Polling myths

Whenever a poll goes up that shows bad news for someone you get the same sort of comments on social media. As I write this piece in May 2017 comments like these generally come from Jeremy Corbyn supporters, but that’s just the political weather at this moment in time. When the polls show Labour ahead you get almost exactly the same comments from Conservative supporters, when UKIP are doing badly you get them from UKIP supporters, when the Lib Dems are trailing you get them from Lib Dem supporters.

There are elements of opinion polling that are counter-intuitive and many of these myths will sound perfectly convincing to people who aren’t versed in how polls work. This post isn’t aimed at the hardcore conspiracists who are beyond persuasion – if you are truly convinced that polls are all a malevolent plot of some sort there is nothing I’ll be able to do to convince you. Neither is it really aimed at those who already know such arguments are nonsense: this is aimed at those people who don’t really want to believe what the polls are saying, see lots of people on social media offering comforting sounding reasons why you can ignore them, but are thinking, “Is that really true, or is it rather too convenient an excuse for waving away an uncomfortable truth…”

1) They only asked 1000 people out of 40 million. That’s not enough

This question has been about for as long as polling has. George Gallup, the trailblazer of modern polling, used to answer it by saying that it wasn’t necessary to eat a whole bowl of soup to know whether or not it was too salty, providing it had been stirred, a single spoonful was enough. The mention of stirring wasn’t just Gallup being poetic, it’s vital. Taking a single spoonful from the top of a bowl of soup might not work (that could be the spot where someone just salted it), but stirring the soup means that spoonful is representative of the whole bowl.

What makes a poll representative is not the size of the sample, it is its representativeness. You could have a huge sample size that was completely meaningless. Imagine, for example, that you did a poll of 1,000,000 over 65s. It would indeed be a huge sample, but it would be very skewed toward the Tories and Brexit. What makes a poll meaningful or not is whether it is representative of the country. Does it have the correct proportions of men and women? Old and young? Middle class and working class? Graduates and non-graduates? If the sample reflects British society as a whole in all these ways, then it should reflect it in terms of political opinion too. A poll of 1000 people is quite enough to get a representative sample.

The classic example of this was at the very birth of modern polling – in the US 1936 Presidential election a magazine called the Literary Digest did a survey of over two million people, drawn from magazine subscribers, telephone directories and so forth. It showed Alf Landon would win the Presidential election. The then newcomer George Gallup did a far, far smaller poll properly sampled by state, age, gender and so on. He correctly showed a landslide for Roosevelt. A poll with a sample skewed towards people wealthy enough to have phones and magazines in depression era America was worthless, despite have two million respondents.

2) Who do they ask? I’ve never been asked to take part in a poll!

Sometimes this is worked up to “…and neither has anyone I’ve met”, which does raise the question of whether the first thing these people do upon being introduced to a new person is to ask if MORI have ever rung them. That aside, it’s a reasonable question. If you’ve never been polled and the polls seem to disagree with your experience, where do all these answers come from?

The simple answer is that pollsters obtain their samples either by dialling randomly generated telephone numbers or by contacting people who are members of internet panels. Back when polls were mostly conducted by telephone the reason you had never been polled was simple maths – there were about forty million adults in Britain, there were about fifty or so polls of voting intention of a thousand people conducted each year. Therefore in any given year you had about a 0.1% chance of being invited to take part in a poll.

These days most opinion polls are conducted using online panels, but even if you are a member of a panel, your chances of being invited to a political poll are still relatively low. Most panels have tens of thousands of people (or for the better known companies, hundreds of thousands of people) and 95% of surveys are about commercial stuff like brands, pensions, grocery shopping and so on. You could still be waiting some time to be invited to a political one.

3) But nobody I know is voting for X!

We tend to know and socialise with people who are quite like ourselves. Our social circles will tend to be people who live in the same sort of area as us, probably people who have a similar sort of social status, a similar age. You probably have a fair amount in common with your friends or they wouldn’t be your friends. Hence people we know are more likely than the average person to agree with us (and even when they don’t, they won’t necessarily tell us; not everyone relishes a political argument). On social media it’s even worse – a large number of studies have shown that we tend to follow more people we agree with, producing self-reinforcing bubbles of opinion.

During the Labour leadership contest almost every one of my friends who is a member of the Labour party was voting for Liz Kendall. Yet the reality was that they were all from a tiny minority of 4.5% – it’s just that the Labour party members I knew all happened to be Blairite professionals working in politics in central London. Luckily I had proper polling data that was genuinely reflective of the whole of the Labour party, so I knew that Jeremy Corbyn was in fact in the lead.

In contrast to the typical friendship group, opinion polls samples will be designed so that they reflect the whole population and don’t fall into those traps. They will have the correct balance of people from all across the country, will have the correct age range, will have the correct balance of social class and past vote and so on. Perhaps there are people out there who, by some freak co-incidence, have a circle of acquaintances who form a perfectly representative sample of the whole British public, but I doubt there are very many.

