Saturday, August 16, 2014

Reading the runes for tonight's polls

I've just arrived in Brussels, and I've barely had a chance to catch breath, but this is how I see tonight's polling situation.  If the embargoed Panelbase poll turns out to have the figures that were prematurely tweeted earlier, it could be - pound for pound - the best Panelbase poll ever for Yes.  That's because the firm's methodology has radically changed since the last time they produced a 48/52 split.  But it's certainly one of the two best Panelbase polls ever (excluding the famous one from a year ago which is generally set aside because of the question sequence).  That's problematical for the No trolls who are trying to play this down, because we're not supposed to be in "Yes have failed to make even further progress from their all-time high" territory, but instead in "Darling bounce helps No to decisive lead" territory.  So much for all that, eh?

It may seem bizarre that Blair McDougall, the No camp's embarrassment of a campaign chief, has apparently been going out of his way to remind people tonight that Panelbase actually showed a Yes lead in a poll last year.  The reason he's doing it is presumably to make the new one seem relatively unimpressive by comparison, but in reality it's highly unlikely that the two polls are directly comparable.  Neither the SNP nor Yes Scotland have insisted upon the unusual question sequence that generated that Yes lead in any subsequent poll.

Judging from Twitter Kremlinology, it's probable that the ICM poll shows some kind of reduction in the No lead, although it's difficult to know what to read into that because the last poll from the firm was on the good side for No.

Kenny Farquharson has once again been rather catty tonight, suggesting that the ICM poll for his publication is the only "independent" poll of this evening, as compared to a Yes Scotland poll that was passed on to a "pet" paper.  Does he have a point?  The short answer is "no", and the longer answer is "mostly no".  As long as there was no jiggery-pokery with the question sequence, it makes no difference whether the paying client is "independent" or not - the results of the headline question are equally credible.  The supplementary questions may be leading, but because they're asked later they can't affect whether respondents say 'Yes' or 'No' to independence.  The only problem with having a partisan client is that they may only publish polls that are particularly favourable, so over a period of time we would only get a partial picture.  But if the Yes campaign have indeed been withholding less favourable Panelbase results, that would just make tonight's numbers look even better.

And if there has been a Yes bounce, does that mean that once you cut away all the spin from the unionist media, Alex Salmond actually defeated Alistair Darling in the debate?  Not necessarily.  It seems to me that Salmond was in a no-lose situation, because the debate will have made people think - and we all know which way voters tend to swing in this campaign once they actively start seeking out information for themselves.

32 comments:

  1. Haste ye back James.

    You are coming back?

    ReplyDelete
  2. You do no the field work for this is from June! It seems yes Scotland have been sitting on it since then. Also some leading questions in it. But why sit on the result for two months? Was this done under the old system for panel base?

    ReplyDelete
  3. "some leading questions in it"

    Questions asked before the referendum voting intention question?

    ReplyDelete
  4. From people who participated in the poll, no leading Q's and fieldwork from last week (post debate).

    Indy Q was 1st question

    ReplyDelete
  5. Well Chris, I'll take your word for it, but "Anonymous" has a reputation for being scrupulously honest.

    ReplyDelete
  6. The Yes stall in Ayr today, one of the last Tory bastions in Auld Scotia, was absolutely buzzing today...

    You can sme it in the air...

    And as for the anonymous naysayers above...

    You can smell it in the air too ;)

    ReplyDelete
  7. I have no idea what the results of the Panelbase are and if there is any reason to doubt the results or not. If not then it is indeed a good poll for the Yes campaign.

    My problem is that is the scenario was flipped. A poll, say by YouGov, commissioned by Better Together reported a 'surge' in the No vote then I doubt you would be as accepting of the poll as you are here. In fact in recent weeks you'll report any gain in No as either noise, 'bizarre' sampling or otherwise flawed in some way. Any Yes-positive result is reported as great news for the Yes campaign.

    I know this is a pro-inderpendence blog but what value is there is just spinning No gains as flawed but Yes gained as surges?

