Thursday, February 12, 2015

Housekeeping note about Facebook

Apologies to anyone who has tried to contact me on Facebook over the last few months.  I went on the site about an hour ago in an attempt to contact a friend who I've lost touch with, and realised that the last time I logged in was before the referendum!  I know it must look like I post there almost every day, but it's all rigged up automatically via Twitter and Networked Blogs.

By the way, I don't want to tempt fate, but it's just possible I may have finally solved the intensely irritating problem of preview posts on the Scot Goes Pop Facebook page always displaying the words "No Image, Invalid URL".  We'll soon find out - this post will serve as a test!

*  *  *

UPDATE (6.10am) : This morning's Scottish subsample figures from YouGov are SNP 51%, Labour 22%, Conservatives 13%, Liberal Democrats 10%, UKIP 2%, Greens 1%.  I haven't been keeping track of record highs and lows in individual subsamples, but there can't have been many GB-wide YouGov polls since the referendum to have shown better figures for the SNP than that.  And it's happened in spite of the familiar downweighting of respondents who say they identify most closely with the SNP (in this case 54 SNP and Plaid identifiers were downweighted to count as only 30).

27 comments:

  1. ...and here was me thinking that you were just a rude bugger! Logged in this morning to find that you had accepted my friend request so welcome back to Facebook!

    ReplyDelete
  2. Highest Yougov poll in the past five years, certainly:

    https://twitter.com/StatgeekUK/status/566194906017058816

    Statgeek.

    ReplyDelete
  3. Populus sub-sample: SNP 43, Lab 24, Con 14, LD 9, UKIP 6, Green 4.

    http://www.populus.co.uk/wp-content/uploads/OmOnline_Vote_13-02-2015_BPC.pdf

    Scots down-weighted from 137 to 123, SNP identifiers from 75 to 55. As with their recent polls, some SNP down-weighting, but not to the ridiculous extent seen previously.

    ReplyDelete
    Replies
    1. Since they updated their party identity weighting, the averages are:

      44% SNP
      26% Lab
      15% Con
      9% Lib
      3% UKIP
      2% Green

      Which is more in line with other pollsters. Lib still look a little high though.

      SNP are still being quite heavily downweighted as they are in e.g. Yougov, but less so now as you say.

      Delete
    2. I think that's because Populus have quite an aggressive spiral of silence adjustment. Given the large number of 2010 LD defectors, there are also a large number of 2010 LD "don't knows". This boosts them with Populus both in GB polls and in the Scottish sub-samples.

      Delete
  4. I think the question for Labour now is. Who won't lose their seat rather than who will lose their seat! I would have said Coatbridge would have been one where they had a chance. looking at the figures across the board. I can see any Labour seat a potential loss.

    ReplyDelete
    Replies
    1. Well I reckon Dougy Alexander will survive, given the local SNP party elected to put forward a 20 year old lass with a penchant for saying stuff she shouldn't up against him.

      Delete
    2. I would rather be represented by a real person than by Douglas Alexander, and I suspect plenty of others feel the same way, but we'll see. I think the importance of the candidate can be overestimated anyway. Who can forget Marcus Fox's bewilderment after his ungrateful constituents ousted him in favour of a 24 year old "boy"?

      Delete
    3. @Anon

      Latest poll has an SNP lead of 6% in Dougie's seat. With 2010 weighting and spiral of silence adjustments which help Labour / hurt the SNP.

      What problem do you have with young people and women BTW?

      Delete
    4. I suddenly have a problem with women? Ok.

      The poll that showed the SNP ahead was before the candidate was announced, so we shall see the effect it has. I'd agree with James in the most part that the candidate normally doesn't have a great effect, but I reckon it will in this case.

      Do I have something against young people? In so much as I doubt that a 20 year old who talks of their hatred for others openly, and also talks of wanting to 'nut' others, is possibly not mature enough for the role of an MP, maybe.

      Delete
    5. Does Ian Davidson get a free pass for making much worse comments, simply because he's a "mature" man of 64?

      Delete
    6. Paisley and Ren South is my constituency. My social circles probably aren't representative of everyone, but Alexander is well known here as a Labour bigwig and not well liked. I can't see much of a personal vote for him. That's if people even care about candidates.

      Mhairi is just a normal person and very sincere. The smears against her (the age and the sectarianism card) are unfair although predictable since she is standing against Alexander. The age issue would have been used against our other nominee as he is 23.

      Delete
    7. @Anon

      It was just that your post suggested you thought the candidate was weak next to Alexander because she was a young woman.

