Friday, September 19, 2014

Scot Goes Pop! : The second fundraiser

Help to keep Scot Goes Pop going until the May 2015 UK general election.

Click here to go straight to the fundraiser page.

I'm writing this on what may be remembered as one of the most traumatic days in Scottish history - the day after we turned our back, for the time being at least, on the sovereign right to shape our own political future, and the day on which Alex Salmond announced his decision to step down as First Minister. However, the months ahead are still full of opportunities as we seek to ensure that the London parties do not betray the panic-stricken "vow" they made to the Scottish people just before polling day. If they do let us down, as they probably will, much will depend on whether the SNP can at last make a telling breakthrough on "away soil" at the UK general election in May.

In the months between now and the election, it's more important than ever that we have a strong alternative media in Scotland to put the pro-SNP case, because in the context of a UK-wide vote, the party is usually widely ignored by the London-based broadcast and print media.

Over the years, Scot Goes Pop has carved out a niche role in the pro-independence blogosphere, particularly due to its coverage of opinion polls. It relentlessly challenged the agenda-driven misreporting of polls as supposedly showing that the Yes campaign were not even in the game.

There will not be a huge number of full-scale Scottish polls before the general election, but there will be some, and with your help I hope to cover them as comprehensively as I have done the referendum polls. When there aren't polls around, I'll go back to doing what I used to do - offer general commentary on the political scene in Scotland, the UK, and further afield, and probably undertake the occasional 'fisking' of mainstream media commentators.

In case you don't know about me, my name is James Kelly, and I've written Scot Goes Pop for more than six years. I've also written articles for the International Business Times, Political Betting, Wings Over Scotland, National Collective, Scottish Roundup, Fair Observer and the Eurovision Times. Some of my IBTimes articles were syndicated on Yahoo News in the run-up to the referendum.

Since I ran the first fundraiser, I've practically been blogging full-time. Even if I take my foot off the accelerator a little (as I'll probably have to), it just won't be possible to find the time to keep the blog going properly without a second fundraiser. I'm setting an ambitious target of £5000, because that's the amount I think would just about be enough to keep things ticking over until May, and perhaps a little beyond. But anything over and above that would be enormously helpful.

If the target is significantly exceeded, I'll be able to use some of the money on advertising. In theory it might even be possible to commission our own poll, but that's unlikely because it would need the target to be exceeded by several thousand pounds.

As I said last time, if the fundraiser fails and only raises a small amount that can't possibly make any difference to my ability to keep the blog going, rest assured the money won't be wasted - I'll donate it to other pro-independence alternative media outlets.

I fully appreciate what a difficult time this is to be launching this initiative, when people are feeling heartbroken and have donated so much of their disposable income in an attempt to secure a Yes vote. I feel the timing is unavoidable, though - I didn't want to do it before polling day and potentially divert money that might otherwise have gone to the Yes campaign, but if I wait even a few more days a golden opportunity may be lost, because for obvious reasons the current readership of the blog is much bigger than it has ever been before, or is ever likely to be again.

What I am going to do, however, is run this over a much longer timescale than last time (60 days) - so if you think you might want to donate but would rather wait a few weeks, feel free to do that. And even if you're a regular reader, please don't feel under any pressure to donate AT ALL. I regard you all as friends (with the exception of the trolls!), and that's far more valuable than any donation.

Now more than ever, let's stay strong.

Click here if you'd like to donate.

And now for the good news...

The No campaign have failed to deliver on their boasts of a few hours ago that they were heading for a victory margin of 58/42, or possibly 60/40, or even greater. The BBC are in fact predicting a result of 55/45 (some would say that they authored that result as well as predicting it, but that's an argument for another day). As Murdo Fraser himself said, once the relief wears off, the London establishment will know that's a far, far tighter outcome than they could have originally expected or possibly feel comfortable with. If the Yes vote had been several points lower, and in particular if they had failed to win the symbolic prize of Glasgow, the long-term aim of another referendum in 12-20 years might not have seemed credible. As it is, we have a result that is much narrower than the 1980 Quebec referendum, which as we all know was followed by a very-nearly-successful second attempt just fifteen years later.

That's a consolation for us, but it's also a long-term threat for London, and that's the one reason for thinking we might possibly get some traction in the push for more powers. It's going to be a hell of a hard slog, though, and I think much will depend on whether the SNP can at least make some kind of breakthrough on 'away soil' in the UK general election that is just a few short months away.

One thing that intrigues me is whether the Greens will continue to support independence as a long-term goal, or will argue that the matter has been permanently settled. Although I'm not exactly James Mackenzie's greatest fan, I've been encouraged to see him make a number of comments along the lines of "if not now, next time", and hopefully that sentiment will be shared by many of his colleagues.

