Saturday, May 2, 2015

SNP lead remains rock-solid at 23% in marvellous YouGov poll

Tonight sees the publication of the latest full-scale Scottish YouGov poll, and there are some suggestions that it might be the last Scottish poll from any firm before polling day.  I must say I find that hard to believe - surely given the unprecedented situation, at least one newspaper is going to fancy the idea of a poll on Wednesday, or even on Thursday morning?  Oh well, if this is the last one, the SNP can hardly have any complaints about the state of play it depicts, which is essentially unchanged since the 'second surge' that occurred a few weeks ago.

Scottish voting intentions for the May 2015 UK general election (YouGov) :

SNP 49% (n/c)
Labour 26% (+1)
Conservatives 15% (-2)
Liberal Democrats 7% (+2)
UKIP 2% (-1)
Greens 1% (n/c)

I haven't been able to track down the fieldwork dates yet, although I do know that the poll got underway before the Question Time leaders' special on Thursday night, and the revelation that Miliband would be willing in some circumstances to help Cameron stay in power.  If it turns out that the bulk of the fieldwork was done before then, we may not be much further forward, because that's the only recent event which seemed to have a reasonable chance of shifting opinion in either direction.

If I was going to sound any note of mild concern about the result, it would be over the Liberal Democrat share, which is significantly higher than in any other YouGov poll since the referendum - the previous range has been 3-5%.  There are basically three possible explanations for the jump - a) it's an illusion caused by the margin of error, b) it's a real but small increase in Lib Dem support across Scotland, or c) it's a real increase in support driven mainly by tactical voting, and is therefore heavily concentrated in Lib Dem-held seats.  The first two possibilities wouldn't present a problem for the SNP, but the third certainly would.

UPDATE : There seems to be some confusion over the exact numbers - Anthony Wells at UK Polling Report (who's in a position to know, because he works for YouGov) says it's SNP 49%, Labour 25%, Conservatives 17%, Liberal Democrats 5%.  Those figures are absolutely identical to the last YouGov poll, so I'm wondering if Anthony has got the two polls mixed up.

UPDATE II : And now we have a third slightly different set of figures.  The Sunday Times, who commissioned the poll, say it's SNP 49% (n/c), Labour 26% (+1), Conservatives 15% (-2), Liberal Democrats 7% (+2), UKIP 2% (-1), Greens 1% (n/c).  Presumably these are most likely to be the accurate numbers, so I've updated the start of the post (the Conservatives were previously on 17%).

23 comments:

  1. So Kenny F is right then. The lead narrows (pmsl).

    ReplyDelete
  2. Was this done before Millibland went Dirty Harry on us and the Welsh?

    ReplyDelete
  3. FWIW, the "worm" technique used to measure sentiment while politicians are talking was VERY negative amongst Scottish voters while Miliband was ruling out any SNP co-operation.

    See "Ed's appeal to Scottish voters" (below the 17:04 entry)

    http://www.bbc.co.uk/news/live/election-2015-32543196

    ReplyDelete
  4. If the Lib Dem jump was caused by a tactical vote then I would expect Labour and Tory votes to have dropped. UKIP vote is down there and they wont vote Liberal under any circumstances I would have thought. Lab vote is up +1 and Lib Dem vote up +2 yet UKIP down 1 so it looks round some spurious rounding up or down is going on there.

    ReplyDelete
  5. UK Polling Report has Labour on 25%

    ReplyDelete
    Replies
    1. YouGov and the Sunday Times also have a new (separate!) Scottish poll, that has topline figures of CON 17%, LAB 25%, LDEM 5%, SNP 49%.

      Delete
    2. He also has the Lib Dems on 5%, so my entire blogpost could be fairly useless if that turns out to be true!

      Delete
    3. If those figures are right, they would be absolutely identical to the last YouGov poll, so maybe there's a mix-up.

      Delete
    4. The figures I heard for the Lab 25% poll were exactly the same as for the previous YouGov poll. Not just the Labour figures but the Tories and Lib Dems too. Maybe people could be getting their wires crossed with the last poll?

