I’ve just got back from the BBC after working all night (you may have seen my bald spot sat just to the left of Emily Maitlis’s big touchscreen last night) and am about to go and put my feet up and have a rest – I’ll leave other thoughts on the election until later in the weekend or next week, but a few quick thoughts about the accuracy of the polls.

Clearly, they weren’t very accurate. As I write there is still one result to come, but so far the GB figures (as opposed to the UK figures!) are CON 38%, LAB 31%, LDEM 8%, UKIP 13%, GRN 4%. Ten of the final eleven polls had the Conservatives and Labour within one point of each other, so essentially everyone underestimated the Conservative lead by a significant degree. More importantly in terms of perceptions of polling it told the wrong story – when I was writing my preview of the election I wrote about how an error in the Scottish polling wouldn’t be seen so negatively because there’s not much difference between “huge landslide” and “massive landslide”. This was the opposite – there is a whole world of difference between polls showing a hung Parliament on a knife edge and polls showing a Tory majority.

Anyway, what happens now is that we go away and try and work out what went wrong. The BPC have already announced an independent inquiry to try and identify the causes of error, but I expect individual companies will be digging through their own data and trying to work out what went wrong too. For any polling company, there inevitably comes a time when you get something wrong – the political make up, voting drivers and cleavages of society change, how people relate to surveys change. Methods that work at one election don’t necessarily work forever, and sooner or later you get something wrong. I’ve always thought the mark of a really good pollster is someone who puts their hands up to the error, says they’ve messed up and then goes and puts it right.

In terms of what went wrong this week, we obviously don’t know yet, certainly I wouldn’t want to rush to any hasty decisions before properly looking at all the data. There are some things I think we can probably flag up to start with though:

The first is that there is something genuinely wrong here. For several months before the election the polls were consistently showing Labour and Conservative roughly neck-and-neck. Individual polls exist that showed larger Conservative or Labour leads and some companies tended to show a small Labour lead or small Conservative lead, but no company consistently showed anything even approaching a seven point Conservative lead. The difference between the polls and the result was not just random sample error, something was wrong.

I don’t think it was a late swing either. YouGov did a re-contact survey on the day and found no significant evidence of this. I think Populus and Ashcroft did some on the say stuff too (though I don’t know if it was a call-back survey), so as the inquiry progresses other evidence may come to light, but I’d be surprised if any survey found enough people changing their minds between Wednesday and Thursday to create a seven point lead.

Mode effects don’t seem to be the cause of the error either, as the final polls conducted online and the final polls conducted by telephone produced virtually identical figures in terms of the Labour/Conservative lead (though as I said on Wednesday, they were different on UKIP). In fact, having a similar error with both telephone and online polls is evidence against some other possibilities too – unless by freakish co-incidence unrelated problems with online and telephone polling produced almost identical errors it means things that only affect one modeare unlikely to have been the cause. For example, if the problem was caused by more people using mobile phones, it shouldn’t have affected online polls. If the problem was caused by panel effect, it shouldn’t have affected phone polls.

Beyond that there are some obvious areas to look at. Given that the pre-election polls were wrong but the exit polls were right, how pollsters measure likelihood is definitely worth looking at (exit polls obviously don’t have to worry about likelihood to vote – they only interview people physically leaving a polling station). I think differential response rates is something worth examining (“shy voters”… though I think enthusiastic voters is just as risky!), and the make-up of samples is obviously a major factor in the accuracy of any poll.

And of course, it might be something completely unrelated to these things that hasn’t crossed our minds yet. Time will tell, but first some sleep.


710 Responses to “Back from the election”

1 2 3 15
  1. I was almost afraid to come back to this site today.

  2. I wonder who was the most gobsmacked when that exit poll was announced? There must have been a lot of competition.

  3. Well I am no statistician but the polling figures look spot on actually save a 3%ish swap from Labour to Conservatives. Is it possible that many voters flipped at the last minute amid fears of a woan and tartan army hell bent on a coup and stealing all our (English) money or is it that the current models are simply wrong and were underestimating the Tories all along?

