‘Brave’ pollsters frighten May with hung parliament call
YouGov then said it was 'important to understand the limitations' of their new model.
Britain is heading for a hung parliament as the Conservatives lose their majority and Labour enjoys a surge of support, according to a shock projection from YouGov.
If YouGov is right and the result is replicated on 8 June it would be a disaster for Theresa May, who called the snap general election in a bid to secure a huge majority ahead of the Brexit negotiations.
But after sending shockwaves around the political world last night, the pollsters quickly stressed the drawbacks of their new model for the 2017 general election.
“It is important to understand the limitations of the model results. First, they are estimates of current voting intentions, not a forecast of how people will vote on 8 June,” said YouGov chief scientist Doug Rivers this morning.
“Panellists tell us how they intend to vote, but they may change their minds and we do not attempt to quantify this uncertainty. Second, the samples in each constituency are too small to be reliable by themselves and are subject to more than just sampling error.
“To compensate for small sample sizes, we rely on a model that pools data across constituencies.”
The YouGov analysis for The Times puts the Tories on course to win 310 seats - down from the 330 they held when the election was called, and 16 short of an overall majority. Labour, on the other hand, would see their number of MPs jump from 229 to 257.
YouGov’s chief executive stressed the central projection that “allows for a wide range of error” and that "the Tories could end up with as many as 345 and Labour as few as 223".
Nevertheless it is rare for a polling company to produce a seats prediction and one rival pollster, Britain Thinks founder Deborah Mattinson, tweeted that YouGov had been “brave”.
Others pointed out that YouGov has a mixed track record. Its 2015 general election eve poll had Tories and Labour neck and neck on 34% when the Conservatives won with a clear seven-point lead the next day.
On the YouGov website, Rivers sets out how the new model works by interviews approximately 7,000 panellists about their voting intentions in the election. Over the course of a week, data are collected from around 50,000 panellists.
He states: “The idea behind MRP is that we use the poll data from the preceding seven days to estimate a model relating interview date, constituency, voter demographics, past voting behaviour, and other respondent profile variables to their current voting intentions. This model is then used to estimate the probability that a voter with specified characteristics will vote Conservative, Labour, or some other party.”
Steve Parsons/PA Wire/PA Images.