Although I'm finding this result as difficult to come to terms with as anyone, I'm hoping to carry on with Scot Goes Pop until at least the general election. The money from the fundraiser won't last for much longer, though, so realistically I'd have to run another one. I don't want to try anyone's patience by doing it too soon, but on the other hand I would probably be foolish to leave it too long, because the blog's readership is bound to drop sharply as interest in the referendum subsides. I'm happy to take advice on what the most sensible timing would be.

In the meantime, we can all take pride in a 45% Yes vote that once upon a time was supposed to be impossible. In a long-term sense, the dream lives on.

A triumph of fear over hope

We clearly are now facing a No vote, and as I said in my IBTimes article this morning, that will result in a national trauma that will take a long, long time to recover from.  I don't always agree with Gordon Wilson, but I think he's absolutely right that Scotland's influence within the UK will now vanish.  In the long run, the biggest questions may be for the likes of Kenny Farquharson, who convinced themselves that there was some kind of constructive way of voting No that could move Scotland forward, even in the absence of meaningful pledges from the London parties.

One battle we mustn't lose, though, is the battle for the truth of how this referendum was won by the No campaign.  I've already seen Louise Mensch retweeting a succession of comments from Tory Union Jack waving idealists, trying to weave a narrative that No voters were somehow embracing a positive message about the UK, rather than being terrorised into rejecting independence by the most negative, cynical campaign in modern British political history.

They may have won the referendum, but let's not allow them to steal the truth.  They chose fear, and won with fear, and as a result Scotland and the UK are diminished places.

Thursday, September 18, 2014

YouGov prediction leaves Yes still in the game

I can now say what I didn't dare say before the polls closed - there have been some really disturbing rumours flying around that there was evidence of a very substantial last-minute swing to No. But YouGov have now released a 'final prediction', based on recontacting respondents after they had voted, and although it's certainly not good news, it's nowhere near as bad as the figures that were being rumoured.

Yes 46% (-2)
No 54% (+2)

So obviously if Yes are going to have a chance of winning, we're going to need there to be some kind of systemic problem with YouGov's methodology. But at least, as of this moment, we're still in the game.

Want to help win a Yes vote over the next three hours?

Just thought I'd make myself useful by reposting this tweet from Ross Greer -

Still need Yes volunteers at every local campaign base tonight. Head along or call 0141 221 4767 to find out where to go.

Last night's Survation poll was better for Yes on the unrounded numbers

A small piece of good polling news as we all wait (or hopefully as we get on with the Get Out The Vote operation) - last night's Survation poll in the Record, which was one of only two polls over the last few days to report a Yes vote of lower than 48%, was actually slightly better for Yes on the unrounded figures -

Yes 47.3% (+0.8)
No 52.7% (-0.8)

Changes are from the directly comparable Survation telephone poll published on Saturday.

Meanwhile, because of the rumours that were swirling around about last night's YouGov poll in advance of publication, I was a bit worried that Yes may have only been rounded up to 48% by the skin of their teeth, but in fact that isn't the case, and it turns out that they've made a small gain on the last YouGov poll -

Yes 48.0% (+0.4)
No 52.0% (-0.4)

UPDATE : The datasets for today's Ipsos-Mori's poll have finally been released, and just like Survation, it turns out that Yes did slightly better than the rounded numbers suggested -

Yes 47.4% (-1.5)
No 52.6% (+1.5)

The above numbers should certainly ease any concerns that there was any detectable swing back to No on Tuesday or Wednesday - all of the changes are well within the margin of error, and in any case two of the three polls are showing small shifts towards Yes.

Lastly, I have a new article at the Fair Observer website, on a similar theme to my last-but-one article at the IBTimes.  You can read it HERE.  I was under pressure of time when I wrote this one, as you might just be able to tell!

Final Ipsos-Mori poll puts Yes just 3% from victory

Ipsos-Mori's last word is -

Yes 47%
No 53%

Not quite as good as yesterday's poll from the same firm, but the difference is easily explainable by margin of error 'noise' - ie. if the true position according to Ipsos-Mori's methodology is somewhere between 46 and 50, it would be completely normal to get 49 one day and 47 the next due to random sampling variation.

I hope to God this is the last poll, and we can now let actual votes decide this referendum!

One Opportunity

I have a little referendum day article at the International Business Times, reflecting on the historical importance of this moment - you can read it HERE.

The 2001 Irish Referendum and parallels to the Indyref

A guest post by Scott Hamilton

As we move into referendum day, it is instructive to look at an interesting example from the recent history of constitutional referenda - namely the Irish vote in 2001 on the Treaty of Nice. Whilst the referendum is not exactly like the Scottish independence referendum, there are certain parallels between the campaigns, how they have been run, and how the polls can spectacularly fail to predict the outcome of such a contest even a few days from the vote.

In his excellent book “A Comparative Study of Referendums: Government by the People, Second Edition", Jens Qvortrup describes the campaign leading up to the 2001 referendum and provides some insight into what caused the shock result.  Much of the book is available on the wonderful Google Books website.