  4. Here’s something you may wish to ask your fellow pollsters. Did anyone think to ask: “If you thought there was going to be another coalition, which party would you vote for to avoid it?”

  5. @SORREL – I felt physically sick when I saw it. Irony is, what wouldn’t I give for it to have been totally accurate today!

    RE: EXIT POLLS

    Are these totally anonymous, with a ballot paper being put into a sealed box?

  6. The same problem showed up in recent US elections (2012 and 2014). It’s called “herding”–where all the pollsters compare their results with each other, and they (unintentionally) interpret their data to more or less agree with each other. The net effect of this is that the polls converge (“herd”) toward a consensus that is based more on conventional wisdom than actual data.

    Voters don’t necessarily obey conventional wisdom. Thus, significant polling errors.

  7. AW

    Thanks for your usual excellent piece, you must be exhausted.

    However I think I really do think it was a very late swing actually on the day as I posted at 3.35pm on the last thread:-

    Obviously I am pleased my forecast of the last three or four years has proved to be correct and the Conservatives have a small overall majority. As I had been forecasting for many months now, the election was decided on the day with a late swing to the Tories in the polling booths up and down the country. Anyway that means a hat-trick of correct forecasts, The Euro’s, The Scottish Referendum and now the 2015 GE.
    Don’t blame the pollster’s, they cannot record a very late swing like that. In many ways they got the election correct. Their forecasts for Scotland were almost exactly right, as was their forecast of a collapse of the LibDem vote, and the success of Labour in inner London. I have been accused of being lucky and just guessing the result, not so. In getting to my own forecast I did use the opinion polls, not the headline numbers, but the answers to the detailed questions on things like the economy, Leadership, the NHS, immigration etc, etc. Once it was clear that the Tories had big leads on “best leader” and “best at running the economy” which they maintained over time, I just could not see how Labour could win. That basically was what I called my “gut feel”, developed over many, many, years, having been deeply interested in politics since I was 18 in 1958. I have voted in every election since then although I have never had an opportunity to make any impact since I have always lived in seats with large majorities, either Conservative or Labour. The only time I was voting positively for a party was in the 80’s as I was inspired by the leadership of one woman. So yesterday I just voted for the party I thought most likely to manage the economy better than the others, as I have done for most of my adult life. It would seem many. many others did the same.
    The actual election campaign from both Conservatives and Labour was dismal IMO, being mostly negative in style, with little vision, although I did think Cameron improved towards the end of the campaign. IMO the most significant additional factor in the Conservatives favour during the actual election campaign was the very strength of the SNP in Scotland. Clearly the loss of most Labour seats in Scotland was going to be a heavy blow, and the polls showed little support for the idea of a minority Labour Government propped up by the SNP. Indeed the polls showed the most favoured election result towards the end of the campaign was a Conservative majority. A polling response that hardened my view of the likely election result.
    For those of you who are deeply upset by the result my commiserations, maybe you will have your turn in 2020 although that looks more difficult now, as no doubt the boundary changes will come into force sooner rather than later. I was personally sorry to see Danny Alexander lose his seat in Scotland, I thought IMO he did an excellent job as Chief Sec to the Treasury. Indeed I think the LibDem’s acted very honourably in 2010 in helping to form a stable government. On the other hand I cannot hide my delight at the fall of Ed Balls, a man who refused to take any responsibility for past mistakes, nor do I have much sympathy for Ed Miliband who took his party more to the left IMO, and paid the price, like his Welsh mentor before him.
    AW runs this site well and his summaries are always interesting and informative, so well done and my thanks AW. Generally posters here are polite regardless of their own political views. Occasionally I have been told that my posts are ridiculous or not relevant and of course at the time they may have seemed so. Maybe I will get less of that in future, although I don’t mind anyway.
    David Cameron has won a stunning victory, (remember the current system has an anti-Tory bias). We were told that Cameron could not increase the Conservative share or increase the number of seats. Well I think I’m right in saying he did both!
    We now have a new Government who face the problem of what to do about the political divide between rUK and Scotland, as well as continuing big issues with the UK economy, and our relationship with Europe. Interesting times…………………………………

  8. @PRAWNY369

    YouGov asked the same people before and after, with no change, suggesting it was wrong in the first place.

