Communicate Research’s monthly poll has voting intention figures with changes from last month of CON 35%(-5), LAB 31%(+2), LDEM 20%(+3). The poll was conducted between March 23-25th. The figures appear to show a very sharp drop in Conservative support, but as noted before, the lack of political weighting in Communicate and Ipsos-MORI polls means the results are more volatile and shifts from one month to the next less meaningful. (UPDATE – someone’s pointed out that the Independent’s coverage of the polls that that it was indeed weighted by past vote. I’m trying to find out if Communicate have altered their methodology and will update once I get an answer. UPDATE II – confirmed, Communicate Research polls are now weighted by past vote – I’ll put a proper post up on it when I get the details of their new methodology. What this means is that their future pollls should be less volatile, although it doesn’t mean that we can read much more into the changes this month, as last month’s sample could have just been unusually stuffed full of Tories!).

Meanwhile Populus have released their first poll of Scottish voting intentions ahead of the Parliamentary elections in May. The voting intention figures in the constituency section are CON 14%, LAB 28%, LDEM 15%, SNP 38%, while in the more important regional top-up vote support stands at CON 14%, LAB 30%, SNP 35%, LDEM 14%. On Weber Shandwick’s swingometer this translates into the SNP becoming the largest party with 49 seats to Labour’s 43, with the Conservatives and Liberal Democrats both on 18.

UPDATE: MORI’s monthly figures have also been released…sort of. A note on their website says that “as we are still running methodological tests while merging the Ipsos and MORI field operations, we are not publishing a full Political Monitor in March”. However, they have released voting intentions from a separate face-to-face omnibus survey which uses the same sampling and weighting regime as their normal figures. The topline voting intentions, with changes from the last MORI poll in January, are CON 41%(+2), LAB 33%(-2), LDEM 17%(-2). It was carried out between the 9th and the 15th March.


56 Responses to “Communicate on Britain, Populus on Scotland”

1 2
  1. Anthony,

    Exactly why do you say “the more important regional vote”.

    As the number of FPTP seats is used as part of the calculation of regional seats, they are intrinsically linked.

    Interestingly although there is little change in the Tory and LibDem votes from the last scotsman poll, the WS swingometer now no longer shows the LibDems getting six more seats than the tories.

    Of course as this is the first Populus poll we can’t really tell if anything has changed since late january when we had the last raft of polls.

    In addition the WS swingometer, can’t really factor in the likes of Margo McDonald so the final effects of the minor parties are hard to determine, although ifthey don’t pick up 1% or 2% soon both the SSP and Greens could be facing a wipe out.

    Peter.

  2. Anthony

    Given the volatility of unweighted polls, can I ask what is the rationale, the props and cons of weighting versus not weighting.

    Thanks

  3. Peter, because if you lose lots of consitutencies but retain your list vote, you get back the seats you’ve lost in the top-up (of course, it doesn’t always work like that in practice, but you get the point).

    Simon, that deserves a post in its own right, but in short, political weighting would be a definite good, it means each sample is politially representative and has the same political make up of previouis samples, meaning changes are more meaningful.

    The reason that some companies don’t do it is because there is no concrete measure of political attitudes you can use to weight to, in theory you can weight how people say they voted last time to the actual result of the last election, but we know for sure that many people don’t accurately report how they voted at the last election, so weighting recalled vote to the real would actually produce an unprepresentative sample.

    Populus and ICM estimate target recalled shares of the vote based on past academic studies of how false recall works, but MORI (and presumably Communicate) think the problem is intractable and that past vote is therefore unsuitable to use for weighting purposes.

  4. You would have thought that these poll figures from Populus would translate into a bigger advantage for the SNP than seems to be the case-it won’t take much of a slip in their lead to bring them back to level pegging with Labour. I think the Scottish election is still wide open not least because both the Lib Dem and Tory ratings in this poll look suspiciously low to me. But most of all as I have said before when it comes to the FPTP seats many voters wanting to punish Labour may vote tactically for the main challenger just to get them out. So I do wonder how much weight should be attached to WB’s swingometer.
    As for the Communicate poll well at least it should cheer up Mark Senior who seems to have been rather quiet recently. For all that this poll seems out of line with the weighted polls it may indicate a degree of volatilty in the electorate’s thinking that should put a break on any undue confidence on part of the Conservatives. The opera ain’t over until the fat lady sings. Oh what fun this all is!