All conventional wisdom pointed to a “Yes” vote- this was the position of the Irish Government of and the mainstream political parties, the mainstream media in Ireland and most “establishment” voices.  The “no” campaign comprised active “grassroot” campaigners and a band of organisations united around a simple message- to reject the Treaty of Nice.

My contention here is that the “Yes” in the Irish example provided by Qvortrup has strong parallels with the “No” in the Scottish referendum. In the discussion below, all that is required is to swap “Yes” for “No” and vice versa.

In the run up to the referendum the Irish Government was doing well in domestic opinion polls, with some 86% of the Irish population agreeing that being in Europe was good for Ireland, leading some to believe the referendum was a foregone conclusion- people would, conventional wisdom suggested, get behind the government, and vote “Yes” to ratify the treaty.

The main political parties supported the “Yes” side, bolstered by support from big business, most trades unions, farming bodies and importantly- the mainstream media. This seemed to suggest an easy victory, if only the voters could be trusted to follow the cues and signals given by the parties.

Except that’s not what happened...

The challenge to the orthodoxy came from a varied and determined set of “No” groups whose ground campaign simply outperformed the combined might of their opponents’ entire campaign. The “No” campaign comprised a variety of enthusiastic groups such as “No to Nice”, the Green Party, Sinn Fein, the Irish Campaign for Nuclear Disarmament, the Justice and Anti-Poverty Body Action from Ireland, the Peace and Neutrality Alliance, the National Platform and the Christian Alliance. There are certainly obvious parallels between the Scottish Yes campaign here.

In addition the “Yes” parties proved unable to work together with a general election looming in 2002; it seems that their lack of co-operation was prompted by mutual mistrust and party political rivalries. The “No” side was, by contrast, unhampered by such rivalry.

The “Yes” parties simply fell victim to over confidence and an over reliance on their powerful friends and a powerful media. Opinion polls had predicted an easy win for them, though as Qvortrup suggests:

“…alarm bells should have begun to ring on Saturday 2 June (Scott’s note- the vote was on 7th June) when an Irish Times- MRBI poll revealed that 45 percent intended to vote “yes”, whereas the figure two weeks previously had been 52%; some 28% indicated they would be voting “no”, and increase of 7% on the earlier poll; while the number of those undecided remained the same at 27%. Somewhat surprisingly, the politicians took little notice of the poll.

The “yes” campaign was lackluster and ineffective: the political parties did not present a unified front, other than agreeing on the necessity for a “yes” result; indeed they seemed more concerned with party-political point scoring, than with securing a “yes” vote”.

The “No” campaign was mainly undertaken using good old fashioned door to door canvassing, something that had not been the norm in Ireland for many years. The reader can decide which campaign in the Scottish referendum most closely parallels this approach.

Qvortrup concludes:

“The polls indicated that the Government was within close range of victory. The outcome would- or at least could- have been very different had the “yes” parties been able (and willing) to co-operate, but that was not the case. The ”yes” side was split by internal rivalries between the parties, and the focus on the Tipperary by-election created tension in the camp. Consequently the parties of the “yes” side failed to co-operate. The “no” side by contrast succeeded in maintaining unity- ostensibly because the parties involved were too ideologically heterogeneous to pose an electoral threat to each other”.

So, in summary - the losing side was made up of political rivals who were suspicious of each other, given an election was looming. The losing side let this rivalry negatively affect their campaign and they failed to present a united front. The losing side relied too heavily on a compliant media, and lacked an effective and broad ground campaign. The losing side took solace in the polls, believing that they had the result sewn up long before the vote, “how can we lose?”. The losing side failed to recognise the ground shifting behind their feet only days before the vote, and when they did, they lacked the tools to counter the shift in opinion.

The losing side lost.

More excerpts of Dr Qvortrup’s book are available at Google Books. See page 72 and the available pages nearby for more.

Wednesday, September 17, 2014

Final YouGov poll puts Yes just 2% away from victory

YouGov's final word of the campaign is -

Yes 48% (n/c)
No 52% (n/c)

Meanwhile, there's a phone poll from Survation which shows that the Yes vote has crept up since the equivalent poll at the weekend -

Yes 47% (+1)
No 53% (-1)

I thought we were going to get a rare Scot Goes Pop exclusive tonight, because two different commenters seemed to have leaked the YouGov result several hours early on the previous thread.  But I'm very glad we didn't get the exclusive, because the numbers were less good than we ended up with.  It does mean that I pre-wrote an entire post on the assumption that the leak was correct, and now I'm going to have to cherry-pick the bits that still make sense!  I certainly thought that YouGov was going to be our only real clue before the end of the evening as to whether there had been any last-minute swing, because their fieldwork carried on a bit longer than either Panelbase or Ipsos-Mori. If that had been the case, the message would have been "no, there hasn't", because the sequence of results that YouGov have produced over the last three weeks looks fairly stable within the confines of the standard margin of error - 47, 51, 48, 48.  But now Survation have muddied the waters with yet another completely unexpected intervention, and with fieldwork that is more up-to-date than anyone else's - it took place entirely over the last 24 hours.