  9. Matt M –

    Yes, the interviewer has a ballot box and duplicate ballot papers, and gets people to vote in just the same way as they, um, vote.

  10. First off congratulations to those Tory posters on here who have celebrated victory without the need to gloat. Enjoy the win.

    Secondly thanks for having me on this board. I have thoroughly enjoyed my time on here. However it is time to say farewell. I have lost all faith in the pollsters who despite poll after poll after poll were hopelessly wrong. This was no margin of error. This was a complete failure of the pollsters to come remotely close to getting it right. I can no longer post on this board purely because the polls do not have any relevance any more. How on earth can they be trusted again?

    My final comment is this. It was Agent Sturgeon wot won it. She should be the toast of all Tories and former Labour voters in Scotland may slowly begin to realise that they have been duped and in reality she got what she wanted…a Tory majority by destroying Labour in Scotland and badly damaging them in England. She can now say to the Scottish electorate that the Westminster Government does not in any way represent the will of the Scots. Another independence referendum will now happen sooner rather than later.

    Farewell.

  11. I think there is something fundamentally wrong in who was being polled. The whole world (almost) felt blindsided by this.

  12. Number Cruncher called it right. There were nine pages of explanation on that site about why all the polls were underestimating the Tory lead.

    As a Labour supporter I am glad that I was forewarned to expect an 8 point Conservative victory, so it was not a total shock to me.

  13. @The Other Howard,

    “David Cameron has won a stunning victory, (remember the current system has an anti-Tory bias).”

    Is that really true though? As far as I know, the system was regarded as having a Labour bias because the complex relationship between boundaries, turnouts, votes etc had in recent times resulted in Labour having a better seats-per-vote ratio. Now it’s the other way round and the Tories have the better ratio. It’s not so much about the boundaries as how efficiently each party’s vote is distributed and of course the vagaries of FPTP.

  14. @Anthony

    Not sure if it’s appropriate to comment, but what do you think of Survation saying they won’t change what they’re doing because they did get it right with one unpublished poll?

    Surely they’re taking a chance on that last poll not just being towards the edge of the margin of error?

  15. I believe the relevant phrase is ‘It’s the economy, stupid’.

    Many people may have told pollsters that they were going to vote Lab or LibDem, but when confronted with a ballot paper and a chinagraph pencil, decided that they couldn’t risk having EM in no.10 with who knows how much additional borrowing, and another note in the safe saying ‘there’s no money left’.

    That, and the spectre of the SNP calling the tune to a minority Lab administration, had a telling effect that the pollsters all failed to pick up.

  16. @ Anthony Wells

    Can it be the misunderstanding of the LibDem churn? The misunderstanding of the non-existent collective soul of LibDems? I don’t say that you considered them identical, but here is something interesting. I have very few YouGov NowCast constituencies, but it is clear that they moved differently in different constituencies, but almost identically in the same constituency.

    Anyway, I just want to repeat that Godot was there already, only nobody saw him.

  17. Must admit to being in shock and a little depressed. And I am sure there are many more feeling the same. I feel more than a little let down by the polls but then again the exit poll got it spot on! Forewarned is forearmed; watching the politicians, of all parties, was not a joy last night.

    I think this site is amazing and will continue to come back and see how things are developing, I think the media will move on quite quickly and will not keep us informed about what went wrong this time.

    A big thank you to AW and everyone on this site, may the force be with you (not the dark side).

  18. @ Anthony Wells

    Following on Matt M’s comment

    @Anthony
    Not sure if it’s appropriate to comment, but what do you think of Survation saying they won’t change what they’re doing because they did get it right with one unpublished poll?
    Surely they’re taking a chance on that last poll not just being towards the edge of the margin of error?

