YouGov did not predict that Jeremy Corbyn will lose his seat to the Lib Dems

“The latest YouGov poll forecasts Jeremy Corbyn will lose his seat to the Lib Dems after 36 years!” 

Islington Liberal Democrats, 29 June 2019

At the weekend the Islington branch of the Liberal Democrats claimed that a new poll by YouGov had forecast Labour leader Jeremy Corbyn would lose his seat to them in an election. 

This is incorrect. The poll wasn’t a forecast, YouGov said nothing about the results in specific seats, and you can’t reliably predict how local seats will vote in an election based on a national poll.

The figures are based on a YouGov poll

YouGov conducted a survey last week which asked over 2,000 adults in Great Britain who they would vote for in the event of several different hypothetical scenarios.

One such scenario involved a Corbyn-led Labour Party standing on an election platform of delivering an alternative Brexit deal, which would keep Britain in the customs union. The scenario also had Boris Johnson as Conservative leader, Jo Swinson for the Lib Dems and Nigel Farage for the Brexit Party.

On a national scale, the poll suggested this platform would result in Labour losing 39% of its 2017 voters to the Lib Dems and ending up with only 17% of the vote overall, behind the Brexit Party (19%), the Conservatives (24%) and the Lib Dems (30%).

The first thing to note is that this is not a typical survey of voting intention: the poll unusually set out a 100-word hypothetical scenario before asking respondents who they would vote for. 

It’s also misleading to present any results from this survey as a “forecast”, as the Islington Liberal Democrats did, because it assumes that the given hypothetical scenario will be the case at the time of the next election.

A national poll for local people

There’s another big problem with using the poll to make predictions about specific parliamentary constituencies. 

Flavible, a polling website that is not affiliated with YouGov, extrapolated the survey’s national results and localised them to give a projection for each Westminster seat in England, Wales and Scotland.

For Jeremy Corbyn’s seat of Islington North, the website projected that in the event of the scenario mentioned above, the Liberal Democrats would win 31.3% of the vote to Labour’s 30.1%. 

This would be a significant change from the 2017 election, when Mr Corbyn won the seat with 73% of the vote and the Lib Dems came third with 9%. 

The main problem here is that you can’t reliably convert national polling results into local areas. 

Flavible’s stated methodology is a ‘weighted swing’, whereby changes in support for the parties at the national level are applied to each constituency, with the numbers weighted by things like which way the seat voted in the Brexit referendum and whether it changed hands in either of the last two elections. 

The evidence suggests that this does not always yield reliable results. For example, a localisation of a YouGov poll from 26 June suggested the Brexit Party would win nearly 25% of the vote in Islington North. 

This contrasts sharply with the vote share that the Brexit Party is estimated to have secured among  Islington North constituents in the European elections on 22 May—around 9%.