A ComRes poll in the Independent on Sunday shows the main two parties static. The topline figures, with changes from the last ComRes poll, are CON 37%(nc), LAB 36%(nc), LDEM 14%(-3). The poll was conducted on the 10th and 11th December.

The previous poll from ComRes was the famous one showing a one point Tory lead (as does this one, obviously!). That this one shows the same suggests that one wasn’t a rogue. The poll isn’t, however, completely static – there is a 3 point drop in the level of Lib Dem support. The previous ComRes poll had showed them leaping by 5 points on the back of a sample that contained considerably more 2005 Lib Dem voters than their previous poll – I wouldn’t be surprised if this shift in the Lib Dem level of support has the same sort of reason – we’ll see when the tables emerge.

Meanwhile ComRes also tested four more statements. Unsurprisingly a majority (55%) disagree that taxes would be lower under the Conservatives – few polls ever show many people believing that any party would actually cut taxes. 52% agree that the fall in the value of the pound shows that Gordon Brown’s economic plans probably won’t work – which again, doesn’t tell us a huge amount – people are probably pretty pessimistic about the economy anyway. As with previous polls that have shown public concern over borrowing, 67% agreed that the government was planning to borrow too much. The most surprising of the four statements to me was “The Conservative response to the economic crisis seems to me like a ‘do nothing’ strategy” – normally polls reveal a pretty cynical attitude towards politicians from the public, and the Conservatives hadn’t seemed to have got any obvious alternative strategy across to the public, so I’d expected majority agreement here. In fact only 45% agreed, with 45% disagreeing – suggesting Labour’s line of attack on the Conservatives hasn’t chimed as much as it might have (not that this seems to be helping the Conservatives in topline voting intentions!)

I’m expecting at least one more poll to be released tonight, so we’ll see if they paint a consistent picture.

UPDATE: Tables are here. Looking at the recalled past vote in the survey, the reverse in the Lib Dem support does indeed appear to be more down to the make up of the sample than a great decline in support. In the last poll 12% of respondents said they voted Lib Dem in 2005. In this poll 9% did – so only three-quarters as many Lib Dems. It also makes it look rather less like unambigously good news for Labour. In the last poll 20% of the sample claimed they voted Tory in 2005, 24% Labour. In this poll only 18% said they voted Tory last time, 26% said they voted Labour. No change in the poll is not good news for Labour if the political make of the sample was more Labour to start with.

I should add that I am very confused by ComRes’s past vote weighting and why it varies so much. I did think that perhaps the first table in their pdf results was showing the unweighted recalled past vote, which ComRes then use to generate their target weightings. Having asked Andrew Hawkins about it though he tells me those are the weighted figures, which seems very strange to me.

15 Responses to “A steady picture from ComRes”

  1. interesting that your poll average now has tories below 40%

  2. I am always sceptical of ComRes polls. There often seem to be very wild changes in vote share.

  3. Anthony,

    On the variations in the weightings, if you were under pressure from the recession and phoning a couple of hundred extra people to get consistent weighting would cost £X, then maybe you might save money by saying;

    “hell that’s close enough and will be statistically okay”.


  4. Here’s another poll out tonight:

    YouGov survey for the Sunday Times remaining virtually unchanged from a similar poll last month.

    The poll showed a slight increase from 5% to 6% in the Conservative lead over Gordon Brown’s party, with Tories on 41% (unchanged), Labour on 35% (down one) and Liberal Democrats on 15% (up one).

    By a margin of 54% to 30%, those questioned said they supported the VAT cut – which lasts until January 2010 – and other measures announced in the PBR.

    But more than half (53%) said Labour was mainly interested in winning the next election, rather than saving the economy.

    :: YouGov interviewed 2,098 British adults online on December 11 and 12.

  5. Scotland figures are;

    Labour 37%, Tories 12%, LibDems 14%, SNP 34%, Others 3%.

    I’d have serious doubts about the fall in the Tories in Scotland to 12% and indeed behind the LibDems but Labour ahead of the SNP isn’t unexpected.

    Just as the Tories seem to have stalled nationally so they have fallen back slightly in Scotland but this fall is unlikely.



    on page 16 of 24, it reads total male 116, total female 116 for the tories. This totals 232. HOWEVER comres say the total is 231. you may say this is irrelevant. IT IS NOT. 232= 37.6% i.e 38%; 231= 37.4% i.e 37%

    Can some one please shread light on this as it is important i feel.