  5. Anthony: If Communicate are to become regular pollsters, could you please add a tracker for them on your top right, like with the other ones?

    The poll doesn’t seem excessively out of line. I believe the long-term average of the polls (by either mean or median) puts the parties on: 37, 31, 18. So the “errors” from the average here is: -2, 0, +2 – all well within the quoted 3% MoE.

    However unless Communicate are now a monthly pollster and until they get a track record (aside from one of major volatility), I’ll continue to take theirs with a pinch of salt.

    Any word on todays IPSO-MORI poll yet?

  6. Philip – yep, a Communicate graph in the corner is overdue. I’ll probably put it up next week.

  7. Anthony,

    fair enough, and it might explain why the marginal drop in LibDem support comparedto the last scotsman poll would see the LibDems drop from a predicted 23 seats to only 17.

    A list share of 14% would see them not only probably loose some list seats, but it also just wouldn’t be enough to make up for loosing one or more FPTP.

    On that note i was out canvassing over the weekend in Rossshire (tee-shirt weather in the Highlands in March, roll on global warming) and for the first time in a decade I was actually getting LibDems coming across because they felt we needed a change.

    Up until now we have mainly been battling with Labourfor second place, and the SSP cost us some vote but we are now getting a definite swing from the LibDems too.

    I had been predicting a good second with labour defections and tactical voting from tories, but an 11%+ swing just wasn’t on the cards because the LibDem vote in what is Charles Kennedies westminster seat seemed rock solid.

    Now, although I think a unset is still highly unlikely, certainly at least in Holyrood terms, this might stop being seen as a safe seat, and I wouldn’t have believed that even a year ago.

    Peter.

  8. Nick , Communicate does not especially cheer me up . Apart from the weighting problem , it suffers the same as ICM from such a small sample nominally 1,000 but excluding don’t knows won’t votes circa 700 so M of E is around 4% .
    It is interesting that the 2 weekly German opinion polls out today have sample sizes of 4,154 and 2,508 . There is the constant variance between the 2 with Forsa showing SPD support 4% lower than Emnid and this variance occurs week after week but trends are much easier to pick up than with UK polls .
    Ipsos Mori poll out now but will place no reliance on that either .

  9. Wow, big difference with the IPSO-MORI result over the Communicate one. For the Tory vote at least.

    Relative to the long-term averages that I posted above: Tories +4, Labour +2, LDem -1

    The Lib-Dems seem pretty stable around the 17/18 figure, but I’m finding it hard to guess the real position of the Tories. This is now the sixth poll in little over a month to give the Tories a figure of 40%+ and the third this month to give us exactly 41%. But no poll has yet beaten that figure so by definition its an outlier, but an increasingly common one.

    May should give us some fascinating results at this rate, especially if the SNP can keep up their consistently good results between now and then.

  10. We’re currently predicting:

    SNP – 49
    Lab – 38
    LDm – 19
    Con – 18

    but some of that is because we haven’t examined properly which independents are still standing this time round.

    What’s interesting is that the high SNP support is also being shown in Westminster polls, so our current Westminster prediction is

    Con – 336
    Lab – 230
    LDm – 28
    SNP – 28

    which is a major victory for the SNP and I think quite possible given the position on the ground in Scotland. Even with the Comm Res figure this morning, the current polling figures are indicating a very firm Tory victory AND a lead in the polls of about 8.5%.

    Not sure yet about this month’s MORI figures – we’ll have a careful look before deciding whether to include them in our next forecast.

  11. By the way Anthony, did you know that your RSS feeds are timing out – both the main feed for this page and comment feeds for individual posts.

  12. Before we get too many comments on how out of synch these two polls are; or that one methodology is more reliable than the other, do note that the polls were conducted between two and three weeks apart: Communicate between 23rd and 25th March, MORI 9th and 15th.