4) Pollsters deliberately don’t ask Labour/Conservative supporters

In so far as there is any rationale behind the belief, it’s normally based upon the perception that someone said they were going to vote for x in a poll, and weren’t asked again. As we’ve seen above, it’s a lot more likely that the reason for this is simply that it’s relatively rare to be invited to a political poll anyway. If you’ve been asked once, the chances are you’re not going to be asked again soon whatever answers you gave.

Under the British Polling Council rules polling companies are required to publish the details of their samples – who was interviewed, what the sample was weighted by and so on. These days almost every company uses some form of political sampling or weighting to ensure that the samples are politically representative. Hence in reality pollsters deliberately include a specific proportion of 2015 Labour supporters in their polls, generally the proportion who did actually vote Labour in 2015. Pollsters are required to report these figures in their tables, or to provide them on request. Hence, if you look at last weekend’s Opinium poll you’ll find that 31% of people in the poll who voted in 2015 voted Labour, the proportion that actually did, if you look at the ICM poll you’ll find that 31% of the people who voted at the last election say they voted Labour, the proportion that actually did, and so on with every other company.

5) Pollsters are biased, and fix their figures

Again, this an accusation that is as old as polling – if you don’t like the message, say the person making it is biased. It’s made easier by the fact that a lot of people working in political polling do have a background in politics, so if you want to look for someone to build a conspiracy theory upon, you don’t need to look far. Over the years I think we’ve been accused of being biased towards and against every party at one time or another – when Labour were usually ahead in the polls YouGov used to be accused of bias because Peter Kellner was President. When the Conservatives were ahead different people accused us of being biased because Stephen Shakespeare was the CEO. The reality is, of course, that polling companies are made up of lots of people with diverse political views (which is, in fact, a great benefit when writing questions – you can get the opinion of colleagues with different opinions to your own when making sure things are fair and balanced).

The idea that polling companies would bias their results to a particular party doesn’t really chime with the economics of the business or the self-interest of companies and those who run them. Because political polls are by far the most visible output of a market research company there is a common misapprehension that it brings in lots of money. It does not. It brings in very little money and is often done as a loss-leader by companies in order to advertise their wares to the commercial companies that spend serious money doing research on brand perceptions, buying decisions and other consumer surveys. Voting intention polls are one of the very few measures of opinion that get checked against reality – it is done almost entirely as a way of the company (a) getting their name known and (b) demonstrating that their samples can accurately measure public opinion and predict behaviour. Getting elections wrong, however, risks a huge financial cost to market research companies through reputational damage and, therefore, huge financial cost to those running them. It would be downright perverse to deliberately get those polls wrong.

6) Polls always get it wrong

If the idea that polling companies would ruin themselves by deliberately getting things wrong is absurd, the idea that polls can get it wrong by poor design is sadly true: polls obviously can get it wrong. Famously they did so at the 2015 general election. Some polls also got Brexit wrong, though the picture is more mixed that some seem to think (most of the campaign polls on Brexit actually showed Leave ahead). Polls tend to get it right a lot more often than not though – even in recent years, when their record is supposed to have been so bad, the polls were broadly accurate on the London mayoral election, the Scottish Parliamentary election, the Welsh Assembly election and both of the Labour party leadership elections.

Nevertheless, it is obviously true to say that polls can be wrong. So what’s the likelihood that this election will be one of those occasions? Following the errors of the 2015 general election the British Polling Council and Market Research Society set up an independent inquiry into the polling error and what caused it, under the leadership of Professor Pat Sturgis at Southampton University. The full report is here, and if you have some spare time and want to understand how polling works and what can go wrong with them it is worth putting aside some time to read it. The extremely short version is, however, that the polls in 2015 weren’t getting samples that were representative enough of the general public – people who agreed to take part in a phone poll, or join an internet panel weren’t quite normal, they were too interested in politics, too engaged, too likely to vote.

Since then polling companies have made changes to try and address that problem. Different companies have taken different approaches. The most significant though are a mix of adding new controls on samples by education and interest in politics and changes to turnout models. We obviously won’t know until the election has finished whether these have worked or not.

So in that context, how does one judge current polls? Well, there are two things worth noting. The first is that while polls have sometimes been wrong in the past, their error has not been evenly distributed. They have not been just as likely to underestimate Labour as they have been to overestimate Labour: polling error has almost always overstated Labour support. If the polls don’t get it right, then all previous experience suggests it will be because they have shown Labour support as too *high*. Theoretically polls could have tried too hard to correct the problems of 2015 and be overstating Conservative support, but given the scale of the error in 2015 and the fact that some companies have made fairly modest adjustments, that seems unlikely to be the case across the board.

Secondly is the degree of error. When polls are wrong they are only so wrong. Even those elections where the polls got it most wrong, like 1992 and 2015, their errors were nowhere near the size of the Conservative party’s current lead.

