Two post-mortems into the general election come out today: the pollsters’ examination of how their surveys got the election so wrong, and Labour’s latest internal inquiry into how it lost that election.
The first report, which is the preliminary findings of an independent inquiry set up by the British Polling Council and the Market Research Society, has more surprising information in it than the Beckett report which, when it it published later today, is expected to say that Labour lost because voters didn’t trust the party on the economy, leadership, or immigration. The pollsters seem to have succumbed to ‘herding’, which is when individual companies alter their sampling formulae to ensure they are producing results in line with those published by others in the same field. And those samples contained too many Labour voters and too few Tories.
Of course, the two reports are interlinked because Labour looked at those polls and assumed it would, with one last heave, get into government. But if the finding about the pollsters’ sampling is correct, what those around Ed Miliband studying the figures would have been seeing when they looked at polls telling them that voters didn’t trust them on the economy, leadership or immigration was a rather more sympathetic picture than reality because too many of those voters in those polls were Labour-leaning.
The pollsters seem genuinely keen to alter their methodology to ensure they don’t make the same mistakes in 2020. To appear at all complacent would undermine them commercially. But the same is true for Labour electorally. The question is whether the current leadership will come away from the Beckett report today with the same resolve to change the party.
Comments