  7. Just to add that as well as Scots (or rather voters in Scotland) being more pro Labour on the four questions on two of the questions “Tory taxes” and ” Browns actions” SNP supporters seem more pro Brown than Labour Supporters…

    But we do have more worries about his borrowing, but that’s probably just because;

    “We’re afae mean”…….


  8. Ryans –

    The 116 are rounded numbers. They don’t actually represent 116 individual people, since after weighting each individual person will not necessarily count as 1 person, they will have weighted values of, for example, 0.95, 0.75, 1.23,1.05, etc, etc.

    If ComRes have 116+116 equalling 231 it means those 116’s must have actually been something like 115.7’s

  9. Peter – “On the variations in the weightings, if you were under pressure from the recession and phoning a couple of hundred extra people to get consistent weighting would cost £X, then maybe you might save money by saying; ”

    It doesn’t work like that – phone polls don’t use quota sampling. They phone up and interview around about 1000 people and take who they get (well, they phone the correct proportions in each region, but beyond that)

    The weighting is done – to grossly simplify things – by taking the actual number of men/women/past Labour voters/past Conservative voters etc you got, and then multipling each one by the correct value to bring the overall proportions to the figures you want.

    In other words, it costs exactly the same to type 12% into the weighting program on the computer as it does 9% :)

  10. AW

    In which case why doesn’t the past weighting remain the same and be exactly the same as the last election which would only have 2% more Labour supporters than Conservatives?

    Surely past voting patterns must be one of the most vital predictors of future voting patterns?

  11. Richard – because people do not accurately remember how they voted.

    We know this for certain, since if you take 1000 people and ask them how they voted the day after the election, and then ask the same people 6 months later, you’ll find their answers have changed.

    The main factors seem to be people forget they voted for minor parties and the Lib Dems. People who didn’t vote, claim they did since it’s the socially responsible thing to do (and those people tend to claim they voted Labour) and people who voted tactically for the Lib Dems but actually supported Labour say they voted Labour, rather than Lib Dem.

    In practice this means pollsters need to weight their sample to an estimate of what a representative sample would say they had voted, rather than how people actually voted. The debate is how quickly this changes – ICM and Populus believe it changes very, very slowly – so weight to pretty much the same figures month-in and month-out. MORI believe it has the potential to shift quickly, so reject weighting by past vote.

  12. AW

    Thanks, but surely such a large change in weighting in less than a fortnight is still inexplicable?

  13. Richard, the point of weighting is that your sample is never going to be a perfectly representative group of people. We know how people voted at the last general election, obviously. We know what proportion of different age and class groups to expect.

    But when you actually go and ring up 1,123 names out of the phone book, the fact you chose them at random means they won’t be exactly representative. If your selection process is genuinely random, eventually the differences will average out over many years – which doesn’t help you now. So you count the codgers, Liberals, etc, work out the %s, and compare that to known data, then you weight each group according to the ratio between its size in the sample and its size in the general population.

    This brings up some other problems; for small groups (like, say, RESPECT voters), you might only have five respondents even though they make up a significant percentage of the population. If you weight them up enough to match the population profile, though, you have the problem that one or two people’s views are being taken as representative of 5-10% of the nation. So you need to oversample the tiddlers, and then weight them down to their real size.

    Therefore, the weightings are never the same, because the size of the weighting needed to correct the sampling error is dependent on the size of the sampling error, which fluctuates randomly around the actual value.

  14. I think you are talking at cross purposes. As Alex says, the weightings applied to particular groups will change from poll.

    For example, 52% of the adult population are female. One month the raw sample, thanks to random sample error, might be 50% female, in which case a weighting of 1.04 would be applied to all the women in the sample. The next month the raw sample might be 55% female, in which case the weighting applied to all women would be 0.95. These change according to the make up of the raw sample.

    However, I don’t think that’s actually what Richard is asking about! I think he is asking about the variation in the past vote figures that ComRes are weighting to, not the weightings they are using to achieve it. Assuming we are reading the tables correctly, 2005 Lib Dems were weighted to a target of 9% of the sample this time, but to a target of 12% last time. I’m sure there is some explanation – but we don’t know what it is.

  15. Anthony,

    “I did think that perhaps the first table in their pdf results was showing the unweighted recalled past vote, which ComRes then use to generate their target weightings. ”

    That’s the only way the % would work, isn’t it? e.g. 20 18-24yr old supporting Labour is only 17% if it is of the weighted amount (117) not the unweighted (85).