    What do you think of their point of not publishing a result just because it looked a bad outlier? I don’t quite expect an answer as it is a competitor.

  19. Richard P

    I don’t know enough about it to argue with you in detail on electoral bias. In any event there is a need to revide the boundaries and i am all for reducing the number of MP’s to 600. All savings are welcome as the deficit still needs to be eliminated.

    I am not a fan of Cameron, nor am I a Tory as those who post regularly know (but I am right of center) but did win a spectacular victory by any measure. Tell me who else has increase the party national share and increased the number of seats?

  20. Mikey

    The Scots have not been ‘duped’. The Scots are not stupid. But the Scots did expect ‘better’ of the English!

    Many ‘ex-Labour voters’ in Scotland have voted SNP before – for example in Holyrood and Local elections.

    And one final thought: what most people south of the Border seem to me to be unaware of is that the main focus of political debate in Scotland is not Westminster but Holyrood. The leaders of four Scottish parties (SNP, Greens, LDs and Tories) are MSPs. Labour have to learn from this.

  21. It’s a bit embarrassing for the polling industry isnt it @Anthony!

    I think that remark from survation?, who said they started to identify this on one poll, weds, but didn’t publish as wanted to go with the group is a disgrace really. Was jaw dropping.

  22. If you look at shares of the vote in constituency after constituency the libdem vote collapsed and tory,labour,ukip and green shares went up but by differing amounts.The key to the election was that in the labour targets as much,if not more, went to the tories rather than labour.

    There are three reasons for this,the economy(past and future),leadership (miliband only neutralised this for labour supporters) but imo the late game changer was the snp threat -principally the biggest contributor was sturgeons wild over the top “lock the tories ” out of downing street playing to try to understandably maximise the snp vote without thought to the consequences in england.

    Pr is now the only way labour can square its circle of simultaneously going left in scotland and right in england.

  23. Didn’t Omnishambles post something the other day about some site that was saying the Cons might actually be, or should be around 7 to 8 percent ahead?

    I live in a what was marginal seat. As someone who swapped from Lib Dem to Con in this election, I can tell you my reason, and I think I may not be alone. In 2010 Lab held power and warned us all how frightening and terrible a Con gov would be, talk of “savage cuts” etc they really painted it as doomsday, but now on reflection over the past 5 years, actually none of that doom and gloom has happened.

    I notice that in many of the marginals Tories won from Lab in 2010, there was another swing in favour of the tories there too, and I believe that to be many people like myself who live in marginal seats slightly cautious about voting Con last time because of the fear.

  24. I think the article makes a lot of good points. The concept of the shy Tory voter is frankly silly, more like the lazy Labour voter. They are not as committed to voting, it would seem. I know the problem, this is the first year for a while I’ve felt motivated to vote Labour myself. Also, using online or telephone polls attracts those who have made some effort and spent some time to participate. So the sample is more inclined to vote (whatever they say) than the general public a whole.

    Lastly, let’s not see those involved with the exit poll get too smug, as they didn’t make the individual seat calls right at all, they just got the overall picture correct. Gedling was shown as a blue stepping stone to number 10 by the exit poll but Labour won comfortably, which wasn’t good for me as a Gedling voter and someone who had wagered hard cash on the outcome being a Labour win. In the event I needn’t have worried. The same applies to Nottingham South called as a Tory gain almost as a fact prior to the event when it was held easily by Labour.

  25. According to the radio news earlier today, it said Survation had a poll on Tues/Wed which said CON37% LAB 31%, but they refused to publish it as it didn’t agree with previous or other pollsters efforts.

    Groupthink anyone ?

  26. Yep, Number Cruncher’s analysis was spot on.

  27. @casual,

    That really is a total disgrace by survation.

  28. Could there of been lower and higher turn outs between certain groups of people causing the difference between the polls and the results

  29. Congratulations to the Blues, while my side (both the party I support, and the party I tactically voted for) have been crushed – at the end of the day, the people have spoken.