There shouldn't be any dismay at all that they've produced a slightly lower Yes vote than the phone polls we've had from ICM and Ipsos-Mori, because different methodologies produce different results (as we know in this campaign of all campaigns).  It is absolutely correct to compare this poll solely with the last Survation phone poll, and doing so paints a good news story for Yes - there has been a pro-Yes swing since the weekend, albeit well within the margin of error.  There is certainly no evidence there of any last-minute swing to No.  My only slight doubt is due to the fact that the previous Survation phone poll was commissioned by the No campaign, who seemingly only made the decision to publish at the very last minute for tactical reasons.  So we can't entirely exclude the possibility that the No campaign have commissioned several Survation phone polls, and withheld all the others because they were better for Yes - in which case the trend in tonight's poll obviously wouldn't be quite so good.  But that's just wild speculation, and there's no point worrying about something that is utterly unknowable at this stage.

Unless we get any more surprise polls at a few minutes' notice, the last clue as to whether there has been any late swing will come in one further Ipsos-Mori poll, which is rather disrespectfully being published tomorrow morning while people are in the middle of voting.  But obviously 'clue' is the operative word, because unless there's a very big shift from today's poll conducted by the same firm, any change could just as easily be a margin of error illusion.  If we can dodge that bullet and convince ourselves there has been no late swing to No, then it seems to me the scenario for reconciling the polls with our hopes for a Yes victory would be as follows -

1) That phone polling is more reliable than online polling.

2) That ICM, with their "gold standard" reputation, are the most accurate phone pollster.

Both of the above held true in the AV referendum - there's certainly no guarantee that they will again, but it's an encouraging thought.  Under that scenario, it's perfectly credible to believe that Yes are at the higher end of the range that the polls are suggesting, ie. 49% - because that is what ICM showed in their only phone poll of the campaign.  If so, the race is close enough to be decided one way or the other by the ground operation tomorrow.

Alternatively, the polls could be structurally wrong, just as they were in 1992, in which case all bets are off.  But remember, it's completely 50/50 as to whether that would be a good thing for Yes or a good thing for No.

Basically, the outcome is unknowable.  And at the end of a long campaign that has seen voters being constantly bullied by having one-sided polls shoved in their faces, we should probably regard that as a truly wonderful destination to have arrived at, however scary it might feel.

Scotland stands on the brink of independence tonight as Yes vote surges to 49% in astonishing TELEPHONE poll from formerly No-friendly firm Ipsos-Mori

These are the extraordinary numbers that will be beamed to STV viewers across the country at 6pm...

Should Scotland be an independent country?

Yes 49% (+7)
No 51% (-7)

I don't think it's possible to overstate just how unbelievable a turnaround this represents for Ipsos-Mori, so often the most No-friendly polling firm in this campaign, and indeed very often the most No-friendly by quite some distance.  I vividly recall sitting in a strange location one evening at the start of March (long story) and feeling my heart sink to the floor as I spotted a tweet from Tam Britton about an Ipsos-Mori poll that, entirely counter-intuitively given the trend from other firms, saw the Yes vote fall to just 36%.  After I'd reflected on it for a few hours, my reaction on this blog was : "I think we can now safely say that Yes will not be in the lead with Ipsos-Mori by September.  But the big question is, do they need to be?"  Well, I may be proved right about the first point (there's one more Ipsos-Mori poll to come tomorrow), but it certainly isn't going to be by much.

For my money, this is the most crucial of today's polls from a psychological point of view, simply because of the "Hi, I'm John MacKay" factor.  If STV had been telling their viewers about a race that was close, but which No appeared to be winning, it might have depressed the Yes vote slightly tomorrow.  As it is, the message voters will be getting is that this is practically a coin-toss, and that (fittingly) Scotland's future is in Scotland's hands for the next 24 hours.

Technically, the fieldwork for this poll can be regarded as slightly earlier than both the Panelbase poll we've already seen, and the YouGov poll which is due tonight. because although it started on the same day as the other two (Monday), it finished a day earlier (Tuesday).  So if the YouGov poll turns out to be significantly less good for Yes, there may be speculation over whether YouGov have detected a very late swing that Ipsos-Mori missed, or whether it's just a question of differing methodologies.  If it's the latter, then there's certainly a case to be made for putting more faith in a telephone pollster that doesn't weight by past vote recall at all, thus opting out of the whole minefield of deciding exactly how that should be done.

* * *


MEAN AVERAGE (excluding Don't Knows) :

Yes 48.1% (+0.7)
No 51.9% (-0.7)

MEAN AVERAGE (not excluding Don't Knows) :

Yes 43.6% (+0.9)
No 47.1% (-0.3)

MEDIAN AVERAGE (excluding Don't Knows) :

Yes 47.8% (n/c)
No 52.2% (n/c)

(The Poll of Polls is based on a rolling average of the most recent poll from each of the pollsters that have been active in the referendum campaign since September 2013, and that adhere to British Polling Council rules. At present, there are seven - YouGov, TNS-BMRB, Survation, Panelbase, Ipsos-Mori, Opinium and ICM. Whenever a new poll is published, it replaces the last poll from the same company in the sample. Changes in the Poll of Polls are generally glacial in nature due to the fact that only a small portion of the sample is updated each time.)

