Remain to win by 8.5% – plus or minus 6

This blog has been updated to include the latest poll by Populus


Take your pick. If the telephone polls are right, then Remain is heading for victory, possibly by a comfortable margin. If the online polls are right, then the result could be extremely close, and we may not know the outcome until breakfast time tomorrow.

Here are the final polls, excluding don’t knows:


ORB/Telegraph: Remain 54%, Leave 46%

Survation/IG index: 51-49%

ComRes:  54-46%

Ipsos-Mori:  52-48%




Opinium: Remain 49%, Leave 51%

TNS: 49%-51%

YouGov:  51-49%

Populus: 55-45%


Those figures incorporate the chosen turnout filter for each company. This sometimes makes a big difference. Here are two sets of figures from TNS and ORB; I have underlined the ones they chose to make their headline projection, on which they wish their accuracy to be judged

  • ORB: All voters: Remain 51%, Leave 49%, certain to vote: 54-46%
  • TNS All voters: Remain 49%, Leave 51%, certain to vote, 46-54%
  • In other words, there is little difference between TNS and ORB’s initial figures, when they count every respondent who takes sides
  • But if we count only those who are certain to vote, then a chasm opens up, with ORB reporting an 8% Remain lead, and TNS an 8% Leave lead.

Confused? You should be. Online and telephone polls have mostly been telling different stories; moreover, the polls can’t agree on whether, or how, to filter their figures to allow for differential turnout.

Indeed, TNS has muddied the waters even more by changing the basis of its headline figure. Last week, it stressed an 8% leave lead, based on those it regarded as likely voters. This week, it looks at first sight as if the Leave lead has collapsed to 2%. Not so. The Remain-Leave gap is precisely the same as last week, according to both ways of calculating the numbers; but this time, TNS has preferred the measure that points to a close outcome rather than a big Leave victory.

To predict the outcome, then, we must do two things: assess the state of play before the start of voting, and judge whether there will be an on-the-day shift. Neither can be done with absolute certainty. Here, with no money-back guarantee, is how I see it.

The simplest way to estimate the eve-of-referendum vote shares is to average the figures from the eight polls listed above. This poll-of-polls gives us:

Remain 52%  Leave 48%

Such a calculation, however, offers spurious precision. As I think the online surveys may be overstating support for Brexit, I reckon that the likely Remain vote ahead of today’s vote was 51-55%, with 45-49% plumping for Leave.

However, if the outcome is very close, then two groups of voters not covered by the polls might tip the balance: the 23,000 Gibraltarians with the right to vote, and possibly 200-300,000 expatriate voters living abroad. (Sadly we shall not know afterwards what the true number is, for these votes will simply be incorporated into the counts at the local authorities where the expatriates previously lived.) Both groups are likely to vote mainly Remain. This could add 0.2-0.3% to Remain’s percentage and so widen Remain’s lead, or narrow Leave’s lead, by around 0.5 percentage points

Now to the possibility of an on-the-day shift. YouGov’s on-the-day Scottish referendum poll, it became clear that more voters were making a last-minute switch from pro- to anti-independence, and that the anti-independence supporters were slightly more likely to vote at all. Together these factors moved the dial two percentage points, from a 52-48% lead for Better Together in YouGov’s previous survey, to 54-46% on the day – close to the 55-45% result.

I believe it’s likely, though not certain, that there will be a similar on-the-day shift today to the status quo. Answering a pollster can be done cost-free. Casting a vote is a decision with consequences. We know that some voters are torn between heart and head: the emotional pull of Brexit versus the worries of what might happen to jobs and prices. Do some people respond to pollsters with their heart, and then vote with their heads?

My judgement is that, if there is an on-the-day effect, it will help Remain rather than leave; so the overall Remain share will be 0-2 points higher than it was yesterday.

Let’s assume the polls haven’t screwed up completely, and the true eve-of-referendum position, including Gibraltar and expatriate voters, was Remain 51.2-55.3%, Leave 44.7-48.8%. Adding in on-the day effects that hover between neutral and a 2 point lift for Remain, the final UK result should be somewhere in the range of Remain 51.2-57.3%, Leave 42.7-48.8%

This gives us  a mid-point prediction of an 8.5% lead for remain, or a majority of around 2.5 million of votes cast. But don’t be surprised if the gap is less than one million – or as much as four million.  And if the phone polls have been systematically overstating support for Remain throughout the campaign, then a victory for Brexit is perfectly possible.

My apologies if that is not precise enough for you. If you need a more exact forecast, I suggest you toss a coin or ask an astrologer.

Has there been an anti-Brexit surge?

The headline on my blog three weeks ago applies once more: “Remain is still ahead. Probably”. The two latest polls both confirm the weekend surveys. Brexit has lost ground in the past week. ORB’s poll for the Telegraph suggests an eight-point shift in the lead (from “Leave” ahead by one point to “Remain” ahead by seven), while the latest YouGov/Times poll finds a five-point shift (Leave’s lead down from seven points to two).



So it does seem that the surge to Leave around ten days ago has been reversed. But is the current position a comfortable “Remain” lead or a narrow “Leave” lead? Here are three reasons why we can’t be absolutely sure.

*1. Any poll, however well conducted, is subject to sampling fluctuations. YouGov has conducted three polls in the past seven days. They have produced a two-point Leave lead, a one-point Remain lead and, now, another two-point leave lead. In a real, millions-of-votes contest, these are very different results. But in a world of polling samples they are statistically much the same. All three polls are consistent with a 50-50 division. We can’t tell whether the move to Remain reported in the weekend polls has continued or stalled.