    Democracy is a wonderful thing, and while the Conservatives celebrate tonight I take solace in the fact nothing lasts forever, and one day they will be crushed by my team.

    So, what’s the solution to the polling errors? Just ask NumberCruncher and Professor Fischer for the answers? :)

  30. Well – all’s not well that predicts wrong but in the end perhaps the UKIP vote was softer Tory voters who in the end knew what they preferred if it was not exactly what they wanted. we should not discount the distorting effect of FPTP which has favoured the Conservatives through the eighties precisely because the Lib/SDLP vote was so evenly distributed. in a sense this election merely unwound the 1997 result in England.

    Wiser heads will counsel more wisdom than I have….

    I make one other observation of modern politics – the pressure placed on losing parties to enter immediately into a leadership election is one consequence of OMOV within all the political parties. In the aftermath of a bad loss it is often hard for any party to imagine the sort of leader who might be best placed to fight and win an election as yet five years away. both Labour and the Conservatives have not made a great fist of it in recent times. It leads one to wonder if there might not be a better solution for all parties but I very much doubt anyone with ambition would share my reservations.

    I love this site – I’ve learned much and hope to learn much more. I enjoy the polls and the rhythm they bring to our public discourse – as ever for good and bad.

    The conservative Party will enjoy its victory but victory always brings hubris and to those who say Labour cannot win in England or the LibDems cannot recover – I still say politics always surprises – who in the gloating gales of 1992 would have thought 1997 might happen and that Labour would govern for 13 years.

    There are two profound questions that are left hanging – the Union of the UK and its part in the EU. These are left unresolved and like Mr Wilson before him Mr Cameron has managed to finesse them as party issues for the present to his part’s advantage but the future beckons. Meanwhile the planet spins and there will be another economic recession or disaster sometime down the line – there always is….

  31. Well I said that the tories would win by 12 overall (to myself, last week), but i didn’t tell anybody …

  32. well, at first glance it seems that the real defeated persons were the poolsters. but could it be that at the last minute many people particularly in england decided to vote conservative because they thought it was better to have Cameron than labour plus the snp ? we may never know, but given that the exit poll was right couldn it be that the others were also right but a crazy shit happened in the last 24 hours ?

  33. The funny thing is that for long enough plenty people here were saying ”Where is the crossover we have been expecting? There is no sign of crossover in the polls, but there should be”

    Or in other words people who knew enough to suspect that crossover should show up, to some degree or another, but made the (understandable) decision to follow the verdict of the polls rather than their own ‘gut feeling’ or expectations that crossover must happen to some degree based on precedent.

    Just as some degree of crossover was expected – but then discounted for the most part because the polls didn’t seem to show it – the feature of recent elections where the pollsters consistently underestimate the Tory vote by a couple of points was discussed but often dismissed, perhaps again because of a lack of apparent crossover showing up.

    Perhaps the message in a UKPR sense is that people on here who know about polls can justifiably have more faith in their understanding of trends and forseable patters than in the figures in the polls themeselves. Or put simply, if you know some degree of crossover to the governing party is inevitable AND the governing party is Tory, whose support the polls will inherently underestimate by a couple of points anyway, you could justifiably have dismissed the final polls and concluded that a 5 or 6 point gap (3 or 4 from crossover, 2 from underestimating) between Tory and Labour was the likely truth based on the parties looking in the polls to be neck and neck with no crossover.

    Of course this is all easy as 1 2 3 in hindsight. I did not see it coming. I thought the crossover would be perhaps 1 point and perhaps also a 1.5 underestimate, so a total 2.5 point advantage to Conservative, enough to be largest party but dethroned by Labour and SNP.

    Anyway, I think that people who know about polls should trust in their expertese to tell them more than the raw data seems to sometimes do!