    The thing is we already know what your anayslis of the polls will be. If ICM shows a gain for No then it's flawed, if it reports a gain for Yes then it's not flawed. We get it.

    You also label anyone who criticises you as a troll. Yet you have no problem going after pollsters personally as was the case with Peter Kellner and you label Blair McDougall as an 'embarrassment of a campaign chief'. Understandable but then you get upset if people have a go at you? Grow a thicker skin.

    ReplyDelete
  8. If ICM showed an increase for No then Blair McDougall and his BT staffers would have been crowing about it.

    ReplyDelete
  9. The field work for this poll was not carried out last week it was in June. The article clearly states this. There may be another panel base poll in the field it it was not this one.

    ReplyDelete
  10. If ICM showed an increase for No then Blair McDougall and his BT staffers would have been crowing about it.

    Of course they would. They're biased as hell. They're running the campaign for No. You'll expect those running the campaigns to dismiss polls that go against them and to crow about polls that go for them.

    But someone professing to analyse polls, even if they have a pro-Inderpendence stance, should at least try to be objective. It's not as if ScotGoesPop was always like this....

    ReplyDelete
  11. Be sensible people, lets see the data before we jump to conclusions of 'was it rigged'. That goes fir both camps,.

    That should be out within a couple of days.

    ReplyDelete
  12. Ryan : Look through the archives if you think anything has changed. It hasn't. And let me just gently say to you that if I really was in need of a thicker skin, the No trolls' posts wouldn't even be appearing here. If you think the level of open debate I allow can be just taken for granted, have a look at Better Nation some time - your patronising little lecture might be better directed at James Mackenzie.

    Anon : Could we please have a link to support this claim about June?

    ReplyDelete
  13. James, I think anon is making the same mistake as some over-enthusiastic Yes tweeters of looking at the last Panelbase headlining in the Herald in June. Unless Smithson is on a massive geriatric trolling exercise (not impossible), the results he broke the embargo with are very similar to the June Panelbase.

    ReplyDelete
  14. There you go. It's in the article.
    http://www.heraldscotland.com/news/home-news/yes-48-no-52-breakthrough-poll-has-yes-at-record-high.24488719

    ReplyDelete
  15. Ryan : Look through the archives if you think anything has changed. It hasn't. And let me just gently say to you that if I really was in need of a thicker skin, the No trolls' posts wouldn't even be appearing here. If you think the level of open debate I allow can be just taken for granted, have a look at Better Nation some time - your patronising little lecture might be better directed at James Mackenzie.

    OK I went thought the article and fair enough. It was like that. For some reason I thought otherwise.

    I don't like 'Better Nation' either. I don't like anything that uniformly takes a side and sticks to spinning for it irrespective of what the evidence is. It's like being shown the colour blue and arguing that it is in fact red.

    I know how I would want to vote, No, if I could (currently living in England) but I know that 'No' has weaknesses. When the polls drifted towards Yes during March/April I didn't dismiss them, I was in fact concerned, and felt Better Together needed to step up their game.

    And that is my criticism here. That No polls will be spun as bad, flawed or not relevant but Yes polls are reported on face value. Polls can be flawed but how likely is it that every poll showing No will be flawed and those showing Yes are not flawed?

    You're hinting here that Panelbase looks to be a perfectly legitimate poll, and that may well be, but if ICM shows a No lead or if it stays the same then will that be legitimate?

    ReplyDelete
  16. Anon : Thanks for setting my mind at rest. You've linked to an article from June, so clearly tonight's poll is new.

    ReplyDelete
  17. PS Looks like ICM is showing a boost for Yes.

    ReplyDelete
  18. James, let the trolls eat their own shit at midnight when the articles are out. The end of the union is coming. A new Scotland beckons.

    ReplyDelete
  19. Scotsman headline: Yes vote advances despite currency fears

    ReplyDelete
  20. James, let the trolls eat their own shit at midnight when the articles are out. The end of the union is coming. A new Scotland beckons.