*2. ORB’s figures have thrown up something curious. They poll 800 people in all, but they apply a strong turnout filter. They base their headline figures on the 500 or so who say they will definitely vote. It’s a number that increases the risk of sampling error. This is the basis on which ORB finds a shift from Leave to Remain.

If we take the views of the 700-plus people in the full sample who takes sides, then a different story emerges. Last week Remain enjoyed  a five point lead; this has now fallen to two points. The real change has been the way ORB’s turnout filter works. These are the proportions of people on both sides who say they are certain to vote. Remain: last week 60%, this week 69%. Leave: last week 68%, this week 64%. Suddenly, “remain” voters now tell ORB they are more determined to vote than “leave” voters.

It’s possible that this is a true and important finding, for until now virtually every poll has found that “remain” voters are more reluctant to make the journey to their polling station. If that has now changed, then the “leave” campaign is in real trouble.

However, it maybe that this is a sampling fluke. If so, ORB’s new headline figures may be overstating Remain’s lead. No such transformation in the turnout pattern is visible in any other poll, including today’s YouGov survey.

*3. That said, the difference between online and telephone polls does seem to have reappeared; and it’s a matter of judgement which system is more accurate. My initial blog explored this at a time when the difference was a chasm, with online polls saying the race was neck-and-neck, while telephone polls generally reported large Remain leads. Then the differences seemed to disappear, with both methods throwing up results ranging from narrow Remain leads to large Leave leads.

Now, it looks as if the difference is back. Fourteen polls have been conducted in the past ten days: seven each by phone and online. The phone surveys divide: four Remain leads, three Leave leads. The online polls divide: one Remain lead, one Level-pegging, five Leave leads. The single online Remain lead has been 1%; three phone surveys have produced Remain leads ranging from 3-7%.



For the reasons I gave when I looked at the online-versus-telephone controversy, my hunch is that Remain entered the final three days of this campaign just ahead. But it can’t be more than a hunch. On Thursday morning I shall assess the final polls and predict the result. Probably.

The final week: safety first

Anybody who says they know for sure how Britain will vote next week is fool or a liar, and possibly both. Since last weekend, we have had ten polls conducted by eight different companies. Five of the polls have been online, the other five conducted by telephone. As a result, we can be certain about some things, but less certain about others.



  • There has been a marked shift since the start of June from “remain” to “leave”. Every polling company finds more support for Brexit than they did in mid-to-late May. Movements in individual polls might reflect sampling fluctuations; but when every polling company, whatever their methods, finds the same direction of movement, this can no longer be reasonably explained by chance.
  • There are significant geographical differences. If the result is close, then London, Scotland, Northern Ireland and probably Wales will vote “remain”, while the rest of England will vote “leave”. Whatever the overall outcome, the political consequences of these divisions are likely to be huge.
  • There is a big generation gap and a big education gap. People over 65 are much more likely to support Brexit than people under 30; and those with university degrees are far keener of staying in the EU than those who left school at 15 or 16.
  • On the other hand, there is NO evidence of a significant gender gap. More women are undecided than men, so might well make up a majority of the late-deciders who could well make the difference next Thursday. However, among those who take sides, men and women currently hold similar views. This is unlike Scotland in 2014, when men voted on balance for independence, while women plumped decisively for keeping the UK together.



  • While the polls agree about the direction of movement, they don’t agree about the scale of the move or the current level of support for “leave” and “remain”. At one of the scale, since mid-May. And in the past six days we have seen figures ranging from a two-point “remain” lead (Opinium) to a ten-point “leave” lead (ORB). I believe the race is currently neck-and-neck, but this is more a best guess than a firm estimate.
  • There are still differences between online and telephone surveys. Overall, these differences are less stark overall than they were; but even when they report similar figures overall, they disagree about the way different groups of voters line up. Online polls tend to report big “leave” leads of 20 points or more among people over 65 and those who voted Conservative in last year’s general election. (There is, of course, a big overlap between these two groups.) Telephone polls tend to show a closer contest among Tories and the elderly. These differences provide another reason to be cautious about interpreting the polling data.
  • We have no real idea how high turnout will be next week. This could affect the result – but we can’t be sure how. Let’s suppose one or two million people make a late decision to go to their local polling station rather than stay at home. And let’s assume they belong to groups who are least likely to vote in general elections. If they are predominantly young people (including those who have registered online in the past few weeks), then this is good news for “remain”. But if they are mainly people who have the fewest educational qualifications, struggle to make ends meet, and live in parts of Britain where general election turnouts are way below the national average, then this will boost the chances of Brexit. Unless, that is, Labour manages to persuade its working-class supporters that it is in their self-interest to side with the overwhelming majority of Labour MPs and vote “remain”.



What, then, will happen in the final week? I argued in a previous blog that past referendums have usually produced a shift to the status quo, because people who decide late have opted for the safety of things continuing as they are, rather than take the risk of voting for change.

I stand by the basic judgement that late deciders will prefer safety to risk. However, I am no longer sure that this will benefit “remain”. In past fortnight the Vote Leave campaign has successfully raised the saliency of immigration, while the Stronger in Europe campaign has so far failed to keep the economic risks of Brexit at the forefront of voters’ minds.

This raises the prospect that the late deciders, who generally follow the news least, and have the weakest feelings about the EU per se, will take sides next week on the basis of “controlling our borders” rather than the dangers that Brexit poses for jobs, prosperity and living standards.