Short version is, yes, the polls could be wrong, but even the very worst polls have not been wrong enough to cancel out the size of lead that the Tories currently have and when the polls have been that wrong, it’s always been by putting Labour too high.

So, if you aren’t the sort to go in for conspiracy theories, what comfort can I offer if the polls aren’t currently showing the results you’d like them to? Well, first the polls are only ever a snapshot of current opinion. They do not predict what will happen next week or next month, so there is usually plenty of time for them to change. Secondly, for political parties polls generally contain the seeds of their salvation, dismissing them misses the chance to find out why people aren’t voting for you, what you need to change in order to win. And finally, if all else fails, remember that public opinion and polls will eventually change, they always do. Exactly twenty years ago the polls were showing an utterly dominant Labour party almost annihilating a moribund Tory party – the pendulum will likely swing given enough time, the wheel will turn, another party will be on the up, and you’ll see Conservative party supporters on social media trying to dismiss their awful polling figures using exactly the same myths.


117 Responses to “Polling myths”

1 2 3
  1. @David Colby

    Why are the French polls so accurate? What are they doing that British polling companies don’t.

  2. @DAVID COLBY

    The first round numbers were likewise exceptionally consistent for Macron, Le Pen and Fillon throughout the final weeks and over 50+ polls. The fact they were also a very good predictor of the final results means this has been skipped over but I imagine that had Le Pen slipped into third then there would have been a fairly sizeable Stewards’ Enquiry into whether the pollsters were massaging their results into line with each other to play it safe!

  3. @S Thomas

    This happened at GE15 I think it was Survation that got the result right but didn’t release the poll because it was out of kilter with other companies.

    Also I do suspect that newspapers or groups that commission polls don’t publish if the result doesn’t chime with their editorial stance.

  4. A slip back for Labour and Jeremy in these latest polls [Snip – please read the comments policy, this isn’t a place for cheering up the troops, etc – it’s for non-partisan comments – AW]

  5. @Sea Change

    “@RAF – On the EU not caring if we Remain.
    That’s clearly not the case from the quotes and actions from both EU bigwigs to their placemen in the UK.
    Brexit has even been referred to as a crime.
    It’s like “Hotel California.” Nobody was ever meant to leave. In fact, there was no legal mechanism to actually leave until the British insisted on Article 50 of the Lisbon Treaty – which Lord Kerr primarily drafted.
    Art 50 is incredibly light on detail. Kerr admits that nobody in Europe thought it was remotely likely that anyone would activate it.”

    I think you misread my post. I agreed that the EU wanted us to Remain pre-Referendum. My point is that they are not seeking for us to do so now; their primary interest being in maintaining the integrity and sustainability of the EU27.

    The purpose of Art 50 was as an expulsion clause. The UK has always had the right to leave by simply repealing the European Communities Act 1972. While the EU never wanted anyone to voluntarily leave the Union it has always been open to any country to do so under their own domestic legislation .

  6. @Sea Change

    “@RAF – On the EU not caring if we Remain.
    That’s clearly not the case from the quotes and actions from both EU bigwigs to their placemen in the UK.
    Brexit has even been referred to as a crime.
    It’s like “Hotel California.” Nobody was ever meant to leave. In fact, there was no legal mechanism to actually leave until the British insisted on Article 50 of the Lisbon Treaty – which Lord Kerr primarily drafted.
    Art 50 is incredibly light on detail. Kerr admits that nobody in Europe thought it was remotely likely that anyone would activate it.”

    I think you misread my post. I agreed that the EU wanted us to Remain pre-Referendum. My point is that they are not seeking for us to do so now; their primary interest being in maintaining the integrity and sustainability of the EU27.

    The purpose of Art 50 was as an expulsion clause. The UK has always had the right to leave by simply repealing the European Communities Act 1972. While the EU never wanted anyone to voluntarily leave the Union it has always been open to any country to do so under their own domestic legislation .

  7. COUPER2802
    I have no idea why the French polls are so consistent and seem to be so accurate, but navel gazing our own polls without examining their accuracy when compared to foreign polls is a bit like talking about how good or bad the NHS is without discussing the health systems in other countries.

  8. JIMJAM

    “she should refrain from bellicose language like Merkel does”

    I think many would find that amusing since they think Merkel was part of the clamour, and was probably directing it.

    I am happy to see what the polls have to say on the subject I suspect some will ask the question “was TM right to say what she did”. We shall see.

  9. The Tories will be very happy to see the latest YouGov with their voting projection back to 48% (+4) with all the others listed, down at least a point. This is the highest lead in YouGov’s last three polls for the Tories. Is this a sign that the EU strategy of interference in the election has back-fired, or just a sign that earlier lead reductions were just poll variations? I guess a few more polls will tell.

    The TNS poll also shows an increase in the Tory vote (+2) and a huge lead over Labour (+24).

    AW
    Many thanks for a very useful reminder to us all about polling and reading the polls.

