu-referendum-how-the-polls-got-it-wrong-again-61639.
- 2.
Of the constituency polls conducted in 2017, three correctly identified the winning party—although given that two of these polls were in Tatton and Brighton Pavilion, this was not an especially onerous task. The poll conducted in Battersea would not have helped identify Labour’s eventual gain; ditto for the one poll conducted in Kensington, which had the Conservatives 17 points ahead in a seat Labour went on to win.
- 3.
P. Sturgis, N. Baker, M. Callegaro, S. Fisher, J. Green, W. Jennings, J. Kuha, B. Lauderdale and P. Smith, Report of the Inquiry into the 2015 British General Election Opinion Polls. Market Research Society and British Polling Council, 2016.
- 4.
For a good discussion on the changes, see Anthony Wells, ‘How the Polls Have Changed since 2015’, UK Polling Report, 27 May 2017, http://ukpollingreport.co.uk/blog/archives/9895.
- 5.
Andrew Hawkins, ‘Why the Huge Disparity in This Election’s Opinion Polls?’, The Times, 8 June 2017.
- 6.
For example, a Survation poll in late May 2017 for ITV’s Good Morning Britain reported that 82% of those aged 18–25 would vote and was greeted with disbelief.
- 7.
Qriously use a technique they call ‘programmatic sampling’, which involves mobile phone users responding to questions whilst using apps. Their poll was published in Wired, which later ran an article examining how Qriously had been so accurate (see Joao Medeiros, ‘Remember the Poll That Predicted the Labour Surge? We Explain the Methodology’, Wired, 9 June 2017, www.wired.co.uk/article/labour-election-result-poll-qriously), based on the fact that the poll had accurately predicted the Labour vote share. The article spent less time on the fact that the poll predicted a two percentage point Labour lead in an election which the Conservatives won by 2.5 points.
- 8.
Will Jennings, ‘The Polls in 2017’ in Dominic Wring, Roger Mortimore and Simon Atkinson (eds), Political Communications: The General Election of 2017. Palgrave Macmillan, 2017. Jennings excludes the Qriously poll from his calculations, but even with this included, the Labour range only rises one point and the conclusion remains the same.
- 9.
‘Daily Mirror GE2017 Poll’, ComRes, 11 May 2017, www.comresglobal.com/polls/daily-mirror-ge2017-poll.
- 10.
Matthew Smith, ‘How Popular are the Parties’ Manifesto Policies?’, YouGov, 22 May 2017.
- 11.
Andrew Grice, ‘Majority of British Voters Agree with Corbyn’s Claim UK Foreign Policy Increases Rise of Terrorism’, The Independent, 6 June 2017.www.independent.co.uk/news/uk/politics/jeremy-corbyn-poll-foreign-policy-terrorism-british-voters-agree-majority-latest-a7776276.html.
- 12.
Matthew Smith, ‘Jeremy Corbyn is on the Right Side of Public Opinion on Foreign Policy: Except for the Falklands’, YouGov, 30 May 2017, https://yougov.co.uk/news/2017/05/30/jeremy-corbyn-right-side-public-opinion-foreign-po.
- 13.
The only conflict on which Corbyn was in a minority was the Falklands War.
- 14.
For just one example, see Amanda Bittner, Personality or Platform: The Role of Party Leaders in Elections. Oxford University Press, 2011.
- 15.
Gideon Skinner and Glenn Gottfried, ‘Theresa May Has Big Lead as Most Capable Prime Minister’, Ipsos MORI, 26 April 2017, https://www.ipsos.com/ipsos-mori/en-uk/theresa-may-has-big-lead-most-capable-prime-minister. There had, however, sometimes been bigger leads—such as Tony Blair’s over William Hague in 2001.
- 16.
‘The Independent Eve of Election Poll’, ComRes, 7 June 2017, www.comresglobal.com/polls/the-independent-eve-of-election-poll.
- 17.
Stephan Shakespeare, ‘A Fifth of Would-Be Labour Voters Want a Tory Government’, YouGov, 16 May 2017, https://yougov.co.uk/news/2017/05/16/one-five-those-who-intend-vote-labour-want-conserv.
- 18.
Peter Kellner, ‘Why is Labour Surging – And How Worried Should the Tories Be?’, Prospect, https://www.prospectmagazine.co.uk/blogs/peter-kellner/why-is-labour-surging-and-how-worried-should-the-tories-be.
- 19.
See, for example, Matt Singh, ‘History Points to a Convincing Conservative Victory’, Financial Times, 24 May 2017; and ‘Theresa May’s Lead Has Dipped But is Stronger than Polls Suggest’, Financial Times, 7 June 2017.
- 20.
John Rentoul, ‘Be Careful with Your Election Predictions, the Opinion Polls Could Be Very Wrong. Again’, The Independent, 29 May 2017.
- 21.
