Polling inquest: YouGov and Co need to provide answers

How did the pollsters get the general election result so wrong? The inquest will be as intense as the one now going on within the Labour Party. YouGov’s Stephan Shakespeare has put his hands up, tweeting: “A terrible night for us pollsters. I apologise for a poor performance. We need to find out why.”

Mike Smithson of Political Betting says: “I got it totally wrong because I believed the polls - particularly the closing ones that we got yesterday morning.” He adds: “This will be costly.”

There are a number of people entitled to say “I told you so”, among them the Daily Telegraph blogger Dan Hodges who called it for the Tories a year ago.

Subscribe to The Week

Escape your echo chamber. Get the facts behind the news, plus analysis from multiple perspectives.

SUBSCRIBE & SAVE

Sign up for The Week's Free Newsletters

From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.

From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.

Sign up

“The basic political trajectory set in 2010 – cuts, hostility to cuts, gradual economic recovery, Tory recovery, Tory victory – is being followed without deviation,” he wrote. “In 12 months’ time Britain will go to the polls. And when it does, Britain will re-elect David Cameron.”

A month ago Adam Ludlow of ComRes highlighted the difference between phone and internet pollsters and showed that in the phone polls “the Conservatives have been leading all year”.

With a week to go before polling day two pollsters actually got it just about right: Ipsos-MORI gave the Tories a five per cent lead while Lord Ashcroft put them six points ahead.

Unfortunately for us poll-watchers, we gave more credence to their eve-of-poll findings, by which time all the major companies were calling Labour and the Tories “neck and neck”.

So what went wrong? Did the public tell the pollsters one thing and then, in the privacy of the polling booth, do another? Or did the pollsters, having spotted developing trends and voter concerns, fail to allow for them in their weighting?

It appears both are correct answers.

On Tuesday this week, the New Statesman’s Stephen Bush wrote an article for the New Statesman which, in hindsight, looks amazingly prescient.

He reported that there was alarm in the Labour camp that in its "mainstay" seats – those won with small majorities in 2010 that it was assumed would stay Labour this time - the "promise" (the number of people contacted by the party who have said they would vote Labour) was not holding up as it should.

The picture was worse in the party’s target seats, reported Bush. Canvassers were picking up Labour voters from 2010 who were now expressing doubts about Ed Miliband. One party strategist told him: “My expectation was that, thanks to the short campaign, those voters would be moving into our column. Instead, they are moving away.”

Looking back, it seems the pollsters picked up on these negatives when they interviewed voters about the issues that concerned them, but failed to carry these findings over to the voting intention results.

• The Mole: Will Cameron come to miss Nick Clegg?

For example, the pollsters saw all along the public doubts about Ed Miliband’s leadership qualities, and yet still had Labour ahead in voting intention polling.

A growing issue in recent weeks was the warning – or “scare story”, depending on your point of view – put out by Team Cameron that a Labour government would be “propped up” by Scots Nationalists who surely had no business governing the English at Westminster.

A YouGov poll for last week’s Sunday Times showed how potent an issue this might be. Forty per cent thought that “Labour would do a deal with the SNP, and it puts me off them”. Among Lib Dem voters the figure was 50 per cent and it rose to nearly 60 per cent among those saying they planned to vote Ukip.

Yet still the pollsters said Labour were on level-pegging with the Tories when it came to voting intention.

Is this where they went wrong? The public will need an answer if they are to have any confidence in future political polling: what, for instance, are we to make of polling by YouGov for The Times Red Box released earlier this week that suggested the majority of Brits would vote to stay in the EU. Is it true or plain wrong? Much hangs on it.

Certainly the pollsters will have to come up with better answers than the one given by one of their most high-profile representatives, YouGov president Peter Kellner, to the BBC’s David Dimbleby in the small hours this morning.

Politicians should not rely so heavily on polls, he said, adding: “We are not so far out as we were in 1992.” Hmmmm.