We should remember that public attitudes to immigration are prompted largely by the rise in insecurity and inequality over the past decade. Secure, well-paid jobs for millions of people are harder to come by, public services are being squeezed, most people in their twenties cannot afford to become home-owners, and so on. We can debate the reasons – technology, the 2008 banking crisis, demographic pressures on public spending, etc.

To many voters, however, globalisation and immigration are major culprits, making life easier for the well-off and harder for everyone else. If Brexit is seen as the route to keeping the rest of the world at bay, and giving Britons better-paid jobs, improved public services and a generally more secure future, then a final-week shift to safety could take late-deciding voters away from the status quo and towards Brexit.

On balance I still expect a shift to the status quo, and a narrow victory for “remain”. But we are dealing in probabilities and not certainties. We have a blurred picture of the present, and can only guess the future. It is perfectly possible that the trends of the past fortnight will continue, and we shall vote to leave the EU.

The big question for the final week is: can Stronger In Europe make its economic warnings of Brexit stick? If they can, they will win; if they can’t, they will lose.

A happy Brexit? No chance

The matter can now be settled. New surveys by Pew Research and YouGov put the issue beyond doubt. If Britain votes to leave the European Union, we shall find life cold on the outside. The notion that the other EU states will be nice to us is for the birds.

On a recent trip to Brussels, I heard much the same as many others: senior officials from both the European Commission and a variety of member states warned that if we vote for Brexit, we will either have to give up free access to the single market – or abandon plans to curb immigration from the EU. Freedom of movement is an essential feature of the single market. If we insist on “controlling our borders” more than we do already, then we shall find it harder selling goods and services across the Channel.

To which Brexit campaigners say: stuff and nonsense. Such talk is crude sabre-rattling. If we opt for Brexit, the other 27 member states will quickly come to terms with our decision. Their companies will not want to lose their easy access to their British customers. They will clamour for no-tariff, barrier-free deal with the UK, however hard we clamp down on EU citizens wanting to work in Britain. Everything will work out fine.


This week’s poll numbers show that this is moonshine. The other countries will have to strike a tough deal, not out of any animus towards Britain but because of the backlash they would face back home if they don’t. Here are some of the key findings.

  • Pew’s ten-country survey finds that Britain is far from the most Eurosceptic country. French voters have a lower overall opinion of the EU. Asked about the way the EU handles European economic issues, people in five of the other nine countries surveyed by Pew are more critical than the British: not just Greece, Spain and Italy, which might be expected, but France and Sweden, too. In none of the ten countries do a majority approve of the EU’s record.
  • Big majorities in all ten countries are unhappy with the way the EU is handling the challenge of refugees. The figures for Britain – 22% approve, 70% disapprove – are close to the ten-country average.
  • In all ten countries, more people want Brussels to shed power rather than continue to the journey towards “ever closer union”. In only two countries – France and Spain, where opinions are polarised – are the numbers even close. The mood across the EU is increasingly nationalist rather than federal; and it’s a mood that extends well beyond the people who vote for explicitly nationalist parties.
  • YouGov’s latest six-country Eurotracker survey helps to explain Pew’s findings. Across all six, voters see the benefits of EU membership accruing not to ordinary workers or families struggling to make ends meet, but to politicians, bankers and “people running big businesses”. EU enthusiasts point to the rights to protect workers and consumers, and the competition rules designed to make big companies behave themselves – but these cut little ice, not just in Britain, but France, Germany, Denmark, Sweden and Finland, too.

The discontent across Europe is palpable. Neither Brussels nor, in the main, national governments command public respect. (YouGov finds that Angela Merkel’s government is even less popular in Germany than David Cameron’s in Britain.)


That is the backdrop against which the rest of the EU will have to grapple with a vote for Brexit. It is inconceivable that Britain will be allowed a “best of both worlds” deal – all the benefits of a close relationship with the EU without the costs and responsibilities of membership. If that, or anything like it, were to transpire, voters across Europe would demand the same for their own countries. In order to fend off domestic political pressures, our current partners would have to hang tough and show their own electorates that following Britain out of the EU would be painful.

In the long run, the best response to the public mood uncovered by Pew and YouGov would be for the EU to reform itself to be leaner and more efficient. Then, the long, slow slog towards respect for “Brussels” could start. Indeed, if the UK votes “remain” on June 23, we could play a big role in promoting a constructive reform agenda. Meanwhile, we should be in no doubt that a vote for Brexit will create huge problems for our partners. And if they don’t look us in the eye when they impose tough terms, it will not be because they are shame-faced but because they are looking over their shoulder at their own nervous and often angry voters.

How history hits Brexit for six

The Brexit camp should enjoy its current slight bounce in the polls, for it may not last. If history is any guide, then “remain” is still heading for victory on June 23. Past referendums in Britain have tended to produce a late move to the status quo. The record from six such contests in the past four decades is striking:


1975 UK-wide referendum on the Common Market

Gallup’s polls during the final five weeks of the campaign showed that support for leaving the Common Market peaked point with two weeks to go – though the lead for “staying in Europe” was still a massive 28%. Gallup’s final eve-of-referendum poll put the lead for the status quo at 36% (68-32%). The result: a 34.4% gap (67.2% in, 32.8% out)


1979 Scottish referendum on devolution

Just over two weeks before the vote, Mori reported a 28% lead for the pro-devolution camp (64-36%). With one week to go, the margin was 20% (60-40%). The swing to the status quo accelerated in the final week. Mori’s eve-of-referendum predicted a 50-50% outcome. The result: a narrow 51.6-48.4% lead for devolution – a margin too small to reach the winning line set by Parliament, which insisted that 40% of Scotland’s entire electorate should support devolution; on the day only 33% did so.


