Friday, June 27, 2014

Why are YouGov so obsessively secretive about the most important detail of their referendum polling?

I'm indebted once again to Calum Findlay for pointing out something that I hadn't noticed - YouGov are repeatedly showing a higher No vote among people who voted SNP in 2011 than any other pollster.  As an illustration, here are the figures from the most recent poll released by each firm...

No vote among 2011 SNP voters -

ICM : 15%

TNS-BMRB : 13%

Survation : 17%

Panelbase : 14%

YouGov : 24%

The pattern is the same even after Don't Knows are stripped out...

ICM : 16%

TNS-BMRB : 17%

Survation : 20%

Panelbase : 16%

YouGov : 26%

Figures aren't available for Ipsos-Mori, because they don't weight by 2011 vote.  But it's significant that TNS-BMRB are well in line with the other pollsters on this point, because like Ipsos-Mori and YouGov, they're one of the most No-friendly firms at the moment.  So whatever the reason for YouGov producing such a high No vote among SNP voters from 2011, it can't be a factor that is generic to all No-friendly pollsters.  Indeed, it may be a sign that YouGov are No-friendly for a completely different reason to Ipsos-Mori and TNS - which has always seemed intuitively likely, given that YouGov are so far out of step with all of the other firms that use online fieldwork.

As Calum suggests, the most likely explanation is the unique and rather eccentric weighting procedure that YouGov employ.  Like all other BPC pollsters apart from Ipsos-Mori, they weight by recalled vote from 2011, but they make one crucial exception - they separate out the people (or at least some of the people) who voted Labour in 2010 and then switched to the SNP in 2011.  Although SNP voters as a whole are often upweighted from the raw data, the upweighting usually occurs almost exclusively among the group that voted Labour in 2010.  In the most recent poll, that group was upweighted from 55 real respondents to 105 'virtual' respondents, meaning that the referendum voting intention of every person within the group was effectively counted twice.  We can reasonably infer a few things from this -

1) This group are probably producing much more No-friendly figures than SNP voters at large.

2) YouGov are struggling to find enough people in this group, so if the tiny sample they do have are in any way unrepresentative of Labour-to-SNP switchers, that error will be dramatically magnified after the weighting, and the same problem will occur in each and every poll.  It's quite likely that exactly the same people are being interviewed over and over again.

3) If YouGov didn't split SNP voters into two separate groups, the Yes vote would probably be higher in the headline figures, and the No vote would probably be lower.

This is all guesswork, of course, because for some reason YouGov only ever reveal the voting intentions of SNP voters as a whole, after weighting has been applied - there's never a breakdown for the two distinct groups they use for weighting.  It's very hard to understand why they would keep that information a secret, unless it shows such an improbable disparity between the two groups that eyebrows would be raised about the wisdom of the methodology.

To be fair, it's not completely impossible that there may be method in YouGov's madness - they were more accurate than ICM and Survation in the European elections, after all.  But if, for example, the purpose of this weighting procedure is to correct for a Yes-friendly bias that is inherent in volunteer online polling panels, it's very hard to understand why a non-online pollster like TNS would be producing a No vote among 2011 SNP voters that is broadly in line with the other online pollsters.

In any case, what is so special about Labour-to-SNP switchers?  There were two other important groups of switchers in 2011, namely Lib Dem-to-SNP, and Lib Dem-to-Labour (the size of the latter group was masked by the direct Labour-to-SNP swing).  Surely if there's wisdom in separately weighting one of those groups, there must be wisdom in separately weighting the other two groups as well?  Why the inconsistency?

Given Peter Kellner's well-known agenda on the subject of independence, it's sometimes very hard to escape the conclusion that he starts by working out what sort of headline numbers would "feel right" to him, and then works backwards to devise a methodology that will generate those numbers.

Wednesday, June 25, 2014

The mass canvass results

A couple of people have asked me to say something about the results of the Radical Independence mass canvass on Sunday, which were as follows -

Should Scotland be an independent country?

Yes 40%
No 29.5%

And with Don't Knows excluded, it worked out as...

