Why were the polls so wrong?
Latest analysis from the British Election Survey on why the opinion polls were up the creek can be read here.
They think the explanation had less to do with "Shy Tories" and more to do with the polling companies not estimating correctly what proportion of people would vote.
The conclusion reads:
"Our analysis of the post-election BES data makes us much more sceptical about late swing, “don’t knows” and Shy Tories. By contrast, we are leaning strongly towards differential turnout as part of the explanation and think that it’s likely that sampling and weighting also played at least some role."
Which roughly translates as
"There was a problem with people lying to pollsters, but it wasn't so much Tories lying to the pollsters about being Tories as people who described themselves as Labour voters not being honest enough to admit to admit they were too lazy to vote."
They think the explanation had less to do with "Shy Tories" and more to do with the polling companies not estimating correctly what proportion of people would vote.
The conclusion reads:
"Our analysis of the post-election BES data makes us much more sceptical about late swing, “don’t knows” and Shy Tories. By contrast, we are leaning strongly towards differential turnout as part of the explanation and think that it’s likely that sampling and weighting also played at least some role."
Which roughly translates as
"There was a problem with people lying to pollsters, but it wasn't so much Tories lying to the pollsters about being Tories as people who described themselves as Labour voters not being honest enough to admit to admit they were too lazy to vote."
Comments
You will find the answer on facebook and twitter before you find it in the polls.
Agree with you absolutely in that the facts demonstrate something is badly wrong with the assumptions and methods used by opinion pollsters, and a wise person will not trust anything they say until they have demonstrated that they have learned from their mistakes and corrected them.
Which is not to say that attempts to work out what they were doing wrong - like the BES paper I linked to - are of no value.
I presume when you say that "you will find the answer on facebook and twitter before you find it in the polls" you mean don't trust any of them?