    Ken
    There were a number of very misguided ( IMO), but amusing posts last night. I really enjoyed your response to one of them, totally agreed with it, and found it both accurate and very funny.

  10. Great (and timely!) article.

    Anyone have good source articles on “gaming” political systems?

    I ask because I have a background in finance and economics and always disturbed by people taking data and using it (or even more worrying not using it!!!) to make preditions.

    EG1: Trump won US president by “gaming” the system – he won enough marginal seats to win
    EG2: In 2015 SNP won 56 seats with 4.7% of the vote, UKIP won 1 with 12.6% of the vote – SNP were the only party standing for the “big issue” in Scotland and the opposition were split

    The Electoral Calculus website is fantastic but does assume aggregate shifts.

    If you take into consideration UKIP not standing candidates in marginal seat and LibDems insisting on standing candidates everywhere (with the comical exception of Brighton Pavilion!!!) then the “predicted” results of a model that makes a subjective “guesstimate” on the role of Brexit on voting intentions gives a much larger majority for Tories.

    Why are the LibDems continuing with a Proportional Representation (PR) tactic in a FPTP system???

    I expect 2017 GE hindsight write-ups will try to blame the pollsters but its the users of the poll information that will be to blame!!
    If LibDems only win 10-20 seats but pull Labour down to 150-160 and hand the Tories a 150-200 majority then DO NOT blame the pollsters – blame the politicians and their failure to understand the voting system in which they are standing candidates in seats they know they are very unlikely to win! Check again the SNP/UKIP outcome in terms of % and seats from 2015 GE!!!

  11. David Colby/Couper/RP – on French polls, we don’t know. If the French had a magic solution, then we’d steal/borrow it :). My understanding is that the French polls use almost identical methods to those in the UK (though French pollsters tend to weight by an urban/rural divide too, as that’s more important in French politics), so it’s probably more a question of how well French voting intentions are driven by the demographic factors that polls accurately reflect.

    Robert Newark – if ex-pats lied in surveys and consistently told us they still lived at their own postcode I suppose they could sneak into the sample, so we could accidentally be more accurate than we think we are… but I doubt it’s that common.

    Tintinhaddock – young men in manual occupations are indeed one of the trickier groups, but yes, they are in there. Picture it like this: say an internet panel is 500,000. Say young working class men make up 10% of the population. Now, if that group are really hard to reach, maybe they only make up 1% of your panel, that is, you’ve only got 5000 of them. Now think about doing an individual poll of 1000 people – you’d need 100 young working class men, but luckily you’ve got 5000 of them on your panel, so you’re fine.

    Couper2802 – “Also I do suspect that newspapers or groups that commission polls don’t publish if the result doesn’t chime with their editorial stance.” In my experience they really don’t. It may end up getting a lacklustre write up on an inside page rather than a big splash, but the attitude is more that they’ve paid for it, so they damn well want to get a story out of it. On voting intention, which this post is all about really, a lot of the polls are done on a timetable anyway so if the Guardian suddenly didn’t publish the ICM poll one month, or the Standard didn’t publish the MORI political monitor, we’d notice.

    JohnB – yep. An average GB poll is only seeking to measure the overall shares of the vote in the country. They don’t necessarily provide a good guide for support for individual regions (in fact, I’ve always advised people to ignore the regional breaks!), and they can’t really tell us much about differential swing in different parts of the country. It’s not what they are designed to do.

    northernruralmodeoman – big swings. Yes, it probably does make it harder because recalled vote will correlate less well with current vote. I think the big collapse of the LDs and rise of UKIP made 2015 harder.

  12. I think the biggest problem with opinion polls is that a percentage of people will not give a truthful answer and this is particularly true where a voting proposition is portrayed publically as not being virtuous. For example with Brexit, leave voters were (and still are to a degree) being widely portrayed as unintelligent, uneducated and xenophobic. It is not particularly surprising that a small percentage of people would not have admitted to being a leave voter if asked by opinion pollsters. Similarly voting for a left wing party is widely portrayed as voting with social conscience whereas voting Tory is portrayed as voting for self interest. Ask people in the street if they would be willing to pay more taxes for more comprehensive public services and I think quite a few people will say yes that actually would not vote that way in the polling booth. The obsession with pollsters assuming that erroneous polls must have been due to polling unrepresentative samples may actually be completely missing the primary underlying cause of the error.

  13. @MARTIN LLOYD
    I think the biggest problem with opinion polls is that a percentage of people will not give a truthful answer and this is particularly true where a voting proposition is portrayed publically as not being virtuous.

    I think the polling organisations have addressed those issues but even if they haven’t I think the same issues you raise could apply to Corbyn. He is portrayed by many as an incompetent extreme left winger with no leadership abilities, supporter of terrorists and a poor dresser!
    In your analysis are people more or less likely to tel the truth to polling organisations when asked if they would vote for him and his party?