A fourth successive poll within less than 24 hours puts Yes just 2% away from victory

Panelbase's final poll of the campaign :

Should Scotland be an independent country?

Yes 48% (-1)
No 52% (+1)

(UPDATE : The 5% No lead that you might have seen reported from the figures that include Don't Knows has been exaggerated by the effect of rounding.  On the unrounded figures, it's...

Yes 45.4%
No 49.5%)

The apparent swing to No is not statistically significant, and because this takes us back to the position in the last-but-one Panelbase poll (which was conducted at roughly the same time as the famous YouGov poll that had Yes in the lead), the likelihood - but not certainty - is that we're looking at margin of error noise.

Again, when it's so close we're entitled to ask the question - is there any systemic bias in Panelbase's approach that might be understating the Yes vote?  Unlike the online pollsters that reported last night, they do correct for the over-representation of English-born people in their raw sample, so we can't use that as an alibi.  But at exactly the same time as they introduced country of birth weighting, they also introduced a controversial procedure which reduces the reported Yes vote, and which no other firm uses - namely weighting by recalled European Parliament vote.  252 people who recall voting SNP in May (94% of whom are Yes voters) have been downweighted to count as just 214 people.  The logic for doing this is that the Panelbase sample always has a disproportionately high number of people who voted in May, meaning that it's therefore prudent to make sure those people are weighted in line with the result of that election.  But the problem is that if you have far too few non-voters from May in your sample, the logical conclusion to draw is either that the sample is hopelessly unrepresentative of the population you are trying to poll, or that a lot of people are not telling you the truth (or some combination of those two factors).  That being the case, you can't be sure whether weighting by vote recall is eliminating a bias, or introducing a whole new one.

The fieldwork for this poll is a bit more up-to-date than the three we saw last night - it took place between Monday and today.  So it's tempting to see this as a strong clue that we shouldn't expect any significant shift to No in tonight's YouGov poll, not least because on the unrounded numbers the Yes vote is actually slightly higher than it was in the last-but-one Panelbase poll.  The problem is, though, that Panelbase's trend has become completely decoupled from the YouGov trend - Yes first reached 48% with Panelbase months ago, at a time when YouGov were still showing them on 39-42%.  So it's anyone's guess what the "Kellner Correction" has in store for us this evening.

More details and a Poll of Polls update to follow...

Extraordinary : ICM still aren't weighting by country of birth, even though they KNOW they have 8% too few Scottish-born people in their sample

Well, the title says it all, really.  I just don't know what credibility ICM think their published voting intention figures can have when they ask for people's country of birth, and then fail to weight by that measure when it turns out that their sample is completely skewed.  Just 73.6% of their weighted sample was born in Scotland, whereas according to the census results the figure should be roughly 81.5%.  English-born people account for 16.6% of the sample, whereas it should be 9.6%.  You'd be forgiven for thinking that ICM believe this is a trivial matter, but in fact country of birth is still one of the strongest predictors of referendum vote - in this poll Yes lead by 52.2% to 47.8% among Scottish-born people, while No lead by 75.6% to 24.4% among English-born people.

My rough calculation suggests that if country of birth weighting had been applied, the No lead among the whole sample would have been just 51% to 49% - that's before the turnout weighting (which helps Yes slightly) kicks in, and that in turn would have been likely to take us to roughly Yes 50%, No 50%.

*  *  *

YouGov have published a poll for Sky News asking about issues like the currency and more devolved powers.  I'm guessing these may well have been bolt-on questions from the voting intention poll which will be published tonight, in which case it's clear from the cross-breaks that No are likely to have some kind of lead in the headline figures (but probably not a huge one).

UPDATE : It seems from what I've seen on Twitter that YouGov are still polling, so perhaps the 2000+ respondents from the Sky News poll will only make up about two-thirds of the 3000 sample that is apparently expected.  In that case obviously the calculation is thrown up into the air again - it could be better or worse.

Wisdom on Wednesday : Looking at ourselves in the mirror tomorrow

"It's no use in thinking it's too late for changing
No use in thinking that it's not up to you
You must teach, you must learn and to change things be willing
For you are your country and your country is you"

The Corries, in their song 'The Dawning of the Day'.

Tuesday, September 16, 2014

Yes vote soars to new all-time high of 48% in scintillating Survation survey

New Survation poll :

Should Scotland be an independent country?