Yes 57.6%
No 42.4%

Believe it or not, I've only been canvassed once in my life, and that was way back during the 2003 Holyrood election campaign.  A man came to the door without bearing any party colours, and asked a series of studiously neutral questions that didn't give away his affiliation.  (The only possible clue was when he asked "and how would you vote if there wasn't a nationalist candidate in this constituency?" -  an SNP canvasser would probably have been less likely to use the word 'nationalist'.)  It was only after he'd got the treasure trove of information that he'd been looking for that he announced he was from the Labour party, at which point I was of course mildly annoyed with myself for having played ball.  I may have saved them a bit of cash that would otherwise have been spent on a lost cause address!

It's not hard to see how that kind of 'blindfolded' canvassing operation can produce results that are at least as useful as any published opinion poll.  But I don't get the impression that the Radical Independence canvass worked in that way - voting intentions seem to have been recorded after the canvassers identified themselves as Yes supporters and made their pitch.  Of course it's great news if people were converted to Yes on the spot, but it does mean that the voting intention figures can't be considered representative of other people who weren't canvassed.  The only way to wholly square that circle would be to canvass every single adult in Scotland - and the numbers from Sunday illustrate what a gargantuan exercise that would be.  An impressive 8317 people were reached - but that's still only roughly 0.2% of the voting age population.

It's also likely that some people won't have told the truth because they didn't want to cause offence or disappointment to the Yes people on their doorsteps - that would be most likely to manifest itself as No voters pretending to be Don't Knows, but there may also have been some people presenting themselves as more certain for Yes than they really are.  It's also the case that Radical Independence were mainly canvassing in lower-income areas where you'd expect the Yes vote to be higher than the national average anyway.

So when you take all those factors together, probably the best way of putting it is that it would have been quite worrying if the results of the mass canvass hadn't shown a Yes lead - but thankfully they did.

Wisdom on Wednesday : What has the anti-independence campaign got in common with the Tories (apart from the obvious)?

"The Tories, every election, must have a bogey man. If you haven't got a programme, a bogey man will do."

Said by Aneurin Bevan, Labour's Minister of Health between 1945 and 1951, and the father of the NHS (albeit an NHS that looked a hell of a lot more like the SNP government's version of it than the monstrosity that is currently developing in England). Incidentally, Bevan also said this -

"No amount of cajolery, and no attempts at ethical or social seduction, can eradicate from my heart a deep burning hatred for the Tory Party. So far as I am concerned they are lower than vermin."

Hmmm. I wonder what he would make of the semi-blissful marriage of mutual self-interest between Labour and Tory in the anti-independence campaign?

Tuesday, June 24, 2014

Has anyone been "otherwise dickish" today?

You may well have seen by now that there's an expletive-ridden post on Better Nation attacking Stuart Campbell for his supposed "misogyny", and also berating Robin McAlpine for speaking up in Campbell's defence. It's written by the blogger Glasgow Sex Worker, who I have absolutely no problem with - I've read her blog over the last couple of years and found myself agreeing with a reasonable amount of what she writes, and she even left a kind comment here once. But it does have to be said that it's extremely unlikely that James Mackenzie would even have allowed Stuart Campbell's name to be mentioned on Better Nation unless it had been in the context of a full-on character assassination. Subtlety on the Wings question isn't an option in Mackenzie-world - he presents his readers and followers with something akin to the moronic George Bush choice of "you're either with us, or you're with the terrorists". Anyone who doesn't think Stuart Campbell is an unspeakable monster unfit for human interaction is likely to quickly find themselves "unpersonned" by Mackenzie (although I'm not sure how much that matters anymore - he did it to me last year, and only a small handful of others unfollowed me on Twitter in a subsequent show of 'solidarity'). His obsession with RevStu has reached the point where he just assumes that anyone disagreeing with him from a certain angle must be motivated primarily by "Wings love" - in my case it just never even occurred to him that (self-centred though it sounds) I was far more troubled by the deeply unpleasant and untrue things he had said about me than I was about his anti-Wings rants.