  14. If I remember correctly the “don’t know”s are also demographically representative in the French polls (thus they can have two headline figures – in whole population, and among those who choose a party), unlike British ones.

    Wouldn’t that improve precision?

  15. Trigguy

    The answer is it depends on the vote share. Typically for a party on 30-40% of the vote 95% of the samples will be within +/- 3%

    For lower vote shares this margin reduces. around 10% the margin of error is about 1.7%

  16. TOH
    ‘Is this a sign that the EU strategy of interference in the election has back-fired’

    That’s the way it’s being spun but nobody appears to have considered that exactly the same events would probably have occurred even if there’d been no UK election. It just happens to be the stage of the Brexit negotiations that have been reached. I suspect that if an election hadn’t been called there’d be the same righteous indignation expressed by Leave supporters but they wouldn’t have been able to window dress it as ‘interfering in our election’.

  17. So with the release of the Services PMI we have a hat-trick of good economic news for theUK.

    Markit / CIPS UK Services PMI® Service sector growth accelerates in April

    Key findings:

    Sharpest rise in business activity since December 2016

    New work and employment levels expand at fastest pace so far in 2017

    Fastest increase in average prices charged since July 2008

    At 55.8 in April, up from 55.0 in March, the headline seasonally adjusted Markit/CIPS Services PMI® Business Activity Index posted above the 50.0 no-change threshold for the ninth month in a row.

    The robust and accelerated rise in services activity was linked to resilient business-to-business demand, new product launches and, in some cases, another rise in sales to overseas clients. April data signalled that new business growth gained further momentum, with the pace of expansion the strongest so far this year and the second-fastest since the summer of 2015.

  18. STEAMDRIVENANDY

    Interesting point of view. We don’t agree, lets hope there is some polling on the subject which might gove us a clue.

  19. @ Steamdriverandy

    “That’s the way it’s being spun but nobody appears to have considered that exactly the same events would probably have occurred even if there’d been no UK election. It just happens to be the stage of the Brexit negotiations that have been reached. I suspect that if an election hadn’t been called there’d be the same righteous indignation expressed by Leave supporters but they wouldn’t have been able to window dress it as ‘interfering in our election’.”

    So you don’t think Juncker is interfering by using phrases describing TM as “living on another galaxy.”? And Merkel endorsing Macron isn’t directly interfering in another country’s election?

  20. Popping out of lurking, I do every now and again…..

    NeilJ – I think though that generally the left tend to be more bold and proud than the right side of the political divide. Generalization I know.

    I went through a period of either not responding, or downright lying to surveys and pollsters after getting spat in the face during the 1992 election.

    Anyhow, the current voting intention seems to be following a narrative that labour in particular are struggling to break out of. All the elections I’ve watched since ’83 have had a narrative of some description and few seem to deviate from that.

    I have a strange feeling that this election is going be historic in one way or another.

  21. @NEILJ @MARTIN LLOYD

    I don’t think econometric attempts to “solve” the issue of intention to vote/who you would vote for solve the problem that Martin mentions.

    This is a more general issue in social psychology (behavioral economics). I’ll drop some of the jargon but it comes down to two issues:
    1/ “Priming” – if you ask someone “how they will vote” and then follow it with “do you intend to vote” you’ve “primed” them. I know pollsters are aware of this and some are using age and historic voting by demographics to use an “aggregrate” tweak and/or other ways to “solve” the problem but I’m not sure the solutions are the right ones.
    2/ “Political Correctness” – as Martin rightly states, there is a lingering perception that people voted Leave “to make themselves poorer” and are “Turkeys who voted for Christmas”. In the US this meant you might have said Clinton when asked but picked Trump on the day. In the UK for sure this will give a bias to “would you vote Remain in hindsight” and possibly overstates LibDem polling% versus outcome.

    I don’t think econometric “fixes” are the the whole answer. I think sometimes you have to step back from the data/models, consider the issues and where necessary state why you think the outcome may be different to the predictions, then with those stated adjustments conduct scenario analysis to see how those “behavioral” changes
    affect the predictions. If the differences of those changes are large or even more importantly when the outcome is close (eg US presidential election) it would be useful to add a “warning” to reading to much into polling data.
    If for no over reason pollsters at least cover their backs!

    Just my input, very happy to discuss

  22. Thanks @AW, Great post and I always try to steer people towards this during GE campaigns. It has become more difficult to champion polls since the 2015 debacle (and to a lesser extent the US errors).

    Does anyone have word about what the latest Rallings and Thrasher predictions are for tonight? I know that it is likely to be a better picture for Lab here than the GE which is worrying enough.

  23. @NEILJ
    In your analysis are people more or less likely to tel the truth to polling organisations when asked if they would vote for Corbyn and his party?

    Regardless of the generalised public perception of Corbyn as a prospective PM, I personally dont think that expressing an intention to vote Labour is being portrayed as non virtuous so I doubt that Corbyn’s leadership in itself will be a reason for erroneous polls.