Yes 48% (+1)
No 52% (-1)

So all three of tonight's polls are showing a 'statistical tie', meaning that the margin of error leaves open the possibility that either side are in the lead.  However, that only holds true if you look at each poll in isolation - if you take the three polls together, it becomes much more likely that No have a slender lead, albeit only if the polling firms are getting their methodology roughly right.  The latter point has been the million dollar question all along.

Malevolent John Rentoul, he of the colonialist demands that Scotland adjusts its national identity to suit his own personal tastes, is claiming tonight to have "evidence" that the polls are systemically biased towards Yes.  He says that he's spoken to someone who saw the Yes campaign's canvass returns up to Thursday of last week, and that they show Yes to be much further behind.  I'm not going to completely dismiss this idea out of hand, because it's perfectly plausible that the polls are systemically biased in either direction, and although Rentoul is a thoroughly objectionable individual, he does at least tend to apply some rigour to his journalism.  But there's something really odd about the claim, and perhaps someone with more knowledge of how canvassing is collated and analysed can offer a potential explanation.  The one thing we know about canvassing is that people don't tell the whole truth, so Yes canvassers will get better results for Yes and No canvassers will get better results for No.  It does therefore seem fantastically improbable to me that the Yes canvassing returns are worse than the results of two very recent scientifically-conducted telephone polls, including one commissioned by the No campaign itself (ie. the 46.5% Yes vote reported by Survation on Saturday).  Could there be some sort of time-lag with the canvassing figures, with old data only being very gradually replaced as people go round houses again?  Or would the centrally-collated figures always be bang up-to-date?  Are the numbers somehow adjusted to take account of people's dishonesty, so that a best guess of the true position can be arrived at?  Could it be the adjusted guesstimate that Rentoul is talking about, rather than the raw returns which surely must be better for Yes than he suggests?

Certainly on the face of it, there's a lot that doesn't quite ring true about the story, so although it's unsettling, we probably just have to work on the assumption that it's wrong (or that Rentoul is guilty of wishful thinking in the way he's interpreted the information).  So let's look at it the other way around.  Is there any obvious reason for supposing that tonight's three polls might be systemically biased in favour of the No campaign?  Yes, there is, and it's a drum I've been banging for a very long time.  As far as we know, not one of tonight's pollsters has weighted by country of birth.  It's been long-established that ICM have too many English-born people on their panel, and although we don't know that the same applies to Survation and Opinium, it seems a reasonable assumption that it might do, given that it's a problem that's been uncovered by three different online polling firms.  If so, weighting by country of birth would only help Yes by a small amount - but that's OK, because they're only behind by a small amount!

As already stated on the previous post, there's a particular issue with Opinium's decision to weight by recalled 2010 vote, which other pollsters used to do, but stopped after it became clear that a lot of people weren't recalling accurately.  Although the effect is diluted by the fact that Opinium additionally weight by 2011 vote recall, the net result of the two weightings is a reduction in the Yes vote.  If they only weighted by 2011 recall, in line with the practice of all but one of the other firms, the Yes vote would be considerably higher - and indeed there would probably be a Yes lead.

Sticking with the theme of what would happen if you applied one firm's methodology to another firm's data, Survation would be showing an even tighter race if they applied Opinium's very strict turnout filter that only includes people who are 10/10 certain to vote...

Survation (absolutely definite voters) :

Yes 48.6%
No 51.4%

There's the perennial problem with Survation that they've had to severely upweight under-25s and respondents from the South of Scotland electoral region.  When trying to guess how that might have distorted the overall results of the poll, what you're really hoping to see is (paradoxically) a very low Yes vote among those groups, because that maximises the chances that the overall Yes vote has been underestimated due to sampling variation.  In this case, it's a middling picture - the Yes vote among under-25s doesn't particularly look too high or too low, while the Yes vote in the south is maybe on the high side, but not necessarily by that much.

* * *


Swing required for 2 out of 7 pollsters to show Yes ahead or level : 0.5%

Swing required for 5 out of 7 pollsters to show Yes ahead or level : 2.0%

Swing required for 6 out of 7 pollsters to show Yes ahead or level : 2.5%

Swing required for 7 out of 7 pollsters to show Yes ahead or level : 7.0%

* * *


Yes only slip back in this update of the Poll of Polls due to the last ICM poll (which now almost certainly looks like an extreme outlier) being replaced in the sample.  But we presume that Yes are still being understated due to the fact that an ancient Ipsos-Mori poll is still being taken into account.  That problem should finally be rectified tomorrow - unless of course Ipsos-Mori's No-friendly status persists and they fail to converge in the way that YouGov and TNS have.

MEAN AVERAGE (excluding Don't Knows) :

Yes 47.4% (-0.7)
No 52.6% (+0.7)

MEAN AVERAGE (not excluding Don't Knows) :

Yes 42.7% (-0.9)
No 47.4% (+0.3)

MEDIAN AVERAGE (excluding Don't Knows) :

Yes 47.8% (-0.1)
No 52.2% (+0.1)

(The Poll of Polls is based on a rolling average of the most recent poll from each of the pollsters that have been active in the referendum campaign since September 2013, and that adhere to British Polling Council rules. At present, there are seven - YouGov, TNS-BMRB, Survation, Panelbase, Ipsos-Mori, Opinium and ICM. Whenever a new poll is published, it replaces the last poll from the same company in the sample. Changes in the Poll of Polls are generally glacial in nature due to the fact that only a small portion of the sample is updated each time.)

