David Wooding of the News of the World is reporting that the long-awaited Survation poll of Oldham East and Saddleworth has voting intention figures of CON 6%, LAB 31%, LDEM 30% (these figures imply a huge share for “others”, but my understanding is that this is because it was not repercentaged to exclude don’t knows.

If don’t knows were around about 25%, then these figures are the equivalent of something like Labour 41%, Lib Dems 40%. Obviously these figures have a much harder squeeze on Conservative support and, consequently, have the Lib Dems much closer to Labour. Until we see all the detailed tables of the polls tomorrow or Monday, it’s hard to speculate on what might be reponsible for the contrasts (Survation are putting up their tables at midnight, Populus on Monday, ICM presumably Monday).

Meanwhile, YouGov’s weekly poll for the Sunday Times has topline voting intention figures of CON 38%, LAB 41%, LDEM 10% (suggesting that 7% for the Lib Dems in the week was indeed just normal variation within the margin of error).

UPDATE: There are more details of the Survation poll on their site here. There are not full tables, but enough to judge that ICM and Populus are more likely to give us an accurate impression. Survation seem to have suffered from an exceptionally high refusal rate of 47% (perhaps it was their questionnaire, or introduction to it – ICM typically get about a fifth of that) meaning the final voting intention figures were based on just 225 people, giving a very high margin of error.

They haven’t provided their weighting targets, but the weighting regime itself was very simple – age, gender, past vote. Compare that to the more complex regimes used by established phone pollsters, which include social class, tenure, work status and so on. I’d stick with the experienced hands on this one (though as Robin Hood has mentioned in the comments, we won’t really be able to compare these polls against the actual result because the fact these three polls have been published could itself change people’s votes).

There are also some more details of the Populus poll here

65 Responses to “Newcomers Survation show tighter race in OE&S”

1 2
  1. @ Anthony

    C’mon my dear, please give us your professional opinion of the Survation poll. :-)

  2. I don’t even have the heart to put my steel caps in. All you can do is give them kudos for publishing the data in full and on time; suggest they don’t give up the day jobs; and then tiptoe gently away (difficult in steel caps).

    After weighting and removing uncertains, refusers, low likelihood to votes etc their percentage are based on a sample size of 169 (is there a smiley for “ah, bless). Actually their commentary isn’t without merit, but of course it’s based on such a small sample it’s not much of an improvement above anecdote.

  3. @ Roger Mexico

    I particularly liked the question asking people how their family members would vote.

    There was a whiff of desperation about it, in the circumstances. ;-)

  4. @ Amber Star

    I’m no expert, I noticed there weren’t really any crosstabs on the poll. Except it noted that even as voters were planning to vote Labour, they trusted (by a 37%-22% margin) Cameron, Osbourne, and Clegg over Miliband and Johnson on the economy.

  5. The full analysis from Lord Ashcroft is up on his site here:

    ht tp://www.lordashcroft.com/pdf/08012011_oldham_east_and_saddleworth_commentaryandpollsummary.pdf

    It’s all percentages though in the tables though, so it’s difficult to see what level of refusals they got and so on.


    The economic trust question isn’t as odd as it looks. Labour voters prefer Miliband etc by 58% to 12% – it’s Con + Lib Dem + Others + Undecideds that throw it the other way.

  6. @ Anthony

    :-) Thank you for the update giving your take on the Survation poll. I was afraid you’d gone & we’d have to wait until tomorrow for your opinion. :-)

  7. Anthony

    One extra thing to note on the Survation poll is that it has fieldwork dates of December 28th – January 7th (excluding public holidays). Given that public opinion in by-elections is notoriously mobile, this makes it even less helpful.

    The high refusal rate puzzled me too. Part of it must be due to inexperienced canvassers, but I wonder if they tried to verify voters’ addresses to check they lived in the right constituency. This would probably put a lot of people off. Presumably most pollsters just ask if people are in the constituency. If you’re so apolitical you haven’t noticed the usual by-election circus, you’re certainly not going to vote.

    Something to ponder over breakfast: do you know if they are weighting responses according to the part people live in? There seems to be a split between Oldham (traditionally Labour) and the rest (usually Lib Dem).

