Tuesday, January 27, 2015

The media again claim that a deal has been "ruled out" when no such thing has happened

On this occasion, the mainstream media do at least have an alibi for their grossly misleading claim that Ed Balls has "categorically ruled out" a post-election deal with the SNP, because it's true that he did reply to a question (asking if Labour would consider such a deal) with the word "no". The snag is that he then want on to clarify what that reply meant, and it was not in fact the conventional meaning of the word "no", but instead the little-known alternative meaning "Labour would like to win a majority". It's rather like asking me if I would consider staying in my current place of abode, and getting the answer "No, I'm hoping to win the lottery."

* * *

SCOT GOES POP POLL OF POLLS

Absolutely nothing should be read into the drop in the SNP lead in this update of the Poll of Polls - it's just a quirk caused by the previous update having taken account of two full-scale Scottish polls, including the famous Ipsos-Mori poll.  Those two polls have now dropped out, leaving a sample entirely comprised of Scottish subsamples from GB-wide polls (four from YouGov, two from Populus, one from Ashcroft, one from Survation and one from ComRes).  That works against the SNP because YouGov and Populus typically downweight the party sharply in their GB-wide polls, thanks to the use of Westminster-centric party ID weightings.

Scottish voting intentions for the May 2015 UK general election :

SNP 41.3% (-5.8)
Labour 28.2% (+2.6)
Conservatives 16.9% (+2.6)
Liberal Democrats 4.9% (-0.6)
UKIP 4.6% (+2.1)
Greens 3.8% (+0.1)

(The Poll of Polls uses the Scottish subsamples from all GB-wide polls that have been conducted entirely within the last seven days and for which datasets have been provided, and also all full-scale Scottish polls that have been conducted at least partly within the last seven days. Full-scale polls are given ten times the weighting of subsamples.)

16 comments:

  1. Those poll of polls figures would still give the SNP 39 seats according to Electoral Calculus Scotland's prediction calculator - and the Conservatives 3 seats, which seems unlikely. (Though I can see how they might take Michael Moore's seat given the likely Lib Dem collapse).

    ReplyDelete
    Replies
    1. You have to remember that the Conservatives are starting off at an especially low level, so it wouldn't be completely shocking if they picked up a couple of extra seats, especially in the event of a Lib Dem collapse. The Tories had 11 seats here in 1992 and not all those voters have died off.

      Delete
  2. Thanks for all the work you do to collate and interpret all these stats, James.

    A general question about polling: when playing around with Electoral Calculus or whatever I have been working on the assumption that SNP and Slab combined get 70% of the vote and then adjust accordingly. Do you see this as useful or will we see the other parties getting squeezed as the campaign progresses?

    PS

    Good to see a couple of new Scottish contributors setting the record straight on that notorious, pale imitation of this blog Electoral Gambling that you mentioned a while ago.

    ReplyDelete
    Replies
    1. The Greens almost certainly won't poll as well as the above numbers suggest, and UKIP might be squeezed. But I think the Tory vote is solid, and the Lib Dems may recover slightly.

      Delete
  3. Just fyi, James _ I did a new YouGov tonight with predominantly Scottish questions.
    John Bell

    ReplyDelete
    Replies
    1. Thanks, John. There's always a chance it might be an internal poll that will never see the light of day, but hopefully not.

      Delete
  4. I noticed that in the BBC poll of polls they don't allow too many polls from any one company, eg You Gov, to stop their result being overly skewed by one company. Food for thought, James?

    ReplyDelete
    Replies
    1. During the referendum, I used exactly one poll per company, which I think was a good approach that nobody else followed. But I don't think that sort of thing is feasible for the current situation. There's very little Scotland-specific polling, so no format for a Poll of Polls is going to be wholly satisfactory. The idea is just to take all the (limited) available data from the last seven days to get a very rough idea of the state of play.

      Delete
  5. I see Balls, like Dave, is saying he's against Scots having a say in governance of the UK.

    Scottish people are the new 'immigrants / foreigners / bogeymen' it seems.

    Is this supposed to win votes?

    ReplyDelete
    Replies
    1. I think he's saying nationalists like the SNP, Plaid and the Kippers shouldn't have a say. Plenty of scots will still have a say by being in the majority that won't vote SNP.

      Delete
    2. According to the Ipsos-Mori poll, the majority of people in Scotland (52%) are planning to vote SNP.

      Delete
  6. "Is this supposed to win votes?"

    Yes, and it probably will - in England.

    Perhaps this is is a sign that Labour HQ are realising that they have definitely lost Scotland to the SNP, and are going to concentrate their efforts on wooing English voters, including the xenophobic ones.

    ReplyDelete
    Replies
    1. I think it also makes it more likely that Labour will get into bed with the Tories. IF they have decided their cause in Scotland is lost then why not alienate further already alienated people by enabling another Tory government? Clearly their future is even more about winning English votes, so we are going to be ignored by them even more from now on. Taking charge ourselves is the only solution.

      Delete
    2. I think some sort of Lab-Tory coalition likely if the SNP hold the balance of power.

      The SNP and Scottish people have been increasingly demonised by unionist parties; this being ramped up for the referendum. Scots are apparently anti-English xenophobes but at the same time want to live off English taxes. The SNP are this embodied (I know, I know.. we want indy but at the same time apparently want subsidies).

      So, Labour would take a big electoral risk in the SE right-wing swing seats working with the SNP post May 15.

      However, Labour stating an unwillingness to work with the SNP can only harm them further in Scotland as a Lab-SNP partnership of some form is by far the most popular option north of Gretna.

      Anyway, the actions of the unionists on this front can only serve to damage the union even further by alienating Scotland.

      In a healthy family of nations, the winning parties in each home nation would e.g. form a 'one nation one vote' type senate to decide e.g. federal budgets for defence etc. Instead, the views of Scottish people, as expressed in elections, are seen as something which should be suppressed at all costs. We either need to vote for English parties, or shut up.

      Delete
  7. Any chance of a poll of polls graph to see the long term trend? That would give a better feel of what is happening in the polls.

    ReplyDelete
  8. Just so long as it is not a line graph which is bogus unless only one company, using a consistent methodology is present. With multiple different companies in the poll and dropping in and out only a bar graph with one bar per week is valid.

    ReplyDelete