    I think that gap is not insignificant, because earlier in the month coverage of Labour was far more hostile than it has been more recently – not claiming that Labour’s been getting plaudits in the past fortnight, but cash for honours was flooding the airwaves in the earlier part of the month – it hasn’t been this past fortnight; there’s been Blair’s seemingly well-received spot on Comic Relief (which many millions will have watched) and though the Budget hasn’t gone down especially well, it’s still been news focussed on Labour’s agenda.

    Can that explain a 4% difference in the Tory lead? I’d certainly say it was worth 2 or 3% of it, so possibly, yes – and no-one’s surely going to quibble over a 1% margin of error difference are they?

  13. Yes – thanks Peter – it’s because of the flash graph a couple of posts down. I need to replace it with a flat jpeg, but haven’t had the time to do it yet.

  14. Peter O,

    The question that will be asked post may about westminster, is Would the thought of a Tory government make people in Scotland rally to Labour to prevent it, or switch to the SNP.

    Peter.

  15. I don’t think there’s any doubt that a 4% Tory lead is closer to the likely result of the next election than a 11% lead, so, in that sense, I don’t think this can be deemed a rogue poll.

  16. Communicate Research’s monthly poll seem to the lay viewer as amazingly volatile and thus worthless.

    If you average the results out as Phillip has done they are not too far off other pollsters but the purpose of a monthly poll must be to show change not provide an average.

    And Andy polls are conducted to assess the mood of people now not the likely result of the next election.

  17. Peter C,

    Yes, as we get closer to a Westminster election we will see clarification. At the moment it is very likely that the high level of support for SNP in Westminster opinion polls is a by-product of the anti-Labour SNP support for the May elections. Where that anti-Labour vote goes in a Westminster election (if it still exists) remains to be seen.

  18. There is something strange about the WB swingometer. As I am convinced that the Greens get the necessary 7%, I have run it with Greens at 7% instead of 3%, with Lab, SNP and Lib all with 1% less (i.e. 29%, 34% and 13%) and Con unchanged at 14%. The result is 7 seats for the Greens (as expected), but then Lab 42, SNP 48 and Con 14! How come the Conservatives loose 4 seats if I deduct 1 % each from Lab, SNP and Lib?!?

    Re Anthony Wells comment on getting lost constituency votes back through the list, this really depends on where you loose you constituencies. Any Lab loss in Strathclyde is for real (as for the LDs in the Highlands), and cannot realistically be retained through the list.

    There are also other inconsistencies in the WS swingometer. Even with the Populus numbers, the LDs retain Gordon. Is that realistic? And I can’t construe any scenario that lets Independents keep Strathkelvin. Maybe WS could amend the swingometer so that after calculate the constituencies, it allows users to make specific changes before calculating list seats…

  19. Ralph – by their nature opinion polls cannot detect change from month to month in public opinion as the you can never tell whether small changes are genuine or caused by sampling/M of E .

  20. Depends how significant the change is. A 6% swing for example is theoretically within the margin of error, but the odds of both parties being at the edge of their margin of errors is pretty low.

    You’ll never know for certain, but you can make a reasonable guess.

    And once you look beyond just month-to-month, you can see swings well beyond the margin of error. Clearly and polls taken at the moment are getting significantly different results to this time next year. Which doesn’t mean that they represent the next election in another years time or potentially another three we could be getting wildly different results – but they do represent a mood now.

  21. True Philip except as soon as a 6% swing is shown everyone then crys rogue poll to the new or previous poll or both .

  22. A definition of a rogue poll..

    A rogue poll is the one that suggests a result least palatable to the rogue reading it…

  23. in scotland,dont discount lib to labour ,and a small amont of con to labour switchers to prevent an snp win.

  24. Since I believe it’s going to be far more important than anyone will admit, at least in a few seats at the GE and certainly in May, why is no breakdown of “other” supplied with the “headline” results? To know the movements in “other” over the months would be fascinating.

