Media commentators seem to have caught up with the fact that UKIP are challenging for third place in the polls… except that it isn’t quite a fact. As ever, the actual picture is more complicated and whether UKIP are competing for third place, or have secured third place, or are trailing in a very distant fourth place depends on the polling company and their methodology.

There are currently nine polling companies conducting regular or semi-regular voting intention polls, four of them have shown UKIP ahead of the Lib Dems – in some cases like Survation almost always, in some cases like YouGov very rarely. The other five have never done so, and in the case of ICM or ComRes’s phone polls have never shown UKIP even vaguely close to third place.

The perception that they are consistently competing for third place comes from a couple of factors. One is publication bias – polling companies that show UKIP in third or challenging for third tend to include them in their topline figures. Polling companies that have them in a distant fourth don’t really bother. This is understandable enough, but means the good figures get far more prominence than the bad ones (it is further exacerbated by the reporting of polls – UKIP in third place is a news story worth reporting. UKIP back in the crowd with 3% or 4%, as ICM, MORI and ComRes have sometimes shown them in recent months isn’t). The other factor is that UKIP tend to do better in online polls, and online polls tend to be more frequent. Between the start of June and the 25th October when I collected up the data for this post there were 153 voting intention polls, 135 were online. Even putting aside YouGov, who conducted over two-thirds of all the polls in that period, there were 33 online polls compared to 18 phone polls. The polls that are best for UKIP are also the most frequent.

The graph below shows the average score for the Liberal Democrats and UKIP for each polling company between June and October, ordered from those showing the biggest Lib Dem leads over UKIP on the left, to those showing the biggest UKIP lead over the Lib Dems on the right.

At the left ICM and ComRes (phone) are showing the Lib Dem very solidly in third place, with UKIP not far above their 2010 level of support. Populus and MORI have a big gap between the two parties. TNS and YouGov both, on average, have the Lib Dems in third place, but have both on occasion shown UKIP in third place or equal with the Lib Dems. Only Opinion and Survation give UKIP a higher average level of support than the Lib Dems, and only Survation significantly so.

As usual there are different reasons for this pattern. For example, Survation give UKIP by far the highest level of support because they prompt from them in their main question (something I’ve written about at more length here). ComRes’s phone polls probably show the lowest level of UKIP support because they use a harsher likelihood to vote filter for minor parties than for the main three (their topline figures include those who say they are 5/10 or more likely to vote for Con, Lab and LD, but only those 10/10 certain to vote for minor parties).

The main cause of the difference however, as I’ve marked on the chart, seems to be between online and telephone polling. For whatever reason, online polls seem to show a higher level of support for UKIP than telephone polls do. This doesn’t seem to be a “minor party” factor, if you compare “others” in general online polls show more support for minor parties, but take out UKIP from the equation and it vanishes, as I’ve shown in the graph below. For some reason, the consistent difference only shows up with UKIP.

There are various possible explanations for this difference. One is interviewer effect, perhaps people are more willing to admit to an anonymous web-interface that they are going to vote for UKIP than they would be to a live human being on the end of a telephone. The other possibility is that there some sort of sampling difference in one mode or the other. Are the sort of people who would support UKIP systemically less likely to agree to take part in a cold-called telephone interview than other people with the same demographic profile? Or are the sort of people who would support UKIP more likely to join an internet panel than other people with the same demographic profile? We cannot tell.

What do we have to judge by? At the last election all the main pollsters got close to UKIP’s actual level of support, so that doesn’t give us any real clue. Since UKIP’s support started to really increase in late 2011 we have had very few big electoral tests to check by, and very few where both telephone and online pollsters have conducted comparable polling.

There were two polls for the Feltham & Heston by-election, one online one from Survation and one commissioned by Lord Ashcroft and presumably carried out by Populus over the phone. Populus showed UKIP at 5%, Survation showed them on 7%. UKIP ended up getting 5.5%, so hardly a conclusive answer (UPDATE – plus, as I should have spotted, the Survation poll was telephone anyway, so the comparison couldn’t have told us anything about online vs phone polls anyway!)

In the London elections this year all pollsters overestimated support for UKIP slightly, but the only telephone poll (by Populus) showed UKIP candidate Lawrence Webb on 3% the same as online polls by ComRes and Opinium. Besides, by describing themselves as “Fresh Choice for London” UKIP did not put their party name on the ballot paper, which may have depressed their actual vote – we don’t know if pollsters slightly overestimated their support, or if UKIP underperformed by failing to mention UKIP in their party description on ballot papers.