Michael Savage, ‘From NME to No 10 – Could the Youth Vote Win Corbyn the Election?’, The Observer, 4 June, 2017. See also Anne Perkins, ‘Building Your Hopes up, Corbyn Supporters? Have You Forgotten 2015?’, The Guardian, 7 June 2017.
- 22.
See Doug Rivers, ‘Introducing the YouGov Referendum Model’, YouGov, 21 June 2016, https://yougov.co.uk/news/2016/06/21/yougov-referendum-model.
- 23.
- 24.
Despite an 11-point Conservative lead in 2015, Kensington was predicted by the YouGov model to be a toss-up with both Labour and Conservatives on 43%; election day saw both parties on 42%, with 20 votes between them.
- 25.
As the press release rather dryly noted, ‘the results of this survey should be interpreted in light of the … poor forecasting performance of similar expert surveys of predictions for the 2015 general election and the 2016 EU referendum’. See ‘Expert Predictions of the 2017 General Election: A Survey by Stephen Fisher, Chris Hanretty and Will Jennings on Behalf of the U.K. Political Studies Association’, Political Studies Association, 2 June 2017, https://www.psa.ac.uk/psa/news/expert-predictions-2017-general-election-survey-stephen-fisher-chris-hanretty-and-will.
- 26.
See Stephen Fisher, John Kenny and Rosalind Shorrocks, ‘A Post-mortem on Forecasts for the 2017 British General Election’, presented at the EPOP Conference 2017. The one caveat is that the betting markets did marginally better when it came to the Liberal Democrats.
- 27.
This poll attracted relatively little attention at the time, partly because it was not by a BPC member and partly because of its methodology—using an app on mobile phones.
- 28.
See Philip Cowley and Dennis Kavanagh, The British General Election of 2015. Palgrave Macmillan, 2015, p. 239.
- 29.
As they had, albeit with much more limited polling, between 1945 and 1955.
- 30.
Indeed, that scale of error on any one party was exceptionally rare—on a par with the Conservative vote share in 1992.
- 31.
This is discussed in Patrick Sturgis and Will Jennings, ‘Will Turnout Weighting Prove to Be the Pollsters’ Achilles Heel in #GE2017?’, Soton Politics, https://sotonpolitics.org/2017/06/04/will-turnout-weighting-prove-to-be-the-pollsters-achilles-heel-in-ge2017.
- 32.
See Will Jennings, Evidence submitted to the House of Lords Select Committee on Political Polling and Digital Media, 31 August 2017.
- 33.
ComRes, ‘GE2017: Our Take’.
- 34.
Described in John Curtice and David Firth, ‘Exit Polling in a Cold Climate: The BBC-ITV Experience in Britain in 2005’, Journal of the Royal Statistical Society A, 171 (2008): 509–39.
- 35.
John Curtice, Stephen Fisher, Jouni Kuha and Jonathan Mellon, ‘Surprise, Surprise! (Again). The 2017 British General Election Exit Poll’, Significance, August 2017.
- 36.
Some 47% said they opposed the policy, compared to just 28% who said they supported it (and 26% who did not know). In addition, majorities of respondents said that they would be more anxious about getting older, more anxious about securing a future for their children, caring for older relatives and owning a house (52%). See Survation, General Election 2017 Poll, 21 May 2017, http://survation.com/wp-content/uploads/2017/05/Final-MoS-Poll-190517GOCH-1c0d1h7.pdf.
- 37.
Steve Howell, Game Changer. Accent Press, 2018, p. 2.
- 38.
See, for example, Textor’s claim that British polling was ‘hopeless’ (Cowley and Kavanagh, The British General Election of 2015, p. 63) or in 2017 that it was ‘shit’ (Tim Shipman, Fall Out. William Collins, 2017, p. 373).
- 39.
See Shipman, Fall Out, pp. 403–4 and 527.
- 40.
Ibid., p. 527.
- 41.
See Chris Prosser and Jon Mellon, ‘The Twilight of the Polls? A Review of Trends in Polling Accuracy and the Causes of Polling Misses’, Government and Opposition (2018): 1–34. See also the conclusions of the House of Lord Select Committee on Political Polling and Digital Media, The Politics of Polling, Report of Session 2017–2019, HL Paper 106.
- 42.
John Curtice, ‘How Pollsters Have Changed Their Tactics’, The Times, 29 May 2017.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2018 The Author(s)
About this chapter
Cite this chapter
Cowley, P., Kavanagh, D. (2018). Wrong-Footed Again: The Polls. In: The British General Election of 2017. Palgrave Macmillan, Cham. https://doi.org/10.1007/978-3-319-95936-8_11
Download citation
DOI: https://doi.org/10.1007/978-3-319-95936-8_11
Published:
Publisher Name: Palgrave Macmillan, Cham
Print ISBN: 978-3-319-95935-1
Online ISBN: 978-3-319-95936-8
eBook Packages: Political Science and International StudiesPolitical Science and International Studies (R0)