  34. Guido Fawkes has an interesting piece on the Polls: ‘Comment monkeys and Forecast freaks.’

  35. Concerning the supposed “inaccuracy” of the polls, I have a couple of observations:
    1 – Using ukpollingreport running average of Tory 34 Lab 33 the result was actually within the 3% margin of error that is always assumed to be in these polls, so maybe the result, even allowing for the fact that it went to the extreme ends of the error margin in opposite directions, wasn’t so surprising after all.
    2 – Last time around, the polling houses issued a “forecast” rather than a final poll: could this be a reason for the “errors” observed this time around? My personal forecast, using only the ukpr running average was Tory 36 Lab 32, allowing for a frequently observed swing in favour of the sitting government at the eleventh hour (we saw that in 2010 in favour of the then Labour government). Could it be that the extra work done in 2010 wasn’t repeated this time around? Or maybe that the polls influenced voters behaviours?

    One final consideration: I’ve followed opinion polls on this website for the last 5 years (btw: congratulations to the Host, very informative and thorough), and, even when Labour was polling 45% I never had the gut feeling that they were going to form the next government. Nothing to do with policy or who was leader, I must add. I suppose this is not particularly relevant to the polling issue but, I wonder, did the polling houses have a similar “gut feeling”? Maybe, if they did, they could do worse than finding out the reason for it, in order to put on a better display next time around.

  36. Re Anthony’s comment at 4.29 that the exit poll is a secret vote answers the point I made in the last post on the previous thread. I did not understand how a voter could be shy before the election and yet no longer shy after voting.

    IMO this confirms the shy Tory/UKIP theory

  37. Is a difference between an exit poll and any other is that it is sampling a population that ACTUALLY voted, rather than said it would, or it might, or it was likely to…etc etc.?

  38. @ MANINTHEMIDDLE

    While those ‘savage cuts’ didn’t happen from 2010-2015, can’t it be argued the reason for this was the coalition government? With no Lib Dems to moderate, couldn’t the new conservative majority implement a more right-wing government this time around? Just trying to understand those who did go from Lib Dem to Tory.

  39. It seems there is an issue with the polls’ methodology and I think it will be very much in the interests of the various pollsters to sort this out quickly if they are to recover credibility.

    One could say it was at about the maximum of the margin of error but that would be a silly explanation – modern methodologies utilising social media, mobile phones and on-line questioning seem likely to enhance poor responses.

    Having said all that, I think the LibDem collapse was ‘in the bag’ back in 2010 – and the polls got this by and large. Those who voted for LD then were never going to forgive the treachery or arrogance of Clegg. For that he personally has much to answer for in this ‘hat-eating result’.

    On top of that there is no doubt in my mind that the posturing of Sturgeon was not only despicable it also galvanised the english to ensure that the result was NOT going to allow a grouping headed by Salmond to hold a minority Lab party to ransom. This was probably the main cause of the late swing.

    It was, as has been mentioned earlier here, good for the country to see the back of Ed Balls – another non-listening politician. He exhibited an amazing level of arrogance when interviewed at his count some time after midnight when he still insisted the anti-tory parties could – under Miliband’s leadership – form a govt. A contrast with his colleague Harriet Harmon who cautioned reflection.

    This is the first outright Tory victory for 23 years but under the fptp system it is NOT the end of Labour – however without a leader who listens to ordinary folk and one who is NOT tarnished by association with the current grouping it will indeed be difficult to emerge as a viable govt in 2020.

    It is essential that Cameron now delivers on Europe reform or out and an immediate cap to immigration wheresoever folk come from. And he needs to be true to his word regarding the NHS being safe.

  40. @Lahdidah, Clegg called literally everything wrong this election.

    By joining the Government line on playing the “evil SNP, evil Nationalists” line he sunk himself and Labour in England.

    I wish the polls had been accurate, Labour might have played a more defensive game and saved their marginals and been far more modest in their ambitions and the Libs would have known the scale of the disaster coming.

  41. @Killaray45
    Your post yesterday prompted me to read NumberCruncher’s article and I was entirely won over by his logic, such that I placed a late bet on Con maj. There is systematic error here and I think Guido Fawkes’ point (http://order-order.com/2015/05/08/you-comment-monkeys-and-forecast-freaks/#[email protected]/qRGRjgvS9QLyjg) is consistent with what I’ve been saying about the dangers of Left-leaning group-think. Right-thinking group-think is also possible, of course!

