Pollsters blame ‘unrepresentative’ samples for failure to predict Cameron election win
Polling experts have concluded that reliance on ‘unrepresentative’ samples was at the heart of their failure to predict the outcome of the 2015 general election, following an exhaustive inquiry into what went wrong.
Before the vote took place numerous polls were published indicating that Labour and the Conservatives were on practically level pegging, leading most to surmise that the country faced a period of post-election political horse trading to determine the new government.
In the event, of course, the result was nothing of the sort with the Conservatives enjoying a clear (if small) majority with the Conservatives winning 36.9 per cent of votes to Labour’s 30.4 per cent – prompting much soul searching and handwringing by polling experts left with egg on their faces.
At the heart of the problem is the apparent difficulty pollsters had in tracking down Conservative voters who tended to be busier, on average, than their Labour counterparts with the Conservatives 11 points ahead amongst those requiring between three and six calls to pin down.
These findings stem from a poll conducted after the election by The British Election Study and British Social Attitudes Survey conducted by NatCen, which outlined the state of play amongst the parties much more accurately.
Smarting from an avalanche of criticism polling experts have said that they cannot get everything right all the time and point out that they accurately predicted the SNP surge in Scotland and the UK-wide collapse of the Lib Dem vote.