  25. John H , the problem is that the different polls cannot agree on the level of Others support which ranges from 7% in Mori Jan poll to 15% via a typical 10% in an ICM poll . These variations are far higher than simple M of E can explain .

  26. Anthony,
    Unless my vision is even worse than my wife claims it is then what the Independant says about the latest Communicate poll it has just published is plain nonsense. It claims that this poll indicates that Labour is in the lead in the South West where in terms of seats Labour came third in 2005. On the other hand it says the Tories are in the lead in the North of England-an equally unlikely scenario. Is this just a confused muddle by the journalist concerned or did this poll really suggest this?

  27. Greenpousse,

    I don’t think there is anything fundamentally wrong with the WB swingometer. The reason that the tories drop is that as the weakest of the four main parties with virtually no FPTP seats and a small list vote, they are actually vulnerable to the Greens.

    If the greens get 7% then they will get a seat in every region which will be disproportionately at the expense of the closest party to them in terms of regional vote. Which in much of Scotland is the tories.

    Asto predicting for the Likes of John Swinburn, Margo Mcdonald Alex salmond in Gordon, or things like a hospital closure no swingometer can do that as they can only compare current polls with past results in an election and extrapolate uniform swings.

    Thats why I think it is significant that here in Rossshire we have LibDems coming across to the SNP, because on uniform swing that shouldn’t really be happening, but it is.

    Peter.

  28. If the SNP continues to have a good result in Scottish Opinion Polls for a UK General Election then the Conservatives will have very limited advances in Scotland indeed.

    It is almost certain that the Conservatives will gain ‘Dumfries and Galloway’ at the next election but there are other seats in Scotland where they have maintained a firm grip on many 2nd places. ‘Perth and North Perthshire’ and ‘Angus’ could easily become Conservative seats with just a 1.7% swing in ‘Perth and North Perthshire’ and 2.1% in ‘Angus’ from the SNP.

    However, although the Conservatives do seem to be making small amounts progress in Scotland for Westminster polls, the SNP are gaining much more and I think that unless the Conservatives effectively tackle the SNP then there advances into Scotland on Election night in 2009/10 will only be a miserable 1 seat.

  29. Peter Cairns,

    Your feedback from Ross-shire poses an interesting question, given that recent history shows SNP performing better at Holyrood than at Westminster, while vice-versa for Tories.

    What will happen at the next Westminster election in LD held seats where Tories are either second or a close third if the SNP take those seats off LDs in May ?

    Two seats in particular spring to mind, Argyll and Gordon. Argyll has been a three-way marginal. Would losing the Holyrood seat to SNP make it a potential SNP/Tory marginal in 2010 ?

    In the case of Gordon it becomes even more interesting since, should the SNP take this in May, it will be down to personal support for Alex Salmond which could not be repeated in 2010 (unless he were to abandon Banff & Buchan). I doubt the SNP would take it at Westminster, which makes it wide open.

    Separately, how do you think an SNP / Con coalition would go down with the Scottish electorate ? This may sound bizarre, but there are strategic arguments for it, and maths is likely to work.

    Paul

  30. Philip,

    The problem with your way of looking at it is that in the straight SNP Tory seats the SP has already mopped up the anti tory vote,

    If between them the SNP and Tories have over 80% of the vote even if there is a 10% swing to the SNP at Westminster, compared to the last election, in these seats, there just aren’t that many Labour and LibDem votes to give it to them.

    Equally I doubt that in the seats where the Tories are strong in Scotland and where there support generally turns out, there is much in the way of New Tory support unless it’s people who defected to the libDems, but that tends not to happen in seats that the Tories can win.

    My gut reaction is that if the SNP is the biggest party in Holyrood but doesn’t get to form a government then the scots will back them at Westminster because they will feel cheated ( which gives the interesting constitutional possibility of a majority of Scots Westminster MP’s being SNP, which used to be what Labour said was needed to leave the UK).

    Conversely if the SNP for a government in Scotland then a lot of people might view Westminster as Labour v Tory and all this SNP support dissappear like snow of a dyke.

    Peter.

