It’s been a long journey, but we’ve finally arrived at the eve-of-referendum polls. For a lot of the Scottish referendum campaign the discussion about polls was one of right or wrong – we had lots of polls showing the same trend (flatlining!), but showing different absolute figures. Companies like MORI, TNS and YouGov were showing big NO leads; companies like Panelbase and Survation were showing a tight race. Then we had a period of some companies showing a strong movement towards YES, some not, and we have ended up with everyone showing much the same figures (what was the true picture earlier in the campaign we will never know for sure – by definition you can check eve-of-election results against reality, but never mid-term ones). With one MORI poll still to come, here are the YES shares in the latest polls from each company (taking the online and telephone methodologies seperately for those companies who have done both):

Ipsos MORI (phone) 49%
ICM (phone) 49%
TNS (face to face) 49%
YouGov (online) 48%
Panelbase (online) 48%
ICM (online) 48%
Opinium (online) 48%
Survation (online) 48%
Survation (phone) 47%%

Essentially everyone is predicting the same result, the margin of error on most of the polls is around plus/minus 3%, every poll is within two percentage points of the others. This isn’t going to be a case of individual pollsters getting it right or wrong, they’ll either all be around about right or all be horribly out.

There’s a temptation when the polls are like this to say YES and NO are within the margin of error, that it’s “too close to call”. It doesn’t really work like that – these polls are showing NO ahead. The margin of error is on each individual poll, and it’s equally likely to happen in both directions. Hence if the “true” balance of public opinion in Scotland was 50/50 we’d expect to see a random scattering of results around that point, some polls showing yes, some polls showing no. We’re not seeing that. We’re seeing polls randomly scattered around the 48/52 mark, suggesting that’s most likely where public opinion is – a very small lead for the NO campaign.

It’s possible there will be a very late swing, that people will have changed their minds in the last few hours or in the polling station itself. In most polls there really aren’t that many don’t knows left to make their minds up though.

The alternative route to an upset is if the polls are wrong, if there is some systemic issue above and beyond normal random sampling error that affects polls from all the companies. I wrote yesterday about what the potential risks are – the main challenges in my view are first whether people who are on the fringes of society and normally play little part in politics don’t get picked up in polls but do vote; and secondly whether there has been an issue of differential response rate, have the obviously more enthusiastic yes voters been more willing to take part in polling that no voters are?

Personally I’m a little more worried about the latter – I think there’s more chance of the polls ending up underestimating the NO vote than the YES vote, but there comes a time when you just have to trust the data. The polls say the result will be around about YES 48%, NO 52%. We will see on Friday morning.


With what I assume are all Sunday’s Scottish polls in, where do we stand? Looking across the board at all six companies polling, two of them using two different modes, we actually have a broadly consistent picture. Excluding don’t knows, the Yes shares in the 8 different companies/methods are:

ICM (online) 54%
Panelbase (online) 49%
ICM (phone) 49%
TNS (face to face) 49%
YouGov (online) 48%
Opinium (online) 47%
Survation (online) 47%
Survation (phone) 46%

Seven of the polls are clearly clustered around a small lead for the NO campaign, with the one exception that rather odd looking ICM online poll with a smaller sample size than their usual online efforts. A lead of just a couple of points in a single poll is within the margin of error, but in this case all but one poll is showing NO ahead, so I think we can reasonably say that the polls are giving NO a genuine but small lead.

If the polls are broadly correct, and if nothing changes in the last five days, then NO look like they’ll have a narrow win… but of course those are two very significant ifs. It’s certainly possible for a race this tight to change within a few days and there have certainly been occasions in the past when the polls have had a systemic error of a couple of points in one direction or the other.


A round up of various polling bits and bobs from the last few days, starting with a couple of Scottish polls. TNS released their latest figures yesterday, as usual there’s quite a gap between fieldwork and release – it was conduced between 23rd July and the 7th August, so overlapping with the Commonwealth games and almost wholly before the Salmond-Darling debate. Topline figures were YES 32%(nc), 45%(+4), DK 23%(-4), or without don’t knows YES 42%(-2), NO 58%(+2). It’s a shift towards NO, but it may be largely a reversion to the mean – TNS’s previous poll had NO dropping five points, so this is monstly just a reversal of that.

This week’s we’ve also seen the latest data from the Scottish Social Attitudes Survey. This is an academic project, so by the standards of media polling the figures are very old – the fieldwork was between the 12th May and 17th July – but they are worth noting because of the quality of the fieldwork. This was a proper probability sample, something extremely expensive you only normally see for academic and government work. Standard face-to-face polls aren’t really random, they use demographic quotas and cluster sampling (interviewers get sent to a place and door knock to five 3 male pensioners, 2 women under 30, etc); telephone polls are quasi-random – they ring randomised telephone numbers, but not every house has a phone and the overwhelming majority of calls don’t result in an interview; internet polls normally draw a sample from a panel. The Scottish Social Attitudes survey draws a genuine random selection of Scottish addresses and then sends a face-to-face interviewer to call at that address until they make contact and can arrange an interview, the response rate is 57% (phone polls these days are under 10%). The SSA survey found referendum voting intentions in the May-July period were YES 25%, NO 43%, Don’t know 32%, a squeeze question pushed those don’t knows to YES 33%, NO 51%, DK 15%. Excluding don’t knows that the equivalent of YES 39%, NO 61%.