      Delete
  6. Original figures come to 102% so there must be something wrong with them

    ReplyDelete
    Replies
    1. No - that sort of thing often happens because of rounding quirks. The original figures look more likely to me.

      Delete
    2. Thanks James, didn't realise that.

      I would be very happy id they are correct.

      Delete
  7. Times site figures same as above except Tories on 15%

    http://www.thesundaytimes.co.uk/sto/news/article1551990.ece?CMP=OTH-gnws-standard-2015_05_02

    A YouGov survey of 1,162 voters in Scotland puts the Scottish Nationalists on 49%, Labour on 26%, the Conservatives on 15% and Lib Dems on 7%, with Ukip on 2% and the Greens on 1%.

    .



    ReplyDelete
  8. James, re the Lib Dems, I have noticed a similar increase in their rating in long-term averages (10 polls) of the YouGov GB poll sub-samples. Up until the end of March they were averaging about 5% (sometimes less), now they are averaging about 7%. This appears to be a tactical shift from Tories.

    Should caution that Populus haven't picked up a similar movement, and Ashcroft actually had Charles Kennedy's share *falling* from 35% in February to 33% in early April. It could just be a vagary of YouGov's panel having partisan type people on it who may be more likely to vote tactically.

    ReplyDelete
    Replies
    1. Should add to the above that the SNP average score in YouGov GB sub-samples has gone up from ~41% to ~43%. Con (18 to 16) and UKIP (5 to 3) have gone down. Labour's score has been more inconsistent (currently averaging 28%).

      Delete
  9. Weather could affect SNP turnout more heavily, tactical voting could make a disproportionate difference too. With these and a safe margin for error Labour may be in a better position than predicted, something he'd claim as a victory..

    ReplyDelete
    Replies
    1. Weather doesn't affect turnout significantly, not in a general election anyway. It's one of the great myths.

      Delete
    2. Curtice, speaking in 2012:

      "No. We've had one or two general elections when it's been raining in some parts of the country and not in another and there has been no significant variation in turnout. Nobody has ever really done the analysis for local elections. It's one of the most common theories of turnout but nobody has ever found the evidence to back it up. We tend to avoid elections in December and January because snow can make a difference. Just because there's a little bit of rain in the south east people think it's relevant."

      http://www.theguardian.com/politics/reality-check-with-polly-curtis/2012/may/03/weather-local-elections-2012

      Delete
    3. Even if rain affected turnout, why would it affect the SNP vote more than the others? Especially when the SNP have 10 times as many boots on the ground to carry out their Get Out The Vote campaign as the other parties put together...

      The claim that the SNP vote would be more affected by poor weather looks to me like an inaccurate claim being planted in the media to justify a sudden change in Labour's fortunes, for the better. Along with the "coincidence" that there are lots of comments appearing from "anonymous " type posters claiming that large scale postal fraud is impossible, (aye right) despite legal professionals saying that the UK system is open to fraud on an "industrial scale".

      Delete
    4. Met office forecast for here in Dundee on Thurs is cloudy but fine.

      You do know that a guide for Japanese visiting Scotland warned them not to be surprised to see people walk around bare headed while it's raining.

      In growing up in Auckland, NZ which is sub tropical and it really RAINS there I learned I'm not soluble.

      And finally for the indyref I would have crawled naked over broken glass and razor blades to vote Yes. I'm only very slightly less motivated this time but only because we already have an SNP MP here in Dundee East so I won't get the pleasure of voting out an Alexander or a Davidson or a Curran or even a Murphy.

      Delete
  10. Curtice blog about the poll:

    http://blog.whatscotlandthinks.org/2015/05/yougov-snp-surge-holds-steady/

    Indyref2 VI at 47Y, 53N.

    What tactical voting there is appears to be favouring the Lib Dems. About half of Lib Dem support appears to be on a tactical basis, rather than it being their first choice. The thing about that is that it will help them in this election, but if they lose the seats anyway, then what's the point in voting for them again?

    ReplyDelete
  11. http://www.scotsman.com/news/john-curtice-still-hope-for-labour-in-scotland-1-3761699

    John Curtice predicts the SNP will get 44 seats.

    He seems to have added all the subsamples together last week and said "alright then that makes 2000".

    ReplyDelete