1979 Welsh referendum on devolution

With three weeks to go, an Abacus/BBC survey said that 58% would vote for the status quo and against devolution. By referendum week, that figure had jumped to 75%. On the day, 79.7% of voters opted to reject a Welsh assembly.


1997 Welsh referendum on devolution

Three weeks before the vote, Beaufort Research showed a majority of almost two-to-one for devolution. The appetite for change receded significantly towards the end of the campaign; but an eve-of-referendum poll by NOP still indicated a 12% lead for devolution (56-44%). In the event, Wales did vote to set up its new Assembly, but by the narrowest of margins: 50.3-49.7%. Support for the status quo had jumped by 16 points in the final three weeks.


2011 UK referendum on the voting system

Three weeks before the vote, telephone polls showed an average 12% lead for the status quo (46-44%), while online polls showed the two sides neck-and-neck (49% for the Alternative Vote, 51% for first-past-the-post). With ten days to go, support for the status quo had climbed to 60% in telephone polls and 57% in online polls. The final online polls showed a 22% lead for first-past-the-post (61-39%), while the final telephone polls put the margin at 34% (67-33%). The result: 67.9% for the status quo, and 32.1% for change – a victory margin of 35.8%


2014 Scottish referendum on independence

The u-turn pattern here was even more dramatic than in the 1975 referendum on the Common Market. For much of the spring and summer, ahead of the September 18 referendum, support for independence hovered around 40%. Then, during August and early September, the appetite for independence grew, and with around two weeks to go, three polls by different companies showed the race neck-and-neck.

However, in the final ten days, banks and supermarkets warned that independence would mean dearer mortgages and groceries, history reasserted itself and support for the status quo revived. The eve-of-referendum polls showed an average 5% lead for Scotland remaining in the UK. An on-the-day survey by YouGov recorded a further shift away from independence. The final margin of victory for Better Together was 10.6% (55.3-44.7%).


The one exception

Alert readers will have noticed one glaring omission from this list: Scotland’s devolution referendum in 1997. The campaign polls showed little movement in opinion, and all were close to the final result: 74.3% for devolution, 25.7% against. How come?

My answer is that this campaign was qualitatively different from the others, in that the proposal for self-government had been hammered out over a number of years. Virtually every major player in Scottish politics, business, the trade union movement, the churches and civil society generally had been involved in the process, and all bought into the deal. The only significant exception was the Conservative Party; but as they had lost all their Scottish seats in the general election four months earlier, their opposition had little impact.

In short, the 1997 Scottish referendum was, in effect, a mechanism for ratifying a national consensus, not a means of resolving a major national dispute. In these circumstances, we should not be surprised by the lack of a late-campaign swing to the status quo.

Plainly the current referendum on the European Union is a “dispute” rather than a “consensus” referendum. So: will history repeat itself? If it does, then if – as I have argued in recent blogs – “remain” entered the final month of the campaign with a modest lead, then we should expect that lead to widen in the days leading up to June 23.


Why the status quo usually gains ground

The prospects for the next fortnight depend partly on why the status quo has generally gained ground in past referendums. Two reasons seem to have contributed to the historical pattern, and both look likely to apply this time.

  1. Some people who take an interest in the referendum issue are unhappy with the way things are and tell pollsters until the last week or two that they support change. It is a cost-free way to express dissatisfaction. (Much the same often happens in mid-term by-elections, when the government of the day suffers a big adverse swing: voters have the chance to protest without having to worry that their vote will hand power to the opposition. However, when referendum day approaches, they think not just about the status quo and what’s wrong with it, but about the alternative. Some voters decide that, on balance, change carries risks that they would prefer to avoid. They draw back from the cliff edge and vote for the status quo.
  1. Other people have better things to do than follow referendum news at all until the last few days. One of the most significant poll findings of recent weeks has been a YouGov/Times poll which reported that 45% of Labour voters were unaware that most Labour MPs want the UK to stay in the EU. I suspect that pollsters could ask other knowledge questions and find that millions of voters are unaware of many basic aspects of the campaign.


Now, many such people will end up not voting at all. But quite a few will vote; however, they are the kind of people who hold no strong views about the EU either way (or the issue at hand in past referendums). Whatever they tell pollsters some weeks ahead of the referendum – if they respond to pollsters at all – most of them end up deciding instinctively that it is safer to leave things as they are than to vote for change.

These are not completely hard-and-fast categories. Some people will be a mixture of both types. The key thing is that their views are less fixed and more risk-averse than people with strong opinions and a passion for political news.

(A recent article by Daniel Jackson of the Campaign Group, for The Times’s Red Box, makes much the same point in a different way. He divides voters into pioneers (typically young, liberal, pro-EU), settlers (typically older, more tribal, anti-EU) and prospectors. People in this final group tend to be busy, status-conscious, motivated by economic self-interest – and often decide late how to vote. He argues that their worries about prosperity will in the end trump their dislike of immigration, and they will break towards remaining in the EU.)