  31. Nick,

    I noticed your comments re the Communicate Research poll, so I decided to take a look. Yes, the Conservatives do lead Labour in the North of England (29% to 27%); while Labour lead the Conservatives in WALES and the South West (27% to 25%). I’d say its very much a given that the Conservatives are ahead in the South West but the inclusion of Wales gives Labour the edge.

    In the South of England, the Conservatives lead Labour (32% to 21%) and in the Midlands, they lead Labour (29% to 28%).

    The Conservatives lead Labour among AB’s (32% to 22%); C1’s (29% to 27%) and DE’s (26% to 24%); but among C2’s are ahead (30% to 22%).

  32. If its OK for me to ask about it: I”m curious as to the idea that it’ll be better for the SNP, if they win May’s election, to hold back on the referendum until after the next general election.

    If they were to hold it relatively quickly they be going in on the back of an election victory, guaranteed (since if they hadn’t won, there’d be no referendum). People will by and large be positive and optimistic.

    By waiting 4 years they get 4 years worth of time to make enemies through making the tough decisions governing will always give – and Labour 4 years to recover, at least in Holyrood.

    Furthermore if as you say is possible Peter, the SNP doesn’t do well in Westminster, then they’d be going into the referendum on the back of disappointment and defeat. Hardly the best timing.

    Holyrood is a much better proposition for the SNP than Westminster is, and generally will be. So I don’t see the benefit of timing the referendum for after an election where by its nature they’re not as likely to do as well.

  33. Philip Thompson:
    > If its OK for me to ask about it: I’’m curious as to the idea that it’ll
    > be better for the SNP, if they win May’s election, to hold back on
    > the referendum until after the next general election.

    From what I’ve seen, most polls are putting the SNP ahead of Labour (so they’re the majority party) – but they are still short of the 65 seats needed to gain an overall majority in the Scottish Parliament.

    This means that SNP will have to govern in a coalition, with the most likely partner being the Lib-Dems (alternatively, the Labour Party might still be able to govern if they partner with the Lib-Dems). One proviso the Lib-Dems are likely to ask for in return for a coalition is that there is no referendum on Scottish Independance (since the Lib-Dems are against this). So, this would make it unlikely that there would be such a referendum in the next Parliament.

    I think the most interesting part of the Scottish elections will be the 28 days after the election – with the deals in smoke-free rooms. Realistically it could be a choice of an SNP/Lib-Dem coalition or a Lab/Lib-Dem coalition. If it’s not sorted within 28 days…another election!

  34. It’s quite possible that the SNP and Labour would lack a majority even with LibDem support which could leave the Greens or Conservatives as kingmakers. Interesting thought.

    I was interested in Peter’s report of possible LibDem transfers to the SNP. Beating John Farquhar Munro would be a huge upset – can’t really see it. Could be significant in other seats though – notably Argyll & Bute which has a LibDem MSP but which the SNP “won” in the regional vote. Could also have an unintended effect in LibDem/Tory marginals in places like the Borders.

    By the way, I doubt that many Tories would vote tactically SNP to beat the LibDems as Peter seems to suggest. Much more likely to be the reverse. In my view Inverness East which Fergus Ewing holds for the SNP could become vulnerable to the LibDems if the Tory and Labour votes slip significantly which is very likely judging by the 2005 result.

  35. Assuming the SNP becomes largest party, then the LDs have three choices: Coalition with SNP (i.e. vote for Salmond as FM), support SNP minority (i.e. abstain and let Salmond get elected with relative majority) or form a minority coalition with Labour (i.e. vote for McConnell, and presumably supported by the Tories). The last option is really a non-starter, since Labour will be in turmoil if they lose, I’d doubt the LD activist will take it (remember LD switch in Perth & Kinross).

    In case of coalition, SNP and LD would have to agree to an open vote in parliament. In case of outside support this would happen anyway. It would then be up to the SNP to put pressure on the other parties to allow a referendum, and given the popularity of having a refendum in Scotland this shouldn’t be too difficult. In fact the SNP could turn the next Westminster election into a referendum on the referendum…

    Oh, an in a coalition the Nats should insist on an unwhipped vote, I guess there are quite a few LD (and indeed Tories and Labour) MSPs who wouldn’t object to a referendum

  36. Re: Greenpousse

    I very much doubt if any Tory or Labour MSPs would vote for a referendum. They loathe the Nats.