    ReplyDelete
  21. YES to Truth
    YES to Freedom
    YES to Prosperity

    Our purpose is honourable and true, yet they seek to oppose it.

    Pity them.

    ReplyDelete
  22. Anon : Thanks for setting my mind at rest. You've linked to an article from June, so clearly tonight's poll is new.

    Classic. Nice work, Anon.

    ReplyDelete
  23. Personally I wouldn't trust any poll as they're all too reliant on "weighting". There's too much reliance on each company's weighting methodology hence to widely differing results between polling companies.
    However, I'm not sure whether to admire or scorn this sites analysis sometimes. I suppose fair play for a Yes site continually showing the Yes result of a poll in a good light but sometimes you go a bit overboard in trying to polish a turd. I keep expecting to see something like a 10% dip in Yes support in a poll accompanied by something like "Yes vote shows surge(1% statistical variation most likely) amongst snowboarders" or whatever.
    PS I'm a a definite No but appreciate you allow uncensored views from both sides.

    ReplyDelete
  24. James wrote

    " As long as there was no jiggery-pokery with the question sequence, it makes no difference whether the paying client is "independent" or not - the results of the headline question are equally credible. The supplementary questions may be leading, but because they're asked later they can't affect whether respondents say 'Yes' or 'No' to independence. The only problem with having a partisan client is that they may only publish polls that are particularly favourable, so over a period of time we would only get a partial picture."

    Frankly this is the kind of objective commentary we come to this site for and it is appreciated.

    I sense James's objectivity (I mean about polling, I do not expect objectivity on the key question) has been on occasions tested in recent weeks. I welcome its return.

    ReplyDelete
  25. "The only problem with having a partisan client is that they may only publish polls that are particularly favourable, so over a period of time we would only get a partial picture."

    That's something I've wondered about myself and it's a very interesting possibility. I read Bad Pharma by Ben Goldacre which addressed the way pharmaceutical companies hold as many clinical trials as possible, only to reveal the results which favoured their commercial interests while burying the rest. It made me wonder if the same could be done by those who commission polls. I've also noticed that Better Together (Unlike Yes Scotland) have never revealed the results of any polls they have commissioned, something that could be interpreted either way.

    ReplyDelete
  26. ICM Poll:

    Yes: 38 (+4)
    No: 55 (+2)
    DK: 14 (-7)

    Excluding DK:

    Yes: 45 (+2)
    No: 55 (-2)

    Apparently don't knows are leaning 2:1 towards No.

    ReplyDelete
  27. looking at those figures, the first No should be 47'ish? (cos 55 + 38 + 14 = 107)

    and 47 / (47 + 38) * 100 = 55'ish

    ReplyDelete
  28. Ah, my mistake. The first No figure should have been 47. And I should go to bed!

    ReplyDelete
  29. So the disaster for Salmond headlines and coordinated assassination by the BNP media failed.

    Support for Yes up support for exterminating Scotland down.

    What excuses will the Scotophobes come up with now?

    ReplyDelete
  30. "I've also noticed that Better Together (Unlike Yes Scotland) have never revealed the results of any polls they have commissioned, something that could be interpreted either way."

    BT have not had to commission any polls because they've been getting them for free from the taxpayer.

    ReplyDelete
  31. Anonymous:

    "BT have not had to commission any polls because they've been getting them for free from the taxpayer."

    I'm aware of the large scale Ipsos Mori polling, but I was thinking more along the lines of polling to rally the troops. Clearly the SNP released the results of this poll to improve the morale of the Yes campaign while undermining the morale of the No campaign. Yet Better Together have never done this. It's not as if it would even be difficult, all they'd have to do is commission a YouGov poll with a thumping No lead and hand the results over to a No-friendly newspaper. We all know that Better Together commission private polls, yet they always keep the results under wraps for some reason.

    ReplyDelete
  32. "We all know that Better Together commission private polls, yet they always keep the results under wraps for some reason."

    Might be because the polling is designed to locate the items which voters are most afraid of and guide their advertising accordingly?

    ReplyDelete