Future imperfect

The words of those financial advertisements should be heeded: “past performance does not guarantee future results”. Some recent polls have reported a shift towards Brexit. Maybe this will continue, and the safety-versus-risk battle will play out differently than it has in past referendums. Perhaps immigration will trump economics as the issue that sways those who are still not sure how to vote; that would shift the dial towards Brexit. Perhaps some crisis will erupt, to do with terrorism, refugees or the Eurozone, to make the late deciders judge that we are safer outside the European Union. There is always an element of unpredictability whenever voters are asked to decide their nation’s future – thank goodness. It would be a sad democracy that behaved in a wholly deterministic manner.

However, the fact that we cannot be certain what will happen between now and June 23 does not mean we cannot judge the probabilities. And while the record of past referendums does not guarantee a shift to “remain” in the final days of the current campaign, it does suggest that such a shift is more likely than not.

“Remain” is still ahead. Probably.

Here is a mid-course correction. Two weeks ago I argued that telephone polls are providing a better guide than online surveys to the state of play in the EU referendum. Overall, taking phone polls together, I still hold to this, and for the reasons I gave. However, individual phone polls are giving cause for concern, including the two latest, for they are proving extremely erratic.

Within the past fortnight, online polls have been consistent: all have shown the contest close to level-pegging, as they have all year, with both remain and leave on 50%, plus or minus two – hence somewhere between a four-point “remain” lead (52-48% one way) and a four point “leave” lead (52-48% the other way). This amount of variation is what would be expected if the public’s views are not changing, given the laws of probability and the inevitable sampling fluctuations.

In contrast, recent telephone polls are all over the place. Excluding “don’t knows”, they range from a 60-40% lead for “remain” (Ipsos-Mori) to a 52-48% lead for “leave” (ICM). The ICM figures are especially striking. This week, as two weeks ago, ICM conducted simultaneous online and telephone surveys for the Guardian. The two online surveys suggest that nothing has changed: both report a four-point lead for “leave”. The two telephone polls tell a completely different story: a ten-point “remain” lead has been wiped out by 7% swing in a fortnight.

This is more than a matter of curious statistical interest. Ipsos-Mori’s figures led to a sharp rise in the value of sterling; ICM’s figures prompted an equally sharp fall.



Part of the problem with telephone polls is that they are more expensive to conduct than online surveys. As a result, their samples tend to be smaller. They are mostly either 800 (ORB’s polls for the Daily Telegraph) or 1,000 (everyone else). Conventional statistical theory indicates a margin of error of 3-4 points for each side – and therefor a margin of error of 6-8 points on the gap between remain and leave.

In contrast, online surveys generally question twice as many people: 1,600-2,000. This does not eliminate random sampling error, but it does reduce it, and so reduce the risk of freak outliers (or “rogue” polls as they are sometimes called). This goes some way to explaining why the online polls have told a story of broad stability, why the telephone surveys oscillate more, generate dramatic headlines and affect the currency markets.

However, we can go a bit further to make sense of this week’s two telephone polls showing a marked move towards “leave”.



First, ORB. Not only does it start with a smaller sample than other companies; it then reduces the effective sample even more, for it bases its headline figures on those who say they are certain to vote. In ORB’s latest poll, this turnout filter reduces the sample to just 433 (226 certain to vote “remain”, 207 to vote “leave”, leading to a headline figure of 52-48%). The margin of error is more than five percentage points for each side and 10-points-plus on the gap between them.

Is ORB right to use such an aggressive turnout filter? Polling companies disagree, and past evidence is inconclusive on the best way to identify which respondents in any given sample will actually take the trouble to vote. For the record, ORB’s latest figures indicate a 55-45% lead for “remain” among all those who state a preference, and a 54-46% lead if the turnout filter used by some other companies is applied. (ORB asked people how likely they were to vote on a scale of 1-10. 144 people who said they would vote “remain” responded with a score of 7, 8 or 9; just 104 “leave” supporters gave the same range of responses. ORB excluded them from its final in-out figures; some other companies would take them into account and, in this instance, double the “remain” lead from four to eight points – and give the Telegraph a less dramatic story)



As for ICM’s latest figures, something looks odd about its telephone poll. Over the years, ICM has pioneered the use of adjustments to improve the accuracy of its projections – partly by its use of turnout filters, and partly by making judgements about “don’t knows”. ICM suspects that sometimes these conceal shy Tories, and help to explain the way polls, going back to 1992, have often understated Conservative support. ICM has usually been vindicated, outshining most other companies in the 1997, 2001 and 2010 general elections.

However, I wonder whether ICM’s adjustments this time point to problems with its latest telephone sample. Normally the voting adjustment moves the party lead by 2-4 points in the Conservatives’ direction. (Admission: this is not a precise calculation, but my impression from watching ICM surveys over the years.) What happened this time is different. After weighting its raw data to make its sample look like Britain’s electorate, it reports a six point Labour lead over the Conservatives (37-31%, after excluding don’t knows). But after adjusting for turnout and “don’t knows”, ICM reports a four point Conservative lead: 36-32%. Thus ICM’s adjustment alters the party lead by ten points. I cannot recall ICM making anything like this scale of adjustment before.

For comparison, its simultaneous online poll has a more normal adjustment, increasing the Conservative lead from two points (34-32%) to five (36-31%).

It is greatly to ICM’s credit that it hasn’t been tempted to massage its figures. By placing its full data on its site, people like me can look under the bonnet of the headline numbers; so full marks for honesty and transparency. But a ten-point voting adjustment suggests that there is something – how shall we put this? – unusual about its phone sample this time. Some have suggested that phone polls have a problem obtaining a good sample over a bank holiday weekend, especially one that falls at the beginning of half term. I don’t know whether this is true, though I can think of one or two odd results from phone polls in the past conducted over bank holiday weekends.



