An update on the Lord Ashcroft constituency polling from last week. Looking at the tables for Doncaster North – Ed Miliband’s seat – the weighting appears to be seriously askew, weighted as if the seat had a Conservative majority at the previous election. I’ve spoken to Ashcroft’s office and it appears to be a human error – typing the wrong weighting target in the wrong box (it’s something I’ve done myself in the past, though thankfully I’ve not done it and published the results!).

Lord Ashcroft’s team are reweighting and republishing the results, but obviously they are likely to make a significant difference to the findings, probably knocking down the reported level of Conservative and UKIP support and bumping up Labour. I’ll update when the new figures are out.

UPDATE: The corrected version of Lord Ashcroft’s polling is now up on his site here – topline figures in Doncaster North are now CON 13%, LAB 54%, LDEM 4%, UKIP 25%. UKIP are still in second place, but they are now a far larger twenty-nine points behind Labour (and given some of the reaction to the earlier poll it’s worth noting that Labour are even well ahead of the Tory and UKIP vote combined). There is no risk to Ed Miliband here.


123 Responses to “Update on the Lord Ashcroft Doncaster poll”

1 2 3
  1. Comes (phone)

    Con 28
    Lab 31
    Ukip 18
    LD 9
    Green 7

    No SNP numbers yet.

  2. Comres not “comes” (which is Spanish for “you eat”, which sounds like Ukip).

  3. Brown standing down as MP. Will his seat be less safe as a result? Will Labour be more or less popular in Scotland?

  4. Com Res tweet at ten pm.

    Three point Lab lead, very surprising, in some ways after the onslaughts on Ed M.

  5. Ukip seem pretty healthy there.

  6. And the Tories once again failing to keep their heads above the 30% line, although with Labour foundering at 31% they don’t have much to gloat about.

  7. Personally, I would be surprised if there wasn’t an incumbency bonus and a double one in some seats.
    Labour may need over 30 Con seats to become the largest party due to their losses to the SNP being more than cons to the UKIP with a them taking roughly equal of the LDs.
    UNS requires around a 5% swing needed and if PK is suggesting 1% lead needed he is only saying 1.5% or so extra swing required due to incumbency in marginals seats.

    My gut feelings that this is a little too high and gives insufficient consideration to the impact of the Ashcroft’s millions in marginals prior to 2010 which even his Lordship has said can’t be replicated.

    One impact of the potential gains for the SNP from Labour, though, is that it would erode some of the TFTP advantage for Labour.
    I still think a level vote share would give Labout most seats and even most with a small deficit but PK is right that up to a 3% Con lead would see them as the largest party – my guess is 2% also and 1% tight.
    A Labour recovery in Scotland though would take us back towards the 3% number as 10 more holds v SNP means 5 less gains required from the Tories, 16, 8 etc.

  8. Surprising to see the Greens at 7 with Comres after their recent online polls seemed to down weight them by 50%

    Will be interesting to see the tables to see if they have refined their methodology again (this time I would say for the better, as it brings them in line with the other pollsters)

    And to see if they prompted for UKIP or not, or maybe they are doing their online polls differently to their phone polls.

  9. Con -2 Lab +1 LD= UKIP -1 compared with last Comres phone poll five weeks ago.

  10. Wash: a situation in which losses and gains or advantages and disadvantages are equivalent; a situation in which there is no net change.

  11. Tables are here:

    http://comres.co.uk/polls/The_Independent_Political_Poll_2nd_December_2014.pdf

    Con 28% (-2)

    Lab 31% (+1)

    LD 9% (NC)

    UKIP 18% (-1)

    Green 7% (+3)

    Others 7% (NC)

    Changes from previous telephone poll (28 Oct)

    SNP 4% – 41% in Scotland but tiny sample as usual with telephone polls (74)

  12. Thanks Roger

    So they did prompt for UKIP.

    VI before the unknown allocation was
    Cons 28
    Lab 31
    LD 8
    UKIP 19
    Green 6
    SNP 4

    So very little change this time, no 4-5% downgrade of UKIP, no 50% downgrade of the Greens like we saw in their last online poll.

    I can’t figure out how they get from table 3 to table 5 to explain why.

  13. ROGER MEXICO

    I think comres are showing the SNP on 37%?

    Labour and Tories tied at 18% but as you say a tiny sample.

  14. YouGov/Sun poll tonight – Labour and Tories tied on 32%:

    CON 32%, LAB 32%, LD 8%, UKIP 15%, GRN 6%

  15. Allan

    No these are the headline figures (Table 6). ComRes do this rather odd squeeze rigmarole and then take likelihood to vote into consideration, yours is the figure before that (Table 3).

  16. @Allan

    (Comres)

    See table 6

    “Q3/4/5. Voting Intention (including Squeeze & DK/Refuse at Q.4) – Weighted to Turnout values (Q2) Base: All respondents expressing a voting intention and saying 5-10 likely to vote”

    SNP on 41%

  17. Knew Fred

  18. ROGER AND STATGEEK

    Ok thanks for that. Had another look and yup I wiz wrong.

    New thread.

  19. @Roger

    Snap!

  20. Richard

    Here’s how to follow the ComRes calculations. I’ll use the Conservatives as the example.

    The initial question, Q3 goes to people who said they were 5/10 likely to vote or more (the rest are ignored) and found 189 people who said Conservative, but 107 who refused to answer or said don’t know (see page 10)

    The 107 people were then asked how they would vote if they had to do so by law. Figures for this question are on page 13. In this question 40 people said Conservative, but this includes people who said they were 1-4 likely to vote and who are excluded. The number of people who count was only 17 (page 10, the people in the Conservative crossbreak who said Refused and Don’t know). Add these together and we have 207 Conservatives (it doesn’t add up exactly because of rounding).

    Next the figures are weighted for turnout. For this, look at page 6. The 134 Tories who said they were 10/10 certain to vote get a weight of 1.0; the 12 who said they were 9/10 certain to vote get a weight of 0.9 and so on down the line. Add these up and we get a total of 186, which almost matches the final total of 187 on page 21 (again, the different will be rounding)

  21. @AW

    Any correlation with ‘forcing’ people to choose and their choice being more inclined towards the opposition if forced?

  22. It is not surprising, in practical terms, that there has traditionally been an incumbency bonus. An MP has resources to work the seat, many candidates do not. Also, the MP has a longer run in. However, as pointed out above, the tactic of a key seat strategy, where parties pour resources and organisation into the seat, may mean the incumbency bonus is lessened. Also, if as suggested above it has primarily been seen in LibDem seats – well, we know the LibDems surpass the other parties in terms of community organising.

  23. Thanks Anthony

    I’m trying to understand the UKIP prompting vs not prompting, which to me at least on their last polls has resulted in a 4/5% difference, which then disappears with this table 3 to table 6 calculation, but this time there is no 4/5% UKIP increase via prompting, and no 4/5% difference in the table3-6, so is that a phone effect or something else.

    Let me play around and see if I can figure it out where the 4/5 came on the online poll and see why we don’t see it here.

1 2 3