Yes vote increases to 48% in awesome Opinium offering

New Opinium poll :

Should Scotland be an independent country?

Yes 48% (+1)
No 52% (-1)

Remember that Opinium weight by 2010 recalled vote - a procedure that no other firm uses and that decreases the reported Yes vote.

New Survation and ICM polls are also out tonight - full analysis of all three polls will appear HERE.

Yes campaign just 2% from victory in new ICM poll

ICM poll :

Should Scotland be an independent country?

Yes 48% (-6)
No 52% (+6)

The percentage changes are of course probably highly misleading, because the previous ICM poll does now look like a very obvious outlier.  This is the joint second-highest ever Yes vote in an ICM online poll.

New Survation and Opinium polls are also out tonight - full analysis of all three polls will appear HERE.

Shall nation speak truth unto nation?

I have a new article at the International Business Times, on the subject of whether or not the BBC have had a good referendum campaign so far.  You can read it HERE.

Monday, September 15, 2014

Poll of Polls – a longer-term perspective

A guest post by Ivan McKee

With the frequency of polls increasing now it’s easy to get sucked into trying to draw conclusions from each poll release to try and determine what is happening on the ground. It can be more helpful to take a step back and have a look at the trends in the Poll of Polls over a longer time.

Using James’ excellent Poll of Polls methodology I have been keeping track of PoP trends both for the headline numbers and for the Male and Female data subsets. It’s worth sitting back and drawing some conclusions at this stage.

Looking back over time a few trends are perceptible and of interest to us.

1. NO lead in the Poll of Polls (data including Don’t Knows)

This was sitting at 21.6% back in September of last year, quickly dropped to fluctuate within a range of 19% (+/- 1%) from Oct through until Dec, this then broke-out to sit within the range 14.5% (+/- 1%) through the period Jan- March.

In April a further drop to a new ‘normal’ range of 11.5% (+/- 1.5%) was experienced which lasted right through until late August.

Currently the NO lead in the PoP is in free-fall. Almost every new poll is driving it lower and it is currently sitting at 3.5% after the weekend polls.  (Interestingly if you exclude the Ipsos MORI poll from early August the PoP across the other 6 pollsters shows a No lead of only 1.5%.)

2. Undecideds

The next interesting piece of data is what has happened to Don’t Knows.

Despite all the comments at the time about Don’t Knows making their minds up in the earlier part of the campaign the DK as measured by the PoP was remarkably stable at 17% (+/- 1%) from September 2013 right through until Feb 2104. It then dropped into the range 16% (+/- 1%) where it sat right through until August. Since then the DK number has been falling steadily and now sits at around 10.5%.

All of this reduction in headline DK has moved to Yes (of course there will be considerable churn under the surface with N to DK and DK to Y taking place).

This pattern continuing would see a further boost to Yes as the final DKs make up their minds.
The other interesting piece of evidence that supports this picture is the trend in the NO vote in the PoP. This was sitting at 51% in the PoP back in Sept 2013, but hasn’t been above 50% in any PoP since mid Oct 2013. Over that period it has gently drifted down and currently sits at 46.4%. Even allowing for the fact that the winning post might not be 50% (assuming not all DKs vote in the end) NO has to see a reverse in this long term trend to prevent being pipped at the post.

3. Gender Analysis

Perhaps the most interesting piece of the jigsaw. As we all know there has been a significant gap in voting intentions between the genders through the campaign. Analysing the separate Poll of Polls calculated for Male and Female voters shows the following :

• The gender gap in the PoP has closed, but not significantly. Female voters were 19% more NO than males in the autumn of 2013, that gap is now sitting at around 14%.

• The % of undecideds among female voters has always been higher than amongst males. Female DK in the PoP was at around 21% in autumn 2013 and has now fallen to around 12% (compared to male DK which has fallen from 14% to around 9% over the same period).

• The PoP for the male data set now shows Yes in the lead by 3.3%, while for females No is in the lead by 10.4%. There was a theory that the gender gap would close as we neared polling day (I believe there is some evidence that a 19% gender gap closed to a 3% gap by polling day in 2011). This may happen this time, and if it does then we are looking at a comfortable Yes victory on Thursday. It may be the case however that the gender gap persists (although at a reduced level) right through until Thursday.

• The other point to consider of course is the impact of the age demographic on the gender gap. There is a huge discrepancy between voting patterns of over 60s (very heavily No) and under 60s (Yes lead in most polls now). The fact that there are more female voters in the over 60 age cohort than males may have an impact that flows through to the overall gender gap. I haven’t quantified this but would be surprised if it is more than 2 – 3%.