Bottom line: the evidence is not – at any rate, not yet – conclusive that there has been a shift to “leave” in the past fortnight. I believe that “remain” is still modestly ahead and that not much has changed in the past fortnight. But if the next few days see a decline in the “remain” vote is a majority of both phone and online polls, then that judgement will need to be revised. Meanwhile, companies such as ORB that use aggressive turnout filters should start with much bigger samples – at least 1,500 in my view – and so reduce the risk random fluctuations in their results that may have little or nothing to do with what is actually happening in the country.


Can immigration trump risk-versus-safety?

Thank goodness for YouGov and Lord Ashcroft. While other pollsters give us little more than figures for staying in or leaving the European Union, they have been exploring the attitudes that will shape the eventual outcome. YouGov’s poll this week for The Times, and Lord Ashcroft’s latest bulletin provide valuable fresh information.

The most significant concern the perceived risks of Brexit. YouGov regularly asks: How safe or risky do you think it would be for Britain to leave the European Union? Responses oscillate a little from poll to poll, but every time this question has been asked this year, at least 45% have said Brexit would be risk, while never more than 40% have said it would be safe. The latest figures are; risky 45%, safe 38%. The gap has narrowed in the past month, having widened earlier in the spring, but remains significant.

When YouGov asks the opposite question: How safe or risky do you think it would be for Britain to remain part of the European Union?, the latest figures are: safe 49%, risky 35%. The numbers saying “safe” match the highest that YouGov has recorded.

Lord Ashcroft has asked a slightly different question which prompts a similar conclusion:

Which of the following do you think represents the greater risk?

  • The fact that we don’t know for sure what life outside the EU would be like for the UK – 58%
  • The fact that we don’t know for sure how the EU will develop or change if the UK remains a member – 42%

 These figures are plainly good for the “remain” campaign – indeed, probably even better than they look. Both polls were conducted online and, as I argued in my last blog, I believe online polls slightly overstate support for Brexit. If so, then they are also slightly overstating Brexit-related attitudes, such as on whether Brexit is safe or risky.

In any event, these latest findings underpin my expectation that, in the absence of external shocks, such as a new terrorism, refugee or Eurozone crisis, support for “remain” is likely to grow towards the end of the referendum campaign. This is beginning to look like so many referendums round the world. The great majority of times, late-deciding voters plump more for the familiarity of the status quo than the hazards of abrupt change. The latest polls suggest that the ingredients are in place for a similar late move this time.

Could the issue of immigration upset this prospect? On the face of it, both of the latest polls suggest it could. YouGov asked: Do you think there would be more or less immigration into Britain if we left the European Union, or would it make no difference?

  • More immigration into Britain – 3%
  • Less immigration into Britain – 59%
  • Would make no real difference to the amount of immigration -25%
  • Don’t know – 13%

 Lord Ashcroft asked: When it comes to immigration into Britain which of the following do you think lies closest to the truth?

 We’ll never be able to bring immigration under control unless we leave the European Union – 46%

  • We won’t be able to bring immigration under control even if we leave the European Union – 34%
  • Immigration is already under control -7%
  • Don’t know – 14%

 Given that most people want immigration to fall sharply, these figures look like good news for the “leave” campaign. However, I doubt whether the issue has the capacity to help the Brexit cause that much. First, The Brexit camp already has the support of voters who dislike immigration most.

Secondly, it is by no means clear that Brexit would cause a sharp fall in immigration. Indeed some “leave” campaigners such as the Ukip MP Douglas Carswell and the Times columnist Tim Montgomerie, say that they want different immigrants (for example, Indian technology graduates rather than unskilled Romanians), not sharply fewer immigrants in total.

That’s not all. Stephen Booth of the think tank Open Europe has unearthed important data about immigration rates in the countries that different Brexit campaigners wish Britain to emulate, either in relation to the EU or in terms of their immigration policies. In all four cases – Norway, Switzerland, Canada and Australia – immigration rates are around twice as high as in Britain, relative to their population. So much for the proposition that greater sovereignty equals radically lower immigration.

It’s possible to argue that the UK could have a different mix of immigrants outside the EU (that would depend on the precise nature of any deal to have continued access to the EU’s single market). However, on the evidence of the other countries that the “leave” campaign cites, it is NOT possible promise that Brexit would lead to fewer migrants.

Actually, that’s not quite right. I can envisage one way in which Brexit would lead to a sharp reduction in immigration. This would certainly happen if Britain’s economy dived into recession, unemployment soared and far fewer jobs were available for anyone, wherever they were born. Here’s my suggestion for a “leave” campaign slogan: “vote for slump and fewer migrants”. Good luck with that one, Brexiteers.


EU referendum: “remain” on course for clear victory

Welcome to my first blog for The Politics Counter. I hope you enjoy it. If you have received this by email and do not wish to receive further blogs, simply click the ‘unsubscribe’ link below. If you have come to this via the Politics Counter site, and DO wish to received future blogs by email, please complete the “subscribe” form to the right.

 Two rival narratives are competing for our attention in the EU referendum campaign. Online polls say the contest is neck-and-neck, while telephone polls insist that “remain” holds a clear lead. I believe that the telephone polls are nearer the truth. Unless at least two million “remain” supporters are scared into switching sides by some sudden crisis (to do, say, with terrorism, refugees or the Eurozone), the UK will vote by a comfortable margin to stay part of the European Union. This blog sets out my reasons for saying this.