      You don't think a woman, notably a younger one, is any challenge for Doug Alexander?

      Delete
  5. Curious. Feed in the Scottish subsample to EC and it comes up with Lab 1 seat, LibDems 1 seat, SNP 57. Change the LibDems from 10% to 6% and it goes to Lab 2 and LibDems 1, the difference being Coatbridge, Chryston and Bellshill where Others get not only the LibDem vote, but some of the SNP vote!

    ReplyDelete
    Replies
    1. The individual seat figures EC gives for the LibDems are ludicrous though. He has them down below 1% in about 42 of the 59 seats, and way above that in the rest. There's little rhyme or reason to it.

      Delete
    2. @yesindyref2

      There is no way that SLAB are going to end up with only 1 MP in Scotland in May. I would love it to happen for their Toryism of the last 20 years or so, and nastiness in the independence referendum. I just cannot see it though.

      Delete
  6. Some of these polls are excellent - the fullscale Scottish ones, even at our lowest (as an SNP supporter) 10% leads are breathtaking.

    The subsamples are all over the place though. 2 days ago YG had SNP 40 Lab 32, and this record high SNP 51 Lab 22 - of course is great reading, but lots of work to be done to even just get double figure MPs back to Westminister.

    If any new, old or non-members, but supporters fancying helping out campaiging for the SNP then I'm sure all the local areas would love to hear from you. It doesn't even need to be chapping doors, and there's no commitment to a set amount of hours per day/week/month etc - basically just do what you can.

    Apologies for the blatant propoganda, but it's important that if you are able to, we help the SNP get as many votes as possible in May. It doesn't even have to be in an electioneering way either. Just nagging at your pal who may well be an SNP sympathiser to get out and vote on the 7th is just as good.

    Let's do this

    ReplyDelete
    Replies
    1. Seconded. Please get out and help. The leaflet runs are being stepped up. We need volunteers.

      Delete
  7. Opinium / Observer sub-sample: SNP 34, Lab 34, Con 15, Green 9, LD 6, UKIP 3.

    http://ourinsight.opinium.co.uk/sites/ourinsight.opinium.co.uk/files/vi_10_02_2015.pdf

    Some weird shit in the tables - apparently more SNP 2010 are now voting Labour (6) than 2010 Labour are voting SNP (3). Can't quite see how that reconciles to moving from the 2010 result (Lab 43, SNP 20) to the scores given in this poll.

    Perhaps shouldn't be taken too seriously as the Wales sub-sample has Lab 30, Con 30.

    ReplyDelete
  8. I'm not really surprised, because last week Opinium made the eccentric decision to move in the opposite direction to Populus, and introduce a weighting scheme that is very unfavourable to the SNP.

    Intriguing that they've suddenly started publishing their Scottish subsample figures immediately after making that change!

    ReplyDelete
  9. The ComRes one has the SNP at 59%;

    http://comres.co.uk/wp-content/uploads/2015/02/SM-IoS-Political-Poll-15th-February-2015_456794.pdf

    ReplyDelete
    Replies
    1. Thanks. Either it is a very SNP heavy sub-sample or there is false recall going on, because 46 recall voting SNP and only 36 for Labour in 2010.

      It feels like the sub-samples for the irregular pollsters (i.e. everyone other than YouGov and Populus) are becoming less reliable as time goes on.

      Delete
    2. I think the problem is that we pre 2010 we had such a huge tactical vote (anyone but Tory) developing, as evidenced by increasing contrast between Holyrood and Westminster results, followed by a major generational shift associated by the political awakening of the iref. Scotland therefore, in terms of past vote or party ID weighting, is a minefield.

      Delete
    3. I have wondered if some people who voted Yes in the referendum and plan to vote SNP use that as their yardstick? Not SNP in 2010 but are now.

      Delete
  10. 59 means 59... a clean sweep of all the seats... What a wonderful number.

    ReplyDelete
  11. James, I wonder if you have seen this article? I think it may be of interest...

    https://archive.today/twL4H

    ReplyDelete
  12. Regardless of her unhelpful comments, I think selecting a 20 year old candidate, male or female, just isn't too clever.

    There is always going to be votes lost on the 'experience' issue, rightly or wrongly, especially by older voters more likely to vote.

    This is a marginal seat that looks winnable on the current figures,, and would be a tremendous scalp to take.

    This just makes it that bit harder. When the stakes are so high, why take chances?

    ReplyDelete