    I do think that the forthcoming election is going to be extremely difficult to call percentage voting wise if not actual result wise. At individual voter level it is split between who would you prefer to run the country and how strongly do you feel about Brexit and how it should be handled perhaps relative to your normal political leanings.

    I think this unique combination of issues is leading to a far more complex position in each individual constituency than is normal.

  24. @steamdrivenandy

    I suppose it’s a question of thinking about when in the course of negotiations governments usually accuse the opposing leader of being delusional and living on another galaxy.

    Given that that doesn’t really happen, it is fair to draw inferences from the circumstances prevailing here.

    My view is that the only thing that can swing opinion to Remain leaning partners is a collapse in Leave morale and a sense of crisis. Otherwise opinion is just too glacial in its changes.

    The EU is not dumb. It knew that it had the sympathy of about 280 opposition MPs plus an unknown number of biddable Tory MPs should there come to a crunch. That is a big advantage. If it was trying to panic people a st from the Tories it was perfectly logical. It may backfire. But it makes no difference if the Tory majority is 100 or 200. But if it is 40 rather than 80 could make things awkward.

    If you look at what Varoufakis said, you wouldn’t say such machinations were beneath the EU.

  25. Merkel may or may not be behind the Clamour but she has not added to the rhetoric directly.

    Leaders should stay above the fray so they can compromise on what their underlings say earlier. Has TM decided to join the fray publicallly for narrow partisan gain at the GE rather than hold her tongue for the good of the country?

  26. @Couper2802

    “I would content that the political organisers of the parties know best.”

    I don’t know what it’s like in other constituencies or parties but round here the data we collect is pretty ropey.
    Only yesterday I canvassed a street of maybe 120 front doors. Amongst these there were 3 that we had as ‘Against’ or Tory who stated very clearly that they were Labour and always had been (and had various moans about Lab which rather attested to their honesty!). Furthermore, 2 of the 3 are actually paid-up Labour members.

  27. Bantams
    ‘So you don’t think Juncker is interfering by using phrases describing TM as “living on another galaxy.”? And Merkel endorsing Macron isn’t directly interfering in another country’s election?

    Of course Juncker’s comments can impact on our election but those same comments would probably have been made even if there was no UK election. Quite possibly they are seen by those who made them as being in the spirit of open and visible negotiations, rather than the secret and invisible ones the UK Government seems to desire.

    As for Merkel backing Macron, that’s a statement of the blooming obvious and I doubt many French voters are swayed by the German Chancellor expressing her preference for someone who’s openly expressed his admiration of her for a long time.

    TOH

    I doubt polling will give us any answers as it seems the great UK public are intent on believing every inflated and imagined insult and indignity that the Express/Mail/Telegraph can whip up.

  28. @Alan

    Thanks, quoting:

    “The answer is it depends on the vote share. Typically for a party on 30-40% of the vote 95% of the samples will be within +/- 3%

    For lower vote shares this margin reduces. around 10% the margin of error is about 1.7%”

    That sounds about right, assuming all the 1000 are going to vote, but how to the non-voters factor into this? Or are the 1000 already selected as those that say they are likely to vote?

    To put it another way, if I selected 1000 people (and selected them as a good cross-section, as described in the article), but didn’t specify in advance that they should be poeple likely to vote, then, on a turn-out of 60%, I only obtain 600 data points that are going to matter. That seriously reduces the statistics – I calculate that it would then be more like a +/- 4% margin for 95% confidence on a party obtaining about 30% of the vote. OK, not much of a difference, but a difference.

  29. ‘If you look at what Varoufakis said, you wouldn’t say such machinations were beneath the EU.’

    Ah but he has something of an extremely large axe to grind and I suspect that some of his points aren’t issues that are purposeful stumbling blocks, though it’s easy to see how a protagonist can perceive them that way, but are ‘accidental’ results of dealing with a bureaucracy through diplomacy. I get the feeling he’d just love to walk up to one person, have a one to one and sort things out directly but diplomacy doesn’t ever work like that and it’s surely ingenuous to think it would.

  30. steamdrivenandy

    You have ,i am afraid, simply closed your eyes to the facts for partisan reasons.

    1. There was a downing street dinner. It was not part of the negotiating process and private dinners are not usually transparent. The British side kept it confidential even when it became the subject of a detailed eU leak. Almost line by line. Diplomacy suggests that both sides should be in agreement that the terms of a meeting be published. They were not;
    2. not content with that the President of the EU called the British Prime Minister ” delusional” and “living in a different galaxy”Note the British papers and TM are not aaking this up.
    3. Junckers reports to his boss Mrs Merkel who repeats the delusion jibe
    4. Barnier then warns TM about slipping on loose rocks while hill walking and to be careful about her health which could by some be seen to be both threatening and insulting bearing in mind her diabetes

    5. All done in the context of a General election.I accept that JUNckers has little experience of elections but calling the head of one of the parties “deluded” might even have dawned on him to be unwise at this particular time.