Let’s start with the basic numbers. Seventy-nine referendum polls have been published this year. Fifty-six have been conducted online. Most have shown a very close race. If we combine them to create a giant poll-of-polls, then both sides have 42%. The other 23 polls have been conducted by telephone. All bar one have shown “remain” ahead, usually by a significant margin. Their poll-of-polls has “remain” ahead by 50-39%.

The past ten days have maintained this pattern. Three online polls show “leave” on average just ahead by 43-42%; three telephone polls indicate, again on average, to a clear lead for “remain”, by 52-39%.


Before we discuss these differences, a broad point needs to be made about all polls. They face a huge challenge getting anything right. By definition, online polls survey people who have chosen to join a polling panel. In theory, telephone polls are able to reach everyone who has a phone. But they are at the mercy of response rates; and in recent years these have collapsed. Twenty years ago, according to one of our most respected telephone pollsters, they achieved a response rate of around 30%. To obtain 2,000 interviews, they asked their computer to generate 7,000 random residential telephone numbers. These days, to obtain 2,000 interviews, they need to start with as many as 28,000 phone numbers, for response rates have fallen to just 7%.

In short, all polls require research companies to extrapolate from the small proportion of the general public they can reach to the far larger number of people who neither join online panels nor respond to telephone polls. The surprising thing about is not that today’s polls sometimes get things wrong, but that they have such a good record of getting so many things right.

The reason why our leading polling companies are so good, so often is that they go to great lengths to make their samples match the country’s population – by age, gender, region, social class, past vote and so on. Usually this process generates accurate results. But sometimes it doesn’t, either because the silent, unpolled, majority, differs from the poll-friendly minority in some way that is not captured even by the smartest demographic sampling – or because of “mode effects”, in which the way a poll is conducted prompts some respondents to conceal their true feelings.


The debate about the difference between telephone and online polls in the current referendum campaign is essentially a debate between these two explanations. Let us consider them in turn.

Supporters of telephone polls say that their samples are better in the current referendum. Even though their response rates are so low, they still reach a fair number of two key groups : a) the people with little interest in politics who tend to favour the status quo rather than take risks; and b) well-educated social liberals who lead busy lives and are less likely to join online panels. Both groups, for different reasons, veer towards “remain” rather than “leave”.

Supporters of online polls say that their recruitment, sampling and weighting methods ensure a good mix of the politically interested and uninterested. As for the matter of social liberalism, the reason why online and telephone polls sometimes diverge is that the anonymity of online surveys prompts more honest answers. Some people are unwilling to admit their true less-than-liberal feelings on, say, gender or race equality, to a stranger over the phone. By the same token, some people who plan to vote to leave the EU think it more prudent to tell a telephone pollster that they will vote “remain”, for this is the view espoused by the government, main opposition, most business leaders and, generally, much of the establishment.

It is possible that both sides are at least partly right – that, say, online panels generate less good samples but more honest answers. Indeed, when I worked for YouGov, we gathered a lot of evidence from different countries about “social satisficing” – an ugly (and, inevitably, American) term for the habit of some people sometimes being more candid online than when interviewed by a stranger. If I wanted to uncover people’s true views about sex, illegal drugs or tax-and-spend, I would always trust a well-conducted online poll more than a telephone poll.

However, I am not sure that social satisficing explains the difference in the current referendum. Here are five reasons – none of them individually decisive but, together, in my view, persuasive.


Four of the six latest surveys report voting intention. The two online surveys (by YouGov and ICM) both show Ukip on 18%. The two telephone surveys show Ukip on 14% (ICM again) and 10% (Ipsos MORI). My judgement is that the figures in the online polls are too high, while Ipsos MORI’S figure looks too low.

My reason for saying this is that the lower figures are more consistent with Ukip’s performance in this month’s elections. YouGov, uniquely, polled in London, Wales and Scotland. It got all the big stories right – Sadiq Khan winning comfortably in London, the Tories coming second in Scotland, Ukip breaking through in Wales – but in each case it overstated Ukip’s support, and by roughly similar amounts. Overall, for every ten votes that YouGov predicted that Ukip would secure, the party obtained only seven.

YouGov has recognised this point. In its latest poll it has tweaked its methods with the aim of eliminating its overstatement of Ukip’s support. However, even after the change, its 18% for Ukip still looks too high.

The relevance to the referendum is clear. Ukip supporters are almost unanimously for Brexit. If their numbers are overstated, so is support for Brexit. My own judgement is that the latest online polls overstate Ukip’s support by around four percentage points. If we reallocate these to the Conservatives, whose supporters divide roughly evenly on the EU, then this has the effect of increasing the “remain” vote by 2% and reducing the “leave” vote by the same amount. This turns an online-average of a one point “leave” lead of 1% to a “remain” lead of 3%. It doesn’t completely close the gap between online and telephone results, but it reduces the difference. (The opposite adjustment to Mori’s low Ukip figure would reduce its exceptionally large “remain” lead from 18% to 14%, and bring it closer to the figures produced by ICM’s and ORB’s telephone surveys.)


Some weeks ago Populus conducted a comparative study, putting the same questions to online and telephone samples. The results appear in a  report by Matt Singh (whose Number Cruncher blogs are essential reading) and Populus’s James Kanagasooriam. Among the questions it put to respondents in England was whether they considered themselves “more British than English”, “more English than British”, or “equally British and English”. This was a question that the British Election Study had posed to a random, face-to-face sample last summer. The key thing is the number who said they were “more English than British”, for this correlates closely with attitudes to the EU: people who say they are “more English” tend also to be more pro-Brexit (and also more pro-Ukip). These are the numbers found by each method: face-to-face: 24%; telephone 24%; online 32%.

