Polling a changing and complex electorate is a huge challenge

Mike Smithson doesn’t pay me to play safe. Or to put it another way, I write for the prestige of appearing on PB but Mike is more than happy for me to stick my neck out.

And my neck has been right out there in the last few postings: suggesting that the race was “too close to call I said there was no solid evidence for making Jeremy Corbyn favourite. I rejected the notion that Corbyn rallies and CLP nominations were a reliable proxy for the opinion of the 640,000 party members and supporters.  And I argued there were many switchers from Corbyn because disappointment with his performance as leader.

One of my key arguments was that we had had no recent polling. Now we do. A YouGov poll for the Times pointing to a Corbyn landslide even bigger than last time. Adam a polite and comradely Tweeter asked me if, in the light of the poll I stuck by my assertion that Owen Smith could win. I said Yes – and here are my reasons.

The basis of my argument that the race was tight was the recruitment of 120,000 new registered and affiliated supporters by the anti-Corbyn Saving Labour campaign. Now the campaign’s number cruncher Reg Race suggests there might be issues in the way this immensely difficult poll was carried out.

There are what one might call three “killer facts”:

• YouGov had no access the Labour party’s lists of members and supporters to help them construct an accurate sample. This is in contrast to public elections where pollsters use electoral registers and demographic date to try to achieve a representative sample “unlike in public elections pollsters and can therefore not construct an accurate sample.”

• Just 50 of the respondents in the Times poll had joined the YouGov panel since the referendum on June 23rd. As Race points out, “This is precisely the period in which Saving Labour was active,” They recruited 70,000 new registered supporters and 50,000 TU affiliated supporters in a brief period in July and early August. It seems unlikely that many will have been captured by YouGov raising the possibility that their panel of registered supporters was skewed.

• The new anti-Corbyn recruits had been digitally tracked from Saving Labour to the Labour party website. At the party end officials could see where new recruits had come from.

Race a veteran of 30 years of polling experience is sympathetic towards YouGov. “YouGov is highly professional and has a good track record in polling the general public and in previous Labour elections.” But they had a uniquely difficult task of polling the complexities of the Labour selectorate.

For their part, YouGov say their result is robust but “obviously, polling of this nature is incredibly complex and any poll is just a snapshot of opinion at that time. Each poll comes with a margin of error and there is far more room for error with research of this type.”

During discussions between Race and YouGov it emerged that the pollsters revised their methodology following the publication of the Saving Labour projection in late August. Race suggests that action they took might have made their problem worse not better.

What YouGov did was to identify people in their existing panel who said that they had been registered supporters last year and who had been Party members in the group struck off by the Appeal Court decision and added them to the registered supporter group. But these two groups are known to be heavily pro-Corbyn. The effect says Race, was that YouGov might have “compounded their sampling problem rather than solving it.”

He reaffirms his view that “The election is too close to call.” For what it’s worth I agree with him.

Don Brind