  42. As someone said earlier, ‘its the economy stupid’.

    The economy is recovering quite strongly, unemployment is falling rapidly, interest rates and inflation are low, and the deficit is declining, as it has to. That is a pretty potent hand of cards, and when allied with the superior leadership skills of Cameron over the untried Miliband, we can hardly be surprised.

    I did predict a late Tory swing, and a LibDem collapse, but not on the scale we have seen.

    Two other thoughts compelled voters towards the Tories. Firstly, they did not fancy a SNP presence in government, and, secondly, they did not trust Labour not to tax and spend. The moment when Miliband said that they had not overspent when last in power was probably the most significant sentence uttered by anyone in the campaign. The audience response of disbelief could be heard across the nation.

    I suspect that UKIP will be forever consigned to the margins of British politics, with the forthcoming EU referendum their last chance to become an established force. I doubt that they will seize it.

    Labour can only concentrate upon one task, which is acquiring a reputation for financial competence, as they managed to develop in the first years under Blair and Brown. They have to claim the centre ground, and be credible standing on that platform.

    One final thought: if the LibDems are genuinely finished, is there any point in them continuing independently? Is there not a case for merging with Labour to create a party of the genuine centre? With some nice lofty, social democratic instincts. Just a suggestion.

    As for UKPR, what a great site, which I will continue to enjoy. Hats off to those Tory-leaning posters who took an awful lot of flak, and have been quite restrained with their post-election crowing. Respect. No pun intended.

  43. Moving on.

    what swings required for Lab largest party or majority in 2020?

  44. @JOHN B

    But it’s so convenient to be able to blame the ‘stupid Scots’ and the ‘evil Nats’ instead of looking at the failures of your own party.

    On topic, what I wonder is why the UK-wide polls were so wrong and the Scotland-only polls were so right–in the same election. There is something strange about that factor to me.

  45. Earlier this week I posted here the Tories’ canvassing had them on 300 plus seats and doing better in marginals than national polls suggested.

    It seems to me that the polling companies are going to have to survey much larger samples in future and focus more on the marginals if they want to avoid further embarrassment. This was a rerun of 1992.

  46. @Sergio
    Your solutions are unlikely to work. The Ashcroft polls polled marginals and failed to predict accurately. Also there was a large 22,000 poll just before the election that also got the prediction wrong.

  47. @Swamp I think Labour would be need to leading by 30% in the polls to eek a sliver of a majority in 2020 :)

    Seriously, depends how the boundary changes work out.

  48. The VIs put forward by the polls are %s of the total of those who express an intention to vote for one of the parties.
    The vote shares at the election are %s of those who have actually voted.
    There are no grounds (at least I have never seen any) for assuming that these are the same, or even in proportion.
    As Norman Tebbit has consistently said for the past 25 years, the way to win elections is to get your support out to vote.
    Turnout was 66%, hailed as the highest this century, but about the same as 2010 and well below the usual turnouts from 1945 to 2000. Labour got about 9million, or about 1 in 5 of the electorate.
    34% of the electorate did not vote.

  49. Despite other responses earlier, I believe that the weather did play a part in this election with people combining a walk in the Spring sunshine, with a trip to the polling station.

    I don’t buy the ‘It’s the economy’ angle either. I think the fear campaign against the SNP and the realisation of the possibility of Ed Milliband as PM changed voters minds.

    I feel genuinely sorry for who ever takes the poison chalice of the Labour leadership, with a need to move to the left to take on the SNP which will alienate English voters, or a move to the right to pacify them, but which will alienate Scottish voters.
    It seems that English voters are not going to accept a Labour coalition with the SNP so no chance of a minority government, and no chance in the short to medium term of a full majority either !

    What Labour needs now is a caretaker who can take on the SNP. Not an easy job.

1 2 3 15