I see no reason for these differences to flow from “social satisficing”. I doubt that anyone will respond differently to a stranger than to an online survey. It reinforces the point about differences in Ukip support: sampling differences may not tell the whole story, but they tell a significant part of it.

(Before proceeding, and before anyone else points this out, I should add that Populus’s director is Andrew Cooper, a Conservative peer and pollster for Britain Stronger in Europe. Those who think this taints his, and Populus’s, analysis are of course free to do so. But I know Andrew; and I also know from my experience at YouGov that a polling company’s reputation depends on its honesty, objectivity and quest for accuracy. All pollsters should be judged on their merits, not on imagined, and invariably false, assertions of wilful bias.)


Populus has also analysed data from last summer’s face-to-face survey by the British Election Study. This was an old-fashioned, time-consuming and expensive survey. It started with a list of random names and addresses. Its interviewers were asked to speak to them, and ONLY to them. If the named respondents were not available first time round, interviewers were instructed to return at least a dozen times over a period of weeks.

Why is this relevant to the EU referendum? Because, according to Populus’s analysis, there was a systematic difference between those who were reached on the first visit and those who were reached later. The longer it took to get hold of a respondent, the more likely was the respondent to be a socially liberal Conservative, “more British than English” – and more favourable to British  membership of the EU.

Now, it does not necessarily follow that people who are harder to reach in a face-to-face survey are less likely to join an online panel and answer their surveys; but it does provide indirect corroboration of the view that the differences between online and telephone polls are more to do with sampling than “social satisficing”.


National referendums are rare, but we had one five years ago, when the UK voted by 68-32% to retain the first-past-the-post voting system for electing MPs rather than switch to the Alternative Vote. All the final polls pointed correctly to a big victory for the status quo, so their performance did not attract as much attention as they did in last year’s general election.

However, once again there was a clear difference between the final online and telephone polls. The two online polls (Angus Reid and YouGov) predicted a 61-62% vote for keeping the existing voting system; the two telephone polls (ComRes and ICM) had the figure much closer: 66-68%.

On its own, this record is of only limited relevance today. The EU is a vastly different issue, and polling methods have evolved over the past five years. That said, it does look as if the online surveys in 2011 ended up systematically understating the number of (less politically engaged?) voters who preferred the status quo.


Once again, the evidence is indirect, but suggestive. This week the New York Times has done a detailed analysis of the polls in this year’s US primaries. It’s main conclusion is simply stated. On average, the telephone polls correctly measured Donald Trump’s popularity, while the online polls generally overstated his support. Could it be that in both Britain and the US, online surveys are overstating the desire for change? The New York Times surmises that one reason is that online surveys generate more “don’t knows” than telephone surveys. It may be that a number of these don’t knows are people who would take sides in a telephone poll – and do vote, tending to side with the status quo.

If this is right, it sounds uncannily like part of the debate here in Britain, where online polls in the current referendum elicit more don’t knows and – perhaps – understate the fear of change by politically less-interested voters.

There is an alternative explanation which online pollsters prefer: that turnout in the referendum won’t be that high, and that the higher don’t know figures produced by online pollsters are more realistic. According to this view, the telephone polls are including too many less-interested status-quo respondents who will end up not voting at all.

Were this the only evidence about telephone-versus-online polls, I’d be tempted to support that view. But as the US telephone polls have proved generally more accurate at measuring Trump’s support, and the New York Times analysis is so consistent with the evidence of what is happening here in Britain, I’m inclined to side with the analysis of the telephone pollsters.


 Two conclusions in fact. The first is that online polls should not be pilloried for their figures. Given the hazards of extrapolating from the responses of those who join their panels to the rest of the population, they have a remarkable record for accuracy. Sometimes they outperform telephone polls – not just on “social satisficing” issues, but occasionally with superior samples. For example, in 2008, YouGov alone predicted Boris Johnson’s victory; all the telephone polls consistently and wrongly put Ken Livingstone ahead.

On the great majority of issues, online and telephone polls produce comparable results. And because online methods are cheaper, this means for the same money, more people can be asked more questions, and so yield a deeper understanding of the public mood.

However, the EU referendum is one of that minority of occasions when there is a significant difference; so far, telephone polls seem to generate more accurate results. We shall see whether the online polls acknowledge this and make further changes to their methods.

Conclusion number two: I reckon that the current state of opinion among those who express a view is something like: remain 55-57%, leave 43-45%. The gap probably narrows when one takes account of the greater determination of “leave” supporters to turn out to vote, to around: remain 53-55%, leave 45-47%.

However, past referendums in Britain and abroad have generally produced a late shift to the status quo, as people who make up their minds late tend to prefer the familiarity of the status quo to the risks of change. Given the character of the debate about the economic impact of Brexit, a similar shift looks likely this time.

This could lead to a significant majority in favour of the UK staying in the EU. From David Cameron’s point of view, this is just as well. A narrow 51-49% victory would be unlikely to settle the matter for long, or bring the two sides within the Conservative Party together. He probably needs a percentage majority in double digits – 10% or more – to have any chance of marginalising his anti-EU critics inside his party. The good news for the Prime Minister from this analysis is that he is currently on course to achieve it. The bad news is that it could still go wrong for him, especially if some kind of crisis erupts in the next five weeks.