My appearance on BBC Breakfast

A few of you might have seen me on the BBC Breakfast show this morning - I took part in a little five-minute slot with Duncan Hothersall.  Since I started writing this blog a few years ago, I've been contacted four or five times about the possibility of going on TV or radio (two of those were from Argentina!), but it never led to anything.  So it was a bit of a shock to my system to actually find myself on UK-wide television, trying to put the case for Yes just three days before the referendum, and up against an "opponent" who is a seasoned political activist and a very professional speaker.  Just to make it even scarier, Duncan had Kezia Dugdale MSP with him - I'm not sure where my support staff had disappeared to!  I'm happy to report that both Duncan and Kezia are very friendly in real life, and it's just as well I recognised Duncan from his photo, otherwise I'm not sure I'd have ever found the right place.

Ah well, I did my best - and you know you've temporarily hit the big time when Caron Lindsay is complaining about you being part of an "all-male panel" on the BBC!  (It's a perfectly fair point, of course, but not my fault.)  Unfortunately, I can't provide a link, because the Breakfast show isn't on the iPlayer.

Sunday, September 14, 2014

Panelbase poll shows Yes at 50% among the whole sample

Many thanks once again to Ivor Knox for sending me the Panelbase datasets - and the big news is that the weighted results for the whole sample are even tighter than the published results (which are filtered by likelihood to vote).

Should Scotland be an independent country?  (Whole sample, Don't Knows excluded)

Yes 49.8% (+3.1)
No 50.2% (-3.1)

If those had been the headline numbers (and there are two pollsters that don't filter by likelihood to vote) they would have been rounded up to Yes 50%, No 50%.  In absolute numbers, there were 452 respondents after weighting who said they will vote Yes, and 455 respondents who said they will vote No.  Even on the raw unweighted data, which usually favours No more, the position is closer than I can ever remember seeing in a Panelbase poll - Yes 49.2%, No 50.8%.

Incidentally, on the published headline figures that take account of Don't Knows, the Yes vote stands at 46.1%, which is higher than in any previous Panelbase poll - and that includes the poll from last September which is normally disregarded because of an unusual question sequence.

As you'll probably remember, Panelbase made two methodological adjustments a couple of months ago, one of which had a Yes-friendly effect, while the other had a No-friendly effect.  The Yes-friendly one was the introduction of weighting by country of birth - the logic for which is inescapable, given that we know online polling panels have a disproportionately high number of English-born people on them.  YouGov have since followed suit.  But the No-friendly move towards weighting by recalled European Parliament vote is much more controversial, and isn't a procedure that is used by any other firm.  It looks absolutely certain that if the latter change hadn't been implemented, there would be a Yes lead in this poll, because the 268 people who recall voting SNP in May have been downweighted to count as just 220.

Turning to the ICM poll, the most frustrating thing about the datasets is that two different sets of fieldwork dates are given - we're first told that the poll was conducted only on Wednesday and Thursday, but then a period of Wednesday to Friday is given.  The latter version was the one reported by the Telegraph (which commissioned the poll), so hopefully that's correct - obviously the more up-to-date that a poll showing Yes ahead is, the better.

Of course when I first heard that a poll had shown the Yes campaign on 54%, the first thought that entered my head was that they had probably been flattered by the rounding.  But that isn't really the case - on the unrounded numbers, the position is Yes 53.8%, No 46.2%.  ICM weight their headline figures to take account of differential turnout, which in this case helps Yes slightly, but even without turnout weighting the numbers are healthy enough - Yes 53.3%, No 46.7%.

If the No campaign are scrabbling around in the datasets for some hope that the headline numbers may have been distorted, they might focus on the slightly implausible Yes lead in the South of Scotland sample, which has been upweighted almost two-fold.  But nothing else is leaping out at me - ICM presumably realised that they wouldn't get a satisfactory sample of 16-24 year olds, so in a repeat of what they did in a poll earlier this year, they've opted out of the problem altogether by weighting 16-34 year olds together.

Finally, there's big news about Opinium's methodology, which I'd suggest calls into question the credibility of the No lead they reported last night.  Unlike every other pollster, they've weighted their results to bring them into line with how people voted in the 2010 general election.  That's in spite of the very clear evidence that there is a particular problem with false recall in respect of 2010, with many voters wrongly thinking they voted SNP because of what they did a year later.  To be fair, Opinium have additionally weighted by 2011 recall, which dilutes the problem, but it certainly doesn't eliminate it - in this poll, 258 people who claimed they voted SNP in 2010 have been downweighted to count as just 179 people.  There's no real doubt that if Opinium had followed the same recalled vote weighting procedures that most other firms use, the Yes vote in the poll would be even higher than the 47.4% reported.

It looks like neither ICM nor Opinium are weighting by country of birth, which is a deficiency that in all probability is leading to the reported No vote being inflated.