    Your analysis that imaginary insults have been concocted by a right wing press seems a little hard to justify given the above. Sometimes one must move from “My EU right or wrong” and be balanced about these things

  31. Lazlo

    “If I remember correctly the “don’t know”s are also demographically representative in the French polls (thus they can have two headline figures – in whole population, and among those who choose a party), unlike British ones. Wouldn’t that improve precision?”

    ‘Dont know’ (and for that matter ‘Do Not Vote’) really do seem to be self explanatory categories…

    Any party party pursuing such categories as an alternative to trying to convince existing actual voters of other parties to vote for them will always fail spectacularly. As we are seeing.

    On France message is clear: it takes a centrist to defeat the hard right (who are often supported- when push comes to shove- by the hard left).

  32. Does anyone know status on which seats UKIP will (and will not) be contesting?

    If you make some assumptions on where those voters will switch to then makes a considerable difference to predicted majorities:
    FWIW I’m using various assumptions with “mid” of:
    “Tory” areas: 70% Tory, 20% Abstain, 10% Labour
    “Labour” areas: 30% Tory, 30% Labour, 60% Abstain

    As you’d expect the Tories will be by far the largest winner of UKIP not standing but it’s obviously a seat-by-seat analysis.

    Also means that even ignoring tactical voting where a choice is offered a poll score of 8% will provide a much lower % on the day with lots of seats not even contested… less than 5% I suspect!!

  33. S THOMAS

    A clinical analysis of the facts. Excellent.

  34. AW very informative, but:
    “elderly people are less likely to use the internet, but enough elderly people do that you can deliberately recruit them and invite them to surveys in the correct proportion.”
    It seems to me obvious that those elderly internet users you recruit are almost by definition not properly representative of the elderly as a whole.
    You get away with that adjustment because the elderly are only a fraction of the total sample, and your ‘non-representative’ subsample will have more in common with the elderly as a whole than with say the under 25s. But how big is your subsample?

    Consider asking the question “Do you think the internet influences voting choices?” How can your subsample possibly properly represent the views of the elderly in general, who never use it? The latter may give you an opinion, if you could ask them, but it could hardly be an informed opinion.

    One serious variation with age is the “correlation” with education, or with educational qualifications. My BSc degree put me into a selection of about 5% of the population. Now we are looking at degree courses for half the population being educated. Does your sample selection deal with these kinds of differences? I have an older friend with a degree who won’t touch the internet with a barge pole.

    @SD ANDY & Bantams “‘So you don’t think Juncker is interfering by using phrases describing TM as “living on another galaxy.”?”
    Interfering, certainly, but does he know which way?

  35. Anthony,

    Given that we did seem to see a fairly marked difference between Urban and Rural over Brexit is there are argument for trying it here or, as I suspect, have you looked at it internally and found little impact on the result.

    I ask because one of the things that came out clearly from the Indyref was that, quite understandably, English people were more likely to vote remain and perfectly free to do so.

    As they only made up about 10% of the electorate it was difficult to see how it could be decisive in any way, although somewhat embarrassingly a number of my party colleagues decided to grind that axe.

    What did get me thinking was when I looked at the distribution and found that in some areas particularly rural ones, English people made up close to 1 in 4 voters.

    If there was a genuine difference in how they voted then although it might not impact a referendum, it could well start to influence individual constituency results.

    If for example people from England are more likely to vote Tory over Brexit and make up 1 in 5 of a rural constituency then it might decide who wins that seat.

    I’d stress that as far as I am concerned as many English people as want can live in Scotland, move here, live where they like and vote n every election.

    My interest is in whether you can use rural/urban or nationality to better predict the results in individual seats?

    Peter.

  36. I seem to remeber the internet polls were closer to the final result of the 2015 election than the telephoning polls. Is that correct or is my memory playing tricks on me.

  37. @AW

    Thank you for a useful post that reminds us all of some of the basic principles in polling and the need to be wary of the influence of our own wishes etc on how we interpret them.

    Tintinhaddock – young men in manual occupations are indeed one of the trickier groups, but yes, they are in there. Picture it like this: say an internet panel is 500,000. Say young working class men make up 10% of the population. Now, if that group are really hard to reach, maybe they only make up 1% of your panel, that is, you’ve only got 5000 of them. Now think about doing an individual poll of 1000 people – you’d need 100 young working class men, but luckily you’ve got 5000 of them on your panel, so you’re fine.

    However it is with just such groups that the possibility of sampling error can occur, as even with 5000 there is a chance that that group is unrepresentative especially as in this case they can be self-selective. To illustrate my point I, two aspects of my life brings me into contact with this demographic – activists within the Labour party and people I know from going to football. Of the former (roughly 20 souls) they are all actively engaged in politics on line, are pro-Corbyn and are more likely to be on a panel. Of the latter group (I’m, on talking terms with roughly the same number), they are all anti-Corbyn and have a fairly dismal view of Labour in general and whilst being on line I can guarantee you it is not for the purpose of engaging in political polling. Obviously weighting attempts to adjust for this, but is intrinsically difficult to do this accurately when you have a demographic were is is hard to get a representative sample in the first place.