  8. @ Roger Mexico

    “One extra thing to note on the Survation poll is that it has fieldwork dates of December 28th – January 7th (excluding public holidays). Given that public opinion in by-elections is notoriously mobile, this makes it even less helpful.”

    A poll taken over that long a period does not strike me as being the most accurate.

  9. Rightly or wrongly I follow YouGov and I don’t pay overmuch attention to the other polls other than as a rough guide to trends. I suppose that makes me a YouGov junkie ? :)

    I am still feeding YouGov stats into a spreadsheet which calculates a rolling average of the last five polls. This has the effect of smoothing out the peaks and troughs and I don’t mind being shot at if there are objections to my method. Well it works for me, and my hamster has never complained so it must be ok !

    So – the latest averages are:-
    Con 39.4
    Lab 41.8
    LD 8.4

  10. Cozmo,

    Your method is fine if you accept that your average is correct two polls ago not on the last poll. I usually average over 10.

  11. Barry Crombie – “It is not easy to see a manifesto that will work to please activists and Tories unless they reinvent themselves as a centre-right party rather than centre-left.”

    It will be interesting to see how the eventual byelection result will be spun by coalition parties.

    There is clearly an element in the Conservative party who see the LDs as a barrier to them achieving an overal majority and are determined to exert a stranglehold over them through into the next election.

    For the time being Lib Dem stance as both the opposition to Tory in their heartlands, and as the main opposition in Labour dominated areas cannot be sustained.

  12. @Colin Green
    I would not claim that my rolling average are ‘correct’ on any given day out of the last 5. It is simply the nearest I can get to a ‘weekly’ average without doing too much key tapping.

    Over the past couple of months both Con & Lab figures have fluctuated and . However, the LD trend has been relentlessly downwards and I wonder where it will bottom out ?

  13. Survation chief executive Damian Lyons Lowe is qouted in Daily Mail as saying ‘There is little between Labour and the Lib Dems and the fate of the Lib Dems rests in the hands of their Coalition partners. ‘Conservative supporters could yet win this seat for the Lib Dems if they choose to make this a case of the Coalition versus Labour”. Hardly a neutral political stance! Makes you wonder about the the motivations behind this poll!

  14. Phil says:- “And combining the three polls, weighted by size of sample in the proportions 15/23 5/23 3/23 as above, very neatly gives:
    Lab 45%, LD 30%, Con 15%.”

    Phil – thanks for that.

    I think that probably is where things stand as we enter the final week of the campaign.

    Sorry to hark back to the 1980s again, but it does remind me very much of the party configuration during the Chrsterfield by-election of 1984. The Alliance should have been able to completely collapse the Tory vote and seriously challenge Tony Benn for control of the seat – but there were doubts over Max Payne (the Liberal candidate) and so the Alliance’s momentum was very slow and they were never quite able to close the deal. I seem to remember it ending something like 10:45:35.

    Clearly in OE & S there are no doubts over the Lib Dem candidate but there is, however, another serious obstacle to their ability to dent the Labour vote – namely their coalition with the Tories.

    This factor will, in the end, be critical to the failure of the Lib Dems to take the seat.

    Incidentally, the strength of the parties on the ground in OE & S may now be 15:45:30, but on the basis of the uniform national swing (as implied by current nationwide voting intention polls) it should be around 30:45:15.

    That surely is some testimony to the continuing strength of the Lib Dem by-election machine.

  15. Robin Hood – I think that LIb Dems the LD vote may be more resilient in their strong areas than in an average seat. As they have a lot of councillors in OE&S, and were only just beaten in May, that’s one of their fairly strong areas, and it has been for some time

    I expect a lot of the lost support is in the ABTs who were tempted to give the Lib Dems a try after a tired Labour government and a popular campaign. These are now disillusioned with the LDs, and that’s why most are going to Labour or to other ‘left’ parties like the Greens.

    So the places where they’ve recently seen a boost will be the ones I think they will see losses.

    This poll looks rogue. However, for Labour it’s hard to get people out (especially in by elections) so it may give them some help if there’s doubt that the result is in the bag. LD and T voters are more likely to turn out, so I expect that even if the other two polls are accurate, the Lab vote will be lower than shown.

1 2