Moving on, there have been a couple of polls on attitudes towards British participation in air srikes in Iraq. YouGov for the Times found 40% of people approved of the RAF taking part in airstrikes against ISIS, 36% were opposed (tabs here). ComRes for ITV found 45% of people supporting British fighter planes conducting airstrikes against Islamic State, 37% were opposed. Unsurprisingly there was strong opposition to British ground troops being sent back into Iraq – only 18% supported it in the ComRes poll. (tabs here)

Finally we have another academic election prediction site, joining Steve Fisher’s projection here and Rob Ford, Will Jennings et al’s projection here. The latest addition is from Nick Vivyan, Chris Hanretty & Ben Lauderdale at ElectionForecast.co.uk. As I write Steve is predicting a Conservative lead of 4 points at the next election, producing a hung Parliament with the Conservatives the largest party; Rob and colleagues are predicting a 0.7% Labour lead at the next election; Nick, Ben and Chris are predicting a 1.2% Conservative lead, producing a hung Parliament with Labour the largest party.


Tonight’s YouGov/Sun poll has topline figures of CON 35%, LAB 38%, LDEM 8%, UKIP 11%. The three point Labour lead is typical of this week’s YouGov polls, which have all shown 3-4 point leads.

A couple more things to flag up, earlier in the week YouGov repeated their question asking people to put the parties and their leaders on a left-right spectrum. There isn’t much change since it was last asked. Labour are still seen as more centrist than the Conservatives, Cameron a little more right-wing than Miliband is left-wing. Cameron is seen as marginally to the left of his party, Miliband bang in line with his. In that sense Ed Miliband isn’t seen as some wild left winger (certainly not compared to the right-wingness of the Tories), but note that he is seen as far more left-wing than his predecessors: Gordon Brown and Tony Blair were both seen as significantly more centrist than the party they led. There are some very nice graphs of the data here.

While it’s not really about polls regular readers will know my sideline in boundary changes. While the boundary review for the coming election was cancelled the changes the government made weren’t repealed, just delayed. The process will start again automatically in 2015, so the issue will inevitably raise its head after the next election with either the Boundary Commissions starting a new review under the new rules, or the government legislating to change the rules again. Johnston, Rossiter & Pattie – the foremost scholars of British boundary redistributions – have published a new paper aimed at informing that debate, looking at whether slightly increasing the tolerance from 5% to 8%, encouraging the Boundary Commissions to split more wards, or sticking with 650 seats would reduce the level of disruption (spoilers: the first two would, the latter wouldn’t). It’s summarised here, and the full report is here.


After every local election I start with the same warning – local elections are not general elections. Fewer people vote, on different issues, and for some who don’t vote on local issues it’s an opportunity for a mid-term protest vote. They aren’t a prediction of the general election. That does not, however, mean that local elections tell is nothing about the bigger picture and it certainly doesn’t mean they don’t have a massive impact on the wider political narrative and public opinion. What can we pick up from last night’s results so far?

Let’s start with UKIP. At the most basic level UKIP have obviously done very well. As I write the BBC suggest they are getting about 25% of the vote in wards they contest (though much worse in London) and have gained about 100 councillors. Two things to put this in context though – the first is that this is roughly the same level of support as they got at the local elections last year. I don’t really know if that’s good or bad for them. At one level they’ve sustained last year’s advance, it’s not a flash in the pan and they are more firmly establishing themselves as a major force. On the other hand, these are local elections on the same day as the European election… shouldn’t they be improving on last year? Second caveat, their vote is still very evenly spread. They are getting many more votes than the Lib Dems, but far fewer councillors. As was the case last year, UKIP are doing fantastically well at coming second in many places and that doesn’t build the council base they require in target seats for the general election. They have made some breakthroughs though – most notably in Essex councils like Thurrock and Basildon where they took seats from both parties, pushing Labour Thurrock and Tory Basildon into no overall control. They’ve also done extremely well in Rotherham.

Moving onto Labour, their results nationwide seem a little lacklustre. They’ll be pleased with the performance in London where they’ve gained several councils, most notably the Tory “flagship” of Hammersmith & Fulham, but they’ve made only sporadic progress elsewhere. As I write John Curtice doesn’t seem to have produced a projected national share yet, but he’s suggested they are up only 3% since 2010 so it sounds like it could be a very anaemic Labour lead. I think the bottom line for Labour is that their local election performance is much like their performance in general election polls. They aren’t an opposition that is soaring ahead in the opinion polls, they’re an opposition that has a very modest lead in the opinion polls. These aren’t soaring local election gains, they are modest local election gains.

The Tory performance is the mirror image of that. They are losing seats and councils, but not disasterously so. They’ve done badly against Labour in London and will be sorry to have lost Hammersmith & Fulham, but happy they have some gains here and there to weigh against it.

Finally the Lib Dems. Horrid council results have become par for the course for them, and these seats were last fought at the height of “Cleggmania” so losses are to be expected. The party normally comfort themselves with the defence that the party do far better in areas where they have sitting MPs. As usual, this is true in some cases, but not in others. The Lib Dems have held their own in places like Colchester, Sutton and Eastleigh… but they’ve also lost ground in Lib Dem held areas like Richmond, Kingston, Haringey and Cambridge, so it’s swings and roundabouts. Either way it is a continuing hollowing out of Lib Dem support as they retreat back towards islands of support around some of their sitting MPs.

And to the wider impact? To some degree we’re still in a holding pattern until Sunday, but the questions are straightforward enough. For UKIP it is how much of a boost they get from today’s successes (however impressive you think they actually are, the media narrative this morning is all “UKIP SURGE!”) and how long that lasts. For the other three parties it is to what extent they hold their nerve – overnight there were a couple of “Tories should have a pact with UKIP” from the Tory backbenches and a few grumbles from the Labour backbenches, but these were all from the usual suspects, the Rees-Moggs and Stringers of this world, who were saying these things anyway. Wait and see if it holds over the next few days and the Euro elections.