Tuesday, 4 May 2010

"There's something wrong with our bloody opinion polls today..."

This is heretical territory, but if there's one tale that's been repeated time after time by those out on the election trail, from every party, it's that what they find on the doorstep doesn't reflect the opinion polls. Even Jackie Ashley in the Guardian gave a coded warning that things could be much worse for Labour than the polls suggested. The polls, I think, have failed to capture the national mood of anti-politics, and perhaps anything could happen on Thursday. Your guess is no better or worse informed than the foremost TV pundits.

And remember how wrong the polls were in 1992; out of 50 published polls during the election period, 39 predicted a vote share that would have resulted in a hung Parliament, eight suggested an outright Labour win and only three predicted a Conservative majority.

Interesting times.


Anonymous said...

If you remember the polls in the run-up to the 1992 election, you must remember the post-election excuses the pollsters made.

It wasn't that the pollsters were biased nor that their methodologies were wrong in any way. No, it was the Conservative voters who lied to them.

The '92 election was the first I really followed and I well remember the "Not my fault, guv'nor" attitude of the polling companies who were not in the least interested in examining why they had got things so wildly wrong. If the Tories pull off a majority, I expect to hear the same foolish excuses from 1992 regurgitated once more.

Chris said...

The pollsters claim to have fixed the 'shy Tory' problem (by asking "Don't Knows" how they voted last time around, and assuming they'll do the same again). But then what was it the American chap said about 'unknown unknowns'?

The Great Simpleton said...

One of the problems with the Uniform National Vote that pollsters use is it doesn't pick up popular local independents. Given the expenses scandal and the general anti politics sentiment there is going to be a backlash, especially in seats where the incumbent is standing again, especially where they were seen as fiddling their expenses.

The BBC did a radio program on this a couple of months ago and at the top end they reckoned there could be as many as 60 plausible independent candidates standing. They may not get elected but they can certainly affect the outcome in those seats.