    As for the LibDems I am quite sure they would hate to see the SNP become the largest party. My impression is that they are far more comfortable in a coalition with Labour. Also the great prize for the LibDems would be a coalition at Westminster with Gordon Brown. Propping up an SNP administration and facilitating a referendum would poison relations – can’t see Menzies Campbell being happy with that.

    My guess is that the SNP will have to do without a referendum if they are to form an administration. It is the only way they could persuade the LibDems to support them – as it would provide the latter with political cover to do the deal.

  37. Covering a couple of issues.

    The SNP have never governed, and in the past part of the success in attacking independence has been the SNP’s lack of experience in Government.

    It is pretty much recognised that until it has proved itself capable of running the executive The SNP wouldn’t win an independence referendum.

    Support for independence is probably running at 50/50.

    With around 40% for, 40% against and 20% undecided. The reason for the so called fall in support, is that the Populus poll was a three way question, Status quo, Independence or morepowers.

    If you divide eighty by two you get forty, if you divde it by three you get about twenty seven. Some papers like the times have missed that, and others like todays scotsman are deliberately miss-interpreting it.

    What we can be pretty sure of is that about a third will never vote for Independence, a third will always vote for it and a third will go for it if they think that taxes won’t rise or services be cut.

    Thus the SNP has to deliver “Better for Less” for Scotland to be independant, and it needs at least four years to do that.

    If you look at Local income tax , the current policy is freeze the Council tax for three years before introduction followed by at least two years at 3% on earned income in lines with UK tax thresholds.

    That effectively means attacking head on the “Independence means higher taxes” argument.

    Another aspect of the local income tax is the status of the Council Tax rebate.

    The treasury argument is It’s to support the Council Tax so no Council Tax no support. The SNP argument is , to paraphrase ” It’s Ooor F*^&%*@ Money”.

    Now how will this play in the run up to a UK election.

    Does Brown stand his ground and let the SNP campaign on “Browns Penny for Scotland” the Local income tax having to be set at 4% not 3%, or does he give in and let the SNP have the cash.

    Standing firm could cost him seats, but giving in would be a victory for the SNP. In principle Cameron has come out in favour of more powers for Holyrood, so does he say “Yes of course it’s Scotlands money” and try to put Brown on the spot.

    All of the issues of more powers that the SNP have raised are in effect setting out the battleground for the Scottish Westminster campaign, and potentially a hung parliament.

    Finally heres an interesting one for people to number crunch, would Camerons prospects of government actually be better if there was an SNP surge at Westminster rather than a Tory revival north of the border.

    You should never fight a war on two fronts, but that is exactly what Brown will probably have to do.

    Peter.

  38. Obviously it would be better for the Tories I think were Scotland to go independent. But here’s a thought – could the removal of so many Labour MPs, combined with a collapse in Labour support after their period in government provide the platform for the Lib-Dems to finally become the Official Opposition?

    Independence may scupper their chances now at becoming minor coalition partners, but could provide them the chance of replacing Labour as the left-wing party in independent England.

  39. Mark Senior. Yes I agree. My theory is that since some or most of the “other” is inevitably “fringe” it will depend on how confident those “fringe” supporters are in feeling able to state their true preference. To be blunt, I suspect quite a lot of it is BNP support and their public profile varies around the country. That will have a direct effect on stated support in dfferent areas. The sampling of the different pollsters will be to different profiles. This could explain the varying amount of “other”. Nevertheless, the actual percentage of “other” does seem fairly constant pollster by pollster, and an analysis would still be interesting.

