The British Election Study suggests that differential turnout the most likely cause of the GE2015 polling failure
LAB supporters more likely to have overstated certainty to vote
There’s an important paper just out from the British Election Study on what went wrong with the polls on May 7th. Why did it happen?
The report by Jon Mellon and Chris Prosser does not give much credence to the “late swing”, “don’t knows” and “shy Tory” theories and argues that differential turnout was most to blame. Those over-stating their likelihood to vote were more likely to be those saying LAB.
A key part of the examination has been to check whether polling respondents who said they would voted actually did so. They do this by looking at the publicly available marked register. They also looked out for other inconsistencies. Some interesting numbers emerged.
20% of respondents in areas without local elections claim to have voted in them in 2015
3-6% of respondents in the campaign wave of BES polling claim to have voted by post before the postal ballots were actually issued
46% of respondents who could not be verified as registered to vote in June 2014 claim to have voted in the 2014 European Election
In all of these cases, the fibbers leant significantly more Labour than other respondents.
The report goes on:
“The evidence in the BES suggests that the reason for the increased impact of differential turnout is not due to a change in the relative enthusiasm between Labour and Conservative supporters since 2010. 84% of Labour supporters in 2015 said that it was “very likely†that they would vote, compared to 86% of Conservative supporters, while in 2010 the figures were 87% and 90% respectively. Rather the data suggest that the increase in the turnout gap between Labour and the Conservatives can be explained by shifts in party support amongst those who are actually less likely to turnout to vote, even if they say they will. This evidence strongly suggests that differential turnout was a major factor in the polling miss.”
Overall they say “this is relatively good news for pollsters. It should be possible for pollsters to fix many of their by using turnout weighting that accounts for the wider set of factors we have identified.”