    Party strategist use polls/focus groups particularly in marginals and of certain sections of the electorate not only to find out whats going on but to identify and test certain narrative’s that benefit them.

  38. Dave

    Depends whether he said it before or after breakfast.

  39. STEAMDRIVENANDY

    As I say I look forward to polling on the subject and the answers will be interesting regardless of what you and I might think since they will indicate what the voters as a whole think.

  40. “Quite possibly they are seen by those who made them as being in the spirit of open and visible negotiations…”

    This is what I say when I leak private meetings.

  41. @Cooper2802

    This happened at GE15 I think it was Survation that got the result right but didn’t release the poll because it was out of kilter with other companies.

    From memory it was out of kilter with Survation’s other polls too.

    I don’t think this poll was considered correct but was pulled for nefarious reasons.

    I think it was was pulled as it was way out line with all the other polls. I think considering it a strange outlier was probably correct given the circumstances of the time. The fact that it quite close to the result wa my view nothing more than pure luck.

  42. Rob

    My point was simpler, and certainly nothing to do with party politics.

    You have Party X, Party Y, Party Z, and Don’t know. If the Don’t Knows are representative of the population, you can treat them as if they were a separate block of voters, which is not the case in British polling (as they are not representative). This is how the Dutch, and the Germans do it, and if I remember correctly, the French too.

    I was wondering if it increased the precision of the polls.

  43. @ PETER (SNP)

    No input on the rural/urban issue but can I ask your views on Scotland as I don’t really follow what goes in Scotland:

    1/ Is the GE a proxy for IndyRef2 – high/low factor
    2/ Does desired timing of IndyRef2 have a factor – SNP want it sooner, other parties later/never
    3/ Is Brexit even relevant? (no disrespect but with Article50 triggered the only voters that would have Brexit as a “switch” issue would be Yes/Leave (guess that is mostly SNP) or No/Remain (you’d think that would be LibDem but they are polling badly so is “No” more important than “Remain” in regards to the GE)

    Appreciate your thoughts.

  44. ROB SHEFFIELD

    @”it takes a centrist to defeat the hard right (who are often supported- when push comes to shove- by the hard left).”

    I watched the debate ( a riveting boxing match with no holds barred).

    Her Industrial & Economic Policy is pure Tony Benn Corbyn could have uttered every word.

  45. @CMJ

    But it is a perfect example of herding and I bet to this day Survation are kicking themselves.

  46. Redrich. This was exactly my point. The guys I was talking about were not actually young – between about 50 and 60 and they don’t use the internet much or even at all – and certainly would never visit a polling site. Guys of the same age and class who do use the internet and polling sites are a different ‘type’ altogether.

  47. Neilj – nope, was much of a muchness. Taking the average of final phone polls and average of final online polls the figures were virtually identical in 2015. It was the EU ref where the online polls were generally better.

  48. AW – Thanks, must have been the E.U polls I was thinking of

  49. @Rob Sheffield
    Your assertion is pure polemic. The polling evidence doesn’t back you up. For example:
    (1) the polls indicated that the hard-left candidate Mélenchon would have beaten Le Pen if he, instead of Macron, had gone through to the second round;
    (2) the polls for the second round indicate that Le Pen is likely to gain a much larger proportion of centre-right (Fillon) voters than of hard-left (Mélenchon) voters;
    (3) iirc, they also indicate Macron winning at least as many Mélenchon voters as Fillon voters.

    Further, the nature of Le Pen’s economic policy is beyond the scope of a ukpr discussion, but to analyse her economic policy as leftwing is considered in France to be extremely contentious.

  50. YouGov detail up on their website now. Some interesting pointers in the detail:-
    Britain leaving the EU is still seen as the most important issue, NHS is now second followed by immigration and the economy.

    On who is best at dealing with specific problems Labour lead on housing and the NHS. The Tories lead on the rest, and strongly so on the economy and negotiating our exit from the EU.

    On the hindsight question about the referendum Leave now leads by 3% Leaving at 46%. In the last YouGov poll Remain had a 2% lead so this is a large switch.

    The detail questions on the effects of leaving continue to be negative as they always have been.

    Thinking about the general election campaigns and promises from the main parties, do you think the following parties are being generally positive or negative?
    Conservatives 42% rating +11
    Labour 30% rating -10

    Taking everything into account, how good or bad an election campaign do you think the following have had so far?
    Conservatives 44% rating +17
    Labour 22% rating -28
    LibDem 21& rating -17
    UKIP 10% rating -36

    I suspect that the Tories will be happy with the campaign so far. For Labour the NHS and Housing are the two areas they should be pushing hard. Not good news for LD’s and a disaster for UKIP.

1 2 3