  40. Thanks David but I still have my doubts about the reliability of Communicate Research’s polling methods.For the Tories to be 2% ahead in the North of England of all places would appear to imply a truly massive lead in the country as a whole -and that’s not what Commmunicate Research are saying. I suspect that their regional samples could be a bit too small to enable them to extrapolate meaningful figures. Nevertheless it seems that the reason for the dip in their headline Tory lead could be related to the meagre 32% they have ascribed to that party’s fortunes in the South of England.That looks to me to be also highly questionable-what was the 2005 GE figure I wonder? It must have been more than that. Something has gone wrong here.

  41. http://news.bbc.co.uk/1/hi/uk_politics/scotland/6506011.stm

    New poll suggesting a strong Green presense

  42. Nick , the regional vote shares and other subsidiary figures are based on too small samples to have any accuracy . Also although the whole sample will be correctly demographically weighted that will not be the case with the subgroups hence that particular sample from the North may have an abnormally high number of AB over 55’s and the South West an abnormally high number of C2’s .

  43. This May’s elections could be calamitous for the LibDems in the long-term, even though in the short term they do ok. It’s not inconceivable to see them in coalition with the SNP in Scotland and, it has been mooted, in a “rainbow” coalition with the Tories and PC in Wales. IF all this came to pass they would have been in coalition with Labour, the Tories, the SNP AND PC when it suited them and wide, wide open to charges of being all things to all people and power-hungry rather than principled. It could cost them dearly if these charges stuck during a General Election campaign

  44. Or it could benefit them to finally dispel the notion of them being a wasted vote once and for all.

  45. “could the removal of so many Labour MPs, combined with a collapse in Labour support after their period in government provide the platform for the Lib-Dems to finally become the Official Opposition?”

    Not this old 1983/87 chestnut again? . I fail to see how a recovery in the Tory position would lead to the LDs gaining more seats than Labour in England.If anything the evidence points to a Tory recovery damaging the LDs at the polls at least as much as Labour, perhaps even more so.Look at the LD performance in last year’s local election once the Tories started to recover in the polls, added to which there are dozens of LD seats that would fall to the Tories if the current poll figures translated into a general election result.

  46. Anthony,

    The Herald says today that the Scottish Daily Mail also had an opinion poll published yesterday, by Scottish Opinion (a brand name of Progressive Partnership). As I don’t read said organ I was unaware of this (they do not publish their Scottish edition online, as far as I am aware).

    From today’s Herald:

    “A Daily Mail poll conducted by Scottish Opinion gave the SNP a six-point lead over Labour – by 34% to 28% – in the constituency vote, and a 32% to 28% lead in the regional list vote.”

    http://www.theherald.co.uk/politics/news/display.var.1296519.0.0.php

    I am not sure if PP/SO are members of the BPC, therefore I don’t know if the detailed datasheets and methodology will be publicly available. Do you know?

    Where did The Scotsman’s monthly ICM poll go? Should it not have been published today? Call me suspicious, but I suspect there is a rather obvious reason why the strongly Unionist Scotsman chose not to publish it right at the firing-gun of the campaign ;)

    PS. Could you update your excellent “Scottish Voting Intention” page (thank you for the service!!) with both the Populus and the PP/SO ones.

  47. What I was trying to say is that if the SNP emerges as strongest party, the LDs have no alternative to prop up an SNP government in some for or another.

    Of course there will be no agreement on an independence referendum in May/June 2007, that’s why the SNP is planning to delay the actual referendum until after the next Westminster election.

    Meanwhile they need to find popular issues to push Westminster / put pressure on the other parties. I think the issue of direct democracy is a good one (what would the LDs do if faced if a general referendum bill, along the lines existing in most German states?), Peter C suggest the council tax rebate, there are lots more.

    On the back these campaigns and a competent government, the Nats can win Westminster 2007, in which case nobody really can deny them a referendum. Of course, its a big ask (esp. the bit about competent government), but that’s their best bet

  48. Very true Philip, and the advantage of a site like this that different opinions and interpretations of positions can be made. I do feel, though, that if people see the LibDems making deals left, right and centre (literally, politically-speaking) it will be easy to paint them as prostitutes for power, which I’m sure is not how they’d like to be portrayed

  49. Opinion polls measure voting intentions AT THE TIME, not projections for a future general election months or years away.

1 2