Recently, Better Nation has dropped all pretence at being interested in debate, and has barred comments on most posts. Curiously, though, today's post is allowing comments - albeit probably only on a highly selective basis. So far there have been two which defend Stuart Campbell, but it must be severely open to question whether Mackenzie would have let those through if they hadn't been written by a woman (they doubtless will still have infuriated him, but his self-image as a feminist would be inconsistent with silencing someone who is verifiably female). Although he's claimed in the past that individual authors have control over moderation on their own posts, it emerged that Jenny Kemp hadn't been responsible for deleting dissenting comments on her post about domestic violence. We'll see whether any dissenting male voices ever make it through today - my guess is that, at most, Robin McAlpine might be permitted a right to reply (followed by a one-to-one 'tutorial' from Mackenzie, naturally).

Incidentally, you might remember that Mackenzie was strangely indignant last year that I referred to him as the "Better Nation supremo", presumably because it didn't buy into the fiction that BN is still a group blog with no single person in ultimate editorial control. Intriguingly, he's recently started to tacitly acknowledge that the distinction between the blog and himself is largely bogus - the Survation polls he has co-commissioned are now referred to interchangeably as "our polls" and "my polls". And in the STV report on the most recent poll, he was billed as the "Better Nation founder", presumably because that's how he described himself. Nowhere in the report was it mentioned that he's a former media chief for the Greens, or even that he's a Yes supporter - and that shows the potential benefit for the Greens in using Mackenzie as a proxy when commissioning these polls, which is what I strongly suspect may be going on.

Hey now, hey now, what can I say? Nothing's fair in love and war.

Just a quick post tonight because I'm very, very sleepy. One of the unexpected side-effects of the fundraiser last month is that when people ask me what I do for a living, the most logical answer is "I'm a crowdfunded blogger", because that's - albeit only temporarily - my biggest single source of income at the moment. I'm then often asked a follow-up question about the nature of the blog, and that's led on to discussions about politics that in the past I haven't really had. (I've overheard lots of conversations about the referendum over the last eighteen months or so, but very rarely participated in them.)

Yesterday I had a chat with a committed No voter, and it was interesting for a couple of reasons. Firstly (and very unfortunately) it seems that Yes campaigners had chapped on her door and taken entirely the wrong approach - they got her back up by very aggressively challenging her views, and asking her "so you want David Cameron to do X/Y/Z to Scotland?" She said they just weren't interested in listening. It was very dispiriting to hear that, given the fine reputation that Yes canvassers have built up, but I suppose with so many people involved in the ground operation it's inevitable that some will be more sure-footed than others.

The other thing she said was that she was worried about an independent Scotland not being a member of the UN. I did a sort of double-take at that point, because I assumed she meant the EU, but she really did mean the UN. I pointed out to her that practically every independent state in the entire world is a member of the UN, but her response was that there are no guarantees - we'd have to apply, and no-one could possibly know for sure what the outcome would be. I suggested that we could be pretty sure, given the UN's track record of hardly ever rejecting anyone. Although she conceded that the risk was very small, she insisted that uncertainty was uncertainty, and that this was entirely typical of what people were being asked to vote for by the Yes campaign. "Unless we can know..." she said.

What can we do in the face of a belief that even the tiniest and most implausible of risks should be sufficient to automatically preclude any positive decision to make a change? It leaves very little room for discussion in the context of this campaign unless the belief itself can be challenged. I suppose the one possible antidote is to remind people that a modest amount of uncertainty is an unavoidable part of life - as individuals we deal with small risks every day without ending up cowering in a corner, and voting No carries as much uncertainty (probably more) as voting Yes does. But I didn't say any of that, because to someone already exposed to an unwanted hard-sell on her own doorstep, I think it could easily have sounded like more of the same. So I mainly just listened.

Maybe this is an incorrect interpretation, but what I take away is that there are some situations where the most constructive thing we can do as Yes supporters is to simply reassure people that we do respect their own views.

* * *

While I'm vaguely on the subject of the fundraiser, I should take this opportunity to apologise for the fact that the Backers page still isn't up. I've made a start on it, but it's a much, much more time-consuming exercise than I ever expected, because I have to cross-check 169 emails from Indiegogo to make sure that no-one who requested anonymity is wrongly included. I'll finish it as soon as I can - I know it's not the main reason that people donated, but it's important all the same.