Even the worse polling for UKIP shows them increasing their support from the last election, mostly quite significantly, so I think we can be confident they have gained support since 2010. Just how much we can’t really tell – the pollsters range from the relatively modest all the way up to trebling their support or more.

199 Responses to “How much support do UKIP have?”

1 2 3 4
  1. colin

    We seem to be on the same side on two consecutive subjects!

    Now we just need to agree about publicly owned utilities…

  2. Alex

    I think that guy at comet did do his job, he identified a cash surplus that could be extracted if the company went into administration. His job is not to keep the company going but to make money quickly for the owners. I’m guessing that he has achieved this. But your wider point that executives often seem to get paid for failure again and again I would broadly agree with.

  3. The Sheep

    Thanks-still don’t quite get it.

    But what the hell-they wind up with a President.

  4. NICK

    Fear not-normal service will be resumed at some point.

    May making a statement to HoC at 12.30

  5. @nickp


    We seem to be on the same side on two consecutive subjects!

    Now we just need to agree about publicly owned utilities…”

    UKPR’s coalition in the making ! :)

    Colin may disagree on utilities in regard to all types. But may be open to persuasion in regard to some. e.g water. I cannot see how it is sensible for UK water companies to be foreign owned and for it to be so fractured. You end up with massive differences in pricing, overflowing with water in some areas, drought in others and not being able to shop around in every region. If I wanted to change my water/sewerage supplier I cannot do so, like gas or electric. Therefore you might as well just have a nationalised company. The trouble is that it is now too expensive to nationalise.

  6. US Election
    Fox news and their ‘fair and balanced’ view of the polls..

    Watch the first 2 mins – it is everything AW warns us not to do – cherry pick.

    Fox news could be a comedy channel they are totally deluded and irrational

  7. @ Alec

    I agree with your analysis about triple dip risk, although an official triple dip can only happen in April (two consecutive quarters of negative growth). I think it was Colin who also pointed out from the last official growth figures that there appeared to be a downward trend in September with most of the growth coming in July although I was not that good at understanding the stats on a month to month basis (I guess a small drop from the high July figure because fo the extra day doesn’t equate to going backwards).

    I remain convinced that the economy will not pick up before the next election. To be fair a lot of it is out of the government’s hands but doesn’t stop me thinking Labour landslide is on the cards which is more or less what Mervyn King predicted for whatever government won the last election.

    In all honesty I can’t see the road plan for our economy improving ever without some form of protectionism from globilisation whether it be higher duty on imports from countries who can make goods/provide services at a tenth of the price of the UK or simply tax rules that ensure that companies that make a profit in britain get taxed on those profits rather than the money ending up offshore.

  8. Colin,

    This is probably the most straightforward explanation of how it all works:

    h ttp://

  9. Couper2802,

    I will be watching Fox News from midnight tonight. The most entertaining TV of every 4 year cycle. When those guys start frothing at the mouth it’s better than any stand up comedian.

  10. Colin, hopefully this explanation will be a bit less confusing.

    There are 538 Electoral College voters. They are allocated to the states in the same proportion as congress members (2 Senators plus a number of Representatives based on population), with 3 going to DC, which is not a State. So the state delegations to the Electoral College vary from 3 to 55. 269 would be a tie, so 270 is needed to win. EC members are loyal party members chosen to support the candidates*, but they don’t appear on the ballot papers as candidates themselves.

    Each state has it’s own rules for elections, even Federal ones, but in each case the Presidential Candidates will be named on the ballot paper. Some states will allow more minor candidates than others. Some will also count ‘write-ins’.

    All but two** states have a winner-takes-all approach to the election. This means that the larger close run states can have a big effect in a tight election, and that’s why everyone making predictions uses the Electoral College.

    * Despite this, it’s not unusual for one or two EC members to become ‘faithless’ and vote in a different way. Usually it’s a protest vote and would not affect the result, but there is potential for an EC member to cause an upset. There seems to be no legal force to make them vote in accordance with their electorate.

    ** Maine and Nebraska are using a slightly more proportional system. The winner of the overall vote gets 2 (analogous to the Senate delegation), and in each House district there will be EC vote won on the basis of FPTP. However, Maine is very likely to go all Dem and Nebraska all Rep, with perhaps one district in each state that could go the other way.

  11. Ms May on election turnouts: “I never set a turnout threshold for any election and I’m not going to do it now. What I do know is the people who are elected as police and crime commissioners will have something that the current police authorities do not have. For the first time ever, they will have a democratic mandate for the people for the work they’re doing.”

    The Unions will be pleased that the Government will accept any majority strike ballot no matter what the turnout.

  12. R Huckle,

    ‘The trouble is that it is now too expensive to nationalise.’

    Why? How much does an Act of Parliament cost?

    Let’s have some positive thinking: it is too costly for the general population to leave it as a private monopoly

  13. MikeMS

    If you are suggesting a simple act of parliament could be used to transfer a companies status to public ownership with zero (or nominal) compensation for investors in said company that is taking a leaf out of Fidel Castro’s (or more recently de Kirchner’s) book.

    I think once a country starts stealing private assets “for the good of it’s people” it’s the fastest route to getting companies to pull out of investing in any long term projects.

    I wonder how this would affect polling if that became One Nation Labour’s first policy? I suspect we’ll never find out!

  14. R huckle

    It’s only expensive to nationalize if the govt pays compensation, would cost nothing to enact a bill allowing utilities to be nationalized without compensation following a referendum of their customers, that would put some free market fear into those companies

  15. R Huckle

    UKPR’s coalition in the making!

    Well, they’ve managed it in the Netherlands (between right wing ‘Tory’ businessman’s party VVD and Labour party PvdA).

    Imagine John Redwood and Ed Balls in the same Cabinet? :-)

  16. Here’s the gospel according to St Nate of Silver:

    Obama’s base of ECVs that he will win with 100% probability is 217.

    He will also almost certainly win the following states with slightly reduced probability:

    Pennsylvania (20 ECVs) – 99%
    Wisconsin (10) – 97%
    Nevada (6) – 94%
    Ohio (18) – 92%

    That makes 271 and the White House. But we’re not finished yet.

    Iowa (6) – 86%
    New Hampshire (4) – 86%
    Colorado (9) – 82%
    Virginia (13) – 81%

    Now we’re on 303 ECVs, and that’s where I’ve put my money. However, I am prepared to be out of pocket, because:

    Florida (29) – 53%

    could well go blue if voter suppression is beaten, and:

    North Carolina (15) – 28%

    if some figures I have seen regarding early voting are correct could boost Obama to 347.

    After that we are in fantasy land – there are only two other states that the 538 forecast gives as less than 100% for Romney:

    Arizona (11) – 2%
    Montana (3) – 2%

    And there is also a very small chance that one of the congressional districts in Nebraska could go to Obama, so theoretically the President could get up to 361 or 362, but overwhelmingly the three most likely totals are 332, 303 and 347 in that order.

    Speaking of fantasy land, that is the location of anyone who still thinks that Romney can possibly win.

  17. Of course it would be necessary to leave the EU and I think the ECHR as well if you wanted to renationalize even if you were paying compensation, I believe. Martyn would know the relevant treaty articles, but the Lisbon treaty was a thatchrite wet dream as regards privatization and making the reversals of privatization extremely difficult.



  19. I’m not sure if our membership of the WTO would also create complications with nationalizing utilities? Anyone know? Might be necessary to leave that as well

  20. Mitz, you are in fantasy-land. There is no way that Obama is favourite in Florida, has no (none absolutely none) chance in NC, and Virginia and particularly Colorado are coin flips. Ohio is maybe 75-25 for Obama in terms of probability but not over 10-1

  21. Mitz

    You start talking about Arizona and Montana as “unlikely” and rule out Mitt taking everything up to Ohio as “fantasy land”.

    I think it’s very unlikely that Mitt will win but a definite (and unpleasant) possibility.

    The percentages you give are not independent (or likely to be the real numbers, just one persons estimate), the chance that Mitt will win Virginia say, in the scenario that he wins Ohio is probably more than 50% as if he’s won Ohio he’s pretty much going to beat the pollsters estimates of his vote across the board.

    To sum up, watch Ohio, if Obama takes that, as expected, you can call it an early night without sitting up all night glued to the screen. If he loses Ohio, I can imagine it being a rather tense night for everyone.

    Of course there is the possibility of the result hinging on Ohio and regardless of how it went the decision being made in a courtroom at a later date.

  22. R HUCKLE

    As a general principle , I am not in favour of the Government running companies .

    The other side of that coin though, is effective regulation & monitoring of appropriate factors by the State .

    Also , in some instances it is right for the State to fund/partly the service ( Health for example) -but that doesn’t mean the State should provide it.

    Regarding Water-you identify the problem-the absence of a common supply resource across the country.
    Worse-a glut in some areas & shortage in others.

    There is a clear need for a National Ring Main ( I presume basically from North to South) . It should not be beyond the wit of politicians to devise a plan which implements this with the involvement of funding from the State & the Private Companies. It would presumably be a substantial construction project & so helpful to the economy in general.

    It would then remain to decide how to structure the ownership of water supply & retail distribution. The models used in Rail & Electricity no doubt offer lessons in both good structures & bad !

  23. Also, are you claiming there is a 0% probability that NH will go red? It voted for GWB, and Romney is a much more appealing candidate to new england moderates.

  24. Is it just me who thinks Romney will be President elect this time tomorrow?

    See a very good Guardian article on how this is possible.

    The Bradley effect worries me.

    This isn’t the basis for my belief I should add. This is how I see it going. Obama 259 Romney 279.

    Romney will carry Florida, Colorado, Ohio, North Carolina and New Hampshire.

    Obama will carry Nevada, Iowa, Wisconsin, Pennsylvania and Michigan.

    I think some people are ignoring (at their peril) the large and consistent lead that Romney holds with independents. Obama also has leads in early voting in most battleground states but the margin of the lead is less than 2008 with the difference being greater than his overall margin of victory. This will be enough to push Mitt Romney over the top.

  25. @ RiN

    There’s no rule in the Lisbon treaty against nationalisation (there has been some in a number of countries since 2010) providing that ythe country concerned follows the competition rules as prescribed by the Commission. Nationalising without compensation is tricky, because the government can be taken to court. If the government really wanted to do it on the cheap, it would have to impose various regulations (including windfall taxes) and hammer the shares in this way (it could trigger a Commission procedure, but that lasts for ages) until it costs very little to buy it. There would be some side effects though.

    It would have been done easier had the banks been properly and lawfully nationalised, but it’s too late.

    I don’t advocate nationalisation, just showing some ways to do it :-)

  26. Laszlo

    Thanks, I’m just trying to point out that a more left wing labour party would have to be more euro sceptic

  27. Alan,

    I didn’t describe Obama’s chances in Arizona and Montana as ‘unlikely’ – I merely quoted the 538 forecast’s estimate of his chances in those two states: 2%. If you want me to put that into words, I’d go with ‘vanishingly unlikely’.

    The percentages that I give are indeed independent, as they are arrived at by careful analysis of all state polls. Nate Silver has a highly sophisticated (and IMHO reliable) model that has an extremely good track record.

    I agree with you: if Romney takes Ohio then there is a very strong likelihood that all of the polls upon which the data relies are wrong and that by then Romney will already have taken Virginia. But again, the likelihood that an aggregation of all these polls is wrong is very small. About 8%, in fact.

    Again, we agree: it all comes down to Ohio. If Romney fails there, the game is over. If Romney wins Ohio, then it probably means that he has a better chance than the polls suggest in other swing states too, but even then Obama will remain the favourite.

  28. is there going to be a thread on here so we can discuss the US election?

  29. Joe R,

    On what basis do you say that there is ‘no way that Obama is favourite in Florida’? Every recent poll shows that this state is the closest of them all, and several show Obama with a slight lead. Compared to three weeks ago when Romney was the clear favourite there in almost all polls. Early voting in the state suggests that Obama has a better GOTV machine. I in no way implied that I think Obama will definitely win Florida, but there is at least an even chance and 538 puts Obama at 53%.

    You also claim that Obama has absolutely no chance in North Carolina. Really? Well, 538 shows that his percentage chance is small, about 1 in 4. But some early voting statistics suggest that this is wrong: 62.9% of the turnout from 2008 have already voted (as of yesterday) and amongst them Obama has a 16.3% lead. That means Romney will have to outdo Obama at the polls today approximately 3 to 1. This data and much more besides here:

    h ttp://

    You say Virginia and Colorado are ‘coin flips’ – you’re entitled to your opinion, but I would suggest they are both more likely to go Obama than Romney and that your coin is not a fair one.

    Finally, I did not say there is a 0% chance of Romney winning New Hampshire – I quoted Nate Silver’s analysis: he give’s Romney a 14% chance there.

  30. On the US Elections here’s a good take:

    This time tomorrow we’ll know!

  31. KeithP

    The sun has reported today that Yougov are predicting an Obama victory already so something may be posted later today on that subject.

  32. Indeed NickP. Both Sam Wang and Dean Linzer are even more bullish about the probability of an Obama victory than Nate Silver is.

  33. If polls have any validity at all, Obama is going to win.

  34. I wish the BBC would stop saying the “closest race for decades” as if 2000 never happened. It’s unlikely this one is going to be anywhere near that close.

    On a sober note of course, the venerable Nate Silver still gives Romney a 9% chance of winning – small, but not impossible odds.

    And talking of Silver, I strongly recommend his book The Signal and The Noise…

  35. Mitz

    Let me clarify, when I said independent I meant “independent of each other”, not that the analysis had any bias.

    If Romney takes states like NH and Virginia, I’d say that Ohio will be much closer than predicted, to put it another way, I’d be happy to get 12.5:1 on Romney taking Ohio if NH and Virginia both fell to him as I believe his chances in Ohio would have to be significantly reassessed due to those events.

    It’s not only the data which is wrong (due to insufficient sample sizes leading to a large enough MOE) it’s how the pollsters handle the data which could be systematically wrong.

    I guess I had issue with you calling 8% “fantasy land”. I’d probably think that that 8% was a floor threshold and depending on how large a “Bradley Effect” we see it could be significantly higher (it might be zero but it’ll never favour Obama and could well produce the sort of systematic error that would be needed)

    If Romney wins Ohio (and very likely to have taken Florida, NH and Virginia as well), would you still think Obama would win Colorado comfortably? I’d think Romney was perhaps a slight favourite in those circumstances and it might not even be enough for Obama.

    If the Pollsters are wrong and Romney takes Ohio, It could well be (and quite likely) he’l take every state up to and including that (barring some serious ununiform swings)

    I think Obama probably is a strong favourite but it’s not unthinkable that Romney could get a surprise victory. The pollsters would only have to be out by 2% or so and suddenly things get very tight.

  36. Alan,

    You make an important point about individual state results not actually influencing each other per se, but nevertheless being reflections of the over all mood and so likely to be in line with each other, at least up to a point. But I do think that it is slightly more complicated than that.

    Florida and Ohio, above all the other states, have a growing notoriety for voter suppression, court actions, ‘lost’ voter registrations etc. If we wake up tomorrow morning still not knowing who has won it will be because of a contested result on one or both of them. If Romney takes both, then that is likely to indicate an overall bias in the polls making many of the other key states tighter, but I would maintain that Obama would still be enough of a favourite to carry VA, WI, NH, NV, IA, CO etc. and fall over the 270 line.

  37. yes, this was never going be a landslide. If Obama gets much more than 300, that would be eyebrow raising. As it would be if Romney won by more than a sliver.

  38. @ Nick P

    I’m still waiting for the Cameron ‘internal’ inquiry results of the Cruddas ‘dinnergate’ scandal which revolved around the luring of huge donations out of business people, in order for them to gain inner access to the top for influence over policy.

    What happened to all that?

  39. I think we know what internal inquiries are worth.

  40. this telegraph article says ohio may be closer than obama supporters think.

  41. paul

    So Romney expects Obama’s supporters to stay away from the polls?

  42. Paul,

    Note how pretty much the entire language of the article is people giving their opinions based on feelings. The only ‘facts’ are centred around a few Republican internal polling stats (that are all massive outliers from the rest of polling, even that which has leant Republican all along). The Democrat campaign organiser who rates his confidence as “six and a half out of ten, no better” is basically saying to supporters of Obama “get out and vote otherwise this could slip away”. The Dem’s worst enemy is complacency.

    Oh, and a tight race will always make a more exciting story than a shoo-in, n’est-ce pas?

  43. I’m hoping for a romney win. That would give the best chance of a decent democrat president in 2016. Whoever wins this one will lose the next one without a shadow of a doubt

  44. Regarding US Election. I think people are forgetting about momentum.

    The momentum is with Obama by the movement in the polling averages over the last few weeks. In fact if you look at rcp (which is republican leaning) it is almost a case in point on how to win a tight election. Obama is moving ahead in the national and state polls at just the right time.

    The republicans are relying on systematic bias in all the polls and that is a long shot.

  45. RiN,

    Can’t possibly let that go unchallenged!

    Victory tonight will present the winner with much less of a poisoned chalice than that which was given to Obama in 2008. And yet despite everything he seems likely to get a second term.

    If the GOP loses, they will rip themselves to shreds. I see no reason to believe that they will be able to come up with a credible candidate in 4 year’s time.

    Win or lose tonight, whoever the Democrat candidate is in 2016, they will have every chance of prevailing.

  46. RiN

    We have a decent democratic president now. In fact, when history looks back on his TWO terms he will be judged one of the best. IMO

  47. Michael @ 12:03,

    Who is going to take Virginia? In your numbers, you seem to have given it to Romney, but if it goes to Obama the final ECV score will be O: 272 – R: 266.

  48. Mitz,

    Please accept my apologies. I have given Virginia to Romney but missed it out in my prediction list :) .

  49